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

Re: Voltage doubler (fwd)





---------- Forwarded message ----------
Date: Thu, 15 Apr 1999 07:39:25 -0400
From: David Sharpe <sccr4us-at-erols-dot-com>
To: Tesla List <tesla-at-pupman-dot-com>
Subject: Re: Voltage doubler

Tesla List wrote:
> 
> Original Poster: David Trimmell <davidt-at-pond-dot-net>
> 
> Hello, I was wondering if anyone on the list, who has a good understanding
> of what is actually going on within a voltage doubling circuit, could
> explain it to me.  
------------------------------------------------
SNIP
------------------------------------------------ 
> David Trimmell
> 
> PS. I am using the doubling circuit that Dave Sharpe and John Freau have
> used on their tube coils, i.e., cathode grounded on the diode.

David, ALL

This is one of the studies to perform once my new controller is on line.
In my system the diode energy storage capacitor is in series with tank
circuit HV input and the diode shunts (in parallel) with tank HV input
and cathode (filaments) of tube.  In most normal configurations, with
the fils being ground referenced, means the ANIODE of the half-wave
doubler circuit (HWDC) diode is connected to ground.

This circuit actually is NOT a voltage doubler in the strictest
sense.  It has been termed in the past a LEVEL SHIFTER, in that the
input voltage to the tubes (under no load conditions) on a high value
resistive load (near open circuit conditions) appears as a sine wave
level shifted upward or downward with the 0 voltage point as referenced
to ground is shifted +/- to a value approaching Vpeak of
input voltage.  Vpeak + Vpeak approached output of a doubler.

Under actual operation, when the AC wave goes negative, the doubling
energy storage capacitor is charged to Vpeak and energy is stored
magnetically in the powering transformer.  At the instant the AC wave
crosses 0 and goes positive; the diode commutates off, the tubes
start to conduct.  At this moment, the capacitor as well as stored
energy in transformer will discharge with voltages additively combining.
On oscillioscope, the current on the tubes appears as a vertical spike,
approaching 2X Ipk plate, exponentially decaying to Ipk, then following
AC self rectification wave from 90deg to 180 deg of remaining tube
conduction.  This spike is extremely fast (<50us) and appears to depend
on switching speed of HV diode.

The SWAG I was looking at (and crude modelling done in EXCEL) concerned
two possibilities:
1.  Capacitor sized to be resonant with powering transformer at
    50/60 Hz mains input    where P = 377 * C * Vin^2
    Vin is RMS input in this case
2.  Capacitor sized for energy storage to tube based on 8.33msec
    discharges * 50/60hz operation.
    Where P = E * N       P = nameplate of powering transformer
                          N = 50/60
			  E = joules E storage
    E = 0.5 * C * Vin^2
    Vin is PEAK Vinput in this case (since capacitor is charged to 
    Vpeak prior to discharge)
I believe in retrospect that 2 is probably closer to correct then
1, although no testing and validation of this theory has been performed.

Also, I suggest use of a bypass capacitor of 0.1uF across HV line
from output of this circuit to toobs and ground to prevent HV kickback
from damaging the transformer HWDC circuit. Also use a self discharging
capacitor out of a Microwave Oven for level shifter energy storage
capacitor, this capacitor could storage dangerous or lethal charges
under certain conditions.  Or install a bleeder resistor across
capacitor.  SCR,IGBT snubber capacitors (1-2kVAC rated), series
connected for proper powering voltage are excellent in this
application.

Corrections, comments welcomed

Regards

DAVE SHARPE, TCBOR
Chesterfield, VA