Openreach FTTP Cable Bend Radius

Can someone in the know please advise of the acceptable bend radius for the cabling used by Openreach:

Between the CBT & CSP
Between the CSP & ONT

Thanks!

In reply to a post by uniquorn:

Can someone in the know please advise of the acceptable bend radius for the cabling used by Openreach:

Between the CBT & CSP
Between the CSP & ONT

Thanks!

I don't have the answers but its worth saying there are two types of connectorised cable that can go between the CBT and CSP (flat and round) so the max bend radius for each are probably different.

Maybe you could explain more about why you need to know as that may help others to better answer the question.

Edited by deleted (Thu 11-Aug-22 16:03:03)

If they still use OFS EZ Bend fibre cabling between the CSP and ONT. That's G657.B3 which has a bend radius of max 2.5mm.

If someone has had OR FTTP installed recently, a picture of the internal fibre jacket markings will be super useful in working out what type it is, hence bend radius.

Although too much of a bend in an optical cable will introduce losses , ultimately if the received power level is sufficient the service works , if the losses introduced are too great the service won’t work…..getting out a protractor to measure a bend radius is an exercise in futility , if light levels have been measured and are within spec

In reply to a post by uniquorn:

Can someone in the know please advise of the acceptable bend radius for the cabling used by Openreach:

Between the CBT & CSP
Between the CSP & ONT

Thanks!

In general with fibre…

If you’re installing your own containment / conduit keep the bends as gentle as possible - if not for the bend radii of the fibre - it will make it easier for the engineer to pull through the fibre if any 90-deg. elbows are sweeps rather than sharp bends without stressing the fibre (big cause of cable fails) by pulling it too hard.

If looking at cable pathways, avoid any external / sharp corners if possible.

There is special BI fibre that Openreach use for MDU type installs - OFS InvisiLight - as said, but best to plan the engineer will just install regular ordinary “inside-out” fibre cable (Dexgreen for example) that will be installed which has an (inner white sheath) outer cable diameter of approx. 3mm. Therefore a minimum installation (short-term) bend radii of around 10x that or approx. 30mm and a static/installed (long-term) bend radii of around 20x or around 60mm. Data sheets in the Dexgreen link above if you’re interested.

Should have said (for the avoidance of doubt) what’s noted above is for “internal” cabling - from CSP to your ONT location.

From the CBT to CSP is Openreach territory- so you should *not* ordinarily need to concern yourself - unless your install is somehow ‘special’. 😂

Edited by Pheasant (Sat 13-Aug-22 13:30:11)

I've just realised I put down 2.5mm - it's 25mm!

I do agree with Iniltous though - if levels are acceptable, then it's all fine. I would imagine this question is being asked for laying ducting/conduit somewhere. If that's the case, then sweeping bends the best (as someone else said in the thread).

Hehe, yeah I thought that may have been a typo. The ITU standard is apparently 5mm radii for G.657.B3 but bear in mind that is for the fibre itself, not a built-up cable construction with buffer and various layer of jacket materials surrounding the fibre.

Manufacturers are always going to be more conservative than the extreme limits of the standard anyway. I generally fallback to a minimum radii of 10x outside cable diameter which has been a conservative but safe rule of thumb since Adam was a boy - well pre BI fibre! 🤣

Thank you for these replies folks, very helpful.

I asked this in the context of planning the external cable run for an upcoming installation. It is not feasible to use a ground level entry point at the front side of the house, as is the case for the existing copper line which is presented from an underground duct - the internal walls there have since been clad in decorative panelling, and disturbing this is a last resort.

My intention was to ask the engineer to install the ONT in an upstairs airing cupboard which also houses my patch panel, LAN switching, and router (the current FTTC modem being located in that downstairs front room with the NTE5 socket.)

This airing cupboard shares the same external wall as the existing copper entry, albeit more towards the centre of the house, and on the 1st floor rather than ground level. Though, the challenge that I'm mindful of is that there is an externally protruding chimney breast between where the cable would surface from underground ducting, to where it would need to be run to enter at the airing cupboard.

Reading your thoughts on minimum bend radius has left me a little concerned that the 4 90-degree bends required to navigate this chimney breast will likely be too tight.

I am considering a plan B which would involve running along either inside corner of the chimney breast and through the soffit in order to get around it internally via the attic - though the attic is not floored, so I presume that the OR engineer will not go in there, thus necessitating me to install ducting with fish tape.

I will try to get a photo today - would appreciate your thoughts on the idea.

My ask was in the context of planning to have the engineer run cabling as depicted in red:
https://pasteboard.co/vG7mmf4DsjEf.png

For reasons mentioned above, the ONT cannot go on the inside of where the existing copper cable enters the building on the ground floor, hence the need for a run to the 1st floor.

Though, the previous replies have me concerned that the horizontal corners of the chimney breast will be a problem.

I am now considering an option shown in blue, where the chimney breast is circumvented internally via a sweeping duct in the attic, between soffit holes either side. My thinking here is that the vertical corners up and down the side of the chimney breast could feasibly have a much larger radius than the horizontal ones required to track its outer perimeter.

Does this seem like a workable approach?