Thanks guys for the comments. Barry....it's definitly some form of 'latch up' which is occuring....meaning the power to the chip is being routed internally which 'holds' the circuit in a non operating state...like a relay in which the normally open contact is wire 'OR'd' to the coil...once the coil is activated, it latches itself in that state till the power is interrupted. The PIC is 'Nanowatt' technology, it will not actually heat up much, I don't think. Once this 'latch up' thing has occured, the PIC is unresponsive to any input, including a hard reset on the MCLR pin, which is supposed to force reset regardless (if enabled, which it is) Interrupting the power supply to the chip, and only that, will effect a restart. Joe...yes there were a couple of pins set as inputs without being tied anywhere, I sorted that but it made no difference. The interrupt service routine invoked at 1second intervals (by TMR1 ) runs happily for days, flashing the watchdog light...it has never frozen. It is only when the interrupt service routine attempts to decrement the 'down counter' register, the contents of which are then displayed on the LCD, that the problem occurs...and then only after (in microcmputer terms) a long time. The exact same problem occurs on a professional development board with a 40pin DIP Chip as opposed to the TQFP SMD device on the target system, making it unlikely that it is related to power integrity or 'hardware' What is confusing is the malfunction occurs completely at random, only when certain software routines are invoked, and results in the chip not responding to MCLR reset, which has never happened in my experience, till now :-) On Fri, Jan 25, 2013 at 10:05 AM, Barry Gershenfeld wro= te: > "Latch up" means something different to electrical engineers. If the > processor isn't overheated, then it's not "that kind" of latch up. But, > check for that, anyway. > > Most other ideas I have...interrupts shut off...executing a > SLEEP...executing off into space...these should all be handled by the > watchdog. > > Unless you get stuck inside the interrupt routine with the watchdog being > reset continuously. Otherwise, I can't think of anything else. > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 Perry Curling-Hope Research and Development --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .