The following was sent to the SAS (SOCIETY FOR AMATEUR SCIENTISTS) Amateur 
Scientist's Forum email (see http://www.thesphere.com/SAS/) and the PSN
mailing list.

--------------------------------------------------------------------------
Greetings,

This is my first message to this list so I apologize for being so negative, 
but here it goes...

In the last SAS Amateur Scientist's Forum email (Vol 3 #1) I read two 
messages about this article, here are my thoughts on it.

I think the April 1996 Scientific American article on a New Backyard 
Seismograph has been a big disservice to the amateur seismology community. 
If the article had been labeled as a strong motion sensor I wouldn't have any
problems with it. I hope to get a hold of some of the IC sensor chips myself
someday and make a strong motion sensor using these chips. The problem is 
these chips are designed to detect a car crash not the very small movements 
that most events make. Someone on the Public Seismic Network mailing list 
(see http://www.seismicnet.com/maillist.html) once calculated that it would take a
local (under 100km) quake of magnitude 5 or larger before you would get a 
useful signal out of this type of sensor. If you do not live near active 
faulting you are not going to pick anything up. You will be spending a lot of
time and money for nothing. If you do live in a place where you have large 
earthquakes (like myself, I live in the San Francisco Bay Area) this sensor 
should be your second sensor, first make a more sensitive seismometer like a 
Lehman or Shackleford-Gundersen. Even in areas with large events you can wait
a long time between events that will be large enough to pick up. I have been
recording local and teleseismic events for over 2 years with sensor in my 
house and monitoring USGS telemetry signals, and so far there have been only 
one event that I could barely feel. I would be surprised if this type of 
sensor would have pick up anything. One more thing about this type of sensor.
Given that you will only be recording large local events I would suggest you 
have all of system/electronics on some sort of battery backup system or UPS. 
If the event is large enough you may loss AC power and loss the data before 
it can be stored.

Now for some good news. You can pick up earthquakes at home. You just need to
make a more sensitive device. Both the Lehman sensor (see July 1979 SA) or 
the Shackleford-Gundersen (SG) sensor (Sept 1975 SA) will record a lot more 
events  then a sensor based on a ADXL05 accelerometer IC chip. This making
the hobby of seismology a lot more enjoyable. I have found that the SG sensor
design makes for a very good seismometer for recording teleseismic (distant)
events. I have compared seismographs from my SG sensor to broadband 
commercial sensor, run by the Berkeley University near me, and am amazed at 
how similar they are. Not bad for a sensor that costs under $200.00 for the 
parts compared to a STS-1 or STS-2 Streckeisen seismometer costing a lot more
(I think I read that they are around $10K). I hope to have a Web page showing
the differences between the different sensor on the Redwood City PSN Home 
page soon.
    
For more information on making a seismograph system please see my PSN home 
page at http://www.seismicnet.com.

Regards,
Larry Cochrane
Public Seismic Network
Redwood City, Cal USA