Part Number: LM2588
I have a question on an issue I am seeing here. We’ve been using this part on a flyback for quite a while now (on the order of 5 to 10 years) and we have had a rash of failures occurring in S.E. Asia on the secondary rectifying diodes. I believe we’ve determined it is voltage stress on the diodes.
I have determined that there is no margin on the voltage rating of these diodes and they need to be upgraded from 100V to 200V, but I think they are failing when the controller gets unstable when it gets warm.
We have the frequency pin open such that we want it to switch at 100kHz. When we apply some heat to it, the frequency starts to change. The fequency increases to 200kHz. But I think the real problem is when we get it even warmer, it looks like the switch turns on and off right away and then back on right away. Pin 5 of the controller or the collector of the transistor is turning on the first time slowly (slow dv/dt), doesn’t ever fully turn on, and then turns off. The second time it tries to turn on is where it looks like the transistor wants to turn on and the dv/dt is much faster. This frequency is around 5Mhz. And it’s that second turn on that causes excessive voltage spikes on the secondary.
We have been able to duplicate this consistently on multiple boards. The next steps I think I’ll take is to actually look at the primary side current and then bread board the flyback outside our board, but from there, not sure where to go.
The thought on the frequency shifting is that with the frequency pin open, maybe that is an issue. But we are following the datasheet. For the high frequency switching, it seems like the control is not telling the transistor to turn on the first time because it never fully turns on and the dv/dt is slow.
I was wondering if TI has seen an issue like this before or could give any insight on what might be happening.
Any help would be much appreciated.
Thanks,