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

Gap Losses

    [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]


I changed the Title to reflect the direction this discussion is going in
----------
> From: Tesla List <tesla-at-poodle.pupman-dot-com>
> To: Tesla-list-subscribers-at-poodle.pupman-dot-com
> Subject: Re: Top Toroid
> Date: Tuesday, February 04, 1997 12:25 AM
> 
> Subscriber: unknown-at-apc-dot-net Mon Feb  3 21:52:48 1997
> Date: Sun, 02 Feb 1997 01:22:01 -0800
> From: Open Minded <unknown-at-apc-dot-net>
> To: tesla-at-pupman-dot-com
> Subject: Re: Top Toroid
> 
> Tesla List wrote:
> > 
> > > Subject: Re: Top Toroid
> > 
> > Subscriber: sgreiner-at-mail.wwnet-dot-com Fri Jan 31 23:10:25 1997
> > Date: Thu, 30 Jan 1997 14:39:43 -0800
> > From: Skip Greiner <sgreiner-at-mail.wwnet-dot-com>
> > To: tesla-at-pupman-dot-com
> > Subject: Re: Top Toroid
> > 
> > Tesla List wrote:
> > >
> > > Subscriber: MALCOLM-at-directorate.wnp.ac.nz Wed Jan 29 21:02:26 1997
> > > Date: Thu, 30 Jan 1997 11:02:08 +1200
> > > From: Malcolm Watts <MALCOLM-at-directorate.wnp.ac.nz>
> > > To: tesla-at-pupman-dot-com
> > > Subject: Re: Top Toroid
> > >
> > > Hi all,
> > >          Thought I might add a brief note to yesterday's diatribe
> > > fyi.....
> > >
> > > >      I think we are generally agreed that the output voltage is
> > > > going to be a function of the energy that gets to the secondary at
> > > > the end of the first primary ringdown, and the total secondary
> > > > capacitance including top load. From first principles, this turns
out
> > > > to be:  Vo = Vcap*SQRT(Cp/Cs).  As Bert Hickman has showed however,
> > > > this ideal will never be reached because of significant losses in
the
> > > > gap. I think in some instances, around 80% of the energy gets
> > > > transferred. Papers by others use the formula: Vo =
Vcap*SQRT(Ls/Lp).
> > > > This is true because Ls*Cs = Lp*Cp   (basic tuning requirement).
> > >
> > > <snip>
> > > Worth noting that if the prim and sec coil geometries are identical
> > > (same height and diameter), the ratio SQRT(Ls/Lp) reduces to a
> > > straight turns ratio (Ns/Np). If you want convincing, do the algebra
> > > on Wheeler's inductance formula. Note also that this does not take
> > > k into account either. It can only be true for a lossless case (no
> > > gaps or other losses in either coil), no matter what k is set at.
> > > A key point from this is that the turns ratio doesn't count in
typical
> > > Tesla Coils. A cruel irony of this is that you can make Ls/Lp
> > > arbitrarily high, but the lower the surge impedance of the primary,
> > > the higher the gap losses.
> > >
> > > Malcolm
> > >
> > Hi Malcolm and all
> > 
> > I have been following with great interest the various posts on gap
> > losses but not always accepting the conclusions put forth. I assume
from
> > your answer above that as the surge impedance goes down then the surge
> > current increases. Are you then saying that gap losses go up due to
this
> > increased current? Doesn't this assume that the "resistance of the gap
> > while it fires remains constant?
> > 
> > My experience with synchronous gaps leads me to a different conclusion.
> > I feel quite sure that the current in a firing gap is a function of the
> > voltage in the cap as well as other factors. I note that the contacts
in
> > my gaps run much cooler when the gap is firing near the peak of the
> > mains voltage. When firing ahead of the peak (by adjusting the gap
> > position with respect to the motor shaft) I notice that the gap pins
get
> > much hotter. I attribute this increase in gap heating to increased
> > resistance in the gap due to a lower current/voltage. I believe that
the
> > gap resistance is dynasmic and definitely increases at lower currents
> > which can definitely be controlled when using a synchronous gap. My
> > primaries usually are 3 to 7 turns and I think on the low impedance
side
> > and therefore have low surge impedance. Still at 1800va input I have no
> > trouble using .25" brass contacts in the rotary and they give very long
> > life.
> > 
> > Flames and comments will be appreciated
> > 
> > Skip
> 
> 
> Skip -
> 
>   What might be happening is that early firing on the 60Hz waveform
> leaves magnetic energy stored in the transformer's core. Hence, you
> are not only dumping the capacitor's energy, you are also dumping
> the residual energy in the transformer as well. If the gap fires
> late, then cap voltage goes down, but the transformer is now 'cooling
> off' with less energy available.
> 
> - Brent (bturner-at-apc-dot-net)

Hey, let me get in here and stick my foot in my mouth!
Could it be that when there is early firing you have *more* sets of gap
sparks for the duration of time that the electrodes of the rotary gap are
lined up? To put it another way: if the gap fires more often, won't it
naturally get hotter? 

When you synchronize the rotary gap to fire at the peak of the capacitor's
voltage charge won't there be fewer sparks during the time the gaps are
lined up? The initial spark across the gap will conduct more energy, but
more energy *conducting* does not necesarily mean more heating of the gap. 

The temperature of the gap is a function of heat _losses_, and when the
capacitor initially discharges at its peak, I think we will all be
surprised to find that this is *not* where the gap wastes the most energy.
In fact, my gut feeling is that this is actually where the gap functions
most efficiently. 

I would not be at all surprised if testing were to show that when the
breakdown voltage of a gap goes above a certain value, its actual
resistance per firing is *lower*. Sort of like when certain materials
become super-luminant and increase in efficiency when pushed past a certain
limit.

Tesla's researches into luminant gases were, in my opinion, more than just
a side issue. He was studying plasmas. Why? Aren't spark gaps deeply
involved with plasmas? He mentioned several times that an excited gas could
carry currents much more efficiently than the best copper conductor. I
don't think he was just referring to his light experiments, but also to
what he had learned about spark gaps. 

This could explain some of the advantages that rotary spark gaps exhibit,
and maybe lead someone to explore if energy efficiency would increase if a
rotary spark gap were to have electrodes that were large in conduction
area, but small in width. This type of electrode, set up to fire *once* at
each peak charging point of a capacitor might produce some very interesting
pulses! I believe the electrodes for such a rotary spark gap would probably
be pulled much further away from one another than is currently the custom.

By the way, did you ever wonder why Tesla was interested in operating some
of his Tesla coils from special AC generators that he designed that allowed
him to use primary frequencies that were way above 60Hz? Could it be that
at a frequency of say 400 Hz one might be able to increase the firing rate
and still be operating the rotary gap in a synchronized fashion such that
each firing would be at a large peak value?

OK, now that I have firmly placed my foot in my mouth, let's see what the
experts say... This should be fun!

I believe someone out there at one time or another used fiber optic cable
to help see what was going on when the rotary gap fired. Does this
individual remember if there was a correlation between the firing angle and
the number of times the gap fired during the time the electrodes were lined
up? 

Conjecture is fine, but data always brings out the truth.

Fr. Tom McGahee