Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: [EFM] Moving forward on extended temperature range optics




Howard,
I agree with Jonathan that an optional PIC would be a task for the group.
Its seems a reasonable compromise.

I want to describe the reasons why hardening is a requirement in response to
your email asking for feedback.

An inside ONT is not viable for FTTH for the existing residences in the USA.
Would subscribers want to allow someone to drill a hole in your outside
wall, run fiber and put a box in your closet, then run the metallic wire
back out to your NID? Since we are all engineers we would probably say that
we would do it just to have the privilege to be a first adopter and the
chance to pay more for higher speeds, more channels. Most people would not
want to do this work.  In addition, when it fails and you need your service
back, you allow the service tech to arrange a visit to fix it and you stay
at home for a certain time and meet the service tech to let them in your
house. Or, you can have an outside ONT and the service tech stays outside
and fixes the box for you to get your service up as fast as possible whether
you are at home at the time.  

A controlled environmental vault for housing an ONU is really only
cost-effective for a thousand homes not for a neighborhood of 200 or a
street of 16 homes.  You can choose to cheapen the optics by not hardening,
but it still destroys the economics for the operators for getting fiber
deeper.  I would bet hardened optics  would not cost more than drilling
holes in walls or digging holes in the neighborhood.

Labor always will go up while equipment prices typically go down over time.
Hardening optics makes sense compared to more labor today and into the
future.

What about reliability? If you want to put telephony on fiber deeper, the
vendor better qualify parts that withstand extended temperature over long
periods of time. The enterprise market for Ethernet NICS can accommodate a
lifetime of 3 years since speeds increase, not so for access. The naysayers
on FTTH say that putting optics at every home is setting operators up for a
maintenance nightmare compared to a metallic network termination. While we
would save maintenance cost by not having to fix the copper when it rains,
but it may get all the cost back replacing failed optics at the ONT.
Advocates for FTTcurb suggest its better to have a node with only one
expensive/hardened optical PMD shared among many homes with a protected
fiber link to deal with optics that fail a lot in the outside plant.  

I have a growing concern about how feasible the current 802.3ah draft
optical specifications are today over a extended temperature range. Can FTTH
optics be qualified and hardened while still keeping cost reasonable?
Nothing I have heard in this discussion leads me to believe companies who
make optics for the enterprise Ethernet market are eager to come to the
plate and help with compliance to an extended range. I suggest, making an
attempt at compliance testing and writing specs to allow cost reduction is
one giant step toward feasibility for our specs. If we confront extended
temperature and work thru compliance tests, we may find out now that we need
to loosen specifications and the promise of EFM can be strengthened.  

Hardening is really a solid requirement.  When IEEE said they would do
hardened optics for EFM, I took notice and told others that these engineers
are going to really address the optical access market. I observe that taking
on temperature  that has not been done in the enterprise specs to date shows
a serious dedication to addressing a new access market.  I think those in
the group have the expertise to take this issue head on.  I personally don't
know what other standards group would be better to work the issue in a
liaison.

I have seen this group rule many topics out of scope over 2 years.  I am not
going to support weakening extended temperature objectives.  The 802.3ah
participants will be famous if specs developed can break the cost barrier
for hardened PMD's that is required for FTTH.  The low-cost reputation of
Ethernet enterprise ports could then really be extensible to the access
market.  

-Kent

> -----Original Message-----
> From: Jonathan Thatcher 
> [mailto:Jonathan.Thatcher@xxxxxxxxxxxxxxxxxxxx] 
> Sent: Thursday, January 23, 2003 12:56 PM
> To: Howard Frazier
> Cc: stds-802-3-efm@ieee.org
> Subject: RE: [EFM] Moving forward on extended temperature range optics
> 
> 
> 
> Howard,
> 
> I think that there is one additional essential task:
> 
> C) Provide an ***optional*** PIC for each PMD indicating 
> operation over the "extended temperature range."
> 
> jonathan
> 
> | -----Original Message-----
> | From: Howard Frazier [mailto:millardo@xxxxxxxxxxxxxxxxxx]
> | Sent: Thursday, January 23, 2003 9:19 AM
> | To: stds-802-3-efm@ieee.org
> | Subject: Re: [EFM] Moving forward on extended temperature 
> range optics
> | 
> | 
> | 
> | 
> | I think that the essential tasks are to:
> | 
> | A) Ensure that all of the Active Optical Input and Active
> | Optical Output 
> | Interface
> | parameters in clauses 58-60 can be met, and the corresponding links 
> | function
> | properly, across an "extended temperature range" of operation.
> | 
> | B) Define what an "extended temperature range" is, and place this
> | definition in
> | an informative annex (Annex 66A) of P802.3ah.
> | 
> | If we can do this, we will have satisfied our objectives and
> | all of our 
> | prior
> | motions on the subject, according to my interpretation.
> | 
> | I believe that we are prepared to do this, and we should do this,
> | without further
> | delay.  We will then have a follow on task to prove that optical 
> | components and
> | links can simultaneously satisfy A and B above, and meet the 
> | 5 criteria.
> | 
> | We are past the point of deciding "what we are going to do".
> | Our job is to
> | carry out our decisions, and to prove that we have done so to the 
> | satisfaction
> | of our Working Group and our Sponsor.
> | 
> | Howard Frazier
> | Chair, IEEE 802.3ah EFM Task Force
> | 
> | Bruce Tolley wrote:
> | 
> | >
> | > Piers
> | >
> | > I am not exactly sure why you felt compelled to disagree
> | with what was
> | > essentially an invitation to a meeting, but here goes
> | >
> | > 1) Network operator requirements
> | > Yes not all the network operators from every region of the
> | world are
> | > coming to our meetings, but I think it speaks to broad market
> | > potential to listen to the customers who care enough to come and 
> | > participate in the debate.
> | >
> | > Yes, network operators want all kinds of things and often 
> different
> | > things, but we are discussing optical PMDs across extended 
> | temperature
> | > here. Let's not cloud the issue. We do not have goals to
> | define fire
> | > safety or 48V DC power over fiber optic cabling.
> | >
> | > 2) Scope
> | > We have a goal that defines the scope. Just because we have
> | not done
> | > things in the past, does not mean we cannot do it in this
> | project if
> | > it is within our charter as defined by the PAR and our objectives.
> | > 802.3 never worked on an electrical power spec and it is now 
> | > completing (rapidly I hope) the  DTE power.
> | > We define many interfaces and performance parameters in our 
> | documents
> | > some of which are not exposed as external interfaces to end
> | customers
> | > or testable by end customers.
> | >
> | > 3) Interpretation of Past motions
> | > The thread of motions shows that we are trying to fulfill the
> | > objective but we are not quite sure of the path to success.  My 
> | > personal opinion is that if the extended temperature ranges 
> | are only
> | > informative, we will not be fulfilling the objective. Some
> | good work
> | > has gone into the draft, but we still have some real
> | technical work to
> | > do. The Task Force voted down the motion that said P802.3ah would
> | > define two sets of optical PMDs but gave us no clear 
> | direction on how
> | > to move forward.
> | >
> | > Thanks
> | >
> | > Bruce
> | >
> | > At 03:54 PM 1/23/2003 +0100, piers_dawe@xxxxxxxxxxx wrote:
> | >
> | >> Bruce, Brian and Richard,
> | >>
> | >> I'd like to point out where this chain of thought goes wrong,
> | >> especially as the logical disconnects have been repeated 
> | later in the
> | >> thread.  See below:
> | >>
> | >> > -----Original Message-----
> | >> > From: Bruce Tolley [mailto:btolley@xxxxxxxxx]
> | >> > Sent: 16 January 2003 19:59
> | >> > To: piers_dawe@agilent.com; stds-802-3-efm@ieee.org
> | >> > Subject: [EFM] Moving forward on extended temperature
> | range optics
> | >> >
> | >> > Piers and all
> | >> >
> | >> > I gave my self the action to help move forward the
> | outstanding issue
> | >> > regarding extended temperature ranges for P2P and P2MP optics
> | >> >
> | >> > We have an objective to include in our specification of PHYs, 
> | >> > support for extended temperature range optics
> | >>
> | >> Yes.  Support, not mandatory requirement.
> | >>
> | >> > The task force has in the past passed motions to specify EFM 
> | >> > optics at -40 to +85 C
> | >>
> | >> To specify the optics, not the temperature.  This is very
> | clear from
> | >> the January and March 2002 motions.  See e.g. the March
> | Joiner motion
> | >> "The basis for the first draft of the 802.3ah 
> 1000Base-LX extended
> | >> temperature objective be met with text that uses 
> 1000Base-LX 5 km 
> | >> single mode specification (clause 38) as the starting 
> | point with the
> | >> following changes and additions:
> | >> - Informative temperature range -40-+ 85 deg C
> | >> etc
> | >>
> | >> and
> | >>
> | >> January Motion #11
> | >> Motion: to create informative annex to address environmental
> | >> considerations.
> | >> Mover: Chris DiMinico
> | >> Second: Alan Flatman
> | >>
> | >> > Network operators have on multiple occasions communicated the 
> | >> > requirement for extended temperature solutions.
> | >>
> | >> This is where the logic really falls apart.
> | >>
> | >> First, is it not just a small subset of network operators
> | (US ones)
> | >> who aren't installing much FTTB.  Other network 
> operators may have
> | >> different physical strategies, climates, and requirements.
> | >>
> | >> Second, network operators need many things; working capital, fire
> | >> safety, an electricity supply...  It does not follow 
> that 802.3 is 
> | >> bound to provide any of them.  Environmental 
> requirements such as 
> | >> these are out of scope of this standard - that's why 
> | temperature is
> | >> to be addressed in an informative annex.
> | >>
> | >> Of course, customers will impose environmental
> | requirements in their
> | >> procurement specs - and Telcordia specs for example are
> | effectively,
> | >> procurement specs.
> | >>
> | >> > As recently as the Vancouver meeting, several box vendors 
> | >> > (including me)b communicated the requirement for extended 
> | >> > temperature range optics.
> | >>
> | >> Same lack of connected logic.  Someone's need doesn't mean
> | that 802.3
> | >> is bound to supply.  Temperature specs are available in 
> the market
> | >> from other sources, who have more expertise in the matter.
> | >>
> | >> Piers
> | >>
> | >> > We need
> | >> > to agree on a path to move forward.
> | >> >
> | >> > So if interested parties want to forward to me their email 
> | >> > addresses, I will host a conference call next week 
> dedicated to 
> | >> > this issue. I think we
> | >> > need to focus on a test specified in each PMD clause, to 
> | agree on the
> | >> > ranges for OLT and ONU optics, to consider the 
> possible special 
> | >> > case of bidis that include 1550 nm DFBs, and to 
> identify any PMD 
> | >> > that might only
> | >> > need to be supported at standard, commercial temperatures.
> | >> >
> | >> > thanks
> | >> >
> | >> > Bruce Tolley
> | >> > Cisco Systems
> | >> >
> | >> >   At 04:11 PM 1/8/2003 +0100, piers_dawe@xxxxxxxxxxx wrote:
> | >> >
> | >> > >G.983.3 refers to ETS 300 019.  This is a very readable
> | series of
> | >> > >documents from the European Telecommunications Standards
> | >> > Institute, giving
> | >> > >a classification of environmental conditions, e.g.
> | weatherprotected
> | >> > >locations, non-weatherprotected, underground.  It uses four
> | >> > classes of
> | >> > >climatic conditions:
> | >> > >         "applies to most of Europe"
> | >> > >         extended
> | >> > >         extremely cold
> | >> > >         extremely warm dry
> | >> > >
> | >> > >And even better, up-to-date drafts are available on the
> | web, e.g. at
> | >> > >http://webapp.etsi.org/action%5COP/OP20030321/en_3000190104v0
> | >> 20101o.pdf .
> | >> >
> | >> >It is not the business of 802 to pick between these
> | classes but we can
> | >> >refer the readers of our standard to this information.
> | >> >
> | >> >ITU-T and ANSI T1 do not have similar documents.
> | >> >
> | >> >Both G.983.3 and refer to IEC 60721, classification of
> | environmental
> | >> >conditions.
> | >> >
> | >> >IEC 60721-3-4 - Ed. 2.0  Classification of environmental
> | conditions
> | >> - Part
> | >> >3: Classification of groups of environmental parameters 
> and their 
> | >> >severities - Section 4: Stationary use at non-weatherprotected 
> | >> >locations  1995-01 is available for CHF99 at 
> | >> >https://domino.iec.ch/webstore/webstore.nsf/artnum/019208 .
> | >> >
> | >> >Piers
> | >>
> | >>
> | >> Bruce Tolley
> | >> Senior Manager, Emerging Technologies
> | >> Gigabit Systems Business Unit
> | >> Cisco Systems
> | >> 170 West Tasman Drive
> | >> MS SJ H2
> | >> San Jose, CA 95134-1706
> | >> internet: btolley@xxxxxxxxx
> | >> ip phone: 408-526-4534
> | >
> | >
> | >
> | > Bruce Tolley
> | > Senior Manager, Emerging Technologies
> | > Gigabit Systems Business Unit
> | > Cisco Systems
> | > 170 West Tasman Drive
> | > MS SJ H2
> | > San Jose, CA 95134-1706
> | > internet: btolley@xxxxxxxxx
> | > ip phone: 408-526-4534
> | >
> | >
> | >
> | >
> | 
> | 
> | 
>