AM2634-Q1: QSPI-D1 pin's resistance decrease to 7k after low temperature test

Hi Will,

Could you tell us a bit more about the test? How low was the temperature the board was being tested at?

Also, could you provide a diagram of the QSPI memory and how it is being connected to the SoC please?

Best,

Daniel

Hi Daniel, 

They tested the board under -40 degree environment. 

As for the QSPI connection, I need to check with customer if they could provide it or not. But from my memory, it is very similar as our CC EVM board. 

Anyway, before it, please help check if it possible that the resistance of pin will be changed under low temperature condition and better can provide the characteristic diagram. Because from on-site debugging, if the pull-down resistance reduce to 5k but not 10k, the failure board can work again even though the voltage level of QSPI-D1 pin is still 1.2V (I think the logic will be 0 because this voltage is lower than logic '0' voltage threshold?). 

Thanks, 

Will 

Hello Daniel, 

Any updates? 

Thanks, 

Will 

Hi Will,

Could you describe exactly how you are measuring the internal resistance of the pull? have you also measured this resistance during reset and normal operation?

Hi Will, 

Also, a few additional requests:

• what is the state of the IO? weak pull up/down or tristated?
• does the issue show at low temp exclusively or show after going through a room temperature temp cycle (room-low-room)?

Best,

Daniel

Hi Daniel, 

we measured the internal resistance of pin to VSS, and removed all peripheral circuits (i.e. external resistance and Flash memory device). And yes, we measured the resistance both power-off, reset and normal operation. 

Regards, 

Will 

Hi Daniel, 

Daniel Bermudez said:

what is the state of the IO? weak pull up/down or tristated?

Tristated, since we removed all peripheral circuits (i.e. external resistance and Flash memory device).

Daniel Bermudez said:

does the issue show at low temp exclusively or show after going through a room temperature temp cycle (room-low-room)?

Also show after going through a room temperature. I was on-site and sure this is the case. 

Regards, 

Will 

Hi Daniel, 

Any updates? Customer is pushing us for the answer, or we will start FA process. 

Thanks, 

Will 

Hi Will, 

And how are you measuring resistance?

Is it by forcing current and measuring voltage or applying voltage and measuring current? Applying voltage and measuring current is always preferred.

Either way what is the applied voltage/current?

Best,

Daniel

Hi Daniel, 

We use a multimeter to measure the pin resistance. From my understanding, the multimeter is applying current on the pin and test the voltage to get the resistance value. Not very sure the applied voltage/current. 

Anyway, the resistance of regular device is 49k with this same measuring method, however the failure device is 7k. 

Regards, 

Will 

Hi Daniel, 

I think we should provide a preliminary answer to customer. Or could you suggest some test need to be done before we can determine the root cause of this problem? Do we have tested the temperature characteristic of pin resistance? If have, please share it as well (if possible).

Thanks,

Will  

This discussion is being continued offline and will be updated once a resolution is found.