determining smps transformer type and primary turns.

[Salvador12]

surely a scope is needed and my friend hes an old electrician has one we actually measured the gate waveforms with the old transformer arrangement and they were nice. i actually have a 7815 regulator to be sure i have some headroom.also i have reostats for deadtime and frequency , i have increased the deadtime resistance a bit bigger than what is used in the schematic to give me some safety.
well I doubt the transformer rearrangement directly caused the fets to go bad.
I think the heat is to blame for their destruction , the question is were did the heat came from , most likely overcurrent in the primary because previously the fets were dead cold yet the secondary voltage decreased upon a heavy load and the maximum output power was never higher than about 500w, ususlly less from my voltage and resistance calculations , this time under the same load the secondary voltage didint drop one bit.

maybe i should just disconnect the feedback IC? well i will ge new parts and measure the supply voltage under the same load , im talking about the IC supply voltage from the 15 volt regulator , if it will be about 15 volts or close i can assume even without scope that the waveforms are good.

still i wonder what made them so hot , if they would have had shootthrough I think they would ahev destroyed themselves alot faster and with enough deadtime it shouldn't be the case, prety much im thinking either overcurrent or somehow they were driven not fully open , but if im correct mosfet's gate capacitance doesnt increase with high load on the drain source channel does it ? and if it doesnt then why suddenly the IC's couldnt drive the mosftets nicely if they did so before all the time.

 

[FvM]

Provided there's no overcurrent due to massive core saturation (unlikely if the core is calculated correctly), the transistor overheating and damage is most likely cause by overvoltage transients and avalanche breakdown. In a bridge circuit, overvoltage can only happen with bad layout or insufficient bus bypass capacitors.

Another possible problem is exceeding the maximum Vds dv/dt during turn-on.

 

[E-design]

The problem may be due to oscillations caused by bad layout and poor feedback compensation like mentioned by other members.

 

[E-design]

A simulation model of your system seems to indicate possible oscillatory behavior due to incorrect feedback compensation. This may be aggravated with bad circuit layout.

 

[Salvador12]

ok so heres how far I am with this smps , ive put it in the amplifier chassis which is self made welded from aluminum to act both as the structural chasis and also as the heatsinking elements of all the major heat dissipating parts.
a bit increased deadtime , a transformer with 12 turns on primary of parallel litz wire and also I added an old CPU aluminum heatsink to the transformer using aluminum cooking foil as the thermo paste and it works , I've had my highest load of about 20 ohms at which the output was 155v DC giving me about 1.2KW of output power, I'd say not bad, everything gets warm ofcourse but with my heatsinking reinforcments works well, even the mains rectifier has a heatsink now cuz under heavy loads it gets hot too, the mains cable gets warm and the wall socket plug too.
of all the things getting hot the switching mosfets are among the cooler ones so i guess that is a good sign.

I am now making the second transformer using the F1 67 ferrite type , because i will have two half bridge smps on a single board , a question I might want to add is can I drive two IR2110 from a single SG3525AN chip? Given all 3 IC's have the same power supply.

I will add some pictures later if I'll take ones.