??? 05/24/11 07:40 Read: times Msg Score: +1 +1 Good Answer/Helpful |
#182354 - Tolerances Responding to: ???'s previous message |
Not all chips are identical.
There are always production variances. That is the reason why the datasheets contains things like minimum, typical and maximum values for timings, currents, voltages, temperatures, ... So when you have a design where four chips works and one chip doesn't work, you should be worried. It may be a broken chip. But only a fool would just assume that. A careful person would instead worry that the hw or sw may push the envelope, having one or more parameters outside that specification. The manufacturer only promises that the chip will behave as expected when all - yes every single one of them - of the parameters are within the specifications. As soon as ony one of them are off, you get situations where some chip fails and some chip doesn't fail. Or all devices may seem to work until you change another parameter (even within the allowed range). So you may find that having a higher or lower temperature or a lower voltage or a slightly slower rise time on the supply voltage would suddenly make one or even all four of your remaining chips also fail. Basically, it's time for you to start ticking off every single parameter in the datasheet and make sure that you really do have your design up to spec. |
Topic | Author | Date |
programs not working in 89s51 | 01/01/70 00:00 | |
Debugging time! | 01/01/70 00:00 | |
Tip | 01/01/70 00:00 | |
migration document | 01/01/70 00:00 | |
Up Counter | 01/01/70 00:00 | |
And what do you deduce from that? | 01/01/70 00:00 | |
read or write | 01/01/70 00:00 | |
Have you really read the datasheets?? | 01/01/70 00:00 | |
. | 01/01/70 00:00 | |
Don't be too hasty to blame the ICs... | 01/01/70 00:00 | |
an answer in the form of an anecdote | 01/01/70 00:00 | |
.. | 01/01/70 00:00 | |
Yes, you are! | 01/01/70 00:00 | |
Tolerances | 01/01/70 00:00 | |
It will be embarassing | 01/01/70 00:00 | |
Thanks friends... | 01/01/70 00:00 |