rotary gap electrodes


I'm asymptotically approaching the state of rotary gap construction known 
as completion.  I thought I would ask for others input before proceeding.

The design of my rotary has two 1" thick 14"x14" pieces of plexiglass to 
hold the bearings and act as side shields.  The stationary electrodes are 
going to be mounted to the plexiglass and poke into the box where the 
disc is rotating.  I am using 1/8" tungsten TIG welding rods for 
stationary electrodes.  The input power will be less than 10KVA.

My questions are: How much cooling of the stationary electrodes do I 
need to do?  Will putting the W (tungsten) rod into a piece of 5/8" brass 
rod 2" long be enough to prevent the rod assembly from getting so hot 
that it melts the plexiglass?  I have also considered some sort of 
non-thermoplastic insulator to surround the brass.  Another option is a 
1" rod of Al with the W rod in it.  The Al could have cooling fins turned 
on it too if necessary.  I don't figure ohmic heating will be much of a 
factor, but the arc will be hot and so heat the electrodes.

Any input would be appreciated.