After stable Voltage output my MOSFET drain and high side of half bridge is in bursting mode

[Vikas ji]

Hello everyone I am new here.
I have problem in half bridge SMPS in which my output voltage is 23v stable which I set through optocoupler feedback. But after output voltage stable when input given around 120 v AC the drain and High side gate goes in bursting mode and not making own smooth.

I just want to know in half bridge SMPS is normal or it is a problem.

Dummy load is 100k and pwm frequency 50khz,
Deadtime 900ns,
The drain waveform and I have is

 

[betwixt]

The 1K resistor at the top of the schematic probably should be across gate and source pins, not gate and ground. With high voltage on the drain you are probably exceeding the drain-gate limit.

Brian.

 

[Vikas ji]

Thanks for your reply.
It is only test purpose before I connect MOSFET with it. It is used to discharge high side so I can get pulse in CRO.
Then after that everything ok on this predriver module I removed this 1k resistor and connect the MOSFET and so on circuit.
SMPS circuit working fine .
But when I check the drain point it shows this type of waveform like sine wave not own it is ok or not

 

[Vikas ji]

Reference circuit is this which I using

 

[FvM]

The converter is running with effectively no load, resulting in low duty cycle. Without knowing the transformer properties, it's hard to say if the waveforms are o.k.

 

[Vikas ji]

I make the pwm stable by using 48v zener diode. So the output is approx 50v and pi filter capacitor at output is 1000uf. But after stable voltage current rising slowly without load and watt also increasing.

 

[cupoftea]

..From scm in top post, it looks like you have 1000uF just downstream of the output diodes...that should not be there.

 

[Easy peasy]

you have put a dummy load on the converter ( the zener ) to " stabilize it "

actually at no load - your previous condition - one expects the control to bounce around a bit - as when power goes to the output it is stored as a higher than target Vout on the cap, and it takes a while for it to bleed down to where the control ckt starts the pwm again . . .

--- Updated Sunday at 1:43 AM ---

@cupoftea - it has a soft start - but you have spotted exactly why these types of designs fail after a variably short period in the field - even with soft start.

on the plus side - he has removed the need for output snubbers on the diodes ( ha ha )