• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 101 subscribers
  • Views 180 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

AM4378: VDD_MPU_MON not working as PMIC feedback

Berkay Somali
Prodigy 170 points
Part Number: AM4378
Other Parts Discussed in Thread: TPS65218D0, AM3358,

Hi, 

We have a custom board with AM4378 processor and a TPS65218D0 as our PMIC. In a previous design with AM3358 processor, we had tied the DCDC2 FB line to the VDD_MPU_MON pin of the processor and it worked fine, all the rails of PMIC were producing correct voltages. 

However with our new design with AM4378, when DCDC2 FB is tied to VDD_MPU_MON PMIC shuts itself down due to overvoltage. However, when we tie the FB to the output of DCDC2 directly it works fine. See below image for when its tied to VDD_MPU_MON.. Purple line is the MPU voltage and green is the current. 

So we started to believe that the VDD_MPU_MON is not working as it should in AM4378 processor in our batch or in general. We would like to know if you can populate this problem in your environment. 

Also, we need to know if this situation may have caused any damage to hardware (MPU, DDR, etc.) because we are having some issues with DDR on these boards now after the voltage issues has been resolved with direct connection of FB to DCDC2 output. 

Thank you in advance for your quick responses. 

  

  • 0
    TI__Guru*** 136978 points

    Hello Berkay Somali, 

    Thank you for the query and inputs.

    Let me check internally. I have a follow up question 

    However, when we tie the FB to the output of DCDC2 directly it works fine. How did you terminate the  VDD_MPU_MON ?

    Also, we need to know if this situation may have caused any damage to hardware (MPU, DDR, etc.) because we are having some issues with DDR on these boards now after the voltage issues has been resolved with direct connection of FB to DCDC2 output. 

    Can i assume that you have some boards that you modified before powering and they are working fine.

    Do you see any difference in behaviour like current drawn or startup between these boards ?

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    Prodigy 170 points

    Hi Sreenivasa, 

    Please see schematic pictures below. Originally we had a jumper at R50 pads and R44 jumper was not placed, tying VDD_MPU_MON directly to FB2. This setup has worked in our previous board with AM3358 but resulted in DCDC2 rail overshoot voltage with AM4378. We have then replaced the jumpers as seen in below pictures and can read the correct MPU voltage now.  

              

    Unfortunately, all the boards have been tried to be started-up at least once in the first situation, so they were modified afterwards. After modification the processors "seemed to be" working fine but we are having problems starting Linux, DDR is failing at memorytest etc. So, we are suspicious about any possible damage during initial firing. We are now in the process of replacing processor, DRAM and PMIC at the same time to see if they will behave different but of course something else may happen during replacement too. 

  • 0 in reply to Berkay Somali
    TI__Guru*** 136978 points

    Hello Berkay Somali, 

    Thank you for the inputs and noted.

    I did check internally and i understand that the VDD_MPU_MON performance was not consistent over different devices.

    Is this the first design you have based on AM4378 ?

    Hopefully the device performance should be better after replacement. Please update and based on the results we can discuss net seps.

    Have a great day and a good weekend.

    Regards,

    Sreenivasa

  • 0 in reply to Kallikuppa Sreenivasa
    Prodigy 170 points

    Hi Sreenivasa, 

    Yes its our first design based on AM4378. With further research on this we found out this issue was noted in the Errata as well. 

    I wish we had seen this earlier, this costed us our prototype boards and so much time is lost. 

    We think this has damaged the MPU and DDR of our boards because they are not working consistently.

    I believe TI needs to revise the product datasheets on such important things. 

  • 0 in reply to Berkay Somali
    TI__Guru*** 136978 points

    Hello Berkay Somali, 

    Thank you for the inputs and understand your concern.

    Your feedback is noted.  We will internally review your suggestions for improvement.

    Regards,

    Sreenivasa