Re: MOSFET dynamic ripple filter for variable supply


"ehsjr" <ehsjr@xxxxxxxxxxxxxxxx> wrote in message
news:Nsfth.3964$o31.1983@xxxxxxxxxxx
Paul E. Schoen wrote:
I have a high current DC supply consisting of a 0-140 VAC variable
transformer driving the primary of a transformer with 20-0-20 VAC output
at about 8 amps. I have the output switchable for FWCT or FWB with some
50,000 uF capacitors to get about 0-25 VAC or 0-50 VAC at nominal 10A or
5A, with occasional surges up to 3x. It works OK for rough work, but the
regulation and ripple are very poor.

I designed a circuit which provides some output voltage load regulation,
but especially very good output ripple of only about 20 mV P-P at 8
volts out into 0.5 ohms load (16 A). At 100 ohms load the output voltage
is 11.6 VDC, and at 10 ohms it is 11.2 VDC, and at 2 ohms it is 10.17
VDC. Not bad for a simple supply. I also added optional current
limiting, which actually provides current regulation by using a
transistor monitoring the current sensing shunt, but I may replace the
transistor with an SCR to latch the output off, rather than burn up lots
of power in the MOSFET.

The MOSFET burns about 25 watts with the 0.5 ohm load, 11 watts with the
2 ohm load, and less then 3 watts for 10 ohms. These figures are all for
16 VAC from the transformer. These are all values obtained using a
simulation in LTSpice. The ASC file follows.

I have not built it yet, but I think it is a workable design. However,
it is somewhat dependent on the characteristics of the MOSFET, as
discussed some time ago. I tried several models, and all worked, with
some adjustment of the bias resistors. I'll let you know if and when I
actually build it.

Paul


[snip ASCII file]

How are you going to get 16 amps out
of the regulator with only 8 amps
in?


In a FWCT configuration, each arm of the transformer contributes 16 amps at
50% duty cycle, so it effectively sees 8 amps. The actual RMS current
through the transformer is about 40% higher than the DC output, or about 11
amps, because of the capacitor charging to peak. But transformers can be
overloaded by 40% for intermittent duty, about 5 minutes on and 5 minutes
off. At my maximum output current of about 35 amps, the transformer will
see the equivalent of 25 amps, or about 3x overload. At that level it will
only be on for about 10% duty cycle, perhaps 10 seconds on and 2 minutes
off.

Paul


.

Quantcast