chuq@plaid.UUCP (02/02/87)
Thanks to everyone who responded to this! For everyone else who is interested, here is a summary of the responses I got. chuq --- From: kontron!brad (Brad Yearwood) Scientific American published a book or two collecting highlights from C.L. Stong's long-running Amateur Scientist column. The book(s?) covered up to perhaps 1970. There were one or two articles about home seismographs in there. The book is probably out of print now, but many public libraries have it. It also has an article on a nifty and dangerous home atom smasher powered by a Van de Graaf generator. I recall that the seismographs were really suitable only for house-dwellers. At least one relied on a long horizontal pendulum arm suspended from a fencepost-like vertical post driven (and probably concreted) into the ground. It might be interesting to think about combining more modern sensor technology (cheap laser interferometers, Kynar piezo film, ...) with the pendulum. The big problem seems to be finding a location not swamped by traffic noise. From: pyrtech!trwrb!trwspp!spp2!stassen (Chris Stassen) I think that Edmund Scientific Co. used to sell seismographs for ~$200, but everything that they sell is more expensive than it's worth. I don't know if you could set one up in your home, though. You may be able to set up the recording apparatus in your home, but the sensor certainly has to be on solid ground. Also, it's a bit difficult to weed out the stuff caused by garbage trucks (and other heavy things) driving by, unless you have several sensors widely spaced. From: karsh@geowhiz.UUCP (Bruce Karsh) Sure, you should be able to put together a home station for not too much money. The Geophysics Dept. at UW Madison puts portable seismic station all over the world. They are quite expensive and highly accurate, but you don't need all the professional features in order to get pretty nice seismic data for home use. What you need for a home digital seismic recording station is: _______ ---------- ----------- | | | | | | | |----------------------| |----------| | | | | | | | _______ ---------- ----------- Geophone Cable Amplifier Computer 1) Geophone Geophones are widely available with lots of different price/performance characteristics. They come in 2 basic types: accelerometers and velocity transducer. The accelerometers are used mainly for large motion. The velocity transducers are used for recording smaller motion. For home use, I think velocity transducers would be more fun. In fact, both types are pretty much the same, they differ only in their frequency response, and in the size of the force for which they operate well. Geophones which have good low frequency response are, in general, larger, heavier, and much more costly than higher frequency phones. (However, there is a relatively recent technique called "force-balance", which has a lot of promise for reducing the cost of low frequency phones). Since you are in an earthquake active area, you don't need low frequency phones. A 8hz phone will probably do just fine. I don't remember how much these phones cost, but I think that they are considerably under $100. With the slowdown in the oil exploration industry, I expect that the cost of geophones has gone down considerably lately. 2) The purist would have coupling transformers on both ends of the cable, but you probably don't need them for home use. Use insulated twisted pair wire for the cable. 3) The amplifiers are the hard part. They should have diferential input and lots of gain. The best way to choose the gain for home use, is to choose enough gain that the cultural background noise level is one or two counts high on your computer's A/D converter. How much gain this is depends on how much cultural noise there is, (busy streets, pumps, etc., are really bad), and how large the motor constant on your geophone is. (The motor constant is the number of volts per velocity). You are in a pretty good situation if you can feel the earthquakes. You will be able to get by with much less exotic amplifiers than if you were in a less seismicly active region. Consider some sort of lightning protection in your amp design. Consider some scheme to keep the amp's dc offset low. (I.e., < 1 a/d convertor count). Capacitive coupling works. Alternatively, you can remove the dc offset in software on your computer. Most triggering algorithms require low offset voltages to work well. 4) The computer should contain an a/d convertor and graphics. Sample the signal at around 100 to 200 samples per second. A good triggering algorithm is, for each sample, take the ratio of the sum of the absolute value of the previous 100 samples to the sum of the absolute values of the following 100 samples. If the ratio exceeds 4, you have a genuine seismic event. (Or a truck, or a pump, or ...etc). Disclamer: I hope this information helps. I am not really an expert in seismic data acquisition devices. I spent a few years making seismic processing software. There are people on the net who know a lot more about seismic data acquisition devices than I do. Hopefully, they will correct the mistakes in this posting and offer some more suggestions. (Lee, Ben, are you listening?) I think that a home seismic recording station could be a lot of fun. With the hardware outlined above, you would have a better seismic recording station than many 3rd world governments have. (And they paid a lot more for theirs than you would for yours!). >From: yerazuws@rpics.RPI.EDU (Crah) Back in the mid-60's, Popular Science Magazine printed an article and plans on how to build your own seismograph. The most expensive part seemed to be the geared-down motor to turn the paper chart. The other parts were plywood, steel wire, and a big coffee can full of concrete. Also a BIC pen. It looked kind of like this: (side view follows) |--| | |\ | | \ | | \ | | \ | | \ | | \Steel wire | | \ | | \ | | \ | | \ | | \ | | \ | | \ | | pivot |"""|Coffee can full of concrete | |<------------| | | | rod |---|-|Bic pen | | | | | O slowly rotating drum The whole thing is about four feet tall, and is sensitive to P-waves running into/out of the page, and horizontal S-waves running left-right plane of the page. So, they recommend you build two at right angles to get everything except vertical S-waves (which are harder to detect). The drum is usually on a threaded rod so that you get more than one rotation's worth of data on a sheet. Quiet traces will be written as nice helicies and when the earth shakes, you'll be able to record it. Have an *accurate* clock nearby, so you can mark the start point of any trace to within a second or two (by hand-swinging the pendulum and writing the time). There are amateur groups who share seismological information- with multiple stations and accurate timing, you can get a good fix on the actual location of the fault. A local universities' geology department can probably get you in touch with the proper group. It works on the principle of inertia, so the post and drum *must* be solidly anchored to the earth. Bolting it to the basement floor is good. You vary the length of the steel wire with a turnbuckle to vary the stability of the coffee-can pendulum. If the wire is too long, the pendulum will droop too much at the center (straight-to-drum) position and it will be insensitive (as well as having the pen lift from the paper on a real shock). If the wire is too short, the pendulum will be unstable and will fall to one side or the other. You have to hit it just right- and it helps to have the BIC pen on some sort of hinged arm to let it track the drum surface rather than digging in/lifting free. Dig up the article for all the gruesome (but very inexpensive) details. From: res@sdiris1.UUCP (res) There is no reason why it can't be done, except the cost of the needed recording drums, and such... the sensors can be homebuilt, if you are good at electronics, but that 24 hour drum chart recorder is not cheap, unless you have one handy already. I am currently using a military surplus seismic motion detector for quake alert (cost about 125 bucks, surplus) and you can make a wake-up alarm with a 6 dollar radio shack car vibration alarm, a sonalert, and a 9 volt battery which will go off at any even slightly feelable quake. Another home method is to use a phonograph cartridge with a mass clipped to the place where the needle would rest, and feed it to an amp with about a gain of 500 or so, then to a meter or chart, as you like. The mass is only a few grams or so. By the way, if there is a mail-list for seismic news, I would like to be placed on it... I am a volunteer with the San Diego Office of Emergency Management, and like to follow news about research into quakes and such. Chuq Von Rospach chuq@sun.COM It's only a model...