forward converter design

this is vijay,I'm working on forward converter topology(conventional tertiary winding topology).I'm using tl494 ic for driving the main mosfet.the problem is i want just to test the circuit ,no feedback control is required .what i want is just to turn on the primary mosfet with some duty factor say 0.45
I'm using IRF640 mosfet
rating 18A,200V,180mohms Rds(on)

circuit design specifications:
please i want an idea how to drive the mosfet using tl494
and also any infomation of sites for finding useful information on simulation of forward converter topologies(third winding reset,active clamp reset)and also for forward converter with synchronous rectification:confused:


2005-06-21 1:00 pm

First of all for driving a mosfet using the old TL494 you need a push-pull driver circuit, which could be easily implemented with few bipolars (e.g. BC337;BC338) or you could use a dedicated mosfet driver or a driver transformer.

I think that the first solution is more convenient for you.

From my experience I can tell you that is better to set the output transistors of the TL494 in the emitter-follower configuration. Another good ideea is to connect them in parallel.

For the simulation part try :

If the MOSFET doesn't have to be isolated from the TL494, a simple complementary PP driver will do it. Heck, the chip might be able to do it naked, I forget what kind of output circuit it has.

If it does have to be isolated, at worst you can drag out a transformer driven by a comp PP stage. That's what I use for driving 6 x STW11NB80 from 5 to 130kHz (it should do 200-300kHz fine, I just have to set the oscillator higher to reach it).

TL494 Forward Converter


While you're at it, why not just do a push-pull design for the main power transformer? I have used this double-ended PWM Chip for a variety of DC-DC converters, both step-up and step-down. Since the forward topology calls for a tertiary "reset" winding (to keep the core from saturating), why not just replace this with a center-tap primary, and use a second IRF640?

You will get nearly twice the power density for the same core used. Or, conversely, you could go with a smaller core.

Anyway, a push-pull would add only a few more parts and nearly double your power density. And, the IRF640's Vds is high enough to work with a 48V input.

As for driving the '640 from the '494, you could use a dedicated MOSFET driver IC, as mflorin said, like the MC34151 or MC34152 driver ICs. Or you could do an emitter-follower, using the 494's internal uncommitted driver as the upper drive transistor, and a PNP with proper biasing for the lower driver, giving a pseudo-totem-pole output from the '494 to drive the IRF640.

Also, synchronous rectification lends itself nicely to the push-pull topology, too.

See Chryssis' or Pressman's or Brown's SMPS design books.

Think about the center-tap push-pull design, though............

Steve :cool: