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.

TPS65220: PMIC Device failing to initiate power

Part Number: TPS65220

Tool/software:

Hi,

We are using the TPS6522053RHBR IC in our board with 3.3V input. 

We have referred to the AM64xx eval board for schematics.

But the PMIC is not powering up on our board. We have checked the status of the few control pins as are as below:

VDD1V8 - 1.8V

VSEL_SD/VSEL_DDR - 3.3V

MODE/RESET - 3.3V

MODE/STBY - 3.3V

nRSTOUT - 0V

nINT - 3.3V

But we are seeing none of the outputs are coming up. Please find the schematics page attached.

PMIC-Page.pdf

Please help us with the debug points if any.

Thanks & regards,

Nanjunda M

  • Hi,

    Thanks for reaching out. Here are a few observations on the schematic:

    • TPS6522053 has the Bucks configured to operate in Quasi-fixed frequency and high bandwidth. This configuration requires a minimum of 30uF output capacitance (after de-rating). 47uF output capacitance per Buck is typically used. 

    • If the AM64 RESETSTATz pin is driving the PMIC RESET pin, then the pull-up resistor is not needed. R33 should be uninstalled. The processor starter kit has an external pull-down resistor to keep the pin low during power-up and until the processor is out of reset. 

    • Could you share a power-up scope capture with a ~4ms time scale showing the following signals:
      • VSYS
      • BUCK2
      • LDO3
      • BUCK3

    Thanks,

    Brenda

  • Hi Brenda,

    Thanks for the response.

    1. Is the capacitor requirement applicable to all the 3 buck converters?

    In our case, we have provision for Buck1 - 31uF, Buck2 - 4uF, Buck3 - 19uF as of now. We can increase the Buck1 & Buck3 to 47uF+. But there is no provision for Buck2. Is that ok?

    2. We checked the PMIC with RESET pin pulled low as well. There is no improvement in the situation.

    3. We have gone through, and the configuration matches the requirements. 

    4.  Please find the image (ch1- Vsys-3.3V, Ch2 - Buck2, Ch3- LDO3, Ch4-Buck3)

    Please help us debug the issue further ASAP.

    Thanks & Regartds,

    Nanjunda M

  • Hi,

    Yes, the same output capacitance requirement is applicable to the three Bucks. They all require a minimum of 30uF (after de-rating). 47uF is typical.

    Could you isolate the PMIC from the processor and any load?

    Thanks,

    Brenda

  • Hi Brenda,

    Thanks for the clarification. We will try with parallelling the capacitors for this.

    We cannot isolate the PMIC from load since most of them are directly connected. 

    Regards,

    Nanjunda M

  • Hi,

    We will keep this E2E on hold while you fix the output capacitance and collect the new waveform. 

    I strongly recommend isolating the PMIC even it is requires uninstalling the processor and peripherals in one PCB. 

    Thanks,

    Brenda

  • Hi Brenda,

    We added the capacitors to meet the 47uF output capacitance requirement for Buck converters.

    But there is no change in the PMIC behavior. Please find the image (ch1- Vsys-3.3V, Ch2 - Buck2, Ch3- LDO3, Ch4-Buck3)

    Meanwhile we are trying to assembly a PCB with independent PMIC IC & related supply. We will let you know once we have the assembled board.

    Please let us know if anything else can be checked.

    Thanks & regards,

    Nanjunda M

  • Hi,

    In the meantime, could you share the layout of the PMIC?

    Thanks,

    Brenda

  • Hi Brenda,

    Please find the layout images.

    Top:

    GND1

    SIg1

    GND2

    PWR1

    PWR2

    GND3

    SIG2

    GND4

    Bottom

    Please let me know if any clarifications.

    Regards,

    Nanjunda M

  • Could you share the PMIC layout file by private E2E message?

  • Hi Brenda,

    We have 2 observations.

    1. In the new board where we mounted only the power circuit and no load

    - We checked the output without the capacitors (min 47uF)- output is 0 on all of the PMIC outputs

    - After we added additional capacitors, we observed proper voltage outputs

    2. In the old board, without isolating the load (processor and other components), we just increased the output capacitors on bucks to around 70uF.

    But still, we did not see any output voltage.

    Can you please let us know how to proceed further.


    Thanks & Regards,

    Nanjunda M

  • Hi Nanjunda,

    We will review the information and provide an update within the next 24-48Hrs.

    Thanks,

    Brenda

  • Hi Brenda,

    Any update on this?

    Regards,

    Nanjunda M

  • Hi Nanjunda,

    Thanks for your patience. We are reviewing the layout and expecting to provide feedback within the next 24Hrs. 

    Thanks,

    Brenda

  • Hi Nanjunda,

    Here are the main layout mismatch that we identified. I recommend referring to the layout guidelines (section 7.4) in the datasheet. Let us know if there are any questions. 

    • The VSYS pin does not have an input cap. 
    • The FB_Bx pins were not routed as a trace from the local output cap to the corresponding FB_Bx pin as documented in the PMIC datasheet.
    • The Buck2 and Buck3 output capacitors were placed close to the FB_Bx pin instead of the inductor.
    • The AGND pin should not be connected to the thermal pad of the PMIC. This VIA on the AGND pin must be connected to the internal ground plane   of the PCB. 

    Thanks,

    Brenda 

  • Hi Brenda,

    Thanks for the review.

    Is there any workaround for this?

    Thanks & Regards,

    Nanjunda M

  • Hi Nanjunda,

    Layout errors typically require a new PCB revision. This will still be needed not only to fix the items highlighted in the previous message but also to add the 47uF output capacitance per Buck that was highlighted during the schematic review.

    1. In the new board where we mounted only the power circuit and no load

    - We checked the output without the capacitors (min 47uF)- output is 0 on all of the PMIC outputs

    - After we added additional capacitors, we observed proper voltage outputs

    What are the differences between the new and old boards? Based on this message, even with the non-optimized layout you are still able to power-up the PMIC with the correct voltages after adding the 47uF output caps are added. Am I understanding this correctly? If the issue happens after connecting the load/processor then I recommend having a full design review from the processor team so they can review the connections to the processor and peripherals. 

    Thanks,

    Brenda

  • Hi Brenda,

    I understand we need to make new version of the PCB, but that takes some time.

    W.r.t other board, which is working, we have not put any loads on the output. It is having only the input supply and TPS65220 and required circuits (inductors, caps, resistors). 

    It does not contain the processor or any other related circuits on the board.

    What could be the reason for this?

    Regards,

    Nanjunda M

  • Hi Nanjunda,

    What could be the reason for this?

    Could you clarify this question? It seems like you have a working PCB with the PMIC only (even with non-optimized layout) and the issue happens when the processor is connected. What would be your question?

    Here is an FAQ from the processor team related to design reviews:

    Let us know if there are any questions.

    Thanks,

    Brenda