THE EDENVALE SEISMOGRAPHS

The station at Edenvale, South Africa, E28d 8' 20", S26d 09' 15"
has sensors to detect the vertical velocity of the Earth's
surface and its velocity in an East-West direction.  The sensors
are firmly mounted on a concrete floor which in turn rests
on a rocky outcrop.  Even so, the floor flexes appreciably when
one stands within one metre of the sensors.  When surrounding
ground is saturated with rain water, the floor tilts
far enough to bring the horizontal detector to a limit stop.
Microseisms are continually recorded.  These are related to weather
conditions. Microseisms are a fertile ground for study by persons
interested in atmospheric as well as ground movement.

The station has been in continuous operation with short inter-
ruptions for adjustments and modifications since 1989.  Numerous
distant earthquakes and many hundreds, perhaps thousands of local
earth tremors have been recorded in this period.  A number of
mining operations within 200 km of my station cause these tremors.

The horizontal instrument consists of a Golytsin (Swinging
gate) pendulum.  Its lower bearing is a steel ball rolling on
a hardened and polished vertical steel plate.  The upper
bearing is a thin wire under tension. At the free end of the
pendulum sits a lead seismic mass and a transducer consisting
of a pancake shaped coil moving between the poles of a magnet.
The coil consists of 2000 turns of thin (48 SWG?) copper wire
salvaged from an automobile high tension ignition coil.
The coil is wound on a former which has aluminium sides in order
to provide eddy current damping.
The pendulum is definitely underdamped.

The coil's output is approximately a microvolt per micron per
second, which is representative of ordinary seismic motion.
The pendulum's natural frequency is 0.07 Hz.  The horizontal
instrument is most useful for detecting distant events.  It
also responds to local tremors but these register more strongly
on the vertical instrument, which has a higher frequency.

The vertical instrument consists of a seismic mass sitting on
the end of a semi-rigid beam which is attached to a 12" steel
rule, the other end of which is firmly attached to the steel
base of the instrument.  Its natural frequency is 1 Hz.
The mass is linked to a moving coil electromagnetic transducer
without introducing friction.  The transducer has a magnet
from a large moving coil speaker, with the annular gap widened
to accommodate the moving coil, which also consists of 2000
turns of fine copper wire.  The coil is wound on a brass former
to provide eddy current damping.  A copper former would have
made a much more effective damper, due to the lower resistivity
of copper.  I find it necessary to provide oil damping as well.

Both instruments feed into DC amplifiers with appropriate
filtering.  The outputs of these amplifiers operate side by
side penmotors which cause ink stylii to register the traces
on a single rotating drum.  The drum speed was originally
1000 mm per hour but I have recently changed it to 500 mm
per hour in order to conserve paper and pens. I have no
intention of converting to digital recording, which would be
the way to go if one reported events to a scientific body.

My scientific reporting is confined to lunar and planetary
occultation timings and as well as a large number of variable
star observations.  I do not report magnetic and seismic data
because South Africa has two agencies, each with a number of
outstations, so my reports would simply duplicate theirs.
The geologically quiet astronomical complex at Sutherland is
host to overseas seismic organisations which use using state
of the art detectors such as (I think) a rock strain guage
seismograph and a cryogenic superconducting microgravimeter.

There is much room for improvement.  The amplifiers do not
have common mode rejection and are consequently unstable and
inefficient.  At my age (79) I do not enjoy building electronic
circuits and feel that it would be unfair to ask associates to
help me with what is essentially a recreational activity.
I also need active filters to define the passband frequency
more sharply.

The writing stylii are ordinary fine fibre tipped pens which
are very robust but give an unacceptably wide trace. Originally,
I used India ink in ultrafine draftman's tubular pens.
The traces were very fine but the wires in the pens tended to
stick, requiring more maintenance than I was prepared to give.
The sticking can be avoided by lifting the pens periodically
to exercise the wires but this would add a complication.

I would like to convert the horizontal instrument to a
displacement sensor, using light dependent resistors but in a
displacement instrument, pendulum drift can be a problem.  Drift
can be overcome but that would again be a complication.

The whole station except for electronic components and fasteners
is built from discarded material.  Even the penmotors would have
been home made, had suitable units not been found in scrap boxes.

If colleagues on PSN wants more information, I'll be glad to
supply it but I leave for Australia in two days time for a 1-2
week visit and will not be on PSNet for that period.

An ASCII based diagram of the vertical sensor is attached.

Cheers all,

Danie Overbeek  Tel 27+ 11 453 6918
The following section of this message contains a file attachment
prepared for transmission using the Internet MIME message format.
If you are using Pegasus Mail, or any another MIME-compliant system,
you should be able to save it or view it from within your mailer.
If you cannot, please ask your system administrator for assistance.

   ---- File information -----------
     File:  991113.sei
     Date:  14 Nov 1999, 17:25
     Size:  820 bytes.
     Type:  Unknown