In a message dated 05/08/2005, charles.r.patton@........ writes:

This has  two problems.
1) Geometrically the difficulty is to have an output for any  direction 
movement. It would be easier to have two coils set for orthogonal  output then do 
the vectorial addition in the PC (which would take two A/D  channels also) . 
The PC could also give direction at the same time.  
It can be done with analog multipliers also if you're a bit  of a 
masochist.
    However, it is fairly easy to make an orthogonal  sensor using a square 
end mirror and four 8 sq mm photo diodes. I can  get a noise level of about 15 
nano metres. Maybe a square hole in an optical  shutter with a light on one 
side and the photo diodes on the other?

2)  Having the magnet on the boom makes you very subject to variable 
magnetic  fields all around, so then it becomes were you measuring a 
magnetic field  variation or an earth movement.
    Unless the field is enclosed, you will pick up a  lot of AC power 
interference. Even with just a soft iron weight, you will  pick up some unwanted 
magnetic noise. 

So I  would suggest you set up your Foucault pendulum with two orthogonal 
coils  standing up, and put the single magnet pole on the floor pointing directly 
 into the bottom of the bob.  
Charles Patton
    ?? My magnets all come with two poles. You could  use two rectangular 
coils and two sets of N+S NdFeB bar magnets pairs, at right  angles on the end of 
the bob. Maybe with soft iron plates top and bottom and  mild steel bolts to 
enclose the fields?
 
    I would prefer to put the coils on the end of a  brass bob weight and sit 
the magnets and the soft iron plates comfortably  on the ground.
 
    Chris Chapman








In a message dated 05/08/2005, charles.r.patton@........ writes:
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>This has=20
  two problems.
1) Geometrically the difficulty is to have an output for=20=
any=20
  direction movement. It would be easier to have two coils set for orthogona=
l=20
  output then do the vectorial addition in the PC (which would take two A/D=20
  channels also) . The PC could also give direction at the same time.=20
  
   It can be done with analog multipliers also if you're a b=
it=20
  of a 
masochist.
    However, it is fairly easy to make an orthogona=
l=20
sensor using a square end mirror and four 8 sq mm photo diodes. I=20=
can=20
get a noise level of about 15 nano metres. Maybe a square hole in an optical=
=20
shutter with a light on one side and the photo diodes on the other?
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>2)=20
  Having the magnet on the boom makes you very subject to variable 
magne=
tic=20
  fields all around, so then it becomes were you measuring a 
magnetic fi=
eld=20
  variation or an earth movement.
    Unless the field is enclosed, you will pick up=20=
a=20
lot of AC power interference. Even with just a soft iron weight, you wi=
ll=20
pick up some unwanted magnetic noise. 
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>So I=20
  would suggest you set up your Foucault pendulum with two orthogonal coils=20
  standing up, and put the single magnet pole on the floor pointing directly=
=20
  into the bottom of the bob.  
Charles Patton
    ?? My magnets all come with two poles. You coul=
d=20
use two rectangular coils and two sets of N+S NdFeB bar magnets pairs, at ri=
ght=20
angles on the end of the bob. Maybe with soft iron plates top and bottom and=
=20
mild steel bolts to enclose the fields?
 
    I would prefer to put the coils on the end of a=
=20
brass bob weight and sit the magnets and the soft iron plates comfortab=
ly=20
on the ground.
 
    Chris Chapman