[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Spark gap



Hi Gary,

I'm not sure I'm following you, and I don't see where I suggested that smaller current supplies suffer.  I was trying to point out that static gaps don't perform well with very high powered (beefy) power supplies.  Pole transformers, potential transformers, even MOT's, do much better using RSG's over static gaps.  I'm not certain just where to draw the line, but I would put a 200mA PSU on the beefy side of the line.
I must have then misread your sentence then. 200mA is riding on the edge of the fence and yes, static gaps have their limits.

Measuring the actual BPS rate on a static gap has always been difficult.  The only way I know to measure it is by using a storage scope and counting the bangs in some time interval and doing the math.  Have you actually scoped and measured your BPS?  I have, and this is how I arrived at the numbers I cited.  Particularly when using LTR caps with a static gap, non-linear ferro-resonance effects cause totally unpredictable secondary currents far in excess of the faceplate rating to occur, and I don't believe that any method exists to reliably calculate BPS.
I have scoped the gap but have not gone as far as to store the data and run the numbers for bps. It certainly would be difficult to measure and your method is perfectly acceptable. I have a new data acquisition tool that can capture 14,400 samples per second. I should attempt a method for recording my gap using it. I understand the gap is chaotic, but we have to look at the average over a decent time frame. I realize the bps will be up and down from event to event. However, I expect bps to be different from one coil to the next and it sounded like a blanket statement of 150 to 300 bps for static gaps, and that just doesn't sit well with me. Maybe I misinterpreted that as well.

Let's first recall that in the tables which recommend the optimal capacitor size based on the transformer, there are separate columns for static gap and for sync RSG, the sync RSG using a much larger cap.  With my 15/60 NST and a static gap, I found that a .02uF cap worked best.  Anything larger caused excessive current draw and ran poorly.  But the same NST with a 120BPS sync RSG ran best with a .04uF cap.  I wish that I could explain why that a SRSG allows a larger cap to be used without a lot of hand waving, but it is a well documented fact.  Do you disagree?
It's not that I disagree, it's maybe that I view the cause differently? With your 15/60 and static gap, the .02uF cap worked best and larger sizes caused excessive current draw and ran poorly. I would certainly think so. Even .02uF is pushing it. The time to get to peak firing voltage to arc the gap is getting excessive at 18ms! Try a .03uF and you are nearing half a second! That's the problem I see and it makes perfect sense to me. The rotary has the option at firing at a much lower voltage (about 14kVp with the .04uF cap). This is "why" the rotary can use the larger cap size. Now the static gap could also be made to work, but the total gap is going to have to get really narrow. Static gaps are fixed and usually the only way to narrow the gap is to tap inward on pipes (which increases thermal losses).

However, you say larger cap sizes allow a larger bang size. In general, yes, but not always. Your 15/60 you might calc at 12 joules if you didn't consider the voltage at the cap at each firing of the RSG at 120bps. Energy drops down to 3.92j! The .02uF cap is up at 4.03j. So, this is part of my problem. To state there's a higher bang energy is not correct (and I've heard that stated many times but without any data to support it). I think the larger cap size performing well is for another reason such as arc dynamics and transfer functions.
Spark length is all about using the biggest bang possible.  And it guarantees that
each bang is identically and maximally size
I assume you're questioning the part about spark length being correlated with bang size?  I think it was one of the folks (maybe Finn?) with a DRSSTC who had the ability to vary bang parameters independently, found that varying just the bang BPS varied the streamer brightness but the length was relatively unaffected.  But varying the bang size directly affected the streamer length.  So, optimizing a TC to have the largest bang size at a relatively slow BPS will yield the longest, thinnest sparks, while choosing a small bang size at a very high BPS will yield shorter, but brighter sparks.
Using a larger cap doesn't always correlate to a larger bang size. BPS "must" be slowed in order to accomplish that and the transformer "must" be able to supply the current, if not voltage drops and bang size is not increased. I agree that a larger bang size will yield longer sparks in most cases. I don't agree that adding a larger cap will do it unless the transformer is beefy enough to supply the current. In your case with the NST, it doesn't increase bang size, however the RSG allows the firing anyway, so something else is the cause.

Am I making any sense?
Take care,
Bart

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla