Re: Racing Spark Prediction

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: dest <dest@xxxxxxxxxxxxx>

Hallo John.

> Original poster: FutureT@xxxxxxx

> Seems that when using the aiding and bucking inductances method
> for measuring k, along with the typical formula that's generally used
> (I can't remember it offhand), the higher couplings give wrong results.
> It makes the k seem tighter than it really is.

something like "measure the series inductance twice, once with one
coil reversed (a1 and a2), subtract one result from the other, and
solve the resulting expression for M.", then

M=|a1-a2|/4

?
yeah, i`ve read about this gotcha in archives, but i think this method
prone to errors with any value of k - coz Lpri is about 3 order of
magnitude smaller than Lsec.
Marco Denicolai had used this method while playing with his Thor, he
make it even simplier - used only one msmt instead of 2 - measured
only a1, then

M=|a1-(L1+L2)|/2

: D

> Using the 60 Hz AC current method you mentioned, then all is
> accurate for k measurements.

so no problemo here : )
i prefer measuring instead of calculating k not because i don`t trust
JAVATC or INCA for example - no, just coz i know for sure, that it would
be impossible for me to wind a primary exactly flat, or exactly at 30
deg, etc, etc, so at the end i would not have rights to compare my real
transformer with simulated one, that`s all % )