Re: Secondary Topload ( was New formula forsecondaryresonant frequency)

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

Original poster: "Kurt Schraner by way of Terry Fritz <twftesla-at-uswest-dot-net>" <k.schraner-at-datacomm.ch>

Hi Malcolm, John, all,

uncompromising examples is someting I like - thanks for your one,
Malcolm: the physical interaction seems obvious! - John, i don't
have at hand the two equations by F. Lowenstein shown in the TCBA
Newsletter for April, 1987, because I don't own the TCBA
Newsletters. But I "think" having a graph or a table, you can
always convert it to a set of equations, with reasonable
precision, by means of modern, statistical curve-fitting tools.
My point was having a DESIGN-tool for planning new TC's, which is
more appropriate than current practice (DESIGN as opposed to
SIMULATION, of something already specified). But, what I'm just
doing now, in order to design a new Tesla-coil, of REASONABLE
CONVENTIONAL GEOMETRY, meaning a toroid something like 0.4 <
CoilLength < 1 and/or 2 < Coil Diameter < 4, is like:

1.) Make a first guess, using Wintesla, Ed Sonderman's, or my own
procedure, which presume the additive behavior of Cmedhurst and
Ctop (by Bert's formula or something else).

2.) Take your design to Terry's great E-TESLA 6 and estimate, how
the potential and E-field stress distribution around your
coil-design behave. Get the more realistic C's and Fres's. Try
ways of reducing E-field stress by adding corona toroids, - or
remove them, if not effective (...done, i.e. with B&W).

3.) Add the (more problematic!) power supply/ballasting/sparkgap
system, with appropriate (measured) parameters to the design, and
move to simulation by Pspice (Microsim).

4.) Iterate the procedure, till within the restrictions of your
available parts, and design goals.

5.) Physically build the TC, after design of construction: points
6.) ... X.), test it points X.) ...Y.), and compare what you got,
to what was planned. Record the experience.

The most important part, in this context, seems point 2.): the
transfer from design to simulation. If the estimation of a
combined C for the secondary + topload could be supported by a
more suitable "mixing rule" (of course in form of a set of
equations) of i.e. Cmedhurst and Ctop, the DESIGN-procedure might
gain in quality and precision. If my statement is showing only
ignorance of what is already present practice: please teach me
the better way!

Cheers,
kurt Schraner



Tesla list wrote:
> 
> Original poster: "Malcolm Watts by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <m.j.watts-at-massey.ac.nz>
> 
> 
> There is a problem with that approach that's been known about for a
> long time. It is best illustrated if you take, say, a 6" major
> diameter torus, plonk it on top of, say a 2" secondary, deduce its
> capacitance by one means or another, THEN put it on top of a 12"
> secondary. You will end up with a large disparity in the capacitance
> value of the terminal between the two situations. In fact, the 12"
> coil will hardly notice it's there. The terminal is effectively
> shielded by the windings. For a long time now I've regarded the
> system as a whole rather than trying to make the bits fit.
> 
> Regards,
> malcolm
> 
Tesla list wrote:
> 
> Original poster: "John H. Couture by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <couturejh-at-worldnet.att-dot-net>
> 
> Kurt -
> 
> What I was trying to point out was that your measured 69 Khz was very close
> to the JHCTES output of 69.59 Khz. In the past some coilers said that their
> measured frequency was much different than the JHCTES and, therefore, the
> program was incorrect. It now appears that the coilers inputs or
> measurements where incorrect.
> 
> The JHCTES program does not use Bert Pool's equation. I found this equation
> did not give accurate enough toroid capacitances compared to measures
> results. The JHCTES program uses instead the two equations by F. Lowenstein
> shown in the TCBA Newsletter for April, 1987. In one of my books I show a
> capacitance graph based on these equations for different size toroids. These
> graph capacitances can be used in the JHCTES program. The graph has the
> advantage that you do not need to do any calculations to find the toroid
> sizes when you know the capcitance values.
> 
> It should be noted that any computer TC program will only be approximate
> because the outputs are dependent on the accuracy of the inputs. At present
> it appears that the ion cloud and other effects would not change the outputs
> by more than +/- 5 percent.
> 
> John Couture
<snip>