Part Number: MSP-FET
Hi folks/ti,
I've seen other users complaining about failed MSP-FET programmers in the past, nothing in those posts has helped me, but I've not really had a significant issue until more recently.
Approx. 5 years ago I designed and built a "one off" piece of test equipment, including the mechanical, electronics and software components. The test equipment loads firmware to a PCBA that goes inside one of our products as well as perfoming some rudimentary functionality tests. It's all driven by a Raspbery pi 3B and uses the MSP-FET to program an MSP430 microcontroller on the PCBA being tested.
In the interveing years I think we have managed to have one MSP-FET fail which isn't bad . . . but more recently we have had them fail frighteningly frequently, I have a pile here on my desk of 7 units, at least 4 are very recent.
Obviously I couldn't let this continue and had to get to the cause of the issue. So to keep Production fed with tested boards (they have unique serial numbers so can't simply have their firmware loaded elsewhere) I started testing boards.
I found that when boards started to fail if I stopped testing for 30 mins or an hour when I came back boards would start passing again . . . for a while, then start failing again. By testing for a while, then pausing for a while then testing again I could continue without killing a MSP-FET. I repeated this process for a couple of days to prove consistency. No failed MSP-FET units . . . so I came to the conclusion that the MSP-FET was getting hot and if continued to be used would then become damaged, by pausing for some time I assumed the MSP-FET was cooling and then start working again. Happy in my assumption I designed and 3D printed a holder for the MSP-FET to get it a little elevated in its enclosure and then cut a hole in the front of the enclosure and added a fan and finger guard.
The MSP-FET no no loger fails at all and I have tested hundreds of boards with no significant pauses required. Happy days . . . I'm happy to take it on the chin that we have hurt these programmers, they are not dead, they are seen by MSP430Flasher, they can have their firmware updated, they just no longer see the target micro.
I can't in good conscience send the recent ones back to our supplier . . . what I'd rather do is fix them.
My ideal scenario from this post is that someong at ti takes my description of the fault and my fix and identifies what component/components has most likely been damaged, if it's replaceable and I can get replacement parts I can then fix all these MSP-FAT units and use them in other projects and spare test equipment that we will be building.
I'll remain optimistic for the moment 
Sorry for the long post, thanks for reading.
Simon G
