Unoffical empeg BBS

Quick Links: Empeg FAQ | RioCar.Org | Hijack | BigDisk Builder | jEmplode | emphatic
Repairs: Repairs

Topic Options
#248442 - 03/02/2005 17:03 Fiber networking question
eliceo
enthusiast

Registered: 18/02/2002
Posts: 335
I currently have some fiber that runs between two buildings. The distance has to be less than 1000 feet, Honestly Im not sure why they ran fiber instead of just cat 5e. A guess would be around 700 feet. I have two cables that run into 3com switches . The connections are labeled:
100 BASE-FX
100Mbps FD FC

The fiber goes into a leviton cable thing and then comes out into 3com switches at both ends. This is 62.5 micron fiber.

I have a few 3com 4228 switches that have GBIC inputs. If I get GBIC inputs that support fiber will the existing fiber be able to connect at Gigabit speeds?

Will this fiber even support gigabit connections?

Any help is greatly appreciated.

eliceo

Top
#248443 - 03/02/2005 17:18 Re: Fiber networking question [Re: eliceo]
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
You should be well within distance even for the less expensive multimode fiber. (disclaimer, distance is also governed by laser/transmitter power, but your distance is well below what typical equipment can serve).

Quote:
Will this fiber even support gigabit connections?


Fiber is protocol-agnostic, so assuming you can mate up the right patch cables, GBICs and such (that match whatever class of fiber you have in place), I can't think of a reason why not.

Here's a colorful, usable fiber tutorial: l http://www.arcelect.com/fibercable.htm
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top
#248444 - 03/02/2005 18:17 Re: Fiber networking question [Re: jimhogan]
mlord
carpal tunnel

Registered: 29/08/2000
Posts: 14478
Loc: Canada
I think that 62.5 micron multi-mode fibre is normally rated up to 220m distance for gigabit transmission speed.. somewhat less than 1000 feet.

This could be an issue here.

Cheers


Edited by mlord (03/02/2005 18:18)

Top
#248445 - 03/02/2005 18:34 Re: Fiber networking question [Re: mlord]
muzza
Pooh-Bah

Registered: 21/07/1999
Posts: 1765
Loc: Brisbane, Queensland, Australi...
62.5 micron multimode 1000BASE SX has a range of 220m
50 micron multimode 1000BASE SX has a range of 550m
9 micron singlemode 1000BASE LX has a range of 5000m

they would have used fibre to electrically isolate the buildings. That is, you could end up with a potential difference between the buildings which is enough to fry the network gear. This would be even worse if you had a lightning strike nearby.
_________________________
-- Murray I What part of 'no' don't you understand? Is it the 'N', or the 'Zero'?

Top
#248446 - 03/02/2005 18:38 Re: Fiber networking question [Re: muzza]
SuperQ
addict

Registered: 13/06/2000
Posts: 429
Loc: Berlin, DE
you can use 1300nm LX gbics on 62.5 micron fiber to get a distance of 1804 ft (550 m)

http://networking.smsu.edu/general/info/CiscoGBIC.htm

LX Gbics are more expensive, but probably the solution to the problem.
_________________________
80gig red mk2 -- 080000125
(No, I don't actually hate Alan Cox)

Top
#248447 - 03/02/2005 19:54 Re: Fiber networking question [Re: mlord]
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
Quote:
I think that 62.5 micron multi-mode fibre is normally rated up to 220m distance for gigabit transmission speed.. somewhat less than 1000 feet.

I should wait to post and give people with a clue a chance to answer first I thought it was 1000 feet regardless if transceivers would do it.

Now Muzza's electrical potential I had not even considered/realized. The only time I've done this was not for isolation, but just to achieve/overcome ethernet/copper distance limits. I will need to grab more clue if ever I do this again!
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top
#248448 - 03/02/2005 20:26 Re: Fiber networking question [Re: jimhogan]
bonzi
pooh-bah

Registered: 13/09/1999
Posts: 2401
Loc: Croatia
Quote:
Now Muzza's electrical potential I had not even considered/realized.

Heh, veeery elongated opto-couplers.

For similar reason (elimination of ground loop) optical digital output from enhanced empegs is preferable for coaxial.
_________________________
Dragi "Bonzi" Raos Q#5196 MkII #080000376, 18GB green MkIIa #040103247, 60GB blue

Top
#248449 - 03/02/2005 21:10 Re: Fiber networking question [Re: bonzi]
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
Quote:
For similar reason (elimination of ground loop) optical digital output from enhanced empegs is preferable for coaxial.

I think I am having a very not good day (or week)! When I read...

That is, you could end up with a potential difference between the buildings which is enough to fry the network gear. This would be even worse if you had a lightning strike nearby.

I thought he was saying that you could get a potential difference *with* fiber, and that you might have to consider that. Thus my "Wow, Huh?" edit: instead of seeing that Muzza was answering the original question "why did they run fiber?", although I will say the times I did this it was distance, not lightning, that led me to it. So, while I think I remain convinced that glass is nonconductive, I gotta read others' posts more carefully.

This will teach me to try to goof off at work -- read more carefully... focus....focus!

I need to go back to what I know about -- Bibles and cysts!


Edited by jimhogan (03/02/2005 23:56)
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top
#248450 - 04/02/2005 00:41 Re: Fiber networking question [Re: jimhogan]
muzza
Pooh-Bah

Registered: 21/07/1999
Posts: 1765
Loc: Brisbane, Queensland, Australi...
years ago i worked for a company as some sort of (not very good nor experienced) IT tech. We moved into a new manufacturing plant and I was involved in the process very late in the design. Although i did a moderate job of designing the wiring of the main office (two floors, 5 departments, 50+ people) it wasn't until later that we were informed that the construction offices needed network connection.
(We'd come from an ad hoc thin net that disobeyed every rule about that kind of media - runs greater than 200m, several repeaters, many, many, many machines on several segments.)
We ended up drawing a cat5 cable thru ground conduits and bunny hopping to each successive factory. There were 5 factories in all, making quite a massive haul in all.
And electrically a minefield

I would have liked to be in on the design earlier to have put in 100mm conduits in a mesh between all factories (with draw strings) and running fibre to each of the factories.
A router (rather than the bridge we got) would have been good too.

ah the memories /shudder/
_________________________
-- Murray I What part of 'no' don't you understand? Is it the 'N', or the 'Zero'?

Top
#248451 - 04/02/2005 01:50 Re: Fiber networking question [Re: muzza]
lectric
pooh-bah

Registered: 20/01/2002
Posts: 2085
Loc: New Orleans, LA
I ran fiber between our three building simply because the cat5 they were using kept getting wet and rotting. Fiber has been stable for >5 years so far where copper was good for <2.

Top
#248452 - 04/02/2005 05:25 Re: Fiber networking question [Re: jimhogan]
bonzi
pooh-bah

Registered: 13/09/1999
Posts: 2401
Loc: Croatia
Quote:
So, while I think I remain convinced that glass is nonconductive...

I was worried for a moment...

Quote:
I need to go back to what I know about -- Bibles and cysts!

_________________________
Dragi "Bonzi" Raos Q#5196 MkII #080000376, 18GB green MkIIa #040103247, 60GB blue

Top
#248453 - 04/02/2005 13:51 Re: Fiber networking question [Re: muzza]
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
Quote:
ah the memories /shudder/

For a while I worked in a 30 story bank-owned building in Portland where the old pneumatic tube system had been abandoned and appropriated on an ad-hoc basis for fiber by the outfit I went to work for.

My most satisfying fiber experience was around '92 when trying to connect 2 buildings downtown here. We were looking at a pretty cheezy rooftop 2Mbps wireless solution (based on Arcnet cards, IIRC) but then were informed that it would violate code (the uglier of the 2 buildings was on the margin of a historic area).

When it looked like we were stumped, an enterprising contractor spotted a chopped 50-pair cable in the basement of the ugly building and followed it into a conduit under a bridge/overpass. It looked like it was ancient, from the days when AT&T fielded a standing army, launched a fleet of ships, conducted its own atmospheric testing, and put conduit wherever it felt like.

We started to ask around about that conduit, but then realized that asking around was a completely stupid approach. So we all stayed very late one evening, that contractor got into a couple J boxes, and fixed up the 50-pair as a pull string and -- voila! Five times better than what we'd hoped. Still in use today I'm told.
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top
#248454 - 04/02/2005 14:03 Re: Fiber networking question [Re: jimhogan]
pgrzelak
carpal tunnel

Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
...It looked like it was ancient, from the days when AT&T fielded a standing army, launched a fleet of ships, conducted its own atmospheric testing, and put conduit wherever it felt like.

...Ah, the good ol' days...
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs

Top
#248455 - 04/02/2005 14:12 Re: Fiber networking question [Re: muzza]
ashmoore
addict

Registered: 24/08/1999
Posts: 564
Loc: TX
My place of work is quite enlightened. Rather than pay a monthly fee to connect two buildings about 3-4 miles apart, they got a really nice 100Mbps microwave setup.
The ultimate in electrical isolation
Great until a huge construction crane was erected in its path, now the datarate drops when the crane swings sideways
_________________________
========================== the chewtoy for the dog of Life

Top
#248456 - 04/02/2005 15:07 Re: Fiber networking question [Re: ashmoore]
eliceo
enthusiast

Registered: 18/02/2002
Posts: 335
I took another look at the fiber. I was hoping I would be able to figure out the distance from the numbers on the cable, but it looks like they switched rolls a few times, so the numbers that I have been able to read, mean nothing. Unless I open up all the points and am able to read them all. I think that one cable has 12 wires inside, does it take 2 of the 12 to make one fiber connection?

Thanks for the info so far.

We also have a 100mbps microwave setup, I can't wait till that one gets obstructed either.

Top
#248457 - 04/02/2005 15:55 Re: Fiber networking question [Re: eliceo]
wfaulk
carpal tunnel

Registered: 25/12/2000
Posts: 16706
Loc: Raleigh, NC US
Yes. There is a transmit fiber and a receive fiber.
_________________________
Bitt Faulk

Top
#248458 - 04/02/2005 22:15 Re: Fiber networking question [Re: eliceo]
muzza
Pooh-Bah

Registered: 21/07/1999
Posts: 1765
Loc: Brisbane, Queensland, Australi...
how many of the 12 fibres are being used?

if they've installed a 12 strand and are only using two, you could up the datarate by using several fibres. This would easily speed up your connection to near gigabit. (12 strands giving 6 circuits, each running at 100Mbps duplex. 1.2gbps) .Hardware may not be cheap, but certainly cheaper than running a new single mode fibre and buying new hardware.
_________________________
-- Murray I What part of 'no' don't you understand? Is it the 'N', or the 'Zero'?

Top
#248459 - 04/02/2005 22:59 Re: Fiber networking question [Re: muzza]
eliceo
enthusiast

Registered: 18/02/2002
Posts: 335
well I have 2 cables that each have 6 pairs. Two pairs go to one drop off, the third goes to another drop off. Im not even sure if the other cable goes to anything (it appears to stop mid way, like they ran out of money)

Top
#248460 - 05/02/2005 00:40 Re: Fiber networking question [Re: eliceo]
muzza
Pooh-Bah

Registered: 21/07/1999
Posts: 1765
Loc: Brisbane, Queensland, Australi...
well if we just look at the cable we know is terminated, does the cable come into the first drop point and have all 6 pairs terminated. Two of these pairs are used at the first drop and another pair is used at the second drop. Essentially so the cable loops through the first drop and the whole cable continues on its merry way to the second drop. Is that the case?

If it is, you could load balance three pairs at each site/office/building and get increased bandwidth that way.
_________________________
-- Murray I What part of 'no' don't you understand? Is it the 'N', or the 'Zero'?

Top