TPS54332: Problem with overcurrent protection in inverting setup

Rafal Lochowski

Part Number: TPS54332

Hi, I'm trying to use the TPS54332 in inverting setup. This is my schematic:

VCC: 14-15v, Vout: -12v @ 1A

It works pretty well - the output is pretty clean (ripple < 80mV), the Ic' temperature is under control (<120C @ 1A load).
Unfortunately it seems, that the overcurrent protection is unreliable - a short-circuit at the output leads to damage of U1 (internal short-circuit of VIN and PH) and D1 (internal short-circuit). The damage occurs randomly, not at every attempt. I tried to use a bigger diode (60V/5A), but it doesn't help at all.

I've got some experience with this Ic in "normal" configuration (positive output) and never had such problems before.

Is there a way to improve the protection of the circuit?

over 3 years ago

0 Zixu Wang over 3 years ago

TI__Expert 8100 points

Hi Rafal,

My colleague Ruby will give you response later!

BRs

Zixu

0 Ruby Li over 3 years ago

TI__Expert 7520 points

Hi Rafal

Here is the case, the input voltage is 15V max, and the output voltage is -12V, as a result, the voltage stress is 27V for the input. When doing the hard short test, there may be voltage spike in the input side which would damaging the device. Would you please do me a favor to capture waveform of the input voltage, inductor current, PH voltage and output voltage?

Thanks.

0 Rafal Lochowski over 3 years ago in reply to Ruby Li

Prodigy 20 points

Hi Ruby and Zixu, thanks for prompt reply!
I also tested the setup under 12v input - unfortunately it acts the same.
Here are the waveforms I managed to capture:

1. Input (CH1) & Output (CH2) during short-circuit

Indeed there's a spike on the input. Do you think I should play with large capacitor in the input? 220uF? Input to GND or Input to Output?

2. PH voltage (CH4) during short-circuit

Unfortunately I don't have a proper equipment to capture the inductor current waveform.

0 Ruby Li over 3 years ago in reply to Rafal Lochowski

TI__Expert 7520 points

Hi Rafal

The voltage spike does harm to the device, adding a large input cap may help here. I would like to confirm here that how long the hard short condition will last? Will it happen in the real application? Do you have the layout file?

Ruby

0 Ruby Li over 3 years ago in reply to Ruby Li

TI__Expert 7520 points

Hi Rafal

I would also like to confirm here, is the real application has input 15V max and output 12V? If so, the spike is a high risk here.

Thanks.

Ruby

0 Rafal Lochowski over 3 years ago in reply to Ruby Li

Prodigy 20 points

Hi Ruby,
Actually I can't be sure how long the short might last - the real application is a consumer device with exposed pins.
Please find the layout below - it's a little more complex, so I added the schematics too. FBE2 & FBE3 are shorted.

0 Rafal Lochowski over 3 years ago in reply to Ruby Li

Prodigy 20 points

I can lower the input voltage to 14V, but there is also +12V adapter in the system, so can't go lower. The safety margin is still low - just 2 volts...
Or I can add another regulator reducing the input to 5v, before the -12v stage.

What U think?

Thanks
Rafal

+1 Ruby Li over 3 years ago in reply to Rafal Lochowski

TI__Expert 7520 points

Hi Rafal

Reducing the input is a good idea from my side, just give margin to this device.

Ruby

0 Rafal Lochowski over 3 years ago in reply to Ruby Li

Prodigy 20 points

OK, thanks!
BR

Rafal

Power management

Power management forum

TPS54332: Problem with overcurrent protection in inverting setup