Re: specification for the uniform field region

[Richard Jones <richard.t.jones@uconn.edu>]

Jim,

I can imagine a number of definitions for the effective field boundary, 
all of which correspond to the hard edge approximation in the 
step-function limit.  I agree that the particular choice you made is 
specifically suited to our requirements, but at the same time it seems 
difficult for a vendor to guarantee because it involves integrals along 
long curved paths through the magnet whose functional shapes themselves 
depend on integrals of the magnetic field.  This means that the shape of 
the effective field boundary is a highly non-local function of the 
global field.  Compare that with the Mainz spec that involves integrals 
along straight lines.  I can imagine that just about any vendor could 
imagine how to determine that his magnet meets that spec.

-Richard Jones

Jim stewart wrote:
>
> Hello Richard
>
> The effective field boundary is the width of the magnet if you are 
> using a hard edge approximation.  This means you need to integrate the 
> field and then divide the integral by the central field for the 
> effective length.  Therefore I have specified for every z the central 
> field, the point where the field has dropped by 3x10^-4, and a point 
> in the fringe field.
>
> I am not sure if this is sufficient but I cannot imagine how any 
> magnet could fulfill these requirements and not be fine for us. 
> Requiring that the EFB is a straight line parallel to the pole axis is 
> a pretty strong boundary condition for the shape of the fringe field. 
> What would you propose?
>
> Jim
>