[Prev][Next][Index][Thread]

Re: Sizing the toroid to the coil, was Effects of the size oftoroids

> Original Poster: FutureT-at-aol-dot-com

Hi John,

I agree that the bang size is essential for toroid sizing. But bang size also
relates a coils capability to process the bang and is directly related to the
coils size and dimensions as you know. I calculate spark length based on the
coils dimensions, which equate to it's power processing capabilities. CSelf is
taken into account because it tells the tell of the coils size. My CSelf factor
is also based on 75 coils I grabbed info on.

2.8 x CSelf was the result of the best performing coils (performance is
power in
vs. spark length - as we both agree on) and the relationship of Ctop to Cself.
Yes, this was an average, but more importantly was how little the relationship
varied on these best performing coils. One can also see the relationship in
inductance if we go the other route. The CSelf value is used in my spark length
equation as well as bang size and more importantly, the effective EPS
("effective" energy per second - which is based on the energy available at the
bps and cap size in relation to the input power). This equation takes into
consideration the total capacitance and top capacitance as 2 seperate entities.
The top capacitance is a required value for spark length. I like it because
at a
glance, coilers with under-developed Ctop always post small spark lengths
compared to what input power says they "should" acheive and my equation says
"yes, it should only be this small".

However, this is by no means my final for this equation and I am hindered
to send
it out there since I've made quite a few changes to it since those 97 coiler
stats and I'm still not sure if it is relatively accurate. There are coils
where
my equation just blows and doesn't represent reality. I like your equation
"immensly" because it takes into account the "actual" input power (if we
use your
equation correctly) and converts to spark length. Your spark lengths are
near to
actual, but anyone can go overboard where it become rediculous. This can happen
with my equation as well.

I would like to take it a step further and create something that can't become
rediculous. This is harder than it looks. I guess if we had data on coils built
out of the realm of "normal size", we could get there. But your equation
and mine
are both based on real world coils and in the most part, the coils are
similar in
size. In most instances where coilers present their stats and arc lengths, our
equations are within only a couple inches of each other, but sometimes,
there are
situations where a coiler is having problems based on simple components, toroid
sizing, tuning, coupling, etc... CSelf is just one area that helps me consider
the top load portion of a coil. Of course, "experience" is always the driving
factor that says yes - this is the coilers problem or no - it's something else.

So, yes I agree about the importance of bang size and this is "most definately"
taken into account. But also, CSelf  and Ctop are necessary because the
capacitance of Ctop and the ability of the coil to charge Ctop is "all"
imporant
if we are talking spark lengths. I bascically use the total capacitance of the
entire system and then equate how Ctop "weights" the coil for spark length.
Heck,
it's almost like the variac post where there is a "sweet spot". I think
there is
a "sweet spot" for Ctop and the total capacitance. My personal sweet spot is a
2.8 factor, but this will vary from coil to coil by some degree - I just wish I
new what the variance actually is.

Bart

> I size the toroid based mostly on the TC "bang size" (joules of energy
> per gap firing).  I may then modify this to make the proportions more
> reasonable if needed.  The toroid can also sized to the spark length;
>
>     Spark length = 10X  to  12X  toroid minor diameter (thickness)
>
> For instance if the spark will be 40" to 50" long, then the toroid should
> be 4" thick.  However higher breakrates can tolerate a smaller toroid,
> because the sparks coalesce easier at higher break-rates.  Also, most
> folks use a smaller bang size at higher break-rates, so a smaller toroid
> is required unless they use a breakout point.
>
> Of couse in general, a smaller toroid will tend to give more individual
> streamers for a given coil.
>
> I don't think the toroid size should relate to Cself, but rather more to
> the power input or to the bang size especially.  This way coils that
> are abnormally small or large for their power inputs will still work well.
>
> In many cases you can get away with using a Cself relationship
> however.
>
> Cheers,
> John Freau