program counter corruption by EMC

sandy wilson · Joined: 25 Feb 2004 Posts: 28

Hello Everybody,

I was reading in an EMC book that it is good practice to set all none used memory locations to a NOP command. If the unused blocks are continues place a reset command in the last location. This is so that if the program counter is corrupted by an EMC spike and jumps into an unused memory location it will enact the NOP and once it reaches the end of the block will reset the PIC.

Does anybody know how to do this and know of any other good EMC practices?

Thanks

Sandy Question

burnsy · Joined: 18 Oct 2003 Posts: 35 Location: Brisbane, Australia

One good tip is not to have the tracks on your MCLR pin too long. I had huge problems for a while at a customers premises. My project kept reseting at different times, which I eventually tracked down to my ICSP clip (and cable) attracting spikes from a great big noisy fluro nearby.

And yes I had decent caps to ground on the mclr pin. I have been careful from then on, about the length of the track from my processor to my icsp connector.

Also, I have that great big noisy fluro lamp as standard test equipment in my workshop now..
_________________
This is the last code change until its ready....

rwyoung · Joined: 12 Nov 2003 Posts: 563 Location: Lawrence, KS USA

If you bulk erase a PIC before programming it gets filled with something like 0x3ff. While that is not the NOP command I think it translates into "W=W+255" which is just as good as NOP
_________________
Rob Young
The Screw-Up Fairy may just visit you but he has crashed on my couch for the last month!