atcomsystems.ca/forum Forums Wiring - Telecom Carriers - Trunks & Services Outside Plant OSP Construction grounding between buildings

yes

How about the power fact after it leaves the fiber ?

Fiber is a to b.

There are other elements at the outer building that you have to consider. Right ?

They all have a effect even thought fiber is being used.

Fiber is a transport of the siginal from the ksu
to keysets at outer bldg.

If a electrode takes a strike its not fiber
(antenna,switch,router,security system are all
common to power)

I seen it happen all is protected, wrong, every element was not was properly protected and it must be.

Ive seen lightning punch a hole in the roof and use the bldg steel as a path and take out everything electrical but that has nothing to do with connecting two bldg s

as far as the path between blogs then yes fiber is 100%

To sum it up, like everyone has pointed out, do it right and you should come out okay.

Fiber is glass. Non-conductive. But what is on each end is power driven. Therefore the issue.

Skip has seen what the true power of lightning is. No matter what you do, if lightning wants you, IT GETS YOU !!! But you can reduce the risk by doing it right.

Cost of this project can be somewhat scary but what will the long term cost be for damaged equipment and lose of production?

Good Luck.

Bravo---Thanks Ken well put.

Ed, I'm going to steal this, sorta. I was taught that you should NEVER ground both ends of a cable shield, if they were in different buildings because of circulating ground currents caused by differences in building grounds. Comments, education, whatever? And anybody else, jump in here, this should be anm interesting discussion! John C. (Not Garand)

Dunno, John. All that I've ever followed was Bell System's Blue Book of Construction Standards. I am sure that installation standards may vary throughout the country. Technically, if the buildings' electric services are bonded and grounded properly, I would tend to think that a ground loop would be a moot point.

I will agree that I've taken apart sheath bond connections in splice closures, especially on long aerial runs, and have seen some minor sparking. I always assumed that this was simply due to voltage induced from the power being carried on the same pole run. Since aluminum is a fairly poor conductor, I figured that was the reason. Even though it was bonded, it could still pick up induced AC.

I think John has a valid point. Even with properly bonded electrical services the shield would be another parallel current carrying path. Whether that presents a problem is another matter though.

I do know that the specs for a service from the pole to an indoor terminal, the building side shield should only be bonded to the stub splice closure (if used) so it ends there. The terminal is bonded to the electrical ground but not the shield. Reason being what comes from the outside (lightning, power crosses) stays outside.

You would be surprised how many Verizon guys don't know this.

-Hal

I agree with both Ed and Hal. Bell always wanted grounding AND bonding. This may have been due to the lead sheath days. The ground loop is a fact, even if we were taught that "ground is ground the world 'round".

I have "cured" problems by "un-grounding" (Is that a word?) a cable on one end. In the early '90s Sprint started the isolation of C.O. grounds and bonding everything in a C.O. to a "PANI" grounding bar (Producers-Absorbers-Neutrals-Inductors). This was during the era of changing from mechanical/analog to digital switches.

So, by current standards at Embarq, the sheath is not grounded to the same grounding point as the C.O. protectors, the grounds on the outside plant are bonded to the MGN, and the C.O. has a separate grounding field. You will find this consistent with RUS as well.

Ed, your box. Hal, your code book. John, I'll have to e-mail you later.

There are two issues here - (1) Grounding for safety and (2) Grounding for performance.

In the old (BSP, GTEP) days of Panel, Stepper and XBar the equipment was so robust that performance was never in doubt. The emphasis was on safety - hence grounding both ends of a cable.

With digital systems, performance suddenly becomes important. When you're dealing with a 3 or possibly 5 volt difference between a "1" and "0" then a 2 volt difference in ground potential suddenly becomes crucial. This applies to both TDM telecom systems as well as Data products.

I recently saw an interesting problem. A customer had two buildings, less than 100' apart. The computers in Building "A" could log on to servers in Building "A" but not to those in Building "B" and vice versa. It turned out that the contractor had not followed the EIA/TIA 607 spec - He had not tied the ground bars in the two closets together. The two buildings were served by two different power generating stations. When it was tested the ground potential was found to be over 3 volts! As a result, each building could function independently but not together as a complete network. A 2/0 cable was pulled between the two buildings and the problem went away.

I should also say that the data tie between the two buildings was a Cat 6 cable. Because of the short distance between the two sites, it was felt that the expense of fibre was not justified. If an optical feed had been put installed, I believe the problem would never have been discovered - At least not immediately.

Ken, I too was raised on "ground is ground, the world round" - but I've come to believe it just ain't so anymore.

Sam