[email protected] said: > One interesting note however. Years ago we had a standard old 4040 ripple > counter in our shop that displayed a low occurrence of jitter of several > times it's input frequency period at it's lowest frequency output (Sort of > what you are describing below). I wish I had the numbers handy, but the > output would be good for most of the time, then every once in a while it > would jump to a longer delay. It was hard to catch with a scope, but when > we measured every single pulse width it showed up fairly well. The high > speed clock (A TTL OSC in a can) never skipped, as far as we know. We never > did figure that one out. From what I remember we switched IC manufacturers, > and the problem when away.
I expect ever old timer who did serious glitch chasing 25-30 years ago has similar stories. Mine involved a dual port RAM, 16x4. I think it was a 74S189. It was made by many companies. Chips from one vendor just occasionally screwed up. We were using them in FIFOs on ethernet boards. (They were probably used other other places too. I was worked on networking.) Back then, most large electronics companies had incoming inspection and qualification groups. I remember stories of somebody contacting that group for help and getting a response of roughly "That's why we didn't qualify them." (We were a research group. We purchased small batches of chips wherever we could get them without going through the official qualified list.) -- These are my opinions. I hate spam. _______________________________________________ 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.
