• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 98 subscribers
  • Views 503 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.

[C6746] characterizing power consumption

Naoki Kawada
Guru 19890 points

Hi,


Just a short question. My customer is wondering if TI must have built your own board and validate the power consumption to issue the power consumption summary like this :

And now I'm getting some questions related to the test environment.

From my understanding, TI does not make such a test environment for power consumption summary sheet and this sheet has been made by considering the silicon/device characteristic. Is my understanding correct ?

Best Regards,
Kawada

  • 0
    TI__Mastermind 35645 points
    Hi,

    We will work on this post and will let you know the update shortly.

    Thanks & regards,
    Sivaraj K
  • 0
    TI__Mastermind 43660 points
    Kawada,

    Yes, TI had built its own board (power characterization board) to validate the power consumption and issued the power consumption summary based on the measurements from power characterization.

    Your understanding is wrong. TI has followed all the required test environment for the power characterization.

    Regards,
    Senthil
  • 0 in reply to Senthilkumar Srinivasan
    TI__Guru* 83955 points

    Kawada-san

    To expand further on Senthil's response. It is not really clear what you meant on test environment, but as Senthil mentioned, for this device family we have an internal validation and debug board that is used for silicon bring up and verification, this was also used for power characterization.

    The power estimation spreadsheet is built upon modular characterization of leakage and activity power , done with internal only test suites for power for each module , the testing is done over samples representative of the worst case spectrum of the process and it is done on the specified datasheet voltage and temperature ranges.

    The test board and test cases are internal only and are not intended to be shared with customers. However they should be rest assured that the spreadsheet is built upon actual silicon/power characterization.

    Hope this helps.

    Regards

    Mukul

  • 0 in reply to Mukul Bhatnagar
    Guru 19890 points

    Sivaraj and Mulul,

    Thank you so much for your help.

    Ok, understood. In fact, my customer's goal is validating that their application does not violate the junction temp requirement under their system operation. The ambient and case temp has been verified on their system -- ambient : 65*C and case:75.1*C. If the actual power consumption has been verified, they can determine the exact junction temp because TI provides thermal resistance information in the datasheet. Right ?

    But the problem is , they are saying they can't measure actual power consumption on their target board because power lines are actually shared with various devices.  It seems there is no reasonable test point for multi-meters. So now they are considering to use power spread sheet to estimate the power consumption for countermeasure. The above confirmation was raised as one of questions.

    Ok now, please let me confirm the usage of power spread sheet. I think I know the basics for power spread sheet, but now I have some questions about some parameters for the spread sheet.

    Please take a look at the following snippet:

    As you see, there is a parameter for junction temp.

    Question1:
    What value should we use for this parameter ? Can we use a value including some assumptions ? For example, like this case : The case temp is validated on their system so putting a little bigger value than validated case temp for this parameter.... Correct ? Please note the preset for Typical 300/456Mhz is using 25*C for this value. 

    I'm not sure the above assumption is correct or not, but anyway, something value will be set to junction temp parameter. And after some other parameters are correctly filled (for example, module usage, CPU/module utilization etc..), the spread sheet calculates the active/static/total power consumption. The following the result from Typical 456Mhz.

    From my understanding, these values can be referred as inputs to "Case Temp Measurement" tab in the spread sheet. 

    Question2:
    At this point, which value should be go to #3 & #4 on "Measurements from Actual Application" ?

    Sorry to ask you so much, but I will appreciate if you would answer to the questions soon.

    Best Regards,
    Naoki

  • 0 in reply to Naoki Kawada
    TI__Guru* 83955 points

    >> If the actual power consumption has been verified, they can determine the exact junction temp because TI provides thermal resistance information in the datasheet. Right ?

    [MB] Correct

    >>What value should we use for this parameter ? Can we use a value including some assumptions ? For example, like this case : The case temp is validated on their system so putting a little bigger value than validated case temp for this parameter.... Correct ?
    [MB] Ideally you would have power measurements from a real system/PCB and along with which you will have ambient and case temperature. The case temperature tab allows you to calculate Tjunction, assuming worst case leakage/static power - as #4 maybe the static power measured on a typical/nominal unit in customer system, and #7  gives the worst case leakage possible - since customer will not always get units that are representative of worst case leakage, this spreadsheet tab allows you to comprehend this by replacing "measured" static power with worst case static power. Since the customer does not have the ability to isolate the device power , they will need to (as you said) use the power estimation spreadsheet and roughly estimate their junction temperature in the main tab to slightly higher than what they see their case temp to be.

    You can refer to section 4 and 5 of the thermal application note.
    This is mentioned in the power consumption summary wiki
    processors.wiki.ti.com/.../OMAP-L138_Power_Consumption_Summary
    Hope this helps.

    Regards
    Mukul

  • 0 in reply to Mukul Bhatnagar
    Guru 19890 points
    Hi Mukul,

    Thanks for your answer. I just forwarded the same to the customer. Following this scheme, they might be able to determine their system does not violate the junction temp requirement....

    Thank you.
    Naoki