Cytron Technical Forum

This is an experimental circuit, using a small transformer to drive P-channel MOSFET, using
a LM2576 adjustable regulator.

Powered by 12V electronic transformer, having a cooler fan for the P-channel MOSFET,
and input voltage can be as high as 60 volts (and more). 50V here for the test.

Output is adjustable. Power test was done with 12V/50W halogen bulb, 4.16 Amps current.
Actually it can not startup at full voltage, since the LM2576 has overcurrent protection which still works
even through the control transformer. To start the 12V bulb, at first voltage must be reduced, then increased again to 12V.
For smaller bulb, 35W, this works automatically. Like a built-in softstart!

The coil is not well matched neither the control transformer, and the gate drive is crude, just the transformer secondary.
Efficiency is about 70%. The TO220 MOSFET is mounted like DPAK, to give best contact for the cooler.





the TO220 power diode also needs a small cooling grid at 4.16 Amps, and gets hot!

The connections are hard to see ! Was the MOSFET underneath the fan?

Why are there 3 coils? What are the other two for?

Allen

yes the MOSFET is under the fan.

One coil is not enough, not matching, so two coils in series. They are not made for LM2576,
unknown value.

Schmematics attached yes hard to see from the prototype photo.

Any ideas how to improve the circuit?

You can build one LC meter to measure the inductance of your coil. http://electronics-diy.com/lc_meter.php

OK yes I should build LC meter, capacitors I can measure with my DMM, but coils = just guesswork.
I made TL494 switching power supply! The coil heats up considerably, even if I have rewound it, having more windings now.

50V -> 12V, have done a test yesterday loading it with 6 Amps but the static cooler is too small for that.

For LM2576, frequency can not be changed/adjusted...



Using a power PNP here 2SA1492, and base drive = 47 Ohms + 250 Ohms resistor bank, they start to smoke at higher voltages so must replace with power resistors.

Rebuilt this circuit on smaller space.

LM2576 supplied by 2x 1.2K 1W resistors voltage divider.

Damping resistor for the gate drive (primary) = 470R now.

Improving, really happy to see that