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

Re: distributed capacitance measurement? Isotropic?




From: 	Thomas McGahee[SMTP:tom_mcgahee-at-sigmais-dot-com]
Sent: 	Saturday, September 06, 1997 10:50 AM
To: 	Tesla List
Subject: 	Re: distributed capacitance measurement? Isotropic?


> > 
> > 
> > From: 	Malcolm Watts[SMTP:MALCOLM-at-directorate.wnp.ac.nz]
> > Sent: 	Thursday, September 04, 1997 2:48 PM
> > To: 	tesla-at-pupman-dot-com
> > Subject: 	Re: distributed capacitance measurement?
> > 
> > Hi Jim,
> >  
> > > From:   bmack[SMTP:bmack-at-frontiernet-dot-net]
> > > Sent:   Wednesday, September 03, 1997 10:09 PM
> > > To:     tesla list
> > > Subject:    distributed capacitance measurement?
> > > 
> > > To All,
> > > 
> > > How the heck do you MEASURE the distributed capacitance
> > > of a secondary?
> > 
> > Measure its inductance with an L meter, measure f then calculate
Cdist
> > from that. I don't know of any method of measuring it directly.
> >  
> > > I have a several methods to determine f-res and Q, but don't
> > > have a clue how you can separate the system into L and C for
> > > true empirical measurement.
> > > Unless you are calculating it from the approximate known value
> > > of L when resonance is determined? Thats what I use,-but that's
> > > NOT a measurement. 
> > 
> > Agreed, but it has to be correct. There is at least one
distributed 
> > model around now that agrees with the lumped parameters :)
> > You could try a capacitance meter hooked between the base and
ground.
> > You would have to compensate for extra lead capacitance. Guess I 
> > should try this myself.
> > 

You would be measuring something more akin to the isotropic
capacitance of the coil surface (much as in measuring the isotropic
capacitance of a topload). 

The distributed capacitance can only be determined indirectly using
resonant frequency and L.

Also, the frequency at which the capacitance meter operates may skew
the results. I would imagine that if the poor capacitance meter was
using a frequency equal to the self-resonant frequency of the coil
that it would give a large error in measured capacitance. There are
several different methods used by capacitance meters to measure
capacitance. Large values of C can be measured using a fixed current
and measuring the resultant ramp time for the voltage to reach a
given value. Smaller capacitance values are generally measured by
passing a fixed frequency signal of known amplitude through the
capacitor under test and measuring the resultant AC current. When
measuring pure capacitance that works just fine. Unfortunately, it
fails miserably if there are other factors involved, such as
inductance or excessive leakage. At resonance the AC current flow
will be QUITE different than that at any other frequency!

An interesting side issue here is what effect the isotropic
capacitance of the coil has on the total distributed capacitance.
Isotropic measurement would not measure inter-wire capacitance, which
definitely has an effect on the total distributed capacitance. The
question is how the isotropic capacitance of the coil relates to the
total distributed capacitance. Is it small compared to the total
distributed capacitance, or does it predominate?

Malcolm, I know you have done a fair amount of work with tight-wound
and also with space-wound coils. Have you noticed anything odd about
the total distributed capacitance of space-wound versus tight-wound
Tesla secondaries? Is there any correlation between distributed
capacitance and exposed wire surface? Or is it primarily an
inter-turn phenomenon? What does that have to say about the relative
merits of tight versus space windings? My own work has led me to
believe that the exposed surface area of a coil is very important,
especially in regard to the electrostatic phenomena that accompany
Tesla coil operation. Probably much greater in its overall EFFECTS
than the inter-turn capacitance. By the same token, the proximity of
the inside of the coil windings and the coil FORM result in
interactions that greatly affect the Q of a coil. The losses are
probably capacitive in nature, in that energy is capacitively coupled
to the material that makes up the form.

For what it's worth...
Fr. Tom McGahee



> > > Another indirect method might be to use lisajous figures 
> > > on a scope to determine the 45 degree point in a LR circuit
(the R
> > > being a variable) at a low and high freqency ( below Fr).  the
> > > reactance difference would represent it's distributed
capacitance,
> > > at the higher freq used. For best results, get close to Fr.
> > > As you know, all electronic component values are frequency 
> > > dependent-even resistors!  The problem gets worse as one
progesses
> > > upward into hf -the uper limit for Tesla coils.
> > > 
> > > The last paragraph is pure speculation on my part. Has anyone
tried
> this?
> > 
> > No. One has to be careful about how the measurement is taken
because
> > mounting the coil upright and connecting the base to a low
impedance 
> > turns it into a quite different animal from a non-grounded coil. 
> > 
> > Malcolm
> > 
> > 
> > 
> 
>