In a message dated 02/09/2005, ian@........... writes:

Hi,
I'm treating this mailing list  the same way I treat coffee time at work with 
my fellow engineers, as a forum  for throwing up dumb ideas with the hope 
that about 1 in 10 fly (a reasonable  rate).
Hi Ian,
 
    Great! This was what I was also trying to do!

Specifically:

The long length is to amplify the effect of very  small tilts to the point 
that the effect is measurable.  10 m is a figure  chosen for easy calculations.

If the signals from the 2 piezo  transducers are differenced then drifts 
and tempco's should cancel  out.
    This involves two assumptions. Firstly that the two  sensor locations 10 
m apart do actually track in temperature to much better than  0.1 C Deg. The 
second is that both sensors have the same TC. In my experience,  these are 
difficult / near impossible. Thermostatting a 10 m tube to a few 10s  of milli 
degrees is likely to be rather expensive.

The  discussions here on how to measure the effect the easy way seemed to 
be  quite complicated.
    Do enlighten me, please! Capacitative level sensors  are quite standard 
technology.

Longer  periods are good!  During the recent hurricane, my medium period  
sensor was useless whilst the professional long period devices rolled  on.
    What sensors are you running / observing /  commenting on?

Having  said all that, there's at least a 90% chance that my idea is 
indeed  dumb!
Pass the coffee...
Ian


Since you are planning to use an open trough, tilts  / seismic waves which 
are slower than the natural period should be recorded OK.  Faster signals 
develop travelling surface waves from both ends of the tube which  are likely to be 
incoherent. See the Washington traces from the 26  Dec Sumatra quake. How are 
you planning to receive / analyse / suppress  these faster surface travelling 
wave signals?
 
    My philosophy is to design out problems where  possible. Pass the 
Aspirins...
 
    Regards,
 
    Chris Chapman





In a message dated 02/09/2005, ian@........... writes:
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000=20
  size=3D2>Hi,
      I'm treating this mailing l=
ist=20
  the same way I treat coffee time at work with my fellow engineers, as a fo=
rum=20
  for throwing up dumb ideas with the hope that about 1 in 10 fly (a reasona=
ble=20
  rate).
Hi Ian,
 
    Great! This was what I was also trying to do!
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000=20
  size=3D2>Specifically:

The long length is to amplify the effect of=20=
very=20
  small tilts to the point that the effect is measurable.  10 m is a fi=
gure=20
  chosen for easy calculations.

If the signals from the 2 piezo=20
  transducers are differenced then drifts 
and tempco's should cancel=20
  out.
    This involves two assumptions. Firstly that the=
 two=20
sensor locations 10 m apart do actually track in temperature to much better=20=
than=20
0.1 C Deg. The second is that both sensors have the same TC. In my experienc=
e,=20
these are difficult / near impossible. Thermostatting a 10 m tube to a few 1=
0s=20
of milli degrees is likely to be rather expensive.
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>The=20
  discussions here on how to measure the effect the easy way seemed to 
b=
e=20
  quite complicated.
    Do enlighten me, please! Capacitative level sen=
sors=20
are quite standard technology.
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>Longer=20
  periods are good!  During the recent hurricane, my medium period=20
  
sensor was useless whilst the professional long period devices rolled=20
  on.
    What sensors are you running / observing /=20
commenting on?
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>Having=20
  said all that, there's at least a 90% chance that my idea is 
indeed=20
  dumb!
Pass the coffee...
Ian


    Since you are planning to use an open trough, t=
ilts=20
/ seismic waves which are slower than the natural period should be recorded=20=
OK.=20
Faster signals develop travelling surface waves from both ends of the tube w=
hich=20
are likely to be incoherent. See the Washington traces from the 26=20
Dec Sumatra quake. How are you planning to receive / analyse / suppress=
=20
these faster surface travelling wave signals?
 
    My philosophy is to design out problems where=20
possible. Pass the Aspirins...
 
    Regards,
 
    Chris Chapman