Earlier I asked about strong motion sensors, both for local
seismic data recording, and also for triggering gas shutoff. 
Someone sent me the URL for an eval kit from analog devices:

http://www.analog.com/techsupt/eb/EB-232_B.pdf

This kit looks really interesting, it has a simple RS232 
connection and comes with software that graphs acceleration.
Would definitely be easy to experiment with. It looks like 
this chip is being used for seismic switches

http://www.analog.com/iMEMS/products/ADXL202_top.html

Here's an article about using these for seismic recording
http://www.analog.com/publications/whitepapers/products/Seismart.html

This article talks about using these low cost accelerometers
in large arrays. This sounds really interesting. Does anyone 
know if these are really being used for strong motion arrays? 
Is this $15 accelerometer really good enough to gather useful 
strong motion data? It's definitely an interesting concept, 
using low cost parts in large numbers. It will soon be 
possible to ethernet-enable these things for really cheap, 
like $15. I'd like to explore this idea further, so I'm 
going to get the eval kits for the ADXL202 and ADXL105 and 
see if I can get one of these individual accellerometers on 
the web.
 
I have on order some SIMM-sized microcontrollers that have 
ethernet, I/O processor and RTOS, plus serial and parallel
ports, CAN bus, and 1-wire microlan. Each module has its 
own IP address and implements TCP/IP, PPP, HTTP, FTP, and 
TELNET. Imagine if you could build a node of a strong 
motion array for $50, with all of these internet protocol 
interfaces. You could then plug them into ethernet 10-base-t 
concentrators, and have really easy ways to get at the data 
from large numbers of sensors. You could plug them into any 
existing network and browse their data at any time from 
anywhere in the world. You could telnet to them, transfer 
files, and even use ppp for communications. Does this sound 
interesting to anyone? Has anyone done something similar?

I'm thinking about taking one of my SIMM-sized boards and 
trying to interface it to one of the low-cost analog devices 
accelerometers. The cool thing is that when I'm done, it will 
have its own IP address and will be hanging off my local 
10-base-t LAN which routes to the Internet. You will actually 
be able to browse the sensor itself. The total cost of the 
prototype will be about $60. If the data is useful I can
make a large array for really cheap. In fact I can plug 
them in anywhere that I can find a 10-base-t network. I 
guess that would make it a world wide array. Things that 
make you go hmmm ...

  -- Doug (seismic newbee but networking professional)

PS. Please share your opinions on the Analog Devices 
    accellerometers. Are these good enough to provide 
    useful srong motion data in large arrays? Or are 
    they really only suited to emergency switching? 


_____________________________________________________________________

Public Seismic Network Mailing List (PSN-L)