Hello,
I came across the forum thread (linked below) and believe I may be seeing this issue. I’d like to know more about the voltage characteristics (voltage levels, durations, etc.) that expose the memory corruption failure. Do you have this information? If so, please share it with me and I can model some tests accordingly.
thanks
Hello Jeff,
I don't know what happened exactly in that thread, but Flash corruption in my experience is almost always due to running the device out of specification. In that case, the customer was running the part on a battery and it probably was slowly falling below the 1.8V VCC limit. Most of the F1xx and F2xx devices do not have integrated SVS to protect against something like this.
These are the key factors to consider:
You must stay within the recommended operating conditions.
Another common violation is system frequency vs supply voltage. Here is the graphic for the MSP430F149. To run at 8MHz, the MSP's VCC must be at least 3.2V. To write Flash, VCC must be at least 2.7V.
I know this graphic isn't the clearest in the F1xx family devices, but it was improved in later generations. Here is a similar graphic for reference for a MSP430F22x2 MCU.
Again, when running outside of these parameters, correct operation cannot be guaranteed. Hope this helps.
Thanks,
JD
**Attention** This is a public forum
