• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 62 subscribers
  • Views 407 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS650864: Questions about MPSoC power consumption

Geonwoo Ban
Intellectual 270 points
Part Number: TPS650864
Other Parts Discussed in Thread: INA226

I have a problem using MPSoC with TPS6508641 and I ask questions.
There are two products that operate normally, and for bad board, the current consumption appears abnormally high.

For debugging, I removed the power sequence output sequentially.
Initially, after removing the output of the BUCK1, I measured the consumption current to verify the difference between the normal and bad board, both of which were the same.
Next, after removing the output of BUCK2, I measured the consumption current, and the results were similarly the same.
And then, after removing the output of the BUCK5, I measured the consumption current and the current of the bad board was confirmed to have risen rapidly.

The problem here is, I don't know if the MPSoC's power block using 0.85V as the power source is the problem or PMIC's BUCK2 0.85V output is the problem.
(0.85V is not used other than MPSoC supply power.)

I would appreciate your help.
Thank you.

  • 0
    TI__Genius 13216 points

    Hi,

    Please clarify the issue. When Buck5 is disconnected from the MPSoC, the current consumption of the entire board rises? Also if Buck5 is where you are seeing the issue please create a current loop for the output current of the inductor and include scope shots of the current.

    Thanks,

    Daniel W

  • 0 in reply to Daniel W
    Intellectual 270 points

    Hi,

    As mentioned above, the current consumption is the same compared to the normal board until Buck5 is removed. And when Buck5 is removed, the normal board is measured at 24V/0.17A and the bad board is measured at 24V/0.28A. 

    For normal board, there is little difference in current consumption when Buck2 and Buck5 are removed, respectively. In fact, the Xilinx Vivado Implementation Report shows the estimated current consumption value for each MPSoC core voltage, and the current consumption value for the problematic 0.85V is very small.

    However, for bad boards, I think there is a problem with using that 0.85V as MPSoC core voltage because the current consumption increases when Buck5 is removed. But there is no problem with the operation of the product. I wonder why only the current consumed is measured highly. And I wonder if the cause of the problem was MPSoC or PMIC.

    From a PMIC standpoint, I'd like to hear advice that can help me analyze the cause.

    Thank you.

  • 0 in reply to Geonwoo Ban
    TI__Genius 13216 points

    Hi,

    Please share a schematic of the pmic portion of your design. If this needs to be done privately please send me a friend request.

    Also, in order to localize this problem, it may be necessary to measure the current in each power source to understand where the current is increasing when you remove buck 5.

    Thanks,

    Daniel W

  • 0 in reply to Daniel W
    Intellectual 270 points

    Hi,

    Attach the Ultra96 V1 schematic. Our board produced the same.
    And the current of each power rail cannot be measured. Each Buck did not connect parts such as the INA226, and even if I measure it in another way, I think the measurement values are relative because each person has different uses and the areas that use each power rail source will be different. I think it's meaningless because it can't be an absolute criterion.

    That's why I compared the current difference between the normal and bad board. On a normal board, the current when the Buck5 was removed did not differ much from when Buck2 was removed because the 0.85V load current is very small, whereas a bad board suspected Buck2 output due to the increased current.

    Thank you.

    Ultra96 V1 Schematic.pdf

  • 0 in reply to Geonwoo Ban
    TI__Genius 13216 points

    Hi,

    This seems most likely an issue with the system external to the pmic. To localize it you could try turning each buck on by itself and seeing which one is pulling different amount of current on the good and bad boards. Then if this is successful evaluate the subsystems the buck is powering and any possible power faults.

    If this is not successful, one way to create a loop to measure current is to stand an inductor up on one end and attach a wire from the high side of the buck to the other pad for the buck

    Thanks,

    Daniel W