• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 79 subscribers
  • Views 692 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.

msp430f2132 - system resets after waking-up from lpm4

Stephane16371
Prodigy 235 points
Other Parts Discussed in Thread: MSP430F2132

Hi,

I am using the MSP430F2132 with a 32768Hz watch crystal to clock ACLK (clock module: LFXT1 ) for timerA and the internal 8MHz (clock module: DCO) as MCLK/SMCLK.

I use:

- active mode,

- no watch dog,

- LMP3,

- and LMP4,

Everything works fine excepted when waking-up from LPM4 (via IO interrupt). Actually:

- if the time between entering LPM4 and activating IO pin (to wake it up) is less than 2minutes, the device behave correctly (program re-starts where is it was before going into LPM4),

- but if the time between entering LPM4 and activating IO pin (to wake it up) is greater than 2minutes, then the device resets!! (program starts at the beginning)

I have read on the web (http://permalink.gmane.org/gmane.comp.hardware.texas-instruments.msp430.discuss/23770) that LFXT1 may have problem to restart, but I do not understand why and where to find useful informations.

Thank you in advance for helping.

Stephane

  • 0
    Guru 227245 points

    LPM4 disables ACLK and (depending on configuration) the crystal that drives ACLK. It is possible that an extended period of deactivated crystal loads the crystal capacitors with leakage currents, so that it has some problems to restart. (Re-)starting the crystal may take up to several 100 ms.
    It would be important to know whether the MSP resets immediately or whether there is some delay because your code doesn't properly cope with the still inactive crystal.
    You may try outputting an 'I'm alive' signal that is output when the device wakes up, before proceeding with anything. Then move this signal output along the code until it isn't output anymore.

  • 0 in reply to Jens-Michael Gross
    Prodigy 235 points

    Thx for your feedback.

    I finally found the issue. ;-)

    This is a ultra low power application powered with a small cell battery. In fact the uC powers on/off a second device via a fet switch. My mistake was to buffer the power supply of this second device with 2x 22uF capacitors. After waking up from LPM4, the uC powers on the second device. Two cases:

    - less than 2 minutes, the capacitors buffer are still charged enough => no problem,

    - more than 2 minutes, the capacitors buffer are discharged. When powering on, the cell battery charges the capacitors but cannot handle the required current. For a short time the battery voltage drops drastically. This drop is detected by the msp and resets it.

    My solution is to move the 2x22uF to battery side instead of being on second device side.

    Cheers,

    Stephane

  • 0 in reply to Stephane16371
    Guru 227245 points

    Ah, an inrush current vs. supply output impedance problem. Nasty.

    Don't remove all capacitance form the second device. You should keep some µF. Just ensure that the battery side can provide the inrush current. 4.7µF vs. 47µF is a good relation. (no matter what the battery condition is, the voltage drop will then below 10%)

**Attention** This is a public forum