Info

Screen Shot 2014-05-10 at 8.02.13 PM
This is it.
Up until now the only people who were able to see data coming out of the seisMeowmeter was myself and anyone else I showed it to. Today, all that changes. Now you too, can see a stream of seismic data in the Calgary area!

Here are the links:
For the Raw Output, click here
For the Frequency Domain, click here


There are a few caveats to be mindful of…

1: This is not in any way, shape or form, a professional seismic activity gathering device. It was never designed to be one. It is simply a neat experiment to see if any kind of ground movement can be detected.

2: This seismograph is not a very sensitive one, so don’t expect to see an earthquake from the other end of the world to show up here. World survey seismometers have sensitivities that are magnitudes higher than geophones. Expect to see, however, the effect of cars driving by, people walking/biking/running/rollerblading etc.

3: This project is not done. There’s still a bit of work that needs to be done such as hardware tweaking and applying DSP algorithms. Often, the feed will be down as I continue to work on the project.

4: The stream is not exactly “real-time.” The ADC samples the geophone 500 times per second but the data can only be uploaded to Plotly’s servers every 50 msec so 1 sec worth of information that you see on the stream means about 40 or so seconds have passed in real life.

5: The FFT graph is not streaming. It updates every 40 sec to 1 min. You’ll have to keep refreshing if you want new FFT data.

I used the Plotly APIs for the graphing/streaming services. It’s a really, really neat framework that allows you to upload all sorts of graphs online. The seisMeowmeter is connected to a Raspberry Pi which takes in the data and uploads to Plotly via a python script.

photo-3
First Impressions
The first thing I noticed about the output was that it seemed to be picking up a lot of noise:

Signal when not much ground movement is occuring

Signal when not much ground movement is occuring

Looking at the FFT, I suspect that the seisMeowmeter is picking up noise from the power mains. Whether it is EM noise or actual vibrations remains inconclusive. However, considering that the line connecting the geophone to the conditioning circuit isn’t shielded, I strongly suspect that I’m picking up EM noise. I guess it’s time to start thinking about using an instrumentation amplifier.

EDIT: After connecting the circuit ground to mains earth, the signal became much clearer so it looks like the circuit has been picking up mains hum.

FFT capture

FFT capture


Helpful People
I really have to thank the folks over at Plotly. They’ve been so helpful in helping me out in getting this stream going! They’ve responded to my emails and worked out any issues quickly. It also turns out that one of their engineers worked with seismological data gathered around here. Crazy!