MC33887 Short-circuit FAULT with higher input voltage

[Tilen]

Hi NXP,

I have your KIT33887EKEVBE dev board kit. I found a problem when testing the current capability of the MC33887 with higher voltages. For load testing I'm using an adjustable electronic load and the power supply used is well bellow any current limit.

With 12V input the MC33887 works as expected, I can increase the load up to 5A before I trigger any FAULT condition.

But when I start to increase the input voltage any higher than 16V I get a FAULT condition with the outputs shut off. At 18V I can't even get 2A to the load.

We would like to use the MC33887 up to 30V/3A. The datasheet states all parameters are specified up to 28V and up to 40V with derating. Is this normal behavior for this IC? Or do we have a faulty IC?

Thank you.

Best Regards,

Tilen

[diazmarin09]

Hello Tilen, 

Please share your results after the IC is replaced. That would be appreciated. 

Regards,

David

[Tilen]

Hello David,

I think I found the problem. It has to do with the FB (feedback) circuitry. On my board there is no FB resistor (that's how it was from the original packaging) and only the FB capacitor is placed. When I removed the FB capacitor everything started working as expected. The high pitched noise I mentioned that was coming from the board is now also gone.

I guess since there is no load resistor on the FB pin that the higher input voltage messed up the internal current source for the FB circuitry?

I have now placed FB cap back on the board and added an 150R load resistor on the FB pin and everything works fine.

BR,

Tilen

[diazmarin09]

Hello Tilen,

I am glad to know that you solved the issue. Thank you for sharing the solution with the community.

The resistor on the FB pin should be placed at board based on the user guide. Thank you for your contribution.

Regards,

David

[Chase121]

Me too have KIT33887EKEVBE dev board kit and i was having this problem couple of months ago and i mail to the community and they solved my problem. UPSers

[diazmarin09]

Hello Tilen,

I hope all is great with you.

The following information would be helpful before I can draw any conclusions about a faulty IC:

1. Are you controlling the load via PWM?
2. Can you share the jumpers configuration?
3. It is possible to share an image of the voltage vs the current?

Regards,

David

[Tilen]

Hi David,

1. No PWM. Just simple ON/OFF control.

2. Jumper configuration in the attached picture.

3. Yellow waveform is input voltage and green waveform is the output current. In this test only one output channel is used (OUT1). The load positive lead is connected to OUT1 and the negative lead goes directly to power supply ground. OUT2 is unused.

As can be seen from the waveform the MC33887 goes to the FAULT state around 25V. The load current is constant at 3A. I got the same result from both OUT1 and OUT2. The bottom side of the board is also attached to cooling fins so the IC does not go beyond maybe 10°C ambient during this short testing cycle. I also noticed an audible noise coming from the board when the input voltage was above ~20V, I'm guessing this is probably the bootstrap supply switching noise coming from the bootstrap MLCC capacitor?

If I reduce the current I can to a higher higher input voltage before getting to the FAULT condition and vice versa.

BR,

Tilen

[diazmarin09]

Hello Tilen,

It looks like the board has a 5V regulator on it. Check the regulator output for the 5V signals as that could go bad as the input voltage is raised. If D1 or D2 or the EN pin does not get the logic level 5V needed.

Also, try testing with no load to see if it still faults at higher input levels. The only thing that would cause a fault other than D1 D2 not being at the correct level is overtemperature, Short-circuit, or undervoltage.

It is possible that IC is bad but from the description you gave I would think the 5V regulator is bad.

Regards,

David

[Tilen]

Hello David,

The 5V regulator is OK. I measured the output during testing and it is at 5V the whole time.

Undervotlage: The power supply output voltage does not dip during testing. Have confirmed this.

Short-Circuit: The load is OK, so no possibility of a short.

Overtemperature: If anything it does look like it goes in to overtemperature condition just based on the time in which it goes to fault. It seems to work a few seconds at say 2.8A (@28V) but then goes to FAULT. But overtemperature is also very unlikely because during testing at say 3A/28V the IC just gets a little warm and nothing more. This can also be confirmed when lowering the voltage to 12V where it can handle 5A with no problem and during this time it gets considerably hotter than in the higher voltage tests and still it does not go into any fault condition.

For some reason is does not working with higher voltages. I will the replay the IC with a new one to see if that helps with anything.

BR,

Tilen