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

Modeling a magnifier




----------
From:  FutureT-at-aol-dot-com [SMTP:FutureT-at-aol-dot-com]
Sent:  Sunday, February 22, 1998 11:08 AM
To:  tesla-at-pupman-dot-com
Subject:  Re: Modeling a magnifier

In a message dated 98-02-22 00:41:18 EST, you write:

<<  
> > The higher coupling coefficient in the magnifier transformer is only a
> > consequence of the splitting of the secondary coil. The actual coupling
> > coefficient, considering the transformer and the third coil, is as low
> > as in a conventional coil.
 
> Applause sign, studio audience applause, fanfare!
> Camera one fade to black, camera two go wide on 3,2,1...
  
Antonio, Robert, all,

This concept has been also postulated by Lou Balint of PA who
has been researching magnifiers for a number of years.  I've posted
the results of some of Lou's work on this list from time to time over
the past year but never received many responses.  I'm glad to see
further interest arising in this area.

> > And this is more polemic:
> > The dynamic behavior of the magnifier system is practically identical to
the
> > one of a conventional coil (with that model, exactly), and so there is no
need
> > for different operating frequencies and special spark gaps (as I see
commented 
> > in several places).

This has been suggested by Lou Balint, Duane Bylund, and myself
(the tuning part), (and probably others) for some time.  As far as
tuning is concerned, it is clear from experiment that the primary
must be tuned to the combined secondary and resonator L and C. 

Regarding the use of special gaps, I have used both special rotary 
gaps and normal rotary gaps both on magnifiers and on classic coils 
and have seen little difference in performance.  Scope tests show 
the performance of series quenching rotaries to be only marginally 
better than regular rotaries.  More on K below.

> > The advantages of the magnifier are that the high voltage terminal can
> > be moved away from the primary circuit, and that a more compact and  
> >predictable primary circuit can be used.
 
> Applause sign, studio audience applause, fanfare!  Roll commercial in 
> 3,2,1...

I speculate that Tesla may have gone to the magnifier design as
a way to fit the coil into a reasonably sized building. 
 
> > What do you think?
> > 
> > Antonio Carlos M. de Queiroz
> > http://www.coe.ufrj.br/~acmq
 
> Antonio, 
>snip>
 
Then Robert replies:

> BUT, given that HV resonator K factors to the primary are the same then 
> in maggy systems as classical two coil systems, how can we explain 
> the benefits apparently seen by the magnifier experts on this list by 
> employing super short rotary break dwell times.  At first blush the 
>traditional explanation has always been by saying you have much higher K, so
> the energy transfers in shorter time, so you have to put out the fire in the
gap 
> quicker.  But I think Carlos has correctly pointed out that the K to 
> what we must in my opinion call "the high voltage resonator" no matter how
you 
> slice it,  is in fact NOT HIGHER. The normal method of measuring K in 
> a maggy is to only measure the coupling coefficient seen between the 
> primary and the segment of the HV resonator that is magnetically 
> coupled to that primary.  I believe this technique to be erronious 
> and misleading.  Not that this number isn't an important 
> specification of such a system, but the conventional K to total HV 
> resonator spec should also be measured , recorded and considered as 
> important as well. 
 
It is interesting to note that although the K can perhaps be looked
at in this way, actual scope observations of the quenching of a
magnifier do show the beats occuring in agreement with the tight
coupling of just the primary and secondary.  The beats do not fit
the case of a loosely coupled system, i.e. the energy is transfered
quickly, in mathematical agreement with the closed coupled case.
I only did this observation and measurement once, so I could be
mistaken is suppose.  I generally don't like to be too sure until I've
checked something a few times but I'm posting this result anyway
because it seems to me like an important finding that others may
want to verify.  Such a quick energy transfer still demands that
special attention be paid to quenching requirements.  Quenching
may improve in a large magnifier (compared with a small one), due
to heavier spark loading and lower frequency operation.

> Just a related point.  If you take a classical Tesla coil and start 
> unwinding the secondary from the top down, stopping at intervals to 
> make K measurements, you will find your K factor increasing steadily as you 
> remove wire, until most of the secondary has been removed, then the 
> trend will reverse itself as you continue to remove the last of the 
> secondary.  That point where the K has maxed out represents the right turns 
> ratio and coil geometry IMO for a good maggy driver coil, if you are 
> going to use that partially unwound coil. 

This is very interesting, can you give an approximate proportion of
winding length that you found to be best for the secondary?  Did you
wind up with 30%, 40%, etc., of the original length for example.  This
may vary maybe from system to system.  But I'd be interested to 
hear of an example of what proportions you found to be good in a
particular system.

> Chances are you will want 
> to rewind a new one with heavier wire.  This changes the TPI figure 
> and the geometry and there you are back to experimenting again.
 
> Maybe the unwashed masses representing the classical croud among us should
> try rotary gaps that quench in tens of microseconds rather than the hundreds
of
> microseconds which is normal on most of our classical coils. 
> Certainly the multi-swept gaps of the maggys perform the 
> function of putting out the fire quickly by virtue of increasing the 
> approach and departure velocities of the electrodes greatly over 
> simpler 2  or 4 gap rotary designs.  Could the extra trouble 
> involved in the manufacture of such superior quenching, shorter dwell gaps
lead
> to better operation of classical coils if employed? 

I've posted numerous times that in all my tests, I've seen only a 
slight difference in quenching capability and coil performance 
between multi-swept gaps, and regular rotary gaps, when used 
in either magnifiers or classic TC's up to 5kW.

Also, I find that rotaries usually do not quench by stretching the arc.
The gap generally quenches while the electrodes are still lined up in 
a good system.  For instance the dwell may be 200uS, but the arc may 
quench in 40uS because the energy is all gone by that time in many
closely coupled systems.  In most systems that I have personally
scoped, the actual quench occured well in advance of the mechanical
dwell time.  This explains why many coilers obtain good results using
static gaps on large systems...the gaps quench due to excellent spark
loading from the toroid.  When the energy is gone from the primary, and
too little is left in the secondary (because it's been disappated by toroid
sparks) to transfer back to the primary and keep the arc alive, then it
quenches.  Adequate spark loading can probably do more to quench 
a spark gap than most special gap designs can do, IMO.  More 
research is needed.  I would like to hear from anyone who may have
scoped their quench times and compared them with the mechanical
dwell times.  Be sure to allow for electrode overlap as they fly past
each other when figuring out the mechanical dwell time.

> I have to admit 
> that I haven't 'dwelled' too much on such details through experimantation in
> actual hardware yet.   One correlation I think I have made however is that
shorter 
> gap dwell times seem to allow a classical system to support a more massive 
> topload C.  

This would seem reasonable, since more energy is available with a
shorter quench time.  I think we should define our terms for quench
time and dwell time also.  I generally refer to quench time as the actual
time the gap arc is in existance, I use dwell time to refer to the 
mechanical dwell time of the gap; the two can be totally unrelated.
I have seen large magnifiers having a mechanical dwell time of 200
uS, and I'm quite sure they are quenching at about 45 to 100us, 
maybe someone will check this out on a large system.

> I expect an intelectual firestorm may result from this topic.  Flames 
> or positive critisizm welcomed.

Robert, BTW I've obtained 22" sparks from a classic tube coil using
(1) 833A, in steady running, at 15A, on a 120V line, using an MOT
with voltage doubler.  Thanks for your inspiration and info in that area.

Comments welcomed,
John Freau

> Robert W. Stephens
  >>