Re: Choice of Diode (fwd)

[Steven Roys <sroys@xxxxxxxx>]

---------- Forwarded message ----------
Date: Sat, 3 Apr 2004 20:17:17 -0800
From: Jim Lux <jimlux@xxxxxxxxxxxxx>
To: High Voltage list <hvlist@xxxxxxxxxx>
Subject: Re: Choice of Diode (fwd)

Most of the HV switching supplies I've worked with (non-flyback, non
multiplier type) use a series of secondary windings (chosen for the reverse
breakdown voltage of the diodes) each with their own full wave doubler (2
diodes, 2 capacitors), and all the doublers hooked in series.  This means
that any one winding has limited dv/dt, limited voltage, etc.  Since run of
the mill diodes typically are around 1kV, if you put out 20 kV, there would
be 10 windings each putting out 1 kV.

The other scheme is the "TV anode" flyback scheme, where you have pulses
driving a multiplier. The multiplier has really poor regulation (voltage
varies a lot with load), while the multiple windings and diodes has much
better regulation.

The number of turns, number of diodes, and number of capacitors is the same
whether you use one big winding and string the caps and diodes in series, or
break the winding into smaller chunks. The winding divided into smaller
chunks inherently distributes the voltage stresses more evenly  (no worries
about reverse recovery on diodes, capacitor tolerances, etc.)  The only
problem is bringing out more wires from your transformer.

----- Original Message -----
From: "High Voltage list" <hvlist@xxxxxxxxxx>
To: <hvlist@xxxxxxxxxx>
Sent: Saturday, April 03, 2004 7:16 PM
Subject: Choice of Diode (fwd)


> Original poster: Steven Roys <sroys@xxxxxxxx>
>
>
>
> ---------- Forwarded message ----------
> Date: Sat, 03 Apr 2004 20:50:44 +0930
> From: Matthew Smith <matt@xxxxxxxxxx>
> To: High Voltage list <hvlist@xxxxxxxxxx>
> Subject: Choice of Diode
>
> Hi All
>
> Further to the discussion on the Tesla list regarding pulsing car
(automobile)
> ignition coils with an HV supply capacitor and gap, I have discovered that
I am
> lacking a power supply in the 800-1500V range.
>
> Looking at the parts that I have lying around, I have decided to build an
SMPS
> (TL494 or similar) with full voltage feedback using a 1:58 flyback
transformer,
> similar to a TV line-out transformer.  This is rated at 1500V output,
which
> would suggest a convenient 25V drive.  (I have several matrix printer
SMPSs to
> hand with 5V/24V output.)
>
> I'll run at somewhere around 40kHz, as I don't know if the ferrite core is
up to
> being driven at the 100kHz of some of the other switching devices I have
lying
> around.  Switching will be done by an IRFZ48V, which has a very low 'on'
> resistance.  A suitable snubber will be provided since this device is only
rated
> to 60V.
>
> On the secondary side, I am unsure what to do.  Do I make a string of,
say, four
> standard 1000V Schottky diodes as normally used in SMPS duty, or would a
> microwave oven diode be suitable for operation at this frequency?  Also,
would a
> microwave oven capacitor be appropriate after the diode?  Peak voltage, if
run
> at 25V in, would be just over 2100V, so well within range of a cap like
this.
> But what about frequency?
>
> Cheers
>
> M
>
> --
> Matthew Smith
> Kadina Business Consultancy
> South Australia
> http://www.kbc.net.au
>
>