Leave a comment

reverancepavane December 12 2007, 16:37:07 UTC

These are my thoughts immediately upon seeing the problem.
(1) The theoretical physicist in me is treating it as a reverse wheatstone bridge (applying a known potential differance to both points and integrating the resistance over the current path elements).
(2) The computational physicist in me is considering the problem in a similiar manner, except reiteratively analysing the current flow according to the resistance of each path element (effectively the same thing but less accurate), which is not as mathematically accurate, but you are going to lose any precision beneath machine epsilon, anyway.
(3) But the experimental physicist in me is looking for my soldering iron and multimeter, as we speak. I do consider myself to be an experimental physicist, btw. It's more fun.
The motto of this story is that mathematicians shouldn't set puzzles for physicists. We'll just ignore the interesting maths in order to find a workable solution that makes mathematician pull their hair out and say "you can't do that." We've had plenty of experience ( ... )

Reply

kaet December 12 2007, 22:15:03 UTC
Greens functions!

I spent about a week on this very problem, some time in 2005 ish, :). I think google started it.

Reply

reverancepavane December 13 2007, 03:53:37 UTC

Greens functions!
Thankyou very much. I'd almost successfully forgotten that part of my life (which accompanied the worst exam I've ever encountered at university). Admittedly it was only four questions of sixteen or so consecutive parts. Biut who could calculate the Green's function of a typewriter in a filing cabinet [blatant exaggeration warning]. And who would want to? I think people got marks for successfully writing their name. It's the first time I've seen the lecturer cry at an exam when he realised that it might have been considered a "tad difficult" and "unexpected" by his students.
I'll be over here in a ball in the corner if anyone needs me.
PS: Gold, and 10^-2 is Silver (EIA standard anyway).
PPS:

Reply

kaet December 13 2007, 12:18:39 UTC
Oh, they fill me with teh fear, too! That and Gramm-Schmidt orthogonalisation! Well, Gramm-Schmidt is not actually that painful. At Uni they were the kinds of things where examiners took great pleasure in claiming they were testing your knowledge of the Universe when they were actually testing
  1. Can I remember integral tables I could look up were I not in an exam
  2. Can I not drop a minus sign or a factor of two, on something I could feed through Mathematica, were I not in an exam
I guess it's very good training for obscure jobs in Whitehall (where my Uni seemd to excel), where pedantry and a memory optimised for long tables of facts are probably your way to advancement ( ... )

Reply

kaet December 12 2007, 22:37:29 UTC
But the experimental physicist in me is...

... desperately trying to remember the resistor band colour for 10^-1? :)

Reply

robert_jones December 13 2007, 13:31:25 UTC
But are you confident that an infinite grid can be modelled by a large grid? I mean, that's an awful lot of grid you'd be ignoring, even though it is a long way away.

Reply


Leave a comment

Up