Seismic Sensors

[EEVblog]

BTW, yes, the test tank was for precision measurements on all sorts of active and passive underwater sonars and hydrophones etc.
We also has "flow tanks" that were more suited to towed arrays.

There is only so much you can in a test tank, even a large one, as the wall effects dominate very quickly.
Larger scale tests were always preferred but were logistically harder, so the test tank was handy for testing out new ideas and revisions to validate numbers etc.
We had a lake test facility in Jarvis Bay.

[radioactive]

One interesting method i know modern seismometers use is to have a test mass as an lvdt moving element, and using a control loop and a coil to keep the lvdt output at 0, the seismometer output waveform would be the output of the control loop, they use one of these for each axis, buying one of these would most likely be prohibitively expensive yet it would be an interesting (and probably damn hard) project for summer (unless you are in the north hemisphere)

From what I've read so far, this is the approach that interests me the most.  It appears you get similar low-frequency performance to the long-stroke type sensors without the larger mass/coils in the sensor.  @mark03 mentioned this type of sensor earlier in the thread  (e.g. Yuma design).  FBA (force balance accelerometer).  The Yuma design appears to use a differential capacitive sensor for the PID control loop in vertical configuration (FBV).  Sounds like capacitive is probably the best way to do this.

The reason I brought up the video with the laser interferometer measurement earlier in the thread, is that it seems like it might be another way to pick up the displacement.  I was thinking that a voice-coil (speaker) with retro-reflective tape (like in the video) could be used for the correction.  I have no idea if this could be made to work as well as the capacitive type sensor (probably not I'm guessing), but it seems like it would also be fun to play with.  I also ran across something describing Michelson interferometer involving a couple of mirrors / beam splitter to get higher resolution measurements, but that looks like it would be a bit much for a first attempt.

[rhb]

A few comments:

Take a look at this thread:

https://www.eevblog.com/forum/projects/suggestions-for-high-resolution-tiltmeter-(inclinometer)-sensor/msg1531160/#msg1531160

in particular the differential capacitance sensing.  The results achieved are quite amazing.  You could sense the movement of a dielectric core.  So basically a capacitance version of an LVDT.

Making an LVDT would not be hard if you can wind a coil and allow generous clearance for the core.  There are lots of plans around for coil winding machines.

Dead CD/DVD drives are an excellent source of beam splitters and first surface mirrors for a Michelson interferometer.  There are laser diodes which have enough coherency length to serve for a seismometer.  IIRC these are the ones with a photodiode in the back and it requires a circuit to stabilize the frequency of the diode.

A cheap He-Ne laser will have a coherency length of 8-12" which is enough for a small ring gyro.

Finally there is the La Coste suspension.  There's a picture of the first LaCoste seismograph here:

https://web2.ph.utexas.edu/utphysicshistory/LucienLaCoste.html

I recall seeing a version made using 3 screen door springs.

[diegogmx]

the problem i see with the interferometer approach (at least the open air one) is that you may very well measure air temperature fluctuations too, just thinking out loud but maybe if you use a REALLY high resolution and dynamic range sensor you could send the beam into a reasonably short fiber and measure a "linear sagnac effect" also using an interferometer
also if you want to sacrifice some resolution i figure maybe a laser diode / photo diode like the one in the applied video would suffice
the other disadvantage i see from a design like that would be that the low frequency response would be pretty pretty low since the sagnac is a consecuence of velocity, and at low frequency and low amplitude velocity is really low

i figure the highest resolution approach would be a michelson interferometer in vacuum with the mirror in a suspended test mass, pretty much like LIGO but without seismic isolation

[rhb]

Once you start playing around measuring motions at sub-micron levels, *everything* becomes an issue.

One of the interesting aspects of a seismograph is that an amateur can build a research grade instrument.  It's challenging project because there are so many factors which must be taken into account.  But the cost is low if you have a suitable location and the necessary skills.

Where I live the traffic noise would be pretty bad, so I'll probably content myself with measuring deflection of the ground as the water level in the lake a mile from me changes.  No real value to that, but I *am* a retired geophysicist.  But then there are probably very low magnitude earthquakes associated with sudden loading from major rain events. So if I ever get my workshop organized who knows what might happen.

[Rerouter]

You can also go all the way to the extreme with a laser tractor beam, essentially a tiny spherical lens is held in place by the photon momentum, and by measuring the deflection of the light you can measure extremely tiny forces.

[radioactive]

A few comments:

Take a look at this thread:

https://www.eevblog.com/forum/projects/suggestions-for-high-resolution-tiltmeter-(inclinometer)-sensor/msg1531160/#msg1531160

in particular the differential capacitance sensing.  The results achieved are quite amazing. 

@rhb,  thanks for pointing to that thread and the other thoughts.  That simple capacitive sensor is pretty amazing.  I assume that is just water/air (80/1) in the vials based on the Ames description?  I think I might start with this sensor and see if I can get similar results in the garage.

@diegogmx,  thanks for thoughts on the laser.  After watching the appliedscience video, I was thinking that it might be possible to utilize the idea of modulating the power supply + DSP to get some higher resolution measurements.  I'll be doing a digikey order soon.  I might pick up a diode with the integrated PIN to play around with.

@Rerouter

You can also go all the way to the extreme with a laser tractor beam....

Then profit!

[Rerouter]

Sorry, the more correct term was Optical Tweezers, Its how they measure extremly small forces

[rhb]

Spirit levels are filled with alcohol so they won't freeze and break.  Hence the name.  It's probably worth spending a little extra and getting the UK made vials.  The interior finish on the Chinese vials is a bit rough and leads to erratic movement.  Not important in a machinist's level, but JBeale found it problematic for a monitoring application.

If you happen to live near a volcano or a major fault zone you can do serious science by setting up an orthogonal pair on a bedrock mounting.  If you've got a good setup with the vials accurately oriented USGS will be glad to get data from you.  I'd talk to USGS before making the permanent mount.  They may know from experience that there are some wrinkles to getting good data.  Also they'd probably want to do a calibration of your unit before installation.

I dropped my membership in the Seismological Society of America some years back so I don't know the current details, but USGS has a sizable network of tiltmeters all over California.

It's a good bit of work to calibrate and integrate a new instrument, especially a seismograph installation, into the network so to work with them you'd need to be very professional in your dealings with them.  That mostly requires that you read and understand the relevant recent literature.  Those papers will also tell you who to contact.

I strongly suspect that the same circuit and electrode arrangement with a piece of teflon rod would be equally sensitive in a seismograph application.

I'm used to seeing ads in the geophysical journals for microradian tiltmeters for many years and understood how they worked, but I never imagined it was as simple to make one as it is.  The commercial units are quite expensive.

[mark03]

One interesting method i know modern seismometers use is to have a test mass as an lvdt moving element, and using a control loop and a coil to keep the lvdt output at 0, the seismometer output waveform would be the output of the control loop, they use one of these for each axis, buying one of these would most likely be prohibitively expensive yet it would be an interesting (and probably damn hard) project for summer (unless you are in the north hemisphere)

This is known as "force feedback" and is a staple of all modern broadband seismometers, including the amateur FBV instruments mentioned up-thread.  It's analogous to negative feedback in electronic amplifiers, yielding a flat response over multiple decades of frequency.  Older and simpler designs like the Lehman, and I believe also the Benioff mentioned by @GK, are merely second-order damped pendulums whose bandwidth can only be expanded through post-processing.  With force feedback, it is possible to achieve the same (actually, much better) performance at low frequencies, without the heavy seismic mass.

[mark03]

The reason I brought up the video with the laser interferometer measurement earlier in the thread, is that it seems like it might be another way to pick up the displacement.  I was thinking that a voice-coil (speaker) with retro-reflective tape (like in the video) could be used for the correction.  I have no idea if this could be made to work as well as the capacitive type sensor (probably not I'm guessing), but it seems like it would also be fun to play with.  I also ran across something describing Michelson interferometer involving a couple of mirrors / beam splitter to get higher resolution measurements, but that looks like it would be a bit much for a first attempt.

You might have seen the same paper that I recall, where they used the guts of an STS-1 (?) retrofitted with optical position sensing, and DSP to keep track of the fringes using quadrature signals.  My recollection is that they were running this open-loop.  I have been assured by people who know this stuff much better than I do, that a digital control loop (for force feedback) is impractical because any realizable D/A will sacrifice too much dynamic range.  This may be an insoluble issue with the interferometric measurements.  OTOH with sufficiently low measurement noise you may be able to achieve what you want without force feedback... which I think was the point of that paper.

Ok, found the quote I was looking for, from Erhard Wielandt, Design considerations for broadband seismometers (http://ds.iris.edu/stations/seisWorkshop04/PDF/Wielandt-Design3.pdf)

As an example for the way in which feedback can increase the dynamic range, we may consider a system whose displacement transducer has only two output states indicating whether the mass position is “high” or “low”. A suitably designed force-balance system would nevertheless generate a feedback current that is, in the time average, proportional to ground acceleration. The seismic mass would oscillate around its equilibrium position and the feedback signal would oscillate around the value corresponding to ground acceleration. This does not necessarily disturb the seismic record provided that the oscillations have a frequency above the seismic bandwidth or an amplitude below the desired resolution. A system according to the latter condition was in 1924 used by A. De Quervain and Auguste Piccard to stabilize the mass position of a 20-ton mechanical seismograph at the Swiss Federal Institute of Technology, by adding or subtracting every minute a few drops of water from a ballast tank on top of the mass. This system came amazingly close to being a digital seismograph – had they been able to make their feedback loop some 10 000 times faster and to count the number of water drops, then their system would have been the first (and so far only) seismograph with true digital feedback.

A digital feedback circuit oscillating at high frequency is used in modern Sigma-Delta A/D converters. It is also possible to design the feedback loop of a Sigma-Delta A/D converter around a mechanical pendulum as the summing point, thus creating a digital seismometer. The concept of digital feedback cannot, however, be fully realized because a digitally controlled forcer with an adequate dynamic range is not available. One would have to couple an electronic DA converter to an electrostatic or electromagnetic forcer, so the feedback force would still be generated by an analog signal; the dynamic range would be limited by the DAC to something like 140 dB (7 orders of magnitude). In comparison, the combination of an electromagnetic forcer with a solid-dielectric feedback capacitor, driven by an analog circuit, produces feedback currents over a range of 240 dB or 12 orders of magnitude, say from a noise level of 10^-14 A at 0.3 mHz to a clip level of 10^-2 A at 30 Hz. So it appears that the analog feedback loop of VBB seismometers cannot easily be replaced by a digital circuit.

[rhb]

FWIW An interferometer based seismograph produces an FM signal.  The larger the amplitude, the faster the fringes change.  Much better than trying to count fringes to just use regular FM demodulation.

[radioactive]

better than trying to count fringes to just use regular FM demodulation.

I noticed that right away (FM signal) from watching the applied science video too.. for the stable laser power supply.  Still *very* cool response for pretty small measurements.  Later in the video where it was demonstrated that wavelength could be manipulated via power supply, I started thinking about DSP and what might be done with that.  (even though he wasn't very optimistic about it being useful).  As mark03 pointed out, this is probably not something to be optimistic about for a closed loop system.

Thank you both for your inputs.  Much appreciated.  I'm still working on another unrelated project, but I hope to transition to playing around with a seismic sensor soon.  That capacitive inclinometer from the other thread is going to be the first thing I want to experiment with I think.

Sorry, the more correct term was Optical Tweezers, Its how they measure extremly small forces

@Rerouter,  sorry I thought you were being sarcastic or something.  That is a *very* interesting video.  Also wayyy out of my capabilities..

[rhb]

Also look into the Zeeman effect with lasers.

http://wanda.fiu.edu/teaching/courses/Modern_lab_manual/zeeman_effect.html

HP uses this in their optical measuring system to derive absolute position from the interferometer fringes.  There's an HP application note describing it in detail.

[max_torque]

Maybe I should try and ask if it's still there and get a tour?

It would be interesting to see that!  It's a field (should that be a "lake"?) i know nothing about and looks to have quite a few interesting challenges and solutions

[Doctorandus_P]

Several years ago there was a post on hackaday about a project of a home made raster tunneling microscope.
I believe it used an isolation scheme built from several layers. A heavy plate (granite?) on 2 meter long springs, with some extra dampening hardware added, but also with half filled bycicle inner tubes. (Edit: Triggered my memory from a post below me).
He had to make do with local situations (urban area?) but managed to get quite decent results.

[rhb]

A large box of sand on partially inflated inner tubes is a popular isolation table design in the hologram community.  It's cheap and much easier to move than a block of stone of similar mass (around 3000 kilograms for a 4' x 8' x 2' table).

[Doctorandus_P]

Over the years several intersing projects have been hacked together and posted via Hackaday.
It may have someting you like.
https://hackaday.com/?s=seismograph

[3roomlab]

that yuma2 LTspice model is fun to play with

[rhb]

It occurs to me that a survey volume of professional seismograph designs would be very interesting.   Unfortunately, I cannot think of a single earthquake seismologist whom I know.  A search with Amazon turned up nothing but reflection seismology books and SSA has no publications except the bulletin and research letters.

Earthquake seismology and reflection seismology have almost no contact.  I was a member of the Seismological Society of America for a number of years but dropped most of my professional society memberships when I retired as it costs quite a lot.  And in my case I simply could not deal with managing 37 publications even if many of them were quarterly.  I'm getting ready to do a major purge of journals from my library as I am running out of space.

I already have far too many projects, but the SSA bulletin and Seismological Research Letters finally got scanned so members have electronic access.  While I was a member they were just starting on making it available electronically.  But had not yet scanned back volumes.

The Society of Exploration Geophysicists which is the major reflection society routinely assembles important papers into review volumes on various topics.  It's very convenient to have all the major papers in a single volume so you can read a citation just by flipping pages.

If anyone reading this has a serious interest in seismograph design I'd like to urge joining SSA and collecting all the papers on seismograph design and testing into a review volume available through a book on demand service.  I'd buy a copy in a heartbeat.  Designing a new seismograph is now pretty rare.  Most networks settle on an instrument and use that same instrument throughout the network.  That greatly simplifies the process of calibrating the installation as you only have to model the earth coupling at that location.  Getting an amateur station accepted into one of the official networks would be real badge of distinction.