• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 79 subscribers
  • Views 1232 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

MSP430F2370: Reset pin not being pulled up to VCC, appears there is leakage current into MSP pin

Chad E
Prodigy 105 points
Part Number: MSP430F2370
Other Parts Discussed in Thread: MSP430G2432

We are seeing an intermittent issue on our product where the RST/NMI pin appears to be pulling something on the order of 10+ uA into the IC. The design uses a 100k pullup to 3.0 Volts and a 0.01uF cap on the reset line. There is not other control of the reset. 

On some startups, the RESET pin appears to be pulling a large amount of current. The result is a voltage drop across the 100k pullup that varies from small up to and included 1.4 Volts, equivalent to 14 uA into the MSP! Once the issue is seen, subsequent power off and startups do not clear the problem. When left off for a long time (overnight) there is some evidence that the problem does not occur on the first power up after that.   

The same issue as been reported on a few threads, such as this one of April 20, 2017. "MSP430G2432: RESET pin drop issue on MSP430G2432". That thread said that it was resolved offline. I would like to know how that was resolved, because I am having the same problem.

 

FYI, I have looked for flux contamination and possible shorting on the PCBA, and have found no evidence of either.

 

Thank you!

Chad

  • 0
    Prodigy 105 points
    More information:

    I left the system off for about an hour. When I came back and plugged in the battery, it powered up and the reset pin was at 2.94 Volts, about where it should normally be. After some time (10 to 60 seconds , unfortunately I did not track), the voltage suddenly dropped to 2.0 Volts, and the problem was back. It now sits at 2.0 Volts.
  • 0 in reply to Chad E
    TI__Guru 56685 points
    Hello Chad,

    Typically, we recommend a 47k ohm resistor for the RST pullup. Can you do a quick exchange to see if this helps with your issue?
    your capacitor value is too high as well, which could be causing the MSP430 to stay at a bad voltage as it tries to dissipate. The upper limit for this cap on RST is 2.2nF.

    What does the voltage going into DVCC look when this happens?
  • 0 in reply to Jace H
    Prodigy 105 points
    Thanks for the quick response. I already tried to lower the pull-up resistance, to 40k and then to 15k, without improvement seen at subsequent power ups. I increased the capacitance to 22 nF with no improvement (with 100k pullup), but as you noted I need to go the other way and decrease it. I will try that. VCC remains clean in all situations, at both the second time scale and micro/millisecond timescale, as far as I have seen on the scope.

    MSP is powered from a lithium polymer battery through an LDO (ADP121) that can source up to 150 mA continuous. The output of the LDO has 2.2uF of decoupling capacitance attached.

    I will try to lower the capacitance and let you know if I see any improvement. Thanks again
  • 0 in reply to Chad E
    Prodigy 105 points
    I have confirmed that both VCC and the voltage on Reset pin drop quickly, to 50 mV or less within seconds of power off. I thought that the MSP might possibly be back-powered at the Reset pin by some unexpected droput of VCC, so I looked at that early on but found no evidence of VCC hovering at any voltage other than 3.0 Volts and off (0.0 Volts).
  • 0 in reply to Chad E
    Prodigy 105 points
    Replaced the capacitor on RST with a 680 pF (down from 10,000 pF). Still using 100k pull-up. When powered up, the pin came up cleanly to 2.90 Volts. It stayed there through three or four power-down cycles. I thought that the problem had disappeared. About 10 minutes later, I come back and RST is sitting at 1.90 Volts. It's back!!
  • 0 in reply to Chad E
    Genius 3850 points
    the chip might be defective?

    My reset circuit is typically 10k/0.1uf - no issues whatsoever. Unless I have a special reason to do otherwise, I stay with that combination.
  • 0 in reply to Chad E
    TI__Guru 56685 points
    Chad,

    What are your decoupling caps on DVCC?
    Are there any other pins directly connected to some sort of power source?
    When you are testing power dropout, are you removing the battery or using a bench supply and turning it off? can you test with a bench supply?
  • 0 in reply to Jace H
    Prodigy 105 points
    Chad,

    What are your decoupling caps on DVCC?
    Decoupled with 2.2 uF, a mix of three 0402 ceramic caps. As sated above, VCC appears steady at 3.0 V at all time scales.
    Are there any other pins directly connected to some sort of power source?
    No. I assume you are thinking that the MSP is being back-driven somehow. I can find no evidence of that. When battery is disconnected, all power is off (0 volts).
    When you are testing power dropout, are you removing the battery or using a bench supply and turning it off? can you test with a bench supply?
    Battery is being disconnected, sometimes for hours. There is evidence that RESET will start up at a normal voltage after a prolonged battery disconnect, but it then quickly decays down to 1.9 to 2.5 Volts within minutes. Subsequent quick power off/on cycles show the reset remaining at the low voltage.
    I just now tested it with an external DC supply in place of the battery, and there is no difference.
  • 0 in reply to Chad E
    TI__Guru 56685 points
    Chad,

    Can you try the test with an LPM3 code example?
    Also, is there anything else on the board that could be causing voltage drop? It doesn't make sense to me that RESET line is lower the VCC since the RESET line should be tied to VCC. Are you willing to share your schematic?
  • 0 in reply to Jace H
    TI__Guru 56685 points
    Chad,

    Any results on your testing?

**Attention** This is a public forum