seismograph

'Electronic Goldmine' had some surplus semsors a while back.

Geophone ? Centered about 16 hz.

Mic input flat at that frequency ?

Greg

Very simple. Attach a piece of music wire to a high point, dangle a weight from the wire. Have two LEDs and two photocells where the light beam is half-blocked by the weight. The light beams are at right angles, so you sense N-S and E-W movement. The photocells put out an analog signal depending on how much light they see. You can put fixed magnets around aluminum vanes on the weight to damp the system. This is basically how the pros do it.

Jon

You could make a couple of LC oscillators, where the Ls were near the pendulum, like a metal detector. Half pot cores maybe. The result would be two audio frequencies, right into the sound card.

Won't do the Z axis as easily.

Hey, a full pot core would be super sensitive to variation in the gap between core halves. That must be useful for something. I bet you could resolve nanometers.

Geophone, but it looks like they are out of stock again.

Here is one for $39:

ph).

es, so

ed text -

For $100 you could get a laser,some mirrors, photodiode and make an interferometer. (not easy though.)

George H. Personally I think the OP needs to do more research.

text -

I once clearly resolved 90 nm steps with a commercial LVDT; that's, like, 1/10 of a wavelength of NIR light. I bet the pot core could do better.

A parallel-plate capacitor would be interesting, too.

With mK temperature regulation, perhaps.

Say, Sloman has experience with those kinds of systems, perhaps a collaboration is in order?

Tim

That's always a problem with nanometer measurements. But the seismograph is sort of AC coupled anyhow.

Those are almost mutually exclusive. A piezoelectric geophone is the closest approximation. See ghost detectors and paranormal activity sensors for geophone details.

The typical sound card craps out below about 20Hz. You can butcher the input circuit to remove the input coupling capacitor to get response down to DC. etc...

Hang heavy magnet from ceiling and suspend it over two 90 degree pickup coil. Point one coil N-S, the other E-W. Leads from coils go to L-R inputs of sound card. The longer the string or wire, the better. However, that will only give horizontal deflection in 2 axes. If you also want vertical deflection, add a spring at the ceiling end, and add another coil below the magnet. The 3rd audio input can go to a USB sound dongle.

What's that in bricks, bread boxes, cubits, or other common units of measure?

Do you want 3 axis output, or a vector sum of all three?

If your goal is to just have a sensor check out

$49 for a USB sensor and some software

If you want to get into the nuts and bolts of building one perhaps start with the one at SparkFun

You can buy geophones new from RT Clark for not much more than the mystery surplus junk. You need to build a LNA.

If you expect to "see" long distance earthquakes, you need very low frequency response. There are hacks to lower the resonance of the geophone with a negative resistor. Long distance sensing is teleseismic. Most geophones are for local shaking. That is where the negative resistor scheme comes into play.

Check out this paper:

RT Clark is cool. If you need some data not on the website, they will send it to you. They don't mind small orders.

Right now they are at the SEG convention in Vegas.

If you are going to use the geophone more like the manufacturers spec sheet, they will list a termination resistance. Be sure to at least consider the resistance of the low noise amp. These low noise op amp or instrumentation amps are quite low impedance all by themselves.

There are plenty of designs on the internet for geophone amps, though none are particularly impressive. I just shake my head at the designs with jfet inputs since you will be plopping a resistor across the geophone anyway. LTC and AD have very low noise (thermal and 1/f) amps.

I think I sound card will not work well. I'd suggest a DSA, but you want to go cheap. If you look at ebay item 261072944165, these sound cards come apart easily. Maybe you can mod one to go to lower frequency. They work find on linux and windows. The converter in the one I bought is from C-Media.

es, so

d text -

RV Jones made a tiltometer/ seismograph with a capacitance micrometer.

There's a graph showing the bi-daily tilt of his lab due to the nearby tides coming in and out.

OK a quick scan to try and find what resolution he quotes... "Our best sensitivity gave an rms limit of 5x10^-12 mm for a recording time constant of 1 second. and our best stability a few times 10^-9 mm per day."

Wow! that was ~1970.

George H.

.highlandtechnology.com  jlarkin at highlandtechnology dot com

so

text -

How did that work? In 1970, he could have used a mercury pool as one electrode.

ight

ngles, so

he

ap

com

oted text -

.highlandtechnology.com  jlarkin at highlandtechnology dot com

Parallel plate capacitor.

I?m not sure if those pictures will come out.

He says it took him a while (years) to report results because the electronics was mcuh better than the mechanicals that he was making.

George H.

(...)

There's an app for that: etc.

The MEMS accelerometer inside the iPhone is not hugely sensitive, but it a good job of displaying internal noise and detecting banging on the table. That should be sufficient sensitivity for detecting an end of the world event.

so

text -

Jones was a genius at flexure stages. A whole lot of the basic flexure technology came out of his lab.

Cheers

Phil Hobbs

On a sunny day (Tue, 06 Nov 2012 09:06:07 -0800) it happened Jeff Liebermann wrote in :

Iphone is expensive, I bought an ebay item 160837973888: MPU-6050 3 Axis gyroscope + 3Axis accelerometer module

and wrote software in C for it to interface it to the PC par port. I have not contemplated using it to detect earhquakes though. You can look up the MPU6050 datasheet to check the sensitivity.

This is from the datasheet:

ACCELEROMETER SENSITIVITY Full-Scale Range AFS_SEL=0 +-2 g AFS_SEL=1 +- 4 g AFS_SEL=2 +-8 g AFS_SEL=3 +-16 g ADC Word Length Output in two's complement format 16 bits Sensitivity Scale Factor AFS_SEL=0 16,384 LSB/g AFS_SEL=1 8,192 LSB/g AFS_SEL=2 4,096 LSB/g AFS_SEL=3 2,048 LSB/g Initial Calibration Tolerance +-3 % Sensitivity Change vs. Temperature AFS_SEL=0, -40°C to +85°C +-0.02 %/°C Nonlinearity Best Fit Straight Line +-0.5 % Cross-Axis Sensitivity +-2 % ZERO-G OUTPUT Initial Calibration Tolerance X and Y axes +-50 mg 1 Z axis +-80 mg Zero-G Level Change vs. Temperature X and Y axes, 0°C to +70°C +-35 Z axis, 0°C to +70°C +-60 mg

Not sure that is enough for earth quake detection. Works great in an auto pilot control system.

angles, so

text -

Incwww.highlandtechnology.com  jlarkin at highlandtechnology dot com

Yikes, is he crushing the mica dielectrics to change the capacitance?