UCC28950: ucc28950 low side driver damaged

Hello Rob

The UCC28950 OUTx pins are intended to be connected to the input pins (LI or HI) of an external driver exactly in the way you describe. I assume that the UCC28950 and the driver are using the same VDD and Ground. If they were not then it could be possible that the OUTx pin gets driven outside its AbsMax ratings (VDD+0.4V and -0.4V).

Is VDD stable under no load ?

As a diagnostic tool you could try increasing the series resistor from 1 Ohm to 1 kOhm. The LI and HI inputs to the driver are high impedance so there should be little voltage attenuation caused by using 1kOhm but the higher resistance will prevent any high currents that might cause damage.

Finally - it would be worthwhile just to double check that the circuit you are working on is actually the same as your schematic says it is - there may be an open or short circuit that you haven't noticed.

Regards

Colin

VDD is the stable and is supplied by a AUX power. UCC28950 and H-bridge driver use the same VDD and ground.The drive energy is small and should not influnce the Aux power.I use the same drive circuit and same parameters with UCC2895.For UCC 2895,It has no problem.The difference of UCC2895 and UCC28950 is drive capability.The damage occurs when SMP power up in no load.UCC2895 has no burst mode and so duty cycle is relative large for UCC2895 in no load.I never meet such things before.Main IC is seldom damaged in application if design properly. I have a question.I cannot see detail circuits of UCC2895 and UCC28950. Has UCC28950 and diode clamp protection circuits ,which ensure drive port voltage range clamped from VDD and GND .Has UCC2895 's drive port diode clamp protection circuits ?

Hello Rob

The UCC28950 drive circuits are intended to be connected to an external driver in the way you have used them so there is no problem with the loading of these pins. The outputs will have the usual diodes to protect against ESD events.

You can increase the minimum Ton in burst mode by using the resistor at the TMIN pin.

is the OUTx switching pattern correct and stable during burst mode operation before the part fails ?

I would check if there is any unexpected voltage at VDD or OUTx pins during start-up.

Regards
Colin

Hello,Colin
The UCC28950 pcb board is newly assemblied. Before whole SMPS's powerup,I check UCC28950 control board and ensure all two low side driver have proper waveform. Because whole SMP power up time is very short, I am not certain what happen in drive port.
Protection Clamped diodes are not for ESD.They are fast recovery diode which clamp port voltage in safe range if port voltage rings.

Hello Rob
The OUTx pins should never see significant low impedance ringing because they are signal level outputs.
In any case that doesn't solve your problem. I would look for unexpected voltage spikes at startup on all of the pins of the UCC28950 - and on the driver too.

The UCC28950 is a very popular PSFB controller and I have not come across your problem before so I expect that we can identify and correct the issue because I am confident it is not an IC problem.

Can you email your PCB layout to me at colingillmor@ti.com and I will have a look to see if there are any recommendations I can make.

I'm going to be Out Of Office for the next two weeks but if you can get the layout to me in the next couple of hours I should be able to look at them before leaving.

Regards
Colin

Hello colin
I will double check the problem myself first.When UCC28950 's drive port fail,the H bridge driver is OK.I guess high drive capability has more possibility to cause voltage rings when drive a high impendace load because fast drive may cause parasitic inductance to ring.If drive port have no clamp protection, these rings may causes overvoltage and negative voltage which damage IC drive ports