On Mon, 18 Apr 2011 06:28:24 -0700, Winston
wrote:

wrote:
On Sun, 17 Apr 2011 13:33:11 -0700, Winston
wrote:

Bob Engelhardt wrote:

The "Build Your Own Magnabend" section is available and it's very
interesting.

Alan reveals some very interesting stuff here that I suspected but
did not know for su

* The 'E' core is more efficient than the 'U' core.

* Clamping force for the 'E' core falls off significantly on the ends
because of the narrowed middle pole.

* Regular door hinges are a practical substitute for the complicated
biaxial set used on the commercial model.

Thanks again, Bob!

--Winston

Good stuff and very useful but I'm a bit worried by the simple
statement that the E core is more efficient.

A lot depends on both the assumptions made and the detail
design. The E core design may not necessarily the most efficient.
However the differences are not large enough to be important in
home brew projects.

Alan Unless you have very good milling facilities then I suggest
Alan that you go for a U-type magnet body. Although this design
Alan is not quite as efficient as an E-type body...

Given the inverse - cube character of magnetic attraction vs.
distance, it seems axiomatic that the two additional gaps
of the U core design should result in higher magnetic
reluctance than that of the E core, all else being equal.

Alan could be wrong on that point, I figured that he is in a
position to know, however....

--Winston

Just a few comments on magnetic attraction.

There is a comprehensive discussion in
http://en.wikipedia.org/wiki/Magnet

A brief summary may be useful.

Attractive force varies as square of the field strength

For a NS magnet of fixed field strength the far field strength
varies as cube of distance so the attractive force varies as
sixth power of the distance.

Close to end of long magnet (i.e.in small air gaps) the field
strength only varies as square of distance - so the force varies
as the fourth power of the distance.

These comments define "magnet field strength" as Amps/metre i.e
fixed coercive force. The resultant Flux density B is then
controlled by the cube or square relation.

With real magnets coercive force is a function of working flux
density so this is only true for an ideal square loop magnet.

This is a fair approximation for rare earth magnets but with
alnico or ceramic magnets field strength drop is even faster
because the working coercive force drops as the flux density
decreases

Jim