In a message dated 07/08/2005, skmort@.......... writes:

Question,,,   I understand that the 1 sec AS-1 vertical uses  an amp-filter=20
which compensates for the signal rolloff so that it works out to  about 25 o=
r=20
30 sec.,,,,   is the schematic for it, or a similar  amp-filter available=20
anywhere?
Hi Stephen,=20
=20
    Have a look at _http://jclahr.com/science/psn/as1/as1_resp/index.html_=20
(http://jclahr.com/science/psn/as1/as1_resp/index.html)  and  in particular=20=
Fig=20
2 for the AS-1 characteristics. Is this adequate for your  purposes? I don't=
=20
know of a published circuit.
    I can't answer your question directly since the  answer depends on your=20
sensor type. Distance measuring sensors may be  easier to compensate than=20
velocity sensors.
=20
    For extending velocity type sensors look at  Roberts, P. M. (1989). =E2=
=80=9CA=20
Versatile Equalization Circuit for Increasing  Seismometer Velocity Response=
=20
Below the Natural Frequency=E2=80=9C, Bull. Seism. Soc.  Am., 79, 1607-1617=20=
This is in=20
effect a particular type of bass boost amplifier,  which keeps the phase cha=
nges=20
in check to give a fairly flat output through fc.  The circuit has high gain=
=20
at very low frequencies, which may or may not give you  noise problems. One=20
'fix' is to fit an extra 3 pole high pass filter at  your low frequency limi=
t.
=20
    The other way to do it is to set up a narrow band  pass amplifier with a=
=20
particular low frequency limit of maximum gain, which then  has a gain at=20
higher frequencies inversely proportional to the frequency.  You need differ=
ent=20
filter laws for distance and for velocity measuring  transducers. You could=20=
also=20
tailor the gain to be constant above the  natural resonant frequency.
=20
    Allan Coleman wrote an article on lengthening the  period of a Lehman se=
e=20
_http://psn.quake.net/lpveloc.html_ (http://psn.quake.net/lpveloc.html)=20
=20
    On second thoughts, the Shackleford-Gundersen  circuit does just this.=20
See _http://psn.quake.net/sgboard.html_ (http://psn.quake.net/sgboard.html)=20=
 and=20
 linked documentation. But it has a distance sensor, so it only requires one=
 =20
integration step.

I'd like  to adapt it to use with my 4.5 sec vertical!!   I can do it after=20=
=20
the fact, using Bob McClures excellent software, but would like to fix the =20
signal before the A/D=20
converter, to save a few  steps! Interesting that they give details on=20
building the AS-1 sensor,  but not the electronics, that I can find?
Thanks in  advance,
Stephen Mortensen


Regards,
=20
    Chris Chapman





In a message dated 07/08/2005, skmort@.......... writes:
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000=20
  size=3D2>Question,,,   I understand that the 1 sec AS-1 vertical=
 uses=20
  an amp-filter which compensates for the signal rolloff so that it works ou=
t to=20
  about 25 or 30 sec.,,,,   is the schematic for it, or a similar=20
  amp-filter available anywhere?
Hi Stephen, 
 
    Have a look at http://jclahr=
..com/science/psn/as1/as1_resp/index.html and=20
in particular Fig 2 for the AS-1 characteristics. Is this adequate for your=20
purposes? I don't know of a published circuit.
    I can't answer your question directly since the=
=20
answer depends on your sensor type. Distance measuring sensors may be=20
easier to compensate than velocity sensors.
 
    For extending velocity type sensors look at=20
Roberts, P. M. (1989). =E2=80=9CA Versatile Equalization Circuit for Increas=
ing=20
Seismometer Velocity Response Below the Natural Frequency=E2=80=9C, Bull. Se=
ism. Soc.=20
Am., 79, 1607-1617 This is in effect a particular type of bass boost amplifi=
er,=20
which keeps the phase changes in check to give a fairly flat output through=20=
fc.=20
The circuit has high gain at very low frequencies, which may or may not give=
 you=20
noise problems. One 'fix' is to fit an extra 3 pole high pass filter at=
=20
your low frequency limit.
 
    The other way to do it is to set up a narrow ba=
nd=20
pass amplifier with a particular low frequency limit of maximum gain, which=20=
then=20
has a gain at higher frequencies inversely proportional to the frequenc=
y.=20
You need different filter laws for distance and for velocity measuring=20
transducers. You could also tailor the gain to be constant above the=20
natural resonant frequency.
 
    Allan Coleman wrote an article on lengthening t=
he=20
period of a Lehman see http://psn.quake.net/lpveloc.html=

 
    On second thoughts, the Shackleford-Gundersen=20
circuit does just this. See http://psn.quake.net/sgboard.html=
 and=20
linked documentation. But it has a distance sensor, so it only requires one=20
integration step.
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>I'd like=20
  to adapt it to use with my 4.5 sec vertical!!   I can do it afte=
r=20
  the fact, using Bob McClures excellent software, but would like to fix the=
=20
  signal before the A/D 
converter, to save a few=20
  steps! Interesting that they give details on building the AS-1 sensor=
,=20
  but not the electronics, that I can find?
   Thanks in=20
  advance,
     Stephen Mortensen


    Regards,
 
    Chris Chapman