Re: visualizing fields near charged objects

[Ludwik Kowalski <KowalskiL@MAIL.MONTCLAIR.EDU>]

John Mallinckrodt wrote:

> With everybody implementing cylindrical coordinate spreadsheet
> solutions, I thought I'd better jump in there as well.  ...

My code uses rectangular coordinates (square cells in 2 dim) for
an object with cylindrical geometry. This is silly. But I am not
sure how to implement the "iterative averaging" for cylindrical
cells. So I have some questions.

For square cells we perform simple horizontal and vertical
averaging between neighbors. (4 cells in 2-dim or 6 in 3-dim).
This seems to be natural for cells of equal volume. But cells
of a cylindrical grid usually do not have that property.

1) Is it appropriate to perform simple averaging for usual
cylindrical cells (dr=const)?

2) If not then what to do? Should it be a weighted average
each time (bigger neighbor has stronger influence)?

I also have a question about the first simulation of JohnD,
the one in which rectangular coordinates were used. But
first let me look at it again; perhaps I will see the answer?
Ludwik Kowalski