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.

TPS659037: Power rail short on AM5718 based board

Part Number: TPS659037

Hi All,

We are using AM5718 processor with PMICTPS6590379ZWSR (OTP: 0X97). We first observed the volatage rail VDDA_PCIE was a short. But since other voltages were proper the DSP was booting up. We continued the operation without rectifying the issue. After 2 days another rails(VDDA_USB, VDDASATA) started showing short. but the operation was continued as our application didnt require SATA and USB. But, now the 3.3 (VDDSHVx) volt is also showing a short. The short is on DSP side as we isolated the PMIC and DSP, Also, the powergood of PMIC is asserted.

What is the reason of such erratic behaviour? Please give us a probable reason and solution if any.

Thanks,

Rajneesh

  • Hi Rajneesh,

    Are you seeing this issue on only one device/boards or you are seeing this on multiple devices/boards? If issue is seen only one device, i would request you to replace this device with another device and confirm that issue is not related to board and it is related to this device. If you confirm this, please contact local TI field team to start the failure analysis process from TI side. Also, send us the PMIC schematic information so that we can review it from our side to see whether these is any application related issues with the schematics. If you are not confortable in sharing your schematics through E2E, please send your schematics through a direct message.

    Regards,

    Murthy

  • Hi Murthy,

     In my schematic the vdda_csi rail is powered by LDOLN and is ramped along with vdda_per,vdda_ddr etc.(as shown in fig. 5.1 of SPRS957D)  But, as per the latest revision of datasheet, vdda_csi should be powered from LDO3 and ramped along with vdda_usb1, vdda_usb2 etc.

    Will it be okay to ramp the vdda_csi first i.e, with vdda_per,vdda_ddr?

    Regards,

    Rajneesh

  • Hi Murthy,

    I am sharing the PMIC and the DSP's power section of my schematic. Please review the schematic and provide your valuable comments.

    TI_SCH_REVIEW.pdf

    Regards,

    Rajneesh

  • Rajneesh,

    I can review your schematic. Can you confirm which TPS659037 device you are using? (The full part number)

    Thanks,

    Nastasha

  • Rajneesh,

    You can disregard the previous message. I see from your original question you are using TPS6590379. The schematic looks okay. Can you verify if this happens on one board or multiple boards?

    Also, POWERGOOD only monitors the SMPS rails, not the LDO rails. By default it only monitors SMPS12, which in this case is the MPU rail.

    Do you have I2c access? If so, can you check is a short is detected on the PMIC interrupt pin? If so, which rails are flagged? Can you read to clear the bit? Does the rail turn back on or stay off?

    Thanks,
    Nastasha

  • Hi Natasha, Thanks for the reply. We have just one board with us right now,. So can’t verify if it is problem with just this board. Actually after isolating the PMIC from the DSP, by removing the feedthrough capacitors, the PMIC generates all the rails. Currently , I dont have access to I2C. We are planning to replace the DSP. Please confirm that the ramping of vdda_csi (powered by LDO_LN) along with vdda_ddr, vdda_debug etc is not an issue. Thanks Rajneesh
  • Rajneesh,

    LDO_LN actually turns on before VDDA_DDR. See the power sequence of the PMIC in the user guide: http://www.ti.com/lit/pdf/sliu011

    Thanks,

    Nastasha