Dear AC-DC Forum,
I'm helping to troubleshoot a design using the ViPER16HN as a non-isolated buck. It's part of a power supply that uses PFC, so the input voltage goes from nearly zero to about 400V peak each AC line half-cycle. When used at 277VACrms, 60 Hz, nearly every unit fails after about one to four weeks in operation.
In this image, Ch.1 is the input voltage (seen at the drain pins of the VIPER). Ch.2 is the output voltage, which is about 11VDC. Ch.3 is the GND pin of the ViPER, with respect to system ground (negatiive pin of a diode bridge). Ch.4 is a differential measurement from the GND pin to the VDD pin. The VDD voltage is cycling back and forth between its operating threshold and its UVLO. The output voltage is fairly steady because the maximum output current is only 16 mA, and a 47 uF capacitor sustains it.
I don't think that the ViPER was designed to have its VDD always cycling back and forth as shown. Could this cause local thermal stress to the die?
When the ViPER fails, the recirculating diode fails as well. It's a 1A, 1000V US1M that should be very robust.
Many thanks,
Chris
