User avatar
jbeale
Posts: 3360
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

earthquake sensing - Rasperry Shake and PS3 Eye webcam

Thu Jan 04, 2018 6:18 pm

I thought this was an interesting result. I have a Raspberry Shake seismometer sitting in my garage which records local earthquakes (and distant ones, if they are large enough). I also have another Pi connected to a Sony PS3 Eye webcam, which I'm using just as a microphone. It sits in a shed in the backyard recording ambient sounds (mostly birdcalls, and traffic noise) as mp3 files.

There was a small M4.4 earthquake near Berkeley, CA this morning at 2:39 am PST. I am about 40 miles away in San Jose, but my raspberry shake picked it up clearly. Plot from R-Shake: https://photos.app.goo.gl/y16nrl8ZJ6qC26jh1

Somewhat surprisingly, the cheap little PS3-Eye device also picked it up. I looked at the mp3 file in Audacity in normal and spectrogram mode: https://photos.app.goo.gl/Uz74C56vgWoIb83p1

The spectrum peak amplitude was around 7 Hz, I did not expect an audio recorder to see much at that frequency. Note, I had done a mechanical modification to the PS3, extracting two of the four mic capsules and putting them on the end of a short cable, so I could install each in a hole in the middle of a US letter size (8.5x11") fiberboard sheet. I was trying to improve stereo separation, but that may also have improved low-frequency response.

User avatar
Gavinmc42
Posts: 2108
Joined: Wed Aug 28, 2013 3:31 am

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Sun Jan 07, 2018 12:32 am

Another use for Eyecams :D
Signal from both is very similar, just wondering if the fibreboard made the difference?
A rigid mount would move with the ground but the microphone film would try to stay stationary
Instead of the diaphragm moving, the microphone shell is moving, the reverse of normal operation?

A big piezo element with a lead weight, soldered.glued in the middle of the disc?
Resonance effects, but with Eyecam microphone there would be less resonance.

I think there is a an earthquake app for Smart phone, is it only accelerometers or does it use the microphone too?
Will be interesting to use with Pikrellcam as the big trucks driving past it is like earthquakes daily :lol:
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

User avatar
jbeale
Posts: 3360
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Tue Jan 09, 2018 10:56 pm

piezos would not be my first choice as they are nonlinear, high impedance, and subject to creep over time. Also mechanically fragile. I remember working with a piezo pressure sensor where the calibration would drift with humidity.

However there are a large variety of DIY seismograph designs online, using a variety of sensors, if you want to roll your own.
Here's one: http://www.instructables.com/id/This-Se ... is-no-toy/

The smartphone apps use only the accelerometer as far as I know. The audio microphone will be much more sensitive to normal ambient noise, than to quakes. I doubt a cellphone mic has much response below 10 Hz. If it did it would probably be useless outdoors, due to wind noise.

User avatar
Gavinmc42
Posts: 2108
Joined: Wed Aug 28, 2013 3:31 am

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Wed Jan 10, 2018 6:22 am

With the Eyecam we have 4 microphones, by mounting them in different directions they could record different types of waves.
Put them inside a sealed anechoic chamber that is solidly connected to the ground.
A network of them could detect the direction and speed if time stamped.

Yep piezo's are brittle, used to use them for Combine Harvester wheat/chaff separation percentage tracking.
Ceramic turntable elements, tricky little things to mount on the sensor plates and they would snap if you touched them too hard.
Once sealed in the plates they a lasted long time, but they were designed to be easily replaced.
Very low cost ;) Analog DSP with opamps/filters/comparators, hate to think about coding that these days.

Job after that was Ground Vibration monitoring for Mining ( explosions) and civil constructions.
3 separate velocity sensors (coil/magnet based), much more expensive and you learn to build tough, trucks ran over these things :lol:
Very weird, you could make them waterproof but outback bulldust still got in :shock:

Next job - PC/DSP based Predictive Maintenance using vibration analysis on rotating machinery.
Heavy iron -TI DSP and PC MS C coding, graph plotting etc.

Pi's could do these things much cheaper now, coding simpler too, plus many more features and a bit more than 32KB or 640KB of sram:D
Still pretty early days for Pi's in Science, starting to appear in Journals/Thesis etc
Still waiting for the Officially approved SciPi Guy T Shirts ;)
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

User avatar
jbeale
Posts: 3360
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Wed Jan 10, 2018 3:27 pm

Yes. USGS puts the 7.6 Honduras quake at 02:51:31 UTC and I saw the signal arriving here in San Jose CA at 2:59:01 UTC with peak indicated frequency around 0.4 Hz but that's only due to the low-end rolloff in my sensor. It was a strong quake but also over 4000 km away from me, so the signal strength isn't so large, and dominated by even lower frequencies that my geophone-based sensor doesn't see. https://photos.app.goo.gl/DdMtlkBhGhgeJEDb2

This "Raspberry Shake" was really intended for more local quakes. A more conventional sensor like a Lehman long-period design (see below) that can see 10 or 20 second period waves would be better at picking up distant quakes. (note: dividing great-circle distance by time, you get around 9 km/sec which is in the ballpark for seismic wave velocity but slightly above the expected 2-8 km/sec range. Of course the waves can travel in a straight line through the earth, not just around the surface, so they didn't really go quite that far.)

Image

BudBennett
Posts: 83
Joined: Fri May 17, 2013 2:45 pm
Location: Westcliffe, Colorado, USA

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Fri Jan 12, 2018 10:06 pm

The quake in Honduras registered 30000 counts on my Raspberry Pi powered seismometer. Myanmar registered as well, but is too far away to be much above the noise floor.
Attachments
20180110Honduras7p6-lorez.jpg
20180110Honduras7p6-lorez.jpg (53.47 KiB) Viewed 2104 times

User avatar
Gavinmc42
Posts: 2108
Joined: Wed Aug 28, 2013 3:31 am

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Sat Jan 13, 2018 11:29 am

BudBennet,
What is your Pi seismo setup?
Seems to me it should be easy to make Pi ones from old model A, B's.
I got enough off those lying around unused now, they need a use instead of just sitting around.

JBeale, how many Pi's do you have doing things now?

Wonder if those shaker LED torches, magnets and coils could be re-purposed?
One of the guys here needed a high mH inductor, the coil from a solenoid got re-purposed ;)
I really don't like winding coils, could a hall effect sensor replace the coil, i2c version?

Hmm, 0.2Hz rolloff, so sampling at 1Hz?
Up to 500Hz for some.
https://en.wikipedia.org/wiki/Seismometer
24bit :o big ask
16bit should be easy enough, any higher and Sigma/Delta needed but they have lower upper freq, which is not an issue :D
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

BudBennett
Posts: 83
Joined: Fri May 17, 2013 2:45 pm
Location: Westcliffe, Colorado, USA

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Sat Jan 13, 2018 1:36 pm

BudBennet,
What is your Pi seismo setup?
Seems to me it should be easy to make Pi ones from old model A, B's.
I got enough off those lying around unused now, they need a use instead of just sitting around.
I am using a Pi B+. The processor is somewhere between 25-45% occupied with data gathering and a seedlink server. I have another PI sitting on the network that archives the seedlink data on a daily basis. I am using a capacitive displacement sensor instead of an inductive velocity sensor. My bandwidth is something like 0.05Hz to 0.5Hz, with 2-pole software filters at both ends. It is better to sample at higher frequencies than 1 Hz -- I'm sampling at around 19Hz, I think (it's been a while). I described my setup in a previous posting:viewtopic.php?f=37&t=3731&start=25

A 24-bit ADC is easier and cheaper than you think. Check out my Hackaday project: https://hackaday.io/project/25848-a-che ... spberry-pi

User avatar
Gavinmc42
Posts: 2108
Joined: Wed Aug 28, 2013 3:31 am

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Sun Jan 14, 2018 1:28 pm

Thanks Bud, had not thought of capacitor sensor, should not be effected by the magnetic damper or Earth's magnetic field.

Wonder if the PCM/PDM i2s on the Pi could be used as a Sigma Delta clock?
Precision comparators?
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

User avatar
jbeale
Posts: 3360
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: earthquake sensing - Rasperry Shake and PS3 Eye webcam

Fri Jan 26, 2018 7:49 pm

I have not been monitoring this thread and I don't routinely post my data here, but if you are interested, there is a website that plots locations of earthquakes detected by the "Raspberry Shake" sensors online in real time. http://raspberryshake.net/eqview/
There are now more than 500 such sensors in various places around the world. Of course they are not evenly distributed, they are heavily concentrated in "high-tech" areas where there are hobbyists interested in such things (mostly near Coastal US & Western Europe; very few eg. in Africa)

Return to “Other projects”

Who is online

Users browsing this forum: No registered users and 12 guests