2 split into 3 - why on my pole?

My village is currently moving a step further towards its full fibre future, with the cabling and infrastructure being installed. Some of the fibre cables are routed through the existing underground ducting, but most of it is being strung overhead from pole to pole.

Naturally I'm particularly interested in the pole that will serve my premises. I can't quite figure out why they have done things the way they have and I was wondering if anyone here might shed some light on it please?

There are two flat-ish fibre cables feeding "my" pole from an underground duct (along with all the current copper). These two fibre cables go to a fibre joint box thing like the one on the left in this pic ( https://www.thinkbroadband.com/assets/images/news-20... ). From this, three cables come out. One goes to a 12-port CBT on my pole and two go overhead to the next pole along

I'm assuming the bundle within one of the cables has been split at my pole. What I can't figure out is why was this done at my pole and not later on, given what happens later on in the chain. Let me explain.....

As I mentioned, two cables go overhead from my pole to the next one. This next one (which I'll call Pole 2) is only there to change the cable direction, as it has no manifolds or anything. This is repeated at the next pole (Pole 3), which again only helps the cables change direction.

At the next pole down (Pole 4), we finally have some action again, where one of the two cables terminates at an 8 port CBT. From here, the sole remaining cable goes on to the last pole (Pole 5), this one with a tiny 4 port CBT.

What I can't figure out is why they split things up at my pole, resulting in having to have two cables from my pole to pole 2 to pole 3 to pole 4. Why didn't they use one cable instead, and then split it at Pole 4 for the final leg to pole 5?

My only thought is for future-proofing. If they find they need more capacity at Pole 4 and 5 at some point, I suppose they could feed a third cable via the ductwork cable to my pole, and do some creative splicing. It all just seems a bit odd.

Does anyone have any ideas?

And on a related matter, I found this fascinating PDF that shows how these cables are likely to contain three or four 12-strand bundles: https://www.millsltd.com/media/mills_physical_infras...

perhaps access at further down the line poles is not as easy so it was easier to do all the joining at your pole as the splicing team could park safely and easily for a time without inconveniencing others ?

Fewer joints = more reliable.

There's also a case of weighing up the relatively low cost of the extra cable, versus the cost of the track node(s) and the cost of the splicing which might have to be done at height.

I believe the CBTs come with cable already attached. So by attaching the CBT to a pole, and bringing the cable all the way back to an underground chamber, you have no aerial splicing to do at all. And if you can bring it directly back to the splitter node, so much the better.

I would say my pole was the least east to access of all of them due to its location and surroundings, but that's not to say it isn't the answer - what I think of as simple/difficult isn't necessarily how a fibre-installing engineer/planner might see it.

I hadn't thought in those terms. That's really interesting. Thank you.

You're over thinking it.

Sometimes what you see on site is not what the guys doing the planning see from google street view 4,000+ miles away in India. When this is handed over to the guys doing the implementation they just do what the drawings say and have no input into it.

hah! Now that's very true.

In reply to a post by sfo32:

As I mentioned, two cables go overhead from my pole to the next one. This next one (which I'll call Pole 2) is only there to change the cable direction, as it has no manifolds or anything. This is repeated at the next pole (Pole 3), which again only helps the cables change direction.

At the next pole down (Pole 4), we finally have some action again, where one of the two cables terminates at an 8 port CBT. From here, the sole remaining cable goes on to the last pole (Pole 5), this one with a tiny 4 port CBT.

What I can't figure out is why they split things up at my pole, resulting in having to have two cables from my pole to pole 2 to pole 3 to pole 4. Why didn't they use one cable instead, and then split it at Pole 4 for the final leg to pole 5?

What you basically have observed is three individual CBTs (of various port counts, based on how many premises they were designed to serve) going back to a single enclosure for onward splicing....

This black enclosure at "your" pole will be a track node - a joint basically. It will be used to splice together rather than 'split' fibre strands - that is it is joining fibres on a 1-to-1 basis going in and out, as opposed to actually splitting a single fibre into multiple outputs using a splitter assembly (SASA) inside a splitter nodes (looks very much like a track node from the outside, just a touch larger).

Each CBT that Openreach use, as mentioned, comes from the factory pre-terminated and tested with a tail cable assembly (of various orderable lengths) containing a fibre dedicated for each port - ready for Openreach to string up along poles or pull through the duct network. This tail with its individual fibres needs to go to a joint enclosure to be spliced one-to-one onto the cable that comes from the serving splitter node.

As @candlerb has pointed out, the less splices/joints the better from the point of view of installation economy as well as fibre light losses (although quite minimal) and long term durability of the installed plant.

In reply to a post by sfo32:

And on a related matter, I found this fascinating PDF that shows how these cables are likely to contain three or four 12-strand bundles: https://www.millsltd.com/media/mills_physical_infras...

It's going to blow your mind, but Openreach have been using 432-fibre 'ribbon' blown fibre bundles that are less than 10mm in overall diameter as the spine cabling form the head-end exchange. In turn this is spliced in groups/lots of 12 fibres in a ribbon-fusion splicer. They are squeezed into a flat web form and laid out for a 12-way "mass" splice in the fusion splice machine.

That's enough to serve FTTP theoretically to around 13,000 premises from a single cable.

Saves time, money and labour.

Thank you for all this info, and also regarding the fibre cable itself.

I love learning about this stuff and I really appreciate it when people are willing to share their knowledge. It is as though my brain wants to drill down as far as it can and won't rest until it gets fed what it wants.