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

Re: Calculating Capacitor currents

    [The following text is in the "ISO-8859-1" character set]
    [Your display is set for the "US-ASCII" character set]
    [Some characters may be displayed incorrectly]



----------
> From: Tesla List <tesla-at-poodle.pupman-dot-com>
> To: Tesla-list-subscribers-at-poodle.pupman-dot-com
> Subject: Re: Calculating Capacitor currents
> Date: Friday, February 07, 1997 2:26 AM
> 
> Subscriber: Rodney.Davies-at-anu.edu.au Fri Feb  7 00:04:34 1997
> Date: Wed, 5 Feb 1997 23:49:09 +1100 (EST)
> From: Rodney Graham Davies <Rodney.Davies-at-anu.edu.au>
> To: tesla-at-pupman-dot-com
> Subject: Re: Calculating Capacitor currents
> 
> Hi Richard,
> 
> On Tue, 4 Feb 1997, Tesla List wrote:
> 
> > Rod, 
> > 
> > There is absolutely no way to know the  peak current in the tank,
ABSOLUTELY!
> > A close first order pass or stab at it would be to apply the Surge
impedance
> > equation.
> > 
> > Z surge = square root of L/C
> > 
> > This will not take into account any form of circuit loses at all, such
as
> > resisitve and gap losses.  The actual current will always, 100% of the
time,
> > be much less.
> > 
> > The only way to determine the actual peak current value is to measure
it
> > with a current transfomrer while the thing is throwin' arcs!
> 
> Yeh, having analysed and experimented a little further as to how the 
> current is behaving across the spark gap, you're very much correct in 
> saying it's virtually impossible to know what the peak currents are doing

> and when they're occurring...
> 
> Well, luckily enough, I know just where to get a current transformer 
> from, I'll give it a go, but I'm expecting extreme fluctuating values...
> 
> Thanks Ruchard,
> 
> Cheers,
> Rod

Rod,
Another thing to keep in mind when using a current transformer is that they
are exactly that: a transformer. As such they are usually designed for
50/60Hz operation. Which capacitor current is decidedly *not*. The peak
nature of the current being removed from the capacitor during discharge
will make it hard to get an accurate measurement unless the current
transformer is designed for operation in the 100s of KHz region. That is
because it has to respond to the *effective* frequency of the pulse, not
the repetitive frequency of the gap firing. In other words, it must
adequately respond to the narrow pulse *width*. But the heavy peak currents
can saturate the core of the current transformer, and inductive effects
caused by the current transformer itself can seriously distort the output
of the current transformer.

Even if the current transformer were connected to a recording oscilloscope
the readings would have to be *interpreted* more than anything else. Little
things like scope probe capacitance can have large effects on accuracy at
high frequencies. Measuring the voltages and currents in the primary and
secondary circuits of an operating Tesla coil is *not* and easy thing.
Which is one reason why it is hard for any of us to say with absolute
certainty that we *really know* exactly what is going on in the Tesla
circuit. This makes it hard to come up with absolute formulae. Look at all
the discussion about what is going on in the spark gap. Really good
physical data would help, but it is very hard to come by!

Don't be despondent over this state of affairs, though. I'm sure that the
use of a current transformer will give you useful comparative data that you
can use locally to tweak your own machine for optimum performance. Just
don't think that you have an absolute reading when you look at your data.
The data you get will be useful, but it is probably not going to be
accurate in the strict sense.

Fr. Tom McGahee