- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Hi Team,
Our company uses TI AM3352 MCU as the main control chip of our equipment. The problem we are encountering now is that after the devices are started normally, the chip will have a reset if the devices communicate with external devices or host computer software through the network port. Not all devices have this problem. For example, if 100 devices are produced, there may be 2-5 devices with this issue.
Troubleshooting process:
1) Software troubleshooting
At first, I suspected that the code itself had uncontrollable problems such as memory overflow. For the troubleshooting, I conducted the coverity scan and modified the places that may involve memory overflow. Although all serious warnings had been resolved, there were still reset-related problem. I conducted the code version management and used the program of March 2019 to search, but there was still a reset problem. I cut the code to the minimum system, leaving only some functions of the network port. And all other threads were blocked with all reading and writing operations of 'nor' and 'nand' disabled. There were still reset issues.
2) Hardware troubleshooting
I tested the power-on sequences of nand, nor, ddr, etc., and no abnormal waveform was found. As for the replacement of the devices, the PHY chip, nand, etc. were replaced, and the reset problem still existed. Finally, a main chip was replaced by a new one, and the reset issue disappeared.
3) Pressure test
Pressure tests of DDR reading and writing operations were conducted, but no abnormalities were found in the devices with the problems. The loopback test was performed on some registers of the main chip, and the setting was not successful, and it failed to proceed normally.
4) Performing cross-validation
I used the main chip of the problematic device on a non-problematic device and the network port communication of the non-problematic device started to embrace a reset.
So far, due to the limitations of my technical level and means, I have not been able to confirm the root cause. I posted this issue for technical support. The pictures attached are the main chip's power-on waveform and internal version diagrams. The PMU used TI's TPS65217CRSLT chip.
Kind regards,
Katherine
