To: psn-l@............. From: shansen@........ (shansen) Subject: Seismometer turns into microbarograph Having discovered this group a few months ago, I figured it was time to put something into this forum. Building a seismometer has been one of those things I've been intending to do for a long time now. However, other priorities and interests (mostly in the area of vacuum apparatus) have persistently nudged out the moving earth detector. Anyway, I now have a PC-Labs card, the SDR program and most of the makings of a Shacklesford/Gunderson sensor. To keep things from moving along smoothly, a recent conversation with a prof friend of mine got me diverted toward building a microbarograph for detecting very slight changes in atmospheric pressure (microns of Hg pressure range) with periods of 1 to 10 seconds (1 to 0.1 Hz). Such microbarographs have been used to detect infrasound signals created by nuclear weapons tests, volcanic activity and meteorites. There are a wide variety of detectors ranging from the simple (appropriately housed loudspeakers) to fancier detectors such as capacitance diaphragm gauges, capacitor microphones, sensitive strain gauge-based pressure sensors, etc. These are operated in a differential mode with, for example, one side of the sensor connected to atmosphere and the other connected to a volume which, in turn, is connected to atmosphere by means of an adjustable, low conductance valve (like a needle valve). Thus, while one side of the sensor can respond quickly, the other has the "leaky volume" which gives the device a high pass characteristic. Very long duration changes, such as normal atmospheric pressure variations of meteorological origin, will not register. For my device, I'm using a low pressure differential Honeywell piezo transducer which has a sensitivity of 1 volt per inch of water (i.e. about 0.54 volts/Torr). I'm also trying a moving coil speaker. The volume is an old propane tank (barbeque grill size) and the leak is a cheap needle valve. Sensitivity for either should be adequate to produce a signal with a 1 micron Hg pressure variation. (For reference, I think that 1 inch in altitude corresponds to about 5 microns Hg pressure difference.) These specs are no wheres near what the microbarographs used for nuclear test ban compliance testing can achieve...something like 0.0001 micron Hg resolution from 0.02 to 5.0 Hz. But, my friend noted that his 1970s device at Penn State would detect strange pulsations, both periodic and aperiodic, with amplitudes of 5 to 20 microns Hg. He never pinned down the origins. So, to make a long story not too long, I was wondering if anyone on this list had toyed with this sort of device. I'd be happy to hear from folks with some real experience. And, if there's any curiosity, I'll be pleased to keep you informed of what comes of this. Then I've got that stack of stuff that still needs coaxing into the shape of a seismometer. Steve Hansen . Date: Wed, 3 Jul 1996 03:22:56 -0700 (PDT) From: David Josephson Subject: Re: Seismometer turns into microbarograph To: shansen On Tue, 2 Jul 1996, shansen wrote: > compliance testing can achieve...something like 0.0001 micron Hg resolution from > 0.02 to 5.0 Hz. But, my friend noted that his 1970s device at Penn State would > detect strange pulsations, both periodic and aperiodic, with amplitudes of > 5 to 20 microns Hg. He never pinned down the origins. > > So, to make a long story not too long, I was wondering if anyone on this list > had toyed with this sort of device. I'd be happy to hear from folks with some > real experience. And, if there's any curiosity, I'll be pleased to keep you > informed of what comes of this. I keep getting random calls from people who find out that I make capacitor microphones, and ask me for just this sort of thing. The catch is that capacitor microphones that go down to 0.5 Hz typically also go up to 18 kHz and cost $2,000 by the time you're done, because they are made to have 120+ dB dynamic range and function from -40 to +50C, have no drift or phase shift, etc. I have been thinking of building one for this meteorological infrasound, as a lot of people have been wondering about some of the barely-audible artifacts that seem to exist. The Taos Hum is one of these, and I will never forget the strange gurgling whistles I heard for most of a day in Wyoming, a few dB above the loudness of the blood in my ears, breakfast and the lizards scurrying around. So, please keep us posted. Microphones exist with a noise floor around -10 dB SPL, or about 3 microPascal. When I'm more awake I'll figure what that is in microns Hg. . Date: Wed, 03 Jul 96 07:28:33 cst From: "jmhannon" To: shansen@........ (shansen), psn-l@............. Subject: Re: Seismometer turns into microbarograph Steve, I have done some experiments with a Motorola pressure transducer similar to the Honeywell unit. A few years ago now there was an eclipse that went throught the midwest. Forrest Mimms was interested in people looking for atmospheric pressure effects from the eclipse. Mainly periodic ripples in the height of the earths atmosphere caused by the moon's shadow cooling off a portion of the atmosphere. I got the 0-5 psi differential transducer and sealed off one port. I had to seal the port just before the data run because the pressure was droping throughout the day and would have pegged out my A/D if it wasn't centered at the start. I think I saw the effect in the data but I never went back and processed the data to see for sure. To be used for continuous data collection some means of equalizing the long term pressure variations needs to be added to the sensor. A slow leak? I haven't figured out how to do this in such a manner as to have control over the low frequency cut off of the sensor. You could use some sort of computer controlled feedback I suppose. Supposedly you can see some earthquakes with this sort of sensor. Years ago in Popular Electronics the Carl and Jerry column described a tornado detector made from a goldfish bowl. This was a capacitor microphone that used the variations in capacitance to control an oscillator. Does anyone remember Carl and Jerry? Jim Hannon ______________________________ Reply Separator _________________________________ >So, to make a long story not too long, I was wondering if anyone on this list >had toyed with this sort of device. I'd be happy to hear from folks with some >real experience. And, if there's any curiosity, I'll be pleased to keep you >informed of what comes of this. >Then I've got that stack of stuff that still needs coaxing into the shape of a >seismometer. >Steve Hansen . Date: Thu, 04 Jul 96 01:55:51 -0500 From: Jim Esler Subject: Seismo-goodies to sell, Chart recorders and accelerometers For anyone wanting to home brew a seismograph or anyone interested in accelerometers :D I have a dozen Chart Recorders for sale. These chart recorders were to be used as Medtronic ECG recorders, but the project was cancelled before these particular recorders were installed. THEY ARE SPANKING BRAND NEW!! Each will come with: schematic, black anodize frame with paper holder/feeder, pen motor, approx +-2.5 dc mv signal for full scale in either direction, synchronous motor for driving paper @ 25 or 50 mm/sec, power supply (for +-15v dc to power pen motor) - in chassis, a 1 ohm 10 watt resistor to regulate the power to the pen heater. Medtronic paid over $700.00 each just for the pen motor, I will sell them for $100.00 each (I will cover shipping in the continental USA). I have used them in the past for building seismographs that have recorded quakes over 9,000 miles from my home in Minnesota (on the frozen tundra). If you replace the synchronous paper motor with a 1/3 rpm synchronous motor, the paper speed is reduced to 1 meter/hour. If you would like to learn how to build a great seismograph, let me know. I also have accelerometers for sale. These were to be used to modulate the paced rate of a pacemaker. These micro-machined accelerometers were made by IC Sensors Inc. I am selling them for $40.00 each (including shipping and documentation/schematics etc). Original cost = $108.00 each Specifications: Range +- 2 G Freq response 0-350 hz Mounted Resonant Freq Min 1500 HZ Sensitivity(mv/g) (min/Max) 1.75/3.0 Built-in overrange stops These are micromachined accels to provide a mass supported by four thin "bridges". Each bridge contains a piezoresistor to create a resistor bridge. As the accelerometer is moved the piezoresistor flexes, changing the impedance of the resistor bridge, An excellent amplifier to use with this accelerometer is the PGA204 made by Burr-Brown, as they document how to connect the PGA204 to a piezoresistor accelerometer. Each accel is housed in a surface mount chip carrier. Each pin has a 25/1000 inch spacing, I have been soldering wire-wrap wire to the surface mount pads. I also have a single ADXL05 to sell as well. (Scarce as hens teeth!) I am also about to build the Fraser-Smith reciever. For others about to do the same, you may want to call Burr-Brown and get the spec sheets on a filter chip they make, part number UAF42. With 2 or 3 of these parts you can build a VERY GOOD low pass filter. They will also send you a nifty program (free, but you must ask for it) for the PC which will graph the filter response, and tell you the size and placement of resistors/caps for a given filter type/characteristic. The UAF42 cost about $9.00 each. For an A to D, I will be using Advanced Microdevices AD7715 ($6-$12/each). It is a 16 bit A to D with a built in low frequency filter, and programable gains of 1,2,32 and 128 X. All this from a single +5v supply... Please call me or use snail-mail, as I very infrequently check my e-mail. Jim Esler 4639 Chatham Rd. Mpls., MN 55421 1-612-571-6372 . From: "The Allen Family" Date: Thu, 4 Jul 1996 10:27:59 -0400 In the late 70's I spent some time building a baraometer, purified the mercury from old mercury relays, played with glass tubing for days, learning to draw it, and shape it. I spent many weeks with a small teflon coated magnet floating on top of the mercury, and a Hall Effect sensor reading the effects of the magnet. t took months and luck to realize that what I had was a Seismometer. That led, of course, to other things, like seismometers, and low pass filters, and op amps. It was great, because at that time, I was teching in a large manufacturing plant, and I learned more about op amps, and the great amount of noise at the low end, then I ever did in class. The crowning thought, of course, was saying to myself, 'If it's that easy to build a seismometer by accident, then building one on purpose should be a piece of cake. I learned! Of course, I now have these two jars of mercury kicking around the garage. Just try to get rid of mercury. ---------- Date: Thu, 04 Jul 1996 23:59:28 -0700 From: Larry Cochrane Subject: Fireworks recorded with my Lehman Greetings, Every year the city of Redwood City puts on a 4 of July fireworks show. This years was the best I have ever seen. They shoot them off about 1/2 of a mile from me (using the 1 sec per 1000 feet rule used for lighting) and my Lehman recorded some of the bigger explosions. The event file can be downloaded from: ftp://psn.quake.net/quakes/9607/fireworks.lc1 This event file shows the last 5 minutes of the grand finale. At 4:53:34 UTC (or 9:53:34 local time) you will see the result of one of the largest firework shell I have ever seen. It almost reached +- 200 counts on my 12 bit A/D converter. When I was in high school, back in 1969 or 1970, I helped in a small way to product the 4 July fireworks display for the Rose Bowl in Pasadena Ca. It was very interesting to see, and help build, some of the displays. I remember being elected to drive with (I didn't drive) one of the two or three trucks with all of the fireworks for the display in them. The drive was from out in the desert, where they where made, to the Rose Bowl, about a 3 hour trip thru a lot of towns knowing quite well that if we got in an accident we could take a small block with us. All I can say is they sure have come a long way with the design of the firework shells. The grand finale was spectacular this year. Regards, Larry Cochrane RWC PSN . Date: Sat, 6 Jul 1996 14:16:19 -0400 From: shansen@........ (shansen) Subject: Seismometer turns into microbarograph - Part 2 I was pleased by the number of responses regarding the infrasound detector note I posted a few days ago. A couple things to add at this time: I've listed a couple Web resources on the Links and Resources page of my vacuum ("the Bell Jar") Website. The address for that page is: http://www.tiac.net/users/shansen/belljar/links/ It's toward the bottom of the list. One of these resources has a brief bibliography that includes some material on infrasound of meteoric and volcanic origin. I'm also putting together a larger bibliography based on what I've been able to pull out of the American Institute of Physics' PINET search utility and from a document on micromanometers that was published by the NBS (now NIST) in 1970. (W. G. Brombacher, "Survey of Micromanometers", NBS Monograph 114, 1970). A couple folks asked about the particulars of the design. Since I don't have it together yet I'm hesitant to get any more specific than I was in the previous note. When I do get some results, I'll place a description and diagram on my Website. The particular pressure sensor I'm planning to use (it's on order) is a Honeywell-Microswitch amplified differential sensor 163PC01D75. It's rated at +/- 2.5" water with a sensitivity of 1 volt/inch water. That's about 0.54 volt per torr. Honeywell's pressure sensor Website has details. It's at http://www.honeywell.sensing.com/prod_fam/pressure.htm. Steve . Date: 07 Jul 96 15:29:36 EDT From: Robert L Barns <75612.2635@..............> Subject: NYC area roster Hi, I finally got a reply to my request for amateur seismologists in the NYC area! Thanks to Larry Cochrane for his reply to my questions and the fascinating account of recording the fireworks. Bob The complete roster of amateur seismologists in the NYC area (all who have responded to my message of 17 June '96) -=-=- Nick & Sophie Caporossi, East Brunswick, NJ 08816 (7/4/96) nickcap@............. "Nothing in operation. Compiling info to build station. Will use computer data logging." -=-=- Bob (Robert L.) Barns, 63 Martins La., Berkeley Heights, NJ 07922 908-464-6785 'phone Mon. thru Sat. 9-12, 1-4 & 8-10P, Sun. 9-12 & 1-4P 75612.2635@.............. Cutting metal to build a Lehman. Will use computer data logging. Years in seis. 0.2. Amateur radio call KB2IKC -=-=- . Date: 09 Jul 96 09:50:09 EDT From: Robert L Barns <75612.2635@..............> Subject: reply to 7/5/96 Nick & Sophie Caporossi, FROM: Robert L Barns, 75612,2635 TO: Nick & Sophie Caporossi, INTERNET:nickcap@............. DATE: 7/9/96 10:06 AM Re: Reply Nick & Sophie, I agree that building a seismometer does not look difficult. I have a Lehman about half built--another 2 weeks may see it working. The Lehman was described very well in Sci Am mag. in July 1979 pp 152-161. I'm making a few modifications but would expect one built as described would work well. Your suggestion of getting the FAQ by Cochrane is a good one. I have it and it is useful. I got a brochure of the AS1 seismograph offered by The Amateur Seismologist for $450. It is a 3 sec vertical and includes the electronics and software for input to a printer port on a PC. I think that the price is reasonable but I think that 3 secs is a little too short. You can request the brochure from seismo@............ I have a couple of catalogs of used seismograph dealers. Their listings of long period sensors carry astronomical prices and the cheap ($15-$50) ones are very short period geophones. Bob Date: Wed, 10 Jul 1996 10:32:09 +1200 From: dann@........ (David A Nelson) Subject: nz quake hi all, well things are still moving down here in the deep south.... Mt Ruapehu is still pumping out heaps of ash in a renewed eruption seq. on Mon morning (local 080796) and White Island Volc. in the Bay of Plenty is also showing signs of warming up, and the earth is still quaking: U.T.: 1996 July 9 2136 NZST: 1996 July 10 9.36 a.m. Lat, Long: 41.90S 172.81E Location: Below St Arnaud Focal depth: 139 km Magnitude: 5.3 Felt in Picton, Blenheim, Motueka, Nelson. cheers Dave Dave A. Nelson Green Is., Dunedin, South Is.. New Zealand. Date: Wed, 10 Jul 1996 09:12:19 -0700 From: Jeff Batten Subject: Re: reply to 7/5/96 >Date: Tue, 09 Jul 1996 08:46:04 -0700 >To: Robert L Barns <75612.2635@..............> >From: Jeff Batten >Subject: Re: reply to 7/5/96 > >At 09:50 AM 7/9/96 EDT, you wrote: >>Nick & Sophie Caporossi, >>FROM: Robert L Barns, 75612,2635 >>TO: Nick & Sophie Caporossi, INTERNET:nickcap@............. >>DATE: 7/9/96 10:06 AM >>Re: Reply >>Nick & Sophie, >> I agree that building a seismometer does not look difficult. I have a Lehman >>about half built--another 2 weeks may see it working. The Lehman was described >>very well in Sci Am mag. in July 1979 pp 152-161. I'm making a few >>modifications but would expect one built as described would work well. >> Your suggestion of getting the FAQ by Cochrane is a good one. I have it and >>it is useful. >> I got a brochure of the AS1 seismograph offered by The Amateur Seismologist >>for $450. It is a 3 sec vertical and includes the electronics and software for >>input to a printer port on a PC. I think that the price is reasonable but I >>think that 3 secs is a little too short. You can request the brochure from >>seismo@............ >> I have a couple of catalogs of used seismograph dealers. Their listings of >>long period sensors carry astronomical prices and the cheap ($15-$50) ones are >>very short period geophones. >>Bob >> >> >> I have been beta testing the upgraded AS1 Vertical seismometer for a few months. As seismometers go, it is a order of magnitude better than a Lehman horizontal seismometer. The AS1 consists of a vertical sensor (3 sec) and a low pass filter. Bandwidth out to 20 sec. I have a Lehman logging data right next to the upgraded AS1. A lot of distant quakes I record with the AS1 do not even show up on the Lehman. With the AS1 I record distinct P, PcP, S, 20 sec surface waves, etc. A lot of distant quakes I record with the AS1 do not even show up on the Lehman. The AS1 inputs the data through the serial port. . Jeff . Date: 10 Jul 96 14:10:54 EDT From: Robert L Barns <75612.2635@..............> Subject: reply to Jeff Batten FROM: Robert L Barns, 75612,2635 TO: Jeff Batten, INTERNET:batten@............... DATE: 7/11/96 1:45 AM Re: Copy of: Reply to Jeff Batten Below is my reply to a note describing Jeff's very good experience with the Amateur Seismologist's (818-249-1759) AS1 vertical seismograph system. He said that the AS1 sees events which his Lehman, operating sidi-by-side, misses. Jeff, Thanks for your very interesting message about the AS 1. I think that many on the PSN list would be interested so you should consider posting your experiences there. Since I live on the East coast, I'm only interested in teleseismic events. I have a few questions: 1. What are the characteristics of your Lehman? Is it as described in the Sci Am article of July 1979? Period? Sensitivity? In principal, I see no reason that a vertical should be more sensitive than a horizontal. Clearly, the two types will see different motions. The P wave is a compressional wave and should have no vertical component. Since the time between the arrival of the P and S waves is a direct measure of the distance from the seismic event, I would think that detecting the P wave is highly desirable. The S wave is transverse and hence can be polarized. The vertical should not detect a horizontally polarized S wave. Both the Love and Rayleigh waves have both vertical and horizontal motions and should be detected by either type of sensor. I find, in Bolt's book (Earthquakes & Geological Discovery), no use for the Love and Rayleigh waves, although they are much stronger than the P and S waves and they are pretty to look at. When you say that you record distant quakes with the AS1 that don't show up on the Lehman, are you looking at only the Love and/or Rayleigh waves or are you referring to the P and S waves as well? The article on the Lehman shows very nice P, S, Love etc. waves from an earthquakes in Turkey and the Galapagos Is. recorded in Virginia. 3. Is it possible that the Lehman is missing some events because it is overdamped? 4. Since the Lehman has a dipole directional pattern and the vertical is omnidirectional, are the events missed by the Lehman due to this effect? 5. You log data 'through the serial port'. The brochure says that it uses the printer port (parallel)?? 6. I would like to write my own software. Is the info that comes with the AS1 software sufficient for this? 7. Could you send me a recording from the AS1 which shows (at least) the P and S waves? 8. Your e-mail address looks like you have something to do with the global positioning system. Is there a resonably priced GPS timing reciever available? If so, I'm sure that the PSN would like to hear about it. Bob Date: Thu, 11 Jul 1996 01:31:34 -0700 (PDT) From: Ken Navarre Subject: Re: reply to Jeff Batten On 10 Jul 1996, Robert L Barns wrote: > 7. Could you send me a recording from the AS1 which shows (at least) the P and > S waves? Or maybe... upload a seismogram to the PSN web site (maybe with the extension of .AS1) so we could all take a look with WINQUAKE. If the file format is compatible... Will WINQUAKE display a file saved by the AS-1 software??? I'd like a sensor that I could run from my laptop.I also want a compatible file format to what "the big kids" use so I can use the same software for all files. Ken Date: 11 Jul 1996 08:06:48 U From: "Phil Giannini" Subject: Orientation To: "The PSN" X-Mailer: Mail*Link SMTP-QM 3.0.2 content-length: 1154 REGARDING Orientation Hello Fellow Enthusiasts, My name is Phil Giannini and I live in San Francisco. For several years and more especially since the 1989 quake I have been interested in that phenomena we call an earthquake. Over the past several months, and with a ton of help from Larry Cochrane, I have finally managed to put together a monitoring station that has given me some very pleasing results. But, as all experimenters, there is always that need to improve the system. One of the sensors I have is a Lehman type device that is oriented in a north/south direction. I was wondering if anyone on the PSN knows if this instrument should be oriented to true north or magnetic north. In the bay area the difference between true and magnetic north is about 17 degrees which is not that much. However in some locations this difference can be quite large. Is there a standard practice when it comes to sensor orientation? Currently my instrument is oriented to magnetic north. Any comments will be appreciated. Thanks, Phil . Date: Thu, 11 Jul 1996 20:41:28 BST From: Peter Styles Subject: Re: Orientation To: Phil Giannini cc: The PSN Phil You should orient it to true North as magnetic North can be all over the shop depending on where you reside. This is not too difficult if you have a compass where you can set the magnetic variation so that the compass needle points to true North. 17 degrees which is actually quite large. Pete Styles Date: Thu, 11 Jul 1996 16:17:36 -0700 From: "Staniforth, Charles" Subject: home assembled seismograph anyone in Seattle area who has made own simple seismograph and willingt advise a neophyte. Date: Fri, 12 Jul 1996 21:59:57 -0700 (PDT) From: Ken Navarre Subject: Re: home assembled seismograph To: "Staniforth, Charles" cc: psn-l@............. > anyone in Seattle area who has made own simple seismograph and willingt > advise a neophyte. I'm not in Seattle, Charles. but it's pretty simple to build one. Another option is that you may be able to snatch the telemetry from the airwaves if you've got a scanner. Check out the FAQ for the frequencies to search. The mail list loaded with good folks with lots of different experiences building sensors. Ken Date: Fri, 12 Jul 1996 22:07:57 -0700 To: psn-l@............. From: Larry Cochrane Subject: Re: questions and A/D card Update Hi Everyone, At 04:23 PM 7/12/96 -0700, Steve Hammond wrote: > >Thanks Larry for the FAQ post. > >I need an update on the status of your data collection card. Funny you should ask... I was just getting ready to announce that the first board shipped yesterday! I have two more boards that will be shipped on Monday. The A/D board works great, I had no problems bringing up all of the boards (so far) and there are no jumpers or any part layout problems. I also updated my web page for the board at http://psn.quake.net/atod.html with some features / specs for the board and pricing information. >I have a >fellow in Belgium that I speak to from time to time, and so far we have >got his Lehman and a vertical running using two DR110 cards and Pete's amp >design (which I built and shipped to him) along with some coils I also >built... I know, now we have bad data being collected on both sides of the >world. Anyway-- Good old Boon called me kuz he still can't get an >Internet address in Belgium and wants to buy another data collection card. > >Boon doesn't have a lot of electonices knowledge, so I need to know how >much does he need to do (know) to get your card connected and working? All he has to do is be able to read English (or as well as I can write it) and read the setup.txt and SDR.DOC files that is included in the current SDR package (SDR16.ZIP). You may have to make a cable for him if he can't solder. The board has all of the inputs and outputs thru a 37 pin "D" type connector. Too connect the analog inputs to a preamp/filter board you need to solder to the mating connector (I supply a mating connector and plastic hood). >Will it work on a 386 using Eurp. 230V power. Sure, the card uses DC voltage from the PC power supply. As long as his PC runs on this power the card will work just fine. >The amps I built are running >on Ni-Cads and Boon swaps them out when they get low. What kind of power >requirements does your collection card have. The board plugs in into the PC and uses the PC power. It's not designed to run on batteries. I haven't measured the DC current on the board, I would guess its about the same as a PC-Labs card, under .5 amps for the +5 vdc and under 100ma for the +-12 vdc. > What is the total cost for the >one without the radio interface and the one with? 12 bit version: With WWV correction $195.00 without $180.00 16 bit version: $245.00 without $230.00 These prices are for the SDR version. I will also have a version that will only work with EMON. This board will not have any of the timing circuitry need by SDR so it will be a little cheaper. I can't support this version until Ted Blank makes a new version of EMON that will support my new card. >Do you have a BOM of all >the parts needed to connect it that I could look at? Check out the SDR.DOC file (http://psn.quake.net/software/sdr.txt) and the other text files in the SDR package. >What kind of a radio >setup would he need in Belgium to connect the card to a WWV receiver? I >have a simple world radio here. I'm not sure if he could use our WWV system at 5 10 and 15 Mhz, I think the reception would be to weak that far away. For a description of how the WWV option works read the WWVCORR.TXT file in the SDR package. Does anyone know if there is a WWV equivalent in Europe? >Would that work using the microphone jack? The earphone/speaker jack would work better ... The receiver should have digital type tuning so that the reception doesn't drift. Radio Shack has one digital portable model for under $100.00. >If so, how does your card react when the single gets weak as mine does >from time to time here at the house? The system can still keep fairly good time because when SDR is lock to WWV it determines how far the reference oscillator is off from the 4.0 Mhz. SDR compensates for the frequence error (of the crystal oscillator) by adding or subtracting time (in 1 MS steps) at a interval necessary to make the system look like the reference is at 4.0 Mhz. Now, when reception is lost SDR will use the last Add/Drop time interval until it locks up to WWV again. Depending on how long the reception is lost and the amount of temperature change in the room where the computer is keep the drift could be anywhere from 10Ms to over 100Ms per day. The temperature stabilized reference option (another $85.00 more) will keep the drift to a few milliseconds per day. Regards, Larry Cochrane Redwood City, PSN . Date: Sat, 13 Jul 1996 03:05:20 -0700 (PDT) From: david Subject: ULF Quake precursors - book review Over the years we've seen a lot of talk from people who claim to have secret recipes for eq prediction based on electric or electromagnetic precursors. Most of them are demonstrably acausal (like frequency shifts in navigation beacons) or at best highly speculative. People started to pay more attention after Tony Fraser-Smith's paper was published after the 1989 Loma Prieta quake; there have since then been at least three major conferences on the topic (one in Southern California in 1992, one in Japan in 93 and one at UC Berkeley last fall). A book has been published of the papers presented at the Japanese conference, "Electromagnetic Phenomena Related to Earthquake Prediction" and all the main players in this field were there, it seems (Fraser-Smith, Varotsos from Greece, Jack Dea from the Navy lab in San Diego, Van Bise and Rauscher, representatives from China, Japan, and lots of other places). There are enough serious papers presented from an experimental viewpoint to keep a hundred expert hobbyists busy, if no one duplicates anyone elses's work! There seem to be two major methods: signals generally noise-like in nature (but sometimes quite narrow bandwidth) that increase due to signals generated by rock stresses or other phenomena related to that, and perturbations in flow of other signals through seismically active areas. There are survey papers predicting the noise signals generated by various rock fractures, and papers showing what sort of E and H fields would be generated at different distances from those signals flowing in a conductive body. There are papers proposing methods where these signals and fields are modified and propagated by various ionospheric conditions, and papers simply reporting on the methods used by various researchers -- generally in quite enough detail that a competent experimenter could duplicate the work. The work that seemed the most surprising to me was that reported by Van Bise and Rauscher. The main shortcoming of most of the EM field work reported before this has been the extremely sparse data set. Fraser-Smith was working on something else entirely, and was only interested in overall noise level from 0.01 to 10 Hz, more or less. OK - he saw an increase in this band but what was it? Van Bise and Rauscher have been doing the logical thing of sampling the entire electromagnetic (initially E and H fields, now primarily orthogonal components of H) spectrum from 0.01 to 100 Hz or more (up to 50 kHz). There are discrete peaks in this spectrum associated, they claim, with specific areas of seismic and ionospheric disturbances. Within the 3 to 4 Hz window, for instance, they find gradually increasing levels (up to 20 dB higher than the background spectrum) for several weeks before a quake, then the signals abruptly stop 12 to 72 hours before the event. Within the Pacific coast region, they claim that the center of this signal energy corresponds to the latitude of the epicenter! The paper reads, "As noted above we have found a dominant signal of 3.2 Hz in the Portland area, 3.4 to 3.6 Hz in the San Francisco Bay area, 3.8 Hz in the Los Angeles area, and a 3.88 Hz signal in the San Diego area." They say that 300 km is the limit of their detection range. It's hard to nail down the intensity of these signals from the data presented, but it looks like around 0.1 nT rms, which is certainly detectable with a good coil system and easily with a low noise fluxgate or optically pumped magnetometer. The book is available from the publisher, Terra Scientific, in Tokyo. It's about $100 for 677 pages hardbound. I mailed them a request with my credit card info and the book arrived a month later. Their address is Terrapub, 302 Jiyugaoka Komatsu Building, 24-17, Midorigaoka 2-chome, Meguro-ku, Tokyo 152 Japan. Fax +81-3-3718-4406. . Date: Sat, 13 Jul 1996 13:49:17 -0400 From: shansen@........ (shansen) Subject: Seismometer turns into microbarograph - Part 3 Here's a quick update on the microbarograph. I haven't received the Honeywell piezo pressure transducer yet. However, I did acquire an MKS differential capacitance manometer, Type 223. The particular unit has a full scale of 10 Torr. This corresponds to an output of 5 volts. The pressure/voltage relationship is linear, i.e. 1 Torr is 0.5 volts, 0.1 Torr is 0.05 volts, etc. The usable resolution is 1/10,000 which means that it is good to 0.2 microns Hg. To quickly try it out, I just connected a 2' piece of 3/16" id automotive neoprene tubing to the reference side of the transducer. I terminated this with a 6" length of 0.005" id capillary tubing. Air slowly leaking into this tube will equalize long term pressure changes, resulting what is effectively a high pass filter. (In the permanent setup, this will be replaced by the tank and needle valve. The adjustable needle valve will permit the time constant of the chamber to be set to any appropriate value.) While I haven't measured it, the time constant seems to be in excess of 10 minutes. The measurement side of the transducer is just open to the atmosphere. Right now I have the transducer on a shelf near an open sliding door. For the temporary readout, I have an MKS display with a graphical output. The time resolution is limited to about 100 milliseconds by the display software. What I am seeing is as follows (based on all of 3 hours of observation): First there are short period variations (0.1 to 1 sec) of about 10 microns peak to peak. Sometimes this seems random, at other times it has appeared to be more periodic. Next there are longer period variations which are sort of periodic. The period is about 3 minutes and the total pressure change is on the order of 20 to 30 microns. Through most of the morning we've had the remnants of Bertha rolling through. Before and during there were wider ranging pressure fluctuations (up to 100 microns) with a period of 10 minutes or so. Things are quieting down now and this effect is much less pronounced. I'm in the process of getting a 1 Torr full scale capacitance manometer of the same type. That will improve the absolute resolution by another order of magnitude. Next steps are to interface the transducer to the PC Labs card (at the moment still in the anti-stat bag) and try the SDR software. I've also got the tank and its valving done. I have to install it in a thermally insulated container. Sometime in the next couple weeks I should have the Honeywell transducer. I don't expect anywheres near the performance but it will be interesting to compare the devices. On a totally different note, Shawn Carlson had an interesting article in this month's Amateur Scientist in "Scientific American." It involved the detection of ~micron motions using Hall transducers. At the end he mentioned a seismograph using a suspended magnet with a Hall device to detect the motion of the magnet. I remember Forrest Mims III mentioning this sort of device a few years ago in either "Radio-Electronics" or "Science Probe!" Has anybody tried this? How about long term drifts due to temperature changes? Steve Date: Sun, 14 Jul 1996 08:54:57 -0400 From: shansen@........ (shansen) Subject: Alternative Detectors At the end of my note of yesterday which gave an update on my microbarograph activities, I threw out the following on Hall effect devices as seismic detectors (I repeat it for those who don't care about microbarographs and might not have gotten to the end of the note): "On a totally different note, Shawn Carlson had an interesting article in this month's Amateur Scientist in "Scientific American." It involved the detection of ~micron motions using Hall transducers. At the end he mentioned a seismograph using a suspended magnet with a Hall device to detect the motion of the magnet. I remember Forrest Mims III mentioning this sort of device a few years ago in either "Radio-Electronics" or "Science Probe!" Has anybody tried this? How about long term drifts due to temperature changes?" I took a poke through my junk pile as I remembered that I had some Hall devices. What I have are a small quantity of UGN 3503U's. I believe these are (were?) made by Sprague Electric. The device, which incorporates an amplifier with the Hall device, has a linear output (i.e. it's not a Hall switch). There are 3 pins: I assume they are power, common and output. Does anyone have any information on this part and how to connect it? I believe one of the above mentioned magazines discussed this device, perhaps 10 years ago. While rummaging through the junk, I dug out a Radio Shack piezo speaker (cat. 273-091B). This consists of a round ceramic element within a plastic housing. I glued a #8 flat head screw to the center of the element (the head of the screw just fit through the aperture) and mounted it with a weight fairly rigidly suspended from the screw (hanging down of course). Connected directly to my PCL-711S card (using the PC Scope software) it turned out to be quite sensitive to house vibrations, walking, etc. As a potential tends to build up across the electrodes, I stuck a resistor of a few 10's of k across the device to keep the trace centered. Anyone have any experience with this sort of approach? Steve . Date: Sun, 14 Jul 1996 22:18:08 -0700 From: Larry Cochrane Subject: SDR user mailing list and new SDR version Hello, I created a new mailing list on my system for SDR users. The list name is SDRUSER. If you are using SDR, or will be using it soon, please drop me a note and I will place you on the list. You can also subscribe by sending a message to listserver@............. with the body of the message: subscribe SDRUSER I added several members and sent this to the SDRUSER list the other day: I have setup another mailing list just for SDR users. If you get this message you are already on it. The list name is SDRUSER and you can send message to the list by sending email to sdruser@.............. I added some of the PSN-L users who I know are running SDR now with a PC-Labs card, or, if you ordered one of my A/D cards and will be running SDR soon. I will be using this list to announce new releases of SDR and also sending them to you as a MIME attachment. You can use this list to report problems and make requests for new features. If several people like a new feature that someone suggested I will put that request at the top of the list. I will also be using this list to start a FAQ for SDR. If you would like to leave this list send a message to me or to the listserver@............. with the body of the message: LEAVE SDRUSER or UNSUBSCRIBE SDRUSER. I have a new version of SDR ready. This version 1.6 has several important features: This version checks for low disk space every 30 minutes. If the free disk space is less then 8 Megabytes per channel, about 1 1/2 days worth of recorded data per channel, a Low Disk Error message will be displayed on the top left side of the display and the alarm will start to beep. If the disk space gets below 100K bytes SDR will no longer save any data to disk. The upper left hand corner now displays the free space next to the H=Help message. I added this test because of a problem Ken Navarre had when he upgraded to 1.5. He created a new directory for SDR and forgot to erase the record files in the old SDR directory. This caused is hard disk to fill up. He then had some problems with is hard disk. So to prevent this from happening I added the test for low disk space and to stop saving data if it gets to low. Another "feature" I added to SDR is to check that you are running the right tsr version. If you see the message Error: Wrong TSR version... This means you are not running the TSRxxx.EXE that came with the current SDR.EXE program. Note: this test does not check to see if you are running the right tsr with the A/D board you are using. There is no way for me to test to see if the board is mine or a PC-Labs 711s. The boards are NOT compatible so you must run the right tsr. I added this test because of an early 1.5 release where I forgot to include the PC-Labs tsr (oops! This version now has both versions). What some of you did is use the the one that came with 1.4. Luckily here was no problems doing this, this time. I added a check for the proper version number for both SDR and the tsrxxx because the data structures used to transmit data between sdr and the tsr may change between releases. If one where to run an old tsr with a newer sdr and the data structure changes who knows what could happen. You can download SDR16.ZIP from my Web page (http://psn.quake.net/software/sdr16.zip) or using FTP. That's it. Please give this version a try and let me know if you have any problems. Regards, Larry Cochrane Redwood City PSN Date: Wed, 17 Jul 1996 07:41:03 -0700 From: Ray Gilman X-Mailer: Mozilla 2.0 (Win16; I) Subject: Seismometer Damping I am considering an electro-magnetic approach to damping the natural oscillations of my seismometer but so far have not been able to generate enough force so far. Can someone give me advice on electromagnetic damping , i.e. what orientation, materials , field strength etc. From: Ray Gilman To: psn-l@............. Subject: Printed Circuit Fabrication I have recently aquired the equipment and experience to fabricate printed circuit boards. I would be willing to fabricate small prototype runs for barter or cash. I can do double sided through hole or smd. I cannot do flow through holes. I can also do board layout from schematic, for a bit more though. Date: Thu, 18 Jul 1996 22:43:17 -0700 From: Larry Cochrane Subject: Re: AS1 Ad Greetings, I thought I would add my .02 cents on this ad. This weekend I sold a 16 bit A/D card and Amplifier/Filter board to someone who has a system from them. Roger told me that the system as only an 8 bit A/D converter chip. His system is several years old, so maybe they have upgraded to a better A/D chip. If the system still uses (Jeff let me know if this is wrong) an 8 bit chip then I would stay away from this system. IMHO an 8 bit A/D chip does not have the dynamic range need to record quakes without running into saturation problems. My 16 bit A/D card has over 100 times more dynamic range then a 8 bit chip. The 12 bit version has 16 times more dynamic range. Roger plans to use the sensor from the package but will be replacing the electronics. The other thing I would point out, or ask, is what kind of software do you get with the system. Can the event files be analyzed with some software like QuakeVw or my WinQuake. Does the software have the ability to determine distant using P and S markers with travel-time tables? How about magnitude and spectral analysis using FFT (WinQuake and do both)? And, is the file format of the event files made public like the PSN format? Now, I do see a system like this being good for someone who is looking for a turn key system. But part of the "fun" of this hobby is making the sensor yourself. I find it very rewarding that I can make something sensitive enough (and I'm NOT very mechanically inclined) to pick up events anywhere on the earth. Just my .02 cents.... Regards, Larry Cochrane Redwood City, Cal, PSN At 09:16 AM 7/18/96 -0700, Jeff wrote: > >Here is the Ad for the AS1 > >> THE AMATEUR SEISMOLOGIST >> 2155 Verdugo Blvd. #528 >> Montrose, CA 91020 >> 818-249-1759 >> seismo@............ >> >> > Add deleted . Date: Fri, 19 Jul 1996 13:29:44 -0700 From: Jeff Batten (by way of Larry Cochrane ) Subject: Re: AS1 Ad X-Info: Evaluation version at sys5.webtronics.com X-Info: Public Seismic Network Mailing List X-ListMember: psnarchive@............. [psn-l@.................... I received this today. I think Jeff meant to send it to the list. I will be making a responce later today. Larry --------------------------------------------------------------------------- Bits, Bits and More Bits. Just my .05 cents worth. Larry wrote: >I thought I would add my .02 cents on this ad. This weekend I sold a 16 bit >A/D card and Amplifier/Filter board to someone who has a system from them. >Roger told me that the system as only an 8 bit A/D converter chip. His >system is several years old, so maybe they have upgraded to a better A/D >chip. If the system still uses (Jeff let me know if this is wrong) an 8 bit >chip then I would stay away from this system. IMHO an 8 bit A/D chip does >not have the dynamic range need to record quakes without running into >saturation problems. What limits the dynamic range of a Lehman horizontal seismometer is the micro seismic 6 second background ( or noise in your amp) , and the power supply rails of your amplifier. This dynamic range is determined by the sensor and amp. Not the number of bits. The AS1 uses a vertical sensor and records events much better than Lehman horizontals. For sensors (Lehmans) used by amateurs 8-10 bits is all that is really needed. If you are using a 10K broadband sensor than more bits is better. The number of bits only determines the resolution for a given voltage span. The AS1 has no problems with dynamic range. It records local quakes and teleseismic events just fine. Check out the Web Site. http://www1.primenet.com/~seismo/ > My 16 bit A/D card has over 100 times more dynamic >range then a 8 bit chip. 100 ??? 16 bit A to D. 65536/256 = 256 Better resolution > The 12 bit version has 16 times more dynamic range. Resolution 4096/256=16 OK >Roger plans to use the sensor from the package but will be replacing the >electronics. ..> Roger is using the obsolete horizontal system. He should get himself the AS1 Vertical. : ) >The other thing I would point out, or ask, is what kind of software do you >get with the system. Data logging and Seismogram plotting. Jeff . Date: Fri, 19 Jul 1996 21:54:57 -0700 To: psn-l@............. From: Larry Cochrane Subject: More on bits and dynamic range of systems. Hi, Jeff Wrote: >What limits the dynamic range of a Lehman horizontal seismometer is the >micro seismic 6 second background ( or noise in your amp) , and the power >supply rails of your amplifier. >This dynamic range is determined by the sensor and amp. Not the number of >bits. and >For sensors (Lehmans) used by amateurs 8-10 bits is all that is really >needed. If you are using a 10K broadband sensor than more bits is better. >The number of bits only determines the resolution for a given voltage span. Sorry Jeff, but in all do respect this is simply NOT TRUE! Some basics: Lets start with what dynamic range means. To keep it simple dynamic range of a A/D system is the lowest signal to the highest signal a system can handle. Dynamic range is usually measured in Decibels or db. To calculate db you would use the following formula: nDB = 20 log Eo/Ei where nDB the number of decibels corresponding to the voltage gain or loss, Eo = voltage output and Ei = voltage input. If we take a look at my Amp/Filter card the Eo = +- 10 or 20 volts and the Ei is around 1 millivolt (this is the amount of background noise you get if you short the input). This gives a dynamic range of around 86db Eo = 20 /Ei = .001 = 20000, Log 20000 = 4.3 * 20 = 86db Earthquake Dynamic Range: Earthquakes have huge dynamic ranges from small ones under 1ML to the big ones that the ML scale can't even handle. For each ML there is 10 times more amplitude (~32 times more energy). The difference between a ML of 2 to ML 6 would be the equivalent of a voltage gain of 10000 (10*10*10*10) or 80db. ML 1 to a 7 would be 120db (I think the ML scale bottoms out in the high 6's). Sensor Dynamic Range: This is a hard one. It depends on a lot of things like how well the sensor is made (internal noise), the sensor location, and what type of events you will be receiving. The sensor location has to do with the local ground noise and the micro seismic 6 second background noise. The type of events also depends on your location. If you are "lucky" to live near active faulting you will be seeing the full range of events, from the little ones to the BIG ONES. If you are receiving teleseismic (far) events only you probably will not be seeing large dynamic ranges. On my Lehman the background noise is a round 20 microvolts (using a 6 pole 10 hz low pass filter). As an experiment I connected my o-scope across the pickup coil (a 240vac relay coil) and moved the boom trying to simulate a large local quake and was able to generate about +- 200 millivolts. This would give a dynamic range around 86db. I'm sure it would be higher if I didn't live in a very noise location. Amplifier / Low-pass filter Noise: This would be the electronics that connects between the pickup coil (or other type of sensor) and the A/D converter. This is a little easier to measure. Short out the input and look at the background noise at the output. Then apply a signal to the board and find out how much voltage swing you can get out of it before it saturates. As I calculated above, my board has a dynamic range of about 86db. Any amp/filter board with a very low noise input amplifier should be able to get this type of dynamic range. My board uses a LT1007 with a noise figure of 60 nanavolts p-p between .1 and 10hz. The Analog To Digital Converter: Like the Amp/Filter electronics it is easy to calculate the dynamic range of a converter based on the number of bits. To calculate the dynamic range of a A/D convert you would use the following formula: Total Counts / background noise counts = Counts, then Log counts * 20 # of bits Total counts Dynamic Range Voltage resolution (note 1) -------------------------------------------------------------------------- 8 256 48db 39.0 millivolts 10 1024 60db 9.76 " 12 4096 72db 2.44 " 14 16384 84db 610.3 microvolts 16 65536 96db 152.6 " 20 1048576 120db 9.5 " Notes: 1) Based on a +- 5 volt (10 volt) maximum input for full output count. 2) Background noise count for 8 to 12 bits usually would be 1. On higher bit counts chips the background noise of the A/D chip should be considered. Over All System Dynamic Range: Just like a chain is no better then the weakest link, the same can be said for any A/D system. If you have a sensor that only as, lets say 60db of dynamic range, and the amplifier and A/D converter has 80db range then the overall dynamic range will be set by the sensor or 60 db, you will be losing 20db. If the sensor and the amp have a 86 db range and the A/D converter only has 48db (8 bit system) range then overall system dynamic range will be set by the A/D converter or 48db. You will be losing 36db of dynamic range. What all this means: What all this boils down to is how well your system will pick up events. If you have a system with low dynamic range, your system will saturate very quickly with even a moderate local event, if you have the gain turned up to pick up teleseismic. If you have the gain set lower so the moderate events don't saturate, you will be missing the teleseismic events. With higher dynamic range systems you will be able to pick up both type of events with out any problems. I wrote: >> My 16 bit A/D card has over 100 times more dynamic >>range then a 8 bit chip. Jeff Wrote: >100 >??? 16 bit A to D. 65536/256 = 256 The 256 is theoretical, in practice the 16 bit chip I use has a internal noise of +- 3 max counts (not bits). The 16 bit version has about 10 times the dynamic range over the 12 bit version. That's it. I know we have a lot of engineers and PHD's out there. If I made any mistakes on my math, or other statements, please feel free to correct me. Regards, Larry Cochrane Redwood City, Ca USA PSN Date: Sun, 21 Jul 1996 14:14:08 -0400 From: Tinkler@....... Subject: Electronic Seismometer Has anyone built the seismometer that was in the Amatuer Scientist colume of Scientific America, that used radio freqs. and a short pendulum. Looks to be a real easy as well as a real small unit to build. Just getting into this, Have always waned to build a seismometer, but thought I was about the only one around until I found this Htlm. Any Help would be great, thanks Dave tinkler@....... . Date: Sun, 21 Jul 1996 21:59:08 -0700 To: psn-l@............. From: Larry Cochrane Subject: Re: Electronic Seismometer Hi Dave and group, If you mean the Shackleford-Gundersen seismometer (I call it a SG sensor because I can't pronounce or remember there names) from the Sept 1975 SA article I have one running. All of my event files on my system with the file extension of .lc3 are from my SG sensor. I really like it for detecting teleseismic (far) events. It has a very good low frequency response to pick up the long period waves or phases. My sensor has a pretty sharp cutoff at ~2 hz so it doesn't do very well for local events, but, this also filters out the local ground noise so I can pick up the teleseismic events better. I have a Web page for the sensor at http://psn.quake.net/sgsensor.html. The only problem with the article is the electronics. I changed the op-amps to LF412's so I could re-scale the resistors and capacitors to eliminate the need to tilt the sensor to bias the electronics. When I get some time I will sit down a draw up a schematic of my changes to the circuit and place it on the PSN home page. Just to let you, and other members of the list, know what type of sensor corresponds to the file names, here they are: File Name Sensor *.LC1 -> My N-S Lehman. *.LC2 -> Short period vertical sensor (Geophone type). *.LC3 -> My SG sensor. *.CGP, *.JSA, *.CMM, *.JCP *.JSB are all USGS sensor that I receive using my telemetry board (http://psn.quake.net/telebrd.html). Regards, Larry Cochrane Redwood City, Ca PSN At 02:14 PM 7/21/96 -0400, Dave wrote: >Has anyone built the seismometer that was in the Amatuer Scientist colume of >Scientific America, that used radio freqs. and a short pendulum. Looks to be >a real easy as well as a real small unit to build. > >Just getting into this, Have always waned to build a seismometer, but thought >I was about the only one around until I found this Htlm. > > Any Help would be great, >thanks Dave > tinkler@....... > . Date: 22 Jul 96 10:38:46 EDT From: Robert L Barns <75612.2635@..............> To: PSN mail list Subject: Finding old books Hi, As a beginner, I thought that it would be helpful to have a copy of CF Richter's 'Elementary Seismology" (1958). Our local library got it for me on interlibrary loan within about 5 days. I also tried some local used book dealers with no luck except that one of them suggested calling 516-734-5650. I did and told them (Peter Hennessey Booksellers) what I wanted. They look around and then send a description and a price--no obligation if you don't want it. The quote arrived yesterday. Hardcover, very good condition, newly rebound, $175 (that's one hundred seventy five)! I don't want one that bad, but if you have one, treasure it. I found the one on loan not very interesting, incidentally. For no known reason, another book accompanied the loaner. It is 'Anatomy of Seismograms' by Ota Kulhanek, Elsevier, 1990. 'The primary goal of this book is to present in a rather tutorial form all the necessary information and techniques pertinent to essential seismogram interpretation.' It looks quite useful. Barnes (why can't they spell) & Noble gave me a quote of $127 for this. I'll wait until I have some traces to interpret. Bob Barns Berk. Hts. NJ 908-464-6785 . Date: Mon, 22 Jul 1996 16:46:53 -0400 From: 73140.162@.............. Greetings, Just located your group, am sure I'm in the right place. Finally, I can talk to others with similar interests as mine without getting that "you are weird" statement. Coming from the world of criminal justice (did you know that there is a market for a seismic sensor which can tell the difference between wind or heavy vehicles and the signature of someone climbing or cutting on a chain link fence surrounding a prison AND work reliably in the real world), I am cramming to get up to speed on measuring quakes. In reference to the Aug. Amateur Scientist article on sub-micron movement and hets, I have the data book from Honeywell and the ss9 series of sensors are a new improved line with better temperature stability then older sensors. The data book also describes how to use the sensor in a bipolar arrangement for increased sensitivity of linear motion. In effect, you could have a sensor centered between two rare earth magnets mounted approx. 1cm apart or less. As the sensor moves back and forth between the two magnets, it's output would vary by several volts depending upon supply voltage. A very small gap between magnets would allow you to measure very small changes in position, probably well under a micron before alot of amplification. This could make a VERY sensitive linear direction sensor for some type of siesmograph. For Steve Hansen, I am very interested in the microbarograph, especially using a speaker as the sensor, please tell me more. I want to mount a Hall Effect Transducer on the cone and use it to measure voice coil travel. Are there other ways? Give me the details if you please. Ben Date: Mon, 22 Jul 1996 13:53:48 -0800 From: lahr@................. (Dr. John Lahr (USGS-GI Seismologist)) Subject: Re: ALTERNATE DETECTORS/HALL EFFECT TRANSDUCERS Speaking of criminal justice uses of seismology, as Ben just did, I though some might be interested in the use of seismic techniques (modified to record sounds) to locate gunshots. A small network was run near East Palo Alto in 1992/93 and demonstrated the feasibility of doing this with inexpensive hardware. You can read about this at: http://quake.wr.usgs.gov/QUAKES/FactSheets/Gunshots/ JCLahr ################################# John C. Lahr ################################ U.S. Geological Survey ############################### c/o Geophysical Institute ############################## 903 Koyukuk Drive ########################################################### ########################################################### P.O. Box 757320 ################################ Fairbanks, AK 99775 ################################# Phone: (907) 474-7997 ################################## Fax: (907) 474-5618 ################################### lahr@................. #################################### Date: Mon, 22 Jul 1996 22:05:37 -0400 To: psn-l@............. From: shansen@........ (shansen) Subject: Re: ALTERNATE DETECTORS/HALL EFFECT TRANSDUCERS/Microbarograph Response to Ben (??) concerning Hall devices and the microbarograph plus a general update on the microbarograph: I looked up Honeywell Microswitch's info on their SS94A1F analog Hall device and captured the data and applications sheets. Thanks. I did get some info from Allegro who has assumed Sprague's line of Hall devices (the ones I have in my mits). No data on the Sprague part and their new one requires a min order of a few hundred parts with a 26 to 32 week lead time. So much for that. The Hall device should make a very nice position sensor for either seismographs or microbarograph diaphragms. At some point in the near future I will get one. Ben wrote: "For Steve Hansen, I am very interested in the microbarograph, especially using a speaker as the sensor, please tell me more. I want to mount a Hall Effect Transducer on the cone and use it to measure voice coil travel. Are there other ways? Give me the details if you please." I have sidelined the speaker sensor for the moment as the sensitivity at low freqs using the output of the voice coil would be quite poor. I am starting with a differential capacitance manometer (MKS type 223) which uses capacitance sensing to determine the displacement of a tensioned inconel diaphragm. The unit I have is a 10 Torr full scale transducer. It has a senstivity of 1 milliTorr. By tomorrow I should have a 200 milliTorr unit in hand. This has a sensitivity of 0.02 milliTorr. The setup is simple: the measurement side goes to atmosphere, the reference side goes to a small (few cc) volume to which may be attached a number of hypodermic tubes of various diameters. This provides a controllable time constant on the reference side, thereby acting as a high pass filter. With a larger needle I can resolve very fine fluctuations. The smaller bores are useful for looking at longer period fluctuations (the former are on the order of a few milliTorr, the latter on the order of tens of milliTorr). I had tried a larger volume (a few liters) but the temp effects were too great. In fact, just touching the cylinder lightly would result in large pressure shifts. Interestingly, one would expect that a warm hand on the cylinder would cause the pressure to rise - Boyle's law as you heat the air. However, the first thing to happen is the cylinder expands a bit before the air heats. Thus the pressure drops very quickly by 10s of milliTorr. Interesting effect. One way to get around this is to bury the reference reservoir. I'm not ready to do that yet but the small reference is getting a good layer of insulation. A speaker cone with a Hall sensor, one side as the reference and the other side to atmosphere, should work. Once I have the very precise capacitance device fully functioning I will try other approaches for comparison. In addition to the capacitance manometer, I also have the Honeywell sensor in hand. With lower sensitivity than the cm, it might be useful for measuring the longer period, larger pressure changes. Finally, I tried my PC Labs 711 with the supplied Labtech Acquire program. Didn't work. I've sent a note to Advantech but has anyone in this group experienced problems? (the error code was 1012...the book just says to call tech support). Steve From: "Raul J. Alvarez" Date: Mon, 22 Jul 96 08:47:28 -700 To: psn-l@............. Subject: Laser Interferometer for Seismometer Sensor Hi Folks! My name is Raul Alvarez, and I have been reading the mail here for a few months to get the feel of the group. It seem most of the frequent writers are doing really interesting home building of equipment. GREAT!! I love it... My purpose in writing is to see if anyone has contemplated using a laser interferometer to measure the displacement of a seismometer pendulum. As an experimenter, I know that sometimes there are easier ways of doing things, and probably this is not the way, but it really sounds interesting anyway. This method would measure displacement and not velocity. Any ideas?? Briefly, I planned to use a solid state laser (diode) from all too available 1X CD Rom players or discarded CD players and a basic gate seismometer. Your comments are invited. Thanks for your time. This is a very interesting group. Raul From: Ken Navarre Subject: Re: ALTERNATE DETECTORS/HALL EFFECT TRANSDUCERS On Mon, 22 Jul 1996, John Lahr wrote: > Speaking of criminal justice uses of seismology, as Ben just did, I though some > might be interested in the use of seismic techniques (modified to record sounds) > to locate gunshots. A small network was run near East Palo Alto in 1992/93 and > demonstrated the feasibility of doing this with inexpensive hardware. Yup. And as a followup... The Redwood City Police Dept. has the system up and running. No real hits as yet but the testing period show promise. Ken . Date: Tue, 23 Jul 1996 09:20:12 -0400 To: psn-l@............. From: shansen@........ (shansen) Subject: Laser Interferometer for Seismometer Sensor To Raul Alvarez: You wrote: "My purpose in writing is to see if anyone has contemplated using a laser interferometer to measure the displacement of a seismometer pendulum. As an experimenter, I know that sometimes there are easier ways of doing things, and probably this is not the way, but it really sounds interesting anyway." Such a device was described in Review of Scientific Instruments, Vol 64, No. 5 (May 1993), pages 1337-1346. (Highly sensitive wideband seismometer using a laser interferometer, Akito Araya, Keita Kawabe, et al.) The stated bandpass is dc to 2 kHz. It consists of a HeNe laser Michelson interferometer with a mirror on a suspended copper mass (3" cube). The suspension is 4 ea. 0.1mm dia tungsten wires. A mirror is on one face of the cube and a damping system is located at the opposite end. I passed along a copy of the article to Larry. Having more knowledge of seismometers than I, he might have some addl comments. The same issue has the article "Active control of a balanced two-stage pendulum vibration isolation system and its application to laser interferometric gravity wave detectors." (P.J. Veitch et al., pages 1330-1336). Steve . Date: Tue, 23 Jul 1996 11:19:04 -0500 To: psn-l@............. From: mgray1@................ (Mike Gray) Subject: Laser Interferometer for Seismometer Sensor I don't have the title of the book handy, but it goes something like "Designing and building earthquake and weather sensors". It has a couple of designs for laser interferometers using fiber optic cable and laser diodes. I'll get the exact title and author tonight and post it if no one else comes up with it in the meantime. Mike Gray mgray1@................ . Date: Tue, 23 Jul 1996 12:29:01 -0400 To: psn-l@............. From: shansen@........ (shansen) Subject: Laser Interferometer for Seismometer Sensor Gordon McComb's book "Gadgeteer's Goldmine (55 Space-Age Projects)" has a chapter on a laser fiberoptic seismometer. This design uses 15' of fiber, a laser diode and a phototransistor. The author states that it will pick up local noise, thumps on the base, etc. but there is nothing to indicate how sensitive it is, what the response is, or if it ever actually picked up seismic activity. Unfortunately, as is the case with most hobby books, there are no references. It's an interesting approach and I'd also be interested in knowing if there is any potential for it to be of any serious value. Maybe the other book has more info. McComb's book was published by TAB Books in 1990. Steve . From: David Josephson Subject: Another magnetics book review To: psn-l@............. I read about and sent away for some more magnetics info, this time a books by a small fluxgate magnetometer manufacturer, Jack Janicke. The book is "Magnetic Measurements, A Technical Handbook" available for $23 from Magnetic Research Press, 122 Bellevue Avenue, Butler New Jersey 07405. Unlike a lot of this sort of thing, the book is primarily useful reference information and instructions about how to make magnetic measurements and how to calibrate your magnetometer setup, and has only a few references to the company's product. Especially if you want to look at the magnetic noise signals below 2 or 3 Hz, fluxgate mags are far better than the search coil mags that have been proposed for this. The book shows a circuit for drive and coil matching electronics and a synchronous detector all made with common parts, which can be used with the company's fluxgate sensors, available for $70. Not surprisingly, the book doesn't tell you how to make your own, but does describe all the other kinds of magnetometers. There are simple plug-in equations for lots of the usual magnetic questions that come up when thinking about seismic magnetic signal measurement. If you don't care about the theory and just want to get to the best kind of magnetometer you can make for $100, write to the company and ask just for applications notes 103 through 106, which include updated schematics for the fluxgate device. . Date: Wed, 24 Jul 1996 09:51:02 -0500 To: psn-l@............. From: mgray1@................ (Mike Gray) Subject: Re: Laser Interferometer for Seismometer Sensor Allan, it did not appear that you posted this to the list. This is the book I was referring to. It was published within the last year or two. Some book stores carry it. Gateway Electronics in various locations in the U.S. stocks it and will fill mail or phone orders. If anyone needs the ISBN number send me an email. Allan, what's this thing called "television/boob-tube" and does anyone have enough time to fool with it when there's more important things to do? Mike Gray >From: ACole65464@....... >Date: Tue, 23 Jul 1996 22:00:56 -0400 >To: mgray1@................ >Subject: Re: Laser Interferometer for Seismometer Sensor >Content-Length: 458 >X-UIDL: 838214158.002 > >The book that I think Mike Gray is referring to is titled: > >How To Build Earthquake, Weather, and Solar Flare Monitors. > >By Gary G. Giusti > >Published by TAB Books > >The book contains some interesting projects for the experimenter. >It is good to see all the activity on this group and that the television >coverage of the Olympic Games is not keeping people away from their computer, >actually it has caused me to spend more time away from the boob-tube. > >Allan > > . Date: 24 Jul 96 11:01:48 EDT From: Robert L Barns <75612.2635@..............> To: PSN mail list Subject: Meas. magnetic field variations To those interested in measuring fluctuations in the earth's magnetic field (and other mag. field meas.): Speake & Co. in England makes several very interesting magnetometers and associated IC's for reasonable money. Until now, it has been a pain to buy from England. Abbreviated specs. for one of these are:An output rect. pulse whose freq. goes linearly from 50kHz to 120kHz when the field changes from 0.5 to 0 Oersteads. The case is a cylinder 60mm long by 15mm dia. Reqs. a single 5V supply and draws 12ma. The temp. coe. is 0.003%/deg C. ~$38. An associated chip needing only a few external components integrates the field over ~ 1 sec. 'and gives an extremely sensitive response to tiny variations'. $22. There is now an American distributer:Fat Quarters Software (really!), 909-698-7950 in Murrida, CA. They will send catalog and application notes on request and will sell onsies and twosies. CA residents add 7.5% (phew!) sales tax. I will have to partake when my Lehman (now drawing lines on paper) is collecting quakes routinely. Bob Barns . Date: Wed, 24 Jul 96 14:21:00 cst From: "jmhannon" To: David Josephson , psn-l@............. Subject: Re: Another magnetics book review I came to this conclusion a while ago. However the support electronics are more complex and will require carefull design to achieve a low noise detector. I am looking at using a two hole ferrite bead as a fluxgate detector. This bead is .5 in long and .2 in dia. I drilled out the holes larger to accomodate more wire and require less drive to saturate. The drive coil is wound through the two holes and the sense coil is wound around the body of the bead. You can add pole pieces to the detector to increase it's sensitivity. Jim Hannon ______________________________ Reply Separator _________________________________ Subject: Another magnetics book review Author: David Josephson at ccmgw1 Date: 7/24/96 1:59 PM Especially if you want to look at the magnetic noise signals below 2 or 3 Hz, fluxgate mags are far better than the search coil mags that have been proposed for this. . Received: from [198.68.231.1] by sys5.webtronics.com (NTMail 3.01.03) id psnarchive; Wed, 24 Jul 1996 13:40:54 -0700 Received: (from david@localhost) by IPAC11.ipac.net (8.6.12/8.6.12) id NAA24718; Wed, 24 Jul 1996 13:31:32 -0700 Date: Wed, 24 Jul 1996 13:31:31 -0700 (PDT) From: David Josephson Subject: Re: magnetics To: jmhannon cc: psn-l@............. In-Reply-To: <9606248382.AA838236112@........................> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Info: Evaluation version at sys5.webtronics.com X-Info: Public Seismic Network Mailing List X-ListMember: psnarchive@............. [psn-l@.................... I ordered some of the Speake magnetometers from Fat Quarters too. The data sheets have been printed in Geo-Monitor recently. These will also be more sensitive than plain coils, *I think*. They don't say exactly how these mags work but I believe they're just an LC oscillator with a big high permeability core in the L. A fluxgate in a ferrite core will be very noisy; I think you could find north with it but it's unlikely that you would get to any sort of sensitivity good enough for micropulsation recording. The best these days is done with a special nickel-molybdenum (I think) alloy in a toroid. Art Green at the USGS has delpoyed lots of these in large baseline arrays. I think they get on the order of 0.01 nT/rt Hz noise level, which is probably good enough. What we really need is to find a source of cheap Rb or Cs frequency standards and a way to make them into mags ... the circuit is simple enough. . Date: Wed, 24 Jul 96 16:13:12 cst From: "jmhannon" To: David Josephson , psn-l@............. Subject: Re[2]: magnetics Hmmm, I had not realized that noise would be a problem with the ferrites. I suppose this is due to the crystaline nature of the ferrites. My thinking was that this noise would be at the switching frequency and would be filtered out with the lowpass filter after the sync. detector. BTW The Jameco catalog has some two axis detectors by Precision Navigation. They use the same pricipal detector as the Speake units. We have some here for compasses in an antenna pointing gadget. The coils are very tiny with a very thin wire core. I think the Jameco unit was about $50. Precision Navigation also makes some other detectors. Jim Hannon ______________________________ Reply Separator _________________________________ Subject: Re: magnetics Author: David Josephson at ccmgw1 Date: 7/24/96 3:57 PM I ordered some of the Speake magnetometers from Fat Quarters too. The data sheets have been printed in Geo-Monitor recently. These will also be more sensitive than plain coils, *I think*. They don't say exactly how these mags work but I believe they're just an LC oscillator with a big high permeability core in the L. A fluxgate in a ferrite core will be very noisy; I think you could find north with it but it's unlikely that you would get to any sort of sensitivity good enough for micropulsation recording. The best these days is done with a special nickel-molybdenum (I think) alloy in a toroid. Art Green at the USGS has delpoyed lots of these in large baseline arrays. I think they get on the order of 0.01 nT/rt Hz noise level, which is probably good enough. What we really need is to find a source of cheap Rb or Cs frequency standards and a way to make them into mags ... the circuit is simple enough. . Date: Thu, 25 Jul 1996 09:48:43 +1200 To: PSN-L@............. From: dann@........ (David A Nelson) Subject: re magnetometers Hi all, There are 2-3 of us down here in New Zealand that are playing with magnetometers. We all ahve a deep interest in solar activity and its effect on the earth's magnetic field. The units are able to detect the diurnal variation of the mag. field and are very easily able to detect the "Sudden Impulse" to the mag. field heralding the start of an ionospheric storm and auroral activity. Then the cameras come out to photograph the event. The magnetometers consist of a 5 cm long bar magnet suspended above a "Hall Effect Sensor" These devices are very cheap usually To: PSN-L@............. Subject: Re: re magnetometers I was trying to ignore the messages on magnetometers, but can't resist asking for more information on the system used by David Nelson. $10 is in my price range! Is it the magnets that one can get from an old disk drive? Where can one get a "Hall Effect Sensor" for <$10? Can you describe the circuit and apparatus in more detail: Parts list. Size of legs. Type of suspension (length, material, number of suspension points, etc.) Is there a recording device or just a meter? In the winter up here in Fairbanks, we have little to do other than watch the Aurora, [ :-) ] but one never knows when to wake up and take a look. A magnetometer connected to an alarm might be just the ticket. JCLahr ################################# John C. Lahr ################################ U.S. Geological Survey ############################### c/o Geophysical Institute ############################## 903 Koyukuk Drive ########################################################### ########################################################### P.O. Box 757320 ################################ Fairbanks, AK 99775 ################################# Phone: (907) 474-7997 ################################## Fax: (907) 474-5618 ################################### lahr@................. #################################### . Date: Wed, 24 Jul 1996 21:37:06 -0400 To: PSN-L@............. From: shansen@........ (shansen) Subject: re: magnetometers The suspended bar magnet is a tried and proven method for monitoring geomagnetic effects. The one I am familiar with employs a couple small bar magnets set in a piece of alum "U" channel with a length of steel rod between the magnets. The channel is suspended with a piece of monofilament with just enough torsion to pull the bar a few degrees from north. Positional sensing may be done with an inductor, HET or a pair of photocells (one on each side of a thin piece projecting from one end of the alum channel with a LED above). Several magnetometers of this design have been in operation for years. The results used to be summarized in a newsletter produced sort-of-periodically. I think it was the Very Low Frequency Experimenter's Newsletter, initially edited by Cap Hossfield then by Bruce Wingate. It was discontinued some years back. I've never made one of these but several folks were using them in conjunction with VLF SID (Sudden Ionospheric Disturbance) or SES (Sudden Enhancement of Signal) receiving systems for solar flare monitoring. (SID monitoring is an area I am active in should anyone be interested.) One observer noted the biggest problem with his system was spider webs. A very simple approach is to use one of those see-thru compasses. LED on top, photocell pair underneath. Might be useful in Alaska. Steve . Date: Wed, 24 Jul 1996 22:07:01 -0700 To: psn-l@............. From: Karl Cunningham (by way of Larry Cochrane ) Subject: Re: More on bits and dynamic range of systems. Got this from Karl Cunningham, I really don't have an answer for him. Anyone else like to reply? It seems to me something would get lost by converting the data to logarithmic and then back again for display. Larry __________________________________________________________________________ Larry -- I thought this might be a little too far off topic for the PSN list, but feel free to post it if you'd like to. Thanks for the great treatise on dynamic range of systems. It's something that I've wanted to see for a long while. This seems like it might be a good time to interject something that I've been pondering for a while... Does anyone use logarithmic A/D's for seismic work? How about a system that has an 8-bit A/D with a programmable-gain amplifier in front of it to make up the rest of the dynamic range? If the amplifier could change it's gain in steps of 4:1 over a range of 1 to 65536, this would only require 9 gain steps to cover a 90+ db range with (at worst) 7 bits of precision. And with the 8 bits of the A/D, it would easily do 120db. I haven't thought about the design of the amplifier, but 9 gain steps doesn't seem too hard to do. Maybe two stages, each with five steps from 1:1 to 1:256. The amplifier could automatically change gain to keep its output within some set limits (eg. 2.5 volts to 10 volts for a 10-volt system). The advantage of a system like this would be that it only requires 12 bits plus sign per data point. The major disadvantage is complexity. Maybe the advantage is not enough to make it worthwhile. Just a thought... Karl Cunningham La Mesa, CA karlc@......... . Date: Thu, 25 Jul 1996 02:12:31 -0700 (PDT) From: David Josephson Subject: Re: More on bits and dynamic range of systems. To: Karl Cunningham cc: psn-l@............. I don't know of any strong motion seismographs that use logarithmic a-to-d's, but this technology is common with exploration seismographs where dual slope converters or instantaneous floating point converters are used; 24 bits resolution was common ten years ago, it's probably still not a monolithic solution. Geometrics in Sunnyvale is one source of these instruments (I used to work there, in magnetics). . Date: Thu, 25 Jul 96 14:45:53 cst From: "jmhannon" To: psn-l@............. Subject: Seismograph Sensitivity All this discussion of bits and dynamic range brings up a fundamental question in my mind. Just what sensitivity does an amateur seismograph need? I know this depends on just how quiet a location you have. But in general what level of background "noise" can be expected? This would determine just how good of a detector one needs. One clue I have found so far is the PEPP seismograph has a spec of 100 nm/sec (velocity) for instrument noise. Does anyone have any more information? Jim Hannon . Date: Thu, 25 Jul 1996 11:48:50 -0800 From: lahr@................. (Dr. John Lahr (USGS-GI Seismologist)) Subject: Re: More on bits and dynamic range of systems. Bruce Ambuter designed an Ocean Bottom seismic system in the 1980's that had gain ranging much like Karl is suggesting. It worked very well. Bruce used to work for the USGS, but I'm not sure where he is now. I remember he published his design in an electronics journal at the time. We also considered hyperbolic sine amplifiers back in the 1970's when we were recording in analog mode. If well calibrated, they could significantly increase the dynamic range for digital recording as well. JCLahr . Received: from [198.68.231.1] by sys5.webtronics.com (NTMail 3.01.03) id psnarchive; Thu, 25 Jul 1996 13:18:00 -0700 Received: by stealth.cacd.rockwell.com (1.37.109.14/16.2) id AA038555339; Thu, 25 Jul 1996 15:08:59 -0500 Received: from apollo1.cacd.rockwell.com(131.198.38.6) by stealth.cacd.rockwell.com via smap (V1.3) id sma003844; Thu Jul 25 15:08:39 1996 Received: from pc110417 by apollo1 with SMTP (1.37.109.8/16.2) id AA05839; Thu, 25 Jul 1996 15:12:46 -0500 Received: from cc:Mail SMTPLINK 2.1 by ccmgw1.cacd.rockwell.com id AA838325561; Thu, 25 Jul 96 15:12:11 cst Date: Thu, 25 Jul 96 15:12:11 cst From: "jmhannon" Encoding: 43 Text Message-Id: <9606258383.AA838325561@........................> To: psn-l@............. Subject: RE: [2] magnetics X-Info: Evaluation version at sys5.webtronics.com X-Info: Public Seismic Network Mailing List X-ListMember: psnarchive@............. [psn-l@.................... Please excuse if this is a duplicate, but I have not seen it bounce off the listserver so I am sending it again. JMH Hmmm, I had not realized that noise would be a problem with the ferrites. I suppose this is due to the crystaline nature of the ferrites. My thinking was that this noise would be at the switching frequency and would be filtered out with the lowpass filter after the sync. detector. BTW The Jameco catalog has some two axis detectors by Precision Navigation. They use the same pricipal detector as the Speake units. We have some here for compasses in an antenna pointing gadget. The coils are very tiny with a very thin wire core. I think the Jameco unit was about $50. Precision Navigation also makes some other detectors. Jim Hannon ______________________________ Reply Separator ___________________________ >Subject: Re: magnetics >Author: David Josephson at ccmgw1 >Date: 7/24/96 3:57 PM I ordered some of the Speake magnetometers from Fat Quarters too. The data sheets have been printed in Geo-Monitor recently. These will also be more sensitive than plain coils, *I think*. They don't say exactly how these mags work but I believe they're just an LC oscillator with a big high permeability core in the L. A fluxgate in a ferrite core will be very noisy; I think you could find north with it but it's unlikely that you would get to any sort of sensitivity good enough for micropulsation recording. The best these days is done with a special nickel-molybdenum (I think) alloy in a toroid. Art Green at the USGS has delpoyed lots of these in large baseline arrays. I think they get on the order of 0.01 nT/rt Hz noise level, which is probably good enough. What we really need is to find a source of cheap Rb or Cs frequency standards and a way to make them into mags ... the circuit is simple enough. . Date: Thu, 25 Jul 96 15:15:11 cst From: "jmhannon" To: psn-l@............. Subject: RE [2]: more on bits and dynamic range of systems Please excuse if this is a duplicate, but I have not seen it bounce off the listserver so I am sending it again. JMH A coworker of mine used (years ago) to work on a oil exploration boat doing seismic exploring. He tells me that the sensors there use a variable gain system. There are A/D D/A converters for use in telephone systems that use a logarithmic conversion called u law or a law codecs. They get effective 12 bits out of an 8 bit converter. I have worked with these and they sound quite well. What you loose is the fine detail on large signals. Given the complexity of the circuitry I don't think logarithmic conversion is real usefull for seismographs today. I am looking at an Analog Devices 24 bit sigma delta A/D converter. This device can get up to 20 bits of resolution/dynamic range at the bandwidths needed for a seismograph. It can be connected directly to the sensor I am using and gives an interface with very few parts. No preamp or filters are required. If there is a problem with storing 20 plus bits per data point the processor could do some compression of the data. Even converting it to log data. A floating point number in a computer is essentially a log representation of the data. BTW It is not "easy" to do 120 dB of dynamic range in any electronic system. You start getting into a lot of noise problems. Jim Hannon . Date: Thu, 25 Jul 1996 14:05:53 -0700 (PDT) From: Steve Hammond Subject: Re: More on bits and dynamic range of systems. To: Karl Cunningham cc: psn-l@............. I do have a thought. I ran into a problem when building amps for schools because I never new (at the time of building) how much gain they would need. I got this idea a few years ago when I built a 12-bit A/D card using a National ADC1241 (12-bit +sign). My idea. Pick a number that is below the MAX value of the A/D and AND the bits together. Ignore the sign bit, letting the output of the AND become a line called TRIGGER. Next: In the PSN design created by Pet Rowe, an OP-27 is used in the first stage of the pre-amp (gain 201). In setting the feedback gain, there is a resister pair, 1K/200K, from pin 6 to 2, used in the OP-27 shown in Pete's drawing and also in the Lehman.PCX files on the BBSs. In place of the 1K build the following: Using a 4066 analog switch chip and a TTL 2-plex selector chip to turn the 4066 on and off, label the 2 inputs to the selector chip SELECT 1 and SELECT 2. Tie the 4 outputs of the 4066 together to ground and put one resister between the inputs to the 4066 and feedback loop where the 1K gain setting resister had been. Select the gains by selecting the resister value. I used x1, x10, x100, x500. Use the binary code for SELECT 1 & 2 to set the gain (I also used a 4 way rotor switch here.) ( As a side note, which popped into my thoughts, in the older USGS amp, it used a resister chain starting at the output of the amp (pin 6 in OP-07) tieing the other end of the chain to ground. The Db level was selected by tapping between the different resister up the chain to ground.) Anyway-- We now have a TRIGGER single and SELECT 1&2 to control the gain, what can be done with them. Lots-- But the next part of the problem is a two parter. The hardware and the code. THE CODE: If you look at the data from a seismic event you will note that the upper 4-bits of the 12-bit A/D are tied together and connected to the sign-bit (at least in Jan Froom's design and my design it is, it may not be in other's) so my thought is this. Option 1) tie SELECT 1 and SELECT 2 to data bus bits-13&14 removing the connection to the sign bit. In my design I used a 74LS273 latch and 74LS373's gate between the 13-data-bit A/D and the 8-bit ISA bus and I was able to connect at this point to the high order gate chip. This would then record the gain setting for each event point in the dataset. A pre-process program could be run before running WINQUAKE or QUAKEVIEW to adjust the dataset for a consistent gain setting. With a bit more cool coding, the view program could incorporate the gain into the display and let the user toggel the gain settings. THE HARDWARE: All that is needed is an external connection to the SELECT 1&2 and TRIGGER input lines. IE. a register that could be set on and off to drive and one to read. How about an 8055? Perfect-- Later I might want to control another seismograph. The driver should test for trigger and reset the gain. There would also need to be an interlock line so that the gain could only be changed when the A/D was not selected and transferring data. DRIVER SOFTWARE: The control of the ADC1241 is via assembler OBJ code that is linked to Ted Blank's SDAS OBJ code becoming a custom version of SDAS. Ted's code will support many card's and if you look at the last entry in the SDAS.OPT file you will fine one that is JAN's RTN. That is really an assembler routine that has a common interface to SDAS. What ever you do in "your" section of the code is up to you. So here the control for SELECT 1 and SELECT 2 and the test for TRIGGER can be added. This would also allow for the interlock of the A/D to the code as it's this routine that also performs the START and READ to the A/D (at least in my and Jan's case). THUMBNAIL CODE: The routine should read high and low byte to test for a MAX value and if reached toss the sample and lower the gain. Start the A/D and test for busy until it has a new sample and then get the data. Seems logical-- If the gain steps were X500, X100, X10, X1 that would be 4 sample times to do a MAX gain lowering to 1X of the A/D. THE OTHER OPTION -- is also were I want to be heading. I have a PIC uP and am still looking for the time to play with it. I'm just going to give control of the gain over to the PIC and let it control the gain and justify the data on the fly as it exits the A/D. IE. The A/D puts out a FF0h (+4080) and the PIC looks at the gain (that it normally keeps at 200 but because of an earlier data point now has at 20, and knows to convert it x10 to 9F60H (+40800). It's a simple idea, but I think it will work. Comments-- Regards, Steve Hammond -- PSN BBS San Jose, CA 408-226-0675 On Wed, 24 Jul 1996, Karl Cunningham wrote: > Got this from Karl Cunningham, I really don't have an answer for him. Anyone > else like to reply? It seems to me something would get lost by converting > the data to logarithmic and then back again for display. > Larry > > > __________________________________________________________________________ > > Larry -- > > I thought this might be a little too far off topic for the PSN list, but > feel free to post it if you'd like to. > > Thanks for the great treatise on dynamic range of systems. It's something > that I've wanted to see for a long while. > > This seems like it might be a good time to interject something that I've > been pondering for a while... > > Does anyone use logarithmic A/D's for seismic work? How about a system that > has an 8-bit A/D with a programmable-gain amplifier in front of it to make > up the rest of the dynamic range? If the amplifier could change it's gain > in steps of 4:1 over a range of 1 to 65536, this would only require 9 gain > steps to cover a 90+ db range with (at worst) 7 bits of precision. And with > the 8 bits of the A/D, it would easily do 120db. > > I haven't thought about the design of the amplifier, but 9 gain steps > doesn't seem too hard to do. Maybe two stages, each with five steps from 1:1 > to 1:256. The amplifier could automatically change gain to keep its output > within some set limits (eg. 2.5 volts to 10 volts for a 10-volt system). > > The advantage of a system like this would be that it only requires 12 bits > plus sign per data point. The major disadvantage is complexity. > > Maybe the advantage is not enough to make it worthwhile. > > Just a thought... > > Karl Cunningham > La Mesa, CA > karlc@......... > > > . Date: Fri, 26 Jul 1996 08:27:02 -0400 From: shansen@........ (shansen) Subject: Microbarograph Update and the Atmospheric Zoo I've pretty well got the transducer part of the microbarograph breadboarded and debugged. The 0.2 Torr (full scale) capacitance manometer is hooked up and running but I haven't taken the less sensitive Honeywell transducer out of the box yet. The plumbing has two components: one volume with a long time constant leak and a similar volume with a short time constant leak. To each volume is attached a plastic syringe which can vary the volumes by about 4x. The syringes are used to fine-tune the time constants (the leaks are capillary tubes) - a suggestion provided by Jack Herron. I've been running the long time constant manifold and observing gravity waves and other disturbances with pressure fluctuations in the range of a few milliTorr to several tens of milliTorr with periods of 30 sec to many minutes. I'm hoping for a thunderstorm this PM to see what sorts of turbulance it produces. The short period fluctuations are of much lower pressures and my data recording system doesn't yet have the sensitivity to pick up the sub milliTorr fluctuations. I am building an amp to take care of that. However, it easily displays 1 Hz waves produced by flapping the hall door back and forth and some atmospheric stuff is quite visible. So I can get the thing out-doors I am improvising a housing from an inverted 11" dia. plastic plant pot. The dish that is normally at the bottom to catch water acts as an umbrella to keep rain out of the vent (drain) holes. The only major problem has been with the recording end. I installed a PCL-711S in my PC and it works fine with the simple scope software that Advantech supplies with the card. It hasn't worked with SDR and I'm assuming my PC has a bad case of IRQ conflicts. Per Larry's suggestion I pulled a couple cards out but no-go. Later today or tomorrow I'm putting it into an otherwise empty 486 to see if it plays. In lieu of anything fancy I'm looking at the output with Radio Shack's PC compatible DVM. The final configuration will have the MKS capacitance manometer sensing the short period, low pressure fluctuations. The Honeywell transducer will be used for longer period, larger variations. I just got a number of papers on infrasound from Los Alamos. One of the authors represented, Douglas ReVelle, refers in a couple of the papers to the "Atmospheric Wave Zoo". In a 1995 presentation entitled "Infrasonic Observations of Meteors" he provides the following breakdown of the zoo (abridged from the orig): Zoo Members - Meteorological/Atmospheric Severe weather Weather fronts, jet stream, wind shear, etc. Mountain associated waves Zoo Members - Solid Earth Earthquakes Volcanic eruptions Zoo Members - Oceanic Microbaroms (analog of microseisms): voice of the sea (anyone know anything about this stuff?) Hurricanes Zoo Members - Upper Atmosphere/Astronomical Acoustic aurora and auroral electrojet Meteor-fireballs (bolides) Finally, some gravity waves are shown in a series of plots at http://www.map.ethz.ch/NL3/NL3Finke1.htm. This page describes a microbarograph network in Bavaria. Steve . From: Jack Ivey To: 'PSN-L RECIPIENTS' Subject: Re: Seismometer turns into microbarograph Date: Fri, 26 Jul 1996 09:46:59 -0400 Hi, I am curious how you eliminate dynamic pressure effects due to wind when you are measuring very small static pressure changes. It seems like almost any structure in which you housed the sensor would have the potential to act like a pitot tube. I know that microphones sometimes have a large foam or fabric shield around them if they need to operate in a windy environment. Are you doing anything similar? Jack Ivey ivey@.......... . Date: Fri, 26 Jul 1996 11:53:16 -0400 From: shansen@........ (shansen) Subject: Re: Seismometer turns into microbarograph In at least partial response to Jack Ivey's note: "I am curious how you eliminate dynamic pressure effects due to wind when you are measuring very small static pressure changes. It seems like almost any structure in which you housed the sensor would have the potential to act like a pitot tube. I know that microphones sometimes have a large foam or fabric shield around them if they need to operate in a windy environment. Are you doing anything similar?" At the moment the sensor is sitting in the open, indoors and near an open door. Wind does indeed introduce some effects in the manner you suggest. In it's next phase, the sensor will be housed in an insulated container with a diffuser. My friend at Penn State, when he was working with these things, did some simple studies of wind induced effects and found that things improved when the sensor was covered with a coffee can with holes evenly spaced about the perimeter. Better was with a garbage can over the coffee can. A heavy jacket also did good things (much like the foam ball over a microphone). For high sensitivity sensors it's been found that, for example, locating the device in a wooded area helps. Trees rustle but the frequency range is above that of the sensor. Also, there are techniques using long (up to a thousand feet) perforated or porous hoses connected to the sensor that are used to reduce noise. I'll start running into these issues when I start looking at pressure variations below 1 milliTorr. In looking at larger effects, like long period gravity waves, this does not seem to be a big issue. But, bear in mind that my total experience so far amounts to about 3 weeks. Steve . Date: Mon, 05 Aug 1996 17:59:39 +1200 To: PSN-L@............. From: dann@........ (David A Nelson) Subject: magnetometers hi all, here is the magnetometer project. In the most reprinted from the original article. with a few of my own comments. I am open to further comments or queries. For the radio amateur, however, it is usually a matter of regularly monitoring bands to find out what is happening that day, that time. It is then possible to predict pretty accurately what is likely to happen during the next few hours, taking into account the rotation of the Earth. One cannot help feeling that 21, 24 and 28MHz bands are often open in some directions, to some countries, far more often than the amateurs in the areas concerned (misled by published predictions?) realise! and anything that helps amateurs to become more aware of current ionospheric conditions is thus of interest. Martin Vincent, G3UKV, was attracted to the concept of jam-jar magnetometers as do-it-yourself "'A" or "K"' index monitors by references in GBVR's VHF/UHF column of March 1987 (pp196-7), and also April 1988 (p289), where it was reported that ~100 readers had requested information on the home-brew jam-jar magnetometers of the type used by members of the Aurora Section of the British Astronomical Association. G3UKV writes: I tried the basic version using a beam of light, and its magnified reflection off a mirror glued to a bar magnet, but I found this awkward and very easily upset. I have since been passed details of a "Recording jamjar magnetometer ' as described by H. R. Hatfield published in the February 1983 Journal of the BAA and reproduced in the form of a data sheet. It's a goer!. My version differs in some details of mechanical construction but utilises the same sensor and op-amp electronics (see Fig 1)(the attached bmp file----Dave). "Using an old kilner jar, a plastic rod (to avoid having to drill the glass) to support the Hall-effect device (634SS2 from Maplin's or Radiospares), and ordinary cooking oil to dampen the magnet's swings, I glued the small bar magnet (about 25mm long) to the edge of a piece of perspex (instead of the lead used in the BAA design), immersed in the oil. A centre zero meter is ideal, since it can be zeroed and will indicate positive or negative magnetic-field changes. A normal meter could be used, with adjustments so that under 'quiet' field conditions, it reads centre-scale. Ideally, such an instrument is kept on all the time. For those lucky enough too own a pen-recorder, that would be even better than a meter. The small pcb/psu sits on top of the jar. "The main application, at this time of rising solar activity in Cycle 22, is to observe changes in the earth's magnetic field, indicating the onset of a solar storm, and sometimes giving rise to radio auroras. It's a do-it-yourself 'A or 'K' index monitor!". Notes on the BAA data sheet relating to the electronics include: (1) ... Gain is equal to R8/R3, typically 100. (2) ... R6 is preset and provides coarse control. (3) ... R5 is a 50 Ohm wirewound "Ioudspeaker volume control" and provides fine control, together with the fine setting arm.(you could use a 10 turn trimpot for real fine control----Dave) (4) ... Do not make Sw1 unless output at Pin 6 is positive. Use fine setting arm and R6 and R7 to get about 2 volts-positive output to start with. (5) ... When Sw1 is closed, the output is damped by C1. R8/C1 control damping when Sw1 is open. (6) ... Power supplies must be carefully regulated (by double zener-diode regulation?) ( good 3 terminal regulators---Dave) . The device needs less than 10 mA total and batteries may be used. The voltage is not critical but must be well regulated. (7) ... When setting up: (a) Measure voltage at Pin 2 of sensor. Adjust suspension until it is about +2V. (b) Adjust R6 so that -2V appears on R6 slider. (c) Then make fine adjustment. Note also that all parts, except where indicated, should be non-ferrous. It is also observed on the data sheet that when torque is first applied to the nylon suspension, it will release itself again to some extent, and may well take several hours to settle down check out the attached bmp image file If any of you are not running windows of a graphics package that can handle bmp's let me know and I will see what I can do. Dave Dave A. Nelson Green Is., Dunedin, South Is.. New Zealand. Murphy's 1st Law of Thermodynamics:: Things Get Worse Under Pressure !! . Date: Tue, 06 Aug 1996 00:11:13 -0700 To: PSN-L@............. From: Jeff Batten Subject: Subscribing to Bigquake@NEIS Does anyone know how to subscribe to the NEIS mailing lists? QED or Large Events. Thanks Jeff . Date: Tue, 6 Aug 1996 14:03:30 -0700 From: rayban1@............. (RAY GALLAGHER) Subject: Re: Subscribing to Bigquake@NEIS To: Jeff Batten You wrote: > >Does anyone know how to subscribe to the NEIS mailing lists? > >QED or Large Events. > >Thanks > >Jeff > > MAILING LIST INSTRUCTIONS We have 2 automated mailing lists here at USGS/NEIC, QEDPOST and BIGQUAKE. QEDPOST sends a daily message of the earthquakes located 7 days behind the current day taken from our Quick Epicenter Determinations listing. I t is our most preliminary publication and typically contains 10-20 events per day. BIGQUAKE sends a message whenever a large earthquake release is issued (the criteria varies depending on the location of the event). Gen erally we try to locate quakes larger than magnitude 5.5 within a day of the event. Quakes larger than 6.5 are located within a few hours of the event. To remove your address from the mailing list, send a mail message to either qedpost-request@................ (to get off QEDPOST) bigquake-request@................ (to get off BIGQUAKE) In the body of the mail message, write the following command on the first line of text (NOT THE SUBJECT HEADER - the mailing lists don't care what is in it): SIGNOFF ************************************************************************ ******* To subscribe, send a mail message to either qedpost-request@................ (to get on QEDPOST) bigquake-request@................ (to get on BIGQUAKE) In the body of the mail message, write the following command on the first line of text (NOT THE SUBJECT HEADER - the mailing lists don't care what is in it): SUBSCRIBE ---end--- Take care & stay safe! Ray http://www.netcom.com/rayban1/quakes.html . To: psn-l@............. Date: Mon, 5 Aug 1996 14:18:26 +0000 Subject: ADXL05 seismometer? Hi all. Has anyone been experimenting with the ADXL05 chips in respect of seismometry? I have, following the article in "Scientific American". My initial impression is that the chip is far too insensitive for seismic monitoring, and also somewhat noisy. Does anyone have any other experiences? Cheers. Tim. Northamptonshire, England. ======================================================================= Tim Kearsley tim@.................. ======================================================================= . Date: Wed, 07 Aug 96 08:59:00 cst From: "jmhannon" To: tim@................... psn-l@............. Subject: Re: ADXL05 seismometer? Tim, Yes, the chip is not very sensitive. That particular seismometer was designed to see nearby earthquakes rather than teleseismic events. There is a great interest in discovering some kind of method of forcasting earthquakes and this instrument was intended to be used along with the ELF detector (also described in a Scientific American article) to try and find some correlation between ELF activity and nearby earthquakes. For detecting nearby quakes the ADXL05 should be just fine. It is worthless for distant events. Unless you live in a seismically active region like California it is not worth building. Jim Hannon ______________________________ Reply Separator _________________________________ Subject: ADXL05 seismometer? Author: tim@.................. at ccmgw1 Date: 8/7/96 8:37 AM Hi all. Has anyone been experimenting with the ADXL05 chips in respect of seismometry? I have, following the article in "Scientific American". My initial impression is that the chip is far too insensitive for seismic monitoring, and also somewhat noisy. Does anyone have any other experiences? Cheers. Tim. Northamptonshire, England. ======================================================================= Tim Kearsley tim@.................. ======================================================================= . Date: Wed, 7 Aug 1996 10:53:26 -0600 To: "jmhannon" Subject: Re: ADXL05 seismometer? Jim and Tim: This issue may be moot, anyway. I called Analog Devices last week to check on the ADXL05, and the sales folks said production has been temporarily stopped because of 'yield' problems. They said to call back in a couple of weeks. At the risk of offending Jim, who may already have done this, since the ADXL05 is an accelerometer, its raw output will be low for low frequency waves, such as from teleseisms, since they have lower accelerations. But the intrinsic sensitivity to acceleration should be nearly flat from the natural frequency of the ADXL05 (12 kHz) to DC. To get a better representation of the ground motion response, you may want to integrate the acceleration to velocity, which is what we get from traditional seismometers. Also, Analog Devices has a pretty complete data sheet on the ADXL05, with which one can modify the circuitry from what is in the Scientific American article. You can get a copy from their 'FAX on demand' number, 800-446-6212. You should ask for FAX code 1846 for the ADXL05. The ADXL05 has a buffer amplifier built-in that you can use to alter the type of signal you get from it. The buffer amp is controlled by components attached to pins 8, 9, 10 of the ADXL05. Could you (Jim) tell how you recorded the signals from the ADXL05? The Scientific American article listed several possibilities, but could you share your experience with what worked or did not work? Thanks, Leigh House On Aug 7, 8:59am, jmhannon wrote: > Subject: Re: ADXL05 seismometer? > Tim, > Yes, the chip is not very sensitive. That particular seismometer was > designed to see nearby earthquakes rather than teleseismic events. There > is a great interest in discovering some kind of method of forcasting > earthquakes and this instrument was intended to be used along with the ELF > detector (also described in a Scientific American article) to try and find > some correlation between ELF activity and nearby earthquakes. For > detecting nearby quakes the ADXL05 should be just fine. It is worthless > for distant events. Unless you live in a seismically active region like > California it is not worth building. > > > Jim Hannon > > > ______________________________ Reply Separator _________________________________ > Subject: ADXL05 seismometer? > Author: tim@.................. at ccmgw1 > Date: 8/7/96 8:37 AM > > > Hi all. > Has anyone been experimenting with the ADXL05 chips in respect of > seismometry? > I have, following the article in "Scientific American". > My initial impression is that the chip is far too insensitive for > seismic monitoring, and also somewhat noisy. > Does anyone have any other experiences? > Cheers. Tim. > Northamptonshire, England. > > ======================================================================= > Tim Kearsley tim@.................. > ======================================================================= >-- End of excerpt from jmhannon -- ------------------ Leigh House MS D-443 Los Alamos National Laboratory Los Alamos, NM 87545 (email: house@......... Tel: 505-667-1912; FAX 505-667-8487) ------------------ Date: Wed, 07 Aug 96 13:05:24 cst From: "jmhannon" To: "Leigh House" , psn-l@............. Subject: Re[2]: ADXL05 seismometer? I have not built this device as I live in Iowa and am not expecting very many local events. I have felt one quake here in 20 years. As a member of SAS I have had email contact with the author of the article (Shawn Carlson) about the project. Shawn is limited to 1200 words in the Amateur Scientist column so not a lot of details can be explained. Just about any A/D data acquisition card could be use to work with the ADXL05. The S.A. article described a circuit that combined the output of all 3 axis into a single signal. Because of the way the axis are combined this signal would not be usefull for doing any analysis of the quake but would serve to indicate the presence of an event. This is what you need for evaluating the effectivness of a precursor detector. Analog Devices also has a WEB page from which you can download data sheets on any of there devices. www.analog.com Jim Hannon ______________________________ Reply Separator _________________________________ Subject: Re: ADXL05 seismometer? Author: "Leigh House" at ccmgw1 Date: 8/7/96 11:58 AM Jim and Tim: This issue may be moot, anyway. I called Analog Devices last week to check on the ADXL05, and the sales folks said production has been temporarily stopped because of 'yield' problems. They said to call back in a couple of weeks. At the risk of offending Jim, who may already have done this, since the ADXL05 is an accelerometer, its raw output will be low for low frequency waves, such as from teleseisms, since they have lower accelerations. But the intrinsic sensitivity to acceleration should be nearly flat from the natural frequency of the ADXL05 (12 kHz) to DC. To get a better representation of the ground motion response, you may want to integrate the acceleration to velocity, which is what we get from traditional seismometers. Also, Analog Devices has a pretty complete data sheet on the ADXL05, with which one can modify the circuitry from what is in the Scientific American article. You can get a copy from their 'FAX on demand' number, 800-446-6212. You should ask for FAX code 1846 for the ADXL05. The ADXL05 has a buffer amplifier built-in that you can use to alter the type of signal you get from it. The buffer amp is controlled by components attached to pins 8, 9, 10 of the ADXL05. Could you (Jim) tell how you recorded the signals from the ADXL05? The Scientific American article listed several possibilities, but could you share your experience with what worked or did not work? Thanks, Leigh House . Date: Wed, 7 Aug 1996 21:41:01 -0400 To: tim@.................. From: James Forbes Subject: Re: ADXL05 seismometer? Tim Actually, I am surprised you were able to find the chip. My son and I have been interested in giving it a try. I called the manufacturer (Analog Devices) and they told me that they were suspending production because of some engineering problem. Would not quote a leadtime. Any more where ever you found them? Jim Forbes At 02:18 PM 8/5/96 +0000, you wrote: >Hi all. >Has anyone been experimenting with the ADXL05 chips in respect of >seismometry? >I have, following the article in "Scientific American". >My initial impression is that the chip is far too insensitive for >seismic monitoring, and also somewhat noisy. >Does anyone have any other experiences? >Cheers. Tim. >Northamptonshire, England. > >======================================================================= > Tim Kearsley tim@.................. >======================================================================= > > . From: Tim Kearsley To: psn-l@............. Date: Tue, 6 Aug 1996 08:32:28 +0000 Subject: ADXL05 chips again Hi all. Thanks for the replies to the message re. the ADXL05 chip and its application in sesimometry. You've confirmed my finding that is is essentially a device for monitoring large or close events. Living here in England, we don't get too many close ones! For the group's info., I obtained three ADXL05's from Allied Electronics in the USA, for around $26 each. At that time they had around 400 in stock. Cheers. Tim. G4WFT Northamptonshire, England. ======================================================================= Tim Kearsley tim@.................. ======================================================================= . Date: Thu, 08 Aug 1996 12:34:24 -0700 From: Richard Morgret To: psn-l@............. Subject: Another source To all, Having tried to buy from the source and been turned down I went to our local auto salvage yard for airbag sensors. I believe this is where all the ADXL05 chips went. The price I was quoted ranged from $2 to $5 each with no garantee. But for the price its worth it. Good luck! Dick Morgret . Date: Thu, 08 Aug 1996 19:57:20 -0700 From: Roger Sorensen To: psn-l@............. Subject: Re: ADXL05 seismometer? James Forbes wrote: > > Tim > > Actually, I am surprised you were able to find the chip. My son and I have > been interested in giving it a try. snip > > Jim Forbes > Hi All! Two weeks ago Allied Electronics (800 433-5700) had 400 or so. I don't know if they still have any, but it might be worth a try! Best of luck! Roger (KD6LQE) Chatsworth, CA Date: Fri, 09 Aug 1996 09:34:20 -0500 To: psn-l@............. From: mgray1@................ (Mike Gray) Subject: Re: ADXL05 seismometer? >Two weeks ago Allied Electronics (800 433-5700) had 400 or so. I >don't know if they still have any, but it might be worth a try! > >Best of luck! > >Roger (KD6LQE) >Chatsworth, CA > To find out the current pricing and stock on hand, now 320 or so for the ADXL05, point your web browser to http://www.hh.avnet.com and select Allied Electronics. If you try the auto salvage yard route, do NOT try to disassemble anything near the airbag! I don't know where the sensor is located. If it is near the airbag, serious injuries or death can occur if the airbag is accidentally deployed. Hopefully, the sensor has already been removed. I wouldn't have the salvage yard remove it on the spot because they may not know how to safely do it. Mike Gray . Date: Fri, 09 Aug 1996 10:10:52 -0700 From: Richard Morgret To: psn-l@............. Subject: wrecking yards & airbags X-Info: Public Seismic Network Mailing List Mike the undeployed airbags are worth too much money to laeve in the car and the sensors are wired to plugs and if you can find them are easy to remove. Dick Morgret . Date: Mon, 12 Aug 1996 23:08:13 -0400 From: Tinkler@....... Message-ID: <960812230813_177536596@....................> To: psn-l@............. Subject: PROGRAMING Would any of you just by chance have some examples in basic of how to do a strip type chart recorder, I have been able to do a bar graph display using the joystick inputs on my old XT. I'm hoping to be able to setup some kind of a siesmograph. I am using a older version of the Quickbasic compiler so I am able to end up with an executable file. Any help in this area would be much appreciated Thanks in advance Dave tinkler@....... . Date: Thu, 15 Aug 1996 22:09:13 -0400 From: ACole65464@....... Message-ID: <960815220912_386252848@....................> To: psn-l@............. Subject: Force Balance Accelerometers for Amateur Use Has any one on this network have any experience with force balance (aka force feedback) accelerometers? I am not talking about the normal type of accelerometer, like the one described in the Scientific American magazine's Amateur Scientist column of a few months back, but the type of instruments designed as broad band seismometers. I have made a couple of different types for home use using a chopped light beam configuration. They appear OK for detecting teleseismic events, when recording an acceleration signal, but when I try to intergrate the signal to get a velocity output signal then I get into trouble. I need help in finding a good integrator design. Can any one describe the best way to go about it? I live in a relative noisy area so trying to integrate the accel signal for long periods could be a waste of time. Thanks, Allan Coleman Date: Fri, 16 Aug 1996 00:18:25 -0700 To: psn-l@............. From: Karl Cunningham Subject: Re: Force Balance Accelerometers for Amateur Use X-Info: Public Seismic Network Mailing List X-ListMember: psnarchive@............. [psn-l@.................... At 10:09 PM 8/15/96 -0400, you wrote: >Has any one on this network have any experience with force balance (aka force >feedback) accelerometers? > >I have made a couple of different types for home use using a >chopped light beam configuration. They appear OK for detecting teleseismic >events, when recording an acceleration signal, but when I try to intergrate >the signal to get a velocity output signal then I get into trouble. I need >help in finding a good integrator design. Can any one describe the best way >to go about it? Greetings, Alan -- Glad to hear you've got your force-balance accelerometer running. Mine is still not (other diversions have gotten in the way), although I've got the design done for the displacement detector (light-beam, like yours). I know something about integrators and the theory behind them. Although I've not gotten my prototype working, I've done quite a bit of thinking about it, including the integrator design. I'm not sure I can help, but I'll sure try. I assume you are using the "capacitor-in-the-op-amp's-feedback" type of integrator. Could you describe more about the problems you are having. Also, what time constants you are using? -- Karl Cunningham karlc@......... La Mesa, CA. 116.9775W 32.7705N +308M interested in almost anything to do with science . Date: Fri, 16 Aug 96 07:38:28 cst From: "jmhannon" Encoding: 48 Text Message-Id: <9607168401.AA840199193@........................> To: ACole65464@........ psn-l@............. Subject: Re: Force Balance Accelerometers for Amateur Use X-Info: Public Seismic Network Mailing List X-ListMember: psnarchive@............. [psn-l@.................... Here are a couple of ideas. First the very long time constants necessary require a capacitor with a very low leakage current and high capacitance. Low leakage capacitors like polypropylene are very expensive and hard to get in large values. Using a smaller value requires very careful physical design of the integrator layout to prevent leakage paths. If you use electrolytic capacitors the lowest leakage devices are the wet slug tantalum caps. Again they are expensive and hard to find. We use these for energy storage and pay about $30 for a 500 uf part. If you are planning on digitizing the output of your accelerometer, then I think the best approach is to integrate digitally. Digital integrators have no leakage and can have an infinite time constant. If you are using a PC to log the data all you need is some software to integrate the data. I am working on a feedback seismograph. It uses an optical sensor but the feedback only does the damping and long term drift correction. I will record the displacment output from the device and process it digiatlly to whatever form I need. Could you explain just exactly what trouble you are having with your integrator? If everything is balanced the noise should integrate to zero over a long time. Jim Hannon ______________________________ Reply Separator _________________________________ Subject: Force Balance Accelerometers for Amateur Use Author: ACole65464@....... at ccmgw1 Date: 8/15/96 9:12 PM Has any one on this network have any experience with force balance (aka force feedback) accelerometers? I am not talking about the