• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 79 subscribers
  • Views 1834 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.

LPM3.5 and Vdd

Pibe
Expert 1635 points
Other Parts Discussed in Thread: MSP430F5659, MSP430F6638, MSP430F6659, ENERGYTRACE, MSP-FET

Hi,

Today I found out what seems to be a dependency between LPM3.5 mode and Vdd.  Code that runs without a problem at Vdd = 2.7 VDC will not work at 2.4 VDC (the tipping point is 2.6 VDC).  I checked also with the code provided by TI in the examples and it does the same. Namely, while the LPM3.5 mode is started without problem (Idd = 1.0 uA), then it never comes back (the RTC ISR never runs either).

I read someone recommended to turn off the supervisor (PMMCTL0_L &= ~(SVSHE + SVSLE);) but it does not have any effect at all (or just PMMCTL0_L &= ~SVSLE;).  Any clue?  Thanks.

Using MSP430F5659 on PZ100USB board.

-Pibe

  • 0
    TI__Guru* 78191 points

    Hi Pibe,

    I have the same MSP-TS430PZ100USB and a similar device (MSP430F6638) here - do you think you could let me know which example code to use to reproduce the issue (is it MSP430F665x_LPM35_RTC.c)? In addition, how do you have the jumpers on JP11 (for VBAT connection) and JP3? Basically, let me know the setup and steps to reproduce the issue here so that I can help you investigate.

    Regards,

    Katie

  • 0 in reply to Katie Pier
    Expert 1635 points

    Hi Katie,

    Thank you for looking into this.
    Yes, the TI code I tried is the one named MSP430F665x_LPM35_RTC.c.
    In JP11, Vbatt is connected via jumper to DVcc (1-2).
    JP3 has a connection between 1-2.
    JP1 has also 1-2 connected.
    JTAG option selected with the six contiguous headers.
    I tried with an MSP430F5659 and MSP430F6659.
    I have also tried reducing the value of the internal caps for XT1.

    The MSP-TS430PZ100USB board is rev1.25.

    Regards,
    Pibe

  • 0 in reply to Pibe
    TI__Guru* 78191 points
    Hi Pibe,

    I tried out the same code example that you mentioned on an MSP430F6638 device. However I was unable to reproduce your issue. Even with down to 2.2V (I didn't test any lower) the LED still turned on after 1 minute (the alarm setting) as expected. Please note that the LED was much dimmer at this voltage - could this be part of your issue? If you use a multimeter to measure the voltage at P1.0, you should see it does get set to "high" state after 1 minute though.

    You also need to make sure to do the test with Free Run from the debug session or completely stand-alone for the code to actually enter LPM3.5. Did you have any FET tool etc connected when you did this test?

    -Katie
  • 0 in reply to Katie Pier
    Expert 1635 points

    Hi Katie,

    Thanks for your suggestion. Good point, although I did not rely on the light of the LED. I was monitoring the supply current with a micro-ammeter (the board is running stand-alone) and did not see any change in the total current from 1.0 uA when feeding the board with less than 2.60 V.

    Still, I measured it again because the truth is I do not know how close to the rail the output HIGH is and may had been already below the LED threshold (although the micro-ammeter should still catch some small variation). (EDITED: I should have seen the AM current again, for Ampere sake!). So I removed the meter, connected the power supply directly to the board and just looked at the voltage on P1.0 and now I can see the true threshold is 2.48 V. I cannot go lower than that though.

    I will continue looking into it. Please let me know if you think of something else.

    Thanks again,
    Pibe

  • 0 in reply to Pibe
    Expert 1635 points
    BTW, did you try stand-alone? I mean, with the debugger it runs just fine, even at 2.0 V.

    -Pibe
  • 0 in reply to Pibe
    TI__Guru* 78191 points
    Hi Pibe,

    I tried using the MSP-FET, but connecting with wires only Vcc and GND (no JTAG/SBW connections at all). It still works in this case, and the part isn't in any debug mode (to get it to supply a different voltage, I set the energyTrace preferences to provide 2200mV instead of 3000mV, then launch the stand-alone energyTrace that only requires Vcc + Gnd and no debug session).

    What kind of power supply are you using? Is it a bench supply? Is there a current limit set somewhere perhaps?

    Regards,
    Katie
  • 0 in reply to Katie Pier
    Expert 1635 points
    Hi Katie,
    I will be out of the office until Wednesday. I am using a bench PS from Agilent, I think a U8001 or so. I do not believe this is related to the test setup, but most probably to some missing setting when entering into LPM3.5.
    Talk to you soon,
    Pibe
  • 0 in reply to Pibe
    TI__Guru* 78191 points

    Hi Pibe,

    You can confirm though that you are using the TI example I linked above MSP430F665x_LPM35_RTC.c, completely unmodified? This is the exact same code that I used, so I'm unsure what else could be different other than the test setup. If it was the code, and we use the same code, then we should not be seeing different results.

    -Katie

  • 0 in reply to Katie Pier
    Expert 1635 points

    Hi Katie,

    I can confirm the same behavior with the code your last post points to.

    I am using a PZ100USB board r1.25 with either a 5659 or a 6659, connected to an Agilent U8001A (with the current limit set for the initial tests to 50 mA, and then for the most recent ones to 500 mA), with jumpers arranged as stated in a previous post. I watch the signal at P1.0 with a scope. The board has nothing else connected and works stand alone.

    It may be useful to get the project file too just in case my current project default settings have something to do with it.

    Thank you,
    Pibe

  • 0 in reply to Pibe
    Expert 1635 points
    Hi Katie,

    Any update on this? I would also like to know what revision is the MCU you use for your test. Mine is marked REV A.

    Regards,
    Pibe
  • 0 in reply to Pibe
    TI__Guru* 78191 points

    Hi Pibe,

    Sorry for the delay - I'd been out of office all week until yesterday, and am still getting caught back up.

    I had been testing on F6638 before. So now I found an MSP430F6659 Rev. A to test on my MSP-TS430PZ100USB rev 1.25 target board. Using a bench supply in our lab, I supplied even as little as 1.84V to the device, loaded with the example code msp430f665x_LPM35_RTC. Even with this low voltage I could not reproduce your issue - the LED always came on (though quite dimly) after some amount of time (the time for the RTC alarm). So unfortunately I still cannot reproduce your issue.

    For the project, I simply imported directly from TI resource explorer - when you use resource explorer, it creates a fresh project for the code example - so you could try this in a new workspace and see if you have different results. If you want to, you could alternately try using the cloud resource explorer (at dev.ti.com) to download the project fresh, or use it to import to CCS cloud (dev.ti.com) if you are trying to do some comparison without having to delete or install anything new.

    Regards,

    Katie

**Attention** This is a public forum