MSP430G2755: Failure to run the program after 10 seconds the MCU was powered

Hi 

How many devices have this issue? The external crystal unstable may cause the device doesn't work. Your customer can refer to some oscillator application notes to check it.

Some layout question:

1. the oscillator is closest to IC?

2. The capacitor is closest to oscillator?

3. The value is match the oscillator?

https://www.ti.com/lit/an/slaa225a/slaa225a.pdf?ts=1657095983444&ref_url=https%253A%252F%252Fwww.ti.com%252Fsitesearch%252Fen-us%252Fdocs%252Funiversalsearch.tsp%253FlangPref%253Den-US%2526searchTerm%253Doscillator%2526nr%253D177

https://www.ti.com/lit/an/slaa322d/slaa322d.pdf?ts=1657095981792&ref_url=https%253A%252F%252Fwww.ti.com%252Fsitesearch%252Fen-us%252Fdocs%252Funiversalsearch.tsp%253FlangPref%253Den-US%2526searchTerm%253Doscillator%2526nr%253D177

https://www.ti.com/lit/an/sprab84a/sprab84a.pdf?ts=1657095986041&ref_url=https%253A%252F%252Fwww.ti.com%252Fsitesearch%252Fen-us%252Fdocs%252Funiversalsearch.tsp%253FlangPref%253Den-US%2526searchTerm%253Doscillator%2526nr%253D177

Dear Allen,

Thanks for your reply!

The customer said about 5% G277 may have this problem. The external oscillator used can let most of G277 work in normal.

Do you have some new ideas based on this situation? 

Best regards

Sorry, I mean the G2755. Sorry for the wrong typing.

Hi 

Customer need to check if the unstable external crystal cause the issue. Because after reset the device can work normally, that means the device is fine. Maybe it is in an unstable status at the beginning(maybe the oscillator or some thing else). Do they have some test to find out it?

Dear Allen,

Yeah, that's the confused point. The G2755 can work normally after they repower the G2755. If it can work normally after repowering, the oscillator should be fine? 

The customer had tried to change the different G2755 to different machines to check whether is their machine malfunctioned that caused the failure of G2755. However, the faulty G2755 could not work in different fine machine in normal.

Dear Allen,

The customer had read the 3 discussion you provided before.

If the external oscillator is unstable, does some abnormal waveform will be monitored? 

The customer had tested and found that there were no frequency or amplitude jumping if the external DCO 8MHz is connected after the external passive oscillator failed to be driven by G2755,  and the waveform of crystal oscillator was normal.

Hi

Xenon Li said:

The customer had tested and found that there were no frequency or amplitude jumping if the external DCO 8MHz is connected after the external passive oscillator failed to be driven by G2755,  and the waveform of crystal oscillator was normal.

Can customer try another crystal which needed lower drive current. Seems that driving capability of the device leads to the problem of crystal oscillator start-up.

Hi Allen, thanks for your reply.

The customer said that sometimes the oscillator could not be driven, but most of the time the oscillator could be driven after the G2755 went wrong. And even if the external oscillator could not be driven, the internal oscillator would be selected and start to work normally.

Hi 

You are right. Generally, if the fault is detected and flag is set, the clock will change to internal DCO. 

If at the beginning, customer is using internal DCO all the time not the external crystal, will the issue occur?

Customer can recover the device by reset of power off, will it occur again after that? From your description it happens in the first time. It is weird that the problem only happens once. Could customer find some ways to reproduce it? 

Do they test the supply power when the issue occur? Is there some under voltage happened?