USBDM 4.10.5 debug problem with CW10.3

[lucianon]

Hi,

Debugging the project USBDM_JMxx_V4_10 with an USBDM and CW10.3, when run "bdm_setTargetVdd()" the line VDD3_ON() makes it to jump to myStartup on "ICP.c". If I debug it with P&EMultilink it runs correctly.

I thied to isolate the function VDD3_ON() on main() and it cause the same wrong jump. What is causing this jump to a wrong address? It can have a stack overflow.

Luciano

[pgo · Accepted solution]

Hi Luciano,

No real way to tell (from here):

How is the target being powered?  It might be the case that turning on the Vdd is causing a voltage spike that is resetting the target.  I don't know why it should be different with the two BDMs.

Another thing to try would be to single step at the assembly level and isolate which instruction it jump at.

bye

[lucianon]

PGO, you are right!  There is no bug on USBDM, I know this old JM problem.  If I debug on assembly step level it works because it turn one bit at a time decreasing the voltage spike and EMI thought RESET pin.

Let me tell you my experience with JM families:

I have a lot of product with JM and I already have a lot of problems with the reset pin caused by noise. Freescale fail when designed this JM. Some products the solution was cut off  the pcb track or cut the reset pin to stop reseting by EMI. I know this JM chips are the most EMI vulnerable I ever known. There is no filters enough that can eliminate a reset by high frequency noise transient at this reset pin. The reset track works as an antenna. It works much better if you don´t connect the reset pin to any track. I only solved this problem when I decide to not use this pin anymore.

Do NOT use any RC reset circuitry recommended by manual, do NOT connect any track on Reset pin with JM families and you will see it working much better without accidentally reset events.

thank you for help me to understand this.

Luciano