[SI-LIST] Re: Do you really ship products at BER 10e-xx ?
- From: "Tom Waschura" <tom_waschura@xxxxxxxxxxxxxxxxxxxxx>
- To: "'steve weir'" <weirsi@xxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Tue, 12 Apr 2005 19:14:56 -0700
Steve,
Naturally, all of my discussions have to do with the signal integrity side
of the link rather than your good point about what "upsets serial links for
substantial periods". We're being pressed to measure bit error rates to
10E-15 and at those rates you start worrying more about nearby lighting
strikes and power-outages rather than the RMS noise on a rail.
The s-parameter model approach (e.g. STATEYE.ORG) is fascinating and would
address the linear frequency domain impacts of what was measured (e.g. you
channel). If properly convolved with a model for your transmitter, traces,
launches, pre-compensation, equalization--both intended and not--then you'd
have what your eye might look like at the receiver--by the way, still not
what a digital decision circuit would make of the eye at the receiver--note
the non-linear effects of hysteresis, etc. This has been used a lot in CEI
and 802.3ap to compare and bound channels but I think the final real BER
number would not be taken this way--it would be taken with a measurement.
Tom
-----Original Message-----
From: steve weir [mailto:weirsi@xxxxxxxxxx]
Sent: Tuesday, April 12, 2005 6:30 PM
To: Tom Waschura; si-list@xxxxxxxxxxxxx
Subject: Re: [SI-LIST] Re: Do you really ship products at BER 10e-xx ?
Tom, Dr. Demming would be proud to hear what you have said. If we have
little random variation then almost any sample will provide a reliable
prediction of performance. OTOH...
I think that this goes to Andy's problem. The things that tend to upset
serial links for substantial periods are those that knock out the
PLLs: Reference clock stability, power supply hits, or EFT / ESD
hits. It might be that there is a resonance in the channel. However,
rather than trying to use random patterns to find the effects, I would
prefer to use the S parameter data to locate the peaks and notches and then
issue patterns that excite them.
Regards,
Steve
At 03:57 PM 4/12/2005 -0700, Tom Waschura wrote:
>Many failure types I've seen fall into the category of causing terrible
>error rates, true; however, many others fall into a more subtle
>category--and these are typically much more difficult to diagnose. In
>general, more random jitter is worse as these PDFs have infinite tails.
Any
>frequency null or notch can cause pattern dependent resonances which might
>convolve with other circumstances to cause bit errors. If it takes a PN31
>pattern to stimulate something, the bit error rate might be 1E-9 just based
>on the pattern length. Passing a go/no-go BER test does not give anyone
>insight into what type of bit error is lurking.
>
>Bit error rate contours may help the notion of "waiting for 10E-12
>measurements". This allows a predicted result based on considerably higher
>measurement data. These predictions only follow the random components of
>jitter and would want to be based on real measurement depth at least past
>the test pattern period.
>
>Anyway. Having solved bit error problems in links, components and systems
>for years; I'd hesitate from saying all I needed was a go/no-go test based
>on some rough "terrible BER". This would keep you moving, but your scrap
>pile would grow. I'd say that even just separating out random,
>data-dependent and periodic bit errors in production failures along with
>someone who can diagnose based on this information is a significantly
>net-positive benefit.
>
>Tom
>
>-----Original Message-----
>From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
>Behalf Of Chris Cheng
>Sent: Tuesday, April 12, 2005 2:47 PM
>To: si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Re: Do you really ship products at BER 10e-xx ?
>
>I have heard similar case from a third party and the customer service
>engineer starts to explain, "of course, its cosmic rays" with a straight
>face. I can't say how many people remains sitting on their chairs and not
>flipping over after that.
>But it is not my personal experience so that's why I am curious.
>Thanks for sharing though.
>
>-----Original Message-----
>From: Henson, Bradley S [mailto:Bradley.S.Henson@xxxxxxxxxx]
>Sent: Tuesday, April 12, 2005 2:35 PM
>To: Chris.Cheng@xxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
>Subject: RE: [SI-LIST] Do you really ship products at BER 10e-xx ?
>
>
>This could make an interesting topic. I have to say that in general, I
>have noticed the same trend: Links work so well the BER is hard to
>determine (lots of test time or link-stress)-or- the links are totally
>messed up. However, I did get called in to troubleshoot a Fibre channel
>application that was just marginal on some of the links. By that I mean
>they would almost make the spec 1E-12 BER sometimes, but usually fell
>short. Some days they operated considerably poorer than 1E-12, but not
>pure garbage.
>
>-----Original Message-----
>From: Chris Cheng [mailto:Chris.Cheng@xxxxxxxxxxxx]
>Sent: Tuesday, April 12, 2005 1:49 PM
>To: si-list@xxxxxxxxxxxxx
>Subject: [SI-LIST] Do you really ship products at BER 10e-xx ?
>
>
>I've been shipping Gb/s serial products for a while and have my share of
>fail parts. However, I have yet to see a physical channel that is not
>either working like a charm or just fall on its face and barfing errors
>like crazy. Sure, chips or disk can fail and generates errors but no
>flaky channels that spits an error every other hour or days. To me, the
>channel is either have a BER that is near 1 (barfing errors like crazy)
>or near 0 (never fail, or at least approaching the life of the product
>it is attached to).
>Are we just kidding ourselves with these fancy BER analyzers or jitter
>instruments ? Do you really let a machine runs at say BER 10e-12 and say
>"ah ha, it only fails once a day and let's ship it" ? Is BER really
>meant for IEEE spec committees and not for real engineers who actually
>have to ship a product ?
>------------------------------------------------------------------
>To unsubscribe from si-list:
>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
>or to administer your membership from a web page, go to:
>//www.freelists.org/webpage/si-list
>
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>List FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
>List technical documents are available at:
> http://www.si-list.org
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
>Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>------------------------------------------------------------------
>To unsubscribe from si-list:
>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
>or to administer your membership from a web page, go to:
>//www.freelists.org/webpage/si-list
>
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>List FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
>List technical documents are available at:
> http://www.si-list.org
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
>Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
>
>------------------------------------------------------------------
>To unsubscribe from si-list:
>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
>or to administer your membership from a web page, go to:
>//www.freelists.org/webpage/si-list
>
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>List FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
>List technical documents are available at:
> http://www.si-list.org
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
>Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
