Forum Discussion

sbj's avatar
sbj
Icon for New Contributor rankNew Contributor
3 months ago

Quartus and power domain

We actually one remaining pending issue with the Agilex chip that we can't get resolved and might need some help.    The basic description is that when comparing the power consumption of the Agilex...
Power
Programmable Logic Device
sbj's avatar
sbj
Icon for New Contributor rankNew Contributor
2 months ago

Thank you Farabi for your reponse.

Please allow me some time to provide you a feedback :)

Cheers

Farabi_Altera's avatar
Farabi_Altera
Icon for Regular Contributor rankRegular Contributor
2 months ago

Hello, 

 

Noted. I will wait for your reply. 

 

regards,

Farabi

  • sbj's avatar
    sbj
    Icon for New Contributor rankNew Contributor
    2 months ago

    Hi Farabi,

    here are the answers I received to your questions



    1-can you try power up with slower ramp time? (within timing spec)


    It does not help.

    2- how many boards manufactured? All showing same behavior?

    The behavior is reproducible on all three of our boards. I also observe very similar behavior on the Avnet evaluation board (with a slightly different package). 
    It appears to be temperature-dependent Vcore leakage.

    3- Is it possible to swap unit on the board with another one? 

     to see if the issue follows unit or follows board
    The issue tracks temperature. Internal Agilex temperature sensors show a direct correlation between core rail current and die temperature.

    4- Please send snapshot of the device top marking so I can analyse the production test history. 

    no answer

     

    Cheers

    • Farabi_Altera's avatar
      Farabi_Altera
      Icon for Regular Contributor rankRegular Contributor
      2 months ago

      Hello, 

       

      While waiting the snapshot of topmarking, can you below:

      1- get measurement (resistance value) between VCC to GND (with the FPGA mounted)

      2- All 3 boards showing the same behavior - suspect board design issue. Is it possible to unmount the FPGA and do step 1 above wihout the unit? 

       

      regards,

      Farabi

      • sbj's avatar
        sbj
        Icon for New Contributor rankNew Contributor
        2 months ago

        Dear Farabi,

         

        1) snapshot of the device top marking so I can analyse the production test history. 

         

        See attachement

         

        2) All 3 boards showing the same behavior - suspect board design issue. Is it possible to unmount the FPGA and do step 1 above wihout the unit? 

         

        the Arrow AXE5-Eagle shows a similar temperature/current dependency.

         

        3) get measurement (resistance value) between VCC to GND (with the FPGA mounted)

         

        Do you really need the Vcc-to-Gnd resistance of a board that works enough to let us configure its logic and boot the internal processor?

        No shorts on the board.

         

         

         

        Cheers