[TCML] bipolar coil
bartb at classictesla.com
Sat May 3 17:54:54 MDT 2008
The recommendation of the srsg ltr size (usually 2.5 x Cres) is based on
maximizing the power through the gap. Going ltr is to prevent resonant
rise voltage from going too high and damaging the cap or nst. These
values are based on internally shunted transformers specifically. It
really helps to run simulations on resonant transformer conditions for
visualization purposes. Once a cap is increased and outside resonant
rise, it's then desirable to maximize power within the nst's capability.
Terry Fritz and others did a lot of testing of nst's and developed ther
Both bps and cap size are the two main factors. It's desirable to keep
bps around twice the mains frequency (120 bps) for maximum efficiency.
It's then desirable to increase the bang size (power at the gap at the
time of breakdown) as high as possible for maximum performance. But, you
can only go so large before the cap voltage begins to get very low. In
the end, the cap size tends to be about 2.5 x Cres for srsg operation.
Static gaps force the ltr at about 1.6 x Cres to keep the 120 bps
operation. If you were to try 2.5 x Cres with a static gap, the bps
would theoretically be very low.
Consider the SRSG LTR size as the size for maximum performance. Yours is
one that will not allow this large of a cap. So, you have to make a
trade off and run a smaller cap size. It's not a big deal, but it is
important to stay outside of resonant rise which is why I recommended
something about 1.6 x Cres which will work.
The problem is using only 2 turns per coil. Makes for a sensitive
adjustment, and I stand by that statement. If you wanted to make that
adjustment less sensitive, then you would need to run smaller than
resonance (a really small cap). However, not with an srsg. The cap would
accumulate a voltage much higher than the nst could handle (and you'll
over-volt the hv windings) due to the fixed time between electrode
alignment. If you end up trying a smaller than resonant cap at some
point, use a static gap which can clamp the voltage as allowed by the
electrode size and spacing. Always use a safety gap with any rotary gap
Kris Grillo wrote:
> Yes, this is a twin setup with two separate identical secondaries, each with its own topload and ground, positioned vertically a distance apart from each other. This is not a traditional bipolar design with one long secondary positioned horizontally with a primary positioned around its center.
> The lead inductance is a killer, but with the bases of the secondaries mounted 36" on center apart, there is only so much I can do to shorten the lead length. My 48" estimate might be a little on the optimistic side.
> I think the main problem here is that I don't quite understand why the recommended LTR cap for the srg is 2.6x resonance. I understand why I don't want to be too close to resonance but I don't quite get why a synchronous rotary gap causes the need for it to be so much larger than with a static gap, or how performance would be effected by decreasing the cap size, so I just stuck close to the recommended figure.
> If I can just use a smaller cap, then the primary turns problem wouldn't be such an issue, but I would like to understand this better before I start tweaking this figure.
> bartb <bartb at classictesla.com> wrote: Hi Kris,
> Using the 9/60 NST, SRSG LTR is .046uF. Doubling this for the bipolar
> situation is just too much cap (resulting in less than 1 turn).
> Remember, you don't have to go this big with the cap. Say you use the
> 1.6 x Cres value which is .0265uF. Thus doubling this you would input
> .053uF for the cap size. When you tune, you'll end with 2.2 turns per
> coil. Thus each coil will have 1.245uH for a total of 2.49uH with the
> .0265uF cap. Each coil will be resonant at 410kHz (assuming each twin is
> 2.375" x 11.5", 975 turns, 30 awg, with 3"x12" toroid).
> I'm not trying to suggest that 0.0265uF is the value to use, but simply
> to show that a smaller cap size is needed so that you can transfer more
> turns to each primary coil. The above assumes 1.6uH of lead inductance
> (I used the full 48" lead inductance considering the model needs to see
> all of it). If you can reduce the wiring to bare minimum, you can then
> transfer that inductance into the primary turns as well. Right now, it
> is being shared with the primary and it is a significant at 36% of the
> total inductance.
> For clarification: Your original post identifies twin coils, not a
> bipolar coil. Is this correct? You were using the program in a bipolar
> configuration simply to try to model the situation? That's how I
> The reason I ask, is if this were actually a bipolar coil, I would model
> it very differently.
> Take care,
> Kris Grillo wrote:
>> I have wound the coils and put about 25 coats of poly on them, and made the toploads. In fact, everything is done except for the primaries.
>> I do realize that 1350VA is a huge amount of power for such a small coil. I think I got a little carried away. Originally I was going to use just one of the 60mA transformers, but then I found another one real cheap, so I bought it. I had another 30mA sitting on the shelf and figured what the hell, lets make a monster of a small coil. I think I'll just stick with the original plan of 60mA.
>> Bart, I tried inputing just one of the twins and doubling the cap size. I get a geometry error, less than one turn primary. I also tried it with 120mA, 90mA, 60mA and 30mA, all with the same results. This was with 2x the recommended LTR cap size for a synchronous rotary gap for each of the currents. I tried it with double 2x resonant also, with the same results. I haven't tried tweaking the primary dimensions, except for spacing the coupling. Any suggestions?
> Tesla mailing list
> Tesla at www.pupman.com
> Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
> Tesla mailing list
> Tesla at www.pupman.com
More information about the Tesla