[TCML] bipolar coil

[bartb]

Hi Kris,

I can't give much help because I've never built a twin coil. But, I see 
your trying to use Javatc for a twin application. I'm not sure that's a 
good idea. It's not designed for twin systems and your trying to double 
up the secondaries? I think the resonant frequency is going to be "way" 
off. You will be better off (I think) modeling a single coil (one of the 
twins) in Javatc for more accurate numbers.

I assume both twins will be identical and their bases grounded. If 
that's the case, a single coils model will show you the resonant 
frequency and info. Depending on their proximity, they can add a little 
external capacitance, but still close to a single coil. The primary 
would need to be modeled differently than normal.

I've done a lot of tricks with Javatc with some odd coils, but never a 
twin. Although the single secondary frequency the resonant frequency of 
both coils, the primary inductance would need to be 1/2 of what it would 
normally be for a single coil. I suspect you could double the cap size 
for modeling purposes to find the primary turns needed for a single coil 
(to represent a twin coil inductance). Just a thought, not too sure, but 
I think this would be a better representation if using Javatc for this 
application.

And regarding the primary inductance being 1/2 per coil, this will cause 
the primary tuning to be tight, meaning the tap needs to be spot on. You 
will end up with the primary tuning "forgiveness" half of what it would 
normally be. The primary LC ratio and the fact that each coil will only 
see 1/2 of the inductance is why.

As far as the transformer, the LTR size is based on the total 
transformer impedance (all the NST's paralleled). It does get big. SRSG 
is about 2.6 x Cres and Static Gaps are about 1.6 x Cres as standard. 
But, you don't have to stick to these values as though they are 
critical. I run near 1.4 x Cres with my static gap to take advantage of 
a little resonant charging. The main thing is to stay away from Cres by 
at least a 1.2 factor. However, 1.2 is riding the fence. Better to go a 
little larger on the cap. The very bottom line is the closer you get to 
Cres, the greater the probability of damaging the NST, and gap breakdown 
stability directly affects this (what you can and cannot get away with).

Take care,
Bart

Kris Grillo wrote:
> I'm putting together a small twin coil setup and I'm wondering if anyone could look over these numbers for me. The secondaries are 2" pvc (2.375" od) with 975 turns of 30 gauge and a 3 x 12 toroid each. The power supply is 2x 9KV 60mA and a 9KV 30 mA NSTs in parallel for 9KV 150mA. The gap is a 120 BPS synchronous rotary gap. The primaries are 6 turns of 1/8" OD copper tubing with .25" spacing and inside radius of 2.4375".
>
> Here's where I'm stuck. Javatc is saying that the SRG LTR cap is .1153 uf. Again, this figure seems relatively huge. Why does the cap size go up so dramatically with the sync gap? The primary is going to be really short with a cap that big. I've tried using javatc but I'm not sure how to input the coil specs. I tried inputing the two coils as one long coil and putting a toroid at either end and a single primary in the middle.I estimated about 4' of lead using 1/8" tubing. This gives me a little more than 4 turns on the primary so I'm guessing less than 2 turns per primary.
>
> When it comes to the wiring, I have seen schematics that show the primaries in parallel and ones that show them in series. Which is best, and why? 
>
> Does anyone have any advise on this? I'm itching to use this sync gap I made. In fact, I started this project when I decided to not use the gap on my 6" coil. I've attached the Javatc file for the above described input labeled twin specs. I have also included the Javatc file with just one of the two sets of sec, pri and topload, for measurement. Those are labeled .1166uf.
>
> Thanks,
>
> ~Kris
>
>
>
>
>  
>        
> ---------------------------------
> Be a better friend, newshound, and know-it-all with Yahoo! Mobile.  Try it now.
> ------------------------------------------------------------------------
>
> _______________________________________________
> Tesla mailing list
> Tesla at www.pupman.com
> http://www.pupman.com/mailman/listinfo/tesla