• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Subscribers 81 subscribers
  • Views 949 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.

MSP430FR5989: Sampling Rate doesn't match power consumption

Samuel Pérez
Prodigy 205 points
Part Number: MSP430FR5989
Other Parts Discussed in Thread: EVM430-FR6989, ENERGYTRACE, UNIFLASH

Dear Ti team,

I'm having a problem with the sampling rate, and power consumption on ESI according to this:

I've changed the sampling rate with the values: 6, 10, 25, 50 and 100 Hz and the consuption didn't change. Using FLOW ESI GUI I could find the parameter to change:

    // Set sampling frequency to User selected sampling frequency
    gTSMSettings = ESI_TSM_INITPARAMS_DEFAULT;
    gTSMSettings.startTriggerSelection = ESI_TSM_START_TRIGGER_ACLK_OR_SOFTWARE;
    gTSMSettings.startTriggerAclkDivider = ESI_TSM_START_TRIGGER_DIV_450;       //  450 -> 6Hz 390->10Hz  162 -> 25Hz   78-> 50Hz   30-->100Hz
    ESI_TSM_init(&gTSMSettings);

Can someone help me? Is there any reason or a recommendation? Power consumption is around 6.5uA

Best regards

Samuel

  • 0
    TI__Intellectual 2200 points

    Hello Samuel,

    Just to double check, have you tried this:

    Thanks and best regards, 

    Juan. 

  • 0 in reply to Juan Aparicio Jimenez
    Prodigy 205 points

    Hi Juan,

    I'm using custom PCB just with the hardware needed. For changing sampling rate I've just edited gTSMSettings.startTriggerAclkDivider and power consumption is the same in each configuration.

    Best regards

    Samuel

  • 0 in reply to Samuel Pérez
    Prodigy 205 points

    Moreover I don't understand exactly how the GUI calculates it. ACLK runs at 32768 Hz and the maximun divider is 450, if we divide it, we have 72.81 Hz, how is it possible to "sample" at 1, 10, 15, 20... Hz??

  • 0 in reply to Samuel Pérez
    TI__Guru 71216 points

    Hello Samuel,

    What is the lowest current you are able to achieve?

  • +1 in reply to Dennis Lehman
    Prodigy 205 points

    Hello Dennis,

    I achieve around 7uA, but it doesn't change when I sample to other frequency.

    Best regards

    Samuel

  • +1 in reply to Samuel Pérez
    TI__Mastermind 29105 points

    Hello Samuel,

    Sorry for the delay.  Once you get <10uA, it can be almost anything pulling the additional current.  If the sampling frequency doesn't have an effect, then maybe it's something else drawing power.  

    Also, the table you posted is showing a change of a couple of uA per few hundred Hz change, but the software settings only mention from about 10-100 Hz, so it seems like the current change from small changes in the sampling frequency may be very minimal.  

    What else is the MSP430 doing in software?  Have you tried just only putting the MSP430 in sleep mode to see mode to see the base level being pulled in your system/configuration?   I would recommend starting in only sleep mode and getting the current to where you expect it.  From there, start evaluating the rest of that application.  

    Thanks,

    JD

  • 0 in reply to JD Crutchfield
    Prodigy 205 points

    Hello Crustchfield,

    Thanks for your reply. I missclicked your comment in "This resolved my issue", don't know if you can change it.

    Well in theory it can't be something else, due to the fact that we made a custom PCB explicity to measure this with just the hardware needed. I understand that it's just a few uA, but with the minimal sampling rate we can't achieve that. Software in this part is only in sleep mode, no detection.

    Best regards

    Samuel

  • 0 in reply to Samuel Pérez
    TI__Mastermind 29105 points

    Hello Samuel,

    Can you share your sleep only software?  How are your GPIOs configured?  Are they connected to anything?  GPIO configuration is important when trying to minimize power in LPM modes.    

    What TI EVMs do you have available? Do you have an MSP-EXP430FR6989 launchpad possibly?  We could run your code or the LPM3 example there to see just the MSP430 only power there and then move that software to your board and evaluate any differences.  

    Thanks,

    JD

  • 0 in reply to JD Crutchfield
    Prodigy 205 points

    Hello Crutchfield,

    Here I share my CCS project and schematic, GPIO are output LOW. I also have MSP-EXP430FR6989, I've tried that example and power consumption is 3mA, I noticed that the script doesn't take into account I2C pull up of EVM, so I added 

    P1OUT |= BIT6 + BIT7;                // for I2C pull up. For measuring current consumption, remove the physical I2C cable.

    to the code and power consuption is 34 uA.

    Best regards

    Samuel

  • 0 in reply to Samuel Pérez
    TI__Mastermind 29105 points

    Hey Samuel, 

    I wasn't able to access the Google Drive.  Is it a large project?  Can you just post the .c file here either as an attachment or just paste it using the "Source Code" tool?   

    I think we should get the numbers down on the launchpad then move back to your board.  How are you measuring the launch pad power?  Are you using an external tool or the launchpads Energytrace feature?  

    I'm guessing the additional current is probably coming from the built-on programmer or having the debug circuitry activate.  To measure the power, we should run the MSP430 without the debugger active.  You can either "Free run" the MSP430 while using energy trace or just disconnect the programmer by removing the J101 jumpers and supplying power to the board (or placing current meter between the 3V the EX-FET to the MSP430 side of the launchpad.) 

    I don't have this exact launchpad currently to test with, but we should definitely be able to get it down to 1uA or less with nothing connected to it. 

    Thanks,

    JD  

  • 0 in reply to JD Crutchfield
    Prodigy 205 points

    Hello Crutchfield,

    Maybe if you tell me an email i can share directly, because I added schematics so you can understand everything better Edit: (try here).

    I'm measuring with a Otii, a power analyzer, without the debug circuitry, loading program with Uniflash, removing all jumpers and with Otii i power supply the board.

    Best regards

    Samuel

**Attention** This is a public forum