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

Re: Spark Gap Replacements



Tesla List wrote:
> 
> Subscriber: FutureT-at-aol-dot-com Wed Jan  8 22:52:13 1997
> Date: Wed, 8 Jan 1997 17:46:38 -0500 (EST)
> From: FutureT-at-aol-dot-com
> To: tesla-at-pupman-dot-com
> Subject: Re: Spark Gap Replacements
> 
<SNIP>

> Bert, as you know, your high powered fast-quench work is of great interest to
> me since my work was done at low power;   I was wondering how quickly your
> secondary rang down, after the gap quenched, for instance how many
> microseconds until amplitude dropped to 50% of the initial (after quench)
> free ringdown (I'll need to know your operating frequency again too)?  And
> did the free-ringdown have a steeply-swooping-down non-linear look to it?
>  The reason I ask, is that in my work, at lowest power, free-ringdown was
> linear and gradual, (high Q damped wave), but as I turned up the variac power
> just slightly, the free-ringdown petered out much more quickly due to a lower
> Q resulting from greater spark loading.  (of course we expect all this).  But
> at your power levels, I would expect the free-ringdown energy to drop
> extremely quickly, due to heavy loading and even lower operating Q's.
> 

Well... it's 10 degrees F. on the porch tonight! I fired up the coil
briefly, and it ran better than it has all summer!! Made a few quick
measurements, under "single shot" conditions, using a storage scope tied
to a pick-up plate about 9 feet from the coil. Fo = 90 kHz. The wife
thinks I'm nuts!!

Case 1: Quenched at first notch, no toroid breakout, secondary free
ringing:
The "ring down" time from peak voltage to about 0 (4 time constants) was
about 2000 uSec, implying a Tau of about 500 uSec. The damping factor is
exp^(-Rt/2L) which is the same as exp(-t/Tau), so Tau = 2L/R. L = 73445
uH for my coil. Solving for the "effective" secondary R:

   R = 2L/Tau = (2*73445*1e-6)/(500*1e-6) = 146890/500 = 245 Ohms
       (This is higher than what I saw earlier this year... frozen
ground         making the RF grounding system more lossy??)
 
   Z = Inductive Reactance at Fo = 2*Pi*Fo*L = 41532 Ohms (inductive)

Estimating Q:  Q = Z/R = 42532/245 = 170   

Case 2: Quenched at first primary notch, with toroid breakout:
This is harder to estimate.  While the streamers are pulling energy from
the coil, the first 150 - 200 uS decline very rapidly (almost
linearly!). Then, the coil more slowly rings down as, apparently, the
corona is extinguished and Q rises to its pre-breakout level. Since
"bang energy is not controlled, there's considerable variation from shot
to shot. Picking a typical one, and going from the peak to 10% there are
about 14 RF cycles. Conveniently, the 10% point is also just about where
the streamers are extinguished. Using Duane Bylund's relationship to
solve for Qs when N=14:

    Qs = N/0.7329 = 19.1

This represents almost a 9:1 decline in secondary Q between no-breakout
and breakout case. I haven't made any recent full power tests, but
earlier runs indicated that as more power was used, the PPS rate went
up, and as streamer length and intensity went up, the rate of secondary
voltage collapse became even heavier. I suspect that the effective
secondary Q could easily approach that of the primary (about 11) or
maybe even lower, under these conditions. I'll wait for a warmer day to
rerun these tests, however!

> When your spark dropped to 1/2 the length at the 4th  notch or transfer
> point, did you try retuning to see if this might "recover" some of the lost
> spark length?

Since I've always tuned for maximum spark length based upon the best
quenching I could do, so I didn't try this wrinkle. Unfortunately, with
the cooling air shut off, my vacuum gaps get very hot, very fast at
higher power levels, so I'm not inclined to try running this test...

> 
> >Full input power runs about 3000-3200 Watts (26-27 Amps at 120 VAC,
>  >power factor corrected).
> 
>   >snip
>  >> >> > John  Freau
>  >> > Safe coiling to you as well, John!
>  >> >-- Bert --
> >>John Freau
> 
>  >True first-notch quenching should certainly be achievable with a
> > properly designed rotary gap, and even with an air-blast gap.
> 
> I tend to agree--it will be interesting to see just which designs lead to
> this elusive goal.   Thanks again for all the information.
> 
> > Safe coilin' to you, John!
> 
>  >-- Bert --
>   >>
> Safe and effective coiling!
> 
> John  Freau

Hope this helps... chatter, shiver...  :^)

-- Bert --