[Solved]Calc board based on STCH02. Assemble it, but it does not work properly!

[MHoly]

Worked in burst mode from 50V HV, then lower output voltage till 120VAC input, and almost totally powered off when input VAC above 135V. The ZCD resistive divider is 82 kOhm and 12 kOhm. Transfomer has 100 turns, 7 and 7 turns respectively. Lpri about 2.7 mHn. I tried 1.3 mHn and 0.9 mHn too. The same results.  Why?

Did anyone have some problems like that? Is anyone know how to calc an impulse power transformer for STCH02 design? 

[MHoly]

In additional.

Made recalculations in eDesignSuite. Transformer now: 59;11;11 turns respectively. Lpri = 0.59 mHn (like in calculations). Double checked directions of winding.

The same situation.

Am I correct understanding the QuasyResonant by this PWM-chip? That mean, no matter what transformer you have, chip will turn on MOSFET, wait till current increased to 0.75V, then MOSFET turn off, and by resistive divider, wait for demagnitizing, which chip see like a negative falling edge? and MOSFET turns on, etc. That is a QR mode in heavy load, when feedback by optocoupler is off all the time.

[MHoly]

Solved!

Problem was in my pcb. Was broke or something in feedback on ZCD pin.

By the way. I strongly recommend that chip for new design. I achieved 15W in range 53-45 VAC/ And stability 12V in output.

[MHoly]

Sorry...

53 - 405VAC input voltage range.

[MHoly]

But mostly important. It works on different transformators.

I tried 2 transf. in on same pcb:

1) Lpri=2.7mHn, 100:7:7 turns gap = 0.2mm

2) Lpri=0.59mHn. 59:11:11 gap = 1 mm

It work well.