# Re: Rewinding BIG transformer

```Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>" <jimlux-at-earthlink-dot-net>

The size of the filter cap depends on how much ripple you want to tolerate.
A first order approximation is to figure out the drop in voltage over 1/2
cycle at your current draw.  For 60 Hz, the time is 8.33 milliseconds, for
50, it's 10 ms (heck of a lot easier to do line frequency math with 50 Hz
power...).

voltage drop (i.e. ripple) (in volts) = current (in amps) * time (in
seconds) (8.33E-3)/ (filter cap size in Farads)

Example:  1000 uF, 1 Amp... 1 * 8.33E-3/1000E-6 = 1 * 8.33E-3/1E-3 = 8.33
Volts...

If it were a 1000 V supply, this would mean you'd have less than 1% ripple,
which is pretty good.  If you had a 15 V supply, you'd have about 60%
ripple, which is pretty bad (hence the huge caps on 5V DC power supplies)

Note that running at a higher frequency helps a whole lot.  At 400 Hz, the
half cycle time is only 1.25 ms, cutting the size of the filter cap (for the
same ripple voltage) by a factor of 6 or 8.

These are APPROXIMATIONS!  In reality, you are charging the filter cap
non-instantaneously, and with a finite source impedance, etc., but, they
will get you in the ball park.

----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Saturday, April 14, 2001 10:05 AM
Subject: Rewinding BIG transformer

> Original poster: "Jason Johnson by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <hvjjohnson13-at-hotmail-dot-com>
>
> Hello,
>
> I'm at sort of a dilema. I have had my BIG transformer apart for a while
now
> and I need to get something done with it. The core proved to be impossible
to
> take apart unless I attack it with a hacksaw, and with 5x15 inches of iron
to
> cut through it would not end up very neat. I've decided to therefore wind
it to
> a realtively low voltage, say 5kv, and use low voltage on some projects
and
> experiments and use a voltage doubler (or tripler, or other voltage
multiplier)
> and run with DC for others. The low voltage/doubler configuration can be a
good
> thing IMO, it requires fewer turns, less insulation, less wire, and gives
good
> flexibilty for my experiments. I've aleady got a nice gap in mind that I'm
> quite confident will be able to quench 5kv at 4 amps (think big tungsten
> carbide electrodes, supersonic airflow and 12kgauss magnetic quenching
> triggered gap in a semi-sealed possibly CO2 or H2 environment), for low
voltage
> experiments, and then the high voltage DC really shouldn't be much more of
a
> challenge than a gap for any regular 15kva+ coil (yeah right, thats childs
play
> :-). Its either this or someone can donate a 400hz generator (3 phase
would be
> nice too :-) capable of 24kva. Or I can feed it the square wave output of
a
> high power inverter (or maybe not :-). Cuz 400hz needs many fewer turns
per
> volt for a given core size.
>
> How big do filter caps for DC need to be? Anybody with an equation? Ideas
for a
> voltage doubler with 120hz pulses, schematics for various voltage
> doublers/multipliers/level shifters?
>
> Also, were more tests done on the INDUCTIVE SOMETHING OR OTHER thread from
> awhile back?
>
> Whadda Y'all think? Anybody got any better ideas?
>
> Jason Johnson
>
>
>

```