Popular Discussion
- 6 comments
- 1 comment
Attached Images
Legend
Community Member
Power Integrations Engineer Power Integrations ModeratorTOPSwitch oscillating and output voltage too high
For the first time I have problems with the TOPSwitch HX, I am developing a 48Vdc UPS, I just can not stabilize the power. The tension is very high oscillating around 93 Volts.
R8 and R11 burn in a few seconds;
Do not pay attention about the maximum voltage of the TL431 and added a 33V Zener R8 before more tension remains very high.
I noticed that the tension Bias pins 4 and 5 is high = 35 volts;
When I change the resistors have to change the TL431 and opto who also burn ...
Remembering that I am not using any of the circuits completed by solving the food problem.
Please I need help very urgent, as I have time to deliver the prototype;
Thank you.
Alan Castelano
Gentlemen,
When removed the programming resistors and current-limiting power of TOPSwitch the tension usually raised, but I have a problem, use a second card with 3 LM2576 regulators for DC outputs, when I turn on the power supply with voltage regulator is connected to oscillating between 5 and 23Vdc, when I disconnect the voltage regulators normalized in 57.6 Vdc.
When I turn on the power supply without the regulators and then connect again the voltage is normal and all the power supply works normally, I think this related to the initial transient activation over how to solve this problem? The problem is solved by the programming current and power line?
Thanks for the help.
Alan Castelano | 12 posts Posted: Feb 9, 2012 | 07:54 AM In response to Oscillating when connected with a small initial load.
Change R8 to something like 10 k. I think that the present value results in too much gain which can lead to unstable power supply.
PI-Spock | 695 posts Hi
Schematics hard to read - would be helpful for a higher resolution version but guessing R8 and R13 are the clamp resistors. If this is the case then check that the transformer winding phasing is correct. If the clamp is dissipating too much power then either leakage inductance is too high or secondary trace inductance (loop area) is too big.
Run converter at very light load and capture the drain voltage and current waveforms at both say 20ms/div and 5us div ie I want to see both individual switching cycles and wider to check to see if the part is in auto-restart (periodically hiccuping). Hiccuping indicates that the part is not getting feedback which it should be with the output voltage above normal. Check that the bias winding return is returned (the PCB trace is connected) from the transformer pin the to source pin/ input capacitor negative. Also measure the bias winding capacitor voltage - must be >8V
Report back and we'll go on finding out what then problems are.
I will hand over to a collegue as I'm without Internet access for the next week.
Cheers
Peter
Cheers PI-Chekov
PI-Chekov | 174 posts I am no expert on this. But since you are in a hurry...offering some troubleshooting ideas. I do not have time to go & look for the 274HK chip & the pix is not clear so I do not know what pins 4 & 5 are. Still....
#1Remove all components that do not belong to the basic converter: these are all beyond the output filter. Put a resistive load of value min/max power. Is the converter stable?
Can it be stabilized by varying TL431 compensation?
#2 If so, then add only those components after the filter that add one functional attribute at a time. Maybe on such component will kick in the oscillations....then you know culprit. See if you can go from here on.
good luck.
-bougnoul
bougnoul | 3 posts 