## PSN-L Email List Message

Subject: Re: Geophone Generator Constant, Winquake
From: Christopher Chapman chrisatupw@.......
Date: Tue, 14 Jun 2011 11:45:00 -0400 (EDT)

```

Subject: Geophone Generator Constant, Winquake

Hello PSN,
Question: Does any/all of this seem right or wrong to you ?
The layout of the circuits are as follows:
.. Geophone Preamplifier and Damping circuits X500 no filtering
This sounds rather a high gain. You need HF roll-off to get=20
rid of the 50/60 Hz power line noise as well as other HF noise.=20

.. Equalization Filter and amplifier  X21 filter at (1/18.8) Hz
I am not sure what you are trying to do? If you want to=20
extend the period to 18.8 seconds, you need TWO filter sections=20
in series to give the 1/f^2 compensation with a total gain of x353.=20
Because the gain still increases at longer periods, you need a=20
high pass 2 pole filter set at about 25/30 seconds to control the=20
VLF noise.
3. Antialize filter X1 N=3D8 at 10Hz
Filters with N =3D 4 or 5 are generally adequate
4. DC blocker X1 N=3D1 at (1/100)Hz
.. Voltage Follower X1
.. Distribution Manifold circuit
A. A/D converter X1
B. Analog Meter circuit X20=20
Hs10-1 Vertcal Geophone
35 ohms at operational temperature
=3D 2.99 V/(in/sec)
m =3D 0.35
c =3D 1Hz
ass =3D 33 ounces avd.
Given:
enerator Constant:
..99V/(in/sec)
eophone resistance:
35 ohms
amping resistor =3D 205 Ohms
This sounds to me to be FAR FAR to low. The HS-10 =20
is no longer quoted on the Geospace website. Can=20
someone who uses one confirm what value the critical=20
Preamplifier with gain of 500 and no filtering
You definitely do need to provide high frequency=20
roll off at and above 50 Hz in the pre amplifier.

reamplifier Low Pass Filter N=3D1 of the following
f =3D 1 Meg
f =3D 3uf
ain is X21
hen I Plug in my numbers,
he OVERALL amplifier gain was computed
nly for 1 Hz.
1. minus Voltage loss related to damping resistor
K1 =3D 20 log(Rd/(Rg + Rd))
=3D  -9.888 dbv
2. plus DC voltage gain of preamplifier (no filtering)
K2 =3D 20log((R1 + R2) / R1)
=3D 53.979
3. plus DC voltage gain of filter stage
K3 =3D 20log((R1 + R2) / R1)
=3D26.510
4. minus voltage gain due to filtering at at 1Hz
4 =3D 20log(Xc/Z)
=3D-25.518
c =3D 1/2PiFC
=3D SQR(r^2 + Xc^2)
5. Overall:
=3D K1 + K2 + K3 + K4
=3D -9.888 + 53.979 + 26.510 + -25.518
=3D  45.083dbv or X179.535 AT Only ONE HERTZ
I do not worry about dampend generator constant
Ge ) since it is covered under Item number one as a loss.
This should mean that only at one hertz
y Ge going into the 16 bit A/D is
..99 X 179.535 or 536.80965 (Volts/(inch/second))
his should translate to Volts/(cm/s) like this
e =3D 536.80965 / 2.54 =3D 211.3423819 (Volts/(cm/sec))
oing into the A/D converter which is Volts Per Count =3D (5Volts / 32767cou=
nts)
his means One count is the following:
11.3423819 V/(cm/s) X (32767 Counts/ 5 Volts) =3D 13850.11166 Counts/(cm/se=
c) or
/13850.11166 (cm/sec)/COUNT or 7.220158E-6 (cm/sec)/count or
******************
ne Count equals 72.20158 nano meters per count At 1Hz and only 1Hz
*******************
This gain will have to be recalculated for any other frequency.
also have N=3D8 LPF at 10Hz and a N=3D1 100sec HPF Dc blocker.
oes any of this sound right to you ?
Is this the right way to calculate overall AMPLIFIER gain etc. for WinSDR ?
What is the best way to handle the magnitude correction ?
ike if no correction at all.
t is my guess, by adjusting the generator constant
ecause it might not meet factory specifications.
ll the other numbers seem to be known.
Please do not challenge my damping resistor at this moment.
have chosen a value of 205 Ohms for my own selfish reasons.
I CERTAINLY DO CHALLENGE IT. You will get a VERY DISTORTED=20
CHARACTERISTICS with this low value and a GREATLY reduced=20
sensitivity.=20
DO YOU WANT A SEISMOMETER WHICH ACTUALLY WORKS, OR DON'T YOU?
Regards,
Chris Chapman=20

=
Subject: Geophone Generator Constant, Winquake

Hello =
PSN,

Question: Does any/all of this seem right or wrong to=
you ?

The layout of the circuits are as follows:
1. Geophone Preamplifier and Damping circuits X500 no=
filtering    T=
his sounds rather a high gain. You need HF roll-off to get rid of the 50/60 Hz powe=
r line noise as well as other HF noise.
2. Equalization Filter and amplifier  X21 filter at (=
1/18.8) Hz
I am not sure what you are trying to do? If you want to <=
/TT>extend the perio=
d to 18.8 seconds, you need TWO filter sections in series to give the 1/f^2 compens=
ation with a total gain of x353. Because the gain still increases at longer periods=
, you need a high pass 2 pole filter set at about 25/30 seconds to control the VLF noise.3. Antiali=
ze filter X1 N=3D8 at 10Hz    Filters with N =3D 4 or 5 are g=
4. DC blocker X1 N=3D1 at (1/100)Hz
5. Voltage Follower X1
6. Distribution Manifold circuit
A. A/D converter X1
B. Analog Meter circuit X20=20

Hs10-1 Vertcal Geophone
435 ohms at operational temperature
G =3D 2.99 V/(in/sec)
hm =3D 0.35
Fc =3D 1Hz
Mass =3D 33 ounces avd.

Given:
Generator Constant:
2.99V/(in/sec)
Geophone resistance:
435 ohms
Damping resistor =3D 205 Ohms
This sounds to me to be FAR FAR to low. The HS-10  <=
/PRE>is no longer quoted o=
n the Geospace website. Can someone who uses one confirm what value the critical damping re=
TT>Preamplifier with gain of 500 and no filtering <=
TT>    You definitely do need to provide high frequency roll off at =
and above 50 Hz in the pre amplifier.
Preamplifier Low Pass Filter N=3D1 of the following
Rf =3D 1 Meg
Cf =3D 3uf
gain is X21
When I Plug in my numbers,
The OVERALL amplifier gain was computed
only for 1 Hz.

1. minus Voltage loss related to damping resistor

K1 =3D 20 log(Rd/(Rg + Rd))
=3D  -9.888 dbv

2. plus DC voltage gain of preamplifier (no filtering)

K2 =3D 20log((R1 + R2) / R1)
=3D 53.979

3. plus DC voltage gain of filter stage

K3 =3D 20log((R1 + R2) / R1)
=3D26.510

4. minus voltage gain due to filtering at at 1Hz
K4 =3D 20log(Xc/Z)
=3D-25.518
Xc =3D 1/2PiFC
Z =3D SQR(r^2 + Xc^2)

5. Overall:
K =3D K1 + K2 + K3 + K4
=3D -9.888 + 53.979 + 26.510 + -25.518
=3D  45.083dbv or X179.535 AT Only ONE HERTZ

I do not worry about dampend generator constant
( Ge ) since it is covered under Item number one as a loss.

This should mean that only at one hertz
my Ge going into the 16 bit A/D is
2.99 X 179.535 or 536.80965 (Volts/(inch/second))
This should translate to Volts/(cm/s) like this
Ge =3D 536.80965 / 2.54 =3D 211.3423819 (Volts/(cm/sec))
Going into the A/D converter which is Volts Per Count =3D (5Volts / 32767co=
unts)
This means One count is the following:
211.3423819 V/(cm/s) X (32767 Counts/ 5 Volts) =3D 13=
850.11166 Counts/(cm/sec) or
1/13850.11166 (cm/sec)/COUNT or 7.220158E-6 (cm/sec)/count or
*******************
One Count equals 72.20158 nano meters per count At 1H=
z and only 1Hz

*******************

This gain will have to be recalculated for any other =
frequency.
I also have N=3D8 LPF at 10Hz and a N=3D1 100sec HPF Dc blocker.
Does any of this sound right to you ?

Is this the right way to calculate overall AMPLIFIER =
gain etc. for WinSDR ?

What is the best way to handle the magnitude correcti=
on ?
Like if no correction at all.
It is my guess, by adjusting the generator constant
because it might not meet factory specifications.
All the other numbers seem to be known.

Please do not challenge my damping resistor at this m=
oment.
I have chosen a value of 205 Ohms for my own selfish reasons.

I CERTAINLY DO CHALLENGE IT. You will get a VERY DISTORT=
ED CHARA=
CTERISTICS wi=
th this low value and a GREATLY reduced sensitivity.     DO YOU WANT A SEISMOMETER WHICH =
ACTUALLY WORKS, OR DON'T YOU?    Regards,    Chris Chapman <=
/FONT>

```