From: Arie Verveer Subject: New Zealand Epicentre location. A Journey Date: Wed, 01 Apr 1998 07:50:08 +0800 Hi, I thought it may be fun to identify the location of an event, so here is my journey. I selected David Nelson's recent M5.0 with a probable location SW of New Zealand. The files are 980331B22.DNN and 980331B22.DNZ. Then using the origin time (17:17:51.9) from these files, I loaded in the my data and performed a 1 hertz High pass filter. To my surprise I found many phases but was unable to identify them. Then using Dave's station location / time as a first approximation, I identified the "P" and "LQ" phases. No "S" phase was found but there was a large phase between the P and LQ phases that wasn't account for. That phases was eventually identified via the IDC data base as a "P" wave from Nepal. NEPAL Ml4.0 17:19:18.1 27.90N by 85.43East at 17:30:00 UTC The New Zealand "P" wave was located and the "S" marker was moved until the "LQ" wave was located. A first approximation location was found using the program "Mapit", via the new distance from my location and Dave's estimates of the location. The epicenter location was re-entered into my data and a new distance and time was extracted. The data was updated in both files and the "Mapit" program was run to identified the final epicenter. Though in Dave's files, I found that the time information was out. So entering a time correction of -70 seconds the phases were aligned. Both stations magnitude estimates were within 0.4 mag. the final epicenter ? 51.14South 170.81East at 17:17:52.0 UTC March 31/ 1998 File : 980331b.au3 The journey ends or maybe it should have never started. Arie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "David A. Nelson" Subject: Re: New Zealand Epicentre location. A Journey Date: Wed, 01 Apr 1998 21:43:56 +1200 Though in Dave's >files, I found that the time information was out. So entering a time >correction of -70 seconds >the phases were aligned. Arie explain what u mean by that ????? my clock accuracy is better than 0.5 of a second Dave Co-ordinator: New Zealand Public Seismic Network Dave A. Nelson ZL4TBN 24 Jensen St., Green Is., Dunedin, South Is.. New Zealand. http://psn.quake.net/dave/home.htm IF there indeed are other parallel universes... I can now rest in the knowledge, that in at least one of them, I am filthy rich and drive a red Ferrari _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Edward Cranswick Subject: PSN recordsection of the 29 Mar 98 M6.4 Fiji Earthquake Date: Wed, 01 Apr 1998 05:41:48 -0700 PSN- The 29 Mar 98 M6.4 Fiji Earthquake was recorded by several PSN stations world-wide (NOTE: in my haste to get this on the Web, I absent-mindedly forgot to include several other excellent PSN records of this event: I apologize for that, and I will try to do better next time) and by a portable autonomous digital seismograph (PADS) equipped with a Guralp CMG-40T broadband (0.03-20 Hz) three-component velocity sensor being tested in the basement of the USGS Geologic Hazards building in Golden, Colorado (i.e., the location of the NEIC). The raw, unfiltered timeseries of the P-waves of this event (77 s are displayed) as recorded by 6 PSN stations and by the USGS PADS are shown in the pseudo-recordsection. -Edward -- Edward Cranswick Tel: 303-273-8609 US Geological Survey, MS 966 Fax: 303-273-8600 PO Box 25046, Federal Center cranswick@........ Denver, CO 80225-0046 USA E.M. Forster said, "Only connect". _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Edward Cranswick Subject: Recordsection of the M6.4 Fiji Earthquake Date: Wed, 01 Apr 1998 05:48:41 -0700 PSN- Having forgotten to include the URL of the recordsection website http://groundmotion.cr.usgs.gov/PSN/M6_4Fiji/29mar98.htm I am posting this message again. The 29 Mar 98 M6.4 Fiji Earthquake was recorded by several PSN stations world-wide (NOTE: in my haste to get this on the Web, I absent-mindedly forgot to include several other excellent PSN records of this event: I apologize for that, and I will try to do better next time) and by a portable autonomous digital seismograph (PADS) equipped with a Guralp CMG-40T broadband (0.03-20 Hz) three-component velocity sensor being tested in the basement of the USGS Geologic Hazards building in Golden, Colorado (i.e., the location of the NEIC). The raw, unfiltered timeseries of the P-waves of this event (77 s are displayed) as recorded by 6 PSN stations and by the USGS PADS are shown in the pseudo-recordsection. -Edward -- Edward Cranswick Tel: 303-273-8609 US Geological Survey, MS 966 Fax: 303-273-8600 PO Box 25046, Federal Center cranswick@........ Denver, CO 80225-0046 USA E.M. Forster said, "Only connect". _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Re: New Zealand. Apology Date: Wed, 01 Apr 1998 21:57:50 +0800 Hi Dave, Please accept my apology if I offended you. It was an attempt to locate an epicentre and my only solution to the problem was to change your timing to fit my interpretations. It fitted very well but I was totally incorrect. There is lesson to be learnt, and I've just learnt it. What happened, was I missed the true event that fitted my data. It was an event at 17:19:16.3 UTC 51.842S by 156.7557E that was just north of Macquarie Island plus the Nepal event. I didn't record your event and assumed they were the same. The posting was written as an explanation on how the results were derived and no bad interpretations were meant. So please again, accept my apology and if you ever see me walking across a road, don't hesitate to run me over in your red Ferrari. All the Best. Arie > Though in Dave's > >files, I found that the time information was out. So entering a time > >correction of -70 seconds > >the phases were aligned. > > Arie explain what u mean by that ????? > > my clock accuracy is better than 0.5 of a second > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Edward Cranswick Subject: Re: Hello from Daly City Date: Wed, 01 Apr 1998 10:51:25 -0700 Bob- I apologize for taking so long to get back to you. First of all, please check out the following site http://groundmotion.cr.usgs.gov/PSN/Ogburn/DalyCity.htm which is incomplete, but it gives you some idea of what I was trying to do. I decided to use the opportunity raised by your questions as an exercise in learning more about what is available on the PSN. I found out that there is alot that I didn't know. Larry's program WINQUAKE is really excellent at looking at teleseismic data, something I have had little experience with. I am used to studying events within about a source-depth of their epicenter. John Lahr is used to looking at regional distance records. WINQUAKE has alot of teleseismic phase and traveltime info that I did not recognize at first. Anyway, I then got lost in Web consciousness, and I have run out of time to finish my response to you because I am leaving for Australia for three weeks this evening. I am very interested in your attempt to record in such an urban environment as Daly City, and I will get back to you more about noise and gain levels. By the way, what happened to your vertical sensor? Good luck! -Edward bob ogburn wrote: > > Hi... > > Thanks again for the meeting at Menlo Park, and I believe that USGS has > decided to stay there... > > I have records of what I believe are these events: > 4.8 98/03/05 21:49:40 36.07N 117.65W 5.0 17 mi ENE of LITTLE LAKE > 5.2 98/03/05 21:47:40 36.08N 117.62W 0.7 20 mi SSW of PANAMINT > SPRING > > Since my S-G sensor is 'lightly coupled' through the condo structure, > the records do not look that good, but they have a time signature and > energy pattern that to my untrained eye indicate that the may be these > events. > > The sensor is positioned as an E-W device. > > If I send the records (Ted Blanks format) could you look at them and let > me know what you think? > > Thanks > > Bob Ogburn > (650) 594-9700 X19 > Satake Research Institute > San Carlos, CA -- Edward Cranswick Tel: 303-273-8609 US Geological Survey, MS 966 Fax: 303-273-8600 PO Box 25046, Federal Center cranswick@........ Denver, CO 80225-0046 USA E.M. Forster said, "Only connect". _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Edward Cranswick Subject: PS. Re: poster etc. Date: Wed, 01 Apr 1998 11:33:54 -0700 Sean- I wanted to say that I am very impressed with your seismometer. When you got on the PSN-L at Xmas, I was going through my winter solistice spiritual doldrums and could not keep up with the complex technical background to the device. I am not much sharper now, but I am feeling more confident about understanding the system after you went over it at John Lahr's last fortnight. As I said in my previous email, you should definitely take your machine to the IRIS meeting in Santa Cruz this July. -Edward S-T Morrissey wrote: > > Ed, > I want to thank you for your support and interest in the PSN/school > seismometer project. Its' taken me all week to get back to you > about it, but I came back here to about a dozen "by the end of > the month" things that weren't on my list when I left. > > I noticed some information about an April poster presentation, > which seemed to be an internal USGS affair if I read it correctly, > but I was wondering IF you wanted my poster material to contribute, > or if it was even appropriate. If you want to use it, let me know. > Or if you want to post it near your AGU poster there at the USGS, > let me know, and I will mail it. > > The university here is continuing to support the idea of getting > patent protection for the hardware store seismometer, so that we > can make sure it remains available to everyone. It has been > suggested that if we agree to let a commercial company make the > sensor, that for every unit they sell at a commercial price, they > must make one unit available at cost to the E&O and PSN programs. > How to draw the line may be problematic; some schools DO have > lots of "science" money. Any suggestions? > > So I have been writing up a 15-page or so "Patent Disclosure" about > the seis. One interesting aspect is the question of "witnessing" > the progress of an invention in time; it so happens that an > operating seismograph does it just by recording earthquakes, > which are always unique in time. So the first "witnessed" > existence of the seis is the VBB record of the M7.5 Santa > Cruse Isl. (SW Pacific) last April 20, the day after I made > the prototype, when it was operating on the wooden garage floor. > > I also never got around to asking you want research you were > involved in there other than the PSN support, which I assume > is not specifically supported by the USGS ... or is it? > If you get a minute, fill me in. > > Thanks, > Sean-Thomas -- Edward Cranswick Tel: 303-273-8609 US Geological Survey, MS 966 Fax: 303-273-8600 PO Box 25046, Federal Center cranswick@........ Denver, CO 80225-0046 USA E.M. Forster said, "Only connect". _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Conti Claudio" Subject: TELESEISMIC RECORDING Date: Wed, 1 Apr 1998 20:55:32 +0200 I RECORDED A TELESEISMIC TRACE AT 18 09.50 UTC PKP a.t. 18 09 50 arr.2 SK s 18 19 20 UTC. Small Phase L . Regards Conti Claudio Syracuse (Sicily) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: The April 1st 1998 Seismograph Date: Wed, 01 Apr 1998 11:53:30 -0700 The April seismograph is finally nearing non-completion as of this date. The following is a description of the component parts of this magnificient machine. The base plate measures 3' X 6' and is 3" thick and is solid titanium. Machining of the plate was done with afew dozen diamond bits of various sizes. The urpright mast is of the same material for proper temperature co-efficient effects and was cast as one piece at a local refinery along with the base plate. The boom is made from a long extruded synthetic red ruby bole, of a diameter of 2" and 5' long. Proper cooling was done to alleviate minor 3" cracks across the diameter. The pickup and damping coils are made of solid gold #98 wire, with a thin teflon coating to prevent shorts. The coils themselves are 6" in diameter by 1" thick. The coils are encompassed by two 340 pound neodymium magnets spaced 1/2" apart. Its wise to not wear any kind of metal when servicing these units, and as a precaution food intake is severly watched for iron content. So far 346 seismologists have made the mistake of wearing steel belt buckles while checking out the magnets. The scrap iron is a visually distracting nuisance. One seismologist wore an opposing magnetic belt buckle, and some other physicists tried to spin and levitate the poor fellow over the magnets. The servicing technicians fondly call these magnets "the black holes", although they are gold plated to alleviate oxidation. Other sensors have been comtemplated but they are non-traditional. The leveling screws use rather large geared down motors which are coupled to a computer which has a program to instantly orientate the platform level. The screws rest on diamond squares. The pier is a solid block of optical grade quartz from Brazil and measured 20' X 40' X 10', and was polished by a collaboration of 459 companys in Redeye, California. The El Nino effects accelerated polishing work by the recent constant rains. The finished block measured 20' X 40' X 10', and took 24 years to complete. The polishing process was tedious and called for engineer correction calculations and surface grinding marks directly on the rough block. Subsequent correction erasures gradually polished the block over geological time. Work came to a halt when a child bumped into the optically clear block and threatened to sue them all as a public nuisance. The coil output is fed into a large computer using a 78 bit A to D card with room for 680 more inputs. The computer automatically configures all data and prints out earth maps with all depth features in 20 meg bits per inch clarity. The display continually shows on a giant screen, and with a hand held remote it is capable of displaying the instant location, magnitude and various other data relating to any earthquake in the world down to a magnitude -.00056 with a gain of one. Attenuation efforts are proceeding. Occasionally the women use the device to check out which shopping center is having a sale. The unit is so sensitive that distant supernovas will occasionally show up, due to the EMF aspects between here and there. More work needs to be done to filter out these frivilous solar system disturbances. Astronomers are not allowed into the control room. The housing for this unit emcompasses a two square mile outer shell building. Within the building are other progressively smaller building around the seismo...roughly 340 buildings or insulated "boxes" is a guess. The temperature is maintained to within ..00000023 of a degree by control electronics. Servicing is a open and shut case. Nevertheless, engineers are constantly at work upgrading the system...and hopefully next year we shall be able to report more significant and meaningful improvements. And then..............I woke up. Happy April Fools Day. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Edward Cranswick Subject: Re: Mark & Margaret (Re: NZ) Date: Wed, 01 Apr 1998 13:15:54 -0700 Dave- I have been paralyzed by not being able to find a suitable reference about the quantitative effect on noise and sensitivity of going downhole. These guys, Ralph Archuleta & Co., U.C. Santa Barbara Institute for Crustal Studies: http://www.crustal.ucsb.edu should have the info, but their website server is down. In the meantime, there is no question in my mind that going down about a meter will decrease the noise above a few Hz by a factor of two (sufficiently qualitative?). Even better is to plaster or cement the sensor to competent outcrop. I am glad you had a chance to meet Mark & Margaret and expose them to the antipodal PSN. I apologize for my belated reply. Have continued to be very busy preparing to leave for Australia for three weeks which I will tonight. Mostly I will be going to see my mother in Adelaide. Three years ago, I went with Anya & Olya and had to explain to my aged relatives why I had a Russian wife & stepdaughter -- now I have to explain to them why I don't. -Edward David A. Nelson wrote: > > At 08:36 AM 3/14/98 -0700, you wrote: > >Mark & Margaret & Dave - > > Good to hear from you. I should have left before you split, because > >everyone's questions to you have been redirected to me. I am really > >going to get started on Dave's poster today! So have a good time away > >from the orifice and enjoy NZ! > > Dave - I still haven't forgotten your question about the effect on > >seismic noise of going downhole; I haven't found any good info on the > >subject yet, perhaps, I'll hear something at the SSA meeting that starts > >Monday in Boulder. We are going to have a sort of small PSN dinner at > >Johh Lahr's house on Thursday because S-T. Morrissey will be in town for > >the SSA meeting. Good of you to be such a gracious host to M&M. > >-Edward > > Hi there, > Mark and Marg. left here 1/2 an hr after sending the mail. > they left Dunedin city the following morning heading south down the scenic > coast road. Unfortunately the weathere turned sour so they may have spent > more time in the car keeping dry than out sightseeing. > > ok on the seismometer burial Ed. i am still interested to see if it is > worth puttting the 3 - 4 ft down rather than virtually rite at the surface > as they are at present. > Hey no prob. having M&M here im pleased u let me know they were > coming this way. At least they know a bit more abt NZ tectonics and > seismology. He loved my wall maps of new zealand showing events over the > last 100 yrs or so. and quickly noted and commented on the seismic gap > along the central Alpine Fault in the South Island. This length of fault > ~300 km, is deemed likely to rupture in the next 30yrs odd if the avg > rupture time between events is upheld. > > take care hope u can visit one day u would be most welcome to stay a > few days at home here. can always find a spare bed. > > Cheers > Dave > > Co-ordinator: > New Zealand > Public Seismic Network > Dave A. Nelson ZL4TBN > 24 Jensen St., > Green Is., Dunedin, > South Is.. New Zealand. > > http://psn.quake.net/dave/home.htm > > IF there indeed are other parallel universes... I can now rest in the > knowledge, that in at least one of them, I am filthy rich and drive a red > Ferrari -- Edward Cranswick Tel: 303-273-8609 US Geological Survey, MS 966 Fax: 303-273-8600 PO Box 25046, Federal Center cranswick@........ Denver, CO 80225-0046 USA E.M. Forster said, "Only connect". _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Re: boom centering motor Date: Thu, 02 Apr 1998 12:25:01 -0600 At 06:53 PM 3/30/98 -0600, you wrote: >Meredith, > >The boom centering motor has to turn a fine thread (eg 8-32) screw >with an eccentric (so it won't turn) weight of 10 to 50 grams hanging >on it oriented along the axis of the boom. The Edmund Scientific >catalog has some nice miniature geared-down DC reversible motors >that weigh an ounce or two and draw less than 100 ma. In particular, >look at the motor part number J41331, a 2RPM, 31 inch-ounce laboratory- >grade (=$25) motor that runs on 8 milliamps at 12 V unloaded and is >1.75" dia x 1.75" long. > >THe motor can be mounted over the centerline of the hinges so as >to minimally affect the boom and the feedback coil/magnet. THere is >no problem with powering it via a pair of the fine coiled wires that >convey the signals over the hinges (I have two extra pair installed). >The fine #38 wire that I use can handle 2.5 amperes before it melts, >so a few hundred ma won't be a problem. > >The above motor can be controlled manually with a DPDT switch, and will >run quite nicely on a 9-volt battery; one could also set up an >auto-zeroing system using the integrator (= mass position) output >of the VBB feedback to control it via a meter relay (hard to find a >center-scale meter relay) or electronic comparators and cmos switches. > >I am also experimenting with electronic centering to extend the >operating temperature range. To do this, I integrate the mass position >output again with an extremely long (gt. 1000 seconds) time constant >and apply it through a limiting resistor to the auxillary 25-turn >coil above the main feedback coil. I am using a passive filter (ie >very large capacitors) to avoid adding noise. I have yet to evaluate >the change in the overall response. >regards, >Sean-Thomas Friends, If I understand the problem here correctly, it amounts to a minor trimming of the weight position to keep the horizontal beam of the seismo preperly centered and balanced. Since I think at least the leaf spring is magnetic, let me suggest a possible technique I have been using with some of my experiments. This is to use a ceramic magnet up to a foot away as a cheap and dirty and very easy to implement fine force trimmer. You position the magnet carefully back and forth until the chart output is within scale. In this way, I have been able to make external delicate beam adjustments on somewhat similar instruments mounted inside styrofoam coolers for purposes of temperature stabilization. Of course the magnetic fields involved are fairly linear at some distance and not necessarily much greater than the earth field, and so should not usually be expected to cause unexpected difficulties aside from slight temperature drift in the magnet or beam. Your motor could move the trimming magnet externally. And now my own question which is off topic for the PSN list, but likely to be known to somebody on this list, who can then respond to me privately: Does anybody know an easy way using my PC (and without spending $1000) to take a raw series of bytes representing a 256 step grey scale, like the scan lines of a TV picture and reconstruct an image? Or better yet by using my raw analog scan stream. Optical scanners do this all the time with lots of inscrutable code, but I am not a programmer and am generating the data stream at my own sweet slow rate with no coding attached except for the natural scan line breaks. Whats the simplest easy way for a relative cyber-klutz like me to turn this data into a picture -- for the purpose of my amateur friendly design for a scanning microscope? Or maybe there is just the right list to join, out there somewhere, and somebody here knows where it is. --Yours, Roger Baker _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: boom centering motor Date: Thu, 2 Apr 1998 20:54:42 -0600 (CST) Roger, A known problem with ferro-magnetic (until we find a non-magnetic spring that is not a thermometer) leaf-spring vertical sensors is that at high gain they are sensitive to geomagnetic storms and have to be shielded in mu-metal enclosures. They have even been known to shift in output from other temporary ferro-magnetic disturbances like a large hunting knife, steel-shank boots, and analog multimetres. (also, BTW, geomagnetics is the reason we use moving COIL rather than a much simpler moving magnet design for velocity sensors). I would hope that refinement of the STM design will allow low noise operation and require magnetic shielding, so the internal center- of-mass trimming motor (which WILL cause a temporary transient when it is activated) will be needed. The current STM design with sliding trim masses on the boom requires about 5 grams/centimeter/deg C adjustment when it its near the center of the boom. I will use someting like 32 grams for the moving trim weight, so the 2 RPM motor and the 8-32 threaded rod will adjust for 1 deg C in about 1 minute. The 1 deg C figure is the current operating range for an integrator output voltage limit of 6 volts. (without any other thermal compensation). Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: RADIOTEL Subject: Re: SPIDER WIRE AND STAINLESS FISHING LEADER Date: Thu, 2 Apr 1998 22:47:32 EST TO: ANYONE I would like to purchase the boom suspension wire for a Lehman seismo that I am building. I have read some good things about a product called "Spider Wire" in the PSN archives. Has anyone had definitive experience with this product and if the experience was good what weight Spider wire should I purchase. According to the above referenced archives, this product, which is really not metal, does not stretch and is super strong. I also read about stainless steel fishing leader. Based on your experience what are its properties and what size(s) or weight and brands have be utilized for a boom suspension. Jim Allen Cerritos, Ca. U.S.A. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Francesco" Subject: ANOTHER BIG EVENT IN CENTRAL ITALY Date: Fri, 3 Apr 1998 12:00:30 +0200 7.27,16 UTC Mb 5.0 7.59.54 utc Mb 4.5 Located:=20 43.29N 12.67E =20 43.23N 12.58E =20 Francesco Nucera - Osimo, Italy
7.27,16 UTC  Mb = 5.0
7.59.54 utc     = Mb=20 4.5
 
 
Located:
43.29N   = 12.67E =20
43.23N   = 12.58E =20  
 
 
 
Francesco Nucera   - = Osimo, =20 Italy 
From: Arie Verveer Subject: FIJI Island Epicentre Date: Fri, 03 Apr 1998 21:03:08 +0800 Hi, After my last very poor effort in locating an epicentre, I thought that the FIJI Island quake might be a good event to study. I don't give up that easy. I used four stations each with very well defined "P" and "S" phase and plotted the information using the program "Mapit". Station .bue 10158km from event 30.112N .. 97.891W .lc3 8506km 37.496N 122.241W .ko1 4825km 19.723N 115.991W .au1 6638km 32.008S 116.135E The point where the lines crossed, was difficult to resolve given the large distance between the stations, but the answer was reasonably close. True Event 17.4S 178.2 West Mapit's Location 17.9S 178.6 West The results are not really that bad, considering the error box in locating the stations and difficulty in finding the point at which, the lines crossed. The answer would be a great deal better with close event and a better distribution of stations. Arie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Darrell Collins Subject: Re: weather work Date: Fri, 03 Apr 1998 08:28:47 -0700 Is PSN going to be there?? Subject: Seismic Network April 2, 1998 PRESS RELEASE FIRST SEISMIC NETWORK FOR TORNADOES (SENETOR) WORKSHOP TO BE HELD 6 APRIL 1998 Seismologists and meteorologists will meet together for the first time at a workshop on 6 April at the Center for Earthquake Research and Information (CERI), University of Memphis to explore means of using existing seismic networks to detect tornadoes in real time. The workshop was organized by Dr. Frank B. Tatom of Engineering Analysis Inc., Huntsville, Alabama with CERI serving as host. Representatives from the U.S. Geologic Survey (USGS), the National Weather Service (NWS), the National Severe Storms Lab (NSSL), and the National Aeronautic and Atmospheric Administration (NASA) are expected to attend along with seismic network coordinators located in Tennessee, Alabama, Arkansas, Oklahoma and Missouri. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Charles R. Patton" Subject: Re: boom centering motor Date: Fri, 03 Apr 1998 09:25:44 -0800 Sean-Thomas, With the continuing discussion of temperature compensating the boom I have to ask: In the good old days they had temperature compensated pendulums for clocks. It consisted of brass and steel rods -- you've probably seen it and if you didn't know, would think it was for decoration -- as it was done typically in a grill pattern. I'm pretty sure the steel they used was standard magnetic, but today there are several non-magnetic materials you could use. For instance using brass at 20 ppmTCE and glass at 8.7 ppm TCE you could get positive or negative 12.3 ppm depending on construction. The question is, "Is this enough compensation range if one were using a phosphor-bronze spring?" Another fascinating number is magnesium with a TCE of 29.8 ppm. Invar is essentially zero, but I don't know is its magnetic or not, if not, this gives a combination possibility of almost 30 ppm. If I do my back of the napkin calculations right, for the current steel spring I estimate you are adjusting with an equivalent coefficient for the mass center of the boom of about 10 ppm/ degree C. So this would be in the ballpark. Left field or not, is a different question. Regards, Charles R. Patton _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: SPIDER WIRE AND STAINLESS FISHING LEADER Date: Fri, 03 Apr 1998 17:58:24 +0000 RADIOTEL wrote: > TO: ANYONE > I would like to purchase the boom suspension wire for a Lehman seismo that I > am building. I have read some good things about a product called "Spider > Wire" in the PSN archives. Has anyone had definitive experience with this > product and if the experience was good what weight Spider wire should I > purchase. ..........................snip.............. Hi, I was recommended spider wire, but was unable to find it over here (GB). Erich kindly sent me some USA wire (not spider),and I have since found that pike fishermen use multi stranded steel lines of various breaking strains. They all seem to consist of 7 strands of hard (piano wire?) wire with an outer nylon cover. I am using such a line (24 lbs breaking strain), altho I consider that a 100% nylon line to be more flexible (but it keeps on breaking!). This 24 lbs has 7 strands, each one .004" dia, so it's reasonably flexible, and strong enough. Here, ALL fishing tackle shops stock this pike wire, so you shouldn't have any trouble in USA. If you do, I will send you a some yards, (enough for a few seismos ), if you let me have your mail address, (privately of course). Incidentally, I use HEAVY galvanised wire for most of the suspension run, only using the flexible wire for the LAST 2 inches. In this way, any stretch is reduced to an absolute minimum. I have tried NON steel lines, in this way, in order to get minimum stretch. Regards Albert Noble (England) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bob@................. (Bob Hammond (USGS-AVO Geophysicist)) Subject: Re: SPIDER WIRE AND STAINLESS FISHING LEADER Date: Fri, 3 Apr 1998 09:21:49 -0900 Spider Wire is THE best fishing product for fish with many sharp teeth. I don't doubt it will be very good for boom suspension! Here in Alaska, Spider Wire is available in most fishing isles of drug stores and super- markets. I might also suggest another product for boom suspension: Phillystran is a kevlar guy wire used for antenna tower guying. It is very flexible, very strong, and doesn't stretch too much. It is available from Texas Towers (www.texastowers.com/philstrn.html). Bob APSN Fairbanks _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: STM-8 Electronics Date: Fri, 03 Apr 1998 13:18:09 -0700 Sean-Thomas, >From my limited electronics experience viewpoint, personally I think it is a much bigger problem than the mechanical sensor building aspect. Of course, that angle of viewpoint will vary with the individuals experiences. Overall...either mechanical or electronics can stop the successful completion by interested parties. I do think it would be a shame for either problem to halt the individuals project; as the seismometer has great potential. The seismo is a prototype and therefore, both mechanical and electronics will see a possible ever changing composition, which is normal and desireable for progress. Anyway...as I gather it, the electronics are a composite of borrowed and "standard modules", and (?) therefore are not a part of the ongoing patent process, or are they? Question...would it be better from a homebrew approach, to build and test the individaul schematics boards, rather than a all in one approach? To supplement this; is there simple tests known for each schematic/board? What would be a minimum pre-requistite in regard to electronic test equipment...i.e., voltmeter and DC oscilloscope? With a specific schematic; the amp/demod/filter/DC amp....I do have questions regarding unspecified values. 1. On the right top side, is the unspecified resistor......160K? 2. A1 amp....the prior input + & -, R & C, are they different for the gain of 100? 3. A5 amp....the prior input + & -, R & C, are they different? 4. A6 amp....+ input, opt R range & C value? 5. Schematic has two opposing caps, one under A1, 2 X 10uf, and another of 2 X 47uf between A5 and A6. These are polarized caps I presume? 6. The MC14066B integrated circuit with its criss-cross lines, may need "connecting dots", for absolute clarity for dummys like myself-ha. I'am not real sure...but I don't want to blow the IC, or have a potential unrecognized problem. Right off, I don't see any other problems with interpreting the other schematics. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Norman Davis WB6SHI Subject: Re: SPIDER WIRE AND STAINLESS FISHING LEADER Date: Fri, 03 Apr 1998 12:17:56 I have read some good things about a product called "Spider >Wire" in the PSN archives. Has anyone had definitive experience with this >product and if the experience was good what weight Spider wire should I >purchase. I tried it and even used 3 strands of the 12*24 and it broke. I went back to a gatar string, G I beleave. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: CVW Subject: Where do I start (building a PSN station)? Date: Fri, 03 Apr 1998 17:07:10 -0500 Okay, I finally have the resources to pursue a dream of seismic proportions-- I have a budget of $1000 to spend on equipment to set up a seismograph. Here's what I already have: 1. A PC running Windows 3.1, with standard I/O and sound card, that can be 100% dedicated to this project. 2. Some potential sites on our lot where I can bury sensors. 3. Enough knowledge to be dangerous, based on the PSN FAQ's. Here's what I don't have: 1. A seismometer 2. An op amplifier 3. An A/D converter (but doesn't the sound card work on the same principle?) Here are my ideal requirements: 1. Measure on 3 axes. 2. Data captured in PSN usable format. 3. Durable and reliable sensors. 4. Ability to bury the sensors 50-100 feet away from recorder. So, how close can I come to creating a fully functional station for under $1000? Better to try making home-made seismometers, or buying commercial geophones? (I'm OK with DIY, but my main objective is good measurement.) Any comments would be greatly appreciated. Thanks, Charles van Wynbergen Atlanta, USA _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Terence Dowling Subject: Re: Where do I start (building a PSN station)? Date: Fri, 03 Apr 1998 14:08:13 -0800 CVW wrote: > > Okay, I finally have the resources to pursue a dream of seismic > proportions-- I have a budget of $1000 to spend on equipment to set up a > seismograph. I'm in essentially the same position. If you get any direct replies (not on the net) that you feel comfortable in sharing I'd appreciate it. -- Terence Dowling (408) 536-3856 Adobe Systems Inc. dowling@......... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Seismology Bibliography Date: Fri, 03 Apr 1998 16:18:02 -0800 Hi Everyone, A few days ago Bob Barns sent me a bibliography document he had put together. I went ahead and made a web page out of it and places it on my system. A link to the page can be found on: http://psn.quake.net/infoequip.html and a direct link to the document is: http://psn.quake.net/bibliography.html If anyone has a book or article they would like to add, please send me the info and I will add it to the list. Thanks Bob. Regards, Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Jim Hannon Subject: Re: Seismology Bibliography Date: Fri, 03 Apr 1998 19:29:32 -0600 Larry Cochrane wrote: > > > If anyone has a book or article they would like to add, please send me the > info and I will add it to the list. > > Thanks Bob. > > Regards, > Larry Cochrane > Redwood City, PSN > Here are a few books in my collection that I didn't see on the list. Elementary Seismology Charles F. Richter 1958 W.H. Freeman & Company Earthquake Interpretations A manual for reading seismograms Ruth B. Simon 1981 William Kaufmann Inc. Earthquakes G.A. Eiby 1980 Van Nostrand Reinhold Company -- Jim Hannon http://soli.inav.net/~jmhannon/ 42,11.90N 91,39.26W WB0TXL _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: VRDT amp/demod Date: Fri, 3 Apr 1998 21:09:12 -0600 (CST) Meredith, Answering your questions: 1: it is 160k, as the other switch driver. 2: The amp A1 gets gain two ways: the 40.2k/4.02k input circuit, and the 10k/1k divider in the feedback, for x 100. 3; The inputs to A5 are the same. 4: At A6, the optional low pass filter is 2meg with 0.1 uf (as in the feedback). 5: When opposing capacitors are used, they are electrolytics wired positive to positive. 6: The synchronous demodulator MC14066 (erroneously labeled 4088) is a quad cmos switch connected as drawn. The convention is that leads or wires that cross do not connect UNLESS there is a dot at the crossing point. I will update the schematic and re-post it on the web. It is a recent re-drawing of the 1982 version, so I have not actually made anything directly from it. It had been converted to a printed circuit that works; maybe I should put that artwork on the web? Or make it available by mail or FAX. The oscillator also exists as a PC board. If I put the PC artwork on the web, do you think that the home practitioner could retrieve it and rescale it to dimensions with enough accuracy that the IC sockets, etc, would fit? And would a photo-copied transparency of it be dense enough to print on a photo-resist PC board? Of course, the long term goal is to integrate all of it into one package. The oscillator will drive three VRDTs, but each needs its own amp/demod. But A6 could be re-configured as the integrator for the VBB feedback (with a different amplifier). BTW: none of the electronics is subject to any patent disclosure. Also, thanks for your thoughtful suggestions for the April 1 seismometer. I'm glad you avoided trying to use superconducting magnets; liquid helium would get expensive compared to the two 340 pound rare-earth magnets. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: Where do I start (building a PSN station)? Date: Fri, 03 Apr 1998 20:46:59 -0700 CVW wrote: > Okay, I finally have the resources to pursue a dream of seismic > proportions-- I have a budget of $1000 to spend on equipment to set up a > seismograph. > > Here's what I already have: > 1. A PC running Windows 3.1, with standard I/O and sound card, that can > be 100% dedicated to this project. Opinion: Upgrade to Windows 95 for enhancements. Ditch the sound card if it is separate from the video card and you need a slot for the A/D card. > 2. Some potential sites on our lot where I can bury sensors. Sounds like a geophone preference. The Galitzin/Ewing/Columbia/ Lehman variety would give you the longer period phases OH's and AH's to check out and enjoy. > 3. Enough knowledge to be dangerous, based on the PSN FAQ's. > > Here's what I don't have: > 1. A seismometer > 2. An op amplifier Larry Cochrane has a 3 channel card. Beautifully made. > 3. An A/D converter (but doesn't the sound card work on the same > principle?) Larrys Cochranes A/D card seems to be in preference....via past email recommendations. Beautifully made again. People say they are a reliable item, which is very important. > Here are my ideal requirements: > 1. Measure on 3 axes. This is going to be the biggest time factor if you home brew, say a Lehman coil-magnet affair. The vertical will be the most difficult. > 2. Data captured in PSN usable format. Larrys programs again seen to be in preference. > 3. Durable and reliable sensors. The old and reliable coil-magnet affair is next to impossible to beat for simplicity and least cost factor. > 4. Ability to bury the sensors 50-100 feet away from recorder. Again, this sounds like a preference to geophones. A crawlspace or basement floor are much more accessible and weather protected. > > > So, how close can I come to creating a fully functional station for > under $1000? Realistically, it would give you a preamp and A/D card and maybe 500 for materials for the seismos. 2K$ over a period of time would probably be more appropriate. (I can hear the whistling spears being thrown now-ha.) I'am probably the cheapest person there ever was for this stuff....maybe 8K$ over some 34 years, but in many ways, the commercial stuff is ready made...and you don't have to STRUGGLE with acquiring every piece and part.... Personally I'd start with the Homebrew Lehman type first. Graduation with a STM-8 vertical may well be down the road of time. > > > Better to try making home-made seismometers, or buying commercial > geophones? (I'm OK with DIY, but my main objective is good measurement.) The reward with buying is a quicker result. The reward with building is learning, experience and ALOT of personal satisfaction. Being as you have the dream...you're already well on your way. Most important....have fun and enjoy it all...good and bad. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: RE: VRDT amp/demod Date: Fri, 03 Apr 1998 23:22:20 -0700 Sean-Thomas, OK and thanks for the update on the schematic. A web or mail posting of the circuit board/s would indeed be a major step in accelerating home brew utilization. I for one, would certainly do the best I could in regard to rescaling etc. efforts. Circuit layout is a time intensive subject. I've never done transparencies, but I guess they have been around for years..... Obviously the total design is yours of course. I would think that with a possible OK from you that you maybe beseiged with willing minds toward more resolutions or possiblities. I suppose with multiple IC's available in one package, that the layout could be chopped down, although the accuracy could possibly suffer? Even the oscillator can be replaced from a 2 IC, to a one unit, I think. Anyway (even though I am no engineer), I would think it could be possible to reduce the specific IC's package units to maybe ~ 6 instead of the present ~ 10 or 11. Pure speculation. I think with time that your seismo will be a standard, and the circuit aspects can be enhanced to a degree of more managability for digital dummys like myself. P.S., perhaps next years April 1st seismograph may feature an anti-gravity mass that is stabilized in space and place. Such a device could be envisioned for cracking nuts during festivities with computer control upon command...besides serving as a seismometer?-ha. Hey...imagination knows no cost boundaries or practicality.... in the mind. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Frank Cooper" Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Sat, 4 Apr 1998 01:17:04 -0600 ---------- From: Frank Cooper To: PSN-L@............. Subject: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Saturday, April 04, 1998 12:47 AM I got my Lehman seismograph sensor up and running in January, 1997, using an Esterline-Angus chart recorder and home-brew amp. In april, 1997, I purchased Larry Cochrane's 16-bit A/D card with wwv correction and his amp/filter card. I also use Larry's SDR and Winquake programs in conjunction with his hardware and my 486 dedicated computer. The results have been extremely gratifying. It is pleasing to see earthquake events displayed on my crt and my chart recorder at the same time. Until a few days ago I felt I would never cease using the chart recorder because a three hour event could be displayed on a continuous sheet of chart paper while the computer seismogram was printed out on a single sheet of unimpressive 8 and 1/2 x 11 inch paper. I have been running the chart recorder at 12-inches per hour so a three-hour event would show three feet of continuous recording. This seemed to be a much more useful display than a single sheet of compressed data using winquake. This all changed a few days ago with the revelation that I can expand eq data from winquake and print it on more than one sheet of paper. Attaching the sheets of paper together makes a more readable seismogram as well as one that is very visually impressive --- even more visually impressive than events recorded with chart paper. On the chance that not everyone is familiar with this technique I am going to relate what I did. You will not want to read further if this is old hat to you. The recording part of my seismograph station can be seen at http://members.xoom.com/hatsatv/past.html --- look for Frank Cooper, W5VID. I used the compressed data from my 3/25/98, three hour BALLENY ISLANDS REGION (7.9 mag) event file (980325a.fc1) posted at Larry's PSN files site. I expanded the data, printed it and the result was a six-foot long very impressive record of the event. I then displayed it on the back wall of my seismograph room along with the many chart recordings. After displaying the Balleny file using Winquake, I changed the y-scale from -310 to -125 in order to increase the vertical amplitude for printing purposes. After the event was displayed with the new vertical scale I clicked on the 27:08 time marker on the far left dragging the vertical line to the next time marker 55:08. I unckecked TEXT under VIEW and then printed the expanded data. I closed the displayed event and then clicked on the file 980325a.fc1 again to re-display it. I again changed the y-scale from -310 to -125. I then clicked on the 55:08 time marker dragging the vertical line to 23:08. I again unchecked TEXT under VIEW and printed the expanded data display. I did the same thing for 23:08 to 51:08, 51:08 to 19:08, 19:08 to 47:08 and ended with 47:08 to 15:08. I now had six printed sheets. On the second printed sheet I cut closely along the first time line, matching it to the first sheet, and then taped the two sheet together with non-reflecting scotch tape. The event trace matched perfectly. I did the same for the other sheets. I ended up with a six foot long impressive display of the event that was easy to see changes in wave forms, etc. It is now possible to think about retiring my chart recorder. I displayed another of my event files, 12/05/97, 971205a.fc1, EAST COAST OF KAMACHATKA (7.7mag) on the well next to my BALLENY ISLANDS event and a three floot long display of my 3/29/98, FIJI ISLANDS REGION (6.4 mag). These expanded files from Winquake may eventually replace all of my displayed chart recordings. Larry, please don't tell me all this was in Winquake documentation. Regards, Frank Cooper W5VID Friendswood, Texas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: Where do I start (building a PSN station)? Date: Sat, 4 Apr 1998 10:17:31 -0500 Charles, Since you live in Atlanta, a seismic desert like here in New Jersey, geophones or other devices which can be buryed will be a waste of effort.= = You need something which will detect teleseismic events. This means something with a period longer than 10 secs. Verticals with the same period have a better signal-to-noise ratio but are more difficult. So go= for a horizontal. After you get your feet wet with a Lehman is the time to think about verticals and force feedback rigs. The Lehman (Scientific American, July 1979, pp152-16) is easy to understand and build. Mine has a 30" boom (1/2" dia. aluminum) but I thi= nk that 24" would be better in that it would req. a smaller housing. A "dog= house" to eliminate drafts is essential. This can be made from 2" thick Styrafoam (from a lumber yard) cemented together with Elmer's. Put a heater (say 15 watts) in the top to stagnate the air in the box. Mine usually operates with a 20-30 sec. period. The sens. to teleseisms is proportional to the square of the period. My coil has 15,000 turns. The suspension wire is piano wire and I can give you all that you will ever need. = A sensitive leveling device is a great convenience--either a leg under the base with a screw having 80 tpi or a lever driven by a micrometer hea= d. The meter-movement calibrator described on Larry's home page is also a convenience in checking performance daily. Mine sits (not bolted down) on the basement floor. There is a fair amount of spurious signals from foot traffic in the house but this is no problem because these events are easily distinguished from 'quakes. I get a good recording about every 2-3 days (a dedicated 286 computer running 24 hrs/day). The farthest 'quake I have recorded was from the Indian ocean-very nearly 180 deg. from here. So I can see 'quakes anywhe= re in the world with a magnitude of ~5.5 or greater. Two or three axes woul= d be nice but I miss only a few with only 1 seismometer. I think that buying Cochrane's A/D board (and his amplifier if you don'= t know how to handle op-amps) is the only way to go because it allows the u= se of his magnificent SDR program (free) for recording. This also allows th= e use of the even more magnificent Winquake (also free) program. I operate with a sampling rate of only 5 samples/sec. This is plenty wi= th a low-pass filter at 0.08 Hz and makes the hard disk storage space reasonable (about 2 megs/day). A sharp low-pass filter is necessary to reduce the noise from microseims which peak at about 0.16 Hz. $1,000 is way more than needed. Call (908-464-6785) if you need more details. Bob Barns _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Sat, 04 Apr 1998 10:51:14 -0800 Frank I like your idea. I have a tractor feed dot matrix printer. I wonder if one can suppress the formfeed to get the recording w/o the cutting? Barry Frank Cooper wrote: >..... > line, matching it to the first sheet, and then taped the two sheet > together > with non-reflecting scotch tape. The event trace matched perfectly. I > did > the same for the other sheets. I ended up with a six foot long impressive > display of the event that was easy to see changes in wave forms, etc. It > is now possible to think about retiring my chart recorder. > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Sat, 04 Apr 1998 11:44:59 -0800 At 01:17 AM 4/4/98 -0600, Frank Cooper wrote: > [snip] > Larry, please don't tell me all this was in Winquake documentation. No, but it would be a nice feature to add.... -Larry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Frank Cooper" Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Sat, 4 Apr 1998 13:48:28 -0600 Barry and others, I use a sheet fed laser printer so I don't know if you can use your dot matrix tractor fed printer as you indicated. Think it's worth a try. Perhaps someone else will know? Frank Cooper, Friendswood, Texas ---------- > From: barry lotz > To: PSN-L Mailing List > Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM > Date: Saturday, April 04, 1998 12:51 PM > > Frank > I like your idea. I have a tractor feed dot matrix printer. I wonder > if one can suppress the formfeed to get the recording w/o the cutting? > Barry > > Frank Cooper wrote: > >..... > > > line, matching it to the first sheet, and then taped the two sheet > > together > > with non-reflecting scotch tape. The event trace matched perfectly. I > > did > > the same for the other sheets. I ended up with a six foot long impressive > > display of the event that was easy to see changes in wave forms, etc. It > > is now possible to think about retiring my chart recorder. > > > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: printer seismograms Date: Sat, 4 Apr 1998 23:39:46 -0600 (CST) Frank, Barry, et. al. Re interesting seismograms: Having a paper copy of an interesting event is a real plus for the PSN practitioner. Since I am not using any of the current ADC hardware, SDR, or Winquake, etc, maybe my relatively primitive method might be of use to some in similar situations. I am using the Radio Shack RS-22-168 multimeter, mainly because I haven't had time to actually do something about the many other possibilities I have been evaluating. It provides one channel at 1 sample/second (teleseismic speed) data at 12-bits, with lots of verbiage written to a *.txt file. I use a simple C program to reformat the output into an array that Mathcad can feed on, and use Mathcad's rather flexible graphic capabilities to plot the data, either as a single line compressed time series, or a multiple line "seismogram" of between 1 and 4 hours of data, with 10 to 30 minutes per line. All these graphics print on a single page with any printer. These can be seen at the web site at: http://www.eas.slu.edu/People/STMorrissey/index.html look st the page showing recent data. For the future, I am trying to find or create a multiple channel ADC system that specifically addresses the output of a broadband sensor. Specifically, the output of a single VBB sensor would be digitized at a high sample rate (40 sps or greater) with event detection for logging short period data (1 to 20 hz), from which would be digitally derived (FIR filtered) a parallel 1 sample / second stream continuously recorded for teleseismic events, and even a 0.1 sps stream for VLP (very long period) data. Such digitizers are widely used in the University networks of broadband stations in conjunction with 24-bit digitizers, but are very expensive ($10K or more). I would hope that we can realize a similar capability for the PSN with broadband sensors. An alternate scheme would be to run parallel SDR processes on the output of the same sensor, but with different sample rates. Maybe Larry's ADC board and SDR and Winquake can support such. I am not familiar enough with it to know. If a PSN station wanted to share data with the professional scientific community, a three component broadband system would have channels (using the SEED (Standard for Exchange of Earthquake Data: the definitions/ nomenclature should be adopeted by the PSN) named: BHZ, BHN, BHE, LHZ, LHN, LHE, VHZ, VHN, VHE. Higher sample rates (80 sps) use the SHZ, SHN, SHE channels. There are definitions for almost any physical parameter and sample rate. Regards, Sean-Thomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: RE: VRDT amp/demod Date: Sun, 5 Apr 1998 00:05:48 -0600 (CST) Meredith, I should have thought of the PC artwork long ago. With it, making the VRDT electronics becomes a plug-and-play activity. Since I still have your SASE (and new film for the Polaroid), I will send you copies of the artwork; THe problem is that the original is black lines for circuitry, and I have always photo-reversed it since the PC photo-resists I have used need clear lines for conductors. I can try to photocopy the actual PC printing film to a transparency. Regarding suggestions for changing the circuits: the design is the end result of a long effort to achieve nano-radian stability for the tiltmeters they were designed for, so we might as well expect a similar best performance for our seismometers. IC chips are very cheap, so "simplifying" a circuit to reduce the number used is not worth considering. The individual amplifiers (LM308) actually become cheaper in quantity. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Update PSN San Jose Web site Date: Sun, 05 Apr 1998 00:35:25 -0800 I have updated the San Jose PSN WEB site. http://pw2.netcom.com/~shammon1/psnsj.htm There is a section with a number of photos of the new exhibt being dedicated on April 18th at the Randall Museum in San Franciso, Living with a restless earth. Sveral PSN members donated labor and parts to create a part of the technical features presented. If you are intersted in seeing an well developed display you might take a look. Regards, Steve Hammond PSN San Jose _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Norman Davis WB6SHI Subject: Re: MAKING A VISUALLY IMPRESSIVE SEISMOGRAM Date: Sun, 05 Apr 1998 09:22:01 At 01:48 PM 4/4/98 -0600, you wrote: >Barry and others, >I use a sheet fed laser printer so I don't know if you can use your dot >matrix tractor fed printer as you indicated. Think it's worth a try. >Perhaps someone else will know? >Frank Cooper, Friendswood, Texas A dot matrix printer with tractor feed will only sheet feed when told to. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: Update PSN San Jose Web Site Date: Sun, 05 Apr 1998 11:04:51 -0600 Wow.......the pictures are a impressive graphic expression not only of the display and its educational awareness of earthquakes, but also too me of the larger hidden background of the big generiousity of the people involved in its creation. I for one would say that the PSN, has a great number of people whose contributions are evident and all over the various web sites......ranging from their pictures, programs, references and yes, a big live part is the continuing event and especially the e-mail ongoing educational process contributors. Thanks too ALL! Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Seisguy Subject: Another Quake Hits Central Italy Date: Sun, 5 Apr 1998 13:08:02 EDT Another Quake Hits Central Italy ..c The Associated Press ROME (AP) - Fans fled a soccer stadium in a small Umbrian town and tourists hurried out of St. Francis Basilica in Assisi on Sunday when another sharp earthquake struck central Italy. There were no immediate reports of injuries or damage. The quake has a preliminary magnitude of 4.5, making it weaker than the 5.0 quake that jolted the region on Friday and damaged 300 homes. Friday's quake was the strongest in several months of aftershocks following two killer quakes in September. Some 1,500 fans rushed from Gualdo Tadino's soccer stadium when the quake hit Sunday just two minutes before the end of a minor league match. Sunday's quake did no apparent damage to the Assisi basilica, whose lower- level church was filled with Palm Sunday visitors. The basilica's upper-level has been closed to the public since Sept. 26, when its ceiling collapsed during a quake. Thousands of people have been sleeping in tents or trailers since the first quakes in September punched the region, leaving 10 dead. Many have refused to go back to repaired homes for fear of the aftershocks. AP-NY-04-05-98 1302EDT _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Francesco" Subject: CENTRAL ITALY Date: Sun, 5 Apr 1998 19:31:38 +0200 Another medium quake occurred in Central Italy (Umbria-Marche border = region) 15.52.18 Utc Mb 5.1 Ml 4.7 Many material damnages I.E.S.N. PSN ITALY Francesco Nucera
Another medium quake occurred in = Central Italy=20 (Umbria-Marche border region)
15.52.18   = Utc
Mb 5.1  Ml 4.7
 
Many material damnages
 
I.E.S.N. PSN ITALY
Francesco = Nucera
From: S-T Morrissey Subject: highpass output Date: Sun, 5 Apr 1998 13:11:45 -0500 (CDT) Barry; Regarding accomodating the thermal drift caused by the leaf spring: the current operating range is about + - 1 degree C before the integrator output saturates. Here is a review of the highpass output discussion: (from previous postings) Also: the VBB output now contains the mass position component. To connect this to a recorder or digitizer (unless you have a 24-bit ADC), a long-period high-pass filter is needed, and usually an additional amplifier. I use 1000 uf, NP, into 1 megohm to an amplifier with a gain of 2 as a buffer to remove DC drift; it still passes periods of 10 minutes or more; you may want a shorter period corner with the horizontal. (see the curves of the step calibration on figure 10 of the figures page). With the unit here, the VBB output can range from + to - 6 volts (I have a +-9v regulated supply), which I monitor with a meter near the recorder so I know when the sensor needs to be recentered. It will also have a large drift anytime the covers have been removed for any length of time, and may take 4 to 6 hours to stabilize again. But the normal output from the high pass buffer is microseims of 1 to 10 mv, and the RS digitizer range is only +-200mv. So the high pass connection is a must. This near-DC response is because although the VBB output is flat to velocity from Tn up to the short period corner, it is flat to acceleration below Tn, so tilt, earth tides, acoustic gravity waves, etc, will be present in the output. From a later posting: I looked at the figure I posted of the electronics block diagram for the high-pass output configuration that is needed to remove DC drift and other ULP noise. It is conspicuously missing! oops and sorry about that. I will amend the drawing to show the large capacitor between the VBB output at the displacement detector and the buffer or "line driver" amplifier, which has a 1 megohm input resistor, which I should also show. (the drawing has been corrected to show the high pass output; the response to a calibratoin step also shows the effect of the high pass connection.) Hopefully you will be acquiring data by now. BTW: if your recorder and/or ADC has a very high input impedance (like 1 meg), you can forgo the buffer amplifier,and just install the capacitance. (I use two 2200uf capacitors connected + to + to get the 1100uf non-polarized; for trials, any large value will do) Regards, Sean-Thomas .. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: boom centering for STM-8 Date: Mon, 6 Apr 1998 00:25:50 -0500 (CDT) Re motor zeroing THe first boom centering motor has been installed on the Beta version of the STM vertical seis. It has been kept very simple: The Edmund Scientific motor part number J41331, is a 2RPM, 31 inch-ounce laboratory-grade (=$25) motor that runs on 8 milliamps max at 12 V ( 3ma with a 9-volt battery) and is 1.75" dia x 1.75" long. The centering motor shaft is coupled to a fine thread 6-32 SS rod about 8" long that simply protrudes out from the motor face. The motor is clamped to the outward end of the upper hinge angle by one if its mounting tabs. The weight of the motor is behind the hinge axis, and the threaded screw is on the boom side; they were set up so that they just about balance, so very little mechanical recentering of the boom is necessary when it is installed. The recentering weight is a small piece of aluminum angle (3/4" x 3/4" by 1/2" long, 1/16" material) with a 8-32 hole threaded into the upper half of the angle; the larger hole lets it turn freely, and also be easily relocated along the 6-32 threaded rod. The angle hangs below the rod. (so it won't turn), and the offset keeps the threads engaged; it has a weight of about 10 grams. THere is no problem with powering it via a pair of the fine coiled #38 wires that convey the signals over the hinges (I have two extra pair installed). The above motor is controlled manually with a DPDT switch and powered by a 9-volt battery. I am considering an automatic auto-zeroing system using the integrator (= mass position) output of the VBB feedback to control it via electronic comparators and cmos switches. The problem with this idea is that there has to be a long (electronic) time delay between the comparator (trigger) and the control switch to make sure it does not auto zero during an earthquake. In operating the motor while observing the integrator (mass position) output, one drives the voltage toward zero, but because of the VBB time constant, one has to stop the motor when the voltage has decreased to about half the starting value, and let the integrator catch up. Ie: if the voltage is 1 volt, run the motor until it approaches 0.5 volt, then wait. A more elaborate zeroing system would temporarily switch the integrator to a very short period during zeroing. Now we can proceed with enclosing the seismometer within a pressure tight containment to reduce barometric noise. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Looking for some help... Date: Sun, 05 Apr 1998 23:49:44 -0700 Hi Everyone, I'm slowly getting the next version of WinQuake together. One of the things I want to do with this release is add online help. To do this I will be adding Help buttons to most of the 30 some odd dialog boxes used by WinQuake. The online help will be in HTML rather then in the standard WinHelp format. Dealing with HTML is a lot easier then Microsoft's Help files. What I'm looking for is some help proofreading and editing the Web pages I come up with. You don't need to know HTML, I'll do that part of it. What I need help with is the content like spelling, grammar, suggestion etc. If you would like to help out please contact me at cochrane@............... Thanks, -Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Re: Update PSN San Jose Web Site Date: Mon, 06 Apr 1998 00:03:35 -0700 > > I for one would say that the PSN, has a great number of people > whose contributions are evident and all over the various web > sites......ranging from their pictures, programs, references > and yes, a big live part is the continuing event and especially > the e-mail ongoing educational process contributors. Thanks > too ALL! > > Meredith Lamb > I agree-- You may have noticed that I took John Lear's suggestion by adding a list of work PSN members have helped out to the web site. I started the list with things I remember. I'm sure there are a lot more. What I would like to do is have people send me one liners-- Who was helped and how. I just remembered a couple more. Ted Blank and I provide the amp and code to get two PSN stations going in a couple of middle schools in Russia. Ted traveled there and helped set them up. I also built part of a seismic system for a middle school in Wisconsin.The local science teacher built the seismograph. I also talked to Jan Froom last week, and he told me that he went back to the Jr. High school in Gilroy and spent two more days teaching seismology to 7th graders. Now there is real courage... And Ken N. ran the USGS BBS for more than a year. All stuff that should be added to the list. I need people to speak up. Please send me any info you have so I could add it to the list. Regards, Steve _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Re: Looking for some help... Date: Mon, 06 Apr 1998 00:05:14 -0700 Larry Cochrane wrote: > > Hi Everyone, > > I'm slowly getting the next version of WinQuake together. One of the things > I want to do with this release is add online help. To do this I will be > adding Help buttons to most of the 30 some odd dialog boxes used by > WinQuake. The online help will be in HTML rather then in the standard > WinHelp format. Dealing with HTML is a lot easier then Microsoft's Help files. > > What I'm looking for is some help proofreading and editing the Web pages I > come up with. You don't need to know HTML, I'll do that part of it. What I > need help with is the content like spelling, grammar, suggestion etc. > > If you would like to help out please contact me at cochrane@............... > Larry I'll help. Send me the HTML and I'll proof it. Regards, Steve _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: printer seismograms Date: Mon, 06 Apr 1998 01:51:37 -0700 At 11:39 PM 4/4/98 -0600, S-T Morrissey wrote: >An alternate scheme would be to run parallel SDR processes on >the output of the same sensor, but with different sample rates. >Maybe Larry's ADC board and SDR and Winquake can support such. >I am not familiar enough with it to know. SDR can record up to 6 channels at 100, 50, 25, 10, and 5 samples per second. Currently all channels have to be sampled at the same rate. SDR saves all data to disk creating daily record files in the ~6 meg per channel range at 50 SPS. It does have a trigger mode, automatically saving PSN formatted event files, but it uses the data saved to disk instead of data saved in RAM. When I designed SDR I decided to save all of the data to disk because I live near a freeway and I knew that I could not rely on triggered data. By saving everything to disk I could replay the data and see what I got. Since disk space is pretty cheep now a days, having a disk drive that can save 6 channels for several days is not that much of a problem. Regards, Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Kenneth J. De Nault" Subject: Re: Looking for some help... Date: Mon, 06 Apr 1998 07:40:09 -0500 Larry, I would be pleased to help but will not have time until May. Hope all is well. We have had some lovely weather following the blizzard you missed. Buck is fine and has been visiting more schools. We went to Charlotte, N.C. to race over Spring Break. I did well and finished 3rd. Ken _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: lahr@.................. Subject: http://pw2.netcom.com/~shammon1/psnsj.htm Date: Mon, 6 Apr 1998 08:38:43 -0600 (MDT) Steve, Your web site looks great! Is your section on "PSN outreach activities and link to other PSN web sites resources " available for others to add items or links? You need to add a link there to your pages on the Randall Museum. JCLahr ################################## John C. Lahr ################################# Seismologist ################################ U.S. Geological Survey ############################### Geologic Hazards Team, MS966 ############################## PO Box 25046 #############################/############################## ############################/############################### Denver, Colorado 80225-0046 ################################ Phone: (303) 273-8596 ################################## Fax: (303) 273-8600 ################################### lahr@........ ##################################### http://www.aeic.alaska.edu/lahr http://www.lahr.org/john-jan _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bob ogburn Subject: Re: Looking for some help... Date: Mon, 06 Apr 1998 07:52:56 -0700 Larry, Please place me in the reviewer pool, thanks Bob Ogburn Larry Cochrane wrote: > > Hi Everyone, > > I'm slowly getting the next version of WinQuake together. One of the things > I want to do with this release is add online help. To do this I will be > adding Help buttons to most of the 30 some odd dialog boxes used by > WinQuake. The online help will be in HTML rather then in the standard > WinHelp format. Dealing with HTML is a lot easier then Microsoft's Help files. > > What I'm looking for is some help proofreading and editing the Web pages I > come up with. You don't need to know HTML, I'll do that part of it. What I > need help with is the content like spelling, grammar, suggestion etc. > > If you would like to help out please contact me at cochrane@............... > > Thanks, > > -Larry Cochrane > Redwood City, PSN > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bob@................. (Bob Hammond (USGS-AVO Geophysicist)) Subject: Re: Update PSN San Jose Web Site Date: Mon, 6 Apr 1998 09:39:11 -0800 Steve, Here is a list of PSN-related outreach I have done and would be glad to see it on your webpage: -created a k-12 science and math education project based on in-school seismographs in urban and rural Alaska. Four schools are currently involved. Each school uses EMON and WINQUAKE software. Continue to guide students and teachers in the teaching of seismology as part of this project. -created and maintain the Alaska Public Seismic Network (APSN) webpage. -helped 3 individuals get their own seismograph stations up and running. My caveat is that I do this outreach as a USGS geophysicist as well as a PSN member. Maybe there is a distinction, maybe not. If I wasn't with the USGS, I'd be doing it anyhow. Hope this is want you want. regards, Bob Hammond APSN Fairbanks _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Amateur Seismology Educational Contributors Date: Mon, 06 Apr 1998 13:16:42 -0600 Steve Hammon, or anyone interested in history..... The problem with time passing is that denotations of their specific contributions usually get more or less lost in various ways. This can be unfortunate for those presently interested in history roles of the subject. Yes....it would be wise to do it now, and hope that a list somehow survives longer than normal. Also involved seems to be the fact that historical figures usually would rather not be pointed out....they are givers and not preferential to basking in the spot light. Nature also finds a way to renew its human resource of contributors and is as it should be. I now wish I'd have kept alot of my earlier papers and references back to the 1960's....but I didn't. At that time there was a group of individuals who communicated via mail. Mail is a poor method of reaching out and communicating over the human population, unless the subject material is such that there is no other way. The web sites and email now, are a world apart from mail. The "tools" a contributor now can use is much better....uhh...perhaps with time, more drawings and etc., would enhance the present state of computer e-mail program technology. The only individuals I knew of that were contributors in the 1960's were: Elmer Rexin....amateur water well level indicators of strong world earthquakes. Deceased. Wisconsin. Worked as a maintenance engineer for a shoe company. Communication via Charles Cox at the time. Gerald Shea...a innovative and skilled seismograph user whose approaches were amazing creations of household materials. Indiana. Worked in a educational institution. Robert Barnard...instrumental innovation way beyond normal. Washington. Electronic engineer. Endless contributions. Charles Cox...A real educational promotion person, who devoted more effort to that than even developing his own seismograph. If I gave him equipment...he would give it away to another person for their use. Amazing. Deceased. Wisconsin. I miss him alot. Anyway....life goes on... However, your present group has great contributors all over the place. I've only been tapping into PSN for maybe 6 months...but your educators are in abundance. For example in the past week emails, information has been put forth by.....(do it in one breath-ha), Frank Cooper, Robert Barns, Steve Hammond, Larry Cochran, Sean-Thomas Morrissey, John Lahr, Bob Ogburn, Roger Griggs, Arie Verveer, David Nelson, Edward Cranswick, Roger Baker, Charles Patton, Albert Noble, Bob Hammond, Norman Davis, Barry Lotz......(whew!), besides all the gobs of event seismograms which are teaching tools, and any others I may have missed. My point is, that they are all conveying messages....information, and therefore education. If you were to go back further, you would expand the list a great deal. As a comparison to the 1960's mail and the present e-mail, your email in one week would cover perhaps a years worth of mail in the "old days". Each email alone will take the indivdual an amount of time to compose from their experiences or thoughts, and is really a contribution FOR others. Monster contributions like developing programs, teaching, traveling for such, or even maintaining a computer day in and day out, are really the ultimate in devotion heights probably. So....you have a rich wealth of people....whose contributions I appreciate very much. My problem is that I don't know the total history of the individuals efforts or accomplishments on PSN. It's been around for maybe some odd 3 years. I can read the email and see the programs, but I'am sure that alot was done before that time period. Getting those individauls to do their own versions would be very nice to see.....if they can be coherced or prodded enough-ha. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: temperature compensation Date: Mon, 6 Apr 1998 21:55:44 -0500 (CDT) Charles; There are lots of available materials to try to make a zero temp. coefficient leaf spring. The STS units have such, made with two dissimilar metals; the remaining problem is that the temp. coefficient itself is temperature dependant; however, it does improve the stability by a factor of ten or more. For the homemade sensor, we have to keep focused on the less exotic stuff; I'm drawing the line at what I can find in the McMaster-Carr catalogue and Edmund Scientific. I still "cheat" a bit because I have the resources of my laboratory of the last 28 years, and I AM a packrat. Of course, everyone will make the best of their resources, and for the VBB instrument, temperature is a noise problem, and does not affect the response described by the transfer function at all. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: rayv Subject: Re: Looking for some help... Date: Mon, 06 Apr 1998 22:09:48 -0400 Larry Would be happy to assist. Ray Villemarette >Larry Cochrane wrote: >> >> Hi Everyone, >> >> I'm slowly getting the next version of WinQuake together. One of the things >> I want to do with this release is add online help. To do this I will be >> adding Help buttons to most of the 30 some odd dialog boxes used by >> WinQuake. The online help will be in HTML rather then in the standard >> WinHelp format. Dealing with HTML is a lot easier then Microsoft's Help files. >> >> What I'm looking for is some help proofreading and editing the Web pages I >> come up with. You don't need to know HTML, I'll do that part of it. What I >> need help with is the content like spelling, grammar, suggestion etc. >> >> If you would like to help out please contact me at cochrane@............... >> >> Thanks, >> >> -Larry Cochrane >> Redwood City, PSN >> >> _____________________________________________________________________ >> >> Public Seismic Network Mailing List (PSN-L) >> >> To leave this list email listserver@............. with the body of the >> message: leave PSN-L > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Long Period Horizontal Seismo Drift Date: Mon, 06 Apr 1998 21:25:45 -0600 Sean-Thomas, OK on the vrdt amp/demod circuit artwork. I'll give it a try. I'll be looking forward to results of your boom motor and pressure container info results. Actually...its irritating because for the moment,...I can't think of a question to ask (ha)...and you're going into a realm of instrumentation that very few people have entered, but it will be very interesting to hear about...and perhaps some day to follow. Perhaps...a tough question. With my horizontal Sprengether seis, (like a Ewing/Lehman etc.) they are capable of 20 second periods, and I've seen a period of 60 seconds once. Of course at some point with this they become grossly unstable naturallly, and drift all over the place, no matter how stable the pier or frame is constructed. Even a 20 second period will historically show need for a adjustment perhaps every couple weeks. I guess the real question is what is really necessary and would a shorter period like 15 seconds suffice? Big distant quakes will simply make the mass lag the subtle L wave tilt anyway....I think. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Re: Update PSN San Jose Web Site Date: Mon, 06 Apr 1998 21:32:31 -0700 Hi Bob-- got the not. I'll add it this week. Thanks for the input. Steve Bob Hammond (USGS-AVO Geophysicist) wrote: > > Steve, > > Here is a list of PSN-related outreach I have done and would be glad to > see it on your webpage: > > -created a k-12 science and math education project based on in-school seismographs > in urban and rural Alaska. Four schools are currently involved. Each school > uses EMON and WINQUAKE software. Continue to guide students and teachers in > the teaching of seismology as part of this project. > -created and maintain the Alaska Public Seismic Network (APSN) webpage. > -helped 3 individuals get their own seismograph stations up and running. > > My caveat is that I do this outreach as a USGS geophysicist as well as a PSN member. > Maybe there is a distinction, maybe not. If I wasn't with the USGS, I'd be doing > it anyhow. > > Hope this is want you want. > > regards, > > Bob Hammond > APSN > Fairbanks > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: new help files Date: Mon, 06 Apr 1998 23:18:33 -0700 Thanks John, I placed the documents I have so far on my PSN system. The following link can be used as a starting point: http://psn.quake.net/wqdocs/index.htm I now have 12 people offering to help, so I have plenty of help. -Larry At 09:00 AM 4/6/98 -0600, you wrote: >Larry, > >How about just putting the new help files on your web site with >a heading > 'Preliminary WinQuake help file under review' > 'Please forward suggestions or corrections to ...' >When the next version comes out, you can just change the >title and leave them there. > >Just a thought. > >JCLahr > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: Where do I start (building a PSN station)? Date: Tue, 07 Apr 1998 00:32:48 -0700 Charles, Sorry for not responding sooner. I've been trying too concentrate on getting the next release of WinQuake ready...Some comments below. At 05:07 PM 4/3/98 -0500, Charles van Wynbergen wrote: >Okay, I finally have the resources to pursue a dream of seismic >proportions-- I have a budget of $1000 to spend on equipment to set up a >seismograph. > >Here's what I already have: >1. A PC running Windows 3.1, with standard I/O and sound card, that can >be 100% dedicated to this project. I recommend that people have two PC systems. One is dedicated too data logging, running SDR or EMON, and the other for general computer usage and WinQuake. The data logging system, if it is going to run SDR or EMON, only needs to have DOS running on it. Both program are DOS based so you won't need to have Windows on that system. The main system needs to have Windows, preferably Win95. If you want to get real fancy, you can network the two together so event files will show up on the Windows system's hard disk. This way you won't need to use a floppy to transfer event files between the two systems. >Here's what I don't have: >1. A seismometer You will need a long period sensor. A geophone is fine for local events, but you will need LP sensors to record teleseismic events. To stay under the 1K limit you will need to build one. Or get very lucking in finding a used one. >2. An op amplifier >3. An A/D converter (but doesn't the sound card work on the same >principle?) As you know I have both available. My Web site has the current price and other info about my boards. > >Here are my ideal requirements: >1. Measure on 3 axes. I'm not sure you can get 3 LP sensors going for around $1K. Maybe one or two horizontal. You can buy used 3 axis geophone for under a few hundred dollars, but they won't work very well in your location. >2. Data captured in PSN usable format. You will need to run SDR or EMON for this... >3. Durable and reliable sensors. Thats a hard one... depends on how well you can make your sensor. I find that my SG sensor is more reliable because I don't need to re center it as often. >4. Ability to bury the sensors 50-100 feet away from recorder. The problem with home built LP sensor is they need to need monitored and adjusted now and then. If you bury it, make sure you can get at it pretty easily. > >Better to try making home-made seismometers, or buying commercial >geophones? (I'm OK with DIY, but my main objective is good measurement.) Do too cost, I think you will need to go with the home-made sensor. Again, geophones are not going to work very well in your location. You will be able to see the P wave of large teleseismic events with them, I see then on the USGS geophone sensors I monitor, but all of the low frequency wave information will be missing from your event files. -Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: Long Period Horizontal Seismo Drift Date: Tue, 07 Apr 1998 13:04:25 +0100 meredith lamb wrote: > With my horizontal Sprengether > seis, (like a Ewing/Lehman etc.) they are capable of 20 second > periods, and I've seen a period of 60 seconds once. Of course > at some point with this they become grossly unstable naturallly, > and drift all over the place, no matter how stable the pier or frame > is constructed. .......................... Hi Meredith, I have been considering the possibility of increasing the period of my seismos, by electronic means. (2 hor. currently about 20secs and 1 vert about 1.5 secs). Erich sent me an article published by Seismological Soc. of America, dated Oct 89. Volume 79 Number 5. page 1607........ Article entitled ,'A vertical equalisation circuit for increasing seismometer velocity response below natural frequency'. This describes a cct using LP filter plus an inverter with a summer, in series with SAME again, to allow a 1sec period seismo to have an effective response down to 10 secs! Only 6 op amps required. I am no great electronics wizz kid , but it should be possible, by suitable changes to C in LP filter to extend ANY seismo period, by a useful factor, without too much effort. The article shows the response from a short period seismo, used in Jemez mountains. The response includes 10-20 sec noise which an unconverted seismo would have been transparent to. You will be able to get a copy of this article from SSA no doubt. Not sure if this article has been mentioned on PSN before. Regards Albert Noble (England). _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: lahr@.................. Subject: hand-held accelerometer Date: Tue, 7 Apr 1998 08:52:07 -0600 (MDT) Dear PSNers, Does anyone have any ideas re the question, below, raised by a San Francisco area teacher? I've never tried to measure acceleration while riding a wild ride, as I'm usually just holding on for dear life and laughing. (Hope this isn't too far off topic.) JCLahr > To: evans@.................... > Subject: home-built hand-held accelerometer > > Hi John, > > How is everything going in Menlo Park, and with your > accelerometer project(s)? > > I thought you might have some ideas for this teacher, > who wrote the question below to the pinhole listserver > of the Exploratorium. > > For Great America, I would think that measuring acceleration > to + or - 1 or 2 tenths would be sufficient. Electronic > means would be great, but may be overkill, unless you > have a hand held unit that they could borrow. > > Maybe just a clear plastic tube with two springs and a > weight? > > _______ > | S | > | P | - 1.g > | R | > | I | > | N | - 0.5g > | G | > | ^ | > | mass| > | v | > | S | > | P | + 0.5g > | R | > | I | > | N | + 1.0g > | G | > _______ > > > I think you can respond to pinhole@.................. > but if that fails, please cc me if you write to > Burt Kessler, as I'm curious too and will share your > response with the list. > > Thanks, > JCLahr > ################################## John C. Lahr > ################################# Seismologist > ################################ U.S. Geological Survey > ############################### Geologic Hazards Team, MS966 > ############################## PO Box 25046 > #############################/############################## > ############################/############################### > Denver, Colorado 80225-0046 ################################ > Phone: (303) 273-8596 ################################## > Fax: (303) 273-8600 ################################### > lahr@........ ##################################### > http://www.aeic.alaska.edu/lahr > http://www.lahr.org/john-jan > > > > From john-jan@........ Tue Apr 7 07:43 MDT 1998 > > X-Sender: jlahr@................ > > Date: Tue, 07 Apr 1998 07:43:16 -0600 > > To: "John C Lahr, Lakewood, CO " > > From: The Lahrs > > Subject: accelerometer > > Mime-Version: 1.0 > > > > Date: 6 Apr 1998 20:49:54 U > > From: bcomet@.......... (Burt C. Kessler) > > Subject: accelerometer > > > > One of the new teachers at my school is looking for instructions for > > building some simple accelerometers in preparation for physics day at Great > > America. Can anyone help? > > > > Burt C. Kessler > > > > A decision is an action you must take when you have information so incomplete > > that the answer does not suggest itself. - Arthur Radford > > > > > > -- > > > > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: grc@............ (Gary Chantler) Subject: Re: hand-held accelerometer Date: Tue, 7 Apr 1998 08:23:46 -0800 Hi, Paragliders use variometers all the time. they keep track of place, time of flight, distance, rate of climb, etc - and are light and small as weight is a consideration. They are not to expessive if you get one used...if this helps at all. GRC *********************************************************** ***************************************** Gary R. Chantler Instructional Tech. II Washington State University (509) 335-5353 Webster Rm 922 gchantler@....... *#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#* _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: lahr@.................. Subject: hand-held accelerometer Date: Tue, 7 Apr 1998 11:43:31 -0600 (MDT) I'll pass along John Evans' response on this. Also, Roger Baker suggested this as a good use for the "seismometer" that the Amateur Scientist column by Shawn Carlson in Scientific American carried a few years back. It wasn't sensitive enough to record any but the largest local earthquakes. JCLahr > From evans@.................... Tue Apr 7 10:59 MDT 1998 > Date: Tue, 7 Apr 98 10:01:07 PDT > From: evans@.................... (John Evans) > To: lahr@.................. > Subject: Re: home-built hand-held accelerometer > Cc: evans@.................... pinhole@.................. bcomet@.......... > > John, > > Finally solving the communications problems to get my prototypes into > the field. Took far longer than expected (doesn't it always?) and I > now know much more than I ever wanted to about TCP/IP/UDP and RS-232). > > I like your idea of the spring tube--very direct and visual and easily > built. Could be calibrated against gravity easily too. Might want to > use a chunk of scrap TEFLON for the mass so it will slide easily (add a > bolt through the center to secure the springs and increase the mass). > Teflon (R) works very easily (too easily), but NEVER get it really hot > (in other words, don't try to melt it)--LETHAL gasses result (envision > breathing through Teflon-coated lungs). A good plastics supplier or Si > Valley surplus house could supply bulk Teflon (R) for minor cost (or > free with a little talking?) in the small amounts needed. > > OSH has plenty of springs--you'll want sloppy ones. > > On the electronic side, I have one accel built for outreach--it could > be borrowed. It is nominally 1 V/g, but has a 0-g offset and has a > temperature dependent gain (and a temperature sensor to help). It is > VERY quiet (far quieter than they need on a roller coaster) but is a > bit tricky to use because of the offset and temperature dependence. > Also gets nonlinear somewhere between 2 and 3 g, so the wilder rides > might not measure accurately. > > For a more permanent (and less tricky) solution I suggest that they > talk to Analog Devices (contacts below) which I suspect would comp or > low-cost a few of their ADXL-05 or the newer ADXL-202 accelerometers. > These are 5-g devices. The latter may not be available yet--I have an > engineering sample sitting on my desk here. The -05 is a one-axis > accelerometer and the -202 a two-axis version, each with noise of a few > percent of g peak-to-peak over the frequencies of interest. Plenty > good enough for roller coaster physics demonstrations, and have been > used for that purpose. Would need power, a little wiring, and a > voltmeter to read the output. (Both my ICS-3028 box and the ADXLs are > micromachined silicon devices, and very tough--good for kids to use. > However, none would like being dropped onto concrete, since that type > of shock actually reaches thousands of gs, and these typically are > rated to 2000 g shock resistance (don't try this with an FBS-23!).) > > Analog Devices: > 804 Woburn St > Wilmington MA 01887-3462 > 617-937-1534 jim.doscher@.......... > 617-761-7436 christophe.lemaire@.......... > (You could tell either one that I sent you. My most recent contacts > are with LeMaire, who gave me this -202 sample.) > > Teflon (R): > Surplus house: HalTek Electronics, 1062 Linda Vista Av, Mtn. View > (650-969-0510) (101-->Shoreline-->toward mountains, away from > Bay--> left at first light-->down to Linda Vista Av-->left to > Haltek, which is on your left). They have scrap Teflon (R), > including some large chunks. Alternatively, try Commercial > Plastics and Supply Corp. in Santa Clara near 101 and De La Cruz > (408-988-6500). They might comp a scrap piece big enough (they > did comp part of the Teflon (R) I bought for the last Open > House--a urethane-foam fault model that needed a slippery > lower-crust and slide blocks). > > Good luck, > John > evans@................... > > Cc: Burt C. Kessler > pinhole@................. > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Walt Williams" Subject: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Predicted Date: Tue, 7 Apr 1998 11:42:39 +0000 Hello All, I am a member of the SETI-L reflector. This interesting post is thought provoking. Best Regards Walt Williams, 98.04.07 dfheli@.............. == Cross Posted From seti@....... == ------- Forwarded Message Follows ------- Reply-to: "Ron Blue" From: "Ron Blue" To: Cc: "Ross Tessien" Subject: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Prediction; Date: Tue, 7 Apr 1998 01:55:45 -0400 Ross Tessien has asked for help in monitoring the likely black hole merger estimated to occur April 31, 1998. Any reports should be forwarded to him. Naturally, data before, during, and after would be useful. Also as a intuitive guess, if the timing is good, radio signals from your favorite SETI source could be stronger around this time. Ron Blue >>>>>> -----Original Message----- From: Ross Tessien To: Ron Blue Date: Tuesday, April 07, 1998 12:09 AM Subject: Re: Aether Tectonics: Alert: Merging Black Hole Prediction; >>I wish you luck Ross. Action is a rarer event than the merger of two back >>holes at least you will be able to sleep at night knowing that >>you make an effort. >>Would changes in radio signals occur as two black holes merge. I belong >>to SETI and would be willing to encourage support from the group? >>Ron Blue > >Yes. You ought to encourage any observations possible. The amplitude of >the radio energy is modulating on the order of 18 hours I think it is. You >could get frequencies and directions from astronomy groups and catalogues. >The galaxy name was in the original post, (I don't have it right now), Seyfert galaxy IRAS 18325-5926 >and >you can look that up on the Skyview Virtual Telescope by just typing in the >name, and the coordinates will come up. But I imagine that the guys >pointing the radio antennae know how to locate a galaxy. > >Anyway, any observations ought to be encouraged. > >Personally, I would give it a slim chance (but not zero), that we will >observe something really dramatic. ie, the ejection of a quasar from the >galaxy core. This is because black holes, IMO and according to the >fundamental models of forces means of transferring action via solitonic >interactions, leads to the conclusion that a BH has a core inside, and not >a singularity. Thus, a BH merger interupts the inflow of aether, ramming the >cores, and one or the other ought to breach! > >In any case, I certainly expect that fireworks are to be expected. If one >or the other is simply swallowed, I am not going to cry. But if some >radical things happen that cannot be explained I will have a smile on my >face ;-) > >Ross Tessien > From: baez@............ (John Baez)Newsgroups: sci.physics.research Subject: Merging black holes?Date: 5 Apr 1998 20:31:01 GMT Organization: Department of Physics - Penn State Univ.Lines: 73 Approved: bunn@...................... (sci.physics.research) Message-ID: <6g8pm5$q2t$1@..................> NNTP-Posting-Host: pac2.berkeley.eduContent-Type: textOriginator: bunn@pac2 Xref: oronet sci.physics.research:7648 Is anyone here familiar with the recent observations of the Seyfert galaxy IRAS 18325-5926 which could be interpreted as indicating a massive black hole binary system that would merge on or before April31st? Apparently in December 1997 researchers observed X-ray emissions with a period of 11 hours, significantly shorter than those seen previously in March 1997, and this February they saw a periodicity of 7.8 hours. They regard it as a priori "very unlikely" to stumble like this upon in-spiralling black holes right before they collide, but they raise the possibility because it would be very interesting if true. (Unfortunately LIGO is not yet up, but I hear that some prototype interferometric gravitational wave detectors will be used to try to detect this possible collision.) For more information, see the telegram posted by the IAU: http://cfa-www.harvard.edu/iauc/06800/06835.htmlas well as the following papers: > >>Mike what does you model say about these events? How about your model >Jerry? Would a tide wave of Aether have significant effect on earth even >though the distance is huge? >Ron Blue A post of mine was rejected from spr. this is my retort. In response to your question, it has already been measured that the earth is exposed to incessant free oscillations. These are not of earthquake origin, and so are either due to wind, waves, or gravitational wave excitation. No other sources are known. Will the earth be damaged, I think not. Is it possible that sensitive instruments might be able to detect the signals? I think so. do we have those instruments in place now? probably not but we have some instruments that might be barely able to detect these waves. Later, Ross Tessien I include ****paper abstracts below*****, a Stanford web site, and logic pointing out that you should post the ideas because they are valid and have already been attempted. The researchers, and others, may not be aware of the potential BH merger. It may well be that the amplitude of the resulting waves are below the sensitivity limits of any of these schemes. I am not saying that a run of the mill, "seismograph", has the sensitivity to detect a gravitational wave. I am telling you that the global network of seismographs, when the data is processed collectively create an interference pattern which can detect, in principle, graviational waves. There are a number of experiments that have attempted to detect gravitational waves using seismometers and the earths motions. These are not your run of the mill earth quake monitors or tecniques. Also, there are other superconducting seismographs that are more sensitive than most and perhaps these might detect a signal. The point is, it is unlikely that a black hole merger would occur again in our lifetimes. Thus, it is prudent to act now and ask questions later. If the idea is stupid, then fine, blame it on an ignorant crackpot and you are home free. But the fact is, physicists have indeed attempted to detect gravitational waves using high tech seismometers and by using the motions of the entire earth. I include also a commentary on the earth's free oscillations at the end so that you can see that the geo physicists have a lot of experience subtracting earthquake generated noise. So much so that they can monitor the earths incessant free oscillations. Now gravitational waves, would show up as incessant free oscillations, if they were originating from celestial sources that are orbiting one another **assuming that the waves were sufficiently large to induce the earths motions**. However, remember that when you excite a bell with a frequency that the bell is sensitive to, you are dealing with an underdamped resonance. Therefore, if the frequency of the BH merger oscillations passes through a normal mode of ringing of the earth, then the earth will behave as an underdamped oscillator, and the signal will be amplified. Ross Tessien In the following document, the knowledge of the direction and or timing of the source signal would help in detecting it from the noise. And improvements in the timing accuracies of the seismometers used could further improve the potential for detecting a signal. http://sepwww.stanford.edu/public/docs/sep75/ray1/paper_html/index.ht ml Detecting gravitational waves using seismic data Ray Abma Author has no known email address ABSTRACT Several years of seismic data acquired by the international deployment of accelerometers (IDA) are used in an attempt to detect gravitational waves radiated from low-frequency astronomical binary sources. These data were preconditioned with various combinations of earthquake removal, gain, and tapering. Directionally weighted spectra were calculated for 0.012 to 0.556 milli-Hz (one day to one-half hour periods) in 84 directions, one year at a time, with correction for the earth's rotation. While patterns similar to the synthetic response from a gravitational-wave source were observed, these patterns showed no consistency when compared between years. This inconsistency suggests that the patterns are noise and not the result of gravitational wave excitation by binary sources. And also see: Progress on low-frequency active vibration isolation for gravitational wave detectors. S. J. Richman, J. A. Giaime, D. B. Newell, R. T. Stebbins, P. L. Bender, J. E. Faller (JILA, University of Colorado.) The preliminary stage of a three-stage active vibration isolation system now includes sensors that measure the position of the platform with respect to the ground. We have developed a control strategy in which these new sensors dominate the feedback signal at frequencies below 0.3 Hz, and seismometers mounted on the platform dominate above 0.3 Hz. The preliminary stage now achieves the same isolation performance as previously reported, and is also tilt-stabilized against the first main stage actuators. These new sensors and new flexure ends for the first main stage support springs have allowed all control loops on the preliminary and first main stages to be closed simultaneously. Forced vibration transfer function measurements show isolation factors of at least 3\times10^3 from 1-10 Hz in both vertical and horizontal directions. The results of these measurements show agreement with our computer model of the performance of the stacked system. [K11.04] Low-Frequency Signal Extraction from Interferometers Using Balanced Heterodyne Detection Scheme with Orthogonally-Polarized Signal and Local Oscillator Beams Ke-Xun Sun (Apollonics Technology, 632 Des Moines Place, San Jose, CA 95133) Balanced heterodyne signal extraction schemes are proposed and analyzed for both Michelson and Sagnac interferometers for gravitational wave detection. In these schemes, local oscillator and signal beams are introduced in orthogonal polarization states, and mixed after exiting the interferometers. The local oscillator freqeuncy can be shifted for heterodyne detection by phase modulators. Extended from a previous measurement at a high frequency (91.9 MHz), low frequency signal extraction is experimentally demonstrated with a zero-area Sagnac interferometer. The local oscillator was frequency-shifted by a post-interferometer phase modulator. At a heterodyne frequency of 96 MHz, signals generated by displacing interferometer arm mirrors were detected in the frequency range from 50 Hz to 10 kHz. Detection of signal was robust against laser amplitude and frequency noises. Low noise interferometric seismometer readouts in an active vibration isolation system for gravitational wave detectors. J. A. Giaime, S. J. Richman, R. T. Stebbins, P. L. Bender, J. E. Faller (JILA, University of Colorado.) Within the frequency range of significant servo gain, our actively controlled seismic isolation platform's noise floor is limited by seismometer readout noise. A new Michelson interferometer readout has been designed and tested which should allow a displacement noise spectral density of approximately 3\times 10^-15m/\sqrtHz. Light for 12 readouts is supplied from a quiet laser though optical fibers, and corner-cube interferometer mirrors allow simple alignment and signal stability. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: W6JRF Subject: Re: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Predicted Date: Tue, 7 Apr 1998 15:01:24 EDT Let's see, as I remember, the old grade school mind jogger went something like... "thirty days hath september, april, june and november..........April 31 huh? _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Stephen & Kathy Subject: re SETI Date: Tue, 07 Apr 1998 13:28:58 -0700 If you drop the as off the end of the first link it works and the date there says april 21!!!!!!! If you add ml to the end of the second link it works and you go to the abstract!! SKM Pilot Hill _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Charles R. Patton" Subject: Re: hand-held accelerometer Date: Tue, 07 Apr 1998 13:03:58 -0700 John Lahr posted: > > > One of the new teachers at my school is looking for instructions for > > > building some simple accelerometers in preparation for physics day at Great > > > America. Can anyone help? > > > > > > Burt C. Kessler I'd like to suggest something very simple. Take a piece of plastic strip like a 1" wide piece of a plastic divider from a notebook, or a vertical slice from a plastic milk bottle or pop bottle. A "straw" the diameter of pencil lead from a plastic broom is still another possibility. In any case, use one of these as a horizontal spring. Take a small cardboard box and cut a square hole in it. Tape one end of the "spring" in the box, horizontally across the hole with the end of the spring adjacent to the edge of the hole on the other side. Now depending on the strength of the "spring" you can add weights, such as pennies, marbles, lengths of wire, etc. one unit of weight is so much deflection, and then is marked on the edge of the hole opposite the "spring". 10 units is so much more. If you leave the 10 units attached permanently, you have just marked a scale in 0.1's of a G. Turn upside down and calibrate in other direction. Now you have a plus and minus G meter. Refinements are to cover the hole with Saran wrap in order to prevent drafts from bothering it -- those 30+ MPH winds on the roller coaster can play havoc with the calibration. Cutting the hole with a radius to match the "spring" will improve accuracy. Use of a strip of plastic rather than a "straw" will confine its reaction primarily to be perpendicular to the plane of the "spring", or even better, two "straws" spreading apart to the attachment point -- a triangle when viewed from above. The side rigidity would be excellent this way so sine/cosine vector resolution of forces could be attempted. Cheap, fast, needs no machining, test equipment, electric power, simle to calibrate, and uses materials found around the house. -------------------------------------- | box | | --------------------------- | | | | | | | | | | S | plastic | | | C | "spring" | tape| | A |==X========================XXXX| | L | X | | | E | weight(s) | | | | | | | | | | | --------------------------- | | | | | -------------------------------------- Regards Charles R. Patton _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Pinhole Listserv" Subject: Re: Re: hand-held accelerometer Date: Tue, 7 Apr 1998 15:31:18 -0700 (PDT) No action was taken on your mail by this listserver. This listserver supports the following commands. These commands should be the text in the SUBJECT field of your mail message. Make sure they are in the SUBJECT field!!! subscribe Your address will be added to the list of subscribers. You will then be able to send messages to this list that will be forwarded to all other list subscribers. subscribe digest Your address will be added to the list of subscribers who receive a digest instead of each forwarded message. You will be able to send messages to this list, and will receive a digest of accumulated messages once a day. digests A list of available digests will be returned to you. This command will only work if you are already subscribed to the list. unsubscribe Your address will be removed from the list of subscribers. You will no longer be able to send messages to the members of the list. This command will only work if you are already subscribed to the list. help This help message will be returned. -------------------- Original Message Follows -------------------- John Lahr posted: > > > One of the new teachers at my school is looking for instructions for > > > building some simple accelerometers in preparation for physics day at Great > > > America. Can anyone help? > > > > > > Burt C. Kessler I'd like to suggest something very simple. Take a piece of plastic strip like a 1" wide piece of a plastic divider from a notebook, or a vertical slice from a plastic milk bottle or pop bottle. A "straw" the diameter of pencil lead from a plastic broom is still another possibility. In any case, use one of these as a horizontal spring. Take a small cardboard box and cut a square hole in it. Tape one end of the "spring" in the box, horizontally across the hole with the end of the spring adjacent to the edge of the hole on the other side. Now depending on the strength of the "spring" you can add weights, such as pennies, marbles, lengths of wire, etc. one unit of weight is so much deflection, and then is marked on the edge of the hole opposite the "spring". 10 units is so much more. If you leave the 10 units attached permanently, you have just marked a scale in 0.1's of a G. Turn upside down and calibrate in other direction. Now you have a plus and minus G meter. Refinements are to cover the hole with Saran wrap in order to prevent drafts from bothering it -- those 30+ MPH winds on the roller coaster can play havoc with the calibration. Cutting the hole with a radius to match the "spring" will improve accuracy. Use of a strip of plastic rather than a "straw" will confine its reaction primarily to be perpendicular to the plane of the "spring", or even better, two "straws" spreading apart to the attachment point -- a triangle when viewed from above. The side rigidity would be excellent this way so sine/cosine vector resolution of forces could be attempted. Cheap, fast, needs no machining, test equipment, electric power, simle to calibrate, and uses materials found around the house. -------------------------------------- | box | | --------------------------- | | | | | | | | | | S | plastic | | | C | "spring" | tape| | A |==X========================XXXX| | L | X | | | E | weight(s) | | | | | | | | | | | --------------------------- | | | | | -------------------------------------- Regards Charles R. Patton _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: lahr@.................. Subject: Re: Re: hand-held accelerometer Date: Tue, 7 Apr 1998 17:44:13 -0600 (MDT) "From: "Pinhole Listserv" " "No action was taken on your mail by this listserver." Woops, looks like the pinhole list requires subscription. I'll send the psn suggestions in from my home account, which is subscribed. Thanks for the help! JCLahr _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: VRDT osc ICs Date: Tue, 7 Apr 1998 19:25:54 -0500 (CDT) Meredith, I just checked the Newark catalog re the ICs: (800 463 9275; www.newark,com) On page 812; Harris Semiconductor.: CD4047BE Cmos oscillator $0.87 ea CD4066BE Quad switch 0.72 CD4018BE Counter/divider 0.97 Also: from Motorola: (look in price list) MC14018BCP $1.00 MC14066BCP 0.75. The Statek quartz oscillator was needed for the long-term (weeks to months) stability of the tiltmeters in Alaska, and is probably over-kill for the seismometer; use the 4047; I do believe that mechanical thermal noise will far ourweigh thermal drift of the cmos oscillator (but DO use 1% resistors and a polystyrene capacitor at pin 1 of the 4047.) Also, the last time I needed LM308As, Newark had them even though they were not listed. I will try to see if one of the Analog Devices "OP" series would be an affordable replacement (with FET inputs). regards Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Charles R. Patton" Subject: Re: hand-held accelerometer Date: Tue, 07 Apr 1998 17:46:14 -0700 (My apologies if this went twice, but I received no response from the psn-l@............. address, so I'm re-sending to the webtronics address) John Lahr posted: > > > One of the new teachers at my school is looking for instructions for > > > building some simple accelerometers in preparation for physics day at Great > > > America. Can anyone help? > > > > > > Burt C. Kessler I'd like to suggest something very simple. Take a piece of plastic strip like a 1" wide piece of a plastic divider from a notebook, or a vertical slice from a plastic milk bottle or pop bottle. A "straw" the diameter of pencil lead from a plastic broom is still another possibility. In any case, use one of these as a horizontal spring. Take a small cardboard box and cut a square hole in it. Tape one end of the "spring" in the box, horizontally across the hole with the end of the spring adjacent to the edge of the hole on the other side. Now depending on the strength of the "spring" you can add weights, such as pennies, marbles, lengths of wire, etc. one unit of weight is so much deflection, and then is marked on the edge of the hole opposite the "spring". 10 units is so much more. If you leave the 10 units attached permanently, you have just marked a scale in 0.1's of a G. Turn upside down and calibrate in other direction. Now you have a plus and minus G meter. Refinements are to cover the hole with Saran wrap in order to prevent drafts from bothering it -- those 30+ MPH winds on the roller coaster can play havoc with the calibration. Cutting the hole with a radius to match the "spring" will improve accuracy. Use of a strip of plastic rather than a "straw" will confine its reaction primarily to be perpendicular to the plane of the "spring", or even better, two "straws" spreading apart to the attachment point -- a triangle when viewed from above. The side rigidity would be excellent this way so sine/cosine vector resolution of forces could be attempted. Cheap, fast, needs no machining, test equipment, electric power, simple to calibrate, and uses materials found around the house. -------------------------------------- | box | | --------------------------- | | | | | | | | | | S | plastic | | | C | "spring" | tape| | A |==X========================XXXX| | L | X | | | E | weight(s) | | | | | | | | | | | --------------------------- | | | | | -------------------------------------- Regards Charles R. Patton _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: RADIOTEL Subject: Re: POSSIBLE EARTHQAUKE Date: Wed, 8 Apr 1998 00:16:06 EDT I am testing out my new verticle seismograph. I recorded, in the Southern California area, what appears to be a quake occuring about 17:46:11 UTC. I have checked the finger servers and they do not show anything occuring around that time. Did anyone else record anthing around the above referenced time period? Jim Allen Cerritos, California _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: MC14066 Date: Tue, 7 Apr 1998 23:54:23 -0500 (CDT) Meredith: Another typo !!! : the demodulator schematic lists the 4066 (a quad cmos switch used as a synchronous demodulator) as a 4088, (which is probably a mythical device.) I guess I expected a person assembling the electronics would look up the spec sheets on the devices used to see how they are used. Maybe I should have posted the original 1982 schematics rather than redrawing them. .. Also: the 4018 IS a 4018 on the oscillator here in my hand. I believe the 4016 is a similar device, but not the one used. Jim H. has pointed out an AD698 device that will run the VRDT (probably) all by itself. It is power hungry (+-15V at 15ma, or 0.45watt), as are other options, which is why I did not originally use them. (the whole tiltmeter system had to run from batteries for one year on a total power of 0.6 watts). Of course, this is not a concern for the seismometer. The AD690 costs $27 per; I'm not sure I would use it at this point. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: finger quake Date: Wed, 8 Apr 1998 00:03:22 -0500 (CDT) Jim, if it is any help, here is the latest NEIS list: qk_info@................. Updated as of Tue Apr 7 13:23:40 MDT 1998. DATE-(UTC)-TIME LAT LON DEP MAG Q COMMENTS yy/mm/dd hh:mm:ss deg. deg. km 98/04/04 22:52:35 50.97N 171.25W 33.0 4.4Mb B SOUTH OF ALEUTIAN ISLANDS 98/04/05 08:17:26 63.03N 148.69W 76.0 CENTRAL ALASKA 98/04/05 10:17:14 51.82N 174.84E 33.0 4.7Mb B NEAR ISLANDS, ALEUTIAN ISLANDS 98/04/05 15:52:20 43.02N 12.47E 10.0 4.6Mb A CENTRAL ITALY 98/04/05 20:14:54 37.96N 112.54W 5.0 3.4Ml A UTAH 98/04/05 23:52:26 41.07N 124.01W 30.0 3.4Ml B NEAR COAST OF NORTHERN CALIF. 98/04/06 05:28:52 27.94N 143.60E 33.0 4.9Mb C BONIN ISLANDS REGION 98/04/06 05:44:24 60.31N 153.45W 175.1 4.4Mb SOUTHERN ALASKA 98/04/06 13:10:53 24.40S 66.90W 150.7 4.3Mb B SALTA PROVINCE, ARGENTINA 98/04/06 14:25:04 61.82N 151.10W 67.0 3.6Ml SOUTHERN ALASKA 98/04/06 23:55:16 30.97S 71.04W 50.0 4.5Mb B NEAR COAST OF CENTRAL CHILE 98/04/07 00:27:07 34.26N 141.86E 33.0 4.9Mb C OFF E COAST OF HONSHU, JAPAN 98/04/07 13:12:42 14.27N 91.56W 66.9 5.0Mb A GUATEMALA ps: nothing says that this is complete. Regards, Sean-Thomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: boom centering for STM-8 Date: Tue, 07 Apr 1998 23:31:45 -0700 Sean-Thomas -- A couple of thoughts about automating your boom centering: I think your centering motor is functionally equivalent to a very long-period integrator driving the voice coil through a much smaller value resistor than the normal integrator in your VBB design. The small series resistor from this integrator would allow a greater force to be exerted by the voice coil, and be enough to overcome the mechanical drifts of the system. This new integrator could have a long enough time constant so as not to disrupt the normal VBB operation. Perhaps a period of hours. The op-amp used for it would have to be low drift (chopper?) so as not to introduce significant noise in the bandpass of interest. Another possibility is to use a timer that goes for an hour or two. When it expires, the voltage out of the usual integrator stage is used to determine how long to run the centering motor for, and the integrator polarity determines the motor direction. With a bit of trial and error, an on-time duration for the motor could be determined that would re-center the boom for a given integrator output. Ideally, the motor should re-center the boom slowly enough so that it is below the low-frequency cutoff of the VBB system, and not unduly affect the output. Just my $.02. Karl Cunningham La Mesa, CA. PSN Station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: VRDT Schematic numbers ~ Date: Wed, 08 Apr 1998 02:09:48 -0600 Hi Sean-Thomas, I'am not all up with the electronics designations by various companies with their nomenclature for cmos over the years. I wish they would just say 4047, without the CD4047 or TC4047BP etc. Of course they have to identify their products, and this makes for a stumbling block for people like me. So....we only need a 4047, 4018, & a 4066, plus the LM308's or one of its derivatives, i.e., LM108 or LM208 or LM308 OR use a ? # Fet input type op-amp/s on what is shown so far. Talking about obsolete....I am thinking of using some LH0044 op amps in place of the LM308-ha....mainly because I have them, they may work better ....but its hard to get them anymore. They may work for the instrumentation amplifier also. Electronics is like new model cars every year & the accessories make for another designation ......whew! (I am not recommending the LH0044,..I don't have that kind of experience.) Don't know about the AD698. But Philips has an interesting LVDT signal conditioner/s...the NE/SA/SE5521 which sells for about 5-8 bucks, adj Freq of 1-20k, single or dual supply, 182mv power consumption, oscillator/sine converter/demod/amp in one package. LVDT, RVDT, bridge circuits...???VRDT use??? Interesting...came out in ~1987, by a company but absorbed by Phillips later. Not a crystal oscillator. Suggestion courtesy of my brother Robert Lamb. http://www-us.semiconductors.philips.com ...use search for spec sheets etc. To sum it up for digital dummy builders: beware the letters prior to the numbers, but make sure you don't order the microscopic sizes...ha. I've got afew electronics books....nothing with these numbers of course....uh...the circle dot in the corner designates pin 1 right? And if it has a half moon on one end...the pin to the left looking down at the top is pin 1? Somedays just don't go right....ha. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: VRDT Schematic numbers ~ Date: Wed, 08 Apr 1998 02:09:48 -0600 Hi Sean-Thomas, I'am not all up with the electronics designations by various companies with their nomenclature for cmos over the years. I wish they would just say 4047, without the CD4047 or TC4047BP etc. Of course they have to identify their products, and this makes for a stumbling block for people like me. So....we only need a 4047, 4018, & a 4066, plus the LM308's or one of its derivatives, i.e., LM108 or LM208 or LM308 OR use a ? # Fet input type op-amp/s on what is shown so far. Talking about obsolete....I am thinking of using some LH0044 op amps in place of the LM308-ha....mainly because I have them, they may work better ....but its hard to get them anymore. They may work for the instrumentation amplifier also. Electronics is like new model cars every year & the accessories make for another designation ......whew! (I am not recommending the LH0044,..I don't have that kind of experience.) Don't know about the AD698. But Philips has an interesting LVDT signal conditioner/s...the NE/SA/SE5521 which sells for about 5-8 bucks, adj Freq of 1-20k, single or dual supply, 182mv power consumption, oscillator/sine converter/demod/amp in one package. LVDT, RVDT, bridge circuits...???VRDT use??? Interesting...came out in ~1987, by a company but absorbed by Phillips later. Not a crystal oscillator. Suggestion courtesy of my brother Robert Lamb. http://www-us.semiconductors.philips.com ...use search for spec sheets etc. To sum it up for digital dummy builders: beware the letters prior to the numbers, but make sure you don't order the microscopic sizes...ha. I've got afew electronics books....nothing with these numbers of course....uh...the circle dot in the corner designates pin 1 right? And if it has a half moon on one end...the pin to the left looking down at the top is pin 1? Somedays just don't go right....ha. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Jim Hannon Subject: Re: VRDT Schematic numbers ~ Date: Wed, 08 Apr 1998 06:23:31 -0500 Speaking of identifing chip part numbers. Have a look at http://www.twinight.org/chipdir/ They have an extensive reference of chip numbers and manufacturers. -- Jim Hannon http://soli.inav.net/~jmhannon/ 42,11.90N 91,39.26W WB0TXL _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Canie Brooks Subject: Re: POSSIBLE EARTHQAUKE Date: Wed, 08 Apr 1998 09:24:20 -0700 At 12:16 AM 4/8/98 EDT, you wrote: >I am testing out my new verticle seismograph. I recorded, in the Southern >California area, what appears to be a quake occuring about 17:46:11 UTC. I >have checked the finger servers and they do not show anything occuring around >that time. >Did anyone else record anthing around the above referenced time period? >Jim Allen >Cerritos, California There was this small one..: == PRELIMINARY REPORT OF SIGNIFICANT EVENT == Southern California Seismographic Network operated by USGS and Caltech (This is a computer generated solution and has not been reviewed by a human) Event Date and Time : 08-APR-1998 15:17:27.6 gmt (08-APR 07:17:27.6 pst) Preliminary Magnitude: 2.3 MLG Preliminary Location : 33 deg. 47.8 min. -118 deg. 44.7 min. -6.0 km depth Event ID #: 9050015 , 31 phases used, RMS = 0.48 ERH = 1.09 ERZ = 30.52 Region: Southern California 15 mi. SSE of POINT DUME 34 mi. WSW of LOS ANGELES 42 mi. SW of PASADENA 27 mi. SSE of CAMARILLO (quarry) 11 mi. SW of Malibu EQ (19 JAN 1989, 5.0) 15 mi. S of the MALIBU COAST FAULT > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Canie Brooks Subject: Re: POSSIBLE EARTHQAUKE Date: Wed, 08 Apr 1998 09:40:42 -0700 At 09:24 AM 4/8/98 -0700, you wrote: >------------------------------------------------------------------------ > >At 12:16 AM 4/8/98 EDT, you wrote: >>I am testing out my new verticle seismograph. I recorded, in the Southern >>California area, what appears to be a quake occuring about 17:46:11 UTC. I >>have checked the finger servers and they do not show anything occuring around >>that time. >>Did anyone else record anthing around the above referenced time period? >>Jim Allen >>Cerritos, California Sorry - I mis-read that!! - I thought it was this morning.... Canie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: VBB block diagram Date: Wed, 8 Apr 1998 20:46:38 -0500 (CDT) Jim, The block diagram is intended to illustrate the feedback concepts and was never intended to be a schematic, which is why it is incomplete. I don't actually use it as drawn; I'm recycling generic amp cards. At this point I have assumed that anyone creating the circuit would have enough electronics background to fill in the details. But.. A1 has a gain of 10 but is also a low pass filter. I believe the resistor from the inverting input to ground should be 23.2k for a gain of X20 (with 464k for the feedback R) since the input filter divides by 2. The resistor loading the output is 10k. The feedback resistor for the integrator input offset balance is 2meg. (I actually use a 3amp instrumentation configuration for the integrator). There is no problem with the integrator with temperature. any more than with any other low pass filter. It operates at unity gain to avoid any saturation that would limit the dynamic range. (it pegs out only when the displacement amp is also saturated, but, of course, for the time constant of the integrator). The integral feedback resistor of 107k is determined by the transfer function and shapes the broadband response. Its behavior is not a matter of supplying an independent current, but to shape the response by the ratio of the integral current to the proportional current through the 581k resistor and the differential current through the 24.1 uf capacitors. The feedback elements cannot be considered independently of each other. At least in this VBB configuration. regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Magnetometer going wild! Date: Wed, 8 Apr 1998 19:14:16 -0700 (PDT) Hi All: For most of the day I was fascinated with watching wild fluctuations in geomagnetic field strengths on my magnetometer! Does anyone have a clue to what's possibly causing this? For the last few weeks the readings were sort of erratic... But, today takes the prize! Frank Condon mailto: frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Charlie Plyler Subject: Re: Magnetometer going wild! Date: Wed, 08 Apr 1998 22:55:23 -0400 Frank, GEOS also has had some adnormal readings today. Take a look at: http://sec.noaa.gov/today.html ELFRAD GROUP N 35 43.7336 W 80 48.4224 cplyler@............. Frank Condon wrote: > > Hi All: > For most of the day I was fascinated with watching wild fluctuations in > geomagnetic field strengths on my magnetometer! Does anyone have a clue to -- _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Re: Magnetometer going wild! Date: Wed, 08 Apr 1998 22:35:04 -0700 Frank Condon wrote: > > Hi All: > For most of the day I was fascinated with watching wild fluctuations in > geomagnetic field strengths on my magnetometer! Does anyone have a clue to > what's possibly causing this? For the last few weeks the readings were sort > of erratic... But, today takes the prize! > You're not the only one recording some fluctuations: http://dxlc.com/solar/index.html -- ---/---- Charles P. Watson Seismo-Watch _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: Magnetometer going wild! Date: Wed, 08 Apr 1998 23:07:56 -0800 Wait till Mother's Day, full moon et al (may 10th for you orphans) Charles Watson wrote: > Frank Condon wrote: > > > > Hi All: > > For most of the day I was fascinated with watching wild fluctuations in > > geomagnetic field strengths on my magnetometer! Does anyone have a clue to > > what's possibly causing this? For the last few weeks the readings were sort > > of erratic... But, today takes the prize! > > > > You're not the only one recording some fluctuations: > http://dxlc.com/solar/index.html > > -- > ---/---- > Charles P. Watson > Seismo-Watch > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L -- Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Re: new help files Date: Thu, 09 Apr 1998 08:59:19 -0700 Larry Cochrane wrote: > > Thanks John, > > I placed the documents I have so far on my PSN system. The following link > can be used as a starting point: > > http://psn.quake.net/wqdocs/index.htm > > I now have 12 people offering to help, so I have plenty of help. > > -Larry > > At 09:00 AM 4/6/98 -0600, you wrote: > >Larry, > > > >How about just putting the new help files on your web site with Larry I quickly read over the current HTML and the four help files. I see that you selected a style such as File menu and File List. Notice that menu is lower case. Another way to refer within a sentence is to italic the words and cap always havew two caps.. EM File Menu /EM or EM Save File /EM. re read the file. re as a prefex... the nuns will get you for this one-- How about reload the data. a SEED or PEPP data set volume... What is a SEED and PEPP data set and why do I need to know this? Looks good-- I'll keep reading. Regards, Steve > >a heading > > 'Preliminary WinQuake help file under review' > > 'Please forward suggestions or corrections to ...' > >When the next version comes out, you can just change the > >title and leave them there. > > > >Just a thought. > > > >JCLahr > > > >_____________________________________________________________________ > > > >Public Seismic Network Mailing List (PSN-L) > > > >To leave this list email listserver@............. with the body of the > >message: leave PSN-L > > > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Velocity Period Enhancing circuits Date: Thu, 09 Apr 1998 21:22:04 -0600 Albert Noble, Picked up the article you mentioned afew days back; "A Vertical Equalization Circuit For Increasing Seismometer Velocity Response Below The Natural Frequency" by Peter M. Roberts in Vol. 79, No. 4, pp. 1607-1617, Bulletin of the Seismolological Society of America, in the Central Denver Public Library. I by no means followed all of the article, but it looks like it originally would be of interest to those with the short period seismos, to expand the response out to the intermediate period results. Electronically the original circuit uses LM312 op amps and its possible to use many other op amps of course, as trial replacements. Its nice to see the total circuit (which you have to 2 X) and it seems extremely simple, even for me-ha. Regardless, if you do try it out, or change it, by all means let us know the results. In many ways, however, I'am more drawn to the posted PSN article and circuit, "A Period Lengthening Device, Long Period Velocity Type Seismometers", by Allan Coleman, Nov 1995 that is on the PSN computer. I've managed to go through all of the past email messages over time, but....I don't remember seeing any mention of anyone trying it....nor is there any mention in the member/station lists. A strong part of my own interest is that I do have some 1500 ohm coils that could work-ha. Of course the Coleman circuit is a feedback circuit, and indicates the possibility of a much longer period, than the Roberts circuit. It would be nice to hear from anyone on the email list, who has tried the circuit? And, of course, their opinion? Failing that, I would presume no one has tried it.... people usually want strong confirmation before going ahead on their own. Technically, I couldn't begin to address any faults in either but the appealing idea of checking them out, is there anyway.... Raise the anchor and hoist the sails Mate!....ha. Thanks for the reference, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: ASCII data Date: Thu, 09 Apr 1998 23:54:11 -0700 Hi Everyone, I got this email today. Does anyone have a utility that Galileo can use to save a PSN event file in ASCII format? I get this question now and than and I plan to add this option to WinQuake in the next release, but that doesn't help Galileo now. Adding this feature to WinQuake is pretty easy, but I'm not sure what to do with the header information like file start time, samples per second etc. I can save the data set in ASCII form, but I would think that programs like MatLab and SAP90 (not sure what that is) would also need information like the sample rate and maybe the start time? Is there any "standards" for dealing with this? If not I was thinking of placing the start time and SPS on the first line and then the data set, one sample per line. The other thing that could be done is save the header info in a separate file. Any ideas??? -Larry Cochrane Redwood City, PSN >>>> Reply-To: "Galileo Tamasi" From: "Galileo Tamasi" To: Subject: Ascii data Date: Thu, 9 Apr 1998 17:37:05 +0200 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 ArialI have winquake program. How i can extract data ascii for dinamic analisys with SAP90 ? Thanks. Galileo Tamasi <<<<<<<< _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: More Earthquake Predictions Date: Fri, 10 Apr 1998 00:44:06 -0700 (PDT) Hi All: Here is another Web-Site I came across that is about predicting earthquakes. James Berkland Earthquake Prediction Web-Site http://www.syzygyjob.com/ Frank Condon frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Ted Blank Subject: Re: ASCII data Date: Fri, 10 Apr 1998 06:34:39 -0700 (PDT) One way that is fairly easy is to just use some BASIC code. There is sample BASIC code for loading a PSN file into memory in the file description on the PSN bulletin board. It would look something like this: DIM A(25100) DEF SEG(A(0)) BLOAD "C:\970101AN.XXX",varptr(a(0)) FOR I=100 to 25100 PRINT A(I) NEXT I You could precede the data (which starts at offset 100) by other print statements to print date, start time, etc. in whatever format was required by this program (SAP90). Make sure array A is DIM'ed AS INTEGER. Ted > > Hi Everyone, > > > I got this email today. Does anyone have a utility that Galileo can use > to save a PSN event file in ASCII format? > > > I get this question now and than and I plan to add this option to > WinQuake in the next release, but that doesn't help Galileo now. > > > Adding this feature to WinQuake is pretty easy, but I'm not sure what to > do with the header information like file start time, samples per second > etc. > > I can save the data set in ASCII form, but I would think that programs > like MatLab and SAP90 (not sure what that is) would also need information > like the sample rate and maybe the start time? Is there any "standards" > for dealing with this? If not I was thinking of placing the start time > and SPS on the first line and then the data set, one sample per line. The > other thing that could be done is save the header info in a separate > file. > > > Any ideas??? > > > -Larry Cochrane > > Redwood City, PSN > > > >>>> > > Reply-To: "Galileo Tamasi" > > From: "Galileo Tamasi" > > To: > > Subject: Ascii data > > Date: Thu, 9 Apr 1998 17:37:05 +0200 > > X-MSMail-Priority: Normal > > X-Mailer: Microsoft Outlook Express 4.71.1712.3 > > X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 > > > ArialI have winquake program. How i > can extract data ascii for dinamic analisys with SAP90 ? > > Thanks. > > Galileo Tamasi > > > > <<<<<<<< > > > > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Sam Toon" Subject: Re: ASCII data Date: Fri, 10 Apr 1998 16:29:27 +0100 I have just written a program in Matlab to read in PSN format files. It can also read in the header information if you are using version 5 of Matlab. If anyone wants a copy just send me an email, or it could possibly be put on Larry's website. I haven't thoroughly tested the program! I will also try and write a program to write the data back out so that you would be able to use Matlab to do some processing on the data and save it as a PSN file. Sam ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Sam Toon serr15@......... Microseismology Research Group http://www.pcweb.liv.ac.uk/Microseis/ Dept of Earth Sciences Tel: +44 (0)151 794 5157 University of Liverpool Fax: +44 (0)151 794 5157 P.O. Box 147, Liverpool, L69 3BX United Kingdom ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -----Original Message----- From: Larry Cochrane To: psn-l@............. Cc: g_tamasi@........... Date: 10 April 1998 07:55 Subject: ASCII data Hi Everyone, I got this email today. Does anyone have a utility that Galileo can use to save a PSN event file in ASCII format? I get this question now and than and I plan to add this option to WinQuake in the next release, but that doesn't help Galileo now. Adding this feature to WinQuake is pretty easy, but I'm not sure what to do with the header information like file start time, samples per second etc. I can save the data set in ASCII form, but I would think that programs like MatLab and SAP90 (not sure what that is) would also need information like the sample rate and maybe the start time? Is there any "standards" for dealing with this? If not I was thinking of placing the start time and SPS on the first line and then the data set, one sample per line. The other thing that could be done is save the header info in a separate file. Any ideas??? -Larry Cochrane Redwood City, PSN >>>> Reply-To: "Galileo Tamasi" From: "Galileo Tamasi" To: Subject: Ascii data Date: Thu, 9 Apr 1998 17:37:05 +0200 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 I have winquake program. How i can extract data ascii for dinamic analisys with SAP90 ? Thanks. Galileo Tamasi <<<< _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) To leave this list email listserver@............. with the body of the message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: ASCII data Date: Fri, 10 Apr 1998 08:35:18 -0700 Larry -- I'd go with just including a header that contains the information in question, and put it at the start of the ASCII file. It's easy enough to remove it later with a text editor if it turns out it's not needed. Karl Cunningham La Mesa, CA. PSN station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: John Hernlund Subject: Re: ASCII data Date: Fri, 10 Apr 1998 11:18:11 -0700 (MST) On Thu, 9 Apr 1998, Larry Cochrane wrote: > Adding this feature to WinQuake is pretty easy, but I'm not sure what to > do with the header information like file start time, samples per second > etc. > I can save the data set in ASCII form, but I would think that programs > like MatLab and SAP90 (not sure what that is) would also need information > like the sample rate and maybe the start time? Is there any "standards" > for dealing with this? If not I was thinking of placing the start time > and SPS on the first line and then the data set, one sample per line. The > other thing that could be done is save the header info in a separate > file. Larry, I think getting an ASCII format going is a great idea, especially for transfering files over the net and e-mail! How about making the first line the start date, the second the start time, the third the SPS and then one line for each data reading... The advantage that I can see in doing this is that a person could easily go into the file and edit and/or remove those peices which could then be transferred to many different kinds of software that handle data in that manner... ****************************************************************************** John Hernlund E-mail: hernlund@....... WWW: http://www.mc.maricopa.edu/academic/phy_sci/Geology/hernlund/ http://www.public.asu.edu/~hernlund/ ****************************************************************************** _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: A Magnetic Spring Seismometer Description Date: Fri, 10 Apr 1998 16:48:09 -0500 Friends, A while back I posted a bunch of stuff on magnetic suspension seismometers of my own design. Actually I experimented with a whole variety of good and bad designs. This is a sort of progress report = of making a few stabs at the problem now and then--presented here in the spirit of revealing some of its more promising aspects in case I go on to something else entirely. Generally my best designs have been small and used the repulsion between two magnets as a spring force to balance the force of gravity rather than the metal spring more commonly used in verticals. Pretty quickly I discovered that an oiled steel razor blade edge attracted to another magn= et mounted behind a thin brass strip makes a great pivot mechanism for such = a design, while being rigid enough to only allow only one accurately define= d axis of movement (in other words little off-axis wobble or vibration). =20 Furthermore the Q of such a system is very high, meaning in effect that i= t is very efficient at capturing and holding mechanical energy with low losses. In other words when slightly disturbed, such a system can take three or four minutes to stop oscillating at several oscillations per second. Such high Q systems, even with short periods, are an excellent match for force feedback, because the low losses mean that it can efficiently integrate slow additions of mechanical energy of energy over long periods of time; much longer than its own natural frequency. This is what is needed to sense teleseismic events with physically small instrume= nts. Such a configuration may be built very compactly and, with suitable adjustments, can be used at any angle from the vertical. For example if such instruments are small and easy to build, three of them could be aligned along the edges of a triangular pyramid, and the traces used to calculate quake shocks as three-dimensional vectors of accelerative force. With computing power so relatively cheap, why not? After initial trials with a laser pointer, I decided that a bright red LE= D and phototransistor interrupted by a flag mounted on the beam is small, sensitive, and cheap enough to do the job nicely (at least down to the ambient higher frequency seismic noise levels where I live in the central city near lots of traffic). No trouble seeing lots of these. The obvious objection to such a design is that it is a little bit magnetometer, a little bit gravimeter, and a little bit seismometer. It falls between three stools, so to speak. There can be little objection to an instrument which is both gravimeter and seismometer, so long as it can be held within range of the daily earth tides. A seismometer usually look= s for the smallest possible accelerations, and gravity is physically the sa= me as acceleration, so gravimeters are merely DC versions of seismometers.=20 The only real problem with a gravimeter being used as a seismometer is that seismometers are designed to be so sensitive to acceleration that sl= ow DC drift and earth tides are likely to throw the readings off scale. A Lehman is a horizontally balanced instrument, designed to be nearly blin= d to gravity, and thus does not suffer from these complications.=20 Anyhow, while the same instrument can be both a first class gravimeter an= d a vertical seismometer at the same time if it can be kept within scale, magnetic fields are another thing entirely.=20 They are just plain bad, insofar as weak external magnetic fluctuations come in all frequencies and interact with the rather strong magnetic fiel= d used to balance the force of gravity inside the instrument. Since we are trying to balance and measure these opposing gravitational and magnetic forces to low parts per million if possible, natural magnetic storms can interfere with our data. One design that I have proposed to use strong ra= re earth magnets to balance gravity on the theory that a smaller stronger mo= re concentrated magnetic balancing force would be relatively blind to weak background fluctuations in the earth field.=20 The bottom line here is that assuming temperature can be controlled fairl= y well, relatively small cheap and sensitive instrument can be designed wit= h magnetic springs. I now think magnetic shielding is the best answer to th= is instrument design problem. Doug Crice posted a response some time ago in which he said that several concentric shields of permalloy and iron would probably do the job. I have finally got around to reading a good paper on the subject, and it looks as if the best way to shield an instrument is probably to wind a th= in tube out of a thin sheet (even a foil if it it is good stuff) of some ver= y high permeability ferromagnetic metal such as permalloy or mu-metal. If t= he ends of the tube extend beyond the instrument and the several layers of magnetic alloy are separated with any non-magnetic material such as cardboard before being wound into a tube, then an instrument placed insid= e the tube can be shielded by a factor of thousands. Iron or steel is chea= p but rather poor for shielding purposes, except to reduce fields of thousands of Oersteds down to the point that permalloy can take over. (Se= e W.G. Wadey; Review of Scientific Instruments, Nov. ,1956. p 910).=20 From what I know, one of the best mountings for such a compact vertical instrument would probably be to lower it into the bottom of a vertical ir= on pipe plugged with concrete at the bottom, buried perhaps six feet in the ground. This would tend to be a good seismic platform, while largely buffering temperature changes, (which are usually the main arch-enemy of instrumental accuracy). To reach some sort of closure to this long post, my latest design uses a stack of three ceramic ring magnets mounted on a vertical screw so they m= ay be raised or lowered by turning the screw with their magnetic axis remaining accurately vertical.=20 This force opposes and supports the short horizontal beam. The latter is = a fourth magnet glued to a utility knife blade with a two ounce lead fishin= g weight on top. This horizontal knife edge is attracted to a brass strip backed on the opposite side with two rare earth magnets. This forms a kni= fe edge pivot surface to one side. In other words the little beam is very stout and heavy and rocks up and down on a horizontal axis less than an inch to one side.=20 On the side of the beam away from the brass pivot rest, I looped a strip = of aluminum cut from a drink can so it extends the beam sideways several inches. This extension amplifies the motion of the center of gravity of t= he beam by a factor of two or three. This hold the little aluminum flag that interrupts a beam between the LED and the phototransistor. The latter hav= e most of their integral plastic lens cut away so the emitting and sensing elements can be mounted very close together surrounding the flag (which moves vertically with the beam) for best movement detection sensitivity.=20 Finally, there is a second stack of two ceramic magnets, mounted on anoth= er vertical brass screw, just over the beam. This latter stack is oriented s= o as to try to pull the beam upwards.=20 The end result is that we have a little beam that resembles a magnetic pu= ck rocking up and down on an axis near its side. Below it we have another magnetic puck that can be raised or lowered and which repels it. To the top we have yet a third puck that can also be raised or lowered to attra= ct rather than repel the beam. All these discs are coaxial. By adjusting the repelling and attractive forces, we can push the beam from below while pulling it from above. Some combination of these forces will be found to very nearly balance gravity -- in such a way that the net force which tends to restore the beam position is very weak.=20 As with a Lehman, this delicate adjustment has the effect of lengthening the natural period of the beam and increasing the sensitivity. Even if th= e period is still a second or a half second, which may not seem very good, the beam will move quite a lot in a response to a very small amount of force, which is what we are after (and which is also what the goodness of= a zero length spring is all about). At this point we have a very high Q vertically sensitive spring and mass and sensor combination. The last step is to add force feedback. This is done by winding a wafer of several hundred turns of thin magnet wire wet with epoxy so that it is about the size and shape of one of the ring magnets (about 1/4 inch thick and an inch across). This coil is mounted o= n a thin glass horizontal platform just below the magnet on the beam. Thus the coil is surrounded by the repelling magnet to its bottom and the beam magnet to its top and coaxial with both. In this way, the coil has maximu= m interaction with the magnetic force that supports the beam.=20 It appears that the output from just one little 324 op amp can supply the 20 milliamps at six volts and this is sufficient to stabilize the beam position, even though the resistance of the coil is fairly low at about 8 ohms. Obviously, however, if the forces that need to be balanced are significant, a few hundredths of a milliwatt of power is not going to be able to push around a 2.5 ounce lead-weighted beam in order to faithfully follow and balance the motions from strong or rapid accelerations. But there is no reason why this should disqualify such a setup from following teleseismic events, which are characteristically slow motions with little power involved. I assume that if one wants to extend the flatness of response to strong shocks or higher frequencies, it is best to raise the voltage (currently= 6 volts from four large dry cells) to twelve volts and put transistor power boosters on the zero potential terminal of the force feedback coil and al= so on the variable voltage terminal, but this should be quite easy. Currently, I use a very simple arrangement of a 10 MFD non polarized tantalum and a 1000 ohm pot in parallel to power the coil. These are between the coil and the op amp output. This is not a sophisticated force feedback arrangement at all, and does not reflect stability at high power levels, but it does tame pesky feedback oscillations when small forces ar= e involved.=20 Obviously I have concentrated mostly on the mechanical configuration in this discussion. I=92m a bit over my head in regard to the finer points o= f optimizing feedback circuitry, but a little trial and error along with simple resistor and capacitor and op amp combinations is one way to do th= e job in a workable if non-optimal way. The STM-8 circuit configuration wou= ld undoubtedly work too. Otherwise read Horowitz and Hill=92s =91Art of Electronics=92 or take a course in feedback theory. A few final points. The external dimensions of the sensor in its present prototypical form exclusive of its mounting (chiefly made from scraps of wood and glass and brass ) is about 3x4x6 inches, with its longer dimensi= on being vertical. Thus it is small and could probably be designed to be be lowered in a 4 inch ID pipe. I think it should be possible to wind a reasonably suitable cylindrical o= r rectangular magnetic shielding tube from a strip of permalloy or mu-metal foil if I knew where to get some. Anyone know?=20 My current prototype is mounted on a pine board with the batteries and circuit inside a plastic insulated beverage cooler along with a heavy chu= nk of iron for thermal ballast. In other words so that any temperature chang= e will be very gradual. Nevertheless, the instrument is a gravimeter and th= us very subject to slow DC drift which pushes it off scale. The rough scale recentering adjustment is done by turning the upper screw to raise or lower the upper attracting magnet, while smaller scale changes are made b= y pushing a stack of ceramic magnets around outside the cooler, as I watch the emerging chart on my computer monitor (I use a two channel WinDaq A/= D for charting). The biggest problem that I anticipate with the current set= up is to keep the trace centered. I think if you reduce the sensitivity enou= gh to have the daily earth tides move the instrument full scale (which also implies good temperature control, I think), then you have automatically reduced the sensitivity too much for it to be a very sensitive seismomete= r. So you need some way to recenter the scale, even if the instrument is perfect, unless the A/D output has very high resolution. I don=92t have any sort of formal site, or any formal graphic presentatio= n prepared other than this written progress report, but I can send a digita= l JPEG photo of my instrument in its modest current form (I think Dave Nels= on offered to host a picture on his web site a while back). I certainly don= =92t advocate that anyone should try to duplicate my design exactly in every detail. For me instrument design is always a work in progress, and I neve= r do seem to get to the polished brass knob stage that some find so appeali= ng. What I am most interested in doing is doing is communicating the basic principles of my rather unorthodox instrument, centered around being fair= ly easy and cheap to build by amateurs as a fundamental principle. All comments are welcome, with the caveat that the current configuration is suggestive but not a polished design.=20 It is completely uncalibrated, and I know that it is sensitive chiefly by walking past it on my concrete slab floor while watching the chart, stamping the floor from the other side of the room, and by studying the static influence of small magnets moved about slowly at a distance of a f= ew feet.=20 --Yours, Roger Baker =20 _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: magnetic suspension Date: Fri, 10 Apr 1998 19:34:54 -0500 (CDT) Roger, Thanks for the description of your magnetic suspension seis. An instrument of this type was one of the first that I worked with back in 1969, called the Sprengnether S-7000. It is still manufactured today because of its robust qualities and total lack of spring resonances. It is rather expensive, though, and is about the size of a shoe box. The S-7000 uses a concentric counter-rotating magnet assembly that has a special alloy shunt that compensates the temperature change of the magnet's strength. Since the rotor magnet is completely inside the fixed or stator magnet, it is relatively immune to external fields, at least as a velocity sensor. The standard output of the S-7000 is from a velocity coil, with a period of 1.5 seconds. I have configured an old one for VBB feedback to operate at 150 seconds, which really pushes the magnetic and thermal stability envelope; it is acceptably stable at 30 seconds though, and uses the VRDT with picometer resolution as the displacement sensor, and the velocity coil/magnet becomes the feedback coil. I looked in some of my catalogues for info on magnetic shielding sheeting (permaloy, mu-metal) and currentlu draw a blank. I will make some calls on Monday. If your sensor detects small magnets from a few feet away, it certainly HAS a serious susceptivity problem. Maybe you could consider a design with the RE magnets working from WITHIN a large pot, ring, or U amgnet, which would be an environment much stronger than external fields. Re recording the displacement detector output: I use a 1000 second high-pass output (1000 uf-NP into 1 megohm) to remove the DC components (tides, thermal drift, atmospherics, etc) of the VBB output, which can vary by several hundred millivolts over a day; the "seismic" component runs about 1 to 10 millivolts for the 6-second microseisms, and I have to be good to my cheap 12-bit digitizer. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: equalizer circuits Date: Fri, 10 Apr 1998 20:34:30 -0500 (CDT) Meredith, We have made several evaluations of the P. M. Roberts circuit, especially trying to make a 4.5 hz phone act like an L4-C at 1 hz. It does work, but it is noisy, even with the best selection of amplifiers. (there was a recent BSSA article on the noise limits of seismometers and amplifiers, but I can't find the reference right now; it had an important table that matched the performance of selected amplifiers to the coil resistance of the seismometers). The problem with the equalizer circuit is that the gain of the 1-pole filter is about 20, so the two of them in series (to get the proper slope) have a gain of 400. This means that all the 1/frequency noise of the amplifiers gets amplified and summed into the output. VERY careful selection of the preamp can help, and narrow-band coupling of the stages also helps, but limits the response. I still cannot get very good spectral coherence between an L4 and an equalized 4.5hz phone when operating side-by side on a shake table. And the noise in the output runs at several millivolts in the 0.8 to 5 second periods, and is ten times that if you breathe on the amplifiers (we use LT1024s). So the equalizer is OK for low sensitivity applications, but don't connect it to a 24-bit digitizer! When I find the reference to the other article (probably in THE HEAP at home), I will post it. Maybe someone else has it handy. regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: not FET amps Date: Fri, 10 Apr 1998 20:44:14 -0500 (CDT) Brian, Thanks for the correction about the LM308 NOT being a FET input like the (correct me if necessary) LF356. I tend to get some of the better qualities of devices confused, and as you observe, National about breaks their elbo patting their back over how DC stable the 308 is; it doesn't even have an offset adjustment. It actually has many FET qualities but without the thermal, etc, problems that you mention. I will check out the LM112 a bit more, since the LM308 isn't always available from all suppliers. Maybe it is a bit more ESD resistant than the 308. I've had problems with ESD wiping out all 20 or so 308s in a single tiltmeter system. I appreciate your interest, and will correct the notation on the next round of drawing upgrades. Hopefully the PSN readers of the documentaion will see this. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Surplus store Mu-metal foil Date: Fri, 10 Apr 1998 20:34:24 -0600 Roger Baker, A possible source for mu-metal might be found in electronic surplus stores. These are canned donut shaped which were originally used for donut choke etc. coils...i.e., ferrite predominates now. In Colorado I've seen them in 3 out of 4 surplus stores. They seem to be mostly orange, but sometimes black and mixed in with the ferrite core stuff. The tip off is that on one side, on the rim it has the appearance of being like a grocery can lid edge. Some are made different, and can be pried open, or pulled apart. Inside is mu-metal rolls of foil, usually not too wide...varies ~ 1/4", but they can usually be rolled off readily to whatever surface...and glued...whatever. The prices really run a wide range, from cheap to fairly expensive...50 cents each on up to $3 each, depending on store. These maybe leftovers From the 60's or so. Diameters are ~ 1.25 to 1.50" usually, but some are bigger. I don't know how thick the foil is...I've never got one. For your deal, it may take several, and hopefully you can get them cheap. They are getting scarcer all the time, and never seem to be in much quanity at all. If it doen't look like ferrite, check it out. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Readers Digest Magnet Healing Article Date: Fri, 10 Apr 1998 20:37:02 -0600 Offbeat but interesting article on using magnets for pain reduction. April 1998, pg. 170, Magnets For Pain? Basically 2 institutions used magnets & placebos over pain areas on 50 patients, and the magnet postpolio patients reported a much higher ratio of relief than the placebo patients. No explanations made. Only a short 1/3 pg or so spot, in the news from the world of medicine. I don't think it would be in the Readers Digest if it wasn't a valid report. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: The care & feeding of a Sprengnether vertical Date: Fri, 10 Apr 1998 22:32:34 -0600 Sean-Thomas, Getting around to setting up my 40 odd year old, and abused vertical. As of today I dropped the mass and promptly broke and mangled all spring wires and pivot hinges. Never fear, later in the day, it was up and hanging again. As it turns out, scatch sending the vertical manual, as I do have it in my heap at home (ha). 30 years affect my sloshing memory. The manual didn't really apply much as everything has been moved or out of the factory setting anyway. Late today, it was bouncing around at 17 second period on the mother pier, before I quit for the day, with more stuff, to be done tomorrow. Am finding out that verticals are another world apart from the horizontals, and this "baby" indicates it will need significantly more maintenance care over a period of time. I can see the relativity with your STM-8 design more clearly. Have questions if the good doctor cares to inform the parent: The original pivot post has a groove slot for presumably #36 music wire or .025" diameter. When I got it I found that the previous owner had tried using a bunch of smaller gauge wires close together, and steel flat hinges 3/8" wide and .005" thick at some time or the other. Am using the flat spring steel hinges presently. It seems to oscillate forever...~ 20 minutes, before I can't see any more movement. Anyway....being as the original was wire, would you recommend going back to the wire or leaving it as is? I can understand horizontals, but this "childs", period adjustment kind of throws me. Level the base, and it has a 6 second period, then raise the front screw, and the period increases, but with mass position adjustments. The mechanism seems somewhat like the horizontal, but I'am sure not really grasping the basis with the vertical period scheme? The spring itself does exhibit the ringing visually and sometimes in the audio range when it hits the stops. You've mentioned this before. I installed acouple wood blocks under the boom back hinge for initial setup and as a form of stop for crashes again. Their height is less than the level of the hinges, and doesn't interfere with operation. This model has photocells and a boom motor mass position device, which will eventually be used...if they work...and I inprovise. Might you have any tips to pass along, other than my questions? Would appreciate it. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Al Allworth" Subject: Re: Surplus store Mu-metal foil Date: Sat, 11 Apr 1998 00:16:20 -0700 -----Original Message----- From: meredith lamb To: psn-l mailing list Date: Friday, April 10, 1998 7:41 PM Subject: Surplus store Mu-metal foil >Roger Baker, > >A possible source for mu-metal might be found in electronic >surplus stores. These are canned donut shaped which were I think what you are describing are tape wound torroid cores. The tape is not Mumetal. There are many different alloys used for these cores depending on the intended use. Some of the material is what is known as "square loop" material. It has a memory characteristic and stays magnetized in the direction it was last driven. This would not be suitable for shielding purposes. Other materials are more linear and can be used for making audio or power transformers. Another type has yet another characteristic and is supplied in matched quads for use in magnetic amplifiers. There are many variations of these three depending on the frequencys and power levels they were designed for. I think the best unmistakable source of Mumetal is found in old cathode ray tube shields. The electron beam in a CRT is easily deflected by earth field and leakage flux from transformers and power lines and must be shielded for an oscilloscope to work properly. Some of the newer shields use a newer, different alloy which is 2 or 3 times better than Mumetal. The best shields are made with 2 layers of Mumetal with layer of copper sandwitched between. They are seldom needed and are quite heavy. By the way, the shielding effect of Mumetal is greatly reduced by drilling or cutting it. If you need to cut it after it is heat treated use very sharp shears and minimize the shock imparted to the piece. I don't know the correct proceedure for annealing Mumetal but I know it is done in a hydrogen atmosphere and that the temperature is critical. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bschafer@......... Subject: frankcnd@.......... Date: Sat, 11 Apr 1998 02:52:49 -0500 (CDT) More Earthquake Predictions Hi KFrank, Thanks so much for this site. Might be very interesting. Does anyone know what Berkland's success rate is? Ran across another book, today, and it looks very interesting. I have only read a couple of pages. It is called "The Trembling Earth." It is written by Frederic Golden and is copyright 1983. Bonnie the crafty crafter _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Re: frankcnd@.......... Date: Sat, 11 Apr 1998 01:07:41 -0700 (PDT) Hi Bonnie: Another book that I'm trying to find a copy of is "When the snakes awake." about animal activity as a pre-cursor to earthquakes. Frank Condon frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" >More Earthquake Predictions > >Hi KFrank, > >Thanks so much for this site. Might be very interesting. Does anyone >know what Berkland's success rate is? > >Ran across another book, today, and it looks very interesting. I have >only read a couple of pages. It is called "The Trembling Earth." It is >written by Frederic Golden and is copyright 1983. > > >Bonnie >the crafty crafter > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Norman Davis WB6SHI Subject: Re: Readers Digest Magnet Healing Article Date: Fri, 10 Apr 1998 22:09:36 At 08:37 PM 4/10/98 -0600, you wrote: >Offbeat but interesting article on using magnets for pain reduction. >April 1998, pg. 170, Magnets For Pain? Basically 2 institutions >used magnets & placebos over pain areas on 50 patients, and >the magnet postpolio patients reported a much higher ratio of >relief than the placebo patients. No explanations made. Only a >short 1/3 pg or so spot, in the news from the world of medicine. >I don't think it would be in the Readers Digest if it wasn't >a valid report. Tried one for my back and didn't notice any significant difference. One guy at work says it works great for him. I just don't know? _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Re: Readers Digest Magnet Healing Article Date: Sat, 11 Apr 1998 08:32:57 -0700 (PDT) Hi Meredith: Wouldn't that be known as "bio-magnetism." Also, I've seen some advertisments of matresses with embedded magnets. Very interesting subject matter. Somewhere in my collection of periodicals I recall seeing a 20 year old journal about alternative healing with an article about the use of magnets. I'll see if I can find it and post what it tells. Frank Condon frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" >Offbeat but interesting article on using magnets for pain reduction. >April 1998, pg. 170, Magnets For Pain? Basically 2 institutions >used magnets & placebos over pain areas on 50 patients, and >the magnet postpolio patients reported a much higher ratio of >relief than the placebo patients. No explanations made. Only a >short 1/3 pg or so spot, in the news from the world of medicine. >I don't think it would be in the Readers Digest if it wasn't >a valid report. > >Meredith Lamb > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: frankcnd@.......... Date: Sat, 11 Apr 1998 08:57:16 -0800 Looking for predictors or the gifted? 98% close enuf? http://www.viser.net/~charking/story.html When studying this keep in mind today's "science" is tommorrows witchcraft. The strides in quantum physics is closing in on the bicosmic energies. The gift is something that the scientific method promoted by the 'analyst' (notice the ANAL) lends itself to never having a final answer, just the most convincing argument gathering the largest world of agreement. I've experienced this on both ends of study. When working in the black arts of EMF/RFI shielding, I got 'results' using what's known as the common intelligence, tapping in with stupid things like intention, but producing a shielding that passed all FCC testing for sure. This bothered all my superiors, with their MS degrees, and they spent six months attempting to figure out what I did, instead of enjoying the product that would beat everyone to the market. While intellectually J Offing this, a competitor got their product to the market and blew my co. off the marketing scene. So when you get some project subsystem working, keep going and don't go for the BOOBY PRIZE - Reasons...or results. I've found results more satifying. For Roger's project try reflective mylar (AKA space blanket material) and use the old 555 chips, they can work miracles...Keep it SIMPLE, genius is simple, as the Truth, out their is... bschafer@......... wrote: > More Earthquake Predictions > > Hi KFrank, > > Thanks so much for this site. Might be very interesting. Does anyone > know what Berkland's success rate is? > > Ran across another book, today, and it looks very interesting. I have > only read a couple of pages. It is called "The Trembling Earth." It is > written by Frederic Golden and is copyright 1983. > > Bonnie > the crafty crafter > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L -- Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: ASCII data Date: Sat, 11 Apr 1998 08:05:35 -0700 Hi Larry I believe SAP90 is a dynamic building structural analysis program. Barry Larry Cochrane wrote: > > I can save the data set in ASCII form, but I would think that programs > like MatLab and SAP90 (not sure what that is) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: equalizer circuits Date: Sat, 11 Apr 1998 08:22:37 -0700 Sean Thomas There were a couple of articles in BSSA vol. 82 #2 ,pp 1071-1123 by P.W. Rodgers that talked about it ."Frequency Limits for seismometers as determined from signal to noise ratios". Is this what you were thinking of? Barry > When I find the reference to the other article (probably in > THE HEAP at home), I will post it. Maybe someone else has > it handy. > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: S5000 LP vertical Date: Sat, 11 Apr 1998 12:32:41 -0500 (CDT) Meredith, The S5000 LP vertical IS an entirely different sensor from the S5000 series horizontals; the only thing they have in common is the coil/magnet assembly. I will push your request for materials to the top of the list next week. I will also send you some new hinge material and new spring suspension wire. Getting the seis to perform as advertised is difficult enough with the original design, and probably nearly impossible with substitutions. In particular, if the hinge crossover point moves at all wrt the upper end of the spring suspension clamp, the period will change radically as the boom position changes. About the best period linearity that can be obtained is 3% over a range of + - 2mm at the boom pointer. This means that if the period is 30 seconds at center, it can be 31 or 29 over the operating range. The adjustment of the upper spring attachment sliding assembly for linearity can take days, and is all for naught if some event stresses the hinges after it is done. When these S5000 were used in the WWNSS, they originally tried for a standard of 30 seconds (and a 90 second galvanometer), but eventually backed off to a 15 second standard because of the instability and linearity problems. I have just "tuned up" a set of 4 S5100s (the "high performance" successor to the S5000: it can actually operate at 90 seconds! .... under ideal vault conditions... ) for a project in India. It was neither fun nor rewarding, since I knew that my homemade VBB instrument could out-perform them 10 minutes after it is put on the pier. Re the period adjustment with the front leveling screw: this is the normal way to trim the period: raising the front makes it longer by decreasing the gravity vector wrt the mass and hinge point. Generally the screw is not turned more than 6 turns from where the base is level; otherwise, adjust the main spring tension screw. This is were there are two clamps on the piano wire, one of which must always be tight or the spring will collapse. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Re: frankcnd@.......... Date: Sat, 11 Apr 1998 10:48:55 -0700 (PDT) Hello there: Fear not... I've not ruled out the use of humans as sensitive indicators for earthquake prediction. I'm only stating that animals are also another area that needs more exploration and study. Thank you for reminding me about Charlotte's predictions. Frank >Looking for predictors or the gifted? 98% close enuf? > >http://www.viser.net/~charking/story.html > >When studying this keep in mind today's >"science" is tommorrows witchcraft. The >strides in quantum physics is closing in on >the bicosmic energies. The gift is something >that the scientific method promoted by the >'analyst' (notice the ANAL) lends itself to >never having a final answer, just the most >convincing argument gathering the largest >world of agreement. >I've experienced this on both ends of study. >When working in the black arts of EMF/RFI >shielding, I got 'results' using what's known >as the common intelligence, tapping in with >stupid things like intention, but producing a >shielding that passed all FCC testing for sure. >This bothered all my superiors, with their MS >degrees, and they spent six months attempting >to figure out what I did, instead of enjoying the >product that would beat everyone to the market. >While intellectually J Offing this, a competitor >got their product to the market and blew my co. >off the marketing scene. So when you get some >project subsystem working, keep going and don't >go for the BOOBY PRIZE - Reasons...or results. >I've found results more satifying. >For Roger's project try reflective mylar (AKA >space blanket material) and use the old 555 >chips, they can work miracles...Keep it SIMPLE, >genius is simple, as the Truth, out their is... > >bschafer@......... wrote: > >> More Earthquake Predictions >> >> Hi KFrank, >> >> Thanks so much for this site. Might be very interesting. Does anyone >> know what Berkland's success rate is? >> >> Ran across another book, today, and it looks very interesting. I have >> only read a couple of pages. It is called "The Trembling Earth." It is >> written by Frederic Golden and is copyright 1983. >> >> Bonnie >> the crafty crafter >> >> _____________________________________________________________________ >> >> Public Seismic Network Mailing List (PSN-L) >> >> To leave this list email listserver@............. with the body of the >> message: leave PSN-L > > > >-- >Michael S.Healy internet: sunmanh@............ >1-503-355-3177 (voice) internet: wave3@........... >1-503-355-3367 (fax) (below RIP-aware as well) >1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program > > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Where's Fontana Trend Map? Date: Sat, 11 Apr 1998 10:54:38 -0700 (PDT) Hi all: I was searching all over the USGS (pasadena) Field Office Web-Site for the recent map that shows the Fontana Trend and how it relates to the recent series of earthquakes in the Inland Empire of San Bernardino county. I don't see it listed anymore. I will email them to see if they can repost it for us. Frank Condon frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: frankcnd@.......... Date: Sat, 11 Apr 1998 11:24:32 -0800 When designing electronics for seismo indicators, don't use Y2D non- compliants - ref da problems - http://www.garynorth.com/y2k/index.cfm Frank Condon wrote: > Hello there: > Fear not... I've not ruled out the use of humans as sensitive indicators for > earthquake prediction. I'm only stating that animals are also another area > that needs more exploration and study. Thank you for reminding me about > Charlotte's predictions. > > Frank > > >Looking for predictors or the gifted? 98% close enuf? > > > >http://www.viser.net/~charking/story.html > > > >When studying this keep in mind today's > >"science" is tommorrows witchcraft. The > >strides in quantum physics is closing in on > >the bicosmic energies. The gift is something > >that the scientific method promoted by the > >'analyst' (notice the ANAL) lends itself to > >never having a final answer, just the most > >convincing argument gathering the largest > >world of agreement. > >I've experienced this on both ends of study. > >When working in the black arts of EMF/RFI > >shielding, I got 'results' using what's known > >as the common intelligence, tapping in with > >stupid things like intention, but producing a > >shielding that passed all FCC testing for sure. > >This bothered all my superiors, with their MS > >degrees, and they spent six months attempting > >to figure out what I did, instead of enjoying the > >product that would beat everyone to the market. > >While intellectually J Offing this, a competitor > >got their product to the market and blew my co. > >off the marketing scene. So when you get some > >project subsystem working, keep going and don't > >go for the BOOBY PRIZE - Reasons...or results. > >I've found results more satifying. > >For Roger's project try reflective mylar (AKA > >space blanket material) and use the old 555 > >chips, they can work miracles...Keep it SIMPLE, > >genius is simple, as the Truth, out their is... > > > >bschafer@......... wrote: > > > >> More Earthquake Predictions > >> > >> Hi KFrank, > >> > >> Thanks so much for this site. Might be very interesting. Does anyone > >> know what Berkland's success rate is? > >> > >> Ran across another book, today, and it looks very interesting. I have > >> only read a couple of pages. It is called "The Trembling Earth." It is > >> written by Frederic Golden and is copyright 1983. > >> > >> Bonnie > >> the crafty crafter > >> > >> _____________________________________________________________________ > >> > >> Public Seismic Network Mailing List (PSN-L) > >> > >> To leave this list email listserver@............. with the body of the > >> message: leave PSN-L > > > > > > > >-- > >Michael S.Healy internet: sunmanh@............ > >1-503-355-3177 (voice) internet: wave3@........... > >1-503-355-3367 (fax) (below RIP-aware as well) > >1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program > > > > > > > >_____________________________________________________________________ > > > >Public Seismic Network Mailing List (PSN-L) > > > >To leave this list email listserver@............. with the body of the > >message: leave PSN-L > > > > > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L -- Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: S5000 LP vertical Date: Sat, 11 Apr 1998 11:32:52 -0800 may be old hat to most but here's a ref book - ISBN - 0-07-025209-2 library 94-23110 /CIP "How To Build Earthquake, Weather, and Solar Flare Monitors" by Gary G Giusti and for you Bill Nye fans, the simplest - http://www.pacifier.com/~sunmanh/edd_tws.html Happy Hoppity Day S-T Morrissey wrote: > Meredith, > > The S5000 LP vertical IS an entirely different sensor from the > S5000 series horizontals; the only thing they have in common > is the coil/magnet assembly. > > I will push your request for materials to the top of the list > next week. > > I will also send you some new hinge material and new spring > suspension wire. Getting the seis to perform as advertised > is difficult enough with the original design, and probably > nearly impossible with substitutions. In particular, if the > hinge crossover point moves at all wrt the upper end of the > spring suspension clamp, the period will change radically > as the boom position changes. > > About the best period linearity that can be obtained is 3% > over a range of + - 2mm at the boom pointer. This means that > if the period is 30 seconds at center, it can be 31 or 29 over > the operating range. > > The adjustment of the upper spring attachment sliding assembly > for linearity can take days, and is all for naught if some > event stresses the hinges after it is done. When these S5000 > were used in the WWNSS, they originally tried for a standard > of 30 seconds (and a 90 second galvanometer), but eventually > backed off to a 15 second standard because of the instability > and linearity problems. > > I have just "tuned up" a set of 4 S5100s (the "high performance" > successor to the S5000: it can actually operate at 90 seconds! > ... under ideal vault conditions... ) for a project in India. > It was neither fun nor rewarding, since I knew that my homemade > VBB instrument could out-perform them 10 minutes after it is > put on the pier. > > Re the period adjustment with the front leveling screw: this is > the normal way to trim the period: raising the front makes it > longer by decreasing the gravity vector wrt the mass and hinge > point. Generally the screw is not turned more than 6 turns from > where the base is level; otherwise, adjust the main spring > tension screw. This is were there are two clamps on the piano > wire, one of which must always be tight or the spring will > collapse. > > Regards, > Sean-Thomas > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L -- Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: RE: S5000 LP Vertical Date: Sat, 11 Apr 1998 15:30:35 -0600 Sean-Thomas, Thanks for the rapid reply. Yes indeed I would be most interested in the new hinge material and spring suspension wire. This critter is tending to be much more difficult than I ever thought it would be. Unfortunately the manual lacks all the distance specs of the wires, hinges, etc., and my present setup is not looking proper at all. I do have one additional question, which is turning out to be a key to matching the magnets/coil alignment. The boom hinge spacing between the clamp blocks is unknown....do you know the distance? The manual references a setscrew on a clamping block as the spacer but this doesn't sound right, as the setscrew actually has two diameters with the head and the thread. I suspect the thread and the head would rest on the clamping blocks? Yes I expect a long time and effort will be put into this machine, from what I've seen and you have mentioned. Being as I'am not really going for professional results, it will serve the amateur results as best as I can manage for a velocity instrument, and perhaps down the road as a VBB conversion project. Eventually I'll be getting into your STM-8, but for the moment I'am doggedly pushing to set up what I have. I shall be glad when it is done as its taken months as is just to get where I am. My vault has been done for acouple weeks and the piers with the horizontals are seemingly to be very stable with no resetting of the zero of either after acouple days of settling in. The piers are set on a plastic cover which entirely covers the vault floor, and hence no moisture has been sensed, and little air humidity is deforming the piers apparently as the sides are painted with a latex/acrylic combo. The piers are 5 & 1/2 inches thick, X 32" X 32" and the floor joists barely allow cover removal for the horizontals...maybe not ideal, but they will have to work. Thanks again for the answers and the future materials...I will of course compensate. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: RADIOTEL Subject: FILTER Date: Sat, 11 Apr 1998 18:29:23 EDT TO: LIST PARTICIPANTS I have just connected a Gould DC amplifier to my recently completed short period vertical seismograph (about 4 Sec. period) based on the mechanical layout developed by Sean-Thomas Morrisey. The amplifier has a built in filter that can be set for (1) no filtering (2) 5 Hz filter (3) 15 Hz filter. Since I live in Southern California where I understand there is a lot of quake activity, which one of the above filtering setting would be best? Jim Allen Cerritos, California U.S.A. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: FILTER Date: Sat, 11 Apr 1998 15:44:14 -0700 At 06:29 PM 4/11/98 EDT, Jim Allen wrote: >TO: LIST PARTICIPANTS >I have just connected a Gould DC amplifier to my recently completed short >period vertical seismograph (about 4 Sec. period) based on the mechanical >layout developed by Sean-Thomas Morrisey. The amplifier has a built in filter >that can be set for (1) no filtering (2) 5 Hz filter (3) 15 Hz filter. Since >I live in Southern California where I understand there is a lot of quake >activity, which one of the above filtering setting would be best? >Jim Allen >Cerritos, California U.S.A. I would go with the 15hz setting since you will be receiving local quakes. Local events have information above 5 Hz, and you need some filtering, so I would go with the higher value. At 15 hz you will need to sample the data at 30hz or higher. For teleseismic events you can use the filtering feature of WinQuake to remove any high frequency local noise. Sounds like an interesting Amp / Filter chip. What's the part number? -Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: RADIOTEL Subject: Re: FILTER Date: Sat, 11 Apr 1998 19:01:24 EDT TO: LARRY COCHRANE The amplifier that I am using is called a Gould Signal Conditioner cage with eight plug in dc amplifiers each having two filter settings and a selectable volts full scale from 5 to .025. C and H surplus (Los Angeles area) carries a lot of these in their warehouse. As I remember, I paid about $50.00 for it. I will take your advice and use the 15 Hz filter setting and use a sample rate of 40hz. Regards Jim Allen Cerritos, California U.S.A. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: low-pass filter Date: Sat, 11 Apr 1998 18:12:15 -0500 (CDT) Jim, I assume that the filter you have is a low-pass (to remove high frequency noise). the biggest problem anywhere is "cultural" noise: cars, trains, us,etc. The frequency content of local quakes peaks at around 10hz for the M=3.+ at distances of 10 to 100 km. So there isn't any defintive answer for the trade off between tuning out the cultural noise and seeing the near-field quakes. I can only suggest what I am doing here at the farm where the Beta seis is in the basement, the house is about 10 meters from the street, and the railroad is 1km away. For a countinuous monitor recorder I'm using an old Geotech "portacorder" modified for DC response and recording with a warm stylus on FAX paper at 15mm/min with 2mm trace spacing (gets me 2.5 days on one record). I have the low pass filter set at 5 hz, but I also have about 40db of filtering at 10 hz in the line driver amp from the VBB seis (the input to this amp is used for the 1000 second high-pass to remove DC signals from the recorder and digitizer). So what do I see: the passing cars and my dogs romping through the house make noises about 1 to 2 mm p-p. We had a little quake at New Madrid on Wednesday: a M=3.2 240 km from here: the p phase was about 4mm, and the s was about 8 mm p-p. Then we had a M=2.8 at 265 km from here, and I only saw the s/lg phase. Yesterday the VBB signal gave a nice record of the M 6.1 quake in the SW Pacific, showing the "trains" of 20-second surface waves from the long path through the thin lithosphere of the Pacific plate. So the bottom line is to experiment. Some level of cultural noise is usually present in most seismic data; it is "acceptable" during the day, and generally vanishes at night. Railroad noise comes and goes over a period of several minutes. Farming and construction have the peculiar signature that they disappear from 12:00 to 1:00, for lunch!. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: RADIOTEL Subject: Re: low-pass filter and freq. resp. Date: Sat, 11 Apr 1998 21:42:43 EDT TO: SEAN-THOMAS, et. ALL I appreciate the input on filtering. BTW, I read the post of Sean-Thomas to the list regarding his evaluation of the circuit in the BSSA to increase seismometer velocity response below the natural frequency. Sean-Thomas indicated that in his test using LT1024s that 1f noise was a problem. Are there lower noise chips (CAZ) than the LT1024s that could possibly improve the signal to noise ratio? Would a chip that would automatically compensate for drift (and possible noise) help and if so do you know of one? What other changes would you suggest? If one could overcome the noise problem, this simple circuit seems a natural for my short period vertical. Jim Allen Cerritos, California _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Jim Hannon Subject: Re: low-pass filter and freq. resp. Date: Sat, 11 Apr 1998 21:17:18 -0500 RADIOTEL wrote: > > TO: SEAN-THOMAS, et. ALL > I appreciate the input on filtering. BTW, I read the post of Sean-Thomas to > the list regarding his evaluation of the circuit in the BSSA to increase > seismometer velocity response below the natural frequency. Sean-Thomas > indicated that in his test using LT1024s that 1f noise was a problem. Are > there lower noise chips (CAZ) than the LT1024s that could possibly improve the > signal to noise ratio? Would a chip that would automatically compensate for > drift (and possible noise) help and if so do you know of one? What other > changes would you suggest? If one could overcome the noise problem, this > simple circuit seems a natural for my short period vertical. > Jim Allen > Cerritos, California > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L Jim, A couple of IC's worth looking at are the Burr Brown OPA27 Ultra Low noise OP amp and the Linear tech LT1052 Chopper stabilized OP amp. Jim Hannon -- Jim Hannon http://soli.inav.net/~jmhannon/ 42,11.90N 91,39.26W WB0TXL _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: low-pass filter and freq. resp. Date: Sat, 11 Apr 1998 23:04:02 -0500 (CDT) Jim, I guess that my bottom line and 2 cents worth of evaluation of the Roberts article is that if you have to go that way (equalization circuits), it is frustrating; I am certainly not suggesting it. We have evaluated many amplifiers, including CAZ designs and some of the best AD and Maxim amps, but they really cannot provide much improvement for high gains, like trying to get the sensitivity of a 5500ohm L4 ( about 300 V/m/sec) from a 4.5hz phone, and I don't think it is worth further pursuit. Of course, IF you have a mechanical sensor that can be configured for VBB feedback (a displacement transducer can be installed, the coil has a low enough resistance, etc.) you are way ahead of the equalization game by implementing the triple VBB feedback. THere are fewer amplifiers involved ( 3 (displacement amp, integrator, VBB output follower/line driver) vs 8 for a fully configured equalizer), and we aren't pushing any specs (unless one goes for a 300+ second integrator), and the result is predictable and calibratable. A 4 second sensor can easily operate at 40 seconds or 120 seconds or beyond. I would suggest this for the next step for your instrument. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Magnetic shielding Date: Sun, 12 Apr 1998 01:44:50 -0500 Friends, I found a good web page at Advance Magnetics (www.advancemag.com) and here is part of what they have to say about shielding. It looks like the best way to go is to buy a roll of their foil and wind enough layers to do the trick. The premeability of high nickel alloys is about 300,000 under optimum conditions, but goes down to 100,000 or less in about 100 Gauss fields, but this is still way, way better than scrap iron at about 100. On the topic of low noise op amps, Linear Technology makes about the lowest noise one, which is the LT 1028. It is made to match a very low impedance source like a coil. In this case, the coil should have a resistance of no less than about 100 ohms, up to 1000 ohms or so, and in this region, the 1028 operates very near the theoretical noise level of a resistor's Johnson noise. No transistor can hear the inherent resistor noise below about 50 ohms, so your coil should have at least this much DC resistance, and probably more. (See Horowitz and Hill). It might be better to obtain or wind a coil with a size and resistance optimally matched to such a premium op amp (and the magnet) rather than the other way around. You can even put such ultra low noise bipolar op amps in parallel if you want to match ultra-low input impedance, but no need here. The Linear Technology Corp. data books have great info on circuits that push the limits in various ways with their premium parts like the LT 1028, which you can order in single quantities through Digikey, which is a very useful and reasonable electronic parts mail order supplier. -Yours, Roger **************************************************************************** *Once the offending field source is identified, one practical approach in determining needed shielding is to order a small quantity of heat treated ready-to-use magnetic shielding foil from a shielding manufacturer. It is available for immediate delivery in various convenient widths, lengths and shielding strengths for high or low permeability requirements with a range of electrical conductivities. Foil is easily, quickly cut with ordinary scissors and hand shaped to the desired outline. It is ideal for R/D, hard-to-get-at places, or for small quantity or extremely compact applications. Many shielding problems thus can be solved quickly. After hand shaping around the component to be shielded, the foil can be held in place with simple adhesive tape. Thickness and number of layers can be determined by ordinary trial and error procedure, or a formula to follow may be requested from the manufacturer. Begin by using a single layer and then adding layers until the desired shielding effect is achieved. When using multiple layers in steady fields and at low frequencies, the low permeability layer should be closest to the field source. This tends to increase the flux density shielding capabilities. The low permeability layer diverts the major portion of the field, permitting the high permeability layer or layers to operate in a lower reluctance mode. If you need relatively few shields or are experimenting, foil is the swift, economical solution. Once foil shielding is functioning satisfactorily in either experimental or production applications, it is time to evaluate the economics. The cost of foil versus prefabricated shields for that particular application should be compared. A prefabricated shield is less costly in larger quantities and for certain complex applications. For designing and manufacturing prefabricated magnetic shields in-house, sheet stock may be ordered. Sheet stock may be formed by bending, stamping, drawing, finishing, etc. on ordinary sheet metal equipment and finished by plating, MIL spec painting, etc. For optimum magnetic shielding characteristics, shields must be heat treated after all forming, welding and machining operations. Your supplier will guide you in the use of the various available states of heat treatment, such as the one which permits ease of forming (mill annealed) or the treatment which assures the maximum mechanically stable permeability or the absolute maximum permeability (which is not necessarily stable mechanically or thermally in some high nickel alloys). High electrical conductivity and high magnetic permeability both contribute to the effectiveness of thin foils in fast-rising pulse shielding by reducing the skin depth. Distinctions have lately been made between the case where the foil thickness exceeds the skin depth and where it is greater. This type of shielding against pulse-type interference achieves the highest order of shielding effectiveness generelly obtained by any means. Attenuations between 300 dB and 1000 dB are not unusual. *************************************************************************** _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: equalizer circuits Date: Sun, 12 Apr 1998 09:25:20 +0100 S-T Morrissey wrote: > > > We have made several evaluations of the P. M. Roberts circuit, > especially trying to make a 4.5 hz phone act like an L4-C at > 1 hz. It does work, but it is noisy, even with the best selection > of amplifiers........................ > The problem with the equalizer circuit is that the gain of > the 1-pole filter is about 20, so the two of them in series > (to get the proper slope) have a gain of 400. ................ Hi Sean, I am a little puzzled about your reply. The article's equations required a gain of 20 from each of 2 stages, for a 1Hz seismo's response to be extended down to .1Hz . Your own experiments appear to make a 4.5Hz seismo go down to 1Hz using the SAME amplification....... .....don't the equations require a different set of RC values for your experiment, and therefor a different amplification? If I read the article correctly, any natural period seismo which is NOT 1 sec.requires new amplification figues, based on the natural period of the instrument before equalisation, (which was 4.5 sec. in your experiment)........should not your amplification have been different then?? I daresay I have the 'wrong end of the stick', (or 'the wrong end of the beam') somewhere, but I am puzzled. Regards Albert Noble (England) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bschafer@......... Subject: frankcnd@.......... Date: Sun, 12 Apr 1998 01:24:27 -0500 (CDT) frankcnd@.......... Hi Frank, It seems that I have heard of this book but can't remember where. I will look at Braille and see if they have it. My reader advisor is a really neat guy and knows I like things concerning seismology so he is always willing to try and help. He was the one that found the book that I am presently reading. If I can get a hold of it I will let you know. Bonnie the crafty crafter _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Giovanni Rotta" Subject: NORTH EASTER ITALY - INTERMED. QUAKE Ml. 5.5 Date: Sun, 12 Apr 1998 13:27:08 +0200 It was just under my chair... My sistem tilt out. Many mild and moderate quake now. Giovanni Giovanni Rotta rottag@.......... Via F. Pizzigoni, 10 33010 Resia (Udine) Italy Lat. 46.373N Long. 13.305E QTH Locator JN66PI _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Seisguy Subject: Earthquake Jolts Alps Date: Sun, 12 Apr 1998 12:34:30 EDT Earthquake Jolts Alps ..c The Associated Press LJUBLJANA, Slovenia (AP) - A strong quake shook a broad area of south-central Europe on Sunday, setting Venice and Vienna swaying and toppling dozens of chimneys in Slovenia, where several people were reported slightly injured. The earthquake was centered in the Alps in western Slovenia and had a preliminary magnitude of 5.5, according to Italy's National Geophysics Institute. The epicenter was in the mountains close to Triglav, a peak popular with climbers, about 15 miles from the Italian border. Slovenian radio said the worst-hit area was around Bovec, a town of 1,500 people where the roofs on about 200 houses were damaged. Part of a church steeple in the town of Kobarida also fell, Slovenian national radio reported. One person in Bovec, 53 miles west of the capital Ljubljana, died of a heart attack blamed on fright, officials said. The mountainous area was blocked off by rockslides touched off by the quake, and phone and electricity lines were down, limiting information from the affected area. Sunday's quake also was felt in Italy, Austria, Croatia, Hungary and southern Germany, although no damage was reported in those countries. People in Milan, Bologna and elsewhere in northern and central Italy jammed fire department switchboards with inquiries. Venice, packed with Easter tourists, rocked for 15 seconds, as did the Austrian capital of Vienna, where tall buildings swayed back and forth. Walls of some high-rise buildings shook slightly in Munich, Germany, about 170 miles to the northwest. A devastating quake ripped through the area just west of Sunday's epicenter in 1976, killing nearly 1,000 people in northeastern Italy. AP-NY-04-12-98 1145EDT _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Walt Williams" Subject: SETV- Fwd[2]: SETI Fw: Aether Tectonics Date: Sun, 12 Apr 1998 10:00:53 +0000 Caruso and all, The original post by Ross Tessien contains a typo, the corrected date is April 21, 1998. I would have posted this correction earlier, but was at NAB conference & show for the past week in Las Vegas, Nv. Walt Williams, 98.04.12 dfheli@.............. ================================================ ------- Forwarded Message Follows ------- From: Caruso Stephen To: Walt Williams Subject: RE: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Predicted Date: Tue, 7 Apr 1998 23:57:00 +0300 aPRIL 31ST IS NOT A REAL DATE. ================================================ From: Walt Williams To: PSN-L Mailing List Subject: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Predicted Date: Tuesday, April 07, 1998 2:42PM Hello All, I am a member of the SETI-L reflector. This interesting post is thought provoking. Best Regards Walt Williams, 98.04.07 dfheli@.............. == Cross Posted From seti@....... == ------- Forwarded Message Follows ------- Reply-to: "Ron Blue" From: "Ron Blue" To: Cc: "Ross Tessien" Subject: SETI Fw: Aether Tectonics: Alert: Merging Black Hole Prediction; Date: Tue, 7 Apr 1998 01:55:45 -0400 Ross Tessien has asked for help in monitoring the likely black hole merger estimated to occur April 31, 1998. Any reports should be forwarded to him. Naturally, data before, during, and after would be useful. Also as a intuitive guess, if the timing is good, radio signals from your favorite SETI source could be stronger around this time. Ron Blue ------------------ deletia ----------------- _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Re: Earthquake Jolts Alps Date: Sun, 12 Apr 1998 10:00:00 -0700 Seismo-Watch issued this earthquake fax alert bulletin earlier this morning: An earthquake registering a preliminary magnitude of Ms 5.7 (NEIC) occurred at 10:55 UTC, April 12, in the Italy-Austria-Slovenia border region in southern Europe. The quake struck at 12:55 p.m. local time and was centered about 27 miles north of Trieste, Italy, and just inside the Slovenia border at the southern margin of the Julian Alps, near the small village of Kobarid. The depth was fixed at 10 km for quick processing and we do not have a Moment magnitude for this event as of yet. The quake knocked out power and telephone lines and caused several landslides. Some homes have been damaged a few people were hurt, none seriously. There have been numerous small to moderate aftershocks. The main jolt was felt as far away as Venice, Italy. A devastating earthquake registering M6.5 struck the same region on May 6, 1976, causing widespread damage and killing nearly a 1,000 people. That earthquake had a robust aftershock sequence which included five more quakes measured M5.7 or greater, including two more in the M6 range, a M6.1 on June 17, 1976, and a M6.3 on September 15, 1976. -- ---/---- Charles P. Watson, Consulting Geologist Advanced Geologic Exploration/Seismo-Watch P.O. Box 18012, Reno, Nevada 89511 Voice: 702-852-0992 / Fax: 702-852-3226 mailto:watson@................ http://www.seismo-watch.com _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: equalizer circuits Date: Sun, 12 Apr 1998 12:24:30 -0500 (CDT) Albert and co, You are correct in observing that we HAVE recalculated the values for the equalizer for our particular application to 4.5 hz phones; and we did believe that the resulting reduced gain requirement would spare us the noise problems. But a difference (reduction) of 2.5 in equalizer gain doesn't help much, especially when it is connected to a 24-bit ADC with a LSB of a microvolt. The gain values for the 1hz to 10 sec equalizer calculate up to 22.1 for EACH stage for a total gain of 488. For the 4.5 hz to 1hz we use a reduced gain, but not that reduced because the new target lower frequency is 2 seconds so that it will look like a properly (over) damped L4-C. THe gain is 14 per stage, or 196 total. But the little HS-1LT phone is well over damped, and has an output of about 20 v/m/sec. So it needs a pre-amp as well as an amplifier after the equalizer to get 300 v/m/sec. The gains of these amplifiers total about 20 (the follower is adjustable), for a total gain of about 4000. This actually is not an unreasonable number: it equals 72db, which is what we generally use for the pre-amp between an L4-C and the telemetry VCO in remote stations. But the telemetry itself has a noise floor of about a millivolt, so amplifier noise is not a problem. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: equalizer circuits Date: Sun, 12 Apr 1998 12:24:30 -0500 (CDT) Albert and co, You are correct in observing that we HAVE recalculated the values for the equalizer for our particular application to 4.5 hz phones; and we did believe that the resulting reduced gain requirement would spare us the noise problems. But a difference (reduction) of 2.5 in equalizer gain doesn't help much, especially when it is connected to a 24-bit ADC with a LSB of a microvolt. The gain values for the 1hz to 10 sec equalizer calculate up to 22.1 for EACH stage for a total gain of 488. For the 4.5 hz to 1hz we use a reduced gain, but not that reduced because the new target lower frequency is 2 seconds so that it will look like a properly (over) damped L4-C. THe gain is 14 per stage, or 196 total. But the little HS-1LT phone is well over damped, and has an output of about 20 v/m/sec. So it needs a pre-amp as well as an amplifier after the equalizer to get 300 v/m/sec. The gains of these amplifiers total about 20 (the follower is adjustable), for a total gain of about 4000. This actually is not an unreasonable number: it equals 72db, which is what we generally use for the pre-amp between an L4-C and the telemetry VCO in remote stations. But the telemetry itself has a noise floor of about a millivolt, so amplifier noise is not a problem. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: Morrissey e-mail Date: Sun, 12 Apr 1998 20:38:03 -0700 Thanks Bob (and ST Morrissey for taking the time to write them!), I ZIP the file (62k) and placed it on my system at: ftp://psn.quake.net/info/stm-mail.zip There is a link to the file at: http://psn.quake.net/infoequip.html -Larry At 02:03 PM 4/12/98 -0400, you wrote: >Larry, > I have found the e-mail from ST Morrissey so enlightening on many >subjects that I have collected all his messages into a single file. I >removed all the header junk and many blank lines (to use less paper when >printed). I may have missed a few but I think that it is most of them up >to 4/8/98. > This file (193k, pure ASCII) is attached. If you think that it would be >of sufficient interest to the group, put it on your page somewhere and >announce it. I didn't want to send that monster to the whole list. This >info might be a convenient source for those building his vertical. >Bob Barns _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: George Harris Subject: Magnetic shielding materials Date: Sun, 12 Apr 1998 19:47:01 +0000 For those intersted in the characteristics of magnetic shielding materials the following is taken from a catalog from Spang Specialty Metals (Home in Butler PA - Office and warehouse in Los angeles). The best materials are Permalloy 80 and Mumetal. They are 80% nickel with the remainder iron. To work the best they must be annealed at very high temperature in hydrogen for 4 hours. Spang supplies a material called Mu Guard 80 which is annealed ready for use. It is available in thickness from .001 to .006 in widths up to 6 inches. As was stated previously, it must not be bent or deformed if the maximum shielding is to be achieved. About 6 years ago, I purchased about 3 feet of material 4 inches wide which I remember cost near $50 when I was experimenting with flux gate magnetometers. They seem to be willing to talk about small orders though there is a minimum. George Harris _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: STM-8 Instrumentation Amplifier Date: Sun, 12 Apr 1998 22:48:43 -0600 Hi Sean-Thomas, Think I have missed out on the instrumentation amplifier in regard to the gain, and perhaps any other pertinent data. I know you use 3 op amps, and have stated that about any opamp would work. Might a chopper op amp be preferred over others....and is there a particular opamp/chopper op amp you have a preference for over others, if you had your free choice as opposed to the best of the cheapest? P.S., any progress on the sealed container? Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: integrator amp Date: Mon, 13 Apr 1998 01:14:04 -0500 (CDT) Meredith, I assume that you are referring to the integrator amp; I use a 3 amp instrumentation configuration: 1: because I have a box of them left over from a self-potential experiment (the Greek VAN idea .... nothing but noise) and 2: it lets me try wild ideas, like 1000 second integrators. Actually a single quality amp will do, like the OPA111, for a reasonable time constant. THe integrator gain is one, connected as a non-inverting follower; for lowest noise, the follower connection (from the output to the inverting input) should be a resistor of the same value as the input. Regarding the barometric contaiment: I have been doing an interim experiment to try to counter the buoyancy of the mass/boom with sealed containers (pill jars) hanging out the other side of the hinges from the boom. I've estimated the volume of the mass etc. at about 100 cc, located 38 cm from the hinges, or 3800cm*cc. To counter this, I have three pill jars (with lids sealed with RTV) of 128 cc ea, mounted 8.5 cm outside the hinges, for a volume*distance of 3264 cm*cc. We have had very windy weather lately, and this reduced the wind noise by about a factor of 10. Since it is almost impossible to fine tune the volumes in this scheme, I don't think it is a full solution, so I will pursue the aquarium containment idea. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: We're not in Kansas anymore (Greetings from Oz) Date: Mon, 13 Apr 1998 10:39:18 +0800 Larry- I have spent the last couple of days with Arie Verveer touring Perth and vicinity. I have seen his beautifully crafted S-G paired NS, EW horizontals in the vault below one of the many observatories of the Perth Astronomical Observatory -- where he has coupled the movements of the stars to the movements of the Earth. We spent this afternoon with Alby Judge and his wife at their new homestead on top of the Darling Scarp (where the PAT also sits) that is equipped with a Wilmore vertical sitting on a concrete pad in the bush covered with an old freezer box for protection/insulation -- a major source of seismic noise is the coming and going of kagaroos. Arie and Alby are doing some very interesting work with the public and plan to do more. Are you still awake? -Edward _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: [Fwd: We're not in Kansas anymore (Greetings from Oz)] Date: Mon, 13 Apr 1998 10:41:24 +0800 Larry- I have spent the last couple of days with Arie Verveer touring Perth and vicinity. I have seen his beautifully crafted S-G paired NS, EW horizontals in the vault below one of the many observatories of the Perth Astronomical Observatory -- where he has coupled the movements of the stars to the movements of the Earth. We spent this afternoon with Alby Judge and his wife at their new homestead on top of the Darling Scarp (where the PAT also sits) that is equipped with a Wilmore vertical sitting on a concrete pad in the bush covered with an old freezer box for protection/insulation -- a major source of seismic noise is the coming and going of kagaroos. Arie and Alby are doing some very interesting work with the public and plan to do more. Are you still awake? -Edward From: Karl Cunningham Subject: Re: Magnetic shielding materials Date: Mon, 13 Apr 1998 10:39:47 -0700 A source of magnetic shielding materials is old dead oscilloscopes. Usually, the CRT is surrounded with a drawn and annealed metal shield. You can cut it with a pair of tin snips, but bending it will reduce its magnetic shielding properties substantially. Look at surplus stores for old junk scopes. Karl Cunningham La Mesa, CA. PSN station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: John Taber Subject: Re: ASCII data Date: Tue, 14 Apr 1998 10:34:21 +1200 (NZST) Larry, I don't think there are any standards for storing ASCII versions of binary seismic data, but I think it works best to have the header info followed by the data in the same file. I like having text lables in the ascii version of the header so that it is easy for humans to read and edit if necessary. Here is an example of part of the ascii form of the ah format that is used by a number of seismology groups: station information code: W01 channel: spe type: L4 latitude: -41.000000 longitude: 174.750000 elevation: 0.000000 gain: 59499999232.000000 normaliztion: 1120.369995 calibration information pole.re pole.im zero.re zero.im 5.000000e+00 0.000000e+00 4.000000e+00 0.000000e+00 -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 -1.332900e+02 1.332900e+02 0.000000e+00 0.000000e+00 -1.332900e+02 -1.332900e+02 0.000000e+00 0.000000e+00 -3.769900e+00 5.026500e+00 0.000000e+00 0.000000e+00 -3.769900e+00 -5.026500e+00 0.000000e+00 0.000000e+00 event information latitude: -40.60 longitude: 173.65 depth: 97.6 origin_time: 1991 11 7 16 47 51.75 comment: null record information ndata: 5458 delta: 1.000000e-02 max_amplitude: 1.834650e+02 start_time: 1991 11 7 16 48 5.440000 abscissa_min: 0.000000e+00 comment: null log: null data: 4.535000e+00 1.535000e+00 1.153500e+01 3.153500e+01 etc I'd recommend you include the same parameters in the ascii format that you have in the PSN binary format. That way you can go back and forth between ascii and binary without any loss of information. John Taber School of Earth Sciences Victoria University of Wellington _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Tony Carrasco Subject: VR-60 Date: Mon, 13 Apr 1998 16:02:04 -0700 Cal State LA is looking for any information on a seismic recorded - Sprengnether VR-60, specifically pictures/diagrams. According to them (via the company) we purchased one here at SDSU back in the mid-70's. Unfortunately, if we did purchase it, the recorder and any information about it is now long gone. If anyone can help I'm sure they would appreciate it. Thanks ***************************************** ***** Tony Paul Carrasco ***** ***** San Diego State University ***** ***** tcarrasc@................ ***** ***************************************** || || || /\/\ ///\\\ ////\\\\ /////\\\\\ //////\\\\\\ ______///////\\\\\\\______ _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: benioff seis Date: Mon, 13 Apr 1998 18:12:58 -0500 (CDT) Bob, The Benioff is a rather narrow band instrument, even when it is operated well over damped. In the WWNSS configuration, the peak magnification (of about 150K) is actually at 0.6 seconds when the nominal magnification is 100k at 1 second. At 6 seconds, it is down to about 600, and it is about 100 at 10 seconds, and at 20 seconds, where teleseismic surface wave energy peaks, the magnification is about 15. So it takes a good size teleseism to see much from a Benioff seis. The sensor in the Benioff is a variable reluctance type. Rather than having a coil moving through an orthogonal magnetic field, the coils are wound around the pole pieces of a large H shaped (H is on its side) magnetic structure, where the vertical center bar of the H is the charged magnet. As the suspension moves the top and bottom bars with the coils wound around them, the gap between them, above and below the fixed magnet, changes, causing a flux change in the moving bars, inducing voltage in the coils. When it was first designed (in the 30's) it provided the strongest signal from the poor ferromagnetic materials that were available then. It also has a serious electronic variant that would make me hesitate to use it as a VBB fedback sensor (not that I wouldn't like to try: I have about 6 of them): it has a very large inductance, about 16 henries when all the coils (8) are in series for maximum output. Of course, for feedback, one would use a low resistance parallel configuration. It also has a rather large mass (~15 kg for the "portable" unit (100 kg for the "big" Benioff) ) that would be difficult to work with considering that the generator constant of the coils is rather weak because of the variable reluctance configuration. The output is 135 -150 v/m/sec with the 80 ohm coil connection (where the inductance is still 5+ henries because of mutual inductance). It would probably work with an equalizer circuit at 10 seconds, but this would mostly enhance the 6-second storm microseisms rather than any data of teleseismic interest: teleseismic surface waves run from 15 to 30 seconds. So I guess that if you want some real LP data, you will have to make a broadband sensor. Speaking of which, the VVB here just showed some very large barometric acoustic waves, about 50 mv p-p at 100 to 200 seconds, and the tornado sirens went off, and now the news reports a tornado lifted some roofs, etc, about 2 miles east of here. Regards, Sean=THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: VR-60 Date: Mon, 13 Apr 1998 18:29:37 -0500 (CDT) Tony, Sprengnether Instruments Inc. is still very much in business, and just made another round of VR-60 recorders. If you call them at 314 565 1682 about your missing recorder, they should have the info on it; they have records of everything that they have ever made. Not that they might know where it went, but I'm sure they could sell you a replacement. They should be able to provide a serial number to help your search. If you need additional info, contact me at sean@............ we use the VR60 as a monitor recorder at our broadband seismic stations located in public parks, etc. In addition to being an excellent educational tool, it is an instant diagnostic of whether the station is working; park personnel may not know what the record means, but they call me the minute is stops making wiggly lines. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: frankcnd@.......... (Frank Condon) Subject: Re: Where's Fontana Trend Map? Date: Mon, 13 Apr 1998 19:26:18 -0700 (PDT) Hi All: I've been working with Katrin, the Webmaster at SCEC to get navigation links in place for the chino98.GIF and crafton98.GIF files on their Web-site (now part of TriNet). Wouldn't it be nice if all interested persons could view the information included on those Map files, instead of only us seismic types. Frank Condon frankcnd@.......... "Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" >Hi all: >I was searching all over the USGS (pasadena) Field Office Web-Site for the >recent map that shows the Fontana Trend and how it relates to the recent >series of earthquakes in the Inland Empire of San Bernardino county. I don't >see it listed >anymore. I will email them to see if they can repost it for us. > >Frank Condon >frankcnd@.......... >"Located in the seismic corridor somewhere between Fontana & Mammoth Lakes" _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: S5000 & STM-8 Misc. Date: Mon, 13 Apr 1998 21:00:40 -0600 Sean-Thomas, OK on the instrument amplifier/integrator oneness...terminology. Have never heard of your pill jar barometric approach experiment. Sounds like a possible means in the future for noise reduction with those who may not use the sealed container. I'am presuming they were plastic bottles, and not glass? What if the volume*distance slightly exceeds the mass/boom? Alot of interesting Mu-metal/Permalloy talk. The S5000 uses what appears to be 2 steel plates between the magnets and the spring, as a form of weak shield or field containment. Have you ever used old oscilloscope mu-metal permalloy shields for similar seismo magnet/spring conversions or improvements? For the one I have, immediately after the shield went in, the mass rose....not sure if may have disturbed something else or not....or is repeatable, didn't take the time to remove/reinstall. Could be the shield created more field and raised the spring?...no gaussmeter here. The Mu-metal/permalloy sounds like a possible luxury item for the STM-8, if one is lucky enough to get an old oscilloscope shield per Al Allworth & Karl Cunningham mentions. Found the right clamping block and hinge distance, per using drill bits as spacers, which allowed the coils and magnets to fit. Takes awhile & some tricks to setting these S5000 verticals up. Could use a pulley & rachet mechanism for the spring extension-ha. Think I'am getting the HANG of it-ha. The finer points not just yet. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: magnetic shielding, baro comp. Date: Tue, 14 Apr 1998 00:17:28 -0500 (CDT) Meredith, I guess I have to chime in here about magnetic shielding, etc. My experience with it is that unless one wants to be an expert about it, it causes more problems than it solves. FOr my degree work I rebuilt a 3.2MEV cyclotron with at hugh (13 ton) electromagnet (that ran on 300VDC at 110 AMps, regulated to 0.001%), with a 30" pole face and a field of 13 kgauss across the 10 inch opening. So dealing with this field was a constant problem, and it was very dangerous. It would make a rattle out of a mechanical watch at about 3 feet. We had to operate our TV monitoring cameras via mirrors with telephoto lenses, after wasting a lot of time with trying to shield them with mu-metal, etc. We made magnetic quadrupole lenses to focus the particle beam, and keeping them from interacting with everything else and vv. was an endless problem. So I have been frustrating myself with magnetic shielding ideas for many years. As for any need for such for the STM8 seis, I don't see any, and haven't made any efforts to use any shielding. Because of the propensity of permaloy, mu-metal, etc.,, to become as useful as sheet metal if it is reformed (cut, bent, etc) after final annealing, only a professionally fabricated shield would be worth considering. The STS-1V has such a shield in the form of a box that fits over the sensor, but I don't know of any other sensor using such. The S5100 LP has no shielding, and counts on geometry and design to minimize the effect of the magnets on the main spring. As long as there is relatively little movement of one part wrt another, any magnetic effects would be static. Your experiment with your S5000 demonstrates that the spring IS attracted very slightly by the main magnets, and that this is modified by any intervening ferrous material, as you observed. There is no need to try to modify it, though; there are over 500 of those LPs operating as they came from Sprengnether; and in the later re-design, the shield plates were eliminated. Re the pill jar (plastic) barometric flotation compensation: as I mentioned, figuring out the volume*distance of all the mass + boom + coils + etc. is nearly impossible. And trial + error methods cannot approximate the complex integration. And I believe that other pressure-induced noise sources play a major role, like the breathing of the coil-magnet assembly with pressure change. I'm only seeing 10 to 20 mv of noise, which amounts to about 12 to 24 nanometers of displacement (with 400 mv/micron and 2x gain in the line driver output). I think that wherever the "sealed" wooden shelf board box breathes, it makes minute jets of air that blow the mass around. If I observe the high-gain output of the microbarometer, I don't see a consistently repeatable correlation with the noise. So the "counter-flotation" scheme may have some benefit, but is not a solution. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Re: magnetic shielding Date: Tue, 14 Apr 1998 16:17:36 -0500 Friends, To clarify the recent discussion about magnetic shielding, I think my seismometer definitely needs magnetic shielding while Sean-Thomas's probably does not for everyday use. While his uses the pure mechanical spring force of a steel spring (which is weakly magnetic), mine uses pure magnetic force, and is thus much more sensitive to magnetic fluctuations in the ambient field. Remember we are trying to measure a few parts per million change in the earth's gravity combined, in a seamless way, with some undetermined amount of accelerative force. Our only practical option is to measure this force by exactly balancing it against some other equal and opposite force like a metal spring or magnetic force while assuming that any slight momentary imbalance detected must be due to acceleration, since the earth's gravity only changes slowly and by a very small amount (or pendulum clocks would never work). So in contrast to Sean's design, mine sees such things as as magnetic storms and stray pocketknives much more easily. In addition, ceramic magnets such as those I use ($1.79 for five at Radio Shack) are decidedly sensitive to temperature. So my design has to be thermally regulated, as I am well aware. Currently it tends to drift out of range, but this is probably mostly due to fact that it doesn't currently have a feedback temperature control, as some earlier versions did. I anticipate this will be done in this case by evenly gluing a bunch of resistors and a thermistor to the metal shield and surrounding everything with styrofoam. The tradeoff, as I see it, is that my design is rather small and cheap and easy to build, and as a DC sensitive device, it may also be used as a gravimeter. Since it is relatively small, it can be sealed against barometric changes relatively easily also, by hermetically sealing it inside a glass jug. I have scrounged what is probably a mu-metal or permalloy high nickel alloy cylinder 2 1/2 inch in diameter and about eight inches long, seam welded and lead grey in color-- out of an ancient (circa 1950?) oscilloscope I had gathering dust. (Does anyone have a use for the ancient 3AP1 CRT that fit inside?). I think I can fit a slimmed down version of my currently operating vertical into this shielding tube. For other models, since the foil sold by modern shield suppliers is very thin, it requires little distortion when bent around tubes, which is its intended purpose. I anticipate the current materials would probably do a similar job at a cost of roughly $25. So the bottom line is that my instrument is small and in all liklihood needs to be pampered with a mag shield and most certainly temperature control, and is also a work in progress. Its too early to say how useful my work is at this point, without a good basis for seigmographic sensitivity comparison, but everything is pretty easy to duplicate. I can send anyone a JPEG of the model that I discussed the other day, which will make the mechanical principles reasonably clear, when combined with my recent posting. (Incidently, I'm pretty much off the wall when it comes to protyping technique. In the first go-round, I often try to use carefully cut pieces of window glass and silicone rubber to build everything, as well as other materials when I need to. Protypical devices and instruments built in this fashion may be built very fast (since the thixotropic rubber sets up in ten minutes or so at each stage). The result is very cheap, can usually be made as rigid as needed by proper bracing, lasts forever, and yet may be cut apart and modified very easily with a razor blade. I'm quite serious about how useful this method is for building prototype frames and fixtures and mountings rapidly.) --Yours, Roger _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: amp selection vs noise Date: Tue, 14 Apr 1998 17:42:43 -0500 (CDT) Barry and all, I finally found the article I referred to last week but could not find the reference: "Limits of Sensitivity of Inertial Seismometers with Velocity Transducers and Electronic Amplifiers" by M.A.Riedesel, R.D.Moore, and J.A.Orcutt; BSSA Vol 80, No 6, pp 1725-1752; December 1990. It makes an important contribution in showing the importance of the selection of the proper amplifier for a given coil (with damping) resistance. It shows in figure 3 (which has one axis mis-labeled as "frequency" when it should be labeled "coil resistance") such things as that using a OP-27 rather than a LT1012 (same as the dual LT1024) can reduce the noise when using an input of 400 ohms by an order of magnitude. It also shows that while a LT1028 is very quiet with a 80 ohm input, it is 1000 times noiser than a CAZ type when connected to a very high impedance coil like 30k ohms. It has a very low voltage noise specification, but at the expense of a very high current noise. The paper shows numerous instrumental measurements with various combinations of sensors, including a 4.5 hz geophone. Their bottom line is that for low frequency data (10 to 20 seconds) from small velocity sensors, amplifiers need to be improved by one to two orders of magnitude to resolve earth noise below 10 to 20 seconds. Their recommendation is to use fedback sensors with displacement transducers. However, these are fragile, "power hungry", etc., and not (yet) suitable for such applications as OBSs (Ocean bottom Seismographs). Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "James M Hannon" Subject: Re: amp selection vs noise Date: Wed, 15 Apr 1998 07:24:11 -0500 Sean, You said "It also shows that while a LT1028 is very quiet with a 80 ohm input, it is 1000 times noiser than a CAZ type when connected to a very high impedance coil like 30k ohms." I am not familiar with the term "CAZ", could you explain? Thanks, Jim Hannon _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Dewayne Hill Subject: Re: amp selection vs noise Date: Wed, 15 Apr 1998 08:26:31 +0100 Lately I've seen a few posting about amp's. My I suggest one. Check out the following URL. I use amp and I'm very happy with the preformance. It is very stable (NO drift - none - nata) and to me it seem very quite. I made one modification and that was in the offset adjustment circiut. To allow for a more precise adjustment I replaced the 100k ohm pot with a 2.5k ohm pot and added two 51k ohm resistors, one on the in coming -12v and the other one the in coming +12v. I used all metallized film caps. Here is the URL www.info2000.net/~aloomis/seisfilt.htm James M Hannon wrote: > > Sean, > You said "It also shows that while a LT1028 is very quiet with a 80 ohm > input, it is 1000 times noiser than a CAZ type when connected > to a very high impedance coil like 30k ohms." I am not familiar with the > term "CAZ", could you explain? > > Thanks, > > Jim Hannon > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L -- Dewayne (n0ssy) & Linda (n0uqf) Hill Westminster, Colorado USA 39.83569n 105.06270w. Elev 5660ft. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Velocity Coil Placements on "U" Shaped Magnets Date: Wed, 15 Apr 1998 11:43:21 -0600 Most amateur coils and magnets use coils placed between the poles of "U" shaped magnets...usually alnico magnet varietys. Another type use a speaker type arrangement with the coils entering into the central hole of donut like magnets with a steel core opposite pole field mechanical "intensifer". There are also coils going around the outside of the magnet pole/s ends with whatever shape. My limited experience in comparing coils, suggests that the coil around the outside of the magnet pole/s, would be a more efficient use of say a "U" shaped magnet than one coil between the poles, for a majority of home brew builders. Obviously, one "U" magnet could then have 2 coils for sensing and or possible damping or two outputs if another method of damping is used. My own available coils are usually round relay coils of various wire sizes and resistances. I do also have some square coils retrieved from old meter movement chart recorders. The square coils will crudely fit over the individual poles of some of my magnets, and seem to crank out a voltage much more higher than my round coils between the poles. Roughly the difference in like ohm coils seems to climb up to 10X difference. Only have 3 different square coils, two with 1500 ohm ranges, and they would peak around 2 volts. The 60 ohm coil would peak around 200mv. I do have some round coils that generate well, but they tend to be small gauge and their resistance climbs up into the 5-10 K range, and seem to be rare surplus finds overall. The pole surrounding coils seem to be a much better approach for higher output ratio and less wire useage, perhaps....a best way for managing the coil to fit the op amp desired. Logic says (too me) that the pole covering coils actually are exposed too more total gauss via the coils surface than the generally smaller coils between the poles. The larger signal may also be a nice preamp reduction in gain and noise. Basically any coil can be made with a form and wound. I am only suggesting that if you are going to wind a coil, try a form that will fit over the ends of the poles, and see the results. Or, wind both variations and compare. I'am writing with almost nill authority here, but the subject keeps showing up on my crude results of comparisons. Thanks, Meredith Lamb .... the lunchbucket seismometrist _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Walt Williams" Subject: Radio Emissions, Electrophonics and Earthquakes Date: Wed, 15 Apr 1998 11:02:36 +0000 Hello all, I found an interesting excerpt which I think some members of the PSN-L may find interesting. Please forgive any typos or spelling errors, a finger is wrapped in tape. Best Regards, Walt Williams, 98.04.15 dfheli@.............. http://www.pacificnet.net/~dfheli ===================================================== "Progress in Explaining the Mysterious Sounds Produced by Very Large Meteor Fireballs (bolide)" Excerpts taken from Journal for Scientific Exploration (JSE, Vol 7, No. 4, pp. 337- 354, 1993, Colin S. L. Keay, Dept Physics, University of Newcastle, N.S.W., Australia) ppg 350 "An intriguing phenomenon, which may also result in the direct transduction of ELF/VLF electromagnetic energy into sound, is the correlation often reported between strong radio emissions and subsequent earthquakes (Corliss, 1983). A number of reports mentioning a "rushing" sound preceding earthquake shocks were gathered by Milne (1841). More recent accounts of such sounds may be lacking because of greater incidence of similar man-made artifacts reducing public alertness to sounds of seismic origin. However, audio frequency electromagnetic disturbances associated with earthquakes have been discussed in this journal (Parrot, 1990), while laboratory studies mentioned by Johnston (1987), and others, show that rock fractures generate electromagnetic signals. Cress and his co-workers (1987) recorded signals peaking in the range from 900 Hz to 5 kHz. Field studies conducted by O'Keefe and Thiel (1991) during large rock-blasting operations revealed a series of electrical pulses with a repetition frequency as high as 5 kHz and an amplitude of several volts. The substantially greater energy release in an earthquake could be expected to generate signals many orders of magnitude higher in amplitude. The connection has yet to be confirmed between these experimental observations and the alarm frequently exhibited by animals immediately prior to an earthquake and of course, the sparse reports of earthquake sounds by human observers." and, ppg 351, "Lastly, it is essential for geo-scientists to take reports of audible phenomena more seriously in order that some progress can be made in identifying the physical mechanisms involved. The subject of electrophonic sounds from bolides is now considered respectable within the meteor science community (Keay, 1992b), and a similar shift is now overdue within the communities of auroral and seismic scientists." _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: CAZ amps Date: Wed, 15 Apr 1998 19:23:27 -0500 (CDT) Jim, The CAZ amplifiers are a form of chopper stabilized amplifier that has (theoretically) no 1/f noise. CAZ stands for "commutated autozero", which also reduces the effect of input offset voltages. I believe that they are discussed in Horowitz and Hill (I loaned my copy several years ago ... haven't seen it since). Since they are designed as DC amplifiers, they get very noisy above 1 hz. THey are also a poor choice for low impedance sources. They generally are NOT low power devices. (the BSSA paper by Riedesel et al mentions a GE type icl421) Regards, Sean-Thomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: coil design Date: Wed, 15 Apr 1998 19:24:15 -0500 (CDT) Meredith, Some of the mysteries of coil design are explained in the first appendix of the BSSA article I mentioned yesterday. Full design criteria can be found in the handbook "Reference Data for Radio Engineers" from the ITT press. In general, the output voltage increases with the square root of the number of turns or the resistance, or a 8000 ohm coil has only four times the output of a 500 ohm coil. OR, if I wanted to double the generator constant of my speaker coil of 160 turns, it would need 640 turns. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: Radio Shack Speaker Date: Wed, 15 Apr 1998 19:25:43 -0500 (CDT) Karl, Thanks for the info on your speaker coil. I expect that there would be some variations in such a consumer product. But you list the RS number as 40-1359, which may be a better speaker to use if it has a greater constant. I don't have any catalogue info on it. My original speaker #40-1349 has a constant of 12.988 N/A. The Beta instrument (also a 40-1349) is 11.815 +,-0.02 N/A. Both were measured as installed on the seismometer, using the displacement detector as a null indicator (+.- 5 microns), and 1 and 2 gram weights to balance with a current read by a 4.5 digit DMM. About 2mm of the coil is visible above the face of the magnet pole. The specification sheet lists the flux density as 7.5 kgauss; I can borrow a gauss meter and compare the four speakers I have to see how much variation we can expect. We may also have to pay attention to the position of the coil. Re using the aluminum coil former/heat sink as a calibration turn, I would be concerned about it shorting to the pole in the event of slight mis-alignment. It may not be aluminum if you were able to solder to it, aluminum generally does not solder. I have also found some indication of mutual inductance (transformer coupling) between my 5-turn coils and the main coil, even though the cal coils are about 2mm above the main coil. The problem starts to show up when using sine wave calibrations above 1 hz or so. THis problem could be serious if the two coils fully overlap. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Seisguy Subject: Small event near Dodger Stadium Date: Wed, 15 Apr 1998 20:29:31 EDT I work in a highrise in Encino and didn't feel a thing! Mike ------------------------------------------------------------------------------ --------------------------------------- LOS ANGELES, April 15 (Reuters) - A minor earthquake measuring 3.2 on the Richter Scale jolted Los Angeles at lunchtime on Wednesday, but there were no immediate reports of injuries or damage. The National Geological Survey in Boulder, Colorado, said the tremor at 1:13 p.m. (2013 GMT) was centered three to five miles (five to eight km) north of downtown Los Angeles and was also felt in the suburban San Fernando Valley. A spokeswoman for the California Institute of Technology in Pasadena said the tremor was a fresh earthquake and not an aftershock to the 6.7 magnitude 1994 Northridge quake, which killed more than 50 people and caused an estimated $40 billion in damages. The Northridge quake has so far spawned about 14,000 aftershocks, most of which have been too small to be felt. REUTERS 19:19 04-15-98 _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: Morrissey mail compiled Date: Wed, 15 Apr 1998 21:20:41 -0400 Hi gang, I have learned a lot about many things besides the STM-8 from Morrissey= 's mail. I thought that the information would be more accessible if it were= made more convenient so I collected all the messages up to about 4/8/98 (= I may have missed a few) into a single file. I removed all the header junk= and a lot of blank lines (so it takes less paper to print). = It (62k of text) is avail. at ftp://psn.quake.net/info/stm-mail.zip Larry sez that there is a link to the file at = http://psn.quake.net/infoequip.html Bob Barns _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: VBB calibration Date: Wed, 15 Apr 1998 18:05:35 -0700 Sean Thomas In checking the calibration of my VBB vertical is the displacement gain (V/M) measured at the output of the VRDT or the "mass position" output? I want to measure it at the "broadband output" but I guess this is velocity. Does the "A1" amp have a gain of 10? With my displacement sensor the output of an instrument amp goes thru a highpass filter and directly into the "Broadband velocity output" junction. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Fiberglass-Epoxy Circuit Board Date: Wed, 15 Apr 1998 20:22:23 -0600 Sean-Thomas, Thanks for the BSSA & Reference Data for Radio Engineers suggestions for learning. I'am sure they will give alot more insight. Am looking forward to a copy of the BSSA soon at the library. In regards to your STM-8 and its aluminum angle brackets for the hinges...I've pondered the use of double or triple thicknesses of printed circuit boards with no copper, with support angles as a possible temperature co-efficient improvement on your unit? I suppose the electronics manufacturers may know more of its temperature coefficent, than I have any access too. I presume it is fiberglass and epoxy composition. Obviously, the top boards or hinge attachment sides could be copper plated to the hinges for signal, etc wires also. Just a thought without the tce data. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: VBB calibration Date: Wed, 15 Apr 1998 22:18:48 -0500 (CDT) Barry, The displacement output as used in the VBB transfer function is measured after the gain of 10 amplifier (that is A1) when the feedback elements are disconnected. In my instrument here it is about 370 000 volts/meter (10x 37 mv/micron from the VRDT). An important point of the VBB feedback is that this SAME output that BECOMES the broadband VELOCITY output when the feedback is connected. This is the voltage that needs to be buffered with your high-pass amplifier to remove the DC components of the signal that would saturate your recorder. The actual output of the instrument is then calculated by the transfer function itself, and is of the order of 1600 v/m/sec. Note that the VBB output voltage (befor your high-pass filter) and the mass position output will approach the same value when there is no major input to the sensor. I use the mass position voltage to determine when and how much remote zeroing is needed. I assume that your high pass filter is not within the feedback loop. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: VBB calibration Date: Wed, 15 Apr 1998 20:31:41 -0700 S-T Morrissey Thanks. The next step is outside! Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: A Magnetic Spring Seismometer Description Date: Wed, 15 Apr 1998 23:58:19 -0600 Roger Baker, With reference to your report of 4/10/98, via email. Thanks so much for the JPEG views of you magnetic instrument. I enjoyed the views, and was amazed at the clarity of the side view especially. I can appreciate the idea of your device as one which probably alot of people have speculated and dabbled in occasionally, usually with little sucess, in a variety of designs. Experimentation is usually a delight to see and do....fun stuff. I have to admit I have had to read your progress report several times, to grasp most of it. Anyway...I do have questions and/or speculations (take your choice), to present... 1. If...one were to use much larger diameter ferrite magnets and of course a larger mass, would not the period be lenghtened somewhat...or would it stay the same? 2. The utility knife pivot distance does not look like it could be lengthened, without collapse...or could it? 3. Would a Hall Effect sensor be much better, with its much larger range of motion, than the photocells? 4. Could the lead mass, be instead made of a copper coil, as part of the feedback control and perhaps do away with the coil underneath? Perhaps this way, the response might be more linear to disturbances. 5. Some varietys of stainless steel....#314 I think....will also do an assumed variety of repulsion, to a limited extent. With the mass being part 314 variety and part feedback coil, I would wonder if the magnetic influence noise would shrink somewhat, in comparison to the solid magnet? I have not seen any magnetic repulsion instrument like yours, and that I find it interesting. Keep up the experimentation. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: A Magnetic Spring Seismometer Description Date: Wed, 15 Apr 1998 23:28:58 -0700 At 11:58 PM 4/15/98 -0600, Meredith Lamb wrote: >Roger Baker, > >With reference to your report of 4/10/98, via email. > >Thanks so much for the JPEG views of you magnetic instrument. >I enjoyed the views, and was amazed at the clarity of the side >view especially. Roger also sent me picture of is sensor. I have placed a copy of it at: ftp://drnet.com/users/psn/info/dcp00494.jpg Its a 300K jpg file. I asked Roger about the camera he used to take the photo. Heres what I got back from him; "My camera is a Kodak 210 that sells for about $800 and takes frames with about 800x1000 pixels. This is a little better resolution than most, but others are still better, like the Olympus, which apparently has a big detachable lens for about $1,200." -Larry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: A Magnetic Spring Seismometer Description Date: Thu, 16 Apr 1998 07:49:17 +0100 Larry Cochrane wrote: > > > Roger also sent me picture of is sensor. I have placed a copy of it at: > ftp://drnet.com/users/psn/info/dcp00494.jpg > > Hi Larry, I wonder if Roger could be persuaded to make a sketch, with the main relevant parts marked, and place a copy of that, on web? The photo above is so intrigueing, but I am not sure what I am looking at. (I know what a yellow croc clip is, but not the other items.) Regards, Albert Noble (England). _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: Radio Shack Speaker Date: Thu, 16 Apr 1998 08:31:26 -0700 Sean-Thomas -- I made a typo. The RS cat. # should have been 40-1349, same as yours. I believe the coil form is aluminum, and I used aluminum solder flux. I also tested the constant while installed in the seismometer, using its displacement detector. I determined the spring constant by knowing the mass and measuring the undamped natural period. Your method with the small weights sounds more direct. I'll try it. I'm not sure where the 7.5 kgauss is measured. It may be between the pole pieces, in the gap -- might be hard to measure. I hadn't thought about the mutual inductance problem. But since the feedback coil is used only to create a magnetic field to force the system back to the null position (and not part of a detector circuit), and the calibration coil is also used to create a magnetic field, I'm not sure if it matters. But my knowledge of magnetics pretty slim. (What the world needs is a simple, intuitive way to understand electromagnetics.) -- Karl _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Seismo-Watch in San Jose! Date: Thu, 16 Apr 1998 09:07:33 -0700 It is with great pleasure to announce that on April 15, the San Jose Mercury News San Jose, California began featuring a weekly Seismo-Watch Newspaper Earthquake Report in the Tuesday edition of their Science and Technology section. The Mercury has chosen to show a basic black and which earthquake epicenter map mostly of the San Francisco Bay Area and Central California, from about Sonoma in the north to San Benito in the south, and the Pacific Ocean on the west to the Sacramento-San Joaquin Valley on the east. Weekly earthquake epicenters registering M1 or stronger are plotted on the map and a table summarizes the various magnitudes recorded. A brief discussion highlights the significant activity. An example of their graphic can be seen at the following URL: http://www.seismo-watch.com/EQNEWS/Newspaper/Newspaper.html The San Jose Mercury News is one of the four largest newspapers in the San Francisco Bay area, servicing several communities in the South Bay Area and central California region, with a potential readership of over 780,000 people. Several active faults cut the region, which not only provide abundant backgroud seismicity but a pose a significant risk to the residents from a major quake. While providing an exciting feature to watch from week to week, it is hoped the Seismo-Watch Earthquake Report will maintain an elevated sense of earthquake awareness and in turn, create a population attuned to effective earthquake hazard mitigation strategies. Since 1992, Seismo-Watch Newspaper Earthquake Reports have gained widespread public, professional and governmental appeal for transferring information from the earthquake monitoring centers to the public via a common, low cost media source. Weekly Seismo-Watch Earthquake Reports currently appear in 32 newspapers from 14 core media syndications within California and Nevada, and have a collective potential readership of over 2.6 million people. For more information on how your local newspaper can begin a Seismo-Watch Earthquake Report, see the Seismo-Watch web page or contact: -- ---/---- Charles P. Watson, Consulting Geologist Advanced Geologic Exploration/Seismo-Watch P.O. Box 18012, Reno, Nevada 89511 Voice: 702-852-0992 / Fax: 702-852-3226 mailto:watson@................ http://www.seismo-watch.com _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Re: A Magnetic Spring Seismometer Description Date: Thu, 16 Apr 1998 12:54:48 -0700 At 07:49 AM 4/16/98 +0100, you wrote: >Hi Larry, > >I wonder if Roger could be persuaded to make a sketch, with the >main relevant parts marked, and place a copy of that, on web? The >photo above is so intrigueing, but I am not sure what I am looking at. >(I know what a yellow croc clip is, but not the other items.) > >Regards, > >Albert Noble (England). Roger sent me an illustration of his magnetic suspension sensor. The 10K gif file (he sent me a bitmap and I converted to gif) can be view at: ftp://drnet.com/users/psn/info/magseis.gif -Larry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Answers to Meredith's questions Date: Thu, 16 Apr 1998 15:02:42 -0500 At 11:58 PM 4/15/98 -0600, you wrote: >Roger Baker, > >With reference to your report of 4/10/98, via email. > >Thanks so much for the JPEG views of you magnetic instrument. >I enjoyed the views, and was amazed at the clarity of the side >view especially. > >I can appreciate the idea of your device as one which probably >alot of people have speculated and dabbled in occasionally, >usually with little sucess, in a variety of designs. Experimentation >is usually a delight to see and do....fun stuff. > >I have to admit I have had to read your progress report several >times, to grasp most of it. > >Anyway...I do have questions and/or speculations (take your >choice), to present... > >1. If...one were to use much larger diameter ferrite magnets and > of course a larger mass, would not the period be lenghtened > somewhat...or would it stay the same? > Yes the period would lengthen, but a far better way to lengthen the period is just to adjust the two magnet heights with the screws. This will lengthen the period up to about a second, but since the beam length is so short, as short as half an inch, this is about the limit for stability -- and with 2 oz of weight this in itself represents a very close balance of gravity with magnetism, meaning it moves a whole lot with a small change in vertical force. It isn't designed to have a long period so much as a high Q, because a feedback force can, in effect, stretch the natural period. >2. The utility knife pivot distance does not look like it could be > lengthened, without collapse...or could it? > Yes, you can make the beam a lot longer and stretch the design in a horizontal direction in effect, and some of my earlier prototypes looked this way. And it may turn out that a long beam is better than a short beam, but I didn't like the air resistance reducing Q. But I might be wrong. But whatever you do, you definitly should have a thermal feedback regulated enclosure, to keep it from drifting off scale, and a magnetic shield, so this argues in favor of a smaller size. If you could figure out a way to very slowly add or subtract larger amounts of magnetic force than I do currently, it might be possible to pull it back into range that way, while retaining the teleseismic period sensitivity. In that case the slow thermal drift could be compensated for, perhaps. >3. Would a Hall Effect sensor be much better, with its much > larger range of motion, than the photocells? > No, force feedback can keep the beam nearly motionless, although I am using very little feedback power right now (but this isn't good, its just cheap and easy for right now). And a good optical sensor is very sensitive and small and cheap. Since you have very little motion on the short beam, you want the best motion detection sensitivity possible and a Hall effect transducer isn't really very good for sub-micron displacements. Its only virtue, so far as I can tell, is that it tends to work along any axis of motion. >4. Could the lead mass, be instead made of a copper coil, as > part of the feedback control and perhaps do away with the > coil underneath? Perhaps this way, the response might be > more linear to disturbances. > I wanted to get the electronics and springy wires all off the off the short beam. In fact the beam is very easily demountable, since the only thing holding it in place is magnetic force. Lead weight on the beam, rather than my copper and epoxy coil, also gives the beam less buoyancy in response to changes in atmospheric pressure. >5. Some varietys of stainless steel....#314 I think....will also > do an assumed variety of repulsion, to a limited extent. With > the mass being part 314 variety and part feedback coil, I > would wonder if the magnetic influence noise would shrink > somewhat, in comparison to the solid magnet? > My magnets should be strong and cheap and easily available. Radio Shack has them all including the rare earth magnets used to seat the knife edge for a few dollars. >I have not seen any magnetic repulsion instrument like yours, >and that I find it interesting. Keep up the experimentation. > Sean-Thomas says that the Sprengnether (sic?) uses a magnetic suspension, so at least some configurations of magnetic suspension are quite practical. I have submitted a black and white line bitmap schematic to Larry Cochrane of my arrangement that should make my setup a little clearer, although I'm not sure yet if he can post it, since the non-optimal way I did it made the file a little big. --Yours, Roger >Thanks, Meredith Lamb > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: More Magnetic Spring Seismometer Questins Date: Thu, 16 Apr 1998 22:38:31 -0600 Roger Baker, Thanks for the effort involved in the JPEG's and the magnificient GIF drawing of your instrument. I get the solid impression from the drawing that you have quite a bit of experience along the line, and it was done relatively fast. I sometimes wonder why I keep asking questions, and don't just hit the books more-ha. (Probably everybody is nodding yes-ha) More questions and speculations...... Going back to question 1 involving the size of the mass. If one can successfully raise the weight of the mass and...keep the present ratio of the pivot (Assuming larger donut magnets), the inertia weight would seem to keep it in one place in space better....and....perhaps the period would be longer, and the Q would seem to stay high. I would think that the magnetic noise would also have a hair less impact in disturbances, due to the higher weight. Perhaps the magnets/weight/gravity balance can be made known initially with a plastic cup and pennys thrown in, and later a lead substitute. I know....I'am throwing out the little magnet stuff-ha. Part of the added mass speculation is directly related to the period of the instrument....a very solid one second period would be much more attractive on the seismic use desireability. I suppose there would obviously be a need for an increase in the number of magnets to do the lift work primarily. The power regulation of the LED and the circuit would probably have to be very tightly controlled I would imagine? Do you intend to do regular or sporatic recording of the output at some time? I would think it would be very interesting to see a variety of results from it. Have you tried or considered using eddy current damping instead of a feed back approach? Anyway....the instruments potential capability of seeing a variety of effects, does have ALOT of appeal in it self. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: More answers Date: Fri, 17 Apr 1998 01:12:56 -0500 At 10:38 PM 4/16/98 -0600, you wrote: >Roger Baker, > >Thanks for the effort involved in the JPEG's and the magnificient >GIF drawing of your instrument. I get the solid impression from the >drawing that you have quite a bit of experience along the line, and >it was done relatively fast. > >I sometimes wonder why I keep asking questions, and don't just >hit the books more-ha. (Probably everybody is nodding yes-ha) > >More questions and speculations...... > >Going back to question 1 involving the size of the mass. If one >can successfully raise the weight of the mass and...keep the present >ratio of the pivot (Assuming larger donut magnets), the inertia weight >would seem to keep it in one place in space better....and....perhaps >the period would be longer, and the Q would seem to stay high. >I would think that the magnetic noise would also have a hair less >impact in disturbances, due to the higher weight. Perhaps the >magnets/weight/gravity balance can be made known initially with >a plastic cup and pennys thrown in, and later a lead substitute. I >know....I'am throwing out the little magnet stuff-ha. Part of the >added mass speculation is directly related to the period of the >instrument....a very solid one second period would be much more >attractive on the seismic use desireability. I suppose there would >obviously be a need for an increase in the number of magnets to >do the lift work primarily. > A longer beam could easily increase the period if the weight were mounted a long way from the pivot, but from my point of view, a long natural period is just not that great in itself, when separated from other tradeoffs. However it is very important if you are designing some other sort of instrument without force feedback like a damped Lehman. >The power regulation of the LED and the circuit would probably >have to be very tightly controlled I would imagine? > If you control the voltage, which is pretty easy to do nowadays, the main source of variation of an LED output is going to be temperature, which is another good reason you need a thermal enclosure. >Do you intend to do regular or sporatic recording of the output >at some time? I would think it would be very interesting to >see a variety of results from it. > I see lots of local high frequency urban noise when I don't filter the output, but thats not very interesting at all. I can make charts, but the current problem is that thermal drift throws the thing off scale long before a teleseismic event comes along. It will take a specially made heat control, sometimes called an oven in the trade, that will maintain the magnetic shield slightly above ambient to kill the drift. My earlier posts talk about how to do that. But all this means work and I'm sometimes lazy or easily distracted toward various other interests. This PSN work don't pay enough! >Have you tried or considered using eddy current damping instead >of a feed back approach? > Feedback control is in itself the best damping approach when well done, IMHO. >Anyway....the instruments potential capability of seeing a variety of >effects, does have ALOT of appeal in it self. > Have you ever heard the expression 'Jack of all trades, but master of none.' ;) But enough of all this. Now that I have gotten out the basics, lets have somebody else play with the approach and improve on it, or discredit it, and let me chime in when I think I have something important to add. --Yours, Roger >Thanks, Meredith Lamb > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Re: A Magnetic Spring Seismometer Description Date: Fri, 17 Apr 1998 08:03:34 +0800 >The power regulation of the LED and the circuit would probably >have to be very tightly controlled I would imagine? >>If you control the voltage, which is pretty easy to do nowadays, the main >>source of variation of an LED output is going to be temperature, which is >>another good reason you need a thermal enclosure. Hi Just 1 1/2 cents worth: If the LED illuminated two fibre optics cables, one illuminated the displacement photo transistor and other illuminated a inline photo transistor. If the outputs were differenced then final result should be independent of the "LED" temperature drift. Do you have any "Winquake" type outputs from your device, say 60 minutes worth? It would be interesting to see the frequency spectrum. Great design. Arie. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Jim Hannon Subject: Re: A Magnetic Spring Seismometer Description Date: Fri, 17 Apr 1998 06:28:21 -0500 Arie Verveer wrote: > Hi > > Just 1 1/2 cents worth: If the LED illuminated two fibre optics cables, > one illuminated the displacement photo transistor and other illuminated > a inline photo transistor. If the outputs were differenced then > final result should be independent of the "LED" temperature drift. > If you carry this one step further and have the boom interrupt both beams of light in such a way that one is being covered while the other is being uncovered (pushpull) you have the potential of reducing several sources of noise and increasing the sensitivity to boot. I bought a couple of photointerrupter modules from DigiKey to do just that. But I think Roger is trying to keep things a simple and inexpensive as posiible and the single phototransistor will work. Since thermal control is needed for lots of other reasons using one beam is not such a problem. -- Jim Hannon http://soli.inav.net/~jmhannon/ 42,11.90N 91,39.26W WB0TXL _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: A Magnetic Spring Seismometer Description Date: Fri, 17 Apr 1998 08:29:11 -0700 A while back I experimented with photo-interrupters for displacement detection. Of the schemes I tried, the best involved using a dual CdS cell with a slot cut in a vane so that the outside half of each of the pair was covered. The sum of the two outputs was used as feedback to the LED to keep its light output constant, and the difference between the two was used as the displacement signal. The best I could do was about 25 nanometers RMS noise in a 5Hz bandwidth. Karl Cunningham La Mesa, CA. PSN station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Re: A Magnetic Spring Seismometer Description Date: Fri, 17 Apr 1998 13:23:43 -0500 At 06:28 AM 4/17/98 -0500, you wrote: >Arie Verveer wrote: > >> Hi >> >> Just 1 1/2 cents worth: If the LED illuminated two fibre optics cables, >> one illuminated the displacement photo transistor and other illuminated >> a inline photo transistor. If the outputs were differenced then >> final result should be independent of the "LED" temperature drift. >> >If you carry this one step further and have the boom interrupt both >beams of light in such a way that one is being covered while the other >is being uncovered (pushpull) you have the potential of reducing several >sources of noise and increasing the sensitivity to boot. I bought a >couple of photointerrupter modules from DigiKey to do just that. But I >think Roger is trying to keep things a simple and inexpensive as >posiible and the single phototransistor will work. Since thermal >control is needed for lots of other reasons using one beam is not such a >problem. >-- >Jim Hannon >http://soli.inav.net/~jmhannon/ >42,11.90N 91,39.26W >WB0TXL > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > Yes indeed, Jim has it right. You have as much sensitivity and thermal stability as you need already, perhaps. The phototransistor is put in series with a 10 K 1% metal film resistor, followed by a voltage follower made from a little 'jellybean' 324 op amp, $1.29 from Radio Shack, which I probably use excessively in my designs. I use two 324's in my simple circuit control and output circuit. I fine tune the position of the beam (the screw adjustments are rough and inconvenient) using a magnet outside the Igloo cooler that encloses the whole thing. If the beam is thus raised and lowered slightly, at some point it will let through just enough light that the resistance of the phototransistor equals 10 K and the voltage output will be be at the midpoint-- where the resistor equals the PT resistance, or three volts using my 6 v supply made from four dry cells. Then this midrange output, which needs attention to keep it there, is amplified by ten before it goes into the A/D converter and into the computer. This buffered output voltage is then passed through a resistor in parallel with a largish cap and then into the coil, which has a DC resistance of about 8-10 ohms for force feedback. But the current electronics is very simple and nowhere near optimized, especially in terms of feedback power. My instrument still rings at its natural frequency, somewhat. But there is a big difference with and without feedback in operation, even at low power. But as I said, I'm more interested in other aspects of design than the electronics and invite anyone to improve on what they see. If you wanted more optical sensitivity, you could use a red laser pointer beam, presumably mounted vertically so as to fit inside the shield with a little mirror to aim the beam horizontally at the tip of the flag. A laser beam is inherently more concentrated than an LED, so that a smaller motion of the flag on the end of the beam can influence more photons going into the phototransistor, by at least factor of at least ten, I would guess. And if you want the ultimate sensitivity, you can call in the heavy artillery, so to speak, with a capacitance micrometer. I would post the results if I had any to brag about, so I will describe the general result at present, rather than sending you a JPEG photo of a chart off my monitor screen. I can amplify the raw output and see a lot of high frequency local man-made noise, or I have been putting a simple one pole low pass filter on the output, but have not seen any teleseismic events that I know of. The output into my Windaq A/D converter looks like a jagged line that gradually drifts off scale due to temperature change. I think I have sometimes seen teleseismic response from ocean waves from the Gulf coast but am not absolutely sure. This looked like a slight but distinguishable sine wave with a period of five or six seconds imposed on the slow drift offscale. --Yours, Roger PS--I just read Karl Cunningham's post. You can imagine that if a clunky old CdS cell will give you 25 nanometers resolution, then a very bright concentrated light source near to a good phototransistor having a very small sensitive area will probably do a lot better. I am well aware of the advantages of optical bridge arrangements, they are not hard to homebrew with two phototransistors, and I like to use them sometimes, but the situation does not require one in this case. Or maybe it does, but I just don't know it yet. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: A Magnetic Spring Seismometer Description Date: Fri, 17 Apr 1998 16:11:16 -0700 Roger -- Actually, the dual CdS cell was an improvement in noise over dual phototransistors by perhaps 3:1. And using a laser pointer was worse than the LED. Although the overall light output of the laser diode (averaging the entire beam) was pretty constant, the distribution of light in the beam's cross section wasn't very stable with time or temperature. Optically diffusing the beam and then columnating it again helped, but it still wasn't as good as the LED. I found that with LEDs, the distribution of light across its beam wasn't very uniform (although it was constant with time), so again diffusing and re-columnating was a help. I'm sure this is quite variable between types of LEDs, and yours may be better in this regard. Roger, I don't want to dissuade you from your work on this ... I easily could have missed a much better approach. I only stopped work on the optical detector when it became apparent that making an LVDT would get my instrument running sooner. -- Karl >PS--I just read Karl Cunningham's post. You can imagine that if a clunky >old CdS cell will give you 25 nanometers resolution, then a very bright >concentrated light source near to a good phototransistor having a very >small sensitive area will probably do a lot better. I am well aware of the >advantages of optical bridge arrangements, they are not hard to homebrew >with two phototransistors, and I like to use them sometimes, but the >situation does not require one in this case. Or maybe it does, but I just >don't know it yet. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Francesco" Subject: EMON PRINT Date: Sat, 18 Apr 1998 03:21:37 +0200 Hi all. I use Quakevew with Emon for immediately analysis of the event before to = run Wq. It's possible to print from Qv ? Francesco Nucera - Italy -
Hi all.
 
I use Quakevew with Emon for = immediately=20 analysis of the event before to run Wq.
It's possible to print from Qv = ?
 
Francesco Nucera   - Italy = -
From: S-T Morrissey Subject: optical displacement detectors Date: Fri, 17 Apr 1998 20:22:45 -0500 (CDT) Roger, I am concerned about your statements about the resolution of a vane or light-beam-interrupting type optical displacement detector. While it does exhibit interesting sensitivity, it is not used in sensitive geophysical sensors because of an inherent limitation: the wavelength of light. One of the rude rules of physics is that one cannot measure a quantity with more accuracy than the ruler that is used. In this case, the red LED has a wavelength of about 600 nanometers (nm), and even though you cannot see them on your detector, the shadow line of the vane is a series of diffraction fringes separated by that amount, all superimposed because of the diffuse nature of the LED source (a LED lasar would show them clearly). The photocell output is an average of this output. Even in the natural world, the limit of our visual resolution is the wavelength of light. This is why we had to invent electron microscopes to see sub-micron objects. The world of micro-positioning detectors uses complex optical encoders and digital interpolation to push the resolution, but the best available is 70 nanometers. (from Dynamics Research Corp, model LB with 64x interpolation). (BTW: I send for lots of literature on such detectors for possible use in strainmeters and tiltmeters ... but not in seismometers) But the resolution we need for seismic instruments is of the order 1 nanometer and less. Right now the sensor here has the normal background noise of microseisms of about 3mv amplitude, which at my current gains (3240v/m/sec) is a ground motion of about 1 micron at 6 seconds. On the power spectral density noise model this is about normal at about -140 db. But if we are interested in someting small and local, like a magnitude 3.0 at 200 km, it will have a peak amplitude of about 100 nm (using Mblg = 3.75 + log(distance in degrees) + log A (microns)). But I can easily resolve 0.1 mv at 10 hz, which is a ground motion of 0.5 nanometers, or about -170 db in power spectra. The smallest event I have recorded is a Mb 2.7 at 265 km, which had a peak motion of 34 nanometers at 3 hz. Early seismic instruments DID use optical detectors, but always employed the principal of optical levers, where the ground motion sensed by the seismometer rotated a small mirror that reflected a light beam over a distance of a meter or more to photographic paper for analog recording or to a differential detector (split photocell) for amplification. Currently optical measurement techniques use incremental encoders or lasars and interferometers, but are still limited by the wavelength of the light used. This is why a the best lasar strainmeters are over 500 meters long. A good discussion of optical displacement detectors can be found in "Strainmeters and Tiltmeters", by D.C.Agnew, (Reviews of Geophysics, Vol 24, No 3, pg579-624, August 1986.) It also has an extensive treatment of capacitive and LVDT-type displacement transducers. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Kees Verbeek Subject: little question Date: Sat, 18 Apr 1998 04:12:34 +0200 (CEST) Hello all, I do not want to interfere in the interesting discussion about the new design from Roger, but there is one point in this design wat's very interesting for me.I'm now for sometime busy with a horizontal seismometer that's small (15 cm)but with a long (20 sec) period. I made my first rough instrument and it seems to work good. it's a re-design from the old Wieger-instrument (I just love the old stuff,can't help it) It's a design that's using the unstable principle , also I use 2 pick-up coils , and the magnet is inside and outside the coil, they are realy great ( took them from a scrap heart-plotter. What i need is a magnetic-shield to cover the instrument and there is my question. The mass and pendulum is "floating" on 2 small magnets and 2 very thin springs, there is no friction. The first rough edition i tested with a small mirror and a laser gave me a good performence it's now next to my "good old" Bosch (still working) and i'm still testing it Now busy with the new design and when i've finished it i'll put a photo and the drawing (cad) on the net. Hope someone can help me with a "simple" solution about the magnet-shield. Greetings, Kees Verbeek , Netherlands _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: shammon1@............. Subject: Re: EMON PRINT Date: Fri, 17 Apr 1998 22:48:39 -0700 Yes you can print from QV. The help text is listed on the banner page. Just enter QV without a file name to display it. In short, in DOS mode enter the GRAPHICS command one time. Then display the file. Press the shift-prtsc (print screen) to print the file. Rgeards, Steve Hammond PSN San Jose Francesco wrote: > > Hi all. > > I use Quakevew with Emon for immediately analysis of the event before > to run Wq. > It's possible to print from Qv ? > > Francesco Nucera - Italy - _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Wieger seismometer/seismograph Date: Sat, 18 Apr 1998 10:40:38 -0600 Kees Verbeek, Can't offer help on the magnetic shielding, but do have a question on the wieger-instrument you are basing your design somewhat on. Is it like the Wiechert pendulum, an inverted pendulum, mass on top, pivot down towards earth, springs on the side? Difference in name or totally different? Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Kees Verbeek Subject: Re: Wieger seismometer/seismograph Date: Sun, 19 Apr 1998 01:45:55 +0200 (CEST) At 10:40 18-04-98 -0600, you wrote: >Kees Verbeek, > >Can't offer help on the magnetic shielding, but do have a question >on the wieger-instrument you are basing your design somewhat on. > >Is it like the Wiechert pendulum, an inverted pendulum, mass on >top, pivot down towards earth, springs on the side? Difference in >name or totally different? > >Thanks, Meredith Lamb > Meredith, Indeed, its the same one. I saw one when i was visiting the"dutch seismo survey"(KNMI) (the "small" one, with a mass of 200 kg( the big one had mass of 1000!!!! kg)) and then they explained it to me , because it is a very complicated but smart seismograph , that was a completely mechanical seismograph and no longer in use . I was there to "investigate" and see the "bosch-seismograph" , the one I build after that visit. The "Wiechert" never did left my mind because I did think there was a very clever basic-idea .I think its the only way (because of the inverted pendulum) to let gravity help the system , in all other horizontal seismographs gravity is against the system. The gigantic weights of the old models were in those days necessarily to overcome the friction between the pen and the paper( and all the other mechanical frictions )That part was also a big problem in the "Bosch-model". Nowadays we have other ways to record so we can do with smaller models, but I wanted to look if it was posible to build a small seismograph with a inverted pendulum , I think its going to work!!. But i also did learn to be very reserved and first look at other ways before yelling that the work is done. So I'm now in the last phase of drawing (after a rough working-model) and after that i can make a start to build for real. Let you know how things are going. Greetings Kees Verbeek , Netherlands _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: A Magnetic Spring Seismometer Description Date: Sat, 18 Apr 1998 22:52:29 -0600 Roger, In regard to the feedback coil ..... With the small donut magnets you are using, it maybe possible to use a small speaker (using just the magnet and coil, not the frame), for the upper magnets replacement, and use the coil in the speaker for the feedback coil. Of course the coil would have to be made to NOT move, and still be in the gap of the original speaker, and pointed down toward the mass. Of course, one could make the coil extended down and into the center hole of the mass also. There would be no attachment to the mass for this. Of course this is an old idea, but it might save winding the coil for some. I suppose one would at least need an 8 ohm or larger speaker, which is size adequate to the task. Of course a larger donut magnet unit could take larger sizes of speaker feed back, and one could arrange various kinds of attached or non-attached coils from speakers....either going thru the holes of regular donut magnets, or permanently set therein. Alot of people have small cheap junk type speakers that might work and accelerate completion. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Re: A Magnetic Spring Seismometer Description Date: Sun, 19 Apr 1998 01:39:43 -0500 At 10:52 PM 4/18/98 -0600, you wrote: >Roger, > >In regard to the feedback coil ..... > >With the small donut magnets you are using, it maybe possible >to use a small speaker (using just the magnet and coil, not the >frame), for the upper magnets replacement, and use the coil >in the speaker for the feedback coil. Of course the coil would >have to be made to NOT move, and still be in the gap of the >original speaker, and pointed down toward the mass. Of course, >one could make the coil extended down and into the center hole >of the mass also. There would be no attachment to the mass for >this. Of course this is an old idea, but it might save winding >the coil for some. I suppose one would at least need an 8 ohm or >larger speaker, which is size adequate to the task. > >Of course a larger donut magnet unit could take larger sizes of >speaker feed back, and one could arrange various kinds of >attached or non-attached coils from speakers....either going >thru the holes of regular donut magnets, or permanently set >therein. Alot of people have small cheap junk type speakers >that might work and accelerate completion. > >Meredith Lamb Meredith, et al, The way I make my coils is to take something like a pencil and wind it up with masking tape until two donut magnets will just barely fit over the increased diameter. I cover the facing sides of two such magnets with masking tape before jamming them on separated by a distance of approximately one magnet width. This makes a sort of bobbin on a stick, and you wind the coil between the two masking tape-faced magnets under a few ounces of tension. You need to secure the wire first of course by winding a few turns on the pencil and tape it over the rim of one of the magnets. The wire I use is the finest of the three spools of magnet wire you get for a few bucks at Radio Shack, measuring .010 inches by my micrometer. But in addition, I wind my coils wet with epoxy (don't use the quick setting kind!) so that when they are wound up to the diameter of the magnets, and the epoxy has set, the whole thing can be warmed up to loosten the adhesive on the masking tape and the bare coil removed. Made thus they have a DC resistence of around ten ohms. This is good for a force coil, but too low for a sensing coil. If you wanted to wind the latter this way, it would be better to use finer wire so as to get a resistance of a few hundred ohms to match a low impedence bipolar op amp like the LTC 1028, I believe. Making your own coils has the advantage of being able to custom fit the other parameters of seismo design. On the interesting topic of sensitive displacement transducers, I poked around in the UT physics library today, searching for review articles. Optical interferometry can sense displacements down to about .1 nanometer, as can linear variable differential transformers or LVDTs. For much more sensitivity, way way down to one hundred thousandth of a nanometer, capacitance micrometers are the way to go. The best overall review article I found was 'Microdisplacement Transducers' by P.H. Sydenham, Journal of Physics E, Scientific Instruments, Vol 5, p721-33, 1972. This has the disadvantage of being an old article, but has some good insights. The best non-interferometric optical method they cite was 20 nanometers sensitivity, but an optical lever system is cited that could do one nanometer. I did find one more recent article; "Real-time, high resolution optical beam position-sensing device' in the Jan. 1991 Review of Scientific Instruments, p 217 that uses a dual gap lead chromate sensor to detect a 543 nanometer laser beam displacement of 5 nanometers, said to be limited by the noise of the laser beam. I think somewhere I have seen references of LED-photodiode or phototransistor displacement transducers that can do in the low nanometers, or several times better than the latter. My current instrument has the center of gravity about three-quarters of an inch from the pivot and the flag extending outwards several inches farther on the beam beyond this so this should mean that any motion is mechanically amplified by three or four times. So I don't really know if I am at the one nanometer sensitivity needed for a good instrument, according to Sean-Thomas, but if not I am not far off. Maybe I can find more recent references that discuss the limits to the sensitivity of my type of transducer. Right now I suspect its photon noise between emitter and detector, which is why you should mount the two as close together as possible and drive the LED hard, or so I think.(Believe it or not, there are now many references to single photon counting with avalanche photodiodes operated at rather high voltage in what is known as the Geiger mode.) --Yours, Roger _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: Saving PSN event file to ASCII file. Date: Sun, 19 Apr 1998 00:30:55 -0700 Greetings, I got an other request from someone that wanted to convert a PSN event file to a text file the other day. Since WinQuake is on a feature freeze, until I get the current version documented and released, I wrote a 16-bit DOS program that will save the PSN event file to an ASCII test file or display the data on the screen. I will add this feature to WinQuake in the next release. The program is called Psn2Text and it can be download at: ftp://psn.quake.net/software/psn2text.zip The "C' source code is include in the zip file. From then readme.txt in the ZIP file: PSN2TEXT Program This 16-bit DOS program converts a PSN Binary event file to an ASCII text file or displays the information on the screen. Usage: psn2text [ /d or /h ] InputFile [ OutputFile ] /d = dump or display data only. /h = dump or display header only. If no output file is specified, output will go to the screen. If neither /d or /h are specified, both header and data will be dumped or displayed. [] = Optional. Files in this package: PSN2TEXT.EXE 16-bit DOS Program file. PSN2TEXT.C C source code. PSNHDR.H Header file for PSN2TEXT.C 980403A.LC8 Example PSN event file. OUT.LC8 Output of PSN2TEXT using 980403A.LC8 as the input. format.txt Documents the PSN binary format. readme.txt This file. -Larry Cochrane Redwood City, PSN _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Walt Williams" Subject: Fwd[1]: A Magnetic Spring Seismometer Description Date: Sun, 19 Apr 1998 00:32:46 +0000 Hello Roger Baker, To what recording device do you connect your Magnetic Spring Seismometer (MSS)? Or is this just an experimental instrument for testing? Is your MSS fairly portable, more rugged? Thank you. Walt Williams, 98.04.19 ================================================= ------- Forwarded Message Follows ------- Date: Sun, 19 Apr 1998 01:39:43 -0500 To: PSN-L Mailing List From: Roger Baker Subject: Re: A Magnetic Spring Seismometer Description Reply-to: PSN-L Mailing List At 10:52 PM 4/18/98 -0600, you wrote: >Roger, > >In regard to the feedback coil ..... > >With the small donut magnets you are using, it maybe possible >to use a small speaker (using just the magnet and coil, not the >frame), for the upper magnets replacement, and use the coil >in the speaker for the feedback coil. Of course the coil would >have to be made to NOT move, and still be in the gap of the >original speaker, and pointed down toward the mass. Of course, >one could make the coil extended down and into the center hole >of the mass also. There would be no attachment to the mass for >this. Of course this is an old idea, but it might save winding >the coil for some. I suppose one would at least need an 8 ohm or >larger speaker, which is size adequate to the task. > >Of course a larger donut magnet unit could take larger sizes of >speaker feed back, and one could arrange various kinds of >attached or non-attached coils from speakers....either going >thru the holes of regular donut magnets, or permanently set >therein. Alot of people have small cheap junk type speakers >that might work and accelerate completion. > >Meredith Lamb Meredith, et al, The way I make my coils is to take something like a pencil and wind it up with masking tape until two donut magnets will just barely fit over the increased diameter. I cover the facing sides of two such magnets with masking tape before jamming them on separated by a distance of approximately one magnet width. This makes a sort of bobbin on a stick, and you wind the coil between the two masking tape-faced magnets under a few ounces of tension. You need to secure the wire first of course by winding a few turns on the pencil and tape it over the rim of one of the magnets. The wire I use is the finest of the three spools of magnet wire you get for a few bucks at Radio Shack, measuring .010 inches by my micrometer. But in addition, I wind my coils wet with epoxy (don't use the quick setting kind!) so that when they are wound up to the diameter of the magnets, and the epoxy has set, the whole thing can be warmed up to loosten the adhesive on the masking tape and the bare coil removed. Made thus they have a DC resistence of around ten ohms. This is good for a force coil, but too low for a sensing coil. If you wanted to wind the latter this way, it would be better to use finer wire so as to get a resistance of a few hundred ohms to match a low impedence bipolar op amp like the LTC 1028, I believe. Making your own coils has the advantage of being able to custom fit the other parameters of seismo design. On the interesting topic of sensitive displacement transducers, I poked around in the UT physics library today, searching for review articles. Optical interferometry can sense displacements down to about .1 nanometer, as can linear variable differential transformers or LVDTs. For much more sensitivity, way way down to one hundred thousandth of a nanometer, capacitance micrometers are the way to go. The best overall review article I found was 'Microdisplacement Transducers' by P.H. Sydenham, Journal of Physics E, Scientific Instruments, Vol 5, p721-33, 1972. This has the disadvantage of being an old article, but has some good insights. The best non-interferometric optical method they cite was 20 nanometers sensitivity, but an optical lever system is cited that could do one nanometer. I did find one more recent article; "Real-time, high resolution optical beam position-sensing device' in the Jan. 1991 Review of Scientific Instruments, p 217 that uses a dual gap lead chromate sensor to detect a 543 nanometer laser beam displacement of 5 nanometers, said to be limited by the noise of the laser beam. I think somewhere I have seen references of LED-photodiode or phototransistor displacement transducers that can do in the low nanometers, or several times better than the latter. My current instrument has the center of gravity about three-quarters of an inch from the pivot and the flag extending outwards several inches farther on the beam beyond this so this should mean that any motion is mechanically amplified by three or four times. So I don't really know if I am at the one nanometer sensitivity needed for a good instrument, according to Sean-Thomas, but if not I am not far off. Maybe I can find more recent references that discuss the limits to the sensitivity of my type of transducer. Right now I suspect its photon noise between emitter and detector, which is why you should mount the two as close together as possible and drive the LED hard, or so I think.(Believe it or not, there are now many references to single photon counting with avalanche photodiodes operated at rather high voltage in what is known as the Geiger mode.) --Yours, Roger _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Web Page update (Western australia) Date: Sun, 19 Apr 1998 20:51:53 +0800 Hi, for those who are interested, the web page at http://www.iinet.com.au/~ajbv/ now contains some information and images of the seismograph vault. It also has a simple page for the last local event and a map for the general reader. More to come. Arie Western Australia Station AU* _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: VBB operation Date: Sun, 19 Apr 1998 08:45:58 -0700 Sean Thomas Well I put my VBB vertical in the garage and was attempting to "zero the mass position". I have been having the problem of the system working for a while(what appears to be reasonable ground noise) then the system output goes flat. It appears that either I am saturating the amplified output or the mass is gradually drifting to one end of it's travel. My question is: is the "mass position zero" close to the zero position of the system without the feedback? The zeroing process is time consuming in that one needs to move the zeroing mass and wait for the system to become stable then repeat the process. I have a highpass filter just after the displacement sensor. The other possible problem is that until I am happy with the circuit I have it on a solderless breadboard. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: zeroing the VBB Date: Sun, 19 Apr 1998 13:16:56 -0500 (CDT) Barry; Re: zeroing the VBB sensor: It sounds like you are running out of voltage for your feedback system. When the mass touches, the feedback starts to fight it and makes a lot of noise. I have +,- 9 volt supplies, and my unit goes "dead" at about +,- 7 volts at either output. Having 15 or 20 volt supplies would help. When I have to do a gross rezeroing of the sensor, I disconnect the signal to the feedback coil (the green clip lead if you can see it in the photo of the electronics box) and look at the meter connected directly to the output of the displacement detector. Then what you see is what you have, and you can adjust your trim masses accordingly. Once it is zeroed, reconnect the feedback coil. I actually have three meters connected: one to the detector output, one to the VBB out, and one to the mass position. As you center or zero the sensor, the mass position will slowly try to catch up with the VBB signal (which, remenber, is also the detector output X10 when the feedback is disconnected). If there is a large difference between these when the feedback is re-connected, the VBB voltage will temporarily take off in the "wrong" direction until the integrator catches up. When I use my remote centering (from upstairs here, near the PC and the monitor recorder), I run the motor until the voltage is within about 0.5 volts of zero, then let the integrator catch up. The remote zero is really handy when we get a cold wind that cools off the basement. You are really being brave in putting your sensor outside in the garage with what is probably a serious temperature problem. Until I can reduce the temperature sensitivity by 10 or 100, I'll live with noise of the basement of the house. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: NO BITMAPS, JPGS, etc Date: Sun, 19 Apr 1998 13:55:37 -0500 (CDT) TO ALL: Some of us are not reading mail in a browser environment!!!! Repeat above PLEASE!!! I read mail in a unix environment, either on a Sparkstation or here at home on the old PC. When someone sends a bitmap or JPEG or whatever, it clogs our mail server, and all I get are endless pages of characters and ciphers. The only way out is to kill the mail process, the dialup, etc, then go back in and clear the hugh file out of my mbox. If you want to make an image available, sent the ftp or internet path to it. Thanks, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: optical interferometers Date: Sun, 19 Apr 1998 13:59:40 -0500 (CDT) Regarding "optical interferometry can sense displacement down to 0.1 nanometer" I don't believe that this is the case for displacement measuring interferometers. There are thin-film interferometers that can for example, measure the thickness of mica sheet layers at 2 nm using blue-green light at 500nm and being able to phase lock to 1/250 of lambda because of the fringe separating techniques. There are also many techniques to sharpen the fringes by manipulating the qualities of the silvered surfaces (see "concepts of Classical Optics, by J. Strong, Freeman press, page, 255 &ff, and appendix A and B, which describe many types of interferometers and applications, including polarizations, but always conclude that the resolution of a distance measuring interferometer is about 0.1 lambda, or 200 nm for green light.), as is evident from what is available commercially. Interferometry involves the interference of two wavetrains of light that have traveled paths of different lengths (one is the fixed or reference path), producing bright or dark lines, fringes, or spots, depending on the design, if the waves interfere constructively or not. Interpolation methods can determine fractions of fringes, with phase lock techniques getting down to a hundredth of a wavelength if sufficient time averaging is done. But averaging kills the frequency response. So the manufacturers who push the resolution (Burleigh, Melles Grott, Dynamics Research, etc.) don't claim better than 100 nanometers repeatability, for reasonable frequency response, even though their micro-positioners (stacked piezo, etc) can easily make repeatable displacements of 0.1 nm. Perhaps the best current practitioners of optical detection methods for geophysical instruments are the people at UCSD at La Jolla, and their instruments at the Pinon Flat Observatory. They use interferometers both for the 750 meter strainmeters, but also for the "optical anchors" that reference the surface piers to consolidated bedrock at depths of 10 to 30 meters. They have just published a paper on using single-mode fiber optics as interferometers for the optical anchors. Regards, Sean-Thomas ... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: debris in magnet Date: Sun, 19 Apr 1998 15:22:21 -0500 (CDT) Barry, I only caught a glimpse of your message about stuff in your magnet because the bitmap (Please don't send!) cleared it all out. I find that masking tape can nicely clean out the magnet gap, as well as the gap in the VRDT, where I have found lint and very fine dog hairs (the dogs always supervise at the seismometer!). On occassion, I haven't even able to see anything, but sweeping the gaps (including the inside if the coil former) with tape has solved an annoying glitching problem. Fortunately, my box enclosure has excluded the spiders, etc, that this old house is fond of. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: VBB operation Date: Sun, 19 Apr 1998 15:38:49 -0500 (CDT) Barry, I just noticed that you said that you have a highpass filter "just after the displacement sensor". I hope you mean that it is between the VBB output and your recording system, and not INSIDE the feedback loop. I cannot imagine that the feedback would be stable if the displacement was high-passed before the feedback components, especially the integrator. However, it sounds like your VBB mass position is reacting properly. It is common, though, to need some amount of low-pass filter after the detector to keep the loop from ringing. Mine audibly "humms" at about 300 hz without a small 50 hz filter. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: VBB operation Date: Sun, 19 Apr 1998 15:54:08 -0700 S-T Morrissey You are correct in both statements. It is before the integrator and it seems unstable (drifts). I will relocate it at the "broadband velocity out position". I thought one would put it before the integrator to eliminate the DC accumulation of the integrator. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: integrator connection Date: Sun, 19 Apr 1998 21:18:45 -0500 (CDT) Barry, You mention a "DC accumulation" at the integrator. I hope that you have configured the integrator as a one-pole low pass non-inverting filter as shown in the (incomplete) block diagram of the VBB system, where the 10x detector output is connected via a large resistor (like 1 or 2 megohms) to the + input, the large non-polarized capacitor (like 2 x 200 uf connected + to +) is connected from there to common, and the unity gain feedback is via a resistor of the same value as the input connected from the output to the - input. (I should update that drawing so it is clearly workable). The integrator output voltage should follow the displacement voltage by the time constant of the integrator. If the displacement is zero, the integrator voltage should go to zero. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: integrator connection Date: Sun, 19 Apr 1998 21:43:45 -0700 S-T Morrissey It's amazing when one gets things hooked up correctly! I plan to heavily insulate the outside of the grounded aluminum enclosure with some R30 rigid insulation,maybe double thick. Also I want to fill the inside air space as much as possible with foam, as Larry has mentioned with his SG . I think that may eliminate some of the low frequency component. The response to ambient vibration looks good. I need to wait for an event now. Thanks again. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Walt Williams" Subject: Soul of a Circuit Date: Mon, 20 Apr 1998 08:27:31 +0000 What is it that makes circuits work? See Below | | V Answer: It's smoke. Because once the smoke comes out and is all gone, the circuit doesn't work anymore. ;) Cheers Walt Williams :-{) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: [Fwd: ALERT: PROTON FLARE AND POLAR CAP ABSORPTION ALERT - 20 APRIL] Date: Mon, 20 Apr 1998 12:46:05 -0700 FYI: Received this alert earlier this morning. Looks to be fairly remarkable. -- ---/---- Charles P. Watson Seismo-Watch From: Roger Baker Subject: On Sensitive Optical Displacement Sensors Date: Mon, 20 Apr 1998 15:36:54 -0500 Friends, In my opinion the topic of the achievable sensitivity of sensitive optical displacement sensors is important. Important because of the fact that if it IS indeed possible to use some some combination of easily available photodetector and LED to achieve nanometer sensitivity, (i.e.--four bucks a pair from Radio Shack) then its liable to be a lot cheaper and simpler to build than competing alternatives. And, accordingl= y, it is likely to be a good sensor for small, cheap, easy-to-build seismometers, where you want to detect a DC displacement -- rather than acceleration, for which you want a coil. Coil/magnet combinations tend to lose sensitivity where very slow teleseismic changes are involved, since not as many magnetic lines of force are intercepted per unit of time so t= he physics tends to favor static displacement detectors, I think. This consideration does not apply to differential transformers used as detectors, however, since they are sensitive to static displacements. But the fact remains that capacitance micrometers although somewhat troubleso= me to build, can be hacked by amateurs, and nothing comes close to them in terms of practically achievable sensitivity if one wants to push the limi= ts of displacement detection. By far the best general reference for the latter, including an illustrati= on of the quite accidental detection of an earthquake by such a capactive sensor, is given in the article 'The design and some applications of sensitive capacitance micrometers' by R.V. Jones, Journal of Pysics E, Scientific Instruments, vol. 6, p 589-600. Also I described a hackable version built out of Radio Shack parts in a recent issue of the Society f= or Amateur Scientists Bulletin. Since I love physics and hacking instruments, let me present my analysis = of the situation as gleaned from the literature. Somewhere, there is a discussion of this very topic in the optical engineering literature, but = I have not found it yet. By far the best one source on the leading edge of optical technology I know of are the hundreds of volumes of SPIE, or the International Society for Optical Engineering, each on leading edge topic= s. There are two general alternatives to use when measuring optical displacement. One is interferometry, and the other is what we might call photon counting. I use the latter. Where interferometry is concerned, it is apparent that it is practical to measure down even to .01 nanometer, but the apparatus gets quite expensiv= e and always uses lasers, etc. (see =91Interferometric metrology: current trends and future prospects=92; SPIE, vol. 816, p 2-18, 1987). In other words, this would take a lab full of stuff that would negate the characte= r of hackable amateur science. But microdisplacement sensing by photon counting is a whole =91nother ball game. In this case you have a flood of fairly easily detectable photons totalin= g up to 16 milliwatts of energy, either red or near infrared photons, being emitted by an LED out of an area of less than one square millimeter.=20 This crosses a short gap of several millimeters chopped by a horizontal knife edge mounted on the seismo beam, which decends across the beam like= a horizontal guillotine blade. The detection area of the sensor is a phototransistor with about a one mm surface area, essentially a photodiod= e with a transistor output built in. If the detector is sensitive enough t= o detect one millionth part of the photons being emitted by the LED, it implies that you could choke off all but a millionth of the photons crossing the gap defined by the seismo beam and still detect what manages to get across the gap defined by three horizontal knife edges attached to the LED, beam, and phototransistor, respectively.=20 If you can detect one millionth of the photons crossing the gap, then the one square millimeter area of the LED divided by a million is a beam of light one millimeter across and one nanometer tall, which fits our requirements to be able to detect a displacement of one nanometer.=20 If the knife edge is at the top end of its travel it will only shave off = a millionth part of the top edge of the light beam passing across the gap a= nd we will never see it. However if the knife edge is instead intersecting t= he last millionth of the photons able to pass across the gap, it will result in a detectable output. I think that one can detect one millionth of the output, and the proof is that if one places the phototransistor four or five feet away from the L= ED in a dark room, even with the focusing lens filed off, they can still detect each other even though the photon flux between them is now reduced by a factor of millions. Now as Sean-Thomas has pointed out, light wavelength is a consideration, because light does not really travel in a straight line but rather will tend to diffract around the knife edge defined by the seismo beam. In oth= er words, if everything is set up properly and scattered light is almost totally eliminated, the phototransistor signal as the last of the light i= s blocked off should be a series of diffraction maxima and minima diffracti= ng around the beam=92s knife edge with a spacing determined by the wavelengt= h of light being used. This does not mean, however, that we cannot detect such displacement far smaller than a wavelength of light, if only enough photo= ns are involved to give a detectable signal in the phototransistor. The most hacker-friendly article I have yet found on this topic is titled =91A simple optical transducer for the measurement of small vibration amplitudes=92 in Measurement Science, 1985, p 947 (I think; my photocopie= s do not give the exact origin of the reference). This describes a simple dual photodiode setup sensitive to tens of nanometers displacement, but nowher= e near optimised in regard to trying to maximize displacement sensitivity i= n consideration of the factors discussed above.=20 For those trying to hack my sensor configuration, may I also suggest putting a tiny square of masking tape over the sensitive areas of the LED and phototransistor after cutting away most of the plastic case as per my illustration. Then spray them all over with black spray paint paint to se= al out almost of the stray light before peeling away the tape to make a litt= le window over the sensitive areas and then mounting them behind their knife edges. The latter edges should define and block off the lower 50% of the beam that passes between them. Very probably any work spent getting the three knife edges accurately lined up parallel to each other will be rewarded as added displacement sensitivity. =20 --Yours, Roger _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "David A. Nelson" Subject: Re: PROTON FLARE AND PCA ALERT - 20 APRIL Date: Tue, 21 Apr 1998 09:30:13 +1200 At 12:46 PM 4/20/98 -0700, you wrote: >FYI: Received this alert earlier this morning. >Looks to be fairly remarkable. Charles P. Watson SATELLITE PROTON EVENT ALERT > ISSUED: 16:00 UTC, 20 APRIL > ** EVENT CRITERIA HAS BEEN MET ** >ATTENTION: > > A text-book perfect long-duration class M1.49 proton flare was observed >with a maximum x-ray time of near 10:21 UTC. The event appears to have unfortunate that it waited till the active region had rotated out of view before this event happened. othewise there would have been a change of good auroral activity. An M1.5 flare is quite small and as we move closer towards solar maximum in 2000 these events will become more regular and stronger. On the othere hand there is a large Coronal hole region crossing the face of the sun at present and over the next few days there is a good possibility of auroral activity from it. So people hight than 45 deg latitude (nth or sth) should keep an eye on their nthrn/southern skies for aurora. Aurora produced by coronal holes tend to be less intense than those from flares but they are usually more persistant..... often lasting 3 + days. Cheers Dave Co-ordinator: New Zealand Public Seismic Network Dave A. Nelson ZL4TBN 24 Jensen St., Green Is., Dunedin, South Is.. New Zealand. http://psn.quake.net/dave/home.htm IF there indeed are other parallel universes... I can now rest in the knowledge, that in at least one of them, I am filthy rich and drive a red Ferrari _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: Soul of a Circuit Date: Mon, 20 Apr 1998 14:25:12 -0800 That's 'spirimentin' sayeth T.Edision 10,000th test of light bulb. keep on oh spark trician ! Walt Williams wrote: > What is it that makes circuits work? > > Answer: > > It's smoke. > Walt Williams :-{) > Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: [Fwd: ALERT: PROTON FLARE AND POLAR CAP ABSORPTION ALERT - 20 APRIL] Date: Mon, 20 Apr 1998 14:32:45 -0800 Holy sun screen! Should we get SPF 1000 or what, charles? 8 ) ^ sunglasses > SATELLITE PROTON EVENT ALERT > > ISSUED: 16:00 UTC, 20 APRIL > > /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: integrator connection Date: Mon, 20 Apr 1998 22:24:22 -0600 barry lotz wrote: > S-T Morrissey > It's amazing when one gets things hooked up correctly! I plan to > heavily insulate the outside of the grounded aluminum enclosure with > some R30 rigid insulation,maybe double thick. > Barry, good deal on the STM-8 going online. Question regardingthe grounding you mentioned. Real copper rod grounding, or circuit ground? Electrostatic prevention? Sounds like a good idea with real earth grounding. Don't remember anyone doing this. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: grounding Date: Tue, 21 Apr 1998 00:00:19 -0500 (CDT) Barry, Meredith, Good idea to be concerned about proper grounding and shielding. Grounding can never be taken for granted. Here at the farm, since the seismometer is sort of distributed from the far basement corner to the PC here to the monitor recorder over by the door, I use the cold water pipe as protective ground. The electrical common of the VBB seis electronics is floating and not grounded, but the case is grounded to the aluminum frame of the seis and to the copper pipe, as is the green ground wire in the AC to DC power supply. as is the chassis of the monitor recorder. I have 5 pairs of shielded data cable running up from the basement; I ground the shields only, not the common of each pair, which are in fact all connected to the electronic common in the electronics box. One wants to make protective grounds one-way streets to drain AC noise, static, etc. Avoid closing the loop anywhere, or currents flowing in the ground loop will induce voltage noises in the data. I consider the copper pipe to be the lowest impedance ground available here. The ground of the house AC electrical system is less desirable. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: optical displacement detectors Date: Tue, 21 Apr 1998 00:22:19 -0500 (CDT) Re: optical displacement detectors: I am concerned about recent statements about the resolution of optical displacement detectors that could be used for seismic instruments. I am concerned that those who are trying to build a working homemade seismograph are being misled into thinking that there is a cheap and simple optical method for measuring sub-micron displacements (better than ~10^-7 meter), or even down to sub-nanometer (to 10^-10 meter) displacements. The more serious concern is the implication that ALL of science and engineering (and seismic instrumentation specialists) have been blissfully ignorant or oblivious to this possibility for the past 50 years. Obviously, if we could be doing it (even at any expense), we would be doing it. And the major suppliers of micro-measurement and manipulating tools and hardware would be selling such devices. And, for example, why would we be struggling with 750 meter evacuated tube lasar strainmeters tracking 1/8 lambda interferometers (resolving 1/8 wave of a ~600nm lasar or 75 nm) so we can accurately measure 10^-10 earth strain? If interferometers could resolve 0.01 nanometer, as has just been claimed, our strainmeter could be a few meters long. I find that the claims being made for optical displacement detectors are speculative and not supported by the fundamentals of physics and/or any engineering data. And curiously, the claimed resolution for interferometric measurements improves daily. On 4/16 it was a reasonable "sub-micron" (eg. 10^-7 m), but on 4/18 it jumped to 0.1 nanometer (10^-10 m), and today it is " practical to measure down even to 0.01 nanometer", or 10^-11 meter, which is half of the first Bohr atomic electron-orbit radius of 5.29 x 10^-11m, or ~4 times the Compton wavelength of an electron (2.426 x 10^-12 m), which is the serious limit in electron microscopes (just to put the numbers into perspective). (I also have a serious problem with the curious notion of a "beam of light"... "one nanometer tall" when light propagates as a quantum waveform uniform in ALL directions, and the wavelength is 400 to 700 nanometers, and, btw, photons themselves are not particles.) I urge restraint on the part of anyone considering a plan to utilize such devices for seismic instrumentation. As observed above, if we could, we would, but we cannot, and it is not because we are ignorant and unaware. And if these claims could be supported by actual data, certainly an attempt would have been made. There is a wide range of proven displacement measuring technology for nanometer resolution at reasonable response times (to 1 khz) via reactive and/or resonant technologies (capacitive, LVDT, VRDT) that the home-made seismologist (who, btw, would probably prefer not to be considered an "amateur hacker") can readily apply to a broadband sensor. And Jim H. has pointed out a single IC , the Analog Devices AD 690, that will totally run such transducers (if +,- 15 volts is available) for $27, from Newark, although its noise floor has not been evaluated ... yet. The noise of capacitive transducers (at 1 hz bandwidth) is about 10^-10m, with the best reported as 10^-12 (1 picometer) at diurnal periods. (see D.C Agnew's paper) One of the best LVDTs (used in the Streckeisen STS-1) has a noise floor of 6 x 10^-11 m. The VRDT is not as quiet, but has the advantage of minimal force being applied to the moving element, great ease of assembly and adjustment, and general immunity to capacitance change noise due do lead movement. The VRDT can resolve the minimal earth noise model of 10^-9 m at 1 hz. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Re: optical displacement detectors Date: Mon, 20 Apr 1998 22:36:41 -0700 Excellent discussion Sean! -- ---/---- Charles P. Watson Seismo-Watch S-T Morrissey wrote: > > Re: optical displacement detectors: > > I am concerned about recent statements about the resolution of > optical displacement detectors that could be used for seismic > instruments. --snip-- _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Seisguy Subject: NBC Dateline-Earthquakes Date: Tue, 21 Apr 1998 02:06:43 EDT For those of you interested, Tuesday, April 21 10:00pm PST NBC Dateline (Channel 4 in L.A.) Subject: Earthquakes, with Dr. Lucy Jones Mike O'Bleness Northridge, CA _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Seisguy Subject: Earthquake felt over much of NZ's North Island Date: Tue, 21 Apr 1998 02:10:42 EDT Earthquake felt over much of NZ's North Island WELLINGTON, April 21 (Reuters) - A deep earthquake measuring 6.3 on the Richter scale rocked much of New Zealand's North Island on Tuesday but little damage was reported. The Institute of Geological and Nuclear Sciences said the quake, which occurred at 11.34 a.m. (2334 GMT), was centred about 33 km (20 miles) southwest of Taumarunui, in the central North Island. The shaking, lasting about a minute, was enough to make the city clock in Hastings, 260 km (161 miles) away, chime out of sequence. The quake, which was 245 km (152 miles) deep, was felt as far south as Christchurch in the South Island, and as far north as Hawke's Bay 360 km (225 miles) north of Wellington. The Earthquake Commission said it had received 12 claims for damage, mostly for such things as cracked walls. Some Wellington residents could not contact relatives following the quake because phone lines became overloaded. REUTERS 01:16 04-21-98 _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Re: optical displacement detectors Date: Tue, 21 Apr 1998 05:09:53 -0500 At 12:22 AM 4/21/98 -0500, you wrote: >Re: optical displacement detectors: > >I am concerned about recent statements about the resolution of >optical displacement detectors that could be used for seismic >instruments. > >I am concerned that those who are trying to build a working homemade >seismograph are being misled into thinking that there is a cheap >and simple optical method for measuring sub-micron displacements >(better than ~10^-7 meter),... ************************************************************************** Response: See yesterday's post which described how to detect displacements down to tens of nanometers with very simple equipment which wasn't particularly optimised for detection sensitivity, titled 'A simple optical transducer for the measurement of small vibration amplitudes'; Measurement Science and Technology, p 947, 1985. ************************************************************************** ....or even down to sub-nanometer (to 10^-10 >meter) displacements. The more serious concern is the implication >that ALL of science and engineering (and seismic instrumentation >specialists) have been blissfully ignorant or oblivious to this >possibility for the past 50 years. > >Obviously, if we could be doing it (even at any expense), we would >be doing it. And the major suppliers of micro-measurement and >manipulating tools and hardware would be selling such devices. >And, for example, why would we be struggling with 750 meter >evacuated tube lasar strainmeters tracking 1/8 lambda interferometers >(resolving 1/8 wave of a ~600nm lasar or 75 nm) so we can accurately >measure 10^-10 earth strain? If interferometers could resolve 0.01 >nanometer, as has just been claimed, our strainmeter could be a few >meters long. > ***************************************************************************** Response: There is a world of difference between fringe counting at 750 meters, which is now possible under stringent conditions, and measuring short distances optically. Obviously the optical techniques one uses to count laser fringes at such a very long distance are completely different than those one uses to measure sub-nanometer displacements on the top of lab benches, which would be more suited, in theory at least, to seismometers where the total displacements are only millimeters or less. To quote quite specifically from 'Interferometric metrology: current trends and future prospects' by P. Hariharan, SPIE, vol 816; Intereferometric Metrology,1987; p 2. 'One of the beams is focused to a small spot (~2 micron diameter), while the other illuminates an area with a diameter of about 50 microns. With this arrangement a useful measurement sensitivity of .01 nm has been achieved' (from page 10). The primary reference for this is 'Optical hererodyne profilometer' by C-C Huang, Optical Engineering, 23 (4), 365-370, (1984). Yesterday I had not found the reference above (which claims .01 nm) but was basing my statement on another article entitled 'A complete high performence heterodyne interferometer displacement transducer for microactuator control'; Review of Scientific Instruments, vol.63, Jan 1992, which says in the abstract 'This article describes a laser heterodyne interferometer-based transducer which resolves displacements to .12 nm (standard deviation) locally, in a scanning range of 10 mm or more, in a 1 Hz to 10 kHz bandwidth under normal laboratory conditions.' Another paper, entitled 'Recent advances in displacement measuring interferometry' by N. Bobroff, Measurement Science and Technology , vol. 4, 1993, p 907 -926 affirms .1 nm in its introduction. But as I made clear, I don't think these setups are appropriate for homemade seismometers. I do think my current sensor is very good, although the sensitivity has not been determined. I do believe it is appropriate to analyze what aspects of physics probably limit the ultimate sensitivity of sensors like mine. ***************************************************************************** >I find that the claims being made for optical displacement detectors >are speculative and not supported by the fundamentals of physics >and/or any engineering data. And curiously, the claimed resolution >for interferometric measurements improves daily. On 4/16 it was a >reasonable "sub-micron" (eg. 10^-7 m), but on 4/18 it jumped to 0.1 >nanometer (10^-10 m), and today it is " practical to measure down >even to 0.01 nanometer", or 10^-11 meter, which is half of the first >Bohr atomic electron-orbit radius of 5.29 x 10^-11m, or ~4 times the >Compton wavelength of an electron (2.426 x 10^-12 m), which is the >serious limit in electron microscopes (just to put the numbers >into perspective). > >(I also have a serious problem with the curious notion of a "beam >of light"... "one nanometer tall" when light propagates as a >quantum waveform uniform in ALL directions, and the wavelength is 400 >to 700 nanometers, and, btw, photons themselves are not particles.) ***************************************************************************** Response: Photons share some of the properties of both particles and waves. They can be counted as individual units by both cooled photomultipliers and cooled avalanche diodes, however, if one wants to go to the trouble, and uncooled avalanche diodes would be a good choice if one wants to push the limits beyond what a photodiode will do. I DID go on to say that if one tries to block a light beam with a knife edge, the shadow will be a diffraction pattern with maxima and minima, rather than a sharp edge, which is quite according to optical theory. The real question to ask here is probably how the function of light falling on the sensitive area changes with displacement of the optical sensor, and what is the ultimate limiting factor in terms of instrument noise. The paper I cited yesterday (Review of Scientific Instruments, Jan 1991, p 217) with a resolution of 5 nm, used a dual lead chromate photoresistor, which I suspect does not have a very good quantum yield. The limiting factor in resolution under these conditions was said to the laser noise. In other words, the resolution was about one hundredth of a wavelength of the light used, but the light was very bright and the sensor likely not very sensitive. I presume most junction LEDs are fairly quiet compared to lasers, and photodiodes, are very sensitive indeed to red and near infrared light. So the photodiode (inside the phototransistor chip) sees a series of diffraction patterns as the knife edge on the beam intercepts the beam. The limiting sensitivity in this case would appear to me to be the accuracy of light measurement or photon counting at low light levels when the beam is very narrow, almost completely cut off, and exhibiting diffraction at the knife edge. If not, then can somebody suggest what would be the ultimate limiting factor in terms of the physics involved? We do know that the limit is at least as small as a twentieth of a wave from the article I cited yesterday without using very optimal conditions, and from what Karl Cunninham posted the other day (25 nm). We also know that scanning atomic force microscopes routinely sense such small displacements on the order of one nm and probably less using laser light reflected from cantilever probes. I'll try to find out the exact sensitivity of this arrangement and post it later *************************************************************************** > >I urge restraint on the part of anyone considering a plan to utilize >such devices for seismic instrumentation. As observed above, if >we could, we would, but we cannot, and it is not because we are >ignorant and unaware. And if these claims could be supported by >actual data, certainly an attempt would have been made. > ***************************************************************************** In the spirit of open discussion of technology which is what I love about this list, I don't think you should discourage others from trying out some elements of my design. If so, why? We won't really know until some others try it and see for themselves, right?. Its really easy to build for about $20, not counting the proposed permalloy shield and thermal enclosure. Admittedly, it does drift in its present incarnation, but I have made no claims I can't support. I can send JPEGS of the output trace, which is now a slanted line with lots of seismic noise. Maybe I'll be lucky enough to catch a quake. Response: Aside from slow thermal drift, (this despite being presently enclosed with a large block of iron in an Igloo cooler), and the fact that I must presume also magnetic noise in the absence of a model that fits into a permalloy shield, my instrument seems seems to work well, or I wouldn't have posted what I did. The feedback circuit is not optimised with a power booster, so that it rings a little. It is sensitive enough to see high frequency local seismic disturbances, and slower sine wave disturbances on the order of five seconds as the trace slowly drifts off the chart due to thermal drift. I can send both the current circuit I am using, which is very easy to build, or the thermal feedback circuit I built for an earlier prototype, and which may be of general interest to others wishing to build thermally regulated enclosures, or ovens maintained slightly above abient. **************************************************************************** >There is a wide range of proven displacement measuring technology >for nanometer resolution at reasonable response times (to 1 khz) >via reactive and/or resonant technologies (capacitive, LVDT, VRDT) >that the home-made seismologist (who, btw, would probably prefer not >to be considered an "amateur hacker")... ***************************************************************************** Response: I write a regular column titled Science Hacker, for the Society for Amateur Scientists Bulletin, and nobody has complained about my calling myself an amateur hacker up to now. The editor, Jack Herron, is a member of this list. **************************************************************************** *...can readily apply to a broadband >sensor. And Jim H. has pointed out a single IC , the Analog Devices >AD 690, that will totally run such transducers (if +,- 15 volts >is available) for $27, from Newark, although its noise floor has >not been evaluated ... yet. > >The noise of capacitive transducers (at 1 hz bandwidth) is about >10^-10m, with the best reported as 10^-12 (1 picometer) at diurnal >periods. (see D.C Agnew's paper) One of the best LVDTs (used in >the Streckeisen STS-1) has a noise floor of 6 x 10^-11 m. The VRDT >is not as quiet, but has the advantage of minimal force being applied >to the moving element, great ease of assembly and adjustment, and >general immunity to capacitance change noise due do lead movement. >The VRDT can resolve the minimal earth noise model of 10^-9 m at 1 hz. ***************************************************************************** Response: The article I cited yesterday by R.V. Jones on sensitive capacitance micrometers estimates the short term sensitivity of their capacitance micrometer at 10 to the minus eleventh mm, or ten to the minus fourteenth meters, in their abstract. But it is true that over the period of a day, drift reduces the accuracy to 10 to the minus twelfth mm, in close accord with your information. This is far less than the diameter of a hydrogen atom (see your comments above) and is only meaningful because this small displacement is averaged over a substantial area. Meanwhile, there is what seems to me to be a good analytical comparison of capacitiative and inductive displacement sensors by A.L. Hugill in the Journal of Physics E, Scientific Instruments, vol. 15, 596-607, 1982. The conclusion: 'It is concluded that capacititive transducers are most suitable in applications requiring high levels of accuracy, stability, and discrimination and low power dissipative and excitation forces. In situations where these requirements are not so stringent, both inductive and capacitative transducers are suitable'. ***************************************************************************** > > >Regards, >Sean-Thomas > ***************************************************************************** To sum up: I'm not without scientic credentials. I have had articles published in various peer-reviewed journals as the lead author including the Journal of Applied Physics, the Review of Scientific Instruments, the Journal of Chemical Education, and had THREE articles in Scientific American under the legendary old Amateur Scientist columnist C.L. Stong (on ultrathin layer chromatography, homemade transistors, and an infrared imaging device). Most importantly, I do not make up facts. I ask only a fair trial of my seismometer elements and do not claim to have it fully developed. I presented it quite specifically in the form of a progess report. It is remarkably easy to duplicate in its essentials, and it seems to me that only empirical evidence is a fair basis for passing judgement. --Yours, Roger Baker _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "James M Hannon" Subject: Re: grounding Date: Tue, 21 Apr 1998 10:20:37 -0500 I want to second the comment about not having ground loops. In communication facilities I have worked in there is only one earth ground for the entire building. All sorts of precautions are taken to avoid other paths to ground including things like placing the equipment racks on insulating platforms off the concrete floor. A final test of the building ground system is to open the lead to the single ground and measure the resistance to ground of the building ground system. I have seen this resistance as high as 100000 ohms for an entire building full of communications equipment including connections to hundreds of underground phone lines and data cables. If the resistance is too low we had to go through the building until we found the problem and fixed it. Not only will having multiple gounds cause noise problems. It can increase the chances of lightning damage. I am having this sort of problem at home right now. So far two modems and a surge protector have been destroyed by lightning strikes nearby even though the phone line is underground all the way back to the office. There are separate grounds for the phone and power in my house and I suspect that one of the grounds is inadequate. This causes voltages induced in one system to go to the ground of the other system through my modem or surge suppressor. It is possible that if you bury your sesimograph or ground it too well that all you will do is provide a better path to ground for lightning, right thru you equipment! Jim Hannon _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: optical displacement detectors Date: Tue, 21 Apr 1998 08:50:22 -0700 Just a note about the AD690. We evaluated it at work for a commercial LVDT conditioner product, and its noise floor was about 2 millivolts peak out of 10 volts peak full scale. It appeared to be an output noise effect, independent of input levels, etc. One could increase the gain of the system to get desired resolution, but this noise represents a real signal-to-noise ratio limitation. In constructing my horizontal force-balance, I built an LVDT demodulator using a handfull of IC's that has an SNR of better than 2x10^5 and a noise floor of about 0.3 nanometers RMS, and less in reduced bandwidths. I agree that variable differential reactance transducers are the way to go. I'm experimenting with a differential capacitor transducer that has a noise level of about 1 nanometer peak-to-peak, and I have quite a few ideas to improve it still. This differential capacitor device is easier to build than the LVDT, and seems to have a bit less noise, and the electronics are simpler. Karl Cunningham La Mesa, CA. PSN station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: cas Subject: Paranormal/Magnetic/Earthquakes - Off Topic Date: Tue, 21 Apr 1998 08:57:14 -0700 (PDT) -----BEGIN PGP SIGNED MESSAGE----- In order to conserve list bandwidth I have placed an article that appeared in our local paper at http://www.ncal.verio.com/~cas/ It is somewhat 'off-topic', but similar threads have appeared on this list and might be of interest to some of you. -----BEGIN PGP SIGNATURE----- Version: 2.6.3a Charset: noconv iQCVAwUBNTzB7dE8bAS3aqZxAQEekwP+MzUKNR3lOlCgF51t8d9lY19uxW3qNZX6 meywOD0xbHXzvswHGAFt09HNgxsQKu1LI1T4pDzmUgcrpMGJLjRY7NlDYb2oTA8V 3dB1wwqujw7mvpcFylv1sWrlq2ECa+KE0UEXyqcFv/QsRqKrqsAH2a5BMlbuCj/d x1rahbfDdbY= =QGLT -----END PGP SIGNATURE----- _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Charles R. Patton" Subject: Re: optical displacement detectors Date: Tue, 21 Apr 1998 10:10:37 -0700 For at least one more data point. As Roger said, there are two ways to look at the measurement system. As a fringe counting scheme and as a beam occultation "weighing" system. HP makes systems of both types. Currently they have an interesting one capable of 0.3nm resolution you can look at at: http://www.tmo.hp.com/tmo/datasheets/English/HPE1720A.html The method employed in the system is that a grating scale is illuminated with a laser diode, the diffraction pattern that forms in space several millimeters away is directly imaged on a multiple photocell array that is part of a custom IC with a pitch equal to the diffraction pitch. Patents I have looked at indicate there are three groups of cells -- somewhat like paralleling every third spoke of a picket fence. Then electronic processing derives the position. It may actually be two sets as the output I observed was quadrature. These things are real, and we're using them to replace a bunch of the standard HP interferometers of the fringe counting variety because these have more resolution in a smaller space. Charles R. Patton _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Analog Devices 690...AD690 Date: Tue, 21 Apr 1998 11:21:36 -0600 Minor note, keep seeing reference to the AD690, but have checked twice over time and am not finding it. Curiosity thing. Analog Devices does not give any reference to the AD 690 on its website. Perhaps too new? Not found in Newark catalog per that number. AD598AP is listed at $28.75 in 98 catalog. The price always goes up-ha. There is however 3 LVDT signal conditioners called the AD598AD, a 20 pin CerDip, a AD598JR, a 20-pin SOIC, and a AD698AP, a 28 lead PLCC & CerDip packages, all on page 803 on the bottom. Analog Devices does have data on the AD598 and AD698 on its web site. Presume someday it will show up on internet Analog Devices site. What is the difference between the AD598/AD698 and AD690, being as I have no reference....major, minor? Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "James M Hannon" Subject: Scientific American The New Backyard Seismology, April 1996 Date: Tue, 21 Apr 1998 14:55:05 -0500 Larry Cochrane's comments to the SAS forum on the ADXL05 sensor SA article finally got posted to the NEW SAS forum below is the author of the SA article's response. See http://psn.quake.net/adx05chip.txt and http://web2.thesphere.com/SAS/ The response was posted to the SAS forum. Shawn Carlson - 05:49am Apr 20, 1998 PDT (#3 of 3) I agree with everything Larry says about the sensitivity of the device described this article. Unfortunately, I didn't have the space to place the new sensor in perspective. (God, how I hate the 1200 word-limit!) The article was intended to provide an inexpensive entry-level instrument into the field of seismology, like a relatively inexpensive telescope provides an avenue for amateurs to get their feet wet in astronomy, without the full potential of a top-of-the-line instrument. These detectors are inexpensive, they can be monitored directly by any home PC, and they allow the amateur to sense three axies of motion simultainsly. And, they are, in my view, an important part of the next generation of amateur seismology. The accelerometer chips, which as Larry correctly points out, were first developed as crash sensors, are now being used in many other areas. As a result, market pressures (including for commercial seismographs) are driving developers to create ever more sensitive sensors. In fact, in the time since the article first appeared the sensitivity of the chip used in this project has increased by more than a factor of 50, which reduces the possible detection threshold brings us from mag 5.0 to mag 3.5 or so. This increase in sensitivity was achieved by lowering the noise floor. Other accelerometer chips now boost even greater sensitivity, and the next generation of these silicon marvels will bring even more subtle motions into easy reach of amateurs. Will they ever compete with the most sensitive devices used by amateurs today? I think that they will, and soon. In fact, I suspect that within the next five years, many instruments on the PSN (a WONDERFUL resource for amateurs, by the way) will be based on silicon-chip accelerometers. Shawn Carlson _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: AD698, not AD690 typo Date: Tue, 21 Apr 1998 17:49:37 -0500 (CDT) Meredith, MY TYPO (again): it is a AD698, not a 690. Sorry for the confusion. Did you see Karl C.s review of it (he inadvertently passed on my typo there too: right device, wrong number)? Unfortunately I don't have the literature here at home (some viral flu or cold has laid me low for the last few days). I recall the review from AD comparing the 598 with the 698 says that the demodulation in the 598 is by rectification, so it cannot handle phase errors like VRDTs produce; the 698 uses synchronous demodulation that reduces phase errors. The circuit I use for tiltmeters is also synchronously demodulated with the 4066; you can notice the two amps that drive the switch as "hard" as possible: these are designed to minimize any phase error in the switching. Karl reports a noise level of the AD698 of 2mv with a 10 volt range, which is only 74db, and we need 120db. I guess the chip has too many compromises built in. I don't think Validyne has adopted it for their VRDT pressure transducers. I have bought quantities their osc/demod/amp module for use with the Schaevitz 005MHR LVDT to make 100mv/millibar microbarometers. It also works with the transformer coil based VRDT at 50mv/micron (at a 2mm gap). For the tiltmeter osc/demod/amp (that I use for the seis VRDT,) the tiltmeter noise was less than 0.5 nanoradian at 25mv/microradian. When I use it with the VRDT at 100mv/micron, the ~10 microvolt noise level is 0.1 nanometers. The dynamic range (with a 7 volt maximum) is 117 db. A so barring further evaluation, it looks like the AD698 won't suit our needs. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: grounding Date: Tue, 21 Apr 1998 17:58:53 -0500 (CDT) Jim, You make a good point (which I should have mentioned) about burying the seismometer so that it is the ultimate earth ground for lightning. You can imagine the problem with a seismic telemetry station, with an antenna 100 ft in the air on the highest hill in the county. Often we would use a cheap natural tower: a suitable tree with a 20 ft. pipe in the top to push up the yagi antenna; (Cypress trees are the best, because the limbs grow like ladder rungs. ) Even with the "best" of lightning suppression, we have had some real smokers. At a station near Paducah the ligntning exploded every device in the radio transmitter, and got onto the shield of the seismometer cable and down to the buried L4-C, which was sealed in a 1/4" ABS pipe with the electronics (amp/vco/DC-DC/calibrator) in a 2.5" aluminum tube above it. It blew a hole in the ABS pipe, a hole in the case of the L4, and a big melt/burn in the side of the L4 mass. Curiously, it "went around" the electronics cannister, which was still operational, probably because the rest of the lightning suppression DID work, where there are probably 20 or so Varistors at work in conjunction with delay inductances and gas discharge devices. You also mention the central grounding for comm. facilities. One of my first tasks here in 1969 was to try to salvage a strainmeter project in a lead mine south of here. We had 100 ft fused quartz rods, (actually 2" dia tubes) for the reference lengths, and capacitive displacement transducers for continuous data, and for calibration we had 3-color (Fabry-Perot Mercury line) interferometers that took a photo of the fringe pattern every hour. (In the pre-Nova3 computer days, interpreting these kept several undergrads busy). The signal processing and recording electronics was at the surface in the lift house, about 1/2 mile from the underground sensors. The problem was that we could only get noisy data in the four hour period BETWEEN mine shifts. During the shifts the data was useless. I finally investigated and found that the electric railroad in the mine was leaking about 100 amperes at 300 volts, and the current would peg at 500 amps when a train moved. I could actually see the voltage gradient (400 ft down in the mine where the instruments were installed) with clip leads over several meters with the old Simpson 260 meter. So the mining company ran a continuous INSULATED ground cable (about 1/2" dia) from the surface, down the shaft, through the drifts; and I connected it to copper bars in an array of porous ceramic pots with copper sulfate in them. At the surface, we did as you did: put everything, even the electronics racks, on 3/8" plexiglass insulators, and referenced all grounding to the new cable. Then it looked like our data was "dead", so we turned up the gains and saw the earth-tides, free oscillations, etc, that we were seeking. At the array of porous pots, I had to keep refreshing the copper sulfate, because the ground currents were plating out the copper onto the bars. SO even the installation of the home seismograph needs to be done carefully with respect to grounding and lightning protection. If lightning comes in on the PC modem's phone line, and tried to find ground at a buried seismometer, the main path is through the PC and the ADC card. Insulating the seis from the earth is probably an effective approach, with a ground cable of about #6 size (possibly with a green jacket to mark it as ground) back to wherever the "master ground" is located. And in general, grounding cables MUST be insulated, so that they do not result in multiple ground paths. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Jim Hannon Subject: Re: AD698, not AD690 typo Date: Tue, 21 Apr 1998 18:15:39 -0500 S-T Morrissey wrote: > The circuit I use for tiltmeters is also synchronously > demodulated with the 4066; you can notice the two amps that > drive the switch as "hard" as possible: these are designed to > minimize any phase error in the switching. > > Regards, > Sean-Thomas > I am just about ready to publish my design for a VRDT drive circuit. It features a small number of inexpensive parts, crystal controlled frequency and digitally adjustable phase of the demodulator drive. I am hoping that the phase adjustment will be usefull in "tuning" the circuit for best performance. -- Jim Hannon http://soli.inav.net/~jmhannon/ 42,11.90N 91,39.26W WB0TXL _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: VBB zeroing Date: Tue, 21 Apr 1998 19:30:03 -0700 Sean Thomas Well I'm having problems with the zeroing of the system. The mass position output wants to slowly drift off to the rail. I can zero the instrumentation amp(displacement output) when the integrator etc is not connected. I can (I think) zero the system without the feedback coil in the loop. But I have trouble with the system all connected. BTW the feedback has the correct sign. I tried to connect it the other way and the beam went more quickly to the rail. The system acts like it's in unstable equilibrium. When I am adjusting the mass should I wait more than an hour for the integrator to stabilize? If I set the mass output to slightly positive the voltage will appear stable, but overnight or during work (about 8 hr period) it will drift off. With the output slightly negative it will slowly drift that way. I am wondering if the integrator cap is charging without a discharge path. My instrumentation amp goes thru a single pole highpass filter directly into the integrator. Should there be a load resistor in there? I assumed the IA output was a low impedence path to ground.??? Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Grounds & VRDT Date: Tue, 21 Apr 1998 22:08:38 -0600 Sean-Thomas, Jim Hannon, Karl Cunningham and everyone interested, OK on the vrdt chip number correction....it did, seem like a form of COLD-con-FUSION ,ha. Interesting that there doesn't seem to be any equivalent chip to replace the number used by yourselves. Actually all the talk on the grounding stuff, will be useful to myself and a bunch of people, not really acquainted with all the aspects of it. I'am really draggggging out my seismo/computer preparation stuff, so the ground aspect data will really be of value now. My house was build in 1951, so half of it doesn't have electrical grounds, and when I got around to the room with the computer and other stuff; it really nosedived the glitchs occurance by maybe 99%. The other 1%, was a dying flickering light bulb, making a brief surge till shut off, then of course no more. Very good recent grounding email talk stuff useful to everyone with seismology interest! Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: VBB zeroing Date: Wed, 22 Apr 1998 18:26:10 -0500 (CDT) Barry, I am confused again about your use of a high pass filter, which you say is now at the input of the integrator?? If this is the case, the integrator is not getting the displacement information that it needs, so I would expect that the seis would be unstable. There are NO high pass filters used WITHIN the VBB feedback configuration. The ONLY place one can be used is between the VBB output and whatever you have for digitizing and recording, and the only reason it is necessary is to remove the near DC signals (thermal thru tides) from your (ours too) limited dynamic range recorder or 12-bit ADC, (unless one happens to have a 24-bit ADC that can handle the dynamic range.) You need to make sure your circuit conforms exactly to both of the block diagrams on the web site. Some of the values used are 2 meg for the integrator input resistor to the + amplifier input and the unity gain feedback from the output to the - input. The integrating capacitor is two 200uf polarized low leakage (ie. not TAGs) connected + to + from the integrator + input to common. (unless you have some quality non-polarized capacitors). But if I assume you have it connected correctly, the other problem that hasn't been solved is thermal drift. The units here will drift off scale with about a 2 degC temperature change. The TIME response of the integrator at the mass position voltage output is no longer than the integrator time constant. A drift of an hour to a day is probably thermal. When I zero the seis here with the manual trim weights, I do it quickly, looking at the displacement output with the feedback disconnected, then I get the cover/insulation back in place. Even body-heat radiation will effect the spring, which relaxes as it warms. That is why the motor-controlled zeroing is so handy. I have plans A thru at least P to try to reduce the thermal drift, but haven't tried them. Regards, Sean-THomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: VBB zero revisited Date: Wed, 22 Apr 1998 16:33:10 -0700 To All I was having problems with my first VBB vertical sensor zero. Previously STM had mentioned thermal drift as a concern. At the time I thought that was an interesting "second order" effect. Well..... As Sean Thomas mentioned on his recent posting , it isn't! I thought I would go back to basic principles and check the the magnitude of the drift. The Bouyancy force on an object is the weight of an equal volume of the surrounding media. If we assume a 1 lb aluminum sensor mass @ 168.5 pcf (excuse my use of english units). The volume would be 0.0059347 cf. Air @ 70 degrees F = 7.5E-2 #/cf Air @ 80 degrees F = 7.35E-2 #/cf Therefore the weight change of 0.0059347 cf of air between 70 and 80 degrees F would be 9E-4 lbs or 0.41 grams. If my math is correct, this is a significant weight change to the system. My sensor is indicating this. This morning the mass position reading was +3.47 v (higher position at the coolest part of the day). After work the sensor was reading -1.47 vts(lower position at warmest time). It's time to install the insulation and zeroing motor. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: X-ray interferometer Date: Thu, 23 Apr 1998 11:21:10 -0400 Hi gang, Here's a footnote to the optical interferometer displacement transducer= discussion. Some year's ago (20?), Dick Delattes at NBS wanted to improve the accuracy of the wavelength of copper k-alpha X-rays (1.540562... Angstoms= ). He built a simultaneous optical and x-ray interferometer (using a dislocation-free Si crystal) so that the no. of x-ray fringes could be determined in terms of the optical fringes. As I remember, he used a Lamb-dip stabilized He-Ne laser for the light. The wavelength of this is= known with high accuracy. Building an x-ray interferometer which can be moved in steps small enou= gh to count fringes while maintaining fringe contrast is one hell of a trick--but it worked. I saw this thing--it was in a room in a room in th= e basement (for vibration isolation). All this work was published but I no= longer have the reference. It greatly improved the accuracy of the X-ray= wavelength. Since the x-ray wavelength is about 4,000 times smaller that the optica= l wavelength, an X-ray interferometer makes a sensitive displacement transducer. I once set up an optical interferometer and I shudder at the= difficulites of making a device which is 4,000 times more fussy. I hesitate to recommend this scheme for a seismometer. Bob Barns Mind like a steel trap-rusty and illegal in 37 states. = _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: Sperry tiltmeter Date: Thu, 23 Apr 1998 11:21:07 -0400 Hi gang, I recently had an interesting time exploring a tiltmeter. My friend Bi= ll Scolnik has this tiltmeter and he found a fascinating paper "Use of a bubble tiltmeter as a horizontal seismometer", by Miller, Geller & Stein,= Geophys. J.R. astr. Soc., vol 54 1978 pp 661-668. This describes a Rockwell Intl. Part No. 67380-301 biaxial tiltmeter and its performance a= s a seismometer. The same 'quake recorded with a broad-band seismometer and the tiltmete= r showed essentially perfect correlation--"wiggle for wiggle"--on a teleseismic event. Correlation with a short period seis was also excellant. Amazing stuff from a small device. The transducer in this device is two resistors in a bridge circuit and = a Parr lock-in amplifier was used to amplify the signals. Bill's tiltmeter was obvoiusly not the same in that the output was a DC= voltage. (He managed to get a print of the pin-outs of the connector fro= m Sperry but no other info.) There was no access to a bridge circuit. Bill's device:Sperry Flight Systems Part # 00283, Sensing element Model= B. It has an associated box of electronics and a read-out which is Part = # 02020-00-01. The tiltmeter is a short (maybe 2 1/2" hi) cylinder painted= white about 5" dia. with a hemispherical dome. This is to a 1/2" thick Al base plate and I saw no hope of getting inside. Attached t= o the dome is a fancy, military quality multi-pin connector. There are connections for + and - 6VDC and a ground and signal wire. This is clear= ly a single axis device since there is only one output. = I built a 20" long tilting table with a micrometer head (smallest div. = is 10 microns, I don't have a differential micrometer) as one leg. Mounted = on this, the tiltmeter seemed to output either +3 or -3 V and anything betwe= en was nearly impossible. It seemed to be bi-stable. In desperation, I mounted the tiltmeter upside-down on the table and it worked fine!! Well= , sometimes wild notions pay off. I ran a calibration curve of output voltage vs. tilt (calculated from micrometer settings). The device was powered with two 6V drycell batteri= es and the output voltage was read by a Radio Shack DMM (which has an RS-232= port). Between -2.5 and 0.6V out (angle range of about 4 arc mins), the plot was nearly linear and a least-squares line gave mins of tilt =3D= (voltage + 4.807)/0.7793 = This is 2,700V/radian, a figure I find astonishing. The least count on= the DMM is 0.1mV which is 0.02 arc secs.! ( 1 micro-radian =3D .21 arc secs.) This tilt is what you would get if you placed a 0.001" shim under= one end of a beam 850 feet long. It looks like the time constant is abou= t 1 min. (to reach 67% of the final value). The tiltmeter was placed on the basement floor near my Lehman and connected thru Larry's amp. to the second channel of his A/D and both wer= e recorded with SDR. The signal from each sensor was passed thru a sectio= n of the Rockland low-pass filter set at 0.08Hz. I got a 'quake from Oaxaco Mexico on both the Lehman and the tiltmeter= =2E = The SDR files will be posted to the archives:980203C.RB1 is the Lehman an= d 980203c.RB2 is the tiltmeter. The records are similar but the noise level of the tiltmeter is poorer than the Lehman. P was not vis. in either and S was relatively stronger = in the tiltmeter record. I don't know what caused (maybe foot traffic) the monster spikes before and after the 'quake but the tiltmeter responded mu= ch more strongly . This performance was not as good as that described in th= e Miller, et. al., paper but it is a different device and with a much less sophisticated electronics. The very informative review paper by DC Agnew, "Strainmeters & = Tiltmeters", Rev. of Geophys., vol 24, no. 3, Aug. 1986, pp 579-624 contains a world of information including a detailed discussion of bubble= tiltmeters. (A paper by Morrissey is cited.) Agnew advises not to try = to build your own bubble meter and also sez that tilt-feedback systems work.= = A sensitivity as high as 38,000V/radian is cited so perhaps the figure above is not completely ridiculous. Agnew suggests that bubble position could be sensed with external plates and a capacitive bridge which makes building a bubble much easier and avoids all electrolysis problems. A quick look in Thomas' Register showed that there are dozens of manufacturers of bubble tubes for levels. All this raises some questions: 1. Why was Bill's device built to work upside-down? My guess is that it= was designed for monitoring tilt in buildings or bridges or something. 2. Agnew sez that bubble tiltmeters have been widely used so there must = be some in dead storage. How can us poor people acquire these? 3. Is this the sort of device that the PEPP is working to distribute as= an "electrolytic" sensor? = 4. Why are bubble tiltmeters so much rarer than horizontal pendulum sensors? Bob Barns = PS: I was playing around with a 4.5Hz geophone. I connected it to an amplifier made from an Analog Devices AMP01 instrumentation amplifier wit= h the circuit on a proto board . It seemed to have much more low freq. noi= se than I thought it should. I remembered an admonition in the AD data sheets about avoiding thermoelectric effects when using hi gains so I simply draped a Kleenex over the circuit and the noise went way down. = Draft-free amplifiers is a new concept to me. PPS:OK, so what's the speed of dark? = = = _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Mammoth Times & Seismo-Watch Date: Thu, 23 Apr 1998 09:29:55 -0700 Guys, I am just too wigglie to contain myself. I just received a message from the Mammoth Times editor who has accepted my new and upgraded earthquake report for their newspaper. Check it out here: http://www.seismo-watch.com/EQNEWS/Newspaper/NewspaperGraphics/MTSeismo.New.jpg Oh, golly, this is going to be fun! -- ---/---- Charles P. Watson Seismo-Watch _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: Sperry tiltmeter Date: Thu, 23 Apr 1998 13:40:48 -0700 Bob -- I wonder if an optical or capacitance position sensor could be adapted to a machinist's bubble level, and if it would be sensitive enough. These are available for (as I remember) the <$50 range from the cheapie machinists supply houses (ie. Enco, etc.). As far as the spikes at the ends of the trace.. Sometimes digital filtering can introduce artifacts at the beginning and end of a trace that's been filtered. Is this a possibility? Karl Cunningham La Mesa, CA. PSN Station #40 karlc@....... At 11:21 AM 4/23/98 -0400, you wrote: > The tiltmeter was placed on the basement floor near my Lehman and >connected thru Larry's amp. to the second channel of his A/D and both were >recorded with SDR. The signal from each sensor was passed thru a section >of the Rockland low-pass filter set at 0.08Hz. > I got a 'quake from Oaxaco Mexico on both the Lehman and the tiltmeter. >The SDR files will be posted to the archives:980203C.RB1 is the Lehman and >980203c.RB2 is the tiltmeter. > > The records are similar but the noise level of the tiltmeter is poorer >than the Lehman. P was not vis. in either and S was relatively stronger in >the tiltmeter record. I don't know what caused (maybe foot traffic) the >monster spikes before and after the 'quake but the tiltmeter responded much >more strongly . This performance was not as good as that described in the >Miller, et. al., paper but it is a different device and with a much less >sophisticated electronics. > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: tiltmeters:Sperry, Rockwell,etc Date: Thu, 23 Apr 1998 14:11:07 -0500 (CDT) Bob, THe tiltmeter you initially describe (the Rockwell/Autonetics) is the one that the USGS tried to deploy widely in the mid '70s for earthquake prediction research. It is the one used by Stein et al. as a horizontal seismometer. I was given 10 of these to use at New Madrid, and 10 for the Aleutian project, and a few etcs. The New Madrid project (they just don't work in the alluvium) was moved to So. California to the "Palmdale Bulge" in the western Mojave, and some were moved to the Crustal Deformation Observatory of UCSD at Pinon Flat for comparison with the long-baseline fluid tiltmeters. I guess I have the worlds stock of them here, about 24 sensors. Long ago we gave up on "improving" the Kinemetrics/Rockwell electronics and built our own to reduce the thermal sensitivity by a factor of 100. We have a standard output of 25mv/microradian (25 000 Volts per radian) with a noise level of 0.5 nanoradian. In operation, we use the earth-tide data at 0.1 microradian p-p to verify our calibration; we have programs that can model the tilt tide. THe most stable instrument we ever operated drifted only 0.32 microradian per year, but 1 to 2 urad/year was more typical. The final status of the borehole tiltmeter project is that the instrument lives up to expectations, but the surface of the earth is too unstable, mostly from meteorological influences. This is at the 3 meter depth that we were installing the sensor, which has no mechanical zeroing so has to be installed by hand to an initial level of one microradian. To solve the surface noise problem, we had developed a method to install the sensor in 30m boreholes with a detachable motor-driven pre-leveling device, but the funding ran out when everyone thought that GPS would answer all our crustal deformation questions, which, naturally, it hasn't. UCSD is still operating long-baseline fluid tiltmeters with "optical anchors" to reference the surface piers to bedrock at depths of 10 to 30 meters. THe "final" paper about borehole tiltmeters is: "Shallow Borehole Tilt: A Reprise"; by F.K. Wyatt, S-T. Morrissey, and D.C. Agnew; Journal of Geophysical Research, Vol 93, No B8, Pgs 9197,9201; August, 1988. The current interest in the electrolytic bubble tiltmeter is for correcting the aiming of large telescopes in the period between tens of seconds and an hour, in order to keep galactic images centered on the ends of optical fibers that lead to spectrometers. The natural earth-tides and wind blowing on the telescope, as well as thermal gradients in the support structure are the problem. There are several sources of electrolytic tiltmeters available, but only the Rockwell bubble has the stability and resolution needed for this application.. Regarding your Sperry tiltmeter: I have one here; it has a white ABS end cap cemented to an aluminum plate. It is designed to be mounted with the plate vertical, as on the wall of a tunnel inside a dam. I believe it has a "Frederics" bubble inside, which can produce an output when it is sideways, depending on how it is mounted. Somewhere I have all the literature about it. It was not useful for crustal deformation studies. Your other questions: 2: The Rockwell sensor has been found in surplus stores: they were originally used for the platform "pre-aim" for ICBMs. It is useless without the bridge circuit and the electronics; if you apply an ohm meter to the electrodes, they plate off into the electrolyte and the sensor is immediately ruined; they must be run only with an AC bridge. 3: It is not the sensor used in the PMD seismometer that is being considered/used for the PEPP project. That electrolytic sensor works on an entirely different (proprietary) principle. 4: Why aren't they used for horizontal seismometers: only the Rockwell sensor has a low enough noise level for teleseismic sensitivity. Stiction of the bubble and electrode contact surface irregularities are the main noise in other bubble sensors. And the last time a price was put on the Rockwell sensor, it was g.t. $5000 per, and they aren't made anymore. For a compact horizontal sensor, a miniature "garden gate" design (eg: 5cm boom, 200gm mass, To = 1 sec.) can be made with a VRDT displacement sensor and VBB feedback that, while not meeting the MTQ noise expectation (M*T*Q > 1), can resolve the mid-range of the earth noise models (at PSD of l.t. -150db at 30 seconds). Unfortunately, it is also a very good tiltmeter, and requires a well prepared site. This brings up the next problem: tiltmeters that are NOT seismometers. An interesting converse to the problem is that the current seismic broadband horizontal sensors ARE very sensitive to tilt noise, which even on piers situated on bedrock limits their usefulness. We are evaluating a tiltmeter that is NOT sensitive to horizontal acceleration to try to remove the tilt noise. It needs to be MUCH quieter than the seismometer, which is the challenge. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Larry Cochrane Subject: The S in SG sensor... its a small world.... Date: Thu, 23 Apr 1998 17:54:28 -0700 Hi Everyone, Got this email from Barry Shackleford today. He's the S in the SG sensor name...Portola Valley is only a few miles from me. More info about the SG sensor can be found at http://psn.quake.net/sgsensor.html. -Larry >Date: Thu, 23 Apr 1998 17:37:50 +0000 >From: Shackleford >Reply-To: pecora@.......... >X-Mailer: Mozilla 3.01-C-MACOS8 (Macintosh; I; PPC) >To: cochrane@.............. >CC: pecora@.......... >Subject: S-G seismometer > >Dear Larry, > >I must say that I was surprised and delighted to see that the >seismometer that Jim Gundersen and I put together back in the >'70s still lives on through your excellent implementation and >seismic home page. > >I noticed that you are living in Redwood City. I'm close by in >Portola Valley, working at HP-Labs. As the Japanese say, "It's >a narrow world." > >We should chat sometime. > >Best regards, > >Barry > >Barry Shackleford >Hewlett-Packard Labs, 3U >1501 Page Mill Road >Palo Alto, CA 94304 > >email: b.shackleford@............ >tel work: 650-857-7537 >tel home: 650-234-1208 > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Geophone Help Date: Fri, 24 Apr 1998 05:23:42 +0800 Hi, I'm planning to set up a small station at home to complement my two SG seismometers some, 11 kilometers away. I hope to do some epicenter locating on the larger local events. The problem is that my house is located in a suburb and subject to the typical manmade induced, ground noises. I hope to keep the complexity as low possible and had intended to use a vertical geophone as the sensor. To that end can anyone recommend a good quality 4.5hz geophone? Is a 1 Hz geophone, available? My long term plan is to upgrade my primary station with a sensitive vertical seismometer. Thanks for your time. Arie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: John Hernlund Subject: Re: Geophone Help Date: Thu, 23 Apr 1998 23:15:50 -0700 (MST) Arie, From personal experience I have learned that vertical data is not good in local events. The reason vertical data is bad for local events is because the waves are travelling to your seismometer at a low angle to the surface, and because displacement in P-arrivals is in the same direction as the velocity they will move things from side to side a lot more strongly than up and down, making horizontal sensors more effective for first arrivals. Also, background noise, especially in suburban areas will have a high vertical component, thus masking almost entirely the P-arrival, or at least making it extremely difficult to nail down. Even so, 11 km separation will produce nearly simultaneous arrival times due to the fast speead at which waves travel. The combination of these two problems will defeat your purpose of epicenter locations... ****************************************************************************** John Hernlund E-mail: hernlund@....... WWW: http://www.mc.maricopa.edu/academic/phy_sci/Geology/hernlund/ http://www.public.asu.edu/~hernlund/ ****************************************************************************** _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Re: Geophone Help Date: Fri, 24 Apr 1998 20:46:12 +0800 Hi John Thanks for explaining the problem. Your advice sure makes sense and I'll redirect my efforts. Its a real problem in identifying the real local events against noise and mine blasting (150km) away. One of our seismic active areas is also about 150 km away. Has anyone had any success with identifying the quakes direction with a X,Y,Z axis seismometers? Arie > From personal experience I have learned that vertical data is not good > in local events. The reason vertical data is bad for local events is > because the waves are travelling to your seismometer at a low angle to the > surface, and because displacement in P-arrivals is in the same direction > as the velocity they will move things from side to side a lot more > strongly than up and down, making horizontal sensors more effective for > first arrivals. Also, background noise, especially in suburban areas will > have a high vertical component, thus masking almost entirely the > P-arrival, or at least making it extremely difficult to nail down. Even > so, 11 km separation will produce nearly simultaneous arrival times due to > the fast speead at which waves travel. The combination of these two > problems will defeat your purpose of epicenter locations... > > > John Hernlund _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Abreviations Date: Fri, 24 Apr 1998 13:41:29 +0100 S-T Morrissey wrote: > per year, but 1 to 2 urad/year was more typical. > > > 3: It is not the sensor used in the PMD seismometer that is > being considered/used for the PEPP project. That electrolytic > sensor works on an entirely different (proprietary) principle. > > > MTQ noise expectation (M*T*Q > 1), can resolve the mid-range of > the earth noise models (at PSD of l.t. -150db at 30 seconds). > Unfortunately, it is also a very good tiltmeter, and requires > a well prepared site. Hi Sean and OTHERS, I wonder if you would mind if I mention the subject of abbreviations. Many of us on PSN (some Americans TOO no doubt ), are sometimes no conversant with all the abbreviations that many use on PSN from time to time. I hope that I speak for many folk, who may struggle to follow the gist of a discussion, in which their understanding is further hampered by the use of abbreviations not understood by them, and who do not wish to ask. (I may be speaking for the 'silent majority' out on PSN.) I do not wish to point at your letter as the only example, but merely use it as an illustration of what I mean. For example, I confess that I don't know what "M*T*Q" is, nor can I fathom "PSD of l.t.", (altho' I DO know what "db" is .) "urad", also puzzles me. Maybe I should know what "PEPP" is (already refered to ?), but this eludes me as well........ Aren't I an ignorant cuss! . At times maths are used, and again, I am not sure what the multiplying/division/squares/roots etc. signs used in ASCII mean....perhaps there are standards for these, if so, maybe you/someone can mention a source of these so that I/we may be able to check. No offence is meant Sean or others . I thoroughly enjoy all this discussion on PSN, but it would be nicer if sometimes folk use FULL terms instead of abbreviations, which they may assume all 270 odd PSNers understand, and which many may not. Hope no one minds me bringing this up. TTFN !!! ("Tat Ta for now" ) Regards. Albert Noble (England). _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Michael S. Healy" Subject: Re: Mammoth Times & Seismo-Watch Date: Fri, 24 Apr 1998 09:27:39 -0800 congratulations ! Charles Watson, your wiggling may set up anomalies. -- Michael S.Healy internet: sunmanh@............ 1-503-355-3177 (voice) internet: wave3@........... 1-503-355-3367 (fax) (below RIP-aware as well) 1-503-355-8738 (bbs SysOp) 8N1 FULL DOS terminal program _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: Abreviations Date: Fri, 24 Apr 1998 11:40:47 -0500 (CDT) Albert, A very good point about abbreviations. I sometimes worry about being too "wordy" in spelling things out, but if an abbreviation is gibberish, the communication fails anyhow. re your specific questions: 1. M*T*Q = Mass(in kg) times the period (in seconds) times the Quality factor, which is often considered the inverse of the mechanical damping, but a value of 1 can generally be assumed. 2. g.t.; l.t., etc: = greater than, less than (from old Fortran) 3. urad = microradian 4. PSD = Power Spectral Density: which is a universal scale used to measure or compare of the seismic signal levels from any seismometer system. It is scaled in db = decibels, which, for power levels, is a logrithmic scale of 20 times the log of the signal level. 5. PEPP = Princeton Earth Physics Project, an Education and Outreach effort to get more earth science education in schools, and includes a program to place live seismic stations in high schools. Various "inexpensive" seismometers with digitizers are being considered. As I mentioned above, your point is well taken. I hope this answers some of your questions. Regards, Sean-THomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: Abreviations Date: Fri, 24 Apr 1998 11:40:47 -0500 (CDT) Albert, A very good point about abbreviations. I sometimes worry about being too "wordy" in spelling things out, but if an abbreviation is gibberish, the communication fails anyhow. re your specific questions: 1. M*T*Q = Mass(in kg) times the period (in seconds) times the Quality factor, which is often considered the inverse of the mechanical damping, but a value of 1 can generally be assumed. 2. g.t.; l.t., etc: = greater than, less than (from old Fortran) 3. urad = microradian 4. PSD = Power Spectral Density: which is a universal scale used to measure or compare of the seismic signal levels from any seismometer system. It is scaled in db = decibels, which, for power levels, is a logrithmic scale of 20 times the log of the signal level. 5. PEPP = Princeton Earth Physics Project, an Education and Outreach effort to get more earth science education in schools, and includes a program to place live seismic stations in high schools. Various "inexpensive" seismometers with digitizers are being considered. As I mentioned above, your point is well taken. I hope this answers some of your questions. Regards, Sean-THomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: John Hernlund Subject: Re: Geophone Help Date: Fri, 24 Apr 1998 10:19:30 -0700 (MST) On Fri, 24 Apr 1998, Arie Verveer wrote: > Thanks for explaining the problem. Your advice sure makes sense and > I'll redirect my efforts. Its a real problem in identifying the real local > events against noise and mine blasting (150km) away. One of our seismic > active areas is also about 150 km away. Has anyone had any success > with identifying the quakes direction with a X,Y,Z axis seismometers? I haven't had much luck with just a single station. Since the waves could be arriving from any direction it is difficult to look for some of the characteristic polarization of the waves that would indicate fault nucleation and slip, although a close study of the seismograms is probably likely to produce some idea this would probably require a high-resolution seismograph... There are books out there on some methods for doing this kind of analysis, I remember seeing one written by Bruce Bolt on methods for telling a nuclear explosion apart from an earthquake; maybe that would be useful... ****************************************************************************** John Hernlund E-mail: hernlund@....... WWW: http://www.mc.maricopa.edu/academic/phy_sci/Geology/hernlund/ http://www.public.asu.edu/~hernlund/ ****************************************************************************** _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Quake in Nevada Date: Fri, 24 Apr 1998 10:28:23 -0700 Got several calls from a M4.4 jolt near Hawthore this morning. -- ---/---- Charles P. Watson Seismo-Watch _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "James M Hannon" Subject: Re: Abreviations Date: Fri, 24 Apr 1998 13:32:18 -0500 Sean, Your description of PSD seems to be missing something. Unless it is used differently in seismology. Since it is a density it needs to be "per" something such as Hertz. Normally dB is a ratio so I would expect this to be dBm or with respect to some other signal level. This would make your -150 dB become -150 dBm per Hertz. Taking 20 log base 10 of a signal level in volts does not give you anything that you can use to compare with another instrument. Jim Hannon sean@........... on 04/24/98 11:40:47 AM Please respond to psn-l@............. To: psn-l@.............. psn-l@.............. cc: (bcc: James M Hannon/CedarRapids/Collins/Rockwell) Subject: Re: Abreviations 4. PSD = Power Spectral Density: which is a universal scale used to measure or compare of the seismic signal levels from any seismometer system. It is scaled in db = decibels, which, for power levels, is a logrithmic scale of 20 times the log of the signal level. Regards, Sean-THomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: Re: Sperry tiltmeter Date: Fri, 24 Apr 1998 20:18:58 -0400 Karl, = My remark about finding many makers of bubble levels was intended to inspire someone to experiment with capacitance pick-ups. The Rockland filter used here is an analog filter which should not introduce spikes. Bob Barns = _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Robert L Barns <75612.2635@..............> Subject: Tiltmeters, reply to ST Morrissey Date: Fri, 24 Apr 1998 20:18:56 -0400 Sean-Thomas, Lots of interesting info in your epistle on tiltmeters, thanks. In your reference to the Sperry tiltmeter:apparently I still don't have= it right--it should be used with the base plate vertical??? I can't imagine what a "Frederics" bubble can be and work (at least reasonably well) when mounted 90 deg. from the proper way. I note a mention of the Frederics in Agnew's paper but no description of how it works. Could you describe this more thoroughly or suggest where I can find the= detalis? Also, if I send a SASE, could you send me a reprint of the Wyat= t, Morrissey, Agnew paper? (And maybe a copy of the literature on the Sperr= y if you can find it.) Bob Barns I love defenceless animals, especially in a good gravy. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: PSD definition Date: Fri, 24 Apr 1998 20:06:39 -0500 (CDT) Jim, There I went again: abbreviating a description for the sake of being brief, and causing more confusion, and mis-stating the fact that power ratios are 10 x the log, not 20 x, as for voltage ratios. Regarding PSD (Power Spectral Density): The standard units used for instrumental/earth noise PSD are in DB = 10 * log [(meters ** 2 / second **4) / hz ] A PSD plot is actually a modified plot of the FFT (fourier Transform) of the data sample, where each term of the FFT of the velocity output voltage is divided by the data calibration constant (volts/meter/sec or counts/meter/second). Then for every term of the FFT, this number is squared and then divided by the corresponding frequency in hz. Then the log of each term is multiplied by 10 and plotted on a log-linear scale. For example, if the value of the term for 10 seconds in the FFT is equal to 0.01 millivolts (a representative noise level of the instrument in my basement), and the instrument constant is 3280 volts/meter/second (the Beta version of the STM8), then 0.00001 volts divided by 3280 equals 3.05 x 10^-9; this squared is 9.3 x 10^-18; it is then divided by 0.1 hz, and the log of that value is -16.03, and 10 x that is -160.3 db. (db with respect to 1 meter**2/second**4/hz). This is the value of the PSD before averaging over a 5-point running mean. This approximate estimate can be seen in figure 4a of the .....Data... page on my web site. As you can see, the constants of the instrument, the digitizer (and any filters, etc.) have been deconvolved, so we can now compare this data with that of any other sensor or site. Or we can compare the PSD of the same event between various sites with completely different instruments. I hope this clarifies the PSD picture. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: Tiltmeters, reply to ST Morrissey Date: Fri, 24 Apr 1998 20:52:43 -0500 (CDT) Bob, It may be that your Sperry unit is meant to bolt to a cieling. The one I have here has a pair of blue arrows over the word "TILT", and a fine screw adjustment that rotates it in the direction of the arrows, which is a vertical plane rotating about a pivot pin in the upper right corner. I haven't opened the ABS case to see exactly what sensor is inside, so I may be remembering wrong about the Fredericks bubble. We did evaluate some "high resolution" Fredericks units in the late 70s, but they were not as good as the Kinemetrics/Rockwell (a real thermal problem at the microradian sensitivity). Fredericks current line only indicates a null repeatability of 0.2 arc-minutes. (email: Francis Kull: fkull@.................. The Fredericks sensors are small slightly curved glass vials or tubes, about 1/4" dia and and inch long, with electrodes arranged out each end and one through the center. They are partially filled with a conductive electrolyte that looks like iodine. AC excitation is applied to the end electrodes, and the differential signal is sensed by the center electrode. (The electrodes are platinum, so the price is highly variable.) Different shapes of the vial give different sensitivities. If the vial is rotated 90 degrees about the long axis, it will still work, but with a different (and probably non-linear) sensitivity. Of course, I guess I will have to break open the Sperry sensor to see what is inside. Something to do next week .....? I can send you a copy of the paper if you send a SASE. Even greater fortune would let me find the Sperry literature. Regaards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: VBB Date: Fri, 24 Apr 1998 19:13:42 -0700 Sean Thomas I have uploaded two VBB files for the Hawthorne event. I noticed that the unfiltered event has a large low frequency component. When I ran it thru a single pole high pass with cut off @ at about the natural freq of the sensor I got a great record(flat log FFT response for low frequencies). Is this a coincidence or would a single pole high pass help? Have you noticed a similar response? I noticed your step response plot had higher than predicted response at low frequencies. Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Mark Robinson Subject: Re: Tiltmeters, reply to ST Morrissey Date: Sat, 25 Apr 1998 15:28:22 +1200 S-T Morrissey wrote: > The Fredericks sensors are small slightly curved glass vials or > tubes, about 1/4" dia and and inch long, with electrodes > arranged out each end and one through the center. They are partially > filled with a conductive electrolyte that looks like iodine. > AC excitation is applied to the end electrodes, and the > differential signal is sensed by the center electrode. > (The electrodes are platinum, so the price is highly variable.) > Different shapes of the vial give different sensitivities. > If the vial is rotated 90 degrees about the long axis, it > will still work, but with a different (and probably non-linear) > sensitivity. Of course, I guess I will have to break open > the Sperry sensor to see what is inside. Something to do > next week .....? > These sound to be exactly the same as the electronic level sensors used in Steadicam IIIAs. They may be available from Cinema Products as spare parts. In the steadicam they are driven using a National Semiconductor LVDT (Linear Voltage Something Transform :) chips. Cinema Products address is buried, I can dig it out if any is interested. They are unlikely to be cheap ! regards Mark _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Fredericks tilt sensors Date: Sat, 25 Apr 1998 01:57:08 -0500 (CDT) Mark, Your observation is correct. In fact, the latest Fredericks catalogue has a front page insert advertising the use of their sensors in the Steadycam. And I have little doubt that a LVDT driver chip would make them work nicely, since all they are is an AC bridge. The ones I evaluated here for geodetic studies worked with several AC bridge oscillator/amp/demodulator systems. For short term, relatively large amplitude tilt sensing with 12 arc-second repeatability (the "<0.2 arc-minute" specification from Fredericks), they should be pretty good. For reference, 5 arc-seconds is about 1 microradian; in geodetic studies, we need 0.001 microradian. I don't know what the current costs are from Fredericks, whether a second source might be better. But I don't think they have the resolution/noise level we would want for using them for horizontal seismic sensors. But I haven't run any numbers either. THe main problem is stiction of the bubble against the glass vial and irregular contact between the electrolyte and the electrodes. I have copies of all the research into these problems that was done by Rockwell/Autonetics in the early 70s. Lots of equations. (the references to Cooper et al are in D.C.Agnew's paper on straimeters and tiltmeters). The paper discussed earlier by Stein et. al. used the Rockwell sensor and a rather high quality amplifier/ filter setup.. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Metal Film Versus Wire Wound Resistors? Date: Sat, 25 Apr 1998 16:22:53 -0600 Hi all, Seems like the 1% metal film variety is the builders choice with most circuits anymore. Precision and low temperature drift I think. I hear that wire wound resistors are electronically noisy compared to other varietys....but to what degree I don't know. Have more endless questions..... I do have some precision wire wound resistors which I am hoping to use in the integrator of S-T's circuit down the road of time. These are 100K, with +- .0025% tolerance. I could paraell another to reduce the resistance to 50K if necessary. The resistors seem to be a very big part of the instrumental/integrator op amp accuracy. So...would using these be a bonus or detriment in regard to noise in the circuit? At this time, I could go with one op amp, or go with the "standard" 3 op amp with its 6 resistors, which may be much better in regard to this CMRR stuff? Am seriously considering using the Harris (old Intersil brand), ICL7650, a chopper op amp; because it is so cheap ~$2. Am perhaps overdoing the circuit question, but I think that this part is very important. No engineer experience here. Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "David A. Nelson" Subject: Re: Metal Film Versus Wire Wound Resistors? Date: Sun, 26 Apr 1998 11:42:46 +1200 At 04:22 PM 4/25/98 -0600, you wrote: >Hi all, > >Seems like the 1% metal film variety is the builders choice with >most circuits anymore. Precision and low temperature drift I think. > >Have more endless questions..... I do have some precision wire >wound resistors which I am hoping to use in the integrator of S-T's >circuit down the road of time. These are 100K, with +- .0025% >tolerance. I could paraell another to reduce the resistance to 50K >if necessary. The resistors seem to be a very big part of the >instrumental/integrator op amp accuracy. So...would using these be >a bonus or detriment in regard to noise in the circuit? >Thanks, Meredith Lamb No Meredith its the makeup of the metal film resistors that make them low noise. Low tolerance of ur wire wound resistors DOES NOT imply Low noise. Stay with metal film resistor for the best results. they are sooooo cheap anyway why bother using something else that is a worse performer. cheers Dave Co-ordinator: New Zealand Public Seismic Network Dave A. Nelson ZL4TBN 24 Jensen St., Green Is., Dunedin, South Is.. New Zealand. http://psn.quake.net/dave/home.htm IF there indeed are other parallel universes... I can now rest in the knowledge, that in at least one of them, I am filthy rich and drive a red Ferrari _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: Metal Film Versus Wire Wound Resistors? Date: Sun, 26 Apr 1998 09:05:07 -0700 Meredith -- In seismometers, noise that matters can be low-frequency enough for resistor people to call it drift. I believe a well-made wirewound resistor is usually quieter than a film resistor. But notice the disclaimer WELL-MADE. From what I understand, it's not easy to make them -- some of the problems: Thermal emf's generated due to temperature gradients across the bulk of the resistor (up to 10's of microvolts); Strains imparted to the wire due to stress on the leads (resistance change up to 10 parts per million (ppm)); Long-term drifts due to slow stress relieving of the packaging (10 ppm resistance change); Pickup of stray magnetic fields like 60Hz from nearby transformers etc. due to the fact it's a coil of wire (10's of microvolts & up, depending on the field strength and way the resistor is wound). There are other problems like resistance change with applied voltage, drift due to self-heating, drift due to current-induced migration of the resistance material, etc, etc, etc. And these are in addition to the problems of tolerance and temperature drift. At work, we have some precision dividers made of card-wound wirewound resistors (wire wound around mica cards) that are checked regularly, and are always much better than 1ppm -- year in and year out. But film resistors have their problems too. It wasn't until the last 15 years or so that even the best films have been able to match everyday wirewounds in temperature drift and long-term stability (read "noise" in seismometers). Perhaps someone else knows more about this, but I think the nature of the resistive material in films has more noise than the resistance wire of wirewounds. Some of the best resistors made are metal foil and card-wound wirewound resistors. Wirewound resistors are usually made from Evenohm (sp?) or other proprietary alloy, and connecting to copper leads generates thermal emfs. This is always a problem. Back to reality... I don't think the tolerance of the resistors used in Sean-Thomas' VBB design is very critical. I'm quite sure that 1% or even 5% values would work fine. But low thermal emf generation is definitely a concern, unless your temperature stability is good. Use as few op-amps as possible to get the job done (goes for resistors too). Each one introduces some noise and the fewer the better. You shouldn't need any common-mode rejection (CMR) in S-T's circuit, as (if I understand it correctly) it's all referenced to ground. Chopper op-amps are very good for 1/f noise and temperature drift. But be sure to synchronize the chopper with all other oscillators in your system (use one master oscillator) or you're sure to get intermodulation problems (beat notes between them). And don't let the chopper frequency get near 60Hz or a multiple of it, or you'll have the same problem. The newer chopper op-amps may produce less chopper-frequency noise at their outputs than older ones. Enough for now. Karl Cunningham La Mesa, CA. PSN Station #40 karlc@....... _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Re: Metal Film Versus Wire Wound Resistors? Date: Sun, 26 Apr 1998 12:12:25 -0500 (CDT) Meredith, Karl, and co. I think Karl addresses the problem of resistor noise and thermal drift quite accurately, and thanks Karl for the details. I would sum it up in saying that one has to read the specifications: you get what you pay for. We have bought some "now noise" resistors for as much as $10 per. And chip resistors are a whole different game. We do have some very old standard decade boxes with wire-wound on mica cards as Karl describes; I just assume that they are still good: they agree with the 4.5 digit DMM at room temp; we don't have the money to send them out to be calibrated again. THe few places I have seen wirewound Rs used in our applications are for the bridge resistors in the tiltmeters, where the temperature coefficient somewhat matches ONE of the thermal characteristics of the bubble bridge. Some of the commercial seismic feedback systems use very high quality metal film, selected for each configuration. (The european 1% resistors have so many color bands I never know which way to read them from.) A last note about amplifiers for the integrator: The OPA111 has been used successfully for this, even as an inverting integrator. I don't know what it costs. And my philosophy about costs: cheaper is not better. If a $10 amp will do the job without pushing the specs, it is to be preferred over a $5 amp that "might" work. We are not in a "dollar-saved, dollar-earned" situation, unless one is planning a bake sale featuring homemade siesmometers. If I find a device or hardware that clearly will fill the bill, (and obviously, not a HUGH expense), I buy it; or actually, I buy two or three, especially ICs, since I may ruin one, or the expected performance is not achieved because the one I tried was defective. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Marnie Gannon Subject: LA quakes Date: Sun, 26 Apr 1998 11:56:08 -0600 Hi Charles, Thanks for all your helpful posts on recent quake activity. I do have a question: These 3 recent LA quakes, 10 Miles NE of downtown, are described by Lucy on the Trinet site, as being in the NW aftershock zone of the Whittier Narrows quake. The 3.2 in LA on April 15 was described as being 4 miles NW of downtown and probably centered on the Elysian Park thrust and fold belt, which is also associated with the Whittier quake...Do you think these recent ones are on the same fault zone? Comments? Thanks, Marnie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Re: LA quakes Date: Sun, 26 Apr 1998 11:41:56 -0700 Marnie Gannon wrote: > > Hi Charles, > Thanks for all your helpful posts on recent quake activity. I do have a > question: > > These 3 recent LA quakes, 10 Miles NE of downtown, are described by Lucy > on the Trinet site, as being in the NW aftershock zone of the Whittier > Narrows quake. The 3.2 in LA on April 15 was described as being 4 miles > NW of downtown and probably centered on the Elysian Park thrust and fold > belt, which is also associated with the Whittier quake...Do you think > these recent ones are on the same fault zone? Comments? > Lucy has been looking at earthquakes in Southern Colifornia for a lot longer than me and I would tend to agree with here, sort of. The quake did fall close to the northern extent of the Whittier aftershock zone but I was not aware of the aftershock activity extenting this far past the Romona Freeway. This recent M3.8 triggered up at I10 near Garfield Ave. about 4 miles away. So it is close but not quite. Then again, she may have been talking spatially rather than tectonically. The M3.2 near Dodger Stadium was centered near the intersection of the NW-trending Elysian Park thrust and fold belt and ENE-trending Hollywood fault system. What I have seen of detailed studies of the area is that it is pretty cracked, and the quake could have triggered on either system. I haven't seen the focal mechanism for it yet. As for the association of the Elysian Park thrust and fold belt system with the Whittier fault system... well they both are in Southern California. :> There has been some speculation that the Whittier fault zone extends NW-ward into the Elysian Park thrust and fold belt. They are both thrust-slip fault systems, just the Whittier is more developed - like it should be. But you know, I went to the TriNet page and could find anything from Lucy regarding the Alhambra quake. But I did find a bunch of info from Kate Hutton at the SCEC page: http://flint.gps.caltech.edu/eqreports/comments/April1998.html Moreover, they have a new auto system of determining Mercalli intensities. I'm baffled it registered MMI IV 'cause nothing was shaken from tables or shelves. ta-ta, I'm off to enjoy the nice weather... -- ---/---- Charles P. Watson Seismo-Watch _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Marnie Gannon Subject: Re: LA quakes Date: Sun, 26 Apr 1998 12:52:57 -0600 Charles Watson wrote: >But I did find a bunch of info from Kate Hutton at the SCEC page: Sorry, my mistake. It WAS Kate I was quoting..Thanks for the info. Marnie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Resistors, Fleas, Illusions and Op-Amps Date: Sun, 26 Apr 1998 17:03:44 -0600 Karl, David, Sean-Thomas and everyone, Thanks for the responses. My original msg was only concerned with wire wound resistors use in the instrumentation amplifier portion of the STM-8; i.e., 2 to 6 resistors, not, for the total circuit package...the government could afford that, but not me. The wire wound resistors I am considering using are surplus flea mart type purchases of long ago. All are around 30 years old, and mostly salvaged off big old circuit boards, for like 2 cents each. Nevertheless, they seem to very well built...plastic coil inside, plastic fire resistance cases, springs on each end, air separation of the coil from the case. The place I got them at is no longer in business...they probably didn't know what they had. So....the price is right -ha. I suppose I should have spelled it out better in the original message. I'am fully for surplus scrounging bargains, alot of fun if you strike gold once in awhile. Trivia. Karls assessment of wire wounds was most interesting, I could even check afew responses like the emf with a magnet and some cigarette lighter thermal resistance drift. Saw ~ 8mv with magnet, and perhaps 1/1000th resistance drift. Interesting stuff...kind of wish he would have kept on with the message further. I see no real crisis in using wire wound resistors instead of metal film in any aspect......now. Hmmmm...maybe I ought to check out those old fat wire wounds with low ohmage for use as pickup magnet/coils. Karls chopper op amp logic says it all, in the sense of a chopper op amp is no good for the STM-8 circuit, with its other oscillator running. Scratch the chopper for this specific circuit. Instrumentation amplifiers seem to need a reasonably close resistance matching tolerance especially for the unity or gain of one amplier involved in the STM-8 package. I suppose one could check a number of like resistances and get a fair match, with a suitable bridge or DVM, if they have the stuff, and wanted to go that far. My only reference is those I see in electronics linear manuels, most of which hit around 1% variation or better is appropriate. I know....the STM-8 is meant for electronics professionals, and I'am trying to compete where I don't belong at all. Trouble is, the electronics IS the WHOLE key to using the STM-8 or any other seismographs, the mechanics are interesting and varied but they are usually not complicated or hard to replicate fairly. I think the STM-8 circuit should be more open to amateurs with alittle less experience...it looks too good! I suppose the real solution will be someone coming out some day with the ready made circuit....which is a hard reality for a realistic majority I think. Have checked the OPA111 on the internet. Allied sells it for about ~ $11 to ~ $23, according to offset rating of typical 250 to 500 microvolts in a TO5/TO99 package. The maximum offset of these goes about 3 times higher, per Burr-Brown data. One would need a 10K trim pot also. Perhaps the bigger question is the maximum input offset allowed for the design, or the average temperature drift, and then one could go accordingly to the massive open market variety and pocketbook? Reckon I've gained from the answers you folks from generiously provided. Thanks. Electronics is a real pain.... Meredith Lamb P.S's, Robert Barns,...thanks for the tiltmeter data and grams...very interesting too see. Will we see more? Would like to see a bigger quake and teleseism record personally..... Sean-Thomas,...hope you are totally over the flu like stuff. No fun at all. Aftereffects can drag on for some time.... David Nelson,...hope you are better from the surgery and perhaps hopping around much better. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: STM-8 expectations Date: Sun, 26 Apr 1998 22:15:08 -0500 (CDT) Meredith, (and company): There seems to be too much concern about the electronic requirements for the STM-8 seismometer. I need to point out that one of the attractive features of the triple VBB feedback seismometer is that just as there are no special demands on the mechanical system, other than it is stable, the same is true of the feedback electronics. I am using recycled amplifiers that are more than 15 years old for the prototype here, mainly because I happen to have them leftover from the tiltmeter projects. Everything is based on the LM308A, a very reliable "low noise" but not particularly special amplifier, other than its DC stability is well known. Even where the OPA111 is used as an integrator, it is not even necessary to compensate it. (ie: just ground pin 8). For the feedback elements themselves I am using ordinary 1% components, like resistors that cost less than $1. I have made sure that the feedback capacitor and the integrator capacitors are low leakage, by testing them with a DMM. Again, nothing special. Since the VBB output signal IS in fact the output of the displacement detector, that is where we have to be concerned about electronic noise and drift. But thermal/mechanical noises in the detector itself can be of equal concern. Fortunately, the VRDT has such a large output (50 to 100 mv/micron with a 2mm range), we have lots of signal to work with, so we are well ahead of the electronic noise. But the VRDT, unfortunately, IS a "wirewond" device, so it does have a real temperature problem, especially if the two coils are not very carefully matched (for inductance AND resistance) before it is assembled. "Hand winding" the coils of a VRDT or an LVDT is especially demanding in this respect. Temperature compensation of the VRDT can be done by paralling the coils with a differential thermistor bridge, or shunting one coil with an appropriately thermal responsive resistor. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Equalising article again Date: Mon, 27 Apr 1998 15:13:09 +0100 Hi, Regarding the article in Bulletin of Seismological Soc. of America, Vol. 79, No 5 Oct.1989, " A Versatile Equalization Circuit for Increasing the Seismometer Velocity Response below the Natural Frequency", by Peter Roberts, which was previously mentioned on PSN................ .................... I have made this cct. up to try out on one of my beams. It became a little tiresome working thro' the equations on page 1610.........so I now have a PC software patch that calculates the values for the amplification required for each of the two stages, and also the values of R*C ( R2 and C1 in cct diagram on page1612). Below are some values for some typical seismo configurations. It's now easy to see that if one is greedy and attempts to lower the natural freq. of a seismo by more than a factor of about x5, the amplification required becomes very high, so we should heed Sean Thomas's recent warnings on PSN about the noise problem associated with this method. Noise is also mentioned in the article. The last example serves only to illustrate this. The damping factor (Zeta) in the equations don't have a significant effect on the values for RC, nor gain required. (The table is based on a value for Zeta of .7) RC = value of R2 and C1 in Megohms and Microfarads. A = amplification required from EACH of the 2 stages... .....adjusting R1 in order to obtain this. (As A=R2/R1) (Overall amplification will be A*A) Natural freq. New Equalised freq R*C A of seismo to be obtained (seconds) 1Hz .2Hz (1/5th) 1.59 x10 1Hz .1Hz (1/10th) 3.18 x21.3 .5Hz .1Hz (1/5th) 3.18 x10 .5Hz .05Hz (1/10th) 6.37 x21.3 .1Hz .02Hz (1/5th) 15.9 x10 .1Hz .01Hz (1/10th) 31.8 x21.3 .06Hz (15secs) .012Hz (1/5th) 26.5 x10 .06Hz .006Hz (1/10) 53.0 x21.3 .06Hz .003Hz(1/20th) !! 106 !! x43.7 !! The last example is to illustrate ONLY that the greater the reduction in natural freq.attempted, then the greater the amplification required. (x43.7 amp factor for EACH stage is x1909 total...example ONLY) One can also see from this table that the A value is solely dependant on the reduction ratio. (A = 10 when equalised freq. is 1/5th of natural freq. of beam, and A = 21.3 when 1/10th) TRIALS........... As I have 2 seismos (facing EW and NS) I thought it would be an interesting experiment to equip one beam with this equalising cct. and compare it with a NON modified beam. I reduced BOTH beam periods to 12 secs, and used new cct to equalise to 60 secs. on the EW beam. Amplification required for each stage in order to achieve this is x10, and LP filter is 20 secs. (R*C=20). I cannot see any noise generated by the new cct., but this may well be because my normal LP filters used after the new cct are lowish at 5Hz. Incidentally, the new cct.is just a lash up using 6 - 741 op amps!!! (I ran out of 8 pin sockets, so 2 op amps are in ONE 8 pin socket!) However, storm microseisms etc. on the modified beam are now more frequently superimposed on lower freq. data, whereas the unmodified beam doesn't show these lower freq. to such a degree. In fact, my trace sometimes resembles the trace shown on page 1613 of article (a teleseismic event in Jemez mountains superimposed on 10-20 sec wind/ocean tremor). I sat 4ft away from beams with a magnet in my hands. Once the beams had settled, I turned magnet VERY slowly back and forth (about once per minute) in my hand!!..... ...... (the things we do for seismology) It was very apparent that the equalised beam had the greater o/p of sine wave (well, a rough sine wave), compared to the normal beam, and this seemed to indicate that a 12 second beam equalised down to 60 secs was more sensitive to LF. I shall do some more experiments on this system . If anyone wishes to read Peter Robert's article it is available from the American Seismological Society, and some members of PSN have also obtained copies from their own local University Libraries. As the Author of the article, Peter Roberts, said in his conclusion, to his article................ "The simplicity of the design procedures makes it readily understandable and easily modified and implemented without extensive knowledge of electronics." Now, that last statement, is ME to a T Regards Albert Noble (England) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bob@................. (Bob Hammond (USGS-AVO Geophysicist)) Subject: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 09:02:03 -0800 Do any of you have schematics for building a proton-precession magnetometer? I'm aware of the Sci Am article from 1968 on building such but wonder if there is anything newer (ie, digital)? Or, do any of you have leads on a used Geometrics proton-precession magnetometer for sale or trade? thanks, Bob Hammond APSN Fairbanks _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Instrumentation Date: Mon, 27 Apr 1998 12:00:29 -0500 Friends, Interesting discussion on resistors. Actually, I think the reason metal film resistors are generally to be preferred over carbon film resistors is primarily their lower temperature cooefficient and more precise vlues rather than noise. You can use a DVM to select metal ones that are matched to one part in a thousand, or make your own by soldering two in parrallel. In any case, metal films are indeed slightly less noisier but the difference in price is so trivial that it makes little sense to use carbon (A 50 pack of nickle-chromium film resistors are about $2.50 at Radio Shack). Carbon composition resistors may give noise well up into the microvolt range, but they're rather uncommon nowadays except in old circuits or for power, I think. The noise contributions of 1 k carbon film resistors is about 50 nanovolts, metal film 40 nanovolts, wirewound with crimped endcaps 30-10 nanovolts, and wirewound with welded ends 10 nanovolts. This from p 277, John Linsey Hood's 'Art of Linear Electronics', 1993. Compare with this the low noise circuit by Jim Williams in Linear Technology's Applications guide, 1987, which parallelled three LTC 1028s (their quietest) to an get extra low noise level of 30 nanovolts. Thus we can barely see the inherent noise from a carbon film resistor, even with such a low noise circuit. An op amp like the OPA 111, which is a precision low bias current JFET input device has a noise level of a few microvolts, so with this you would never see the noise in a carbon film I think. The main objections to carbon would then be thermal drift and matching accuracy. At any rate, you want to put your best efforts into the front end where the signal comes in and where noise and drift get in before the signal is amplified so that less demanding electronics can take over. John Hood's book is particularly excellent at discussing the important topic of the inherent properties of individual components and how this physics relates to front end design. **************************************************************************** Recently a retiring professor friend at the University of Texas gave me his copy of 'Building Scientific Apparatus', 1983, by Moore, Davis, and Coplan. Its a great book for science hacking, and reasonably up to date as the second best book that I know about. The best, the standard old classic in the field of homebrew scientific instrumentation in my opinion, is John Strong's 'Procedures in experimental Physics', published in the thirties, I think. Back in that era, everyone tended to build their own equipment, and Strong told you exactly what you needed to know to do it right and all the pitfalls using easily accessible equipment and materials. Another classic in the field of homebrew instrumentation is Robert W. Wood's 'Experimental Optics'. An interesting fellow who roamed the world, and settled down at John Hopkin's University some time around the turn of the century. He invented the spinning parabolic mercury mirror, the quartz ultrasonic generator, and many useful methods and techniques in optics. Wood had the knack of building amazingly sophisticated optical instrumentation with odds and ends and he explains how to duplicate everything yourself in exceptionally well-written detail. --Yours, Roger _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: [Fwd: PR for Gulf Islands earthquake] Date: Mon, 27 Apr 1998 10:29:28 -0700 FYI -- ---/---- Charles P. Watson Seismo-Watch From: Robert L Barns <75612.2635@..............> Subject: Resistors, Fleas, Illusions and Op-Amps Date: Mon, 27 Apr 1998 15:35:44 -0400 Meridith, Sorry, there will be no more 'grams from that tiltmeter--it has been returned to Scolnik and he is using it to observe the tilt of a massive pier on which one of his high precision pendulum clocks is mounted. Maybe you could borrow Sean-Thomas' device and try it. Bob Barns _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Doug Crice Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 13:12:10 -0700 You can buy a used proton magnetometer from R.T. Clark, 405-751-9696 that ubiquitious purveyor of used geophysical instruments including geophones and drum recorders. Probably they won't be cheap enough to tempt you, but there should be more out there with the Cesium mags catching on. As far as schematics, the proton mags haven't really changed much in the last 30 years. Phase lock loops and counters are still phase lock loops and counters. Sure, now you can measure the period instead of the frequency and convert it to gammas (oops, nanoTeslas) with an embedded microprocessor, but so what? The paradigm shift has been to the Cesium mags which are a whole new ball game. I'm sure the geophysics department at the University of Alaska has an old G-816 knocking around, did you take a look at it? -- Doug Crice web site http://www.georadar.com GeoRadar Inc. e-mail dcrice@............ 19623 Via Escuela Drive phone 408-867-3792 Saratoga, CA 95070 USA fax 408-867-4900 Bob Hammond (USGS-AVO Geophysicist) wrote: > > Do any of you have schematics for building a proton-precession magnetometer? > I'm aware of the Sci Am article from 1968 on building such but wonder if there is anything newer (ie, digital)? > > Or, do any of you have leads on a used Geometrics proton-precession magnetometer for sale or trade? > > thanks, > > Bob Hammond > APSN > Fairbanks > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: David Josephson Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 13:42:50 -0700 (PDT) > Or, do any of you have leads on a used Geometrics proton-precession magnetometer > for sale or trade? Suggest calling Ross Johnson at Geometrics, see if he can put together a kit of pieces for an 801 (with tweaking, good to about 0.25 nT) or something newer. This would be called a "base station magnetometer" in their parlance, so sensor heading error would not be an issue. 1-408-954-0522 _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: David Josephson Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 13:48:04 -0700 (PDT) Doug Crice wrote > As far as schematics, the proton mags haven't really changed much in the > last 30 years. Phase lock loops and counters are still phase lock loops > and counters. Sure, now you can measure the period instead of the > frequency and convert it to gammas (oops, nanoTeslas) with an embedded > microprocessor, but so what? The paradigm shift has been to the Cesium > mags which are a whole new ball game. That's all true, but for the frequency band of interest (0.001 - 0.1 Hz, if you take the conventional view) a good proton mag is almost as good. But I think if I were going to build one from scratch, it would indeed be a cesium mag; the electronics are much simpler. Maybe Al would give me back that cesium sensor. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: juggie@................... (Nick Tile) Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 98 21:49 BST-1 In-Reply-To: <199804271702.JAA10806@kiska> Bob, There was a cct for a Proton Precession magnetometer published in a UK electronics journal nearly 20 years ago (Practical Electronics) - you could try phoning them on 1202 841 692, or e-mailing them on: editorial@...................... or visiting their web site: http///www.epemag.wimborne.co.uk. I can't recall the details, but I do recall that it was very large, very cumbersome and involved walking around with two coils, one on each end of a long pole :-) If you want to build a maggie, you could try a search on the web for the FGM-X series of chips and sensors from Speake 1873 811 281, or 1873 810 958 (fax) - they also have an american distributor who is on the web, (Bill Speake isn't YET )and has all their spec sheets and data sheets zipped up in downloadable form - I have a simple Maggie of theirs sitting in here now, showing that things are reasonably quiet, but a couple of days ago they weren't. Nick, Colchester, England _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Erich F. Kern" Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 15:06:09 -0700 Hello, Speake & Co. has a U.S. distributor (yours truly). The data sheets, application notes and price list may be downloaded from our web site at: http://www.dconn.com/fatquarterssoftware When you visit this site, there's a link to Darrel Emerson's web site which will be of interest to those monitoring solar activity. Darrel's posted some very good plots of the geomagnetic field as detected with the Speake sensors and comparison plots from the Tucson Magnetic Observatory during the same 3 day period. Best Regards, Erich Kern ---------- > From: Nick Tile > To: psn-l@............. > Cc: juggie@................... > Subject: Re: Proton Precession Magnetometer circuits > Date: Monday, April 27, 1998 14:00 > > In-Reply-To: <199804271702.JAA10806@kiska> > > If you want to build a maggie, you could try a search on the web for the > FGM-X series of chips and sensors from Speake 1873 811 281, or 1873 810 > 958 (fax) - they also have an american distributor who is on the web, > (Bill Speake isn't YET )and has all their spec sheets and data sheets > zipped up in downloadable form - I have a simple Maggie of theirs sitting > in here now, showing that things are reasonably quiet, but a couple of > days ago they weren't. > > Nick, Colchester, England _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: [Fwd: WARNING: POTENTIAL MAJOR SOLAR FLARE WARNING - 27 APRIL] Date: Mon, 27 Apr 1998 16:17:48 -0700 Looks like it is time to rock 'n roll! -- ---/---- Charles P. Watson Seismo-Watch From: S-T Morrissey Subject: equalization noise etc. Date: Mon, 27 Apr 1998 18:22:03 -0500 (CDT) Albert, Roger, and co. Albert, Thanks for the discussion and calculations of the parameters for equalizer circuits, And putting 2 and 2 together, the timely review of resistor construction and typical noise levels by Roger is definitely appreciated. I would like to again suggest the article "Limits of Sensitivity of Inertial Seismometers with Velocity Transducers and Electronic Amplifiers" by M.A.Riedesel, R.D.Moore, and J.A.Orcutt; (BSSA, Vol 80, No. 6; pp1725-1752, December 1990). [ with the observation that the axis of Figure 3 should be "ohms" and not "frequency"]. This paper gathers all four of our concerns (amplifiers, resistors, seismometers and earth noise ) into a single discussion, with lots of experimental data. While the paper doesn't touch directly on the equalization question, it can provide some insights. The seismometer/amplifier noise figures can suggest the noise we will have to deal with in an equalizer with a total gain of 100 to 400, even with an optimal match between the seismometer coil R and the amplifier input characteristics. It also shows that to "equalize" a 4.5hz phone to the output level of an L4C at 1 hz requires an overall gain of over 10 000, even with high impedance coils. Figure 14 shows that the coherence between a 4.5 hz phone and the L4C will be very poor, like 0.3, at 1 hz because of noise considerations alone. Equalization will not change this, but actually adds to the noise, and the coherence is about the same, which we have confirmed here. this explains why we have had noise problems with our equalizer circuit evaluations, especially in the 1 to 10 second range, where we see 2 to 8 mv of noise, and much more if we look too closely and actually breathe on the input amplifier and resistors, an effect that someone mentioned recently. Regards, Sean-Thomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Instrumentation Date: Mon, 27 Apr 1998 16:40:37 -0700 To All As I was screwing the Aluminum portion of my sensor to the Brass Portion I began to wonder about the temperature sensitive potential that may be generated by the bymetal mechanical connection. Could it act like a thermocouple? I run my ground thru the composite boom from the displacement sensor to the amp. It saves me one wire and grounds the entire sensor. Am I developing a slight potential between ground and each connection? Barry _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: ACole65464 Subject: Book Availability Date: Mon, 27 Apr 1998 23:14:28 EDT Hello PSN Netizens (do you like that word?), Do I remember correctly, but did someone a few weeks ago say that they had difficulty in trying to locate a copy of Bruce Bolt's book titled Earthquakes and Geological Discovery? I came across 4 copies of it on the book shelf in the gift shop of the Nevada State Museum in Carson City, Nv a few days ago, their price was $32 each. I hope this helps someone. Allan Coleman, Edmonds, Washington _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Ted Blank Subject: Quake alert by e-mail to pager? Date: Mon, 27 Apr 1998 22:01:47 -0700 (PDT) A new version of EMON is in the works. One new feature is the ability to execute a BAT file or command (EXE) when a quake file is saved. Since I got this new pager that lets people page me via e-mail, I wondered if there was any way to have a PC compose a message, dial a phone number, send an e-mail message and hang up. I suppose it would have to use an existing ISP account (like this one on netcom). This way the PC could page me if a large quake was recorded. I'm sure someone has tried this before, so if you have ideas you want to share with the group, or send me via private mail, I'd appreciate it. The new version also supports BAT/EXE file execution on a scheduled basis. For instance, if you want to run a BAT file to set your clock by dialing a time service every Sunday at 2:00 am, you can now tell EMON to do that. Up to two separate events can be scheduled in addition to the one which is performed after a quake file is saved. If anyone wants to try out the new version, let me know and I'll send a copy. It's been running for a week on my system. I'm also adding support for a new A/D card for Alby Judge, which is just about ready, just needs some testing. :-) Ted _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Doug Crice Subject: Re: Proton Precession Magnetometer circuits Date: Mon, 27 Apr 1998 22:11:02 -0700 David Josephson said: >That's all true, but for the frequency band of interest (0.001 - 0.1 Hz, if you take the >conventional view) a good proton mag is almost as good. But I think if I were going to >build one from scratch, it would indeed be a cesium mag; the electronics ar e much simpler. >Maybe Al would give me back that cesium sensor. Sure the electronics may be simple, but blowing a glass cesium sensor is right up there with building superconducting gravimeters on the difficulty scale. -- Doug Crice web site http://www.georadar.com GeoRadar Inc. e-mail dcrice@............ 19623 Via Escuela Drive phone 408-867-3792 Saratoga, CA 95070 USA fax 408-867-4900 _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: FFT, grounds, etc Date: Tue, 28 Apr 1998 00:34:32 -0500 (CDT) Barry, I am still puzzling over your question of Apr 24 about high pass filtering of the VBB data, especially at the "natural frequency" of the sensor. Do you mean the mechanical period, or the electronic VBB period? I'm not sure what you are seeeing for the FFT. Is it a true FFT or a Power Spectral Density (PSD)? Regarding the high pass: I use a single pole far removed from the instrument response to remove the near DC data/drift from my recording. The corner is at 1000 seconds. In the plot of the PSD of the calibration step, this results in a steeper slope at the long period response. The slightly higher response near To of my step calibration is because I failed to adjust the Rp or damping resistor when I changed the mechanical period from 2 to 4 seconds, and the VBB response was under damped. Actually, the calibration step only gives the shape of the response. Playing with the math of it only backs out the value of Rp. The only way to truly calibrate the VBB instrument is with a tilt table. Fortunately, knowlege of the mechanical and feedback parameters yields a fairly accurate determination of the response via the transfer function. Regarding your concern today about grounding with dissimilar metals: it all just becomes part of the picture. If the connections are clean and low resistance, the slight EMFs aren't enough to worry about. I very frequently ground aluminum with copper or brass ground lugs. I am curious about your mention of a composite boom. My table of general coefficients of materials generally lists resin-based synthetics as having high thermal coefficients. Like formica is 170 (ppm/ degC), plexiglass is 90, etc,; whereas aluminum is 28, brass is 20, and iron is 12, etc. Maybe you are using a special composite or one with a high coefficient to counteract thermal problems? Regards, Sean-Thomas. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Re: STM-8 vrdt matching Date: Mon, 27 Apr 1998 23:37:05 -0600 S-T Morrissey wrote: > > > Since the VBB output signal IS in fact the output of the > displacement detector, that is where we have to be concerned > about electronic noise and drift. But thermal/mechanical noises > in the detector itself can be of equal concern. Fortunately, > the VRDT has such a large output (50 to 100 mv/micron with > a 2mm range), we have lots of signal to work with, so we > are well ahead of the electronic noise. > > But the VRDT, unfortunately, IS a "wirewond" device, so it > does have a real temperature problem, especially if the two > coils are not very carefully matched (for inductance AND > resistance) before it is assembled. "Hand winding" the coils > of a VRDT or an LVDT is especially demanding in this respect. > Temperature compensation of the VRDT can be done by paralling > the coils with a differential thermistor bridge, or shunting > one coil with an appropriately thermal responsive resistor. > > Regards, > Sean-Thomas > Sean-Thomas, The xfmrers or potential vrdt's I have experience with, most fall with a nice narrow range of like 212 to 225 ohms; with the majority (7) in the 217 ohm setting out of a bunch of some 20. This was a nice surprise. Its easier to match the total resistance than the inductance by far. Of the 7, there were only 2 close inductance matchs of .1mh and another of .2mh, of the 7 total available. The primary would usually range around 4.7 to 5.1mh, and the secondary would range around 58.2 to 60.0mh. An exact match would either be luck of the draw or buying a greater number of the Mouser transformers to convert. This was done with the 14 E core laminations in the coils. I was never successful in squeezing in more than 14 in any coil. I suppose the vrdt is alot like my Sprengnether coils in the sense that it is the most valuable part of the whole unit....replacing them is a very tough deal. Any more tips or tricks, or a reasonable inductance range of match other than perfect? I don't know... Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Resistors & Magnetic Spring Date: Mon, 27 Apr 1998 23:46:32 -0600 Roger, Thanks for the resistor and assorted reference information. Looks like I do have some ~ 10 nanovolt noise wire wounds..wouldn't have found out otherwise probably. Also, I found a round brass weight and by hand juggling it, I could get a more first hand (ha), picture of your magnetic spring device. The combined weight of the brass and magnets mass, came to a lifted weight of about 4 & 1/2 pounds I figure per bathroom scale. Parlor type appreciation approach, but interesting anyway. Makes me wonder how the Sprengnether model approach was done? Thanks, Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Charlie Rond Subject: Re: Quake alert by e-mail to pager? Date: Tue, 28 Apr 1998 02:50:53 -0500 Ted, >A new version of EMON is in the works. One new feature is the ability to >execute a BAT file or command (EXE) when a quake file is saved. Since I >got this new pager that lets people page me via e-mail, I wondered if >there was any way to have a PC compose a message, dial a phone number, >send an e-mail message and hang up. I suppose it would have to use an >existing ISP account (like this one on netcom). This way the PC could >page me if a large quake was recorded. > >I'm sure someone has tried this before, so if you have ideas you want to >share with the group, or send me via private mail, I'd appreciate it. Exciting new features in your wonderful program! The folks at CERI are paged in a similar way when a big quake occurs, or one in the New Madrid area. I don't know the specifics of how the system is triggered, but could find you a contact there if you want any additional info. Charlie Rond -\/\--Public Seismic Network-Memphis rond@................ (901) 360-0302 [ free BBS ] http://gandlf.ceri.memphis.edu/~rond http://gandlf.ceri.memphis.edu/~rond/psn _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: equalization noise etc. Date: Tue, 28 Apr 1998 08:09:25 +0100 S-T Morrissey wrote: . It also shows that > to "equalize" a 4.5hz phone to the output level of an L4C at > 1 hz requires an overall gain of over 10 000, even with > high impedance coils. Hi Sean, I really hesitate to bother you....you seem to spend all your time answering questions on PSN! BUT, my calculations for a 4.5Hz seismo equalised to 1Hz gives an RC value=.318, and an OVERALL amplification of only 79 !! ........using the article by Peter Roberts in American Seismology bulletin, which I posted on PSN yesterday. How does this very LOW figure square with your 10,000 above??? I can well understand noise problems with gains of 10,000, but the Robert's cct. doesn't require such large values, EVEN WHEN equalising down by a x10 factor...(your own 4.5 - 1 is less than x5.) I realise my magnet experiment was unscientific (turning it slowly back and forth), but it did show nice o/p at 60secs, which the other unequalised beam didn't have, and no noise, even with 741's!! Now I know that, once again, I've got hold of the "wrong end of the beam somewhere".......but know not, I. Regards Albert Noble (England) > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Ted Blank Subject: "Quakes" program on Discover Channel Date: Tue, 28 Apr 1998 07:42:56 -0700 (PDT) Last Sunday night Discovery Channel broadcast their program "Quakes". It has a lot of footage taken during quakes, the typical pans and scans of post-quake damage, plenty of footage of burning buildings, etc. However the hightlight was when they were visiting Waverly Person at the NEIC. You see him on tape excitedly pointing to each of a bank of drum recorders as a quake rolls in, saying "Here's the P wave on this one, and here's the P wave over here, we don't even know where this one is yet..." It's clear this is not rehearsed. There's a little bit of animation of plate movement but overall it's pretty weak on the science part in my opinion. Anyway, if anyone would like a copy just send me a tape and a paid return mailer and I'll copy it for you. 954 Foxswallow Ct., San Jose, California 95120. Ted Blank Please allow 4-6 weeks for delivery. Offer not valid where taxed. You may have other rights which vary from state to state. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Karl Cunningham Subject: Re: Quake alert by e-mail to pager? Date: Tue, 28 Apr 1998 08:28:40 -0700 Ted -- My Internet Service Provider has a service (for a fee) that will forward email messages to an alphanumeric pager. I don't know any details, but you might check with your ISP. Karl Cunningham La Mesa, CA. PSN station #40 karlc@....... You wrote: >Since I >got this new pager that lets people page me via e-mail, I wondered if >there was any way to have a PC compose a message, dial a phone number, >send an e-mail message and hang up. I suppose it would have to use an >existing ISP account (like this one on netcom). This way the PC could >page me if a large quake was recorded. > >I'm sure someone has tried this before, so if you have ideas you want to >share with the group, or send me via private mail, I'd appreciate it. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: JHammes@......................... Subject: Re: Book Availability Date: Tue, 28 Apr 1998 08:47:34 -0700 Regarding Bolt's book: it's available from www.amazon.com for $24. Jerry Hammes Palo Alto, California ______________________________ Reply Separator _________________________________ Subject: Book Availability Author: PSN-L Mailing List at CCSMTP Date: 4/27/98 11:14 PM Hello PSN Netizens (do you like that word?), Do I remember correctly, but did someone a few weeks ago say that they had difficulty in trying to locate a copy of Bruce Bolt's book titled Earthquakes and Geological Discovery? I came across 4 copies of it on the book shelf in the gift shop of the Nevada State Museum in Carson City, Nv a few days ago, their price was $32 each. I hope this helps someone. Allan Coleman, Edmonds, Washington _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: meredith lamb Subject: Magnetic Spring Thought Date: Tue, 28 Apr 1998 10:10:12 -0600 Roger Baker, Have a thought for your magnetic spring device. There are obviously mass center magnetic twisting torques when using your flat edge donut magnets per your initial device. Placement of material has to be somewhat adjusted around these by slideing your razor pivot or the positioning of the lifting magnets. How about the use of upward curved magnets (surplus motor magnets) for the lift; and, say a round bar magnet for the mass of appropriate length and diameter? I'am guess that alot of the torque around the mass center would disappear, except of course down toward the pivot direction. Meredith Lamb _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Roger Baker Subject: Yet more instrumentation stuff Date: Tue, 28 Apr 1998 13:22:06 -0500 Well Merideth, et al, There's million different arrangements that should work OK. Mine was simple to set up and works too. Its hard to visualize your exact concept. Let me only say that a stiff razor blade with an oiled edge is is awful good at accurately defining an axis of rotation for the beam, when seated on a thin brass sheet with small rare earth magnets behind the brass to hold this edge in light contact. And the damping forces, which are the enemy of sensitivity in this case are very slight, leading to a very high Q mechanical system. The center of gravity of my very short beam does follow a short radius curve of a little over half an inch--in theory --but the force feedback force keeps the boom from moving much at all, assuming feedback power is sufficient to equal the acceleration forces encountered. That is the reason why I used a two ounce lead fishing weight on the beam -- and my wimpy little hundred or so milliwatts of feedback from just one little 324 op amp just isn't enough to completely damp the ringing from ambient urban noise. So anyone who wants to duplicate my setup should probably add more power, and in fact go for it with my blessing!, but I'm primarily interested in other details at the moment. Like what is the limiting sensitivity of my optical detector? Right now I am vibrating a dummy flag with a piezo transducer and am about to use lock-in to see if the sensitivity goes down to a nanometer DC displacement or better. My apparatus has a 20:1 aluminum lever and a micrometer so I can actually calibrate displacement pretty accurately I think. One thousandth inch displacement on the micrometer is somewhat more than one micron displacement of the flag. Then you vibrate the flag with an AC signal and you can compare the two magnitudes of movement. It looks like an AC signal of three volts on the piezo is equal in amplitude to about one micron of flag movement induced by the micrometer/lever. You can see the one micron displacement sine wave on the scope very sharp and clear, and I imagine that one can probably detect one hundredth or thoudsandth of this amplitude. But that may be premature for me to judge. Its best to use the resonant freq of the piezo. By now reducing the AC voltage into the piezo by a factor of one thousand, and seeing if it is still detectable with a lock-in, it should be possible to see if the sensitivity limit really is a nanometer or less. (Another option is a tuned bandpass filter). Lock-ins are tremendous little gadgets to use when you have a captive modulated signal to detect and want to add another decimal point in your favor to a signal to noise ratio. It may turn out that the most sensitive arrangement is to use some form of modulated LED and lock-in arrangement to detect displacement. Anyhow, this should all give me a pretty good feel for the limiting sensitivity of displacement detection setups with an occulted LED/phototransistor combination. Also, it has become apparent that one must accurately align the knife edges to be exactly parallel if diffraction is to be a limiting factor; otherwise the slope of the cutoff curve will be too gradual as a function of displacement. And so on and so forth; blah blah blah... --Yours, Roger At 10:10 AM 4/28/98 -0600, you wrote: >Roger Baker, > >Have a thought for your magnetic spring device. > >There are obviously mass center magnetic twisting torques when >using your flat edge donut magnets per your initial device. Placement >of material has to be somewhat adjusted around these by slideing >your razor pivot or the positioning of the lifting magnets. > >How about the use of upward curved magnets (surplus motor magnets) >for the lift; and, say a round bar magnet for the mass of appropriate >length and diameter? I'am guess that alot of the torque around the >mass center would disappear, except of course down toward the pivot >direction. > >Meredith Lamb > > >_____________________________________________________________________ > >Public Seismic Network Mailing List (PSN-L) > >To leave this list email listserver@............. with the body of the >message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: more equalization noise Date: Tue, 28 Apr 1998 13:36:44 -0500 (CDT) Albert, I guess I wasn't clear enough. THe gain figure I mentioned is estimated from figure 1 of the Riedesel et al paper., and as I said, it is what is needed to produce the equilavent output of an (their) L4C at 1 hz from a 4.5 hz phone. The gain figure is not just the gain of the equalizer, which in our design (for a corner somewhat greater than 1 second to stay in the more linear portion of the response curve) needs to be 12.33**2, or about 150, but the additional gain needed to get the 1 hz signal level of the equalized 4.5 hz phone up to that of the L4C at 1 hz. The Riedesel study uses special high-impedance coil seismometers, so the L4 output they show in figure 1 is about 800 v/m/sec (I think the Y axis scale of that figure is labeled wrong : it shows a power spectral density, rather than an output voltage due to acceleration). THeir L4 has a 33k ohm coil, and the 4.5 hz L15B has a 960 ohm coil. The output voltage of the L4C at 1 hz is about 10 000 times that of the L15B at 1 hz. I our application, we are targeting the output of a standard 5500 ohm L4C with an output of about 300 V/m/sec, trying to equalize a well over damped HS-1 with an output of about 9.4 V/m/sec at 4.5 hz, or about 0.9 v/m/sec at 1 hz. (I am estimating, since I don't have the figures here). So the equalization flattens the response out to ~1.3 seconds to a level of about 10 v/m/sec, with a total gain of 150. An additional gain of 30 is needed to match the L4C level, for a total gain of 4500. This additional gain can be either after the equalizer, or in a preamp before it. Either way, we have noise problems. They can be lived with for low gain applications, but clearly show up in the data from a 24-bit ADC. BTW: Your rotating magnet is a nice idea to induce a 60-second influence on your seismometer. Could you attach the magnet to the minute hand of an analog clock? One group has calibrated STS-1 sensors by gravitational attraction, by rolling a bowling ball past them on a wooden track. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: more equalization noise Date: Tue, 28 Apr 1998 13:36:44 -0500 (CDT) Albert, I guess I wasn't clear enough. THe gain figure I mentioned is estimated from figure 1 of the Riedesel et al paper., and as I said, it is what is needed to produce the equilavent output of an (their) L4C at 1 hz from a 4.5 hz phone. The gain figure is not just the gain of the equalizer, which in our design (for a corner somewhat greater than 1 second to stay in the more linear portion of the response curve) needs to be 12.33**2, or about 150, but the additional gain needed to get the 1 hz signal level of the equalized 4.5 hz phone up to that of the L4C at 1 hz. The Riedesel study uses special high-impedance coil seismometers, so the L4 output they show in figure 1 is about 800 v/m/sec (I think the Y axis scale of that figure is labeled wrong : it shows a power spectral density, rather than an output voltage due to acceleration). THeir L4 has a 33k ohm coil, and the 4.5 hz L15B has a 960 ohm coil. The output voltage of the L4C at 1 hz is about 10 000 times that of the L15B at 1 hz. I our application, we are targeting the output of a standard 5500 ohm L4C with an output of about 300 V/m/sec, trying to equalize a well over damped HS-1 with an output of about 9.4 V/m/sec at 4.5 hz, or about 0.9 v/m/sec at 1 hz. (I am estimating, since I don't have the figures here). So the equalization flattens the response out to ~1.3 seconds to a level of about 10 v/m/sec, with a total gain of 150. An additional gain of 30 is needed to match the L4C level, for a total gain of 4500. This additional gain can be either after the equalizer, or in a preamp before it. Either way, we have noise problems. They can be lived with for low gain applications, but clearly show up in the data from a 24-bit ADC. BTW: Your rotating magnet is a nice idea to induce a 60-second influence on your seismometer. Could you attach the magnet to the minute hand of an analog clock? One group has calibrated STS-1 sensors by gravitational attraction, by rolling a bowling ball past them on a wooden track. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Steve Hammond Subject: RE: Quake alert by e-mail to pager? Date: Tue, 28 Apr 1998 12:31:01 -0700 OK I'll bite...=20 I envision this as having four components:=20 1) Setting the trigger level and triggering. 2) Determining if there is more data (big event can create 8 or 9 event = files). 3) Creating a message and putting it in a sub-directory. 4) Sending the page message. Detail: (1) This is standard feature and already done by EMON. A test will need = to be added at trigger time if a pager message needs to be sent. Test = =3D true? , save the time_save, event_count_V_save, event_count_N_save, = event_count_S_save, go to step #2 =20 =20 (2) To be sure that you don't miss marginal levels of event data from = distant events, EMON needs to keep monitoring for a period after the = data set is saved. You want to be sure you capture event file number = 2~N. This will take some new code. The code would wait looking at buffer = A and B and count the number of discarded buffers. This could be a new = profile option: Number_of_buffers_until_call xx ; This values = can be determined in minutes based on the time it takes your system to = fill buffer A and B. (A value of 10 on my system would be about 30 = minuets.) Test =3D true? value reached. Go do step #3 (3) Format a message for the pager and save it to disk. time_save, = event_count_V_save, event_count_N_save, event_count_S_save taken from = the initial saved values in step #1. Save the formatted message file to = the EMON event file directory in a plan text format under the name = PAGER.TXT. Exit EMON and invoke PAGER.BAT step #4 and Return to EMON = monitoring =20 (4) PAGER.BAT invoke TELIX and uses the script function of TELIX to = dial the pager number and send it the PAGER.TXT file. This is pager = company dependent item, but I have a Skytel pager and sent the message = 174523231235258 (length in bytes time-6,vdata-3,ndata-3,sdata-3) and = the pager formatted it to look like 1-745-232-3123-5258 which worked for = me. We used the Telix script functions in the BBS days and they are easy to = create and dependable as long as you test for things like busy... These are my thoughts-- Comments? Regards, Steve Hammond=20 PSN San Jose, California -----Original Message----- From: Ted Blank [SMTP:tblank@........... Sent: Monday, April 27, 1998 10:02 PM To: psn-l@.............. Subject: Quake alert by e-mail to pager? A new version of EMON is in the works. One new feature is the ability = to=20 execute a BAT file or command (EXE) when a quake file is saved. Since I = got this new pager that lets people page me via e-mail, I wondered if=20 there was any way to have a PC compose a message, dial a phone number,=20 send an e-mail message and hang up. I suppose it would have to use an=20 existing ISP account (like this one on netcom). This way the PC could=20 page me if a large quake was recorded. =20 I'm sure someone has tried this before, so if you have ideas you want to = share with the group, or send me via private mail, I'd appreciate it. The new version also supports BAT/EXE file execution on a scheduled=20 basis. For instance, if you want to run a BAT file to set your clock by = dialing a time service every Sunday at 2:00 am, you can now tell EMON to = do that. Up to two separate events can be scheduled in addition to the=20 one which is performed after a quake file is saved. If anyone wants to try out the new version, let me know and I'll send a=20 copy. It's been running for a week on my system. I'm also adding=20 support for a new A/D card for Alby Judge, which is just about ready,=20 just needs some testing. :-) Ted _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: barry lotz Subject: Re: FFT, grounds, etc Date: Tue, 28 Apr 1998 15:23:47 -0700 Sean Thomas Well let's see if I can explain. My VBB sensor is very much like your configuration; however, I use a capacitor bridge to measure displacement. The output goes thru an instrumentation amp,a 44 hz single pole low pass, and into the triple connection point. The VBB output goes thru a 4700 sec single pole highpass,buffer, 11 hz single pole low pass, 1 sec single pole high pass, and into the output amp/line driver. The reason I included the single pole high pass with a cutoff freq of 1 sec was,in the FFT response to ambient vibration and a local event (250k away) there was an about 12 db/decade increase in response below 1 sec. It was saturating my output. With this slight filtering I was able to get a flatter response at the low end. The following are my MCAD parameters(sorry about the lack of subscripts):r=260900, t=1.7 sec, m=0.35, c=50E-6, rp=5.8E5, ri=1.07E5, rf=8, ti=39, and tn=91. I have heavily thermally insulated the sensor and circuit. I am getting an about 2 vt swing in the mass position output each day, which I attribute to temperature. I plan to set up a controller to measure temperature, time and mass position every hour for the day to get an idea of it's properties. The one other noticeable property of the sensor so far is that the ambient noise has the typical about 6 sec microseisms but also has an equally large 2 hz component(not seen with my Lehman). Could this be the vibration of the leaf spring? I haven't eddy current damped it yet. Barry PS The response I describe above I believe is an FFT which I get from Larry's fine program. My boom is made of the channel you used but the displacement sensor is a brass unit at the end. Finally, I think the unit maybe under damped since if I remember the damping coef is 0.46. Shouldn't it be around 0.7 like the Lehman? With a higher damping coef, the low frequency response drops off at a little higher frequency than with the lower damping. I could increase the ti to correct this. _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: watson@................ (Charles Watson) Subject: Re: Quake alert by e-mail to pager? Date: Tue, 28 Apr 1998 17:22:37 -0700 WOW! EMON sounds like a dandy program. Where did you say one could get some more information? -- ---/---- Charles P. Watson Seismo-Watch _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: Barrys' VBB Date: Tue, 28 Apr 1998 20:20:10 -0500 (CDT) Barry, I updated the info for the mathcad response you gave me some time ago, which indeed showed that yout VBB was underdamped at 0.48. Generally zeta should be greater than 0.707 (the critical value). The value of Rp is the principal control of the damping. I tried a value of 270k ohms to raise your damping to about 0.78. (conversely, if you raise Rp to 10meg, your seis will actually oscillate at Tn). But I noticed that the output is quite low at 553v/m/sec. I think it is because your Cp is much too large with respect to your mass of 0.35kg. I tried a value of 24 uf instead, and virtually put the rest of the C in the integrator to lengthen it to 80 seconds, to keep Tn at 90. Then the output just about doubles to 1153 v/m/s. I also changed Rp to 1.0 meg, but left RI to 107k. The damping is about 0.78. Of course, the possibilities are endless. I also use several poles of low pass filtering at about 5 hz to get the cultural (=traffic, my dogs, me, etc.; I have a railroad about 2 km away, and a 8-lane interstate about the same) out of the data. Someday I will put the sensor in the 10meter pit at the experimental seismic station in the back field, and reduce the noise filtering. In the meantime, I got a nice record of the Mblg 4.2 near Oklahoma city this AM (1413z), about 1000 km distant. Maybe some of our Colorado PSN stations recorded it? About your 2 hz noise: it may be real, but vertical, probably cultural, but not seen by your other (horizontal) sensor. My spring rings at a much higher frequency. I am unsure about your 1hz high pass corner. I think it will wipe out your broadband response. Regards, Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: S-T Morrissey Subject: multiple responses Date: Tue, 28 Apr 1998 21:39:18 -0500 (CDT) Re multiple posts of mail responses: The multiple posts occurr because I don't always notice the multiple addresses in the mail I receive, especially from England. When I use rhe R command to reply, it replies to both addresses. Of course, I also receive multiple notices from the MAILER, but too late. Sorry for the inconvenience. Sean-Thomas _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: ACole65464 Subject: Looking for the Best Integrator Date: Tue, 28 Apr 1998 23:17:23 EDT To all the PSN electronic wizzards, I would like someone to help me by answering my questions about integrators, as my knowledge regarding electronics is limited. After experimenting with different electronic integrators, used for converting an acceleration signal to velocity, I am wondering if I have finally found the best type for this particular application. I found that using either the "classic" type integrator circuit or a single pole low pass filter always showed a slight instability of the signal trace. Now using a state variable filter (center freq set at .03 Hz for example) utilizing 3 op-amps, I can pick off the band pass signal for a velocity response (drops off @ -6dB above .03 Hz) or I can pick off the low pass signal for a displacement response (drops off at -12dB above .03 Hz). Either response is much more stable than using other single op-amp designs. Below are my two questions: 1. Is the state variable filter a better circuit to use for a more stable response? 2. What should the "Q" value be, assuming I should use this circuit? Any feedback to my questions would be appreciated. Allan Coleman, Edmonds, Washington _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: bschafer@......... Subject: quake detectors? Date: Wed, 29 Apr 1998 00:19:30 -0500 (CDT) I have heard, recently, of an alarm, much like a smoke detector, that can be installed in someone's home to warn them of an impending quake. I think that is gives something like 30 seconds warning before the quake strikes. If this is available, where would I be able to learn more about it? I am also wondering how it would differentia between the vibration of a truck going by as opposed to a very minor quake. Interesting. Bonnie the crafty crafter _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: prewar Subject: Re: more equalization noise Date: Wed, 29 Apr 1998 07:35:44 +0100 S-T Morrissey wrote: > > > I guess I wasn't clear enough. THe gain figure I mentioned is > estimated from figure 1 of the Riedesel et al paper., Hi Sean, It seems that we are talking at cross purposes, your article and the one I refered to by Peter Roberts!! So, I conclude that Robert's cct. was NOT the cause of the noise you refered to in your first comments, when I brought up the subject of Robert's cct., but was caused by Riedesel's designs. > Your rotating magnet is a nice idea to induce a 60-second > influence on your seismometer. Could you attach the magnet > to the minute hand of an analog clock? > One group has calibrated STS-1 sensors by gravitational > attraction, by rolling a bowling ball past them on a wooden > track. As I said, "The things we do for seismology" ........... ......I am going to mount magnet on a stepper motor whose rotational speed be controled by a pot. Then I will be able to examine ALL frequencies from say, 1Hz to .01Hz .etc. Trouble is, were do I get the time to do all these things ?. By the by, you mention in other letter about double posting, and you said, " from England"....I hope that I'm not doing anything wrong this end to cause you this trouble......I HAVE to reply to psn-l@.............. else none of my letters get thro' to PSN (who said,"A damned good thing too" ). Thanks for your time in clarifying the article mix up. Regards Albert Noble (England) > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Arie Verveer Subject: Live SDR Image WEB PAGE Date: Wed, 29 Apr 1998 07:45:57 +0800 Hi, For a limited time only! My web page now contains a "LIVE camera" image of the display screen running Larry's "SDR" program. My employer is running a open day this sunday with a the "Science Theme" I would be most interested in all comments, from "I hate it" to "I really hate it". Web Page http://iinet.net.au/~ajbv I hope it works!!! Arie _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "DAN SANSON" Subject: Re: Quake alert by e-mail to pager? Date: Tue, 28 Apr 1998 15:34:24 -0700 hello, my name Dan Swanson I am new to group. I have a computer pager program that lets you monitor any pager message it does not need to no the pager ID you can select the pager you want and have its messages re sent to you or as many as you pagers you want . it can also dial different paging companies. it also has a feature that allows you to have five triggered inputs N/C N/O .That will let you send a pre typed message, to alert you of the inputs being triggered. i.e. house alarm- fire alarm- door open- the message can be any length the system transmits POGSAG (over 90 percent of pagers) .alphanumeric message or just numerical message. YOU CAN EVEN MONITOR CAL BERKLEY EARLY SEISMIC PAGER WARRING and resound to you or as many pagers numbers you want to have called. if you want more info please contact Dan -----Original Message----- From: Charlie Rond +ADw-rond+AEA-ceri.memphis.edu+AD4- To: PSN-L Mailing List +ADw-psn-l+AEA-psn.quake.net+AD4- Date: Tuesday, April 28, 1998 12:52 AM Subject: Re: Quake alert by e-mail to pager? +AD4-Ted, +AD4- +AD4APg-A new version of EMON is in the works. One new feature is the ability to +AD4APg-execute a BAT file or command (EXE) when a quake file is saved. Since I +AD4APg-got this new pager that lets people page me via e-mail, I wondered if +AD4APg-there was any way to have a PC compose a message, dial a phone number, +AD4APg-send an e-mail message and hang up. I suppose it would have to use an +AD4APg-existing ISP account (like this one on netcom). This way the PC could +AD4APg-page me if a large quake was recorded. +AD4APg- +AD4APg-I'm sure someone has tried this before, so if you have ideas you want to +AD4APg-share with the group, or send me via private mail, I'd appreciate it. +AD4- +AD4-Exciting new features in your wonderful program+ACE- +AD4-The folks at CERI are paged in a similar way when a big quake occurs, or one +AD4-in the New Madrid area. I don't know the specifics of how the system is +AD4-triggered, but could find you a contact there if you want any additional info. +AD4- +AD4- +AD4- Charlie Rond -+AFw-/+AFw---Public Seismic Network-Memphis +AD4- rond+AEA-ceri.memphis.edu (901) 360-0302 +AFs- free BBS +AF0- +AD4- http://gandlf.ceri.memphis.edu/+AH4-rond http://gandlf.ceri.memphis.edu/+AH4-rond/psn +AD4- +AD4- +AD4AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXw BfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8AXwBfAF8- +AD4- +AD4-Public Seismic Network Mailing List (PSN-L) +AD4- +AD4-To leave this list email listserver+AEA-psn.quake.net with the body of the +AD4-message: leave PSN-L +AD4- _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Ted Blank Subject: Re: Quake alert by e-mail to pager? Date: Wed, 29 Apr 1998 06:51:32 -0700 (PDT) The latest version is always on the PSN web page, under Data Acquisition Software. Documentation is available separately or bundled with the code. Ted > > WOW! EMON sounds like a dandy program. > Where did you say one could get some more information? > > -- > ---/---- > Charles P. Watson > Seismo-Watch > > _____________________________________________________________________ > > Public Seismic Network Mailing List (PSN-L) > > To leave this list email listserver@............. with the body of the > message: leave PSN-L > _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: "Phil Giannini" Subject: GIF images Date: 29 Apr 1998 07:45:52 -0800 REGARDING GIF images Hi Larry, I sure like the "GIF Image" feature you have added to the recient files list on your web page. It saves a ton of time when comparing traces. You ALWAYS amaze us all with all your great work. Thanks, Phil SFN, SFZ _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Ted Blank Subject: Re: Quake alert by e-mail to pager? Date: Wed, 29 Apr 1998 08:34:36 -0700 (PDT) I always think new function is easy until Steve designs it right! :-) Sounds good. I found a pager program (cpager95) which bypasses the need to have an ISP and email account, it dials the toll free number for pager access via modem (most companies have one) and sends a text string to your pager. Numeric if you just have numeric pager. It worked on my skytel alphanumeric pager. You can check it out on the web at www.pager95.com. Dan Swanson: your post didn't tell us how to get in touch with you. Ted _____________________________________________________________________ Public Seismic Network Mailing List (PSN-L) From: Doug Crice Subject: Re: Looking for the Best Integrator Date: Wed, 29 Apr 1998 08:19:41 -0700 I have used the state-variable, three op-amp filter for years in exploration seismographs, and they are quite stable and dependable. If you set the damping to 0.5 instead of 0.7, then you can use all three outputs simultaneously (lowpass, highpass and bandpass) with the same gains. A damping of 0.5 gives you a little peak at the natural frequency (or if you prefer, a little ripple in the passband) which I guess makes it a Tchebyshev filter instead of a Butterworth. The way to make damping 0.5 is to make all the gain resistors the same value (and of course equal C's), so the design gets really easy and the corner frequency is 1/2piRC and the passband gain is 1 for all functions. If you use a dual pot for R, then you can adjust the frequency. Then, if you sum the lowpass and highpass outputs (with a classic op amp summing circuit), you get a nice stable notch filter. For those low frequencies, you would need some high-impedance op amps and probably some big, high-quality, non-polarized capacitors like they have in most surplus stores. However, it doesn't feel intuitively correct to call this an integrator, nor does this seem like a good way to make one. Integrators need a slope of 6 db/octave, not 12, and even though the bandpass fits that definition for some range of frequency, it seems like a simple op amp with a nice C in the feedback loop and some R to take care of drift would be a better choice. Most introductry op amp books have an integrator early in the text. -- Doug Crice web site http://www.georadar.com GeoRadar Inc. e-mail dcrice@............ 19623 Via Escuela