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

Re: Tesla simulations vs. scope traces (fwd)




From: 	Scott Stephens[SMTP:stephens-at-enteract-dot-com]
Sent: 	Tuesday, December 23, 1997 1:23 PM
To: 	Tesla List
Subject: 	Re: Tesla simulations vs. scope traces (fwd)

At 10:54 PM 12/20/97 -0700, you wrote:

>I've been working with some PSPICE simulations again, and a few interesting
>things have surfaced.

>1) With a digital scope (500mhz) I captured some traces of my 8X6 coil
>operating and ran the output through a FFT program at work.  The scope was
>triggering on a 4MHz ring-up , followed by a 100KHz sine wave.  

For that test, I would rather use a sweep with a spec-an through the
spectrum in question. FFT's lie, according to the windows function used,
don't they? And a 4MHz ring-up? Isn't the period of the ring-up typicaly >
than Fr for under-damped networks?

>2) WinTesla calculations predicted 91.1KHz secondary frequency.

>3) A simple RLC PSPICE simulation AC analysis predicted 91.458KHz.

AC analysis is the classic 1/2 PI (LC)^.5, right?
I got an accurate PSPICE analysis doing a transient analysis, which includes
a term -(L/2R)^2. 

>4) A distributed ladder network of series inductors simulating the bottom of
>the coil equal to the total secondary inductance/2 then /4 ,/8, etc..with
>capacitances to ground starting at the top equal to total secondary self-
>C/2+Toroid-C then self-C/4,/8, etc...showed a 91.58KHz component for the
>Inductor L7 at the top of the coil, and also 1,2,3,and 4MHz spikes for the
>Inductor L1 at the bottom of the coil...simillar to the scope traces from the
>real coil.

This sounds like your modeling the coil as a transmission line rather than
lumped-elements, which makes more sense considering the percentage of
distributed to total capacitance.

>6)  Both the scope traces and the PSPICE simulations showed the high frequency
>components are of at least 3 to 1 greater magnitude than the approx 100KHz
>component.  I think this may be the secret of the magnifier.  If you tuned the
>extra coil such that it's res freq is equal to the L1 ring-up, you could get 3
>times the output?????
>
>I am building an extra coil for res at 4MHz to test this (without any coating
>so I can add or remove windings for tuning.)
>