Re: Cap Location (fwd) A correction (fwd)

---------- Forwarded message ----------
Date: Tue, 13 Jan 1998 20:05:04 -0800
From: Bert Hickman <bert.hickman-at-aquila-dot-com>
To: Tesla List <tesla-at-pupman-dot-com>
Subject: Re: Cap Location (fwd) A correction (fwd)

Tesla List wrote:
> ---------- Forwarded message ----------
> Date: Mon, 12 Jan 1998 22:30:46 -0700
> From: terryf-at-verinet-dot-com
> To: Tesla List <tesla-at-pupman-dot-com>
> Subject: Re: Cap Location (fwd) A correction
>         As John Couture probably suspected, I miss quoted the numbers for
> resonant rise.  With a 60mA 15KV neon transformer connected to a 0.01725uF
> capacitor (no other load present), the resonant rise of the system raises
> the output voltage by about 40% above what one would expect.  Thus at 120
> volts input, instead of 15KV on the output I may get about 21KV.  This
> stress usually will not kill a neon but doesn't help it much.  The real
> problem may be the input current.  Spice modeling (which has accurately
> predicted other parameters I have measured) predicts the input current to
> the neon will be about 18A and the output current could go to 140mA.
> Apparently, the neon's current limiting my no longer function as designed
> when the output voltage and current are 90 degrees out of phase.  This will
> definitely hurt the neon if the theory is correct.
>         I have only tested this theory at low power due to current limits on
> my equipment.  Now that the probe is working, I can briefly push the system
> to check this theory out.  Since the transformer is brand new (former
> testing did not go well :-(, I want to be very careful.
>         If anyone has any thoughts on neons drawing high currents or
> producing higher than rated output voltages when a capacitor is connected to
> their output terminals, I would be interest to hear them.
>         I haven't investigated power factor correction much but I fear the
> power factor capacitor and the primary winding of the transformer may see
> high resonant currents even though the AC input looks better.
> Much to work on.......
>         Terry
>         terryf-at-verinet-dot-com
> >


The approach works very well. I'm using two 15 - 60's in parallel with
total tank capacitance of 0.021 uF. This places me squarely at 60 Hz
resonance. My total gap setting is 0.54" (18 static gaps at 0.030"
each). The line current jumps up significantly, the transformers and
homemade cpas begin fairly loud "humming" and the gaps begin
sporadically firing at only about 15% of the variac setting. Increasing
the voltage setting causes the gaps to more consistently fire as well as
makinmg them fire more times per half cycle. 

The gap's firing has the effect of lowering the effective Q of the
tranny-tank cap series LC circuit, as well as helping to clamp the
maximum voltage seen by the transformer outputs. Because the tank cap
tends to neutralize the inductive current limiting at resonance, the
tank caps can now recharge much quicker. The result is about 3-4 gap
firings every half-cycle. However, the current (and voltage) stresses
are much greater than the transformer designers intended. Hard-won
experience has also shown that 60 MA transformers are MUCH better at
handling these overcurrent stresses than 30 MA ones. I no longer use any
30 MA trannies after losing four in relatively quick succession.
However, over a two-year period, I've only lost one 15-60.  

With 167 uF of power factor correction, the output short-circuit primary
current draw for the transformer-pair is about 12 Amps. The "running"
current at full power is between 25 and 30 Amps... the difference being
due to the neutralizing effect the tank cap has on transformer current
limiting. However, without PFC's the current draw climbs to 35+ amps
until the breaker blows.

Safe coilin' to you, Terry!

-- Bert --