Hi Hal, >First, we understand metastability. We can measure it and predict it. It it >mattered, we could test each individual part. That was exactly my original point. The discussion got off topic, I just wanted to point out that it is a matter of probability, and the math to calculate this probability of failure is well known and understood. The math says that you cannot prevent a metastable event, even if there are 1000 flip flops. The probability is never zero. And since the processes are stochastic, the first failure could happen within 2 seconds of power on, even with 10000, or a million cascaded flip flops. You can never prevent a metastable event going through all of the flops. You can just make that probability approach zero. Actually, if you had a million cascaded flip flops one of them would probably fail due to other reliability reasons :) >Second, the failure mode is exponential in a parameter we can control. So >given a particular set of parts to pick from, it's reasonable to make a >design with a probability of error small enough so that other things are much >more important. Right! My original point again. The question was: how often do crystal jumps happen. How often do metastable events happen. Both may be just a matter of statistics and probability. If we don't see a crystal jump within say one month of operation, we can probably say the probability of one happening is extremely low. But I have seen jumps happen after 3 months of continuous, documented operation without any jumps, so again we may never have certainty that jumps/metastable events will not happen. We just have to make them happen so seldomly that they won't matter to the application at hand. But again if it jumps after 3 months, no one can guarantee it won't jump after 10 minutes. bye, Said
____________________________________ From: SAIDJACK To: SAIDJACK Sent: 10/16/2008 16:43:33 Pacific Daylight Time Subj: Fwd: [time-nuts] Frequency Stability of Trimble Mini-T >From iPhone Begin forwarded message: From: Hal Murray <[EMAIL PROTECTED] (mailto:[EMAIL PROTECTED]) > Date: October 16, 2008 15:40:12 PDT To: Discussion of precise time and frequency measurement <[EMAIL PROTECTED] (mailto:[email protected]) > Subject: Re: [time-nuts] Frequency Stability of Trimble Mini-T Reply-To: Discussion of precise time and frequency measurement <[EMAIL PROTECTED] (mailto:[email protected]) > We have done some extensive research trying to correlate jumps to external phenomena (radiation, vibration, thermal effects, etc). There is no straight forward correlation. It's similar to asynchronous switching inside a digital computer. You can add levels of flip flops to synchronize across two asynchronous time-domains, ... I think that's misleading in two ways. First, we understand metastability. We can measure it and predict it. It it mattered, we could test each individual part. Second, the failure mode is exponential in a parameter we can control. So given a particular set of parts to pick from, it's reasonable to make a design with a probability of error small enough so that other things are much more important. If it absolutely, positively can't take any hit, then some more work is involved I would say that the first step is to put a number on "absolutely, positively". If you aren't willing to take some risk, you won't get off the drawing board. -- These are my opinions, not necessarily my employer's. I hate spam. _______________________________________________ time-nuts mailing list -- [EMAIL PROTECTED] (mailto:[email protected]) To unsubscribe, go to _https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts_ (https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts) and follow the instructions there. = **************New MapQuest Local shows what's happening at your destination. Dining, Movies, Events, News & more. Try it out! (http://local.mapquest.com/?ncid=emlcntnew00000001) _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
