MP4033 Secondary winding rectifier diode is burning

MP4033. Secondary winding rectifier diode burns after some period of work (~8month). The rectifier diode is used as proposed in datasheet reference schema - MBRS3200 (Schottky 200V/3A). LED strip is 48V, 500mA.
When I started to investigate the issue, oscilloscope measurement showed that diode reverse voltage is quite high, especially when start to dimming the LED. The voltage during dimming, achieves to 300V P-P. When do not dimming, voltage is about 204V P-P.

Question is: How about use fast recovery rectifier diode for higher voltage (lets say 400V/3A, 50ns recovery time), instead of Schottky 200V/3A, that is proposed by reference design? Is it feasible? Or it should be Schottky anyway? Because in all refence schemas you have always Schottky rectifier diode.

Here is scope screenshot when start to dimm with Triac dimmer. The reverse voltage on rectifier diode is 302V P-P.

Hello! Welcome to the MPS Technical Forum!

Can you provide us with more information like input voltage? Would you be able to post your schematic here?

Did this issue happen only on one board?

Hello!
Thanks for the reply.

Input non-dimmable voltage is 230V. The schematic is pretty match the same as reference one in datasheet.

Here are my schematic and scope diagram taken during triac dimming. It shows that first pulse reverse voltage on D8 diode during dimming is much higher (Vpp 306) than the next switching pulses. In addition, next pulses during dimming, are higher than pulses when do not dimming. Not dimming pulses are 204 Vpp, showed on scope diagram on my first post. It looks like using dimmer may cause diode damage.

This issue happens on production items, of course on PCB.

Thanks,
Andriy

Here is scope diagram on D8 reverse voltage during dimming.

Main question is why on reference design is used Schottky diode if it used in boundary conditions and prone to be damaged?
Is it Ok to use fast recovery rectifier diode instead, with higher reverse voltage, 400V, 35ns recovery?

Thanks,
Andriy

Hello Andriy,

Thank you for your patience with this request. I am looking into your questions and will get back to you soon.

One question that I had to understand this issue better, did you build multiple boards with this circuit or was it just one board? If there are multiple boards, how many such failures of the rectifier diodes did you see?

Hello,

There are multiple boards with this issue, about ~30 (this is amount that I aware of, but can be also lamps that customers didn’t send back).

But looking at scope diagram it is obvious why it happens. It because rectifier diode works in boundary conditions.
Does this issue can be fixed by just replacing Schottky diode by fast recovery diode with higher reverse voltage?

BTW, this issue is also reproducible on competitor’s drivers, they are using older version, MP4030 IC.

Thanks,
Andriy