LM5175 Our application circuit is mostly lifted from the application in the data sheet. In this application, a FET transistor switches power from the rest of the system to the regulator. The fault condition appears to be related to having a very fast turn on of the power source to 28VDC. As yet, we have not been able to capture an event when instrumented likely due to the effect of probes slowing the rise time. Our system applies 22 to 32 VDC to the DC-DC system and when the fault occurs, the power draw is very high for a sustained time until something in the system breaks. If the lab power supply is large enough, the system fuse will blow. If not, then the lab supply will limit current by dropping voltage destroying the FET that switches power to the DC-DC converter.
For this fault to occur, something in the LM5175 logic would have the enable the wrong FETs establishing either a direct path from VIN to return. Is there internal logic reset circuits that could be tripped up by a fast rising VIN?
