Re: Magnifier conversion

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

Hi Steve,

At 10:31 AM 4/15/2003 +0100, Steve wrote:
>I wouldn't mind spending lots of money on IGBT bricks, and lots of time 
>designing a flyback converter-based charging circuit, if I thought the 
>OLTC magnifier was going to work well. But at the moment I have no reason 
>to believe that it'll be any better than a 2-coil system. It would be more 
>like "you pay, you don't play." Also, it will be much more complex, 
>because of the new charging circuit, and the main reason I liked the OLTC 
>was simplicity.
>
>I'll see how Terry Fritz's magnifier conversion goes. If it gives longer 
>sparks than the 2-coil system, there may be a case for going ahead.

A good plan.  Although there has been much discussions about 3 coils 
systems, there is still the need for a little more R&D work and modeling 
verification.  OLTCs are not forgiving of trial and error design, so having 
the details and all worked out first is best.  This is why I also choose to 
work on a conventional system first.


>>The peak voltage was 323kV (ideal case).  It was easy to modify my 
>>present MicroSim model for my small coil to add these values and some 
>>real good guesses as to what the actual losses will be:
>>
>>http://hot-streamer-dot-com/temp/NewMag/0413-02.gif
>>
>>Happily, even with the losses (especially the streamer loading), the 
>>values still give very good if not almost perfect C1 to C3 voltage conversion.
>>
>>http://hot-streamer-dot-com/temp/NewMag/0413-03.gif
>>
>>The peak voltage drops to 226 kV with loading.  However, the old model 
>>only got to 129kV!!
>>
>>http://hot-streamer-dot-com/temp/NewMag/0413-04.gif
>>
>>That should give 3X the power to a streamer for a brief instant!!  While 
>>the two coil system gives a lower voltage over many cycles, the magnifier 
>>gives a much higher voltage over fewer cycles.
>
>
>I don't understand what's going on here, surely conservation of energy 
>means that 1/2*C1*V1^2 = 1/2*C3*V3^2 + losses, whatever the configuration 
>of coils. So did that extra 100kV come from a reduction in losses through 
>faster transfer? When I modelled the 2-coil and 3-coil OLTCs, the peak 
>output voltages were almost identical. The 3-coil case was about 5% higher.

A Tesla coil may easily burn up 50+% of it's energy in losses.  Ideally, I 
would get 227kV on the toroid.  However, the gap and other losses (the 
streamer is also considered as a loss here) reduce that ideal top voltage 
to only 114kV.  However, the very fast and lower loss transfer of the 
magnifier seem to reduce that loss drastically giving a toroid voltage of 
188kV!  Here are side by side voltage plots for a conventional and 
magnifier system (the numbers are changed some since I am now modeling with 
the new larger toroid):

http://hot-streamer-dot-com/temp/NewMag/2C-3C.gif

So if the peak voltage really does increase 65% the streamer length 
"should" too.  This coil performance was right at the spark lengths 
predicted by John's formula at 28 inches before.  So maybe now it will do 
46 inch arcs from a 9kV/30mA NST!  But have to actually see that to believe 
it ;-))

In the case of an OLTC with the gap losses already very low, the advantage 
of a magnifier system my not be so nearly as great.  However, primary and 
other losses are still very high, so maybe the magnifier methods could 
still help the OLTC a lot.  Have to study this more in light of the results 
with the conventional magnifier.


Gary Lau wrote:

>While stranded wire will certainly work, depending upon the operating 
>frequency, AC resistance losses will be considerably higher than with the 
>same gauge in a solid conductor.

I now model the secondary wire at 20 ohms and get a peak voltage of 
187.791kV.  I raised the resistance to 80 ohm and the output voltage was 
185.282kV (models are great for these questions ;-)).  So the resistance is 
not a big deal (a good thing since I already ordered the stranded wire 
;-))  Stranded wire is easier to "play with" in this case and I already 
over sized the wire anyway.

I think the key to the low loss is that the peak current in the secondary 
is only 12 amps as opposed to 215 amps in the primary.  A 1.5 ohm spark gap 
resistance in the primary burns a peak of 70kW while the peak power loss in 
the secondary is only  2.9kW!  That is the key here.  By transferring power 
to the higher Z circuits as fast as possible, we can reduce the power loss 
24X !!


John Wrote:

>Building a magnifier is not very critical.  Magnifiers as they are
>commonly built, don't offer any real efficiency benefit compared
>to normal 2-coil systems.
>
>Antonio has done some theoretical work which suggests that
>a special magnifier design may outperform a 2- coil system,
>but this is difficult to achieve, and has never been demonstrated
>in a spark-producing coil.

It is interesting that the large C2 needed for Antonio's design would not 
be available in a small coil, but may be naturally available in a large 
coil.  This might explain why "small" magnifiers just seem like 
conventional coils, while larger coils with considerably more natural C2 
capacitance work so much better.  For a magnifier to work other than a 
conventional coil, C2 has to be a very large capacitance.  Otherwise, there 
is no difference.  In a small system (smaller than Wysocks ;-)), we need to 
add a special C2 capacitor.  In my system, C2 has to be 164pF.  But the 
Cself of the secondary coil and the wire from the secondary to the extra 
coil are only 29.8pF.  That is probably not enough to give any great 
advantage.  Thus I am adding a 134pF 120kV MMC style cap to C2.  If all 
goes well, this should give much higher output power as mentioned above.

The C2 MMC (High Voltage => (HV-MMC)) is easy to make.  It is far easier 
than I thought it would be, so I don't think these "small" magnifiers 
mentioned by Antonio are hard to make at all.  With a large toriod, C2's 
value is pretty high which lowers the secondary voltage (<80kV peak) 
helping to prevent primary to secondary breakdown and putting the voltage 
easily into the HV-MMC range.  At first, I forgot about the other stray 
capacitances so my HV-MMC probably should have had smaller value caps.  But 
I can go down to 100pF right now which should still be fine (it can be 
tapped at 99 points ;-)).  The HV-MMC seems like it will be perfect and it 
can even self heal if it takes streamer hits ;-))  I am also putting in the 
bleeder resistors since they will work just as well in this case.  A 150pF 
cap at 150kV could pack a pretty big punch especially considering it could 
arc 3 inches!!

We'll just have to see...  I guess I am surprised Antonio's system has not 
really been tried before!?  Modeling is certainly suggesting that the 
performance will be far greater!!  Also, the magnifier may be less affected 
by the long streamers de-tuning the system reducing spark length.  The 
magnifier may be able to "drive" the streamers far easier than a more 
tuning sensitive two coil system.  One problem may be getting the gap to 
quench on the first notch.  That usually adds resistance.  But not a big deal.

Cheers,

         Terry