Other Parts Discussed in Thread: MSP-GANG, MSP-FET, MSP-TS430PW20
I have completed three different low volume builds. Each build added more and more protection circuitry along with more care in assembly and handling. Yet, each build gives a yield of about 50% failure. The failure is determined by checking the voltage on pin 10, which is internally connected to the output of the comparator (eCOMP), after programming.
I have done ESD testing on a couple of working boards and they are able to withstand 8 kV direct contact without damage. Before programming all voltage levels are in an acceptable range on all boards. Before programming, pin 10 measures ~1.1V. After programming, a board fails if pin 10 measures 3.3V. On a good board pin 10 measures 0.0V. According to theory, the output of the comparator (pin 10) should not be a logic high, because zero volts is applied (measured at the pin) to the positive input of the comparator and the negative input of the comparator is connected to a DAC whose output is set to 0.75 * Vcc (2.475V). If the code was wrong or eCOMP misconfigured, then no boards would work.
If the MSP430FR2111 is replaced, there is a chance that the board will start working. When the MSP430 is replaced and the board starts working, this makes me believe that the problem is not due to external circuitry on the board or faulty programming. I am not sure what the next step should be in troubleshooting this problem.