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.

MSP430F6736A: voltage at RST pin less than AVCC

Part Number: MSP430F6736A

Hi

i am developing a custom power meter board...one of my tests involves switching a 240V-12V halogen lamp ON/OFF. (The lamp uses a plain simple transformer to drop the voltage).

I am sampling the line voltage (suitably reduced) at SD0P0/SD0N0.

I notice the lamp is giving off a lot of noise that presents itself as spikes (of about 5nS) occurring at about 1MHz for about 50uS whenever it switches ON/OFF.

The noise is not affecting the reading that i am getting from the SD24B (i still get the correct voltage).

I placed a scope probe at pin 9 and observe that the spikes have permeated the AVcc line but they dont appear to cause any issues there.

I next placed a scope on the RST pin and noticed that the spikes are also visible there...BUT...in addition, i noticed that the voltage in between the spikes has now trended downwards to about 2V (this trend was not observed at pin 9)....that low voltage now caused the MCU to reset.

So my question:
1. What is the connection between the noise appearing at the SD24 and the decrease in voltage at the RST pin??
2. Why is the MCU internal voltage, as seen by the voltage at the RST pin (in between the spikes), so much lower than AVcc?

Scope captures can be upload if needed. 

thanks in advance

  • Hello Moshe,

    If the voltage on RST is low and has a lot of noise, then this is obviously an issue. I don't think it's an MCU issue, so much as it is a system level one. The MCU does not supply voltage to the RST line. RST should be pulled up externally and have a filter cap. The filter cap cannot be to large, as it can then interfere with programming and JTAG communication. Figure 2-1 shows the recommended RST pin pull-up and capacitor values. www.ti.com/.../slau278ae.pdf

    Do you have RST pulled up externally? If so, what is it being pulled up to, DVCC? What is that voltage/ noise look like?

    Thanks,
    JD
  • Hi JD

    Thank you for responding.

    (This is an amended reply to you in case you saw the original).
    Yes, the noise is an issue (which i am addressing) but its inadvertently/indirectly highlighted another issue (which i am trying to understand).

    To answer you (& add more questions).
    As per the User guide & datasheet, the RST pin has an internal pull-up that is enabled by default.
    So question 1:
    Why is an external pullup needed if there is an internal one??

    As it is, I DO have an external pullup of 47K with a 2.2nF cap which is pulled up to DVcc. and THAT voltage REMAINS steady at 3.3V (albeit with the same noise).
    As an exercise, i did remove that 47K resistor & saw that the voltage at the RST pin remained at about 2V (in between the spikes).
    [i expected it to be much less in this instance]

    So i can surmise the RST pin is sinking about 27mA when the spikes are around.
    However, when the lamp is NOT operating (& there are no spikes) the RST pin voltage remains at 3.3V. So the RST pin is not sinking any current at that time (which is the usual situation).
    So question 2:
    a) Why is the RST pin sinking current when the spikes are around??
    b) Why is the RST pin NOT sinking 'some' current when the lamp is not operating??

    thanking you in advance
  • Hello Moshe,

    1. Good point about the internal default pull-up on reset. I had forgotten about this feature on most F5xx and F6xx devices. This means that an external pull-up is not required, but it might still be recommended in this case.

    2. Every GPIO will have a very small amount of leakage current, but I don't really think that would be enough to cause the issue here. Is it possible that you are seeing an RC Ripple/averaging effect on the RST line?

    The fact that you see the voltage rise when you add a 2nd resistor pull-up, basically cutting the pull-up R value in half since it's in parrallel with the internal pull up. This is going to reduce the RC time constant on the RST pin, which will probably increase the voltage but show the noise a bit more.

    To test this, you could put a cap 1/2 the size instead of the external pull-up and see what kind of effect you see.

    Thanks,
    JD

**Attention** This is a public forum