Changes to the Shackleford-Gundersen Seismometer
by Larry Cochrane
9/22/96
PENDULUM CHANGES
1: I used a 4 Mhz TTL oscillator rather then building it from discrete parts.
Any frequency around 2 to 6 Mhz will work. These oscillators produce a
square wave, but will work fine as a transmitter. The TTL oscillator
requires 5 volts, so I mount a 7805 voltage regulator, and some bypass
capacitor, near the oscillator can. The pendulum is powered with 12 volts.
Photo
2: To power the oscillator I pass the voltage through the hinge rather then
using wires. I used two pieces of thin brass shim material and epoxy
them to two fiber glass PC boards (without copper). One end of the hinge
(one of the PC boards) is mounted to the frame, the other is used to hold
the pendulum. The two PC boards are separated by about 1/8 of an inch. I
then soldered wires to the brass hinges.Photo
3: I added some mass to the pendulum by gluing two lead fishing weights (60
grams each) to each side of the pendulum just above the transmitter antenna.
Photo
CHANGES TO THE ELECTRONICS
1: I use a LF412 op-amp for the summing amplifier. This op-amp has much better
characteristics than a 741. Almost any modern op-amp will do. I buffered the
drive to the damping solenoid with the other half of the LF412 op-amp. If
you are going to drive a low impedence coil you might want to use a low
impedance drive op-amp like a NE5534. This op-amp can drive loads down to 600
ohms. I currently have the summing amplifier mounted on the sensor base plate.
The processing electronics and power supplies are all located outside the
box that the sensor is in. Schematic of the
summing amplifier electronics.
2: By using a LF412 (or similar op-amp) as an integrator it is possible to
replace the large polarized capacitors with smaller non polarized ones.
I re-scaled the resistors to use a 2.2 mf capacitor for both the feedback
and lowpass filters. Schematic of the signal processing
electronics.
OTHER CHANGES
1: I added about five pounds of lead weight near each leveling screw to add
mass to the system. Photo
2: You MUST make a box that just fits the sensor and fill the inside with
foam rubber to occupy as much of the free space as possible without
touching the sensor. I tried running my sensor without doing this
and had all sorts of problems with thermal noise.
Photo
[ Back ] [ Home ]
Larry Cochrane - www.seismicnet.com/contact.html