Half Bridge SMPS using SG3525+IR2110, Output not changing with duty cycle

Thread Starter

agsuresh

Joined Dec 28, 2023
66
Hi,

I have built an smps, Half Bridge using SG3525 + IR2110.
It works fine - Except for changing(control) the output voltage by changing the duty cycle.
The nominal dc voltage is 270. Half is 135v
The transformer core is E42/21/15
Primary windings : 20
Secondary 4+4 : gets 32v - needed only 22v
Mosfets : IRF840

I can change the output if a low value res is fixed in parallel to the output filter caps. But this wastes power too much.

I have tried practically everything , changing the frequency, snubbers......, inductor on the output after the bridge diode (dual rail output). etc etc. Nothing helps to control the output.

It seems the power is simply dumped into the caps and it stays there. The change is too slow for the dutycycle to adjust it.
Is there any recipe to control the output in a better way than load resistors ?

The only two things I haven't tried is changing dead time and changing the primary turns. But can any of that truly help eliminate load resistors ?
 

Thread Starter

agsuresh

Joined Dec 28, 2023
66
A picture is worth 1000 words. Where is the schematic.
Though the schematic in general is same, I have made some modifications according to the core size + few others to make it even better. The feedback is not connected, but a mechanism to manually change the duty cycle is implemented to verify the actual working voltages and range.
 

Attachments

ronsimpson

Joined Oct 7, 2019
4,664
What load do you have from +70V to -70V? I think your load is very small. My first guess is that the output current is so low that the output is peak rectifying and not finding the average voltage. If this is the case the output will be the peak voltage over 10% to 90%. Increase the current or increase the 20uH to a much higher value for a test.
1777381439481.png

There are two output filter inductors in the output. Are they coupled into one or are they separate inductors?
 

MisterBill2

Joined Jan 23, 2018
27,374
For a transformer with so very few winding turns, the frequency should be several MEGAHERTZ, which I doubt it is. Coore saturation or output saturation will usually cause a lot ofproblems. Your transformer needs more turns to provide more reactance.
 
Top