Passive probing of JTAG programming of a uC

  • Thread starter Thread starter berkeman
  • Start date Start date
  • Tags Tags
    Programming
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
Sign up Log in
5 replies · 2K views
berkeman
Admin
Messages
69,781
Reaction score
25,629
I'm working on an old test fixture that uses JTAG to program a microcontroller (uC) and its Flash memory chip on production PC Board Assemblies. The programming works fine with an old version of the PCBAs, but for some reason it is failing with a new version of the PCBA that should have had no changes to the JTAG connections, uC or Flash memory chip circuitry.

I'd like to passively monitor the JTAG programming stream to compare the passing and failing datastreams. Does anybody know a good tool or Pod or other way to do this? I've used an old HP logic analyzer in the past to do such probing for short datastreams, but the serial-to-parallel conversion is a bit of a pain for a vanilla tool without some formatting capability.

The JTAG programming Pod I'm using is the TI MSP-FETU430IF (and the Gang programming equivalent), and the uC is the TI MSP430F5437A with external serial Flash memory. I do have a TotalPhase Beagle SCI/SPI Serial Analyzer Pod available, if there is some way to adapt it to JTAG...

Thanks for any ideas.

https://msharmavikram.files.wordpress.com/2012/03/msp-fet430uif.jpg
msp-fet430uif.jpg
 
Engineering news on Phys.org

Discussion

berkeman said:
The programming works fine with an old version of the PCBAs, but for some reason it is failing with a new version of the PCBA that should have had no changes to the JTAG connections, uC or Flash memory chip circuitry.

Does it fail all attempts to program them? If so I'd want someone to check the PCB artwork for unintended changes in those areas.

In the distant past I've had issues with changes to decoupling capacitors affecting programming.
 
I am able to program the new devices just fine at my desk using my TI programming Pod. But the programming fails in the Mfg Test Fixture with an error complaining that the device uC JTAG ID is wrong (which I know is not true from my testing at my desk). I suspect some tiny misalignment of the pogo pins in the fixture and the new layout, but visually the connections look good.
 
I've used a Saleae for this. Their software is amazing. They have a bunch of protocols already but it is very easy to add your own as well.

https://support.saleae.com/hc/en-us/articles/208666936-Joint-Test-Action-Group-JTAG-
https://support.saleae.com/hc/en-us/sections/201990583-Supported-Protocols
https://www.saleae.com
 
Last edited:
  • Like
Likes   Reactions: berkeman