I <em>might</em> have found the problem.

Right now I think it is that damn MCLR pin. I went back to a simple "blink" program and everything worked okay for a little bit then just stopped. Since this should be a "no brainer" program even for me I figured soemthing had to be wrong with the PIC.

I restarted it and everything worked fine again for a short time then died again.

I bumped the resistor on the MCLR pin and found the LEDs behaved badly. More moving on that resistor and I could affect the outcome.

So, I checked that whole area out and found there was no power to the resistor so I reseated all of that and things have been running fine now for 5 minutes.

I put back in my original program, but left it just blinking LED's without the multiplexer for the time being. Based on the output I am seeing through the LEDs the program is working as I wrote it.

Now, this raises all sorts of questions again about MCLR. I programmed it with MCLR=ON so there is no way it should have ran when there was no power at that pin, yet it did run for a short time. What's up with that? That pin will drive me nuts yet since it does not seem to behave the way everyone here says it should behave.

I still think I need to know how I incorporate some kind of reset in the program just in case it does crash. How do I do that?

Tomorrow I will wire the breadboard again to include the multiplexer. That should ramp up the heat on the regulator if it continues to behave as it has been. That will be the real test if it can run for 10 hours without burning out or crashing. The way things have been going I don't think I want to hold my breath.

Bart