## PSN-L Email List Message

Subject: Re: plum bob seismometer
From: ChrisAtUpw@.......
Date: Thu, 16 Nov 2006 09:10:34 EST

```In a message dated 2006/11/16, ian@........... writes:

> I'm not convinced that you need to match the period of a plum bob to
> the seismic motion. A plum bob will follow the tilt even if it takes 50 seconds
> to reach its maximum. At the end of the 50 seconds, it will still be
> pointing to the centre of the Earth. On top of that you will have the natural period
> oscillation but this can be filtered out to get the mean value.

Hi Ian,

Are you talking about tilt responses which tend to be dominant for
periods over ~300 sec? At periods below this the tilt signal of a seismic wave is
minor.
Otherwise you have a pendulum system, which if correctly damped will
give an output with velocity above the resonant period, but falling as f^2
below it.

> The real value of having a long plum bob is that the amount of
> displacement of the end weight is much larger and much easier to measure. It also
> helps to reduce the significance of mechanical imperfections of the mount at
> the other end.

The problem with large pieces of equipment is keeping the air column
stable. You could evacuate a tubular vertical column, but it would be costly
and the column could still flex with temperature.

There need not be a sensor sensitivity problem. Photocells can resolve
about 15 nano metres, which is likely to be below the background noise. You
can get down below 1 nm with variable capacitance methods

Regards,

Chris Chapman
In a me=
ssage dated 2006/11/16, ian@........... writes:

I'm not conv=
inced that you need to match the period of a plum bob to the seismic motion.=
A plum bob will follow the tilt even if it takes 50 seconds to reach its ma=
ximum. At the end of the 50 seconds, it will still be pointing to the centre=
of the Earth. On top of that you will have the natural period oscillation b=
ut this can be filtered out to get the mean value.

Hi Ian,

Are you talking about tilt responses wh=
ich tend to be dominant for periods over ~300 sec? At periods below this the=
tilt signal of a seismic wave is minor.
Otherwise you have a pendulum system, w=
hich if correctly damped will give an output with velocity above the resonan=
t period, but falling as f^2 below it.

The real value of having a long plum bob is that the amount of displac=
ement of the end weight is much larger and much easier to measure. It also h=
elps to reduce the significance of mechanical imperfections of the mount at=20=
the other end.

The problem with large pieces of equip=
ment is keeping the air column stable. You could evacuate a tubular vertical=
column, but it would be costly and the column could still flex with tempera=
ture.

There need not be a sensor sensitivity=20=
problem. Photocells can resolve about 15 nano metres, which is likely to be=20=
below the background noise. You can get down below 1 nm with variable capaci=
tance methods

Regards,

Chris Chapman
```