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.

MSP430F135: MEMS oscillator on XT2IN seems to weaken LPM3 power savings

Part Number: MSP430F135

Hej,

I have a MSP430F135 that sports a 32kHz oscillator on XIN/XOUT and a DSC1001 MEMS oscillator at 8MHz on XT2IN that I disable in LPM3 using its STANDBY input.
Prior to entering LPM3 I switch MCLK to LFXT1 and SMCLK as well. Then I put the MEMS into standby.

Now, when I put the application into low-power-mode 3 (LPM3) the above mentioned circuit draws about 340uA.
As soon as I put my finger on the XT2IN, it introduces quite some noise BUT also the current drops to 14uA.

So I suspected the input to be floating and added a 100k pulldown. However, the circuit still draws 340uA instead of the desired 14uA.

Has anyone a clue how to solve this? I'd rather like to not ship a standard-finger with every device..

Thanks in advance, Joern.

  • Hi Joern,

    Thanks for your post. I find this behavior strange and would expect the opposite. When you place your finger on XT2, I would expect that it changes the effective load capacitance on the XT2 crystal which would probably cause an oscillator fault. According to the user's guide, when a XT2 oscillator fault happens and when MCLK is sourced from XT2, MCLK is automatically switched to the DCO for its clock source. Thus, I would expect the results to be opposite where current would jump from 14uA to 340uA due to the DCO powering on. However, you mention that you're switching MCLK and SMCLK to LFXT1.

    Do you know the expected current consumption from the XT2 oscillator? Assuming there's an oscillator fault when you touch it, this could be disabling the oscillator causing your current consumption to perhaps reflect one of the MSP430F135's low power modes. In the thread below, I read where XT2 crystals can consume 340uA or more.

    why device consumption is so high in low power mode when i use external crystal as the device mains-clock?

    Please see Figure 4−3 in the user's guide. It illustrates all the signals that impact the XT2Off signal. Perhaps, your code seems to be disabling XT2 when it's not.

    Regards,

    James

  • Hi James,

    Thank you for taking the time to find an answer to my msps mysterious behaviour. 
    Oscillator fault can be ruled out as I have already properly switched to MCLK:=ACLK which works well.
    The MEMS oscillator (I'm not using a crystal) draws max. 10uA in standby mode which I enable.
    (the MEMS being in run mode consumes 6mA)

    The Figure you suggested to have a look at is indeed a valuable resource.
    I found that in the BSCTL2 register, Flag SELS selects the SMCLK source.
    In my code I kept it on (SMCLK set to XT2CLK) because i interpreted the "LFXT1CLK when XT2 ocillator not present on-chip" as
    "if not enabled". And since I had disabled XT2CLK I assumed this could be left untouched.

    However, thanks to your post I got suspicious, revisited my code and cleared the SELS flag. And voila, current consumption drops
    to 14uA, problem solved.

    thank you,

    Joern.

**Attention** This is a public forum