Re: [EFM] Moving forward on extended temperature range optics
There are some important issues being left out of this discussion. I still
believe that extended temperature range is a requirement, as clearly put out
by item (A) in Howard's proposal.
1) First of all, a big part of the discussion revolves around the equipment
that is located at the CO. It is important to remember that other half of the
gear is located at the customer premises, a place where the operator has no
access for the most part.
The equipment at the customer premises is in a much less controlled
environment than the CO (even considering street shelters). My personal
experience tells me that most people not only don't read manuals, but also
seem to have pleasure at torturing electronic devices. It is not uncommon to
see devices like cable modems and xDSL routers stuck inside cabinets, under
TV sets or even besides a window plenty with direct sunlight and moisture.
Customers also shun at noisy gear; that's why cooling fans are banned from
There is also the problem of the demarc point. Some of our studies called for
an external ONU, at the outside wall of the customer. This is another
location where the temperature range is problematic.
[btw, there is a consequence of another fact: neither engineers or telco heads
are exactly experts on customer devices. It's a very particular environment
where 802.3 has never been before]
2) I'm also concerned about the cost impact on extended temperature range
components. Without the pressure to think about the extended temperature case
(if left as an option), it is highly probable that the standard parameters
turn out to be too strict for the extended temperature range; in other words,
extended temperature components may turn out to require alternative
technology (ex: FP x VCsel), which is more expensive.
On the other hand, if the standard *requires* extended temperature range,
there will be a few desirable side effects:
- more effort is done to make sure that the standard does not impose
unreasonable costs for extended temperature range components;
- larger volumes of extended temperature components will lead to lower costs;
- more effort to develop tech that works at the extended range, which may also
lead to lower costs.