Hello all,
I have a DC-DC design using LM5045 with the HS1 pin sinking current to ground. HS2 pin is working properly. I have replaced the IC once and problem remains. Before replacing again I though to ask the comunity if this behavior was seen before. In order to study the problem I have installed a protection resistor of 8R2 in the drain of the high side MOSFET 1 (see schematics). Without this resistor the MOSFET burns out. With this resistor the gate drivers seems all okay. The side 2 works correctly all time. Even disconnecting the resistive load the problem remains. Its is not related to rise/fall time of transistors (that could trigger a shooting-through in some MOSFETs). As seen in the scope photo, the HS1 path seems to be grounded. In fact, the LM5045 gets hot and the voltage ramp on HS1 seems to indicate some internal resistance with thermal behavior. When the circuit is powered off I did use a multimeter to measure resistence between HS1 to ground (and HS2 to ground) and both shows the same value of 4.4Kohm. It does mean the grounding effect is an active response from the internal circuit and not just a short-circuit (internal or in the PCB). Also, both HS1 and HS2 with same resistence value seems there is no observable difference in the device's circuit. The device seems to operate correctly with all timings and responses as expected.
The COMP input is floating because I have removed the transformer and the whole secondary side is meaningless during this study. Anyway, LM5045 with COMP left open just goes to 50% duty cycle and this is okay to study it's behavior. The schematics is a simplified reproduction of the actual one to show what is in operation during this study. Including the not used circuits would only add noise to this post.
It is important to clarify this behavior happens with or without the load. Therefore the current is not passing through the Lower Side MOSFET 2. The current is also not passing through the Lower Side MOSFET 1. It is off (as seen in the picture as signal B and measured in the circuit, just to be sure).
The voltage on the load shows the correct operation when Upper SIde MOSFET 2 is on and Lower Side MOSFET 1 is on. So, even if HS1 is internally grounded, during this phase the Lower Sidfe MOSFET 1 active also grounds HS1 and therefore the issue does not affect operation.
I can replace the device again but wonder if I made some so ridiculous mistake I can't see. The IC was purchased from Digikey about 6 months ago and I believe it is not a counterfeit.
Any clues?
Thank you in advance.
Best regards,
Augusto Einsfeldt
With protection resistor in place the gate voltages seems to be correct.
The protection resistor voltage shows the current passing in the upper MOSFET.
