PSN-L Email List Message

Subject: Re: folded pendulum
From: tchannel1@............
Date: Sat, 30 Jun 2007 06:51:42 -0600


Chris,    This may be hard to answer.  In this case, the mass would only =
be  an amount required to do the job.
I have read that some people use very little mass, others a lot.  On a =
vertical and on a horz one gets the idea of how much to use.   This a =
bit different, and here is where I could use some direction.

1  If I added no mass, simply the connecting arm, made of the same =
material as the pendulums, and the same length or maybe longer.   This =
arm is the mass, and as such is not much of a stress on the SS? shims =
used as hinges.
My mock up was made like this.   Then I added a large mass, everything =
seemed to work the same. But using a large mass would  put a lot of =
strain on the hinges.

2  About the only advantage I can see with a large mass is that air =
currents would be less likely to move it.

3 It would be easy to construct it either way,  Perhaps no addition mass =
is too little, and an amount which stressed the hinges is too much?  Is =
1/2 that amount proper?    I understand your suggest about the 5% trim =
weight.

Thanks, Ted


  ----- Original Message -----=20
  From: ChrisAtUpw@..........
  To: psn-l@.................
  Sent: Saturday, June 30, 2007 12:13 AM
  Subject: Re: folded pendulum


  In a message dated 2007/06/30, tchannel1@............ writes:


    Hi Chris,  If I understand, you would use the Arm for the Mass or =
the Arm also acts as the mass.
    With the mass already centered, you would only need a small amount =
of mass on a threaded rod to tweak the period?
    Thank You, Ted


  Hi Ted,=20

         That is correct. Making the horizontal arm fairly heavy to act =
as the mass should give it a very high position stability and symmetry. =
There should be minimal thermal expansion problems. You would then just =
move the period trim mass, which might be only a few % of the total =
mass. A small error in the position of the trim mass should be tiny in =
comparison to the total mass.=20

         The lateral forces from the two pendulums subtract, hence you =
need very high stability. I suggest that you first make the frame, the =
pendulums and the arm and then slide a trim block along the arm. You =
might start with about 5% and see if this is adequate or if it could be =
reduced. You will likely need to fit a magnetic damper from the start. =
Remember that the damping required decreases as the period increases.

         If you make the main mass, the trim mass, the mounting fittings =
and the threaded adjuster all out of brass, all the thermal expansions =
should match. The Australians chemically machined a complete square form =
seismometer, including the flexures, out of a solid block of bronze to =
get very high dimensional stability.=20

         There are several interesting papers on seismometers and =
flexures in the publications list at http://www.gravity.uwa.edu.au/ =
There used to be some on line both there and a LIGO in the USA.

         Regards,








Chris,    This may be = hard to=20 answer.  In this case, the mass would only be  an amount = required to=20 do the job.
I have read that some people use = very little=20 mass, others a lot.  On a vertical and on a horz one gets the idea = of how=20 much to use.   This a bit different, and here is where I could = use=20 some direction.
 
1  If I added no mass, simply the = connecting=20 arm, made of the same material as the pendulums, and the same length or = maybe=20 longer.   This arm is the mass, and as such is not much of a = stress on=20 the SS? shims used as hinges.
My mock up was made like = this.   Then I=20 added a large mass, everything seemed to work the same. But using a = large mass=20 would  put a lot of strain on the hinges.
 
2  About the only advantage I can = see with a=20 large mass is that air currents would be less likely to move = it.
 
3 It would be easy to construct it = either=20 way,  Perhaps no addition mass is too little, and an amount which = stressed=20 the hinges is too much?  Is 1/2 that amount = proper?    I=20 understand your suggest about the 5% trim weight.
 
Thanks, Ted
 
 
----- Original Message -----
From:=20 ChrisAtUpw@.......
To: psn-l@..............
Sent: Saturday, June 30, 2007 = 12:13=20 AM
Subject: Re: folded = pendulum

In a=20 message dated 2007/06/30, tchannel1@............=20 writes:

Hi Chris,  If I understand, you would use the Arm = for the=20 Mass or the Arm also acts as the mass.

With the mass = already=20 centered, you would only need a small amount of mass on a threaded = rod to=20 tweak the period?
Thank You, Ted

Hi Ted,=20

       That is correct. Making = the=20 horizontal arm fairly heavy to act as the mass should give it a very = high=20 position stability and symmetry. There should be minimal thermal = expansion=20 problems. You would then just move the period trim mass, which might = be only a=20 few % of the total mass. A small error in the position of the trim = mass should=20 be tiny in comparison to the total mass.=20

       The lateral forces from = the two=20 pendulums subtract, hence you need very high stability. I suggest that = you=20 first make the frame, the pendulums and the arm and then slide a trim = block=20 along the arm. You might start with about 5% and see if this is = adequate or if=20 it could be reduced. You will likely need to fit a magnetic damper = from the=20 start. Remember that the damping required decreases as the period=20 increases.

       If you make the = main=20 mass, the trim mass, the mounting fittings and the threaded adjuster = all out=20 of brass, all the thermal expansions should match. The Australians = chemically=20 machined a complete square form seismometer, including the flexures, = out of a=20 solid block of bronze to get very high dimensional stability.=20

       There are several = interesting=20 papers on seismometers and flexures in the publications list at=20 http://www.gravity.uwa.edu.au/ There used to be some on line both = there and a=20 LIGO in the USA.

      =20 Regards,

       Chris = Chapman
=20

[ Top ] [ Back ] [ Home Page ]