[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [TCML] Is my understanding correct?
Hi Philip,
P Tuck wrote:
JavaTc is a great tool, but the trouble is you can play around just a little
bit too much!
Yes, that is true. There are no limitations and there is a myriad of
coil parameters possible. Even these days, we still cannot identify
"best" coil design for a set of parameters. We can certainly generalize
known/experienced good ond bad parameters, but so much is subjective to
other components in the system as well as personal experience. We are
slowly progressing toward that goal.
<snip>
So finally my Questions are:-
1) Desirable Freq, (high or Low) to give the caps an easier time
Generally, a lower frequency is best. It helps not only with the caps,
but also with the coil itself keeping RF losses as low as is possible in
a given design.
2) Prim Inductance, High or Low to get down gap losses, come to that, what
is High or low on a coil this size ?
There is of course "too high" and "too low". Too high will encounter
unnecessary losses and reduce the emf needed from the primary. Too low
is difficult to predict. Gap losses tend to reduce with some impedance
in the primary, but it's hard to quantify when too low of an impedance
begins affecting the gap. There are plenty of coils running low turn
coils. It depends on the type of gap, energy across the gap, and break
rate. For most medium sized coils, low turns is not always good. Many
have reported better performance with 10 to 20 turns. For high powered
coils, the primary tends to lower turns (high power allows for large cap
sizes which causes primary inductances to end up low). Nature of the
beast thing. Even the high powered coils would likely benefit from a
little added primary inductance, but as long as their gap is quenching
as intended, then there's really no need.
3) Aspect ratio: (4 to 5:1 I know that) BUT one design (the lowest at 181)
uses a 5.5:1 - is this ratio pushing it a bit?
There is plenty of room here. The h/d between 4 and 5 is the common
recommendation. But, you can certainly go above and below this to some
degree and still have a good performing coil. Much of this is turns and
wire size. If you keep say 1000 turns and grow the wire size (versus the
turns), then you'll be better off. Problems occur when the turns are the
"adjustable" parameter. That's when Rac and Rdc losses begin to be seen
due to increased turns needed to achieve the h/d.
I built a high turn coil at 1800 turns (8.5" x 39.75", 24 awg) and I
wasn't impressed (even with a decent h/d of 4.7). I've since reduced
turns and it's performed better (although I still have a ways to go to
optimize this particular coil). My point with this is to keep the turns
constant, and let the wire size grow to accommodate (but also, keep an
eye open on the primary turns). If they go below say 8 turns for
resonance, then you may be better off going up a notch on the wire size
(countering the worst case losses in a given situation).
4) Can a 10K/48ma NST power a 4.5 inch coil (as opposed to the norm of a 4
inch for this NST)
Absolutely! The 48mA is a decent transformer. As Gary mentioned, the VA
is what your looking at for spark length. The 10,000 volt rms number is
important for gap spacing, but the VA is what makes short or long
sparks. Going from a 4" to a 4.5" is no big deal. Your better off with a
wider secondary in most cases (regardless of wire size) as long as you
don't go too wide (which I would say is about 3:1, where sparks fly from
toroid to primary simply due to proximity). There is a mechanical
property to sparks, and it does take a little run time to see what's
going on and what can happen when the top load is too close to the
primary. In other words, wide is good, but there is a thing as too wide.
It's a balancing act. The top load geometry affects this situation.
There are a lot of possibilities out there that could work well with low
h/d coils.
Take care,
Bart
Thanks hopefully in advance
Regards
Philip
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla