TPS65177: TPS65177 without any output failure case

Gabriel Xu

Intellectual 2145 points

Part Number: TPS65177

Hi teams

My customer CVTE is using TPS65177 to power their TCON board in a new project.

This project has MP and shipped more than 100Kuint in past few months. No failure case reported in the past.

Recently several failure cases are reported from CVTE the phenomenon is: TPS65177 didn't have any output at all.

After they replace the chip with a new one the issue fixed.

I have measured all I/O pins the resistance is normal no short circuit or open circuit are found.

I wonder to know is this device has some kind of latch off protection will lead to this issue?

What should I do to further narrow down the root cause? Thanks in advance.

over 6 years ago

0 Aaron Goodson over 6 years ago

TI__Expert 6330 points

Hi,

I have notified our expert regarding this topic. Please expect a response by 8/2/19.

Thanks,

Aaron

0 Gabriel Xu over 6 years ago in reply to Aaron Goodson

TI__Intellectual 2145 points

Hi Aaron

Do you have any update? As this project has already MP so is quite urgent.

0 Liaqat Khan over 6 years ago

TI__Mastermind 18550 points

Hello Gabriel,

My apology for bit of a delay in response. We are not aware of any latch off or other issues that will cause device failure in the applications. Unfortunately, there is not much information (schematics, layout etc?) in this post to review to see what could possibly cause this the failures. If you can provide more information, we will be happy to review to see if we can spot something marginal in the application itself.

You mentioned that when the the TPS65177 chip is replaced on the board, it becomes good. Has customer tried placing the bad chip on a good board to see if becomes bad? Meaning AB swap instead of just replacing chip on band board? if AB swap is done and bad board becomes good and good board becomes bad, you may consider initiating a QA return back to TI so that our team can do FA analysis on the device.

Kind Regards,

Liaqat

0 Gabriel Xu over 6 years ago in reply to Liaqat Khan

TI__Intellectual 2145 points

Hi Liaqat

Thanks for your comments.

I totally agree with you it's a possible quality issue. I already plan to move forward to FA next step.

But before that, I have to verify this issue isn't an application issue.

I have uploaded the schematic as an attachment. Could you help to review it and give me some advice on what I need to further check to narrow down the root cause?3817.原理图.pdf

0 Shinya Morita over 6 years ago in reply to Gabriel Xu

TI__Mastermind 43885 points

Hi Gabriel,

Could you add this schematic review action to QA return system? Then our team will do total support for this issue analysis. Normally this is way we tackle failure issues.

Thanks,

regards

Shinya

0 Gabriel Xu over 6 years ago in reply to Shinya Morita

TI__Intellectual 2145 points

Hi teams

I have returned the NG units to FA team and the bench test result shows that all function is normal in TI'a lab.

The reason why I didn't do the bench test in the field is that I can't find EVM for this device.

Below is the details update about this case:

During powering up procedure:

BUCK1 output VIO PIN37 have ramped up within 3ms reach 3.3V.
BUCK2 didn't use in the customer system.
VGL didn't have any output. I have measured DRVN PIN21 no switching waveform at all
The BUCK1 also been turn off seems the device is been latch off.

I have several questions:

If BUCK2 isn't used in the system. Do we have to disable this channel through I2C or EEPROM in REG00H?
The customer left the BUCK2 SW node and sensing pin (Vcore & SWBK2) floating. Are this configuration is correct? What's the recommended connection for not use channel?
Customer also connect a 100nF cap between PIN 40 and PIN39 (BK1SW and NC pin) are this matter ?
Are this issue may lead by abnormal EEPROM writing? how to verify about this?

The BUCK1 output waveform is shown below Yellow (VIO) Blue (Vin)

E.GC.TCON.43001B维修原理图 (1).pdf