How to make an inexpensive geophone
Kevin Criqui
before. I'm interested in doing a little amateur seismology. Can
anybody point me at either an inexpensive, pre-make geophone or a way
to construct one myself? I intend to digitize the waveform using a
SoundBlaster card in my PC and then do a spectral analysis. Also, any
pointers to appropriate books (I'm an electrical engineer, so I'm no
geology expert) would help.
_kevin
--
| Kevin Criqui ke...@toad.horizon.com |
| Science Horizons, Inc. ...sdd.hp.com!horizon!kevin |
| 710 Encinitas Blvd. #200 (619)942-7333 (w) |
| Encinitas, CA 92024 (619)942-1652 FAX |
Russ Evans
>We just started getting this group locally, so this might have come up
>before. I'm interested in doing a little amateur seismology. Can
>anybody point me at either an inexpensive, pre-make geophone or a way
>to construct one myself? I intend to digitize the waveform using a
>SoundBlaster card in my PC and then do a spectral analysis. Also, any
>pointers to appropriate books (I'm an electrical engineer, so I'm no
>geology expert) would help.
Much depends on what you wish to observe with your geophone (but earthquake
seismologists tend to refer to them as seismometers!). With a little
searching, you can probably find a good "low-frequency" vertical geophone as
used in reflection seismics for US$50 to 100; this would probably allow
you to monitor the shaking from a nearby road in an urban environment, but
in a suitable site, a 4Hz phone will pick up small earthquakes within
50km or so. You don't say whether you are in a seismically active area.
Alternatively, it is remarkably easy to make a functional long-period
horizontal seismometer. Most folks have suitable parts lying around in
their garage. Let me offer a few quick pointers as to how to go about
it. The general idea is to attach a coil and magnet assembly to the
end of a horizontal rod, which swings about an *almost* vertical hinge,
just like a poorly hung garden gate. The coil and magnet assembly can
come from an old loudspeaker - the first trick is to find one where the
gap in the magnet, into which the coil fits, has plenty of slop - keep
reading, you'll soon find out why! The second trick is to make a
friction-free hinge. You can do this with a strip cut from a flexible
sheet material like mylar or thin plastic. Clamp one side of the strip
to the "gatepost" and one side to the "gate"; Let me try out my terminal
artwork - looking down on the hinge assembly
T T
--------/---------- ---------/----------
| / | | / | <-Clamping pieces
--------/---------- ---------/----------
--/------------------------------/----- <-Mylar hinge
-----------/---------- ---------/-----------------
/ | | /
"Gatepost" | | "Gate"
^ ^
^----These things are supposed to be screws!
The axis of the hinge should be a degree or two off true vertical and
should tilt along the axis of the boom; you'll probably want to fix
things so that you can adjust the tilt on the "gatepost" a little. The
horizontal boom is usually L-shaped with a brace across the right-angle
and the longer arm horizontal - say about three feet long and one foot
high. For a junkbox seismometer, you could make it from old timber.
The boom should swing to and fro with a period of some seconds -
ideally about ten. The loudspeaker magnet is mounted on the end of the
boom and the coil fixed to the same platform as the "gatepost", so that
it doesn't foul the magnet and the boom continues to swing freely.
As the magnet moves, it generates an EMF in the coil, which you can
amplify if necessary and digitise or record in some analogue fashion.
That's your basic seismometer! It will pick up large teleseisms -
if you're lucky, you might get a record every few days. As it stands,
this design has the problem that, once set in motion, the boom tends to
swing to and fro for some time. It needs damping. Following the Heath
Robinson school of seismology, this can be achieved by pinning a large piece
of card to the boom; this makes the seismometer respond to every air draft
in the room, but that problem can be overcome by covering the whole apparatus
with a big cardboard box. You need to put the seismometer in a quiet place,
fixed to something well anchored to the ground - e.g. in a basement.
If anyone thinks I'm joking about this contraption, I'm not. The seismographs
of the World Wide Standard Network (the VELA generation) are simply
sophisticated versions of the above, made of carefully matched metals to
avoid temperature expansion problems, and using a paddle in a bath of oil
instead of the air-damper. I'll admit that I've not myself built one
exactly as described, but I have heard reports of people who have built
wooden seismometers and been pleased with the results. A UK magazine,
Practical Electronics, published a series of articles describing a project
down these lines a year or two back (Sorry, I don't have a copy). For myself,
I've built enough long-period instruments on Heath Robinson principles and seen
enough seismometers from the early days, to have every confidence that it
will work! [btw: has anyone in the US got a Mainka working - or a Wiechert?]
The main problem with setting up a seismic station is finding a quiet site
on hard rock. If you're living in a rural area, with a basement hewn
into solid granite, you're well placed. If you're living in downtown LA,
take up stamp collecting. If you want your observatory to report
its readings to one of the agencies(*), you'll need to get a good source of
timing (an off-air signal such as WWV in the US or MSF/DCF in Europe is
probably the easiest way to go). You also need to be able to read the
records properly, which needs a little experience and an understanding of
seismology.
(*) Whether they will want your readings depends on where you are, the
quality of your site and instrumentation, and the reliability of your
assessments.
As far as books are concerned: I recommend Bruce Bolt's "Earthquakes:
a Primer" published by Freeman & Co. It's an inexpensive text of remarkable
quality. Bolt recently retired as professor of seismology and head of the
seismograph station at UC Berkeley. If you would like to look at some
seismograph records, there's a useful book of examples by Ruth Simon, with
explanations of what you're looking at. The exact title escapes me at
the moment - I'm writing from home, and I don't have a copy to hand.
Hope that helps!
Russ Evans
British Geological Survey, Edinburgh UK e_g...@va.nmh.ac.uk
Ronald J Friedel
Kevin,
There was a little article in, I believe, the May issue of Mac World,
that described something called the "Peoples Seismograph Net."
Someone from the USGS in Menlo Park is proposing a network of buried
seismometers connected to computers through A-D converters. They
claimed that the seismometer and electronics would cost a few hundred
dollars.
The USGS would set up the network so that the digitized signals would
be sent by modem to them for their processing and analysis. The signal
would then be available for your own use - to print it, manipulate it
and finally archive it. Software is apparently available to use this
system on Macs, PCs, and Amigas.
I called the telephone number mentioned in the article but the person
was out of the country till the end of May. I have written a letter
to this person to get more info. (Sorry I can't remember his name;
I'm at home and all that other stuff is at work.)
Now to build your own seismometer. In the July 1979 issue of Scientific
American, there was an article in the Amatuer Scientist column telling
you how to construct your own seismometer. This has since been referred
to as being a Lehman seismometer. It is a damped, swinging gate style
long period horizontal device. Improvements to the Lehman device were
published in the Journal of Geologic Education in about 1987.
A person named Gary Barker, a teacher at a science high school in Saginaw,
Michigan, has published articles in the above mentioned journal which
show some recordings from a Lehman seismometer connected to an old Apple II.
He has sent me a number of copies of recordings from his system. I have
one from the Loma Preita event and it sure is nice to have a recording
that can be played back at a number of playback levels so that you can
"see" as much detail as you want.
Ron
--
Ron Friedel
GeoSciences UW-Milwaukee
rfri...@csd4.csd.uwm.edu
Rainer B{rs
[text written earlier]
We just started getting this group locally, so this might have come up
before. I'm interested in doing a little amateur seismology. Can
anybody point me at either an inexpensive, pre-make geophone or a way
to construct one myself? I intend to digitize the waveform using a
SoundBlaster card in my PC and then do a spectral analysis. Also, any
pointers to appropriate books (I'm an electrical engineer, so I'm no
geology expert) would help.
I've never really tried this myself, but theoretically it could be possible
to use the pick-up from a record player as a home-made geophone. I bet You
all know how sensitive record players are to thuds etc... The frequency
respone should be adequate (goes down to < 50 Hz, usually). Getting
the needed sensitivity might be a bigger problem, but sinse You are an
electrical engineer, You might build an amplifier of Your own. Or maybe
Your stereo-set would do?
A definite plus, and a great step for mankind would be the possibility
to have a left and a right channel seismogram! Think of all the possibi-
lities to separate seismic traces on a political bases (left and right,
see:-)
The guys reading rec.audio and rec.audio.high-end might have something
to say about this, though.
If You get anything done in this way, tell us about it, too. Okay?
Richard Ottolini
The reflection seismic industry deploys data collection crews which work
with a couple thousand 5/50 Hertz geophones at a time. The sensor in one
of these geophones is a magnetic mass on spring within a coil. It is about
a cubic inch in volume. They wrap this in tough plastic case with a six inch
metal spike and two wires sticking out. I heard these sensors cost about $10
apiece and are manufactured by companies in Texas. I don't know if they can
be purchased in very small quantities.
When I mentioned this newsgroup to a geophone expert at lunch today, he said
he once directly connected a geophone to the joystick wires of a PC and got
a reasonable signal.
Some hobbist company like Edmund Scientific should package and sell individual
geophones for amateur seismologists.
Reginald H. Beardsley
B{rs) writes:
> I've never really tried this myself, but theoretically it could be
> possible
> to use the pick-up from a record player as a home-made geophone. I bet
> You
> all know how sensitive record players are to thuds etc... The frequency
> respone should be adequate (goes down to < 50 Hz, usually). Getting
> the needed sensitivity might be a bigger problem, but sinse You are an
> electrical engineer, You might build an amplifier of Your own. Or maybe
> Your stereo-set would do?
Theoretically it would NOT be possible. A seismometer requires an
inertial
frame of reference. There is no reasonable way to achieve this with a
phono
pickup. The best places to look for data are the Amateur Scientist
reprints from Scientific American, check your library.
For a professional reference try the following:
Quantative Seismology: Theory and methods, Aki & Richards, W.H. Freeman
1980
Chapter 10 covers the basic theory and includes graphs of typical power
levels for different propagation modes and frequencies. This is stiff
going
if your physics is rusty, but as long as you stay out of the rest of the
book
not too bad.
For earthquake seismology, the band of interest is 0.001 to 100 Hz. For
petroleum exploration it is 10-100 Hz. For engineering it is 10-500 Hz
(very shallow ( ~10 meter depths) work only at high frequencies). Note
that engineering and petroleum exploration provide the signal source and
use
very different detectors from the earthquake case. This is because the
bulk
of the earthquake arrivals of interest are below 1 Hz.
--
Reginald H. Beardsley
ARCO Information Services
Plano, TX 75075
Phone: (214)-754-6785
Internet: dpr...@arco.com