Re: Modeling the transformer

[Terry Fritz <twftesla-at-xxxxxxxxxx>]

Hi Herbert,

At 08:21 PM 11/2/2000 +0100, you wrote:
>Hi,
>does anyone have some explanation, how to determine the values
>needed for Microsim to model the transformer (neon, polepig etc.)
>Microsim version 8 offers 2 types of parts, the XFRM_LINEAR and
>the XFRM_NONLINEAR.
>The models I found for tesla coils, use the linear part.
>This offers the parameters COUPLING,  L1_VALUE and L2_VALUE as the
>only parameters to be chosen, no number of turn ratios etc...
>How do you determine the values you need for your own specific
>transformer?
>I tried some measurements with my transformer using simple RL low pass
>filter
>setup with a signal generator and a scope, but without success. Using
>the inductivity values
>I calculated (0,22H for the primary and 156,24H) for the secondary in a
>very simple
>microsim model (just the transformer connected to the source, loaded by
>a 100k resistor)
>did not result in the correct secondary voltage for this transformer
>displayed by microsim.
>This seems to prove, that my method of measurement is not suitable for
>determination
>of the L1 and L2 values. How else could it be done?
>
>Thanks very much for some advice,
>Herbert
>

Here's how I do it.

Measure the open load current of the transformer.  This will tell you the
primary impedance.  Then you can calculate the inductance needed to get
that impedance with:

Z = 2 x pi x F x L 

So if you get .25 amps at 120VAC (60hHz),

0.25 = 120 / Z    So Z = 480 ohms. 

480 = 2 x pi x 60 x L   So L = 1.273H  Which is the primary inductance.

The secondary inductance is the square of the voltage stepup multiplied by
this inductance:

Ls = (Vo/Vi)^2 x Lp

If we have a 120 to 15000 volt tranny that is:

(15000/120)^2 x 1.273 = 19894H

You then run a simple model with the output shorted and adjust the coupling
to get the correct short circuit current.  Then check the open load voltage
and adjust the secondary inductance up a bit if need.  This is sort of
trial and error but it works very well.

The input and output resistances can be measured as usual.  You may want to
check my two models (center taped) if you have not already and the paper...

http://users.better-dot-org/tfritz/site/MicroSim/MicroSim.html
http://users.better-dot-org/tfritz/site/papers/modact/modact.html

Let me know if you have questions.  I have this data already of a number of
common NSTs.  Hopefully, my new site will be up real soon to get this on
the WEB...

Cheers,

	Terry