TXS0108E: the resistance between B7 (pin13) and GND is 11.1129Ω, which is close to short circuit

user5707026

Part Number: TXS0108E

The customer's board was originally running normally. After restarting, it was found that the card could not be found. After analyzing and testing, we found that the level conversion chip (TXS0108ERGYR) pin is short-circuited. Our test result shows that the resistance between B7 (pin13) and GND is 11.1129Ω, which is close to short circuit
There is an urgent need to understand the mechanism of failure, because this is very important, please try to analyze.
Have you encountered this similar situation before?
Have experience to learn from it, to understand why there is a short circuit between the pin and the ground pin.

Thanks

Elsa

over 5 years ago

0 Dylan Hubbard over 5 years ago

TI__Mastermind 44340 points

Hi,

I won't be able to analyze with the limited info provided. Could you provide more details like a schematic and scope captures from your testing?

0 Nessee LI over 5 years ago in reply to Dylan Hubbard

Prodigy 70 points

Hi, Dylan

Sorry we do not have any scopes , but if you have any advice on figuring this out by using wavescope pls tell me ,and I may use the broken IC do some experiment.

Below is the schematic of this chip on our board. Thank you for reply!

0 Dylan Hubbard over 5 years ago in reply to Nessee LI

TI__Mastermind 44340 points

Hi Nessee,

What all is connected to B7? Heavy loads can cause issues with this device.

0 Nessee LI over 5 years ago in reply to Dylan Hubbard

Prodigy 70 points

Hi, Dylan

This is the topology of our reset system. Pin B7 is driven by MCU or PCIe slot through a logic gate.

1. I think it's ok because we use MCU GPIO as reset pin in push-pull mode, and other chips connected to MCU reset work normally.

2. I suppose this situation with a strong logic gate cannot be called heavy loads . The continuous output current is ±50mA，through power to GND is ±100mA.

3. What kind of issues may occur with heavy loads?

Thanks a lot~

0 Dylan Hubbard over 5 years ago in reply to Nessee LI

TI__Mastermind 44340 points

Hi Nessee,

This short can be caused by an EOS event. My question is invalid since the B7 pin is only used with unidirectional signals as an input. Knowing that, we need to isolate the signals going into that pin as it could be an indicator. LVC has strong drivers which can lead to pretty serious overshoot/undershoot problems. I would use an oscilloscope to see if the signal on B7 (output of the SN74LVC1G08DBVR) has this issue.

0 user5707026 over 4 years ago in reply to Dylan Hubbard

Genius 3145 points

Dears

After 15 Day, Any feedback here?

If any update?

Thanks

Elsa Duan

0 Dylan Hubbard over 4 years ago in reply to user5707026

TI__Mastermind 44340 points

Hi Elsa,

I haven't heard anything back from my last response so I assume they found the cause with my recommended testing.

0 user5707026 over 4 years ago in reply to Dylan Hubbard

Genius 3145 points

Hi Dylan

Can you help share the email for us? Or we can use Dylan.Hubbard@ti.com?
Our customer want to contact with you immediately.
They are the Key Mass Market customer, and will be the Assigned Account customer.

Thanks.

Elsa Duan

0 Nessee LI over 4 years ago in reply to Dylan Hubbard

Prodigy 70 points

Hi, Dylan

Thx for helping. I wonder if there is anything we can do to push this problem going on , or we may do some experiment on the broken IC . Thanks!

Nessee

0 Dylan Hubbard over 4 years ago in reply to Nessee LI

TI__Mastermind 44340 points

Nessee,

You can always start an FA so we can get the failing parts and do additional testing to see where the damage may have occurred.

Logic

Logic forum

TXS0108E: the resistance between B7 (pin13) and GND is 11.1129Ω, which is close to short circuit