----- Original Message ----- 
From: "meredith lamb" 
> 
> A Interesting question (too me), has arisen while briefly
> utilizing some neodymium magnet disks.  The normal coil/
> magnet scheme for seismometers is to use North-South
> poles facing each other; with the coil inbetween.  OK....
> what if same poles are repelling (~ North-North); with the
> coil inbetween.  Is there really any difference?  Is the normal
> attraction of unlike poles really better than the repelling
> variety?  I've made a simple jig with same poles facing each
> other and I don't really see any difference from my previous
> output background in the normal attraction arrangement,
> though admittedly I have not yet made different comparison
> models of this subject.

OK - with in the restriction that I have not formally designed
magnetic circuits - here is some straight scoop.

Our happy magnetic sensor seismometers are basically
fragile, temperamental permanent magnet motor/generators.  
Same game except:
      - we strive to achieve "reasonable" linearity
          ( given the same speed of movement,
            the generated voltage about the same in all
            parts of the zone of movement)
      - supporting structures are very different
           (hanging pivoted support instead of
             rotation through 360 degrees)
      -   and more or much more fragile
               YES!!!
      - we strive to achieve voltage rather than current
          (lots of fine wire)

Our seismometer motor/generator uses the identical effects
  a) a current through a wire in a properly arranged
       magnetic field will produce a force.

  b) movement of a wire "cutting 'magnet flux' "
       will produce a voltage

OK  

Assume more voltage is a good thing to help overcome
 other noise at the very slight speeds of the sensing wires
  caused by  distant earthquakes.

So,  assuming a reasonable structure, we want:
   - lots of length  wire in the magnetic field
      (the more wire - not counteracting eachother-
        the more voltage)
  - lots of magnetic field
       (the more magnetic field, the more voltage)


So - at last - a discussion of the magnetic field.

Lots is good - so putting magnets in AIDING series
     helps give lots of magnetic field ("flux").
Lots is good - so putting iron (permeable to magnetic
    "flux" ) in the return path helps give lots of magnetic field
Lots is good - so putting POWERFUL magnets in aiding series
     is even better

> 
> A untested thought with this arrangement does occur...and
> that is, if the field is repelling, would not the immediate field
> around the two magnets be more expanded; would might
> benefit the coil induction abit more; especially in the plane
> between the magnets outward where the coil is?  This route
> "might" benefit coils which are larger in diameter than the
> magnet pole faces.

Yes - but - but - but - putt - putt - putt

Between the pole faces is where you get any attempt
at linearity.  Computing/measuring linearity outside
the pole pieces gets spooky.

> 
> I think the magnetic field around a normal magnet scheme
> and a repelling scheme would still be kind of a oval figure 8,
> with like a board and iron fileings over the magnets.

Yes - but I contend that your linearity and well as the net
magnetic effect "will go all to pot".

> 
> Their is of course ways to mount magnet disks in the normal
> fashion....but....if the possibility exists that same poles function
> as well; this might make it much easier mechanically to just use
> a backing material, glue and brass screw rods between them
> for adjustable spacing arrangements.
> 
> Another facet is that if they are repelling, the likelyhood of
> crashing magnets together and total damage is reduced more.

True - but I contend this benefit is very small compared to
the "damage" to the sensitivity and linearity of the unit.

> 
> Any comments or actual experience with same subject out
> there?

Comments? - comments? - you should never ask for comments  ;-)
The likes of me will do just that   ;-|

-----------------------------------------------------------

That said, I will suggest that us amateurs who do not want
to build solenoid type magnet structures and coils in
an attempt to reach say 2% linearity can use the following.

An old radar oven has a magnetron.  A magnetron (by definition)
has a magnet.  The unit works best with a strong, reasonably
linear magnetic field.   The unit I picked up has two magnets
5 cm in diameter mounted in an iron housing, with a separation
of 2.5 cm.   The iron housing even has the two opposite sides
open ready to allow the coil access to the field.

I find this ideal for fun seismometers.  
You just:
  1) separate the two parts of the iron housing, 
  2) remove the "tube" of copper and ceramic,
  3) put the iron housing back together,
and you have a strong accessible magnetic field 

You will note that the magnets have a hole in their
centers - this actually increases linearity - the center
is not so strong relative to the sides  :-))

----------------------------------------

Oh golly - one more thing - hang in there

Where to hang the coil in the magnetic field?  ???
Yes - just where?

How to talk about it?
Most of us build a coil in a toroid (a donut).
Let us lay the donut horizontal, and make a 
vertical cut down through the center axis of 
the donut.   We are now looking at two 
smaller circles.  OK?

Let us call the center of one of the circles
  "a" 
and the center of the other circle 
  "b"

I contend, that the  "a" part of your coil should
naturally hang in the center of the magnetic pole pieces.
and the center of your donut should be outside the pole pieces.

----------------------------------------------

Cheers
   Ed Thelen

P.S.   You, the reader, have remarkable patience.
Thank you


_____________________________________________________________________

Public Seismic Network Mailing List (PSN-L)