TPS1663: TPS1663 input short protection. LM5013 failure

Ignasi Saladrigas

Part Number: TPS1663
Other Parts Discussed in Thread: LM5013, , TPS2663, TPS2662

I have been doing short circuits to the power supply first stages of our circuit: A TPS16630 EFUSE at 50V + a LM5013DDAR DC/DC (12V) + a LM5013DDAR (variable output, output by default: 38V).

We want the board to protect itself when a short circuit is applied at the input (this board is part of a modular system: a couple of modules inserted into a backplane board). Doing the short circuit we test the situation in which a module is plugged and working (50V) and then someone add a wrong shorted module to the same 50V supply.

When we have done that:

- The LM5013 at 38V has been destroyed. We observe 'SW' pin, 'input' pin and 'GND' pin internally connected. How can we protect this DC/DC?

- The LM5013 at 12V is ok. When we do the same test several times (4), finally this DC/DC has been destroyed as well. Does it make sense?

- We have monitored the IMON voltage of the EFUSE and we have measured up to 10V (during 5us aprox) with a 5k1 resistor. The documentation says the output of this pin is limited to 5V?

I would appreciate your help.

Many thanks

over 1 year ago

0 Kunal Goel over 1 year ago

TI__Guru* 79181 points

Hi Ignasi,

Is TPS1663 device fine?

Regards

Kunal Goel

0 Ignasi Saladrigas over 1 year ago in reply to Kunal Goel

Prodigy 10 points

Hi Kunal,

Yes, the TPS16630 device seems it's fine.

0 Kunal Goel over 1 year ago in reply to Ignasi Saladrigas

TI__Guru* 79181 points

Hi Ignasi,

I will get back by tomorrow EOD on this.

Regards

Kunal Goel

0 Kunal Goel over 1 year ago in reply to Kunal Goel

TI__Guru* 79181 points

Hi Ignasi,

Please share schematic for review. When you did short at VIN then VOUT was at 50V?

Regards

Kunal Goel

0 Ignasi Saladrigas over 1 year ago in reply to Kunal Goel

Prodigy 10 points

Hi Kunal,

Please found attached an sketch of what we have mounted on the PCB we are using to test this. Please note we have mounted an 820k resistor from pin2 to pin3 of IC19 in order to enable it, as the 3V3 DC/DC is not mounted. The IC31 is not mounted as well (as, again, 3V3 are not there), I have maintained in this schematic for you just for your understanding on how are working both DC/DCs. D3 and D4 are only previsions and are never mounted. R215 is 150kohm and R7 is 3kohm. Finally, in our final BOM we have mounted C168=180pF, but when we had our issues, 56pF was mounted, that is why we have maintained 56pF in this schematic.

Regards,TI_supply_TPS16630_LM5013.pdf

0 Kunal Goel over 1 year ago in reply to Ignasi Saladrigas

TI__Guru* 79181 points

Hi Iganasi,

I will get back by tomorrow EOD on this.

Regards

Kunal Goel

0 Kunal Goel over 1 year ago in reply to Kunal Goel

TI__Guru* 79181 points

Hi Iganasi,

I think because of input short there is lot of current in reverse direction through body diode of eFuse. This can damage eFuse or DC-DC(it may not support high current in reverse direction). Device like TPS2663 with reverse current blocking might be more useful here.

Regards

Kunal Goel

0 Ignasi Saladrigas over 1 year ago in reply to Kunal Goel

Prodigy 10 points

Hi Kunal,

- We are using a board consumption up to 50V/2A, therefore the TPS2662 cannot be used (0.88A). Is there any other way to protect the LM5013?

- What about the IMON voltage measurement I mentioned in my first message. We have monitored the IMON voltage of the EFUSE and we have measured up to 10V (during 5us aprox) with a 5k1 resistor. The documentation says the output of this pin is limited to 5V. Which explanation have yo to that bad behavior?

Best regards,

Ignasi

0 Kunal Goel over 1 year ago in reply to Ignasi Saladrigas

TI__Guru* 79181 points

Hi Ignasi,

Input short is violation of device behavior so device might not work as expected and other functionalities too. I think if current would have flown from VIN to VOUT then you should have seen it limited. Does IMON voltage settles to lower value?

Regards

Kunal Goel

Power management

Power management forum

TPS1663: TPS1663 input short protection. LM5013 failure