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

Re: Cap safety gap (renamed) (fwd)



---------- Forwarded message ----------
Date: Wed, 17 Dec 1997 10:22:06 +1200
From: Malcolm Watts <MALCOLM-at-directorate.wnp.ac.nz>
To: Tesla List <tesla-at-pupman-dot-com>
Subject: Re: Cap safety gap (renamed) (fwd)

All, 
     The following is a bit of a thinkin' aloud ramble as I don't 
have much rotary experience. I stand to be corrected on anything I 
say. In fact, the idea of this piece is to promote some discussion so 
that for those of us who don't know all the answers (me), something 
can be learnt.

> Date: Tue, 16 Dec 1997 14:04:02 EST
> From: Esondrmn <Esondrmn-at-aol-dot-com>
> To: tesla-at-pupman-dot-com
> Subject: Cap safety gap (renamed)

<snip>

> Malcolm,
> 
> I would appreciate your thoughts on this.  The last time I had this coil set
> up (outside), I tried many combinations of primary set up.  This includes
> welder settings, resistance in series with welder, rotary speed and variac
> settings.  I added 600 ohm resistors in series with each H.V. feed line.  No
> bypass caps and no chokes in the H.V. feeds.  With resistive ballast only, I
> get a popping sound coming from the rotary gap that doesn't sound good and the
> gap firing is erratic.  When I add the welder in series with the ballast
> resistance, the popping in the rotary goes away and the system runs much
> smoother.  The discharge sparks are much hotter and louder than with resistive
> ballast only - at the same primary voltage and current.  Is this an indication
> of a resonant charging condition that is allowing the voltage across the cap
> to ring up too high??
> 
> Should I increase the H.V. series resistors from 600 ohms in each leg to maybe
> 3,000 or 4,000 ohms?   At 500 ma (5 kva at 14.4 kv), I will run into a
> dissipation problem in a hurry along with concern about voltage breakdown
> across the resistor.  At 3,000 ohms, the resistor would be dissipating 750
> watts with 1,500 volts across it.
> 
> The best performance in the above testing produced 57" sparks with the welder
> set at maximum, series resistance at 4.8 ohms, variac wide open and rotary
> speed at approx. 400 bps.  Lowering the series resistance or increasing rotary
> speed would cause the cap safety gap to fire.  Primary input energy at this
> point was only about 3.5 kva.  I want to get the system dialed in to where I
> can input all the power I have available - probably about 8kva.  I have
> produced 80" sparks in the past with this system.
> 
> Your idea of lowering the primary Q goes back to the original design where I
> had all the primary components interconnected with two parallel sections of
> RG213.  Most of these recent problems have occured after I replaced all that
> wire with 3/8" copper tubing - raising the primry Q.
> 
> Comments?  Thanks,  Ed Sonderman

First, I would like to say that nirvana has to be when no resistance 
is used at all because it ultimately wastes power and efficiency is 
the name of the game. It is possible this nirvana can't be reached 
with AC charging system but my gut feeling is that it can be.

    This seems to be a problem that is peculiar to asynchronous 
rotaries. IMHO it should never occur with a static gap unless it is 
set wider than the safety gap and I don;t think it should occur with 
sync gaps either. It is undoubtably associated with resonant charging 
IMO as the cap should otherwise be unable to charge beyond the peak 
o/c voltage output of the transformer. Being asked to comment 
definitively on this one is a bit of a problem for me right now 
because I don't have a satisfactory model to explain it - you can bet 
I'm about to start working on it because I am migrating to rotaries 
very soon.
   
    A simulation using a transistor gap might yield some useful 
answers. I know from the last time I did this a couple of years ago 
(but for different reasons), I was able to *quintuple* the output 
voltage from a power supply with a boost/limiting inductor at slow 
enough gap speeds. Perhaps there are other artifacts that should be 
taken into consideration although they aren't obvious right now.

    Things that would be useful to model this include knowing with 
certainty the inductance (and resistive values), and behaviour over a 
wide range of gap speeds. A fundamental consideration that *must* be 
tken into account is that to charge a given value of capacitance to a 
given voltage requires a very exact amount of energy. The only 
possible source of energy apart from the transformer and boost 
inductor is back-feeding from the secondary. This can only be a 
factor if the gap opens when a significant amount of energy remains 
in the secondary (gap opens at a time of minimum primary activity).
Experiments in the past suggest this is unlikely but I can't 
completely rule it out.

     Energy stored in the inductor while the gap is closed is 
inversely proportional to the inductance but proportional to the 
current squared, the nett result being that the smaller the 
inductor, the more energy is stored and the faster it can recharge 
the cap according to LC when the gap eventually opens. Additionally,
if the gap closes while the inductor is still carrying a current
(likely), then it already has some stored energy in it. I am sort of 
assuming that the inductor is recharging a la: Tcharge = L.I/V but of 
course, V isn't constant. If the gap is too slow, the inductor has a 
lot of time to charge itself up to a high energy level. If V is large
at the time, Tcharge becomes small. Conversely, fast presentations 
might not allow it to empty itself. A possible scenario: two 
succesive fires followed by a miss.

    What do the experienced rotary operators think?

Sorry for the verbiage. I want to understand this as much as you do.
Malcolm