usart pin transmits junk values after couple of correct transmission

[sanjib]

Posted on February 07, 2015 at 14:01

I have a piece of code of usart half duplex which was working fine with stm32f030 mounted on the pcb (designed as required for my project) for some time . I am giving external 5 volt supply as I am driving motor too. For controller I am regulating it to 3.3 volt . the command for motor moving send by another f030 controller through usart .

After sometime the Usart communication doesn't happen and it never happens again I  checked the Usart pin for checking what it is transmitting I found it gives some junk value(checked with docklight as well as scope) but it is working fine with the f030 evaluation board please help

[Tesla DeLorean]

Posted on February 07, 2015 at 17:14

And how to propose to have this remotely debugged?

Do you have annotated output from the scope or logic analyzer showing the junk values, ideally on the same time line where it works to the the point where it stops working, or fails? Do you have this all adequately grounded, and the signal wires shielded? What AWG is the wiring for the supplies (power and ground), how long is the cabling/distance between the boards? How well protected is the design to spikes and reverse EMF from the motors, etc. Is the serial connection using a coax, or differential pair? Does the robustness improve if you do?