Sorry for OT, but I in big trouble!
From time to time my devices with above mentioned processors corrupt content of internal program memory. Does anyone see this behavior od 80C51ED2?
Dozsa Gyoergy
P.S. Keil C51 user since '91
Yes, I use reset IC. good, then we have to hunt for the remaining 1% of causes. Which reset IC, partnumber, please. Do you have a decoupling cap directly across Vcc and Gnd on each and every chip on the board?
This is an device in (my) production seven years, without significant modification. In the past two years sporandic errors are reported from customers, eg. flash corruption. is this certain (one) customer(s) or across the board?
I can't reproduce this error in lab, testing with more than 10,000 power on/off cycles. it seems nobody can.
Erik
Yes, RC resets are deadly to all forms of non-volatile data. But any chip with IAP capabilities can corruct the program in case the program runs awol, and the chip doesn't have a hardware-based lock support.
Have you seen any indications of watchdog resets or other troubles that may indicate that you may get a stack overflow or stack corruption? Are you using function pointers?
I can't reproduce this error in lab, testing with more than 10,000 power on/off cycles. [...] it seems nobody can.
Well, at least the usual laboratory power-supply on/off cycle often can't. It can take rather more "creative" power-cycles (bouncing switches, strange loads in parallel to the board, nearly dead batteries, ...) to hit the vulnerable spot of the system.