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

RE: Sphere/Toroid Comparison Chart



Original poster: "John H. Couture by way of Terry Fritz <twftesla-at-uswest-dot-net>" <couturejh-at-worldnet.att-dot-net>


Bart, All -

Regarding your suggestion for putting the resonant frequency in the inputs I
tried this several years ago. I show a design procedure for Tesla coils
using the resonant frequency as a start in the Tesla Coil Construction
Guide. However, I soon found that this put a burden on the designer that was
not realistic in most cases. When you make the resonant frequency an input
you have to be able to predict the resonant frequency. The E-Tesla6 program
can do this much better compared to a designer's estimate. Even a resonant
frequency program will have some inaccuracy. Only a proper test after the TC
is built will give you the actual resonant frequency.

Once you have the actual test frequency you can then adjust the secondary
terminal pf value to give this frequency in the outputs of the JHCTES or
JavaTC programs. It is obvious that with the E-Tesla6 program you will be
able to start with a more accurate TC design.

If you want to take into consideration the effects of secondary inductance
changes due to currents then you cannot use the Wheeler equation. Someone in
the future may want to tackle this problem. The fundamental equation for the
inductance of a wire is
     L = Fl/I
L  = Inductance of a wire
Fl = Flux linkages
I  = Current causing the flux
I would expect it would be a difficult road to get from this equation to the
design of a Tesla coil.


I do not believe that the resonant frequency should be in the inputs for the
JHCTES or JavaTC programs because the only way to determine this frequency
is by test after the TC is built. After you know the resonant frequency then
you can adjust the secondary terminal pf to give you the correct capacitance
so the resonant frequency shows in the outputs. From this info you can then
find how much the toroid isotropic capacitance is reduced when placed on the
secondary coil.

John Couture

--------------------------

---Original Message-----
From: Tesla list [mailto:tesla-at-pupman-dot-com]
Sent: Wednesday, May 09, 2001 7:54 PM
To: tesla-at-pupman-dot-com
Subject: Re: Sphere/Toroid Comparison Chart


Original poster: "Barton B. Anderson by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <tesla123-at-pacbell-dot-net>

Hi John,

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 -
>
> I agree the equations shown on my web site do not represent the table
shown
> below. However, for toroids above 20" large diameter there is a close
> agreement. The graph I show in my books is based on my web site equations.
> The graph covers toroids with large diameters from 10" to 80" and spheres
> 10" to 60". There are straight line curves for 3"-6"-12" and 20" small
> diameter toroids. None of the curves overlap.

I also graphed my data. The linearity is definate. This is good news.

> The equations were shown in the Tesla Coil Builders Association Newsletter
> Vol 6, #2. The graph was made several years ago using the Quattro Pro
> program. The graph shows isotropic capacitances which have to be reduced a
> certain percentage when the toroid is placed on the secondary coil. I
> estimated the percentage reduction should be about 20% but this may be too
> high.

I understand how you came to a percent based on your graph. This is good.
However, 20% may not be too high. I
think it will depend on many variables.

I propose that both JHCTES and JavaTC should change Fr and leave Ctop alone
(Fr is the affected parameter as
you know - regardless of C or L changes - Fr is what counts). Why not make
Fr the changed value. In all
probable systems, it will be less than 20%. You could easily adjust your
graphs accordingly to represent Fr (Fr
"did" change to this value regardless of C or L unknowns).

The programs can be more appropriately representative to real world coils.
Later when C and L are better
defined, the programs can more accurately represent these values and Fr
should then be even better. Just food
for thought - something to chew on. I'm sure you've been down this road
before.

> Terry's E-Tesla6 program will certainly give a more accurate
> capacitance because it takes more of the surrounding variables into
> consideration.

I totally agree!

Take care,
Bart A.