• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 79 subscribers
  • Views 806 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.

Repeated UART failure (MSP430F14x)

stephane adams
Prodigy 20 points

Other Parts Discussed in Thread: MSP430F147, MAX202

We have several hundreds of device incorporating MSP430F147/148/149 running.

Recently, we started experiencing failures in UART1 Rx pin (URXD1). This pin is driven by a logic gate powered by the same 3V3 supply as the MSP.

Symptom is a sudden drop in input impedance of URXD1 pin, giving almost a short circuit to GND.

All other pins of the MSP (including UART0, ADC12, P1, P2, P4, P5 port pins behave normally.

We tried several solutions to avoid the blowup of URXD1  : driving URXD1 through a 4k7 resistor, using diodes to Vcc and GND to avoid any voltage peak at startup, ... but the problem still randomly occurs.

Are there known vulnerabilities in this circuit ?

  • 0
    Guru 227245 points

    stephane adams said:
    Recently, we started experiencing failures

    So what has changed compared to before? You didn't have this kind of problems before but now you have. Something must have changed.

    Generally, all MSP port pins ahve clamp diodes which can source/sink 2mA against GND and VCC. If you ensure that the current is limited to 2mA, the MSP won't take any damage. Your 4.7k resistor should cover any voltage peaks up to 9.4V above VCC.
    Also, you say that the pin is driven by a gate that is powered by the same VCC as the MSP. Then it should be impossible to overload the pin.

    My own setup (working with a gate that is powered by 5V, a cheap MAX202) is a diode between driver and MSP and a pullup resistor to VCC on the MSP side. This turns any source driver into an open collector driver.

    It is possible that the 'gate' (whatever it might be) is driven itself by a voltage outside of its operating range. This voltage might push through to the output. This might happen when the device is powered up (or powerless) but the input signal is already applied. Did you change your conenction equipment? Switch from a real serial port to an USB serial converter on the PC side? Maybe a superimposed potential difference?

    The only reason I can think of that might cause the input impedance towarts a shortcut to ground would be melting the clamp diode to GND (effectively turning it into a shunt resistor). It's something I have seen happening with other diodes. But it's not easily happening. The reason in my cases were high-energy transients (in a switching power supply). I've never seen it on an MSP port pin, and we too built several hundred devices with RS485 and RS232 connections for industrial environment.

  • 0 in reply to Jens-Michael Gross
    Prodigy 20 points

    HI Jens-Michael, thanks for replying. I 100% agree with what you say, we already investigated all these possibilities. We have no explanation for the blowup. We didn't change our design. That's why we start thinking about a bad MSP430 lot. Another possibility might be incomplete control of ESD during manipulation of the chips, but in this case it's very strange always the same pin (UART1-Rx) fails and no other pin.

    I'm pretty sure other users must have experienced similar problems, that's why we posted it here.

**Attention** This is a public forum