| ??? 12/12/07 17:11 Modified: 12/12/07 18:08 Read: times Msg Score: +1 +1 Good Answer/Helpful |
#148106 - Even not in theory, of course! Responding to: ???'s previous message |
Whenever the die is heated (even only during the manufacturing) or other energy is brought into the crystal (programming of flash, EEPROM, etc.), damages occur. Due to the Brownian motion, all kinds of impurities, deffects and even doping gradients have the tendency to turn the very high degree of order of such a die into total chaos. It's only a question of time and as Steve already told, temperature. Remember the Arrhenius equation.
But when we talk about a flash microcontroller, then things are even worse. That the datasheet states "1,000 write/erase cycles" should give you an idea how stressful even only the original programming is. The following datasheet snippet shows the data retention time versus temperature of microcontroller ST2660C (in the absence of data about AT89C2051): 
You can clearly see, that an indefinite lifetime is entirely unrealistic, even at room temperature. Take note, that what you see above is statistical data giving information only about some sort of average of a huge (indefinite) number of lots. How your very individual chip behaves is written in the stars. Christoph said:
Flash memory for example has a guaranteed data rentention period (usually a decade or more, but still finite). Where have you found this specification in the datasheet of AT89C2051?? Kai |
| Topic | Author | Date |
| Life expectancy for 89c2051 | 01/01/70 00:00 | |
| Depends on how cruel you get with your design | 01/01/70 00:00 | |
| No wear mechanism | 01/01/70 00:00 | |
| In theory. | 01/01/70 00:00 | |
| one more issue | 01/01/70 00:00 | |
| Even not in theory, of course! | 01/01/70 00:00 | |
| Not in the datasheet. | 01/01/70 00:00 | |
| datasheet is not the only source of information | 01/01/70 00:00 | |
| FITs, MTTF, data retention bake | 01/01/70 00:00 | |
| Life | 01/01/70 00:00 | |
| Sandy have a look at this , it could give you some | 01/01/70 00:00 | |
Life expectancy for 89c2051 | 01/01/70 00:00 |