Watanabe-san,
As always, thanks for your comments and proposed changes. I am generally open to any proposed changes that helps move our process towards consensus; however in this case, I cannot accept
your view as depicted in your attached file and, while I already noted that the colors are somewhat subjective, I will maintain my original view as noted below:
I have no real issue with adding mechanical robustness to the Table above but I cannot agree with your unsubstantiated claim that GI-POF is better than glass optical fiber in tight
bending conditions. If we want to move toward consensus in our process, we need to back up claims with solid technical data. During the last review cycle, I accommodated your change to remove “A4i” as the GIPOF type with a follow-up question on what is being
proposed and if a specification was available. I received no response and it is not clear to me what GI-POF is now being considered for this application.
I have offered to test the GI-POF and OM3 against key optical, mechanical and environmental requirements but you will not supply samples for testing. I have been unsuccessful at this
point to even buy any A4i GI-POF on the open market for comparison. My OM3 fiber is available for anyone to test and I remain open to testing that fiber at Corning’s test laboratories or at a third party test facility.
Regarding mechanical robustness, I summarize below several studies conducted by Corning over the years to prove mechanical robustness. I can also present this material as a contribution
at a later meeting if needed.
Corning Incorporated has been manufacturing optical glass fiber for over 50. Corning has developed tools and models to help customers understand the impact on fiber’s mechanical reliability
in a wide variety of applications where stress and strain are present. The cornerstone of our mechanical reliability modeling effort is the ability to incorporate tension, bending, various lengths under stress, and multiple stress events into a single model.
The present discussion on fiber in tight bends fits well with our prior work in fiber mechanical reliability.
The square symbols in the Figure below are all the flaws below 560 kpsi on nearly 4000 kilometers of fiber proof tested to 100 kpsi. The probability of finding a flaw weaker than, say,
200 kpsi on a 1 meter length of this fiber is about 1x10-6.
Failure probabilities over a 25 year lifetime are very low.
Consider the following example of an installation with extreme fiber bending due to corners, storage, and staples;
• Twenty permanent 90-degree turns around corners and each bent to a radius of 5 mm. This is equivalent to 5 full 360° turns. Using 3 ppm per turn, the failure probability for corners is
15 ppm.
• Twenty full 360° storage turns at a 15 mm bend radii. This translates to a storage failure probability of 2 ppm (20x0.1).
• Finally, 70 staples causing the equivalent to approximately 3 full turns at 5 mm radius. This translates to a total staple failure probability of 9 ppm (3x3).
The total installed failure probability for this rather severe installation example is less than 30 ppm. One can conclude that risk of fatigue related failures due to bending in installations
is low. Now let us push this example further. Assume an entire hypothetical FTTH network consisting of 100,000 homes installed with the bending scenario described above. The predicted number of fatigue-related breaks is less than 3 for the entire FTTH network
over 25 years of operation.
Another example is our AOC which was designed to withstand multiple knots and sharp kinks during operation:
And finally, to characterize the design for its flexing ability, we designed and built the energy chain setup shown in the figure below. The moving part of a belt-driven linear
stage is such that the energy chain exhibits a 180° turn with a bending radius of only 25 mm. The travel of the stage was 350 mm and the setup ran with a cycle frequency of 1 Hz. The optical cable was fixed to the energy chain at the ends of the chain only
leaving it loose inside the chain, which represents a worst-case routing condition in terms of service durability. We stopped the testing after more than 10 million cycles without a fail. After this test, the cable passed continuous functional
testing:
To summarize, I believe OM3 glass is robust for the automotive application. I would like to ask if you can share similar data to support your claims that POF is superior to GOF in terms
of bend and tensile performance (breaks, deformation and attenuation increase), as well as furnish the multi-gig POF that you are proposing for this standard to be tested by a third party (or Corning if you like). We will gladly supply our fiber to an agreed
3rd party for test.
I will be asking the task force tomorrow to support the short wavelength VCSEL/OM3 PMD variant as the best option for moving our standard forward in a timely manner. I don’t believe
the other variants are quite ready at this time.
Best regards,
Steve
Steven E. Swanson
Senior Standards Manager
Distinguished Associate
Global Technology & Industry Standards
MT&E
Corning Optical Communications
4200 Corning Place
Charlotte, NC 28216
m 607-725-1129
swansonse@xxxxxxxxxxx
Standards are a bridge between markets and technologies; whoever controls the bridge controls the future…
-----Original Message-----
From: Yuji Watanabe <yuji.watanabe@xxxxxxx>
Sent: Friday, June 18, 2021 5:49 AM
To: STDS-802-3-OMEGA@xxxxxxxxxxxxxxxxx
Subject: [802.3_OMEGA] [EXTERNAL]--[802.3_OMEGA] Straw polls (Title is changed)
From: Yuji Watanabe <yuji.watanabe@xxxxxxx>
Sent: Friday, June 18, 2021 5:49 AM
To: STDS-802-3-OMEGA@xxxxxxxxxxxxxxxxx
Subject: [802.3_OMEGA] [EXTERNAL]--[802.3_OMEGA] Straw polls (Title is changed)
Hello Steve,
At the last Interim meeting, you explained that the merit of POF is connectivity and GI-POF loose this merit by reducing its core diameter to the same diameter as OM3.
Your point is correct from point of view of connectivity, however, GI-POF has another merit. GI-POF is harder to break than glass optical fiber in tight bending condition.
Because harnesses are used in order to install cables in automobile, optical fiber cables are combined with copper cables to make harnesses. This is different use condition from data center.
In this situation, risk of optical fiber break should be considered.
I understand that P.5 in your straw polls presentation is your view. I would like to present my view as attached.
I also added the new row of mechanical robustness as a point to be considered.
To TF members,
I would appreciate it if you would consider mechanical aspect when you vote.
Best regards,
Yuji Watanabe, AGC
-----Original Message-----
From: Swanson, Steven E <SwansonSE@xxxxxxxxxxx>
Sent: Friday, June 11, 2021 7:53 PM
To: 渡邊 勇仁(Watanabe Yuji)/AGC/事業開拓・インキュ <yuji.watanabe@xxxxxxx>
Subject: Re: [EXTERNAL]--[802.3_OMEGA] P802.3cz/D1.1 comments posted
Watanabe-san,
Thank you for your comments; I will make the corrections and re-post once I receive all comments.
Best regards,
Steve
Sent from my personal assistant
On Jun 11, 2021, at 5:58 AM, yuji.watanabe@xxxxxxx wrote:
Hello Steve,
My understanding is different form your PPT file in some points.
I would like to ask you below corrections before straw polls.
General: Replace “A4i POF” with “GI-POF”
Slide 2: Replace “VCSEL + A4i POF @ 850nm” with “VCSEL + GI-POF @ 850nm/980nm”
Slide 2: Replace “SiP + OM3 @ 1310nm” with “SiP + OM3/GI-POF @ 1310nm”
Best regards,
Yuji Watanabe
From: Swanson, Steven E <SwansonSE@xxxxxxxxxxx>
Sent: Thursday, June 10, 2021 11:17 PM
Subject: Re: [802.3_OMEGA] [EXTERNAL]--[802.3_OMEGA] P802.3cz/D1.1 comments posted
Sorry for the multiple ping; my link to the straw polls was an older version.
Please use this one:
Steve
Steven E. Swanson
Senior Standards Manager
Distinguished Associate
Global Technology & Industry Standards
MT&E
Corning Optical Communications
4200 Corning Place
Charlotte, NC 28216
m 607-725-1129
Standards are a bridge between markets and technologies; whoever controls the bridge controls the future…
From: Swanson, Steven E <SwansonSE@xxxxxxxxxxx<mailto:SwansonSE@xxxxxxxxxxx>>
Sent: Thursday, June 10, 2021 9:40 AM
Subject: Re: [802.3_OMEGA] [EXTERNAL]--[802.3_OMEGA] P802.3cz/D1.1 comments posted
802.3cz participants,
At our meeting on April 6, I presented the following contribution regarding PMD selection.
I noted that based on Straw Polls conducted at the IEEE 802.3cz meeting on March 17 that no consensus existed to select a PMD for baseline text. We are now 3 months down the road in the development
of our standard, and in my opinion, while we have made some progress, our inability to make progress on consensus on several topics threatens our timeline and the confidence our customers will have on the ability of IEEE 802.3cz to produce a timely multi-gig
optical automotive standard.
I also noted that current options that have had some discussion in our TF could lead to ~50 individual PMDs and in my opinion (again), that is NOT a standard.
We need to make some tough decisions (that is a requirement for developing a standard and our primary job as experts).
At our last meeting on June 8, I submitted a contribution with a list of Straw Polls and asked for your input on that contribution.
* Some are straight votes, some are proposed as “Chicago Rules”
* Is there support in how the polls are proposed?
* Is there support for the list of straw polls?
* Are there changes you would like to see to the straw polls?
* Are there additional straw polls that folks would like to see?
It is important that we get some get some good data on the preferences of the group; I purposely did not include an “abstain” or “need more information” as options because I really want people
to make a choice based on your expert opinion.
Picking one solution for the standard is not the end of the consideration for the other solutions. There are other avenues, e.g., MSAs or other options to recognize technologies.
IEEE is good at doing this; most times we get the selection right, some times we might be wrong. When we get it wrong, we have included other technologies at a later time based on new information
or market demand.
An example is our early work on data center standards:
Initially, we only standardized SR and LR. We now have SR, LR, FR and DR. FR and DR were not included initially based on several factors that then were addressed and included at a later point.
My personal view:
* We need to select a solution that supports our objectives and the CSDs
* We cannot pick them all
* We cannot afford to wait
Best regards,
Steve
Steven E. Swanson
Senior Standards Manager
Distinguished Associate
Global Technology & Industry Standards
MT&E
Corning Optical Communications
4200 Corning Place
Charlotte, NC 28216
m 607-725-1129
Standards are a bridge between markets and technologies; whoever controls the bridge controls the future…
From: ROBERT GROW <bobgrow@xxxxxxx<mailto:bobgrow@xxxxxxx>>
Sent: Tuesday, June 8, 2021 6:09 PM
Subject: [EXTERNAL]--[802.3_OMEGA] P802.3cz/D1.1 comments posted
Colleagues,
The comments received on P802.3cz/D1.1 with proposed responses have been posted to our Task Force web pages (https://www.ieee802.org/3/cz/comments/index.html).
Our Chief Editor, Mr. Torres, presented a report on the comments during our 8 June teleconference (Chief Editor's Report<https://www.ieee802.org/3/cz/public/8_jun_2021/CEReport_3cz_080621.pdf>).
This report also includes the plan for considering these comments, and explanations for the dispositions of some “buckets" of comments as groups. Mr. Torres has also indicated he would be pleased to receive suggestions for improvements to the proposed comment
responses (and encourages review of the "Grammar and syntax" bucket of comments for any improvement in proposed responses).
We will begin comment resolution during our 15 June teleconference where we will also set aside some time for presentations (we have three presentations we were not able to get to during our 8
June teleconference).
Robert M. (Bob) Grow
Chair, P802.3cz Multi-Gigabit Optical Automotive Ethernet (OMEGA) Task Force
________________________________
To unsubscribe from the STDS-802-3-OMEGA list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-OMEGA&A=1
________________________________
To unsubscribe from the STDS-802-3-OMEGA list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-OMEGA&A=1
________________________________
To unsubscribe from the STDS-802-3-OMEGA list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-OMEGA&A=1
________________________________________________________________________
To unsubscribe from the STDS-802-3-OMEGA list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-OMEGA&A=1