OT: Below grade fiber interconnect points

Has anyone ever used a below grade vault for housing fiber cross connects?

We have to move a fiber interconnect facility due to the current building being demolished.
If you have I would be interested in talking to you. If there are more appropriate lists, I would appreciate any suggestions.

Thanks,
-Roy Hockett

Network Architect,
ITS Communication Systems
University of Michigan
Tel: (734) 763-7325
Fax: (734) 615-1727
email: royboy@umich.edu

When you say "below grade vault", do you mean something that's only accessible through a manhole?

I haven't done this specifically, however if the vault does not have a controlled environment, you could be dealing with massive headaches related to dust/dirt contamination, moisture penetration, etc. I work in a large-campus .edu environment, so I'm some of the headaches you're probably trying to avoid. Also, be aware that access to the vault could be an issue. There are OSHA regs related to what sort of training and safety equipment someone who will be working in an underground vault must have.

I'm assuming that the fiber will be cross-connected to a new location prior to the building being demolished.

Not knowing your outside plant or circumstances, would it be feasible to fusion-splice a new tail onto the fiber that was going to the building that's being demolished, or (ideally) pulling a new piece of fiber to the new building, so you don't have to deal with potentially dodgy splices?

jms

Usually it would spliced outside at the manhole where the fiber meet to go in the building. Depends on the way you want to connect them etc.

Thomas L Graves

You can stick a "splice" in a manhole. You don't want a "patch panel"
or cross-connect in that sort of environment, keep that housed inside,
somewhere.

Jeff

Thank you for comments. Let me clarify the situation. We have a building that has been fiber cross connect
location and is being demolished. This location has about 20 fiber cable entering where we patch between
fiber paths. If we relocated these cross connect field to another building and that build is demolished we have
to do this all over again, so the desire was to have an independent facility for the fiber cross connect field, but
I am guessing due to esthetics the below ground vault was selected, we just learned of this selection and thus
my query to this group to find other that have dealt with similar situations and if so, experience base recommendations,
and things to be aware of.

Thanks,
-Roy Hockett

Network Architect,
ITS Communications Systems and Data Centers
University of Michigan
Tel: (734) 763-7325
Fax: (734) 615-1727
email: royboy@umich.edu

If the vault has a controlled environment and access, similar to what you would find inside of a comms room, that's one thing. If it's more like a typical manhole (damp, dirty, dark, possible temperature extremes, other utilities/hazards), then the only thing that should be in there is a water-tight splice case. Fiber patches need to be in a clean environment.

Did this project provide any funds for relocation or replacement of the communications facilities that would be lost due to the demolition? We've gone through this many times on our campus.

jms

Your only other real option would be to deploy a "road-side" cabinet which has environmental controls...

https://www.google.com/search?q=roadside+cabinets&tbm=isch&tbo=u&source=univ&sa=X&ei=f8-EUqHNOdPqkQe32IHABw&ved=0CEsQsAQ&biw=1620&bih=735

Thanks,

Joe

There are a couple of ways to do this. If you're going to keep everything
below grade in a man hole or cable vault of some type, you definitely need
to have it in a sealed splicing enclosure. That means instead of using a
patch panel, everything has to be fusion spliced. You'll get less loss
because it's a direct splice, but changing things later becomes a pain.

The other option is to use an outdoor fiber patch cabinet / pedestal.
There are multiple outdoor rated above ground fiber cross connect cabinets
available that you could probably find an unobtrusive place to install.
Or, there are several manufacturers (3M, etc.) that make fiber cross
connects that fit in a plastic version of a standard telco pedestal which
you could pretty much place anywhere.

Another option is an above ground cabinet. Many telecoms use them.

Thomas L Graves

Here is a link to a Raycom Fosc that has pigtails and bulkheads in it that I'm guessing would suit your needs. We use them underground in vaults a various points where a pedestal doesn't make sense. You need to make sure there is proper drainage in the vault though....

Without knowing more about the physical facility, it's hard to know for sure what you need.

http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&ved=0CGsQFjAJ&url=http%3A%2F%2Fwww.raycom.cz%2Fdl2%3Fid_download%3D164&ei=KS-FUpm3M-PayAGMkIDQDw&usg=AFQjCNGrdt4iDKOmh6CNXG4YtwxZocBJTw&bvm=bv.56343320,d.aWc&cad=rja

Sam Roche - Supervisor of Network Operations - Lakeland Networks
sroche@lakelandnetworks.com| Office: 705-640-0086 | Cell: 705-706-2606| www.lakelandnetworks.com

IT SOLUTIONS for BUSINESS
Fiber Optics, Wireless, DSL Network Provider; I.T. Support; Telephony Hardware and Cabling; SIP Trunks, VoIP; Server Hosting; Disaster Recovery Systems

"The information contained in this message is directed in confidence solely to the person(s) named above and may not be otherwise distributed, copied or disclosed. The message may contain information that is privileged, proprietary and/or confidential and exempt from disclosure under applicable law. If you have received this message in error, please notify the sender immediately advising of the error and delete the message without making a copy."

I remember seeing a below grade vault here in Michigan once. It was
purpose built (years ago for MCI IIRC), but I don't know by whom.
Heavy steel plate door on top, looked like those on major water pipe
vaults. Likely built to similar civil engineering standards. But
this was fairly early in the history of laying fiber, so there are
probably newer standards.

Off the top of my head, it had a lot of things concerned with water and
humidity -- dual redundant sump pumps, dual heaters mounted 6' off the
floor, an environmental monitoring panel, an exterior antennae pole for
out-of-band reporting from the monitoring panel. I didn't have the
opportunity to open the fairly beefy looking power panel, so I don't
know whether there was a dual feed -- but it wouldn't surprise me.

As to cleanliness, it wasn't particularly clean, but not really dirty.
(Much like any exterior shopping center access demark, assuming you've
seen those.)

I also saw a Bell South below grade fiber vault once, but wouldn't
recommend it, as it was full of water at the time.... To be fair, I'm
not sure they had a cross-connect panel in there.

Hi Justin and Roy:

Has anyone ever used a below grade vault for housing fiber cross
connects?

We have to move a fiber interconnect facility due to the current
building being demolished. If you have I would be interested in
talking to you. If there are more appropriate lists, I would
appreciate any suggestions.

When you say "below grade vault", do you mean something that's only
accessible through a manhole?

In the case of CEVs, it's actually a "doghouse," an access-controlled
(cyber-lock or similar cylinder plus alarm system) entrance which also
houses the HVAC machinery. See below.

I haven't done this specifically, however if the vault does not have a
controlled environment, you could be dealing with massive headaches
related to dust/dirt contamination, moisture penetration, etc.

UC Berkeley installed 3 CEVs (Controlled Environment Vaults) below
ground on campus about 10-15 years ago. One of them houses one of the
two main fiber penetrations to campus, including DWDM gear,
patch-panels, border routers, even packetshapers (back when those were
relevant in a large EDU environment), servers, WiFi portals, etc. This
stuff has all been in place for at least 10 years and has worked really
well, modulo the caveats below. Two of the vaults have 6-7 19" telco
racks, and one (the one with the big fiber entrance) also has a 23" rack
in addition to the others.

Caveats:

o It's hard to physically get equipment into the vault. You'll often
need a hand-crane/winch for smaller items (servers, switches, small
routers), and a powered crane (one of those small ones that goes in the
back of a pickup truck) for larger items. We've managed to put things
as big as a Juniper M120 in the CEV, and we have probably gotten bigger
stuff down there. But it's not just as trivial as loading it onto a
hand-cart and wheeling it in.

o HVAC issues happen everywhere, but we've had to completely overhaul
the HVAC on one of the CEVs about 2-3 years ago (right around the time I
left UCB). They were throwing more heat at the HVAC than was specified,
and there were a lot of custom parts for the doghouse enclosure that
weren't available anymore. So you may find that you need to allocate
more budget to maintenance for HVAC (as well as sump pumps, CO sniffers,
etc.).

o The larger of the three CEVs had an encounter with the Highway 24
Freeway overcrossing over Telegraph Avenue and had to be repaired before
installation. (The CEVs are divided horizontally, so that there is a
top piece and a bottom piece that get installed in the ground in
sequence and then are sealed up. In this case the top
piece--specifically the doghouse assembly--got crashed into the bridge.)
It was repaired and has been fine, but you may have to deal with issues
like that.

o Do not wear a skirt, dress, or kilt when entering the CEV. The
updraft of ventilation will cause issues as you climb down. (Sounds
silly, but I have actually had to advise folks to wear long pants based
on the experiences of a manager who was touring the CEV one day.)

I work
in a large-campus .edu environment, so I'm some of the headaches you're
probably trying to avoid. Also, be aware that access to the vault could
be an issue. There are OSHA regs related to what sort of training and
safety equipment someone who will be working in an underground vault
must have.

Yes. We have all had to go through the required OSHA training for
confined spaces and ladder safety. The CEV system must also have a CO
sniffer, sump pump, etc. We had a very lightweight lock-out-tag-out
system that basically involved calling the NOC when entering and exiting
the vault and also logging that information in a logbook physically at
the vault.

I'm assuming that the fiber will be cross-connected to a new location
prior to the building being demolished.

Been there, done that as well. UCB had a big fiber re-splicing about 4
years ago, back when I still worked there. In this particular case, it
went into a regular building, but the setup was similar to that of the
CEVs (it was the redundant entrance to the campus for one of the CEVs).

Not knowing your outside plant or circumstances, would it be feasible to
fusion-splice a new tail onto the fiber that was going to the building
that's being demolished, or (ideally) pulling a new piece of fiber to
the new building, so you don't have to deal with potentially dodgy splices?

So, if it's like what was done at UCB, the fiber was disconnected
(starting at 5pm, since an entire building would be down for _each_
bundle of fiber that was done) from the panel in the old building and
back-pulled to the nearest CV (not controlled environment). It was
re-spliced in that CV and routed to the new building, reterminated and
reconnected to a new switch to bring the building back up. (Some
bundles didn't need to be spliced, since it was a shorter path to the
new building.) This is all doable within a CEV environment as well, and
it requires pretty much the same level of coordination and project
management.

michael

[ 17 pages of caveats elided ]

So, the elephant in the room at this stage of the thing is this:

Why don't you just *put this stuff in a building*, and, y'know, never
demolish it?

Yes, you'll probably have to build it to CO grade standards, but that
isn't exactly rocket surgery, and it seems to me that you peel
3 or 4 layers of crap off the top doing it that way.

Unless you're in, say, the Philippines, I can't see the advantage of
burying all that stuff underground on a campus-scale deployment.

Cheers,
-- jra

From: "Michael Sinatra" <michael@rancid.berkeley.edu>

UC Berkeley installed 3 CEVs (Controlled Environment Vaults) below
ground on campus about 10-15 years ago. One of them houses one of the
two main fiber penetrations to campus, including DWDM gear,
patch-panels, border routers, even packetshapers (back when those were
relevant in a large EDU environment), servers, WiFi portals, etc. This
stuff has all been in place for at least 10 years and has worked really
well, modulo the caveats below. Two of the vaults have 6-7 19" telco
racks, and one (the one with the big fiber entrance) also has a 23"
rack in addition to the others.

Caveats:

[ 17 pages of caveats elided ]

I realize I am wordy, but four bullet points (one of which involves
apparel) != "17 pages of caveats". Nice try. The rest of the email was
inline replies to Justin's points.

So, the elephant in the room at this stage of the thing is this:

Why don't you just *put this stuff in a building*, and, y'know, never
demolish it?

Have you ever been involved in University space wars? Especially in a
new building? The 9-layer OSI model gets pretty top-heavy when you
factor that in. If anything, the caveats helped to keep others from
wanting to use the space.

But I will say that the general difficulty of getting equipment in and
out of the CEVs generally discouraged UCB from doing more CEVs beyond
the 3 originals. That _one_ caveat weighed pretty heavily.

michael

From: "Michael Sinatra" <michael@rancid.berkeley.edu>

>> Caveats:
>
> [ 17 pages of caveats elided ]

I realize I am wordy, but four bullet points (one of which involves
apparel) != "17 pages of caveats". Nice try. The rest of the email was
inline replies to Justin's points.

We're not supposed to use emoticons on NANOG; it's unprofessional or
something. :slight_smile:

> So, the elephant in the room at this stage of the thing is this:
>
> Why don't you just *put this stuff in a building*, and, y'know,
> never demolish it?

Have you ever been involved in University space wars? Especially in a
new building? The 9-layer OSI model gets pretty top-heavy when you
factor that in. If anything, the caveats helped to keep others from
wanting to use the space.

Nope. But this isn't "space". It's "equipment". I assume they're
not moving their 1.5MW gensets around every year either?

But I will say that the general difficulty of getting equipment in and
out of the CEVs generally discouraged UCB from doing more CEVs beyond
the 3 originals. That _one_ caveat weighed pretty heavily.

Yeah; cranes are a bitch. :slight_smile:

You seem to be taking this awfully personally, though, Mike; did you
*set* the policies and procedures I'm scoffing at?

Cheers,
-- jra

...

Yeah; cranes are a bitch. :slight_smile:

No, it is arranging for a rigging crew and the
safety plan reviews for the lift (at least in any
major company/institution which wants to
stay on the happy side of OSHA; and has
consul that suggests that the risks of not
following the process is likely a CEE).

Gary

I am NOT TAKING IT PERSONALLY DAMMIT!!!

Okay, now being serious (note clever way of avoiding using emoticons
while pointing out that I _wasn't_ being serious above), I wasn't
involved in the decision process and didn't have much say in why things
were done. If it looked like I was taking it personally, that was only
because of the bald-faced, yet hidden accusation of my being wordy,
which I categorically resemble.

Okay, now I will really be serious. For the stuff that they did, and
still do, the CEVs _did_ (and do) work. The user interface is a bit
more challenging that a regular building. My understanding is that this
got us out of a lot of political battles, but I was not privy to those
conversations. I think, however, that it may have something to do with
what Roy is going through at UM, especially as he noted that the
decision appeared to have been made at higher levels. In this case,
management did something that's not totally wrong, IMO.

michael

From: "Michael Sinatra" <michael@rancid.berkeley.edu>

> You seem to be taking this awfully personally, though, Mike; did you
> *set* the policies and procedures I'm scoffing at?

I am NOT TAKING IT PERSONALLY DAMMIT!!!

Well, jeez, Loueeze... :slight_smile:

Okay, now being serious (note clever way of avoiding using emoticons
while pointing out that I _wasn't_ being serious above), I wasn't
involved in the decision process and didn't have much say in why things
were done. If it looked like I was taking it personally, that was only
because of the bald-faced, yet hidden accusation of my being wordy,
which I categorically resemble.

Which category?

Okay, now I will really be serious. For the stuff that they did, and
still do, the CEVs _did_ (and do) work. The user interface is a bit
more challenging that a regular building. My understanding is that this
got us out of a lot of political battles, but I was not privy to those
conversations. I think, however, that it may have something to do with
what Roy is going through at UM, especially as he noted that the
decision appeared to have been made at higher levels. In this case,
management did something that's not totally wrong, IMO.

Yeah; I get what you're saying; layers 9 and 10 often overrule layer 8
(that's money, lawyers, and people, respectively :-). I just can't help
but think that the CBA is *really* lopsided against CEVs in nearly every
circumstance; the "wink wink nudge nudge" approach to keeping departments
out of your damn rack room seems the largest weighted item on the list.

Cheers,
-- jra