David Saum DSaum@............
Subject: Damping Resistor


Hi Chris
****I think we agree that the damping resistor is not important for
requency response below the resonant frequency.
Hi Dave,
    Yes, but so what ? If you want a useful seismometer, you need to be=20
able to pick up the P and S waves for the distance timing analysis and=20
this involves getting a reasonably flat response out to 10 Hz for local=20
quakes, or 5 Hz for near regional ones.

***My goal is get AS1 equivalent performance out of a 4.5 Hz geophone,
o I need to go to 1/100 of the resonant frequency, all the way down
o 0.045 Hz in the range of surface wave frequencies.  I am not sure
hat the Roberts circuit is up to that.
    Then I suggest that you do some simple maths. A flat with velocity=20
response would involve a maximum gain of 100^2 =3D 10,000. Then put that=20
gain onto the noise inherent in a 390 Ohm geophone ! And add on the=20
problems of the amplifier 1/f noise and shot noise. Then maybe sit down,=20
have a double Brandy and think of another approach ? Like making a simple=
=20
vertical with a period of about 2 seconds, but avoiding all the noise and=
=20
constructional problems of an AS1 ? Maybe add atmospheric pressure change
compensation ?
    The AS1 had a resonant period of about 1.5 seconds and the response=20
was electronically extended to ~4.5 seconds. The high frequency response=20
was severely limited by the 6 SPS ADC rate to something nearer 1.5 / 2 Hz.=
=20
It had oil damping with about a +/-5 F Deg operating range, which is very=
=20
limiting in practice. I fitted one with magnetic damping and extended the=
=20
response period out to 20 seconds. This enabled me to pick up at least=20
twice as many quakes. But I couldn't do anything about the ADC sample rate=
=20
to get the full range of P and S wave signals.=20
..=20
****I have a very simple circuit that approaches that performance as you
an see by my sample detections, including spectra, listed below the
elicorder plot on my page
ttp://www.infiltec.com/seismo/qm45.htm
 think that you can see from these detections that my 4.5 Hz geophone
ystem has good performance from 1 Hz all the way down to surface waves
ear 0.04 Hz.
    Sorry, but NO, I can't see that, simply because I don't know what the=
=20
envelope / FFT SHOULD look like in sufficient detail. The surface waves=20
often have very large amplitudes, so just seeing them is no indication=20
that their ampitudes are ~correct. Even the origonal AS1 could pick up=20
SOME surface waves! I would have expected to see a strong FFT peak at=20
~20 seconds, but I don't see one. Are you using a 1/f or a 1/f^2=20
amplifier and what is the response period range ?=20
    Regards,
    Chris Chapman












=
David Saum DSaum@............

Subject: Damping Resistor





Hi Chris

****I think we agree that the damping resistor is not important for
frequency response below the resonant frequency.
Hi Dave,=
    Yes,=
 but so what ? If you want a useful seismometer, you need to be 
able to pick up the=
 P and S waves for the distance timing analysis and 
this involves getting a reasona=
bly flat response out to 10 Hz for local 
quakes, or 5 Hz for near regional ones.

****My goal is get AS1 equivalent performance out of a 4.5 Hz geophone,
so I need to go to 1/100 of the resonant frequency, all the way down
to 0.045 Hz in the range of surface wave frequencies.  I am not sure
that the Roberts circuit is up to that.

    Then I suggest that you do some simple maths. A flat with velocity 
response wou=
ld involve a maximum gain of 100^2 =3D 10,000. Then put that 
gain onto =
the noise inherent in a 390 Ohm geophone ! And add on the 
problems of the amplifier 1/f noise and shot noise. Then maybe =
sit down, 
have a doub=
le Brandy and think of another approach ? Like making a simple =
vertical with a period of about 2 seconds, but avoiding all the =
noise and 
constructional problems of an AS1 ? Maybe add atmospheric pressu=
re change
compensation ?
    The AS1 had a resonant period of about 1.5 seconds and the respo=
nse 
was =
electronically extended to ~4.5 seconds. The=
 high frequency response 
was severely limited by the=
 6 SPS ADC rate to something nearer 1.5 / 2 Hz. 
It had oil damping with about a +/-5 F Deg operating range, which is very 
limiting in practice. I fitted one with magnetic dampin=
g and extended the 
response period out to 20 seconds=
.. This enabled me to pick up at least 
twice as many =
quakes. But I couldn't do anything about the ADC sample rate 
to get the full range of P and S wa=
ve signals. 
. 
*=
***I have a very simple circuit that approaches that performance as you
can see by my sample detections, including spectra, listed below the
helicorder plot on my page
=
http://www.infiltec.com/seismo/qm45.htm
I think that you can see from these detections that my 4.5 H=
z geophone
system has good performance from 1 Hz all the way down to surface waves
near 0.04 Hz.

    Sorry, but NO, I can't see that, simply because I don't know what the <=
/FONT>
envelope =
/ FFT SHOULD look like in sufficient detail. The surface waves =
often have very larg=
e amplitudes, so just seeing them is no indication 
that their ampitudes are ~correc=
t. Even the origonal AS1 could pick up 
SOME surface waves! I would have expected to see a strong FFT peak at 
~20 seconds, but I don't see one. Are you u=
sing a 1/f or a 1/f^2 
<=
FONT size=3D2>amplifier and what is the resp=
onse period range ? 
    Rega=
rds,
    =
Chris Chapman