Monday, April 9, 2012

modifications to distortion triggered switch

After I'd mostly finished my triggered switch, I learned a few things before I got around to testing it, which is very good.

Thing one which I learned is that I am going to need much higher clamping forces to electrically join the brass discharge electrode to the aluminum flanges of the switch. There are formulas - arrived at empirically - for the minimum surface area through which you can pass a given amount of "action" (coulombs, charge, energy, area under the curve) with considerable variances depending on whether it's for pulsed power or a buss bar in a substation.

(if I could have machined them out of a single piece of brass, I would have, okay? Maybe on the next switch.)

That epiphany meant I needed to remove the brass screws from the electrodes and drill and tap them for big fat machine screws coming in from the back.

I also machined off the radius from the rear of the electrodes, since it serves no purpose here (they came with it and I was being lazy) and removing it increases the contact surface area.

And finally, I added a relief around the bolt hole to accommodate an o-ring to ensure a gas tight seal, because this switch will be operated (at least at first) at low pressures - "operating on the left side of the Paschen Curve", as the pulsed power boffins would say.

All of these modifications are evident in the photos below:


It should be intuitively obvious to the most casual observer (I just love that snotty phrase) that machining the back of the electrodes changes the gap distance, which implies a change in operating pressure for any given voltage.


How this will sort itself out in forward resistance in the final application remains to be seen. I may either build a new switch, or - more likely - make one new shorter housing piece, keeping the shorter of the old pair, and fabricate six new tie-rods. (I refer to the white Delrin rods which hold the switch together with cap screws - I have called these 'spacers' elsewhere. I shall endeavor to call them tie-rods henceforth, forthwith and like, going forward, man.)

Yeah, one of those end caps could use some cleaning up on the electrode mating surface. That would be the end cap whose center boss did not have to be turned down because I did not mill its o-ring relief too deep. Ahem.

Also, I found after I had machined off the radius of the electrodes, I had to finesse the edge of the gas ports on the inside, as they were covered up by the new base profile of the electrodes.

Really, the only reason this thing isn't fully reassembled and ready to test today, right now (I've got some connection brackets rigged up for it too) is because I don't happen to have the right length bolts handy. I'll be running to the barber and neighborhood hardware store later this morning, after my morning teleconference.

This work was supported by the Joss Research Institute, Laurel MD.

No comments: