TPS2663: TPS26630GER IC is burned out

?? ?

Part Number: TPS2663

Tool/software:

Hi,

We have encountered two incidents where the TPS26630GER IC failed, resulting in a short circuit between pins 1 and 2 (VIN) and pins 17 and 18 (VOUT).
No abnormal surge voltages or inrush currents have been observed on the input or output sides, and all protection features appear to be functioning correctly. However, internal damage to the IC still occurred.
Could you help identify any other possible root causes we should consider?
Additionally, please help verify if there are any potential issues with our circuit design.

1 month ago

0 Quang Bui 1 month ago

Prodigy 20 points

Hi, we have noticed the same issues regarding the TPS26631RGE series: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1538181/tps2663-tps2663-tps26631rge-short-circuit

0 Amrit Jit 1 month ago in reply to Quang Bui

TI__Mastermind 30801 points

Hello,

Welcome to E2E.

Can you check if any transient voltage at input which could violate the abs max voltage?

What's the SHDN enable voltage? Is it 12V as mentioned in net name?

Will recommend you use TVS diode at input to protect against input transient voltage.

Thanks

Amrit

0 ?? ? 1 month ago in reply to Amrit Jit

Prodigy 20 points

Hi,

There is no significant surge voltage detected at the input, and a Soft Start mechanism has been implemented on the input side.

Additionally, the SHDN pin is at 3.3V.

The waveform below shows the measurement result.

0 Amrit Jit 1 month ago in reply to ?? ?

TI__Mastermind 30801 points

Waveform looks good. physical damage happens because of abs max violation. Find the explanation for damage in ICs.

when a maximum limit for either the voltage across, the current through, or the power dissipated in the device is exceeded and causes immediate damage or malfunction, or latent damage resulting in an unpredictable reduction of its lifetime.” Of these conditions, it is the overvoltage that can open unintended current paths such as forward or reverse breakdown of diodes or oxides reaching their breakdown voltage within integrated circuits (ICs). Once an overvoltage opens an unintended current path, the resulting currents can cause damage that includes the melting of silicon; the fusing of metal interconnects; thermal damage to packaging material; and the fusing of bond-wires, leading to electrically induced physical damage (EIPD).

I would advise you to follow below section in datasheet for protection against transients. You can consider putting Schottky at output and TVS at input.

Thanks

Amrit

0 ?? ? 1 month ago in reply to Amrit Jit

Prodigy 20 points

For a DC 48V input and output, what are the recommended specifications for the Schottky diode on the output side and the TVS diode on the input side?

0 Amrit Jit 1 month ago in reply to ?? ?

TI__Mastermind 30801 points

For TVS diode you can use below calculator.

https://dr-download.ti.com/design-tools-simulation/calculation-tool/MD-9eu0p1l7Co/1.0.0/TVS_Recommendation_Tool.zip

For Schottky, B260-13-F which is used in EVM.

Thanks

Amrit

0 ?? ? 1 month ago in reply to Amrit Jit

Prodigy 20 points

Hi,

Based on the designed circuit with an input voltage of DC 48V, please verify whether the parameters shown in the diagram and the recommended TVS specifications are correct.

0 Amrit Jit 1 month ago in reply to ?? ?

TI__Mastermind 30801 points

Icb current you put is 60A. It should be 6A.

Recommended TVS diode looks to be good.

0 ?? ? 1 month ago in reply to Amrit Jit

Prodigy 20 points

Hi,

In the event of a short circuit or overload current, if the TPS26630RGER is shut down via the SHDN pin, will positive and negative voltage transients still occur at the input and output?

0 Amrit Jit 1 month ago in reply to ?? ?

TI__Mastermind 30801 points

yes, FET turn off will create transients.

Both VIN and SHDN will turn off the FET.

Thanks

Amrit

0 ?? ? 26 days ago in reply to Amrit Jit

Prodigy 20 points

Hi,

During the short-circuit test, a transient surge voltage of approximately 76.8V was detected at the input, lasting for about 64 nanoseconds. Would this surge potentially cause damage to the IC?

0 Amrit Jit 25 days ago in reply to ?? ?

TI__Mastermind 30801 points

Hey,

Can you zoom in to the surge instant?

Is VOUT going -ve ?

How far you are doing short circuit from the eFuse?

Thanks

Amrit

0 ?? ? 24 days ago in reply to Amrit Jit

Prodigy 20 points

Hi,

The zoomed-in waveform is shown below.
According to the measurement results, VOUT did not exhibit any negative voltage behavior.
The short-circuit test was conducted very close to the output of the eFuse.

Thanks

0 Amrit Jit 20 days ago in reply to ?? ?

TI__Mastermind 30801 points

Hello,

The spike in input is because of path inductance on input side.

voltage will spike up because of sudden interruption of current and Ldi/dt spike. If this spike goes above abs max value, eFuse will be destroyed.

To limit the voltage surge, TVS+bulk caps are recommended at input.

Thanks

Amrit

Power management

Power management forum

TPS2663: TPS26630GER IC is burned out