[BULK] Re: [TCML] copper tube stationary gap - was: NST rebuild good

[bartb]

Hi Jim,

The bevel sounds like a good idea to me. Anything that will help focus 
the airflow in and around the electrodes is a plus. Back when I tried 
the brass stock, I used apparatus to similarly focus the airflow over 
the brass. But without success. The brass would work fine for a while, 
but the thermal rise could not be stopped and once it was up to temp, no 
amount of airflow helped (sparks fizzled out). This wasn't due to the 
material itself but the fact that it was solid stock. It took longer to 
heat up but impossible to cool down unless I ran really low power. I had 
similar issues with small 1/2" copper tubing. With large enough pipe 
stock and external cooling, a relatively cool thermal balance could be 
reached even at high power.

I've been meaning to epoxy a thermal probe on one of the electrodes to 
gather temperature data for power input and of course performance. It 
probably wouldn't be that difficult to calculate needed cfm for a given 
electrode shape and size for power at the gap (at least based on a 
common configuration). I think most of us still use the trial and error 
method.

Let us know how the new gap works out.
Take care,
Bart



Jim Mora wrote:
> Hello All,
>
> My 8" coil used a single gap of 1" brass dowels on threads (easy adjustment
> and cleaning) ( somewhat obscure RQ) blown by a powerful leaf blower coupled
> to a 12" long 3" to .5 inch nozzle. It quenched a 15KV/120ma very well.
> Since I have acquired a 14400/5KVA PT (GE), it is torn down for a rotary
> retrofit.
>
> I am going to experiment with a static gap much like Bart's which will be
> blown by a pair of side x side pancake blowers (a lot of air). To focus the
> air across the bottom, assuming it will rise just fine, I cut a 45 degree on
> the face of the phenolic which is tapped and threaded to except the copper
> tubes. The clincher is, I beveled the top edge to create a Bernoulli effect
> and the air stays exceptionally well focused across the tubes mounting
> surface rather than traveling up and away before it reaches the length of
> the tubes. I'll post results when it runs.
>
> Jim Mora     
>