Required Materials

Tools

Details

The Enclosure

I built the seismometer in a discarded Lipton Ice Tea mix jar with a small piece of flat glass plate to cover the jar’s mouth. I used a glass jar because:

• I wanted a closed container to eliminate air currents that could give false signals,
• Glass has a low “coefficient of thermal expansion”, which means that it will be relatively immune to changes in dimension due to temperature changes.
• I could see through the glass while installing the seismometer’s guts to make sure the pendulum, sensor, and magnet damper fit and interacted properly—a real plus!
• It was cheap!

 Note: For the instrument depicted in the photos, I drilled a hole in the wall of the jar to pass the three wires for the sensor electronics through it. In a later version, I simply passed the wires out the top of the jar, and laid the glass cover plate over them. This caused no ill effects.

The Torsion Fiber and Suspension Mechanism

I used fiberglass thread that I scavenged from a discarded fiber optic cable as the torsion fiber that supports the pendulum. The fiberglass I used was sort of an insulation or space filler around the actual optical fiber. I just stripped off the outer plastic insulation, and pulled out the fiber glass threads. I wound a few of the threads into a single, thin thread and knotted the ends. I considered using nylon fishing line for this, and it may work. I was concerned, however, that the nylon would stretch, allowing the pendulum to sag and drag against the other components.

 I then soldered some loops made from paper clips onto some small steel pieces. I built two of these for each seismometer. These were used to support the torsion fiber on each end. I attached to small, but powerful, magnets (like the kind sometimes used for earrings and other jewelry) onto the back side of each hook assembly. I then built two similar small steel plates with magnets pairs (but without hooks). These were used on the outside of the jar to hold the hook assemblies in place on the inside of the jars. 

The Pendulum

The pendulum is a copper penny with a strip of metal glued to it.  I also glued a very thin piece of brass (you could use copper or  aluminum foil) to one flat surface of the penny. When assembled, this strip of metal (I call it the “flag” in the photos, as did Roger Baker for his design) serves to interrupt the light beam depending on the position of the pendulum. The flat surface of this metal strip should be parallel with the flat surface of the penny. I then cut a very narrow slot into the edge of the penny  with a small metal cutting wheel mounted in a rotary tool. This allowed the pendulum to be glued onto the torsion fiber with epoxy. When assembled in the jar, the torsion fiber is oriented vertically, while the pendulum is oriented horizontally, and able to rotate around the vertical axis. Aside from providing mass to regulate the period of the pendulum, the electromagnetic eddy currents are formed in the copper in the penny when it interacts with the magnetic field produced by the damping assembly. This slows the movement of the pendulum, allowing the earth’s movement to be more faithfully registered.

The Damping Assembly

The damping assembly is constructed from a strip of soft steel (about 1/16” thick, ¾” wide, and about 2-1/2” long) and two disc-shaped neodymium magnets. Bend the flat piece of soft steel  into the shape of a “U”. Then attach the flat disc-shaped neodymium magnets to each of the tips of the “U” to form a horseshoe magnet. No need to use glue, just let them attach themselves so that their opposite poles are facing each other on the inside of the “U”, but not touching. Be careful—the magnets are quite strong, and can pinch your fingers or shatter if you allow them to snap together. The resulting gap between the magnets shooame-built horsehoe magnet as shown in Figure 4.

The Sensing Electronics

The sensing electronics are assembled on a perf board cut to about 1-1/2 wide and 4” long. Use some enameled hook-up wire to connect the components. The infrared LED and the phototransistor should be facing each other as shown in the diagram below. They should be spaced about ½” to ¾” apart and aligned so that the light from the LED will shine on the phototransistor (you won’t actually see any light because the “color” of the light is beyond the range of human vision—similar to the remote for your television). If you use the same hook-up wire for the leads going to the power supply and the output, you should label them so that you don’t get confused later. You could color-code these wires instead—I typically use red for the positive power supply wire, black for the negative power supply wire, and blue or yellow for the output wire. As you can see in the photo below, I used hot melt glue to secure the LED and phototransistor into place so that they wouldn’t vibrate. You can bend strips of steel into “V” shapes and glue them onto the back of the perf board to allow attachment of magnets. Magnets can then be used on the outside of the jar wall to maneuver the sensor electronics board into alignment with the pendulum flag. Neodymium magnets are strong enough to keep the sensor board from shifting without glue, allowing for later adjustments.

Important Note: I have tried using Cadmium Sulfide photocells (CdS Cells) rather than phototransistors, but got poor results. It turns out that CdS cells are temperature-sensitive, also, which can introduce a some noise. I also don’t believe they are as sensitive as phototransistors (but, hey, I could be wrong).

The Base Plate

The base plate is simply a sheet of aluminum with three holes drilled through it (See Figure 1). I arranged the holes at the vertices of a triangle somewhat larger than the diameter of the jar so that there is enough room to use a wrench or screwdriver to adjust the tilt of the base plate. I then tapped these holes to fit a fine threaded screw. Neither the size nor the threads per inch are really critical. If you don’t have a tap, you could simply epoxy some matching nuts to the plate over the holes. Use double-faced tape to attach the glass jar to the base plate so that it doesn’t slip off and break as you tinker with it.

Final Assembly

 This part requires lots of tinkering to get the geometry right, and to get the pendulum period and damping adjusted properly. The initial set-up may take several hours, and then you need to monitor or record the output for days or weeks to determine if you are getting adequate results.

Lay the jar on its side and brace it so that it doesn’t roll. Find a small stick about ¼” x ¼” x  2” long to use for propping up the damping assembly.  Mix up some epoxy, and using a long stick place a large dab midway down the inside wall of the jar; this is where the rear edge of the damping assembly will be glued. Smear some more epoxy on the 2” long stick and place it on the inside jar wall about ¾” or so from the dab of epoxy. Place the damping assembly in the dab of epoxy, with the forward edge propped on the stick. Make sure there is plenty of epoxy attaching the damping assembly to this stick. Be careful NOT to get epoxy on the magnets—you may need to adjust their location later. Allow the epoxy to cure according to package directions.

Next, tie small loops in each end of the torsion fiber. The loop both of these onto the hooks attached the two torsion fiber supports fabricated earlier.

After the epoxy for the damping assembly has cured, place some double-faced tape on the top center of the base plate, and attach the base of the jar to it in an upright position.

Next, install the torsion fiber by sliding the lower torsion fiber support into the jar, and attaching magnet assemblies to the outside by the mutual attraction of the magnets through the glass. Slide the assemblies down until the pendulum is far enough inside the jar that you can similarly attach the upper torsion fiber support, again using magnets outside the jar for attachment. Slide the magnets around until the pendulum’s penny is positioned between the magnets of the damping mechanism. 

Next slide the sensor electronics board into the jar, being careful not to snag the torsion fiber or the pendulum. Use magnets and a strip of metal on the outside of the jar to attach the board to the jar wall, then slide the board around using the magnets and steel strip until the LED and phototransistor are situated on opposite sides of the pendulum’s flag.

The fiber should be tight enough to hold the penny at nearly a right angle to the wall of the jar. Adjust the leveling screws so that the pendulum tends to rest with about 1/3 of the penny overlapping the magnets in the damping assembly. I found this achieved the best damping—and therefore the best registration of ground motion. The pendulum should swing between the damping assembly magnets and the LED and phototransistor without rubbing any of them. This may take a lot of tinkering. You may also have to disassemble the instrument several times to trim the size of the flag to ensure it doesn’t collide with the perf board.

After you have assembled the instrument, you should be able to bump it gently with a finger and see the pendulum move slightly to one direction, but return rather quickly to its original position. This indicates that the pendulum is being damped more or less the way you want it to be damped.

If you already have your amplifier/filter and an analog-to-digital converter connected to your computer, you can do a bench test by connecting the seismometer to power and to the analog-to-digital converter. If you have constructed the instrument properly, it should register your presence as you walk near the instrument. You weight will cause the floor to tilt or warp (even a concrete floor!), in turn, causing the pendulum to seek a new, lower level. If you get those kind of results, your instrument should be ready for installation.

Carefully carry your instrument to the installation site. After you set it in place, you will have to re-adjust the level screws and possibly the sensing electronics and pendulum by manipulating the magnets on the outside of the jar.  Cover the jar’s mouth with a flat piece of glass, and cover the whole assembly with an upside-down cardboard box to keep out light and convection currents. It's best to cover the instrument with a cardboard box to prevent infrared radiation from light bulbs or other sources from interfering with  the signals.

Here's a seismogram that I recorded with this seismometer, a homemade amplifier/filter, a Dataq analog-to-digital converter, and WinDAQ software:

As you can see, the system provides a very nice response, allowing high-frequency body waves and low-frequency surface waves to be detected.

Useful Links

Larry Cochran’s Redwood City Public Seismic Network:

http://psn.quake.net/

Lots of interesting links and projects. Larry also sells analogue-to-digital converters and special software of his own design for amateur seismologists.

Fred Buenes’ Seismometer Site:

http://www.moonglow.net/seismo/

Fred operates a Lehman-type seismometer.

Sean-Thomas Morrisey’s Site:

http://www.eas.slu.edu/People/STMorrissey/

A Detailed (pre-electronics era) Torsion Seismometer:

http://www.eas.slu.edu/Earthquake_Center/Instruments/torsion_seis.html

Originally from: Anderson, J. A., Wood, H. O., Description and theory of the torsion seismometer, Bulletin of the Seismological Society of America, vol. 15, p. 1-72, 1925.

Roger Baker’s description of an force feedback torsion seismometer.

The Amateur Scientists’ bulletin, Volume 6, Number 2, Summer 1999, p. 6.

Hi-Q Seismometer

http://www.eden.infohwy.com/~rcbaker/

Roger Baker’s very interesting seismometer/gravimeter (This link is broken).

Robert Lamb’s Ste

http://www.geocities.com/ResearchTriangle/Facility/1739/index.html

Many creative approaches to seismometers.

Dataq Instruments

www.dataq.com

Dataq sells analogue-to-digital converters which come with useful software.

Click here to see who's linking to my site.