• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 5 answers
  • Subscribers 17 subscribers
  • Views 2934 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.

Compiler/CC3200: Power Managment

xiansheng zhou
Intellectual 270 points
Part Number: CC3200
Other Parts Discussed in Thread: CC3120,

Tool/software: TI C/C++ Compiler

Recently I evaluated the power consumption of CC3200, but I never tested the current 0.85ma in power measurement.My test current is 1.5ma, why do I test and document different?Can you provide the wifi router environment parameters for testing power measurement?Or can you provide the key parameters in the power measurement project?For example, what are the parameters of LSI_SLEEP_DURATION_IN_MSEC?

  • 0
    TI__Guru 63006 points
    Hi xiansheng zhou,

    Are you using the launchpad?
    Our numbers were measured in Idle (no traffic at all), Connected to an AP that broadcasts its beacons every 102.4ms (this is the most common configuration) with no LSI (i.e. LSI is set to 100msec).
    How are you measuring the current consumption? what is the floor consumption? Can you share a photo of your current profile?

    Br,
    Kobi
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points
    Hi Kobi,
    I use the launchpad to measure power consumption using the HVPOWERMONITOR.
    I need help to understand the result of an experiment with LSI and DTIM versus power consumption.
    In fact, I used power measurement to test the DTIM change test, which is not much change in power consumption. Why did this happen?Are there any examples for reference?
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points
    I understand how the DTIM works. My question is DTIM + LSI. As I originally posted, with an LSI of 300, I expect the CC3120 to wake up every 300ms. If I change DTIM from 1 to 5, I would assume no difference in current consumption, but that is not the case according to the current measurement. That is where I need an explanation.
  • 0 in reply to xiansheng zhou
    TI__Guru 63006 points
    Hi,

    I'm not sure what your question/issue is. Please explain.
    Regarding the LSI power estimation, the following provides the expected wakeup time in your case.
    Please find more info in www.ti.com/.../swra502a.pdf:

    Assuming AP Beacon interval of 102.4ms:
    If LSI=300ms and DTIM=1 (i.e. every 102.4ms) -> device will wake up every 300msec (missing 2 DTIM broadcasts).
    If LSI=300ms and DTIM=2 (204.8ms)-> wakeup every 204.8ms (since the LSI dictates the maximum sleep time)
    If LSI=300ms and DTIM=3 (307.2ms)-> wakeup every 307.2ms
    If LSI=300ms and DTIM=4 (409.6ms)-> wakeup every 409.6ms (LSI is not relevant)
    If LSI=300ms and DTIM=5 (409.6ms)-> wakeup every 512.ms (LSI is not relevant)

    Br,
    Kobi
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points
    Sorry, I didn't say that clearly enough.I use the power meter process to measure.The PDF link you provided cannot be opened. Could you please send it to me directly?For example If LSI = 300 ms and DTIM = 2 (204.8 ms) - > wakeup every 204.8 ms (since the LSI dictates the maximum sleep time), I'm a little don't understand?Could you explain?Why 204.8ms instead of 300msec?What is the relationship between DTIM LSI and power consumption in AP Beacon interval of 102.4ms?For example, If LSI=300ms and DTIM=3 (307.2ms)-> wakeup every 307.2ms Average Current is 0.85ma.
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points
    In your development board to burn the power_measurement project, use the power meter to measure power consumption in ALWAYS.
    Why is it not obvious when LSI=100 to modify the parameter power consumption of DTIM?Under what circumstances do you change the power consumption of the DTIM parameter?
    I tested the power consumption of 1.5ma when DTIM=1 and DTIM=5 in LSI=100.Why do DTIM=1 and DTIM=5 have the same power consumption?
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points
    I see part of the document swra502a. PDF, and I can't understand what it means.For example: AP Beacon interval of 102.4ms LSI=500ms DTIM=2 Actual wi-fi Sleep Time=409.6, I can't understand why the Actual wi-fi Sleep Time is 409.6, not 512?Can you tell me?
  • 0 in reply to xiansheng zhou
    TI__Guru 63006 points
    When DTIM is 2 and AP Beacon Interval is 102.4ms -> we basically need to wake up for the DTIM every 204.8ms.

    LSI is defined as the MAXIMUM sleep interval, so in your case we cannot sleep more than 500ms (we will still be ok up to 5*102.4=512ms, compensating for to the beacon interval deviation).

    Waking up every 512ms is not an option because we need to wake up for receiving the DTIM beacon (there won't be one at 512).
    If you set LSI to 500ms, then we will miss 1 DTIM and wake up only for the 2nd (409.6ms).
    The next DTIM will come after 614.4ms which is too long for the LSI selected.

    Br,
    Kobi
  • 0 in reply to Kobi Leibovitch
    Prodigy 100 points

    Kobi Leibovitch said:
    When DTIM is 2 and AP Beacon Interval is 102.4ms -> we basically need to wake up for the DTIM every 204.8ms.

    LSI is defined as the MAXIMUM sleep interval, so in your case we cannot sleep more than 500ms (we will still be ok up to 5*102.4=512ms, compensating for to the beacon interval deviation).

    Waking up every 512ms is not an option because we need to wake up for receiving the DTIM beacon (there won't be one at 512).
    If you set LSI to 500ms, then we will miss 1 DTIM and wake up only for the 2nd (409.6ms).
    The next DTIM will come after 614.4ms which is too long for the LSI selected.

    Br,
    Kobi

    Hi Kobi,

    Thanks for your explain.

    I had met the same question about wakeup time.

    (1)about the understand

    Following your explain,i can understand the Wi-Fi Sleep Time in the file(swra502a.pdf).But i still not clear about your answer(above).

    If LSI=300ms and DTIM=4 (409.6ms)-> wakeup every 409.6ms (LSI is not relevant) 
    If LSI=300ms and DTIM=5 (409.6ms)-> wakeup every 512.ms (LSI is not relevant) 

    why not 204.8ms? why LSI is not relevant?

    (2)my study about your Analysis

    Hardware :CC3200 LAUNCHPAD

    Fireware: SDK,POWER_MEASUERMENT(always on)

    How can I get the same wifi sleep time from your Analysis?

     

    Best Rergards!

    zx

  • 0 in reply to zx
    TI__Guru 63006 points
    Hi zx,

    If the DTIM interval is longer than the requested LSI - then the LSI is not relevant because we will anyway go to sleep for more than the request (DTIM=5 -> we will wake up every 512 ms).
    In your measurements, it seems that there is no effect to changing the DTIM which is very strange and I'm not aware of such issue. In addition, the value of 204 doesn't make sense for any DTIM values other than 2 & 4.
    Please make sure that the DTIM interval was really updated?

    Br,
    Kobi
  • 0 in reply to Kobi Leibovitch
    Intellectual 270 points

    hi

    I have two questions to ask:

    1. At Beacon=100 LSI=300, DTIM=4 my Actual wi-fi Sleep Time is 102ms.Why do you say Actual wi-fi Sleep Time is 400ms?

    2.Beacon=100 LSI=100 adjust the parameters of DTIM, I measured the Actual wi-fi Sleep Time is 100ms, why adjust the DTIM parameter, the Actual wi-fi Sleep Time has not changed?

  • 0 in reply to Kobi Leibovitch
    Prodigy 100 points

    Kobi Leibovitch said:
    Hi zx,

    If the DTIM interval is longer than the requested LSI - then the LSI is not relevant because we will anyway go to sleep for more than the request (DTIM=5 -> we will wake up every 512 ms).
    In your measurements, it seems that there is no effect to changing the DTIM which is very strange and I'm not aware of such issue. In addition, the value of 204 doesn't make sense for any DTIM values other than 2 & 4.
    Please make sure that the DTIM interval was really updated?

    Br,
    Kobi

    Hi  Kobi,

    Thanks a lot.

    I had recheck my code,

    In the code of Powermeasurement,when i used UDP mode, it seems can not capture the Beacon(but i donot know what happend),which make the data show in above.

    Now i changed to TCP mdoe ,the data (DTIM interval <LSI) follows your explain,but still there were something wrong with the data(when DTIM interval >LSI) 

    The data of  most right column is the  wifi wake up interval(ms) time.

    I had still try to get the relationship between LSI and  DTIM interval (when DTIM interval >LSI) ,whether you advice a code (from the CC3200SDK_1.3.0)? thanks.

    Best Reards!

    zx

  • 0
    TI__Guru 63006 points
    Hi,

    I assume you have already found a solution for this one so I'm closing this thread.
    If the issue is still relevant, please open a new thread.

    br,
    Kobi