FW: [SI-LIST] : METASTABILITY IN FLIP FLOPS

Johnson, David (david.johnson@intel.com)
Thu, 15 Jul 1999 08:38:01 -0700

The below message apparently is in reference to Dr. Howard Johnson's 7/14/99
email newsletter, which probably goes to most SI-List members but not
necessarily all. My own understanding of metastability is that it's
fundamentally a statistical issue. There is no fixed, guaranteed time limit
on how long a flip-flop can stay in a metastable state; there is only a
probability distribution, just as there is no guarantee that flip-flop won't
even experience a hard failure or get hit by a cosmic particle that's strong
enough to "bump" it once it a very great while. What cascaded flip-flops
(or cascaded latches) can do, as Dr. Johnson's newsletter mentions, is
reduce the probability of a metastable state in the second flip-flop to a
level that is more nearly comparable to the probabilities of other "bad
things," such as hard failures.

It's interesting to experiment with just how long a real flip-flop in a real
lab can actually be "persuaded" to stay metastable when deliberately nudged
into such a state in a highly controlled manner. I vaguely recall that
Intel may have done such experiments a long time ago and succeeded, in an
extremely low-noise environment with extremely finely adjustable timing, in
getting a simple "74S" series flip-flop to stay metastable for four hours
(!) before it either snapped back to a valid state or the experiment was
terminated.

And yes, as Dr. Johnson also mentioned, a metastable state isn't always a
"steady" state; it can involve transient or even oscillatory behavior. What
flip-flop cascading relies on is sampling the first flip-flop's metastable
condition with an extremely narrow timing window in the second flip-flop,
such that the second flop isn't likely to go metastable unless the state of
the first flop just happens to be exactly right. In particular, the voltage
level that is likely to be generated by a "steady" type of metastable state
in the first flop usually won't be the level needed to make the second flop
go metastable. Of course, there is always a chance that the first flop will
snap out of its metastable condition just as its output is being sampled by
the second flop, but, hey, that's what statistics is all about (no absolute
guarantees, just reasonable likelihoods)!

---- Dave W Johnson, Intel

-----Original Message-----
From: Shayle Hirschman [mailto:shayle@mho.net]
Sent: Wednesday, July 14, 1999 8:56 AM
To: si-list@silab.eng.sun.com
Subject: [SI-LIST] : METASTABILITY IN FLIP FLOPS

Dr. Johnson,

Are you saying that using two flip flops in series with a clock whose
frequency is too fast to include the extra time allotted for meta state
resolution works successfully since the probability of the output of the
first flip flop changing during the vulnerable input time of the second
flip flop is very low?

That's an interesting idea which, if true, can enable the the clock
frequency to stay optimally high.

I had thought that the meta state of the first flip flop was an unsafe time
to clock into the second flip flop. But you say there aren't any
oscillations and that it will be a valid high or low during the meta state,
possibly changing again at the end of its meta state when it makes its
final decision.

Shayle

*************************************************************
Shayle I. Hirschman, Senior Engineer
Managing Director
Digital Design Solutions
http://www.digital-designs.com
shayle@mho.net
Phone 901/759-1802 Fax 901/759-2324

**** To unsubscribe from si-list: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list, for more help, put HELP. si-list archives are accessible at
http://www.qsl.net/wb6tpu/si-list ****

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****