Angel,

Regarding your other questions:

1: The current folder of drawings and schematics weighs about 6 ounces.
If you send the proper postage to get it to you, I will provide the
envelope. I can also provide 1 meter or so of hinge material (0.005"
bronze, 0.75" wide).

2: Many have experimented with logrithmic amplifiers to expand the
dynamic range of seismic recording. The problem is how to deal with 
the negative side of the signal. Many schemes of the "invert and add"
idea have been tried to handle the bi-polar analog signal, sometimes
even developing two log-scaled outputs. These still have problems with
low level signals, like near zero voltage.

The only successful compression of seismic data has been gain-ranging,
where a comparator senses a large signal and knocks down the gain of the
amplifier by 1/10 and keeps it there for a duration much longer than 
the event. Software can find the gain step and restore the gain. Multiple
gain-ranging has also worked. It has not been without problems, like
dealing with the transient at the gain step. Often it is done at the
ADC converter, where the n-bit mantissa is scaled by augmenting the 
exponent by the inverse of the programmable gain amplifier.  Modern
24-bit EDME digitizers avoid the problem altogether for a price.
But this means digital telemetry, with all its complications. So
multi-level, multi-channel, (40 db separation) analog FM telemetry is 
the simplest solution.

Regards,
Sean-Thomas

_____________________________________________________________________

Public Seismic Network Mailing List (PSN-L)