This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AM5706: PCIe Link Down. Ripple Voltage

Part Number: AM5706

Tool/software:

Hi, I am using AM5706 in my project. And we are facing an issue such that the PCIe link is getting down while working. AM5706 processor is acting as end point and Intel atom 3930 based COMe module  (SoM) act as Root complex, in a different board. And the clock is generated from Module and distributed to the AM5706 board with a clock buffer  through backplane. (image attached)

.

The below images are the probe clock images. It is observed clock of same frequency and voltage while working as well as link down condition.

Then we tried probing the PCIe voltage (which is generated by PMIC TPS659163

And the ripple measured to be 68.6mV.

I tried verifying it with AM5706 datasheet, but no details was provided asn asked tho check PCI Express® Base Specification Revision 3.0. but in that document also I was not able to get the required data.

Please help me solving my issue by

1. Please review the PCIe connection in schematic.

2. Please review the clock and Voltage waveform

3. Please provide acceptable ripple range.

Regards,

Athuljith R

  • Is the PCIe link good and then dropping or never established? 

    One comment on the schematic - when comparing to TI's EVMs, the termination resistor (R1506) is on the other side of the DC-blocking caps (C1056, C1057) - I'm not sure if/how much this will impact clock waveform.

    I'm adding another PCIe expert to help with this thread.

  • Please see the inline comments

    Is the PCIe link good and then dropping or never established? 

    [Aj]  PCIe link is getting established and losing at random timing (sometime hours some time 15minutes)

    One comment on the schematic - when comparing to TI's EVMs, the termination resistor (R1506) is on the other side of the DC-blocking caps (C1056, C1057) - I'm not sure if/how much this will impact clock waveform.

    [Aj] Thanks for the observation, we tried changing the position of resistor, but the results are same.