In a message dated 2007/04/28, gmvoeth@........... writes:

> Why Doesnt anyone ever talk Hall-Effect or Strain Guage or Variable 
> Resistance or magnetic levitation like you take up most the strain with a spring but 
> levitate grain or so with an electro magnet to counter gravity and set the 
> period ?

Hi Geoff,

       You would probably have to put the coil on the arm and the magnets on 
the baseplate. One of the problems is the relatively high drift of the magnet 
strength with temperature. Maintaining the field would take an appreciable 
amount of current from a highly regulated source. 
       Putting magnetic components on a seismometer arm makes it liable to 
pick up  a lot of interference signals. You would need to completely and 
effectively screen the sensor and this is not easy to do, or cheap.
       You can make Hall-Effect sensors, but to get good seismic sensitivity, 
you have to use strong magnets and the movement range is then less then +/-1 
mm. Not a problem for a simple pendulum or a spring accelerometer.
       Strain gauges will not give you the signal to noise ratio necessary.
       Simple optical systems using a miniature tungsten filamant bulb, a 
moving slot shutter screen and two Si large area photocells connected 
differentially can give a better performance. Eg VTD34, BPW34 or larger. For a Lehman, 
you might choose 10mm or longer photo cells - see Silonex. I run the bulbs at 
about 0.8 of their rated voltage to give a very long and stable light output and 
use a voltage regulator + bypass transistor to do it. You need the large area 
photocells to get the S/N ratio up. The tiny 1mm square Si cells do NOT work 
well enough, particularly those in LED form cases with an end lens.  

       Regards,

       Chris Chapman   
In a me=
ssage dated 2007/04/28, gmvoeth@........... writes:



Why Doesnt anyone ever talk Hal=
l-Effect or Strain Guage or Variable Resistance or magnetic levitation like=20=
you take up most the strain with a spring but levitate grain or so with an e=
lectro magnet to counter gravity and set the period ?




Hi Geoff,



       You would probably have to put the coil=
 on the arm and the magnets on the baseplate. One of the problems is the rel=
atively high drift of the magnet strength with temperature. Maintaining the=20=
field would take an appreciable amount of current from a highly regulated so=
urce. 

       Putting magnetic components on a seismo=
meter arm makes it liable to pick up  a lot of interference signals. Yo=
u would need to completely and effectively screen the sensor and this is not=
 easy to do, or cheap.

       You can make Hall-Effect sensors, but t=
o get good seismic sensitivity, you have to use strong magnets and the movem=
ent range is then less then +/-1 mm. Not a problem for a simple pendulum or=20=
a spring accelerometer.

       Strain gauges will not give you the sig=
nal to noise ratio necessary.

       Simple optical systems using a miniatur=
e tungsten filamant bulb, a moving slot shutter screen and two Si large area=
 photocells connected differentially can give a better performance. Eg VTD34=
, BPW34 or larger. For a Lehman, you might choose 10mm or longer photo cells=
 - see Silonex. I run the bulbs at about 0.8 of their rated voltage to give=20=
a very long and stable light output and use a voltage regulator + bypass tra=
nsistor to do it. You need the large area photocells to get the S/N ratio up=
.. The tiny 1mm square Si cells do NOT work well enough, particularly those i=
n LED form cases with an end lens.  



       Regards,



       Chris Chapman