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

Re: scan tesla questions



Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>

Hi Steve,

At 04:31 PM 7/14/2005, you wrote:
Hi again Terry,

> >Secondly (i should have been paying more attention before).  Does
> >scantesla use primary feedback to do its simulations?  Or does it use
> >some other method for selecting the drive frequency?  This could be
> >very important.  Just wondering how the program works in that respect.
>
> It uses primary feedback.  It could easily use secondary current feedback
> too but I would have to change a number in a line...  Let me know if you
> want that.
>

No, i was just making sure that the program was doing the modelling
correctly.  Jimmy thought you might have used something that just
searched for the "best" drive frequency, and that it might not be
necessarily based on primary current zero crossings.

Antonio's original program used a fixed frequency, but I just made it switch the input voltage dependant on the primary current. If I change a few variables, that could easily be secondary current. I need to add an input switch for that...



Is there a possibility to add a "switch delay time" into the program?
Im not sure if it would matter, but if it isnt something difficult to
do, it would be nice to be able to model just one more detail: driver
delay.

The program can do "anything" ;-)) I will put switching delay on the todo list...




> >Anyway,
> >im finding that scan tesla's peak current predictions for my coil are
> >not correct (it predicted 800A, i measured 1100A).  This might be due
> >to a foul Rprimary input.  This run used a K of only .13, which is a
> >big change for my usual K=.24 type stuff!  I think this K is too low,
> >though, as more simulation work shows that a K of about .17-.18 seems
> >to have the same SERT, same bang energy, but less primary current.
>
> It "should" do the currents fine.  Rprimary is pretty important.  Do you
> think the program is not calculating something right or there is something
> wrong?  If there is a problem, I can check it out.

No, scantesla seems to agree with my pspice simulations (it should!)

Yea! Let me know right away if it does not since that would mean I broke it ;-))


,
but they dont agree with my measurements >:-(.  I even tried making
Rprimary smaller (from .1ohms to .05ohms) and it didnt change
noticably.  The coil is wired up with large chunks of copper bar and
2awg battery cable... i suppose the primary resistance could indeed be
quite low!  I will do some more tests this weekend and see if i can
figure out the discrepency here.

>
> The currents seem to work out very closely to the real current for me.

Yeah, it seemed to predict nicely on my medium coil too.  Another
question, did you ever check your RF envelope (primary current) on the
real coil vs the model?  Curious to see if that is working out
correctly as well.  The RF envelope seems to offer some clues as to
how the streamer is loading the coil down...

It seemed to work perfectly for me once I adjusted Rprimary a bit. I was just using "CheapScope" so I did not check it rigorously. I should have gotten like 83 amps peak and that is just what CheapScope said...



Just thought of another question (not really related to the program).
Does your book about sparks mention the optimal rise time for an 11
foot streamer (or leader or whatever the heck we are supposed to call
it)?  I would guess about 175uS, but that is afterall... a guess.

Yes it does, I mentioned the equation a few weeks ago. I forgot it right now 0:o)


I'll try to refreash the ram...


>
> Is the current transformer for the primary current rated up to 1100A ?

I dont see why it couldnt handle that.  Its home made on a core that
should handle about 2V/turn easily at this frequency.  Its got 200
turns on it and its putting out about a 10V signal, so there should be
no saturation at least.  Im not sure what else is involved with CT
"ratings" other than the V*s product.  Current still appears
sinusoidal... in fact, when checked against a pearson at lower power,
my CT has less phase error (mine leads in phase just slightly) and
also gives a nice clean output.

Ok, Here is the "whole" story about that:

http://www.pearsonelectronics.com/current-monitor-products/application-notes.htm

http://www.pearsonelectronics.com/datasheets/technical-literature/Application%20Notes.pdf

The waveforms would probably look distorted if it was saturating though...

Cheers,

        Terry



Steve