The So-Called Don Smith Generator

[Tesluh]

Dezeinstein or anyone who cares to comment,  How would you drive the l1 coil of the Don Smith "white board" device?  He used a bertonee nst that is out of production and I cant find one no matter how hard I try.  And even then I don't think anyone has video of that exact board "in use". 

I have 7 different nsts and fried a couple.  trying to drive it at radio frequencies, 20khz or more but none of my nsts seems to work right for this arrangement.  he has 2 outputs rectified and the spark gap returning through the ground terminal but most modern nsts have gfci on the ground terminal (best I can tell).  I think the one I am trying to use now is tripping the gfci.

looking for 1300v so my output is about 4kv.  my circuit looks just like his, though I calculate about 150khz resonant frequency with .2uf cap across l1 and .047uf across l2.  I could use capacitors to lower it if necessary.

would love to see drawings/details of what you think the most "sure to function" circuit/components would look like to make even a fraction of the power you are describing. 

I am having difficulty picturing how you would connect l3 and l4 step down transformer and charge 48v super cap banks while discharging them with an inverter to the load.

[gyulasun]

Hi Tesluh,

I cannot answer all your questions but I can tell you a few things.

--  you can buy NSTs at ebay without gfi but they are expensive, plus the shipping cost comes to it
    (for instance: http://www.ebay.com/itm/252629542643 ), or just search ebay with words "neon sign transformer no gfi"

-- a cheaper solution would be to buy this HV transformer and some components and build your own "nst":
   http://www.amazing1.com/products/miniature-ferrite-high-frequency-transformer-7kv-10ma-20k-100khz.html
   There is a downloadable PDF file in the page that includes a schematic circuit for this transformer with the other needed components in Page 10,    labeled as 'basic neon 2100'. You can change the operating frequency of the HV output by varying the C5 capacitor across the primary coil (at the transistor collector, the present 1uF gives 36.5 kHz output frequency because the primary coil has 19 uH inductance as per the PDF file, page 9).

-- an even cheaper solution would be to buy a similar HV transformer at ebay and use it in the same schematic of the PDF file. See this offer here: http://www.ebay.com/itm/282215688077 I think you recognize the HV output is the two lightblue PVC isolated wire pair, the thick enameled wire pair is the primary coil for the collector of the transistor and the thin wire pair is the feedback coil to the base circuit of the transistor as shown in the previous schematic.
You need to experiment with the C5 capacitor to change the frequency, of course, and for this you would need a frequency measuring means like a handheld DMM that includes such feature or a scope, etc. If you have an L meter, that would be great to measure this transformer coils and then you could compare them to the inductance values given in the above PDF file in Page 9. But this latter is not so important, the point is you can tune the frequency by changing C5.
Notice capacitor C8 at the HV output of the schematic, it tunes the HV output coil to approximately to the same frequency the primary coil is tuned by C5 and you need at least a 10kV rated capacitor there. The exact pF value for C8 is not so critical (but the HV rating is important) because the HV output coil has high DC resistance with respect to the primary coil for both the ebay and the amazing1.com transformers, so the resonance curve for the output coil is flat. 
It would be wise to run the assembled circuit first from a lower than the 9V indicated in the PDF schematic (say from 5-6V only). If the circuit does not oscillate (i.e. no HV output) then flip the feedback coil ends to get oscillation (oscillation may also depend on the supply voltage, but do not go way higher than 9 - 10V to avoid burning the transformer unless you can measure HV amplitude somehow which is not easy at some ten kHz range).

Hopefully, Mr Dezeinstein returns to this topic and shares his words of wisdom on your other questions.

Gyula

[Tesluh]

Thank you Gyula, you beat me to it:)  I just got this in the mail today! 

http://www.ebay.com/itm/182346077493

12v dc driver and a tv flyback coil.

now the question is, I have a coil with only two output wires, some amount of hv across them (1" streamer), at some kHz frequency (unknown yet).  how do I connect those to this primary tank circuit?  Dons schematics have a two output nst with both leads having diodes on the ends which supposedly puts out hv hf pulsed dc?  The spark gap is connected to the ground terminal which I believe is connected to the center tap on his transformer.  either I take both leads and use diodes on the ends and the spark gap has to go somewhere (earth?) or I need to wind a secondary for the flyback that has a center tapped coil so the two outsides can go to diodes and the spark gap can return to center tap of secondary.  Either I need three connections from the flyback or I need a different schematic to wire the l1 coil.

this is basically a tesla coil.  how would someone drive a tesla coil using a zvs style driver and flyback transformer?  I can buy one from ebay but would have to wait weeks for it to arrive to see how they do it.

[gyulasun]

Hi Tesluh,

Well, your driver and the flyback trafo are nice but you are limited in the flexible changing of the HV output frequency versus the other oscillator circuit and the HV trafo shown in the PDF file I referred to yesterday.

I do not know the inside circuit of the NST what Don Smith used, my best speculation is his NST had a center tapped secondary coil output i.e three output wires and the two diodes worked exactly like a center tapped full wave rectifier circuit works, you are surely familiar with, see this http://www.electronics-tutorials.ws/diode/diode18.gif  and the three wires are A, B and C, C center tap is the common negative (ground symbol). The schema is from this site: http://www.electronics-tutorials.ws/diode/diode_6.html

Now, in your flyback trafo you have only 2 output 'wires', see my attachment 1.

In the second attachment I encircled the positive (red) and the negative (blue) connection points in two Don Smith circuits (they are familiar, aren't they) where I think your flyback trafo outputs should be connected, with the correct polarities. You need to identify the negative HV pin of the flyback trafo, of course but I believe you already did.      

The flyback trafo gives a half wave rectified HV output while the NST output (if the center tapped assumption is correct) gives a full wave output; this means the spark gap can fire once from the flyback output during a full T period (T=1/f) and can fire twice from the original NST output during a full period. This means you need to tune L1 (the primary coil) to the spark gap frequency accordingly.

Gyula

[Tesluh]

Got it! thank you for your help!  Yes, having adjustable frequency would be nice.  I have a source for an adjustable hv transformer with outputs similar to this (like the schematic you posted )that would make tuning easier.  I'll see if I can find the frequency and tune the l1 and l2 to match.