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.

AM335x DS0 Power Consumption Issue on the GPEVM 1.5A

Hi Team,

 Nowadays a customer report the power consumption issue with our suspend mode. They find over 40mW but not 5mW on the product board, SoC power consumption only. So we redid the power consumption test on the GPEVM 1.5A (DDR3). Test result is similar with his production board. 

  On the GPEVM 1.5A, we can capture the voltage of the shut resistor to calculate the SoC power consumption. The test result is as below:

 

Comparing with the power consumption we offered on the wiki,

the vdds_ddr and vddshv6 are the places where the huge numbers come from. 

The software I used is Linux SDK 6.0, as customer want to use the 3.2 version kernel. and from the software side, it went to the suspend mode successfully. 

My questions would be: why there's the huge numbers on the VDDS_DDR, and VDDSHv6? What part could lead to the larger numbers? 

If any test or modification needed, please don't hesitate to let me know! Many thanks!

  • Hi Steven,

    Do they have Vtt termination on their DDR? Do they turn off the Vtt regulator in suspend?
  • Hi Biser,

     The test result above is based on our GPEVM 1.5A, but not customer board.

  • BTW, my cusotmers didn't use the VTT in their DDR3 design. But they got a big number as well.
  • These figures are not consistent with what's reported on the wiki page processors.wiki.ti.com/.../Processor_SDK_Linux_Kernel_Performance_Guide Can you measure on the EVM with a newer kernel to see if this is a software issue? By the way, kernel v3.2 is no longer maintained and is not supported for new designs.
  • Capturing Mike E's response

    Can you confirm the SDK image is for EZSDK 6.0, and is a fresh install?  Do you have access to another EVM to verify the problem is not hardware related?  It would be useful to see your full kernel logs, as well as the console output after resume.  Your high VDDSHV6 number possibly indicates a driver is blocking suspend.

     

    I recently characterized all AM335x SDK versions, from SDK6.0, to most recent, and did not see the issue you’re having (my EVM is rev 1.6A, however).

     

    Here is my measurement data:

    EZSDK 6.0 - suspend

    Power Supply

    Shunt [Ohm]

    Voltage Drop [V]

    Voltage [V]

    Current [mA]

    Power [mW]

    VDD_CORE

    0.05

    0.0000816

    1.024

    1.63

    1.67

    VDD_MPU

    0.05

    0.0000142

    0.933

    0.28

    0.27

    VDDS_RTC

    2

    0.0000437

    1.804

    0.02

    0.04

    VDDS_DDR

    0.24

    0.0000056

    1.504

    0.02

    0.03

    VDDS

    0.24

    0.0000558

    1.809

    0.23

    0.42

    VDDS_SRAM_CORE_BG

    2

    0.0016006

    1.815

    0.80

    1.45

    VDDS_SRAM_MPU_BB

    2

    0.0000136

    1.815

    0.01

    0.01

    VDDS_PLL_DDR

    2

    0.0000015

    1.815

    0.00

    0.00

    VDDS_PLL_CORE_LCD

    2

    0.0000019

    1.815

    0.00

    0.00

    VDDS_PLL_MPU

    2

    0.0000015

    1.815

    0.00

    0.00

    VDDS_OSC

    2

    0.0000015

    1.815

    0.00

    0.00

    VDDA1P8V_USB0/1

    1

    0.0000027

    1.823

    0.00

    0.00

    VDDA3P3V_USB0/1

    2

    0.0000411

    3.320

    0.02

    0.07

    VDDA_ADC

    1

    0.0004570

    1.827

    0.46

    0.83

    VDDSHV1

    0.24

    0.0000190

    3.301

    0.08

    0.26

    VDDSHV2

    0.24

    0.0000048

    3.320

    0.02

    0.07

    VDDSHV3

    0.24

    0.0000082

    3.301

    0.03

    0.11

    VDDSHV4

    0.24

    0.0000056

    3.320

    0.02

    0.08

    VDDSHV5

    0.24

    0.0000163

    3.301

    0.07

    0.22

    VDDSHV6

    0.24

    0.0001032

    3.301

    0.43

    1.42

    Total

    6.9697

  • Hi Mukul & Mike:

      Highly appreciate for your help on this case!

      Sorry for the late response due I was tied to customer issue support yesterday.

      Today I reproduced this issue on  another GPEVM, 1.6A. My latest result is as below:

      I measured the VDDS_DDR and VDDSHv6 power rails, which contribute a lot to the final big numbers.

    1. For the VDDSHv6 power rails, the result is the same.
    2. For the VDDS_DDR, I measure the J41 connector, which marked VDDS_DDR, the result is the same. But for J20 connector, which marked AM335x_VDDS_DDR, it is almost Zero.

    I’m curious of the J41 connector, which marked VDDS_DDR, what’s this power rail used for?

    I try to search the description of the J41 connector on the wiki, but failed as wiki has some problem today, couldn’t be accessed.

      In addition, attached is the full log of booting and suspend process.

      Could you help to let me know your comments on this? Many thanks!

     

    Test result highlight in my test:

     

    resistor

    drop Voltage
    (mV)

    current
    (mA)

    voltage
    (V)

    Power
    (mW)

    VDD_CORE

    0.05

    0

    0

    1.024

    0

    VDD_MPU

    0.05

    0

    0

    1.107

    0

    VDDS_RTC

    2

    0

    0

    1.805

    0

    VDDS_DDR

    0.24

    4.4

    18.333333

    1.492

    27.353333

    VDDS

    0.24

    0.1

    0.4166667

    1.814

    0.7558333

    VDDS_SRAM_CORE_BG

    2

    0

    0

    1.814

    0

    VDDS_SRAM_MPU_BB

    2

    1.6

    0.8

    1.814

    1.4512

    VDDS_PLL_DDR

    2

    0

    0

    1.819

    0

    VDDS_PLL_CORE_LCD

    2

    0

    0

    1.819

    0

    VDDS_PLL_MPU

    2

    0

    0

    1.819

    0

    VDDS_OSC

    2

    0

    0

    1.814

    0

    VDDA1P8V_USB0

    1

    0

    0

    1.808

    0

    VDDA1P8V_USB1

    2

    0

    0

    3.3

    0

    VDDA_ADC

    1

    0.4

    0.4

    1.819

    0.7276

    VDDSHV1

    0.24

    0

    0

    3.32

    0

    VDDSHV2

    0.24

    0

    0

    3.32

    0

    VDDSHV3

    0.24

    0

    0

    3.32

    0

    VDDSHV4

    0.24

    0

    0

    3.32

    0

    VDDSHV5

    0.24

    0

    0

    3.32

    0

    VDDSHV6

    0.24

    1.1

    4.5833333

    3.32

    15.216667

     

     

     

     

     

    45.504633