TOP234G at voltages > 200VAC

You will have to describe the problem so that I can give a specific answer. Since the problem is worse at higher input voltage, you may have a control loop stability issue. If you are using a TL431 error amplifier/reference, for testing purposes, you may want to replace the TL431 with a zener diode. This will decrease the loop gain. If the power supply works properly, you will need to analyze the loop to find a fix.

Z. Cochrane

Hello, thank you for the help. First of all tell you I am the person who posted you yesterday regarding the Coss of the TOP234G. I think I am having problems at high input voltages due to the Coss. At high temperature, 60ºC ambient, the voltage supply works worse than at lower temperature and thermal shutdown happens. Could Coss be bigger at high temperatures?

I do not discard the loop stability issue. Please find attached my feedback circuit. Note that C8 is quite big (220nF). I the past it was 100nF, but in some devices I had problem with the loop stability. When the load changed, a high ripple appeared in the voltage output of 5.3V. The ripple was 50Vpp at 2Khz. Increasing the value of C8, the problem disappeared and the output voltage is constant although load changes.

Which are the differences in the behaviour of the feedback loop at high and low voltages?

My issue is due to I had my design finished 2 year ago and now the design requirements are being increased. My input voltage is 65-250Vac for a 11W load. Is it possible to obtain a correct design using the TOP234G.

Regards

Coss does not change much over temperature. I do not believe that the issue is Coss. From your description, the problem is loop stability. This is a difficult problem to solve if you do not have a loop analyzer.
I noticed from your schematic that the resistor in series with the opto's diode is 430 Ohms. It is difficult to read the value of the voltage source, but it looks like 55 Volts. If this is correct, try increasing the resistor value from 430 Ohms to 3.9 k Ohms. This will reduce the loop gain by almost a order of magnitude. This should help the stability. However without a bode plot of the loop, this is a guess.
You may have other issues that is causing the power supply to overheat. Check the drain voltage and current waveforms to see if the voltage spike is less than 100 Volts (when the switch turns off). Verify from the current waveform that the transformer is not saturating. A saturating transformer can be determined by the current ramp is curving up (it should be a straight ramp without curvature). Also you may need to increase the heatsink size on the device.

Z. Cochrane

Please find attached the feedback circuitry, note that between 5.5V and 5.3V there is a inductance of 3.3uH.

I attched to two graphic form the voltage in the tranformer primery, the firt one at 120VAC and the second at 250VAC. I think there is no saturation in the transformer, what do you think?

I have a zenner to reduce the spike voltage of 150V. and the Vr is 70V. I think is well desinged.what do you think?

The following step will be increasing the heatsink.

regards

sergio

The waveforms look good. Since the bias for the feedback is only 5.5 Volts, the resistor value is good. Using a larger heatsink will help the thermal shutdown issue.

Z. Cochrane