Metastability data

[Altera_Forum]

Where can I find more information on metastability for Cyclone or Stratix devices?

[Altera_Forum]

Metastability data is not available for the later devices. This is because of improvements in the process technology of newer device families reduce metastability so this info is no longer needed.

[Altera_Forum]

So you're telling me Altera devices can never go metastable? If there is even a slight possibility, shouldn't they issue a report with data?

[Altera_Forum]

Hello fellow Ron, 

 Altera has data for the older generation families - FLEX 10K, etc. which I believe are on 0.50 micron TSMC process. The FIT data from this can be used as a ceiling for any new generation families you may use. The metastability rates diminish as process transistors shrink.   -Ron

[Altera_Forum]

I found application note 42 that has the metastability info for older Altera devices at  

http://www.altera.com/literature/an/an042.pdf

[Altera_Forum]

Dr Evil here... Long time listener, first time caller... 

 So I'm involved in, lets say "military" projects. I have some very tight MTBF requirements that I must meet. Are you telling me Altera will not provide me with metastability data to ensure my "laser" works correctly?     --- Quote Start ---  Hello fellow Ron,  Altera has data for the older generation families - FLEX 10K, etc. which I believe are on 0.50 micron TSMC process. The FIT data from this can be used as a ceiling for any new generation families you may use. The metastability rates diminish as process transistors shrink.   -Ron  --- Quote End ---  

[Altera_Forum]

As Ron stated, Altera will provide you with the metastability data of the Flex10K family. Specs for any newer device are better than what they describe in an42 (http://www.altera.com/literature/an/an042.pdf). So, as long as you meet the MTBF specs described in that document you'll be A-O-K. 

 I guess that if you want to be living on the performance edge AND want to meet tight MTBFs I suggest you send a fat check to the characterization team to do the research for you, and another check to their management to re-prioritze this project to go to the top of the list ;-)

[Altera_Forum]

Well... with that kind of logic, I should use 10K timing characterization data. And if I can meet it, I should be fine with my Stratix II designs. 

  Hmmm... Compelling approach! I love it!

[Altera_Forum]

Gentlemen, 

 Despite Dr. Evil's campy banter. He does have a point. I do a fair amount of work with military subcontractors and to be compliant MIL-STD 1629A or MIL-STD 756 or MIL-STD 1388 you need the metastability data to calculate MTBF.   So question isn't who to write the check to but rather how much business does Altera want to lose! Right now your competition will supply the metastbility data.    MM

[Altera_Forum]

This came up in a search now that Altera searches the forums... So I thought I'd post an update in case anyone is looking for current info! 

 Instead of publishing the metastability constant values, Quartus II can now report a design-specific MTBF for newer families, which seems even more useful (with certainly less math required!). See handbook chapter for details: managing metastability with the quartus ii software (http://www.altera.com/literature/hb/qts/qts_qii51018.pdf)