Thanks Alan. The 'spark' is actually blue rather than the white that appears on the video... would that help identify the gas?

I have never built a Marx generator myself but I read that often lenses are placed between the gaps so that some ultraviolet light from one gap is focused onto another gap, so that after the first gap breaks down, the ultraviolet light from the discharge stimulates the next gap to break down, and so on.  If you use a safety gap, place an opaque barrier to block the light between the safety gap and the Marx generator gaps so it doesn't accidentally trigger the Marx generator pulse.

The thing about Marx generators is that they are so easy to build, and given the absence of semiconductors, very robust (if using appropriate capacitors or MMC arrangements). They work well enough without the lenses unless you need that level of precision. The safety gap has not been an issue, since I just have to make the gap narrow enough to  permit a plasma arc to strike when disconnected from the Marx, but wide enough to permit the first Marx gap to trigger before the plasma arc can be established. It's all grist to the mill when you're "learning by playing".

By the way if you want to read about an interesting Marx generator, read about the Z-Pinch Marx Generator for pulse power

https://en.wikipedia.org/wiki/Z_Pulsed_Power_Facility

It's used for nuclear fusion energy research.

"As of 2012, fusion shot simulations at 60 to 70 million amperes are showing a 100 to 1000 fold return on input energy"...

Fascinating... I had no idea that fusion had come so far.

Thanks Dan.

I thought a LOPT was a type of flyback transformer where the flyback works on the principle of storing energy in a flux gap in the ferrite core, but I could be wrong...

You have hit on an interesting point for me. I don't have any expertise in this field so I've often wondered why the word 'flyback' is used in so-called "flyback transformers" and "flyback diodes". Regarding the former, I presumed that it was a linguistic hangover from the use of ferrite core transformers for generating not just the HT for CRTs but also the line-scan sawtooth (and also the frame-scan sawtooth, though it remains a mystery to me how both sawtooths can be generated on the same core). Hence my conflation of the terms "flyback transformer" and "Line OutPut Transformer". The term "flyback diode" remains an enigma to me since I've only encountered them in the context of back-EMF protection when driving inductive loads.

You may have too few turns on the primary and the voltage could be too high on the secondary.  It is best to start with more turns and be conservative so you don't overvoltage the transformer.  With the cold cathode fluorescent load, there can be a continuous discharge and this prevents the voltage from getting too high, but basically once the capacitors on the Marx generator are charged they are like an open circuit, and this could cause the LOPT to arc internally because of the high voltage.  The diodes in the LOPT could be damaged and therefore it may be charging and discharging the capacitors, but you would still see an arc from the HV terminals.

I was copying the pattern of the LOPT that works, but in both cases there is a constraint on the number of turns given the gauge of wire being used for the primary winding. The possibility that the reclaimed LOPT has been damaged by internal arcing hadn't occurred to me because I was able to draw 2cm arcs externally, but this no longer appears to be the case so you may be correct that the device has failed in this manner. I did try it with a Cockcroft Walton voltage multiplier (just in case it was generating AC) but this produced nothing either.

Do you have a safety gap for your secondary for your LOPT?  This might prevent the voltage from getting too high on the secondary. 

I haven't been using a safety gap because the working system I'm copying doesn't use a safety gap, but it would seem prudent to employ one from now on. Thanks for your interest and for your advice Dan.

Your secondary and topload should do fine for first adventures with a SSTC2 circuit.

So please keep us updated on your progress in this thread

Will do Mads, though it's a low priority task so progress will be slow.

I recently bought a tiny Chinese "Tesla Coil" out of interest, and it prompted me to build a bigger version from some old plastic waste pipe, copper wire reclaimed from a discarded CRT-style TV, and an old metal slinky spring for a top load. I boned up on the Slayer Exciter circuit and built one to drive my coil. It resonates, and the coil causes fluorescent tubes to glow, but now I'm interested in getting some sparks from this thing. The dilemma I face is that I suspect that the Slayer Exciter is not powerful enough to get the sparks, and the alternative spark-gap driver would need to be tuned to the coil (which is an inconvenience that I avoided by using the Slayer Exciter). Any guidance for a complete novice would be greatly appreciated.

Misfiring is often due to imbalance in charging resistors/inductors,

The resistors are all 1W 1M 1% metal film. I guess I'm really asking about the mechanism of failure when the main gap gets too large. I would have expected all the capacitors to become fully energized just as before, and so the energy should be finding some easier path than the main gap, but I don't see any evidence of that. Perhaps the capacitors aren't getting to full charge, but I don't see how that would follow from just increasing the main gap. Still perplexed.

Hi guys, I've been busy building my first ever Cockcroft-Walton voltage multiplier and my first ever Marx Generator. I don't know if it's permitted to supply video links here but if so then I can show provide a link to a YouTube video that shows the setup. My question concerns misfiring. I'd like to understand what is happening in the Marx Generator when it occasionally misfires, with the minor spark gaps showing weak sparks (or maybe they are just plasma paths). I can't see any obvious shorting pathways, so I'm wondering where all the energy is going on the misfires. Thanks for reading.

I have noticed this too and it appears to be down to voltage, current and frequency.

Type 1 is usually a discharge of static electricity, or a low frequency spark which is essentially a series of very short discharges.
(Van de Graaff generator, Wimshurst machine, some induction coils)

Type 2 is a DC or high frequency AC arc. From what I have observed, the thin arc and violet colour are associated with low current.
(CRT flyback transformer, ioniser, plasma ball)

Type 3 is a low frequency AC arc with high current. The flame appearance is because the heat caused by the high current pulls it up into a flame shape. The hum or buzz is the 50Hz or 60Hz frequency of the mains.
(Neon sign transformer, microwave oven transformer)

Thanks SB.
That would seem consistent with my experience... I get sparking from a ten-stage voltage multiplier but a violet plasma arc when current limited using ten 720k resistors in series, and I get the cream-coloured flame on a Jacob's Ladder driven by a flyback transformer (the driver draws around 5A). I was wondering what is happening to the air (oxygen and nitrogen molecules) to give these different phenomena.

Hi guys, my name is Andy Kay and I'm from England. I had an interest in electronics from an early age, and after attending university I started work in the industrial electronics field, moving on to telecommunications, and quickly found myself in software development. I'm now retired and revisiting the electronics scene. High voltage has caught my interest, Tesla coils, voltage multipliers, Jacob's ladders, etc. and I'd like to learn more about this stuff. Hoping this is the place to find instruction and inspiration.

Edit (9-May-2018):
I've been here a little while now and I've enjoyed reading the posts. I had tried to sign up to 4HV but got no response, so very happy to have found this site. Just want to say a big thank you to Mads.