Re: Flat conductors (was I need help).

[kowalskil <kowalskil@MAIL.MONTCLAIR.EDU>]

Referring to my measurements JohnM asked (on 2/26/02):

> Your experimental data generally reflect the type of distortion
> that would necessarily be caused by the existence of the
> boundaries, but they also appear to be in pretty significant
> disagreement with the theoretical requirement that the
> equipotentials be perpendicular to the edges of the paper.
> Did you measure carefully right out to the edge to see if
> the lines become perpendicular?

Yes, the lines are perpendicular to the paper borders. I just
conducted an experiment whose purpose was to demonstrate
the opposite. The silver dots were at (6,2) and (22,18); these
are Pasco paper coordinates in cm. With this geometry the
central bisecting equiupotential line is at 45 degrees with
respect to the x and y axes. I expected this line to cross the
paper boundary at 45 degrees. And that what I observed,
all the way up to last cm or less. Then the line turned
toward the margin and intercepted it at about 90 degrees.
Other equipotential lines were also examined carefully
near the boundary (mm by mm) and were found to be
practically perpendicular as well.

If I had more time I would repeat the measurements
after cutting left and right margins to make the a square
sheet. How would the central equipotetential know to
which margin to turn near the corner. The one I observed
turned toward the nearest lower margin. So much so far.

Why should the gradient be parallel to the paper boundary
everywhere, as it seems to be?  I know why the E lines are
perpendicular to conductive surfaces in electrostatics but I
do not know why should they be parallel to the surface
inside a conductor connected to a battery.
Ludwik Kowalski