Racing sparks without terminal sparks / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Racing sparks without terminal sparks

1 2 next

Move Thread

LAN_403

dex

Sun Jan 10 2010, 08:12PM

Registered Member #2566 Joined: Wed Dec 23 2009, 05:52PM
Location:
Posts: 147

Can racing sparks appear along secondary coil if topload has large enough ROC to prevent streamer formation from TC terminal? I suppose the answer is yes ,but I don't know of an experimental evidence supporting this.

Steve Conner

Sun Jan 10 2010, 08:20PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

If the secondary is too short for the ROC of the toroid, but the coil is otherwise well set-up, it'll flash over down its whole length, either down the inside of the tube or along the windings on the outside. I've seen both.

Racing sparks are a slightly different thing, they're flashovers along only part of the secondary's length, caused by an uneven voltage gradient.

dex

Sun Jan 10 2010, 08:32PM

Registered Member #2566 Joined: Wed Dec 23 2009, 05:52PM
Location:
Posts: 147

I am talking exclusively about racing sparks.Not about complete secondary flashover,which can occur instantly or sooner or later after ignition of local racing sparks,or about flashover between secondary and primary.I don't limit my question to toroids.It could be big enough sphere (spheroid) too.

Steve Conner

Sun Jan 10 2010, 08:40PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hmm, well in that case, I'd say that racing sparks don't really depend on the toroid ROC.

I believe that local racing sparks are caused by higher resonant modes of the secondary, whereas the main output voltage is caused by the fundamental 1/4 wave mode. And as you learn in calculus 301, the resonant modes are orthogonal to each other: energy in one doesn't imply energy in the others.

It's harder to run a spark-gap coil at high power without exciting the higher modes, because the excitation from the spark gap is broadband. But a solid-state coil can do it because of the high spectral purity of the inverter output. This is why (IMO) SSTCs don't suffer from racing sparks, but flash over instead.

The Seibt coil demonstrates higher resonant modes:

dex

Sun Jan 10 2010, 08:56PM

Registered Member #2566 Joined: Wed Dec 23 2009, 05:52PM
Location:
Posts: 147

Seibt coil is at least 3/4 wave or even higher resonant modes excited coil.My concern is only 1/4 wave coil,without large tunning error.Someone on TCML suggested that streamer breakout from terminal might be one cause of the racing sparks theoretizing that sudden release of energy sends heavy voltage disturbances and significant content of higher overtones along secondary coil.Resulting in racing sparks.Experimental evidence seems to not support this as the cause.Without doubt racing sparks are the result of too high voltage between turns.So,this is the question.

Steve Conner

Sun Jan 10 2010, 09:17PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, I see your point. I don't think the release of energy due to streamer breakout is that sudden, though. On this old page:

I show scope traces of the output voltage from a coil with and without streamers. The streamers take several cycles to drain the energy. I used a breakout point so I could remove it to suppress the streamers, and you could argue that this gives a more gentle start to the breakout, but Richie Burnett showed that a spark breaking out from the bare toroid behaves similarly to one from a breakout point, in terms of RF loading. Once it's broken out for the first time, the mess of ionized air works as a breakout point for subsequent bangs.

If the TCML guy's hypothesis were right, then arcs from the topload to ground should cause much worse racing sparks, because we know that they dump the topload voltage to ground almost instantly, and therefore should produce much more high-frequency energy. Terry Fritz published scope shots of it, and in the Colorado Springs notes, Tesla writes about using ground arcs to "up-convert" the output of his coil to higher frequencies.

dex

Mon Jan 11 2010, 11:48AM

Registered Member #2566 Joined: Wed Dec 23 2009, 05:52PM
Location:
Posts: 147

Steve McConner wrote ...

If the TCML guy's hypothesis were right, then arcs from the topload to ground should cause much worse racing sparks, because we know that they dump the topload voltage to ground almost instantly, and therefore should produce much more high-frequency energy. Terry Fritz published scope shots of it, and in the Colorado Springs notes, Tesla writes about using ground arcs to "up-convert" the output of his coil to higher frequencies.

Yes. That's the point and this is the evidence from every day coiling practice that racing sparks are NOT triggered by sparks to air .Sparks to ground cause even more dramatic overtone disturbances but racing sparks are absent however.As known that problem is closely related to coupling between primary and secondary coil.If the system is overcoupled the racing sparks can be created.Althought "overcoupling" is not the most appropriate term.With coupling as low as k=0.1 one can get racing sparks too if the bang energy is too high for secondary structure.Isn't that right?But there is more to it!How exactly is too high local voltage gradient among secondary coil turns created when the coils are "overcoupled" is quite a mystery me.

Steve Conner

Mon Jan 11 2010, 12:07PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

OK, so I agree with you and disagree with the TCML guy.

wrote ...
How exactly is too high local voltage gradient among secondary coil turns created when the coils are "overcoupled" is quite a mystery me.

Antonio de Queiroz once showed that the spark-gap Tesla coil can be modeled as a 4th order bandpass filter excited by a step function. Increasing the coupling broadens the passband of the filter, and because the step function contains energy (actually to keep Scott Fusare happy: voltage, not energy :) ) at all frequencies, everything in the passband will get excited.

(Scott: it follows that the PSD of the excitation is equal to the passband of the filter, not "DC to light" as I wrongly said. The spectrum of the excitation voltage is still DC to light, but the PSD is different to the voltage spectrum, because power = voltage * current, and the network only draws current at its resonant frequencies.)

I think racing sparks are caused when the passband gets wide enough to include one of the secondary's overtone modes.

Note the overtone mode frequencies aren't nice harmonic multiples of the fundamental, so they're not excited by square wave drive. A Tesla resonator is more like a bell than an organ pipe. In particular, the toroid lowers the fundamental frequency without affecting the overtones. Paul Nicholson's TSSP is a huge source of information on this.

dex

Tue Jan 12 2010, 12:03PM

Registered Member #2566 Joined: Wed Dec 23 2009, 05:52PM
Location:
Posts: 147

First remark (about modelling SGTC as bandpass filter) has a very little if anything to do with voltage gradients on distributed secondary coil structure.OTOH,the link you provided really deals with effects and simulation of distributed secondary circuit.But,where do you see in it a direct simulation indications that increasing coupling from k=0.1 to k=0.22 significantly affects local voltage gradients?I do think that racing sparks has something to do with constructive interference of very high overtones (higher than 10x fundamental res. frequency) but connection of this thing with increasing coupling (k=0.1...0.2 is the range of interst for coiling) is WEAK.For example it is weaker than the effect of k increase on the increase of the efficiency of energy-transfer from primary to secondary.As k is increased from 0.15 to 0.22 top volts increase by less than 10% in a typical SGTC.That returns us to two already asked questions I opened this thread for:

1.
Can racing sparks appear along secondary coil if topload has large enough ROC to prevent streamer formation from TC terminal?
2.
With coupling as low as k=0.1 one can get racing sparks too if the bang energy is too high for secondary structure.Isn't that right?

Without much philosophy ,I'd like to know of clear experimental confirmation.

Steve Conner

Tue Jan 12 2010, 12:36PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I have no experimental evidence that would confirm or deny either of these.

The only time I ever saw anything that could be described as "racing sparks" was when I put a too-big tank capacitor and toroid on my spark-gap coil.

You can see two small spots of light near the top of the secondary, however others appeared during the run, with the same spacing. Afterwards, regularly spaced burnt spots could be seen on the secondary where the discharges had been. The regular spacing is what makes me think of overtone modes.

The point I'm trying to make is that the voltage profile on the secondary is the sum of the instantaneous voltages due to all of the resonant modes. If any of the overtones are excited, the voltage profile will be uneven, with ripples in it. My hypothesis is that these ripples cause the local racing sparks. If the voltage gradient due to the fundamental was already almost high enough to cause flashover of the whole secondary, then even small ripples could start local breakdowns.

But I have no real evidence for this, I put it forward as a point for discussion.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob