PSN-L Email List Message

Subject: Re: noisy power supply
From: John Hernlund hernlund@.......
Date: Thu, 06 Jul 2000 12:09:09 -0700 (MST)

On Wed, 5 Jul 2000, Angel Rodriguez wrote:
> The new computer that is causing the problem is not even one of the the
> seismo computers, it's just on the same network.  It's an NT
> machine that routes between my local network and the network that I
> get my Internet signal from.  I had to make an 2.4 ghz link to my house
> since the nearest phone line is 6 miles away.
> From your suggestion I assume that it might be that the noise is
> coming the the air as RF interference.  I will try to make some more
> shielding as you suggest.

Yes, the RF is usually my problem.  We are attempting to make electrical
conductivity measurements of mantle materials at high temperature and pressure
(in a multiple-anvil device).  Since most of this stuff has a low
conductivity, we are usually interested in measurements in the kohm-Mohm
range.  Because the resistances are high, the current generated by our
measurements is low, so lends itself to more variation in the ratio V/I.  We
heat our samples using resistance furnaces and up to 800 W of power, which
generates a huge amount of noise.  This problem is amplified by the
neighboring labs which do secondary ion mass spectroscopy, nuclear magnetic
resonance, etc..  We have to shield virtually everything to get a reasonable

> I have a new 3 component broadband that I have kludged into Larry's
> board.  It's flat between 23 second and 40 Hz.  I post the PSN files
> on an FTP at anonymous no password.
> For now I have only been collecting data but I hope to start some sort
> of long term project. If you have any ideas please let me know.  I am
> trying to make the leap between seismometry and seismology.  My
> interest run towards local events since the network run by the
> University of Panama probably can't see events much below 2.5 Md.
> Warmly,
> angel

You can do a lot with a single station I think.  One of the first things I'd
try is to measure the P-S converted waves from the moho and the offset time
between them, which will give you a depth to the discontinuity beneath the
station.  The way it works is that a portion of the P-waves coming up transfer
their energy into S-waves along with a continued P-wave.  The S-wave motion
will have to be in-line with the radial direction back to the focus, which
confines your search for waveforms in the three components.  The vertical
record will show a strong P arrival, and then shortly after the P-S converted
wave will be seen in the horizontal record.  Using knowledge of the two
different velocities then, you can find the depth to the moho.  This is
usually what a lot of the wiggles soon after the first P arrival account for:
the noise and reverberations and conversion of the first wave to other phases 
at discontinuities.  

Another fun thing to try is to find SS and PP precursors, which are phases
that don't bounce off the surface but rather bounce off the discontinuities at
660, 410, and the moho.  Because the PP and SS come in after bouncing at the
surface these will arrive a little sooner.  This will give you depths to
discontinuities at points in between the epicenter and your station.  

Many seismologists are doing both of these types of studies these days, and
some are producing maps of the global variations of the discontinuity depths.
With an array, you can get a bunch of different mid-points and study the
region in more will be interesting to see what comes out of
the upcoming USArray in this kind of study; should be pretty hot.  If you have
access to a research library, see if you can browse through the Bulletin of
the Seismological Society of America (BSSA) which has a lot of interesting
articles and different techniques you can apply to seismic data.  The
seismology stuff is really very easy and simple, but the knowledge you can
obtain about the inner Earth with it is pretty cool.

Good luck!

John Hernlund
E-mail: hernlund@.......



Public Seismic Network Mailing List (PSN-L)

[ Top ] [ Back ] [ Home Page ]

Larry Cochrane <cochrane@..............>