Duct installation: what is the maximum pulling tension on an optical fibre cable?

The winch pulls, the dynamometer climbs, and nothing arrives in the chamber downstream: the cable has stopped two thirds of the way along, inside a bend that looked generous on the drawing. There are two ways forward: stop and recalculate, or add another twenty kilos of tension and find the damage three months later, in an OTDR report. The number that separates them sits in an ITU-T Recommendation, and it is almost always missing from specifications.

Maximum pulling tension is a formula, not a habit.

ITU-T L.100, Optical fibre cables for duct and tunnel application, edition 5.0 approved on 13 January 2024, is the Recommendation for duct and tunnel cables installed by pulling. Clause 6.2.2.1 sets the cable’s rated tensile load, LS:

LS = 1.5 × W, where W is the force in newtons exerted by the weight of 1 km of cable suspended vertically. If the result exceeds 2700 N, the tensile rating should be 2700 N.

On site: a duct cable of around 90 kg/km — a typical weight, always to be checked on the manufacturer’s data sheet — gives W ≈ 880 N, so a rated pull of about 1300 N, a little over 130 kgf. An armoured cable at 200 kg/km would reach 2940 N, but the cap brings it back to 2700 N: about 275 kgf, the absolute maximum on a terrestrial duct cable.

The same clause adds the number almost nobody asks for: the long-term or residual load rating LL is 30 per cent of LS — the force that can remain on the cable after installation, on a vertical section or in a bend where it has stayed under tension. On the 1300 N of the example, that is roughly 400 N.

And it matters more than the first. In Annex A, clause A.3.1, L.100 does not even specify an attenuation change at LS, because that is a short-term load event; at LL it requires no change in attenuation, under load and after the load is removed. It also constrains fibre strain: no more than 60% of the fibre proof strain under LS, no more than 20% under LL for fibres proof tested at 1%. Excessive tension is not felt straight away: it turns into residual strain.

The bend is the other limit, and it is measured in cable diameters.

In clause 6.2.3.1 L.100 defines minimum bending diameters as multiples of the cable outer diameter (OD): 40 × OD under load, that is during installation; 20 × OD or 30 × OD once installed. On a 12 mm cable that means 480 mm of diameter — a 240 mm radius — while pulling, and 240-360 mm of diameter once the work is finished.

This is the number that sizes rollers and sheaves in the chambers: the bending-under-tension test (A.3.3, method E18A of IEC 60794-1-21) uses rollers with a radius equal to half those values, and at the end the attenuation must be unchanged. The fibre bend radius inside closures is a separate matter, and depends on the subcategory chosen.

The duct is “proved” before, not after.

Clause 7.7 is worth a day on site. Before any installation the duct has to be proved: checked that it is clear of debris and not crushed. Then come the duct filling ratio — duct inner diameter against cable outer diameter — and the cables already in place. Clause 7.7.1 closes the loop: the pulling force should not exceed the cable tensile rating, and the use of installation lubrication can be of benefit.

A duct deformed by a heavy vehicle changes the filling ratio and multiplies friction exactly where tension is already high. Table A.10 gives the plate/plate crush test loads: 1.5 kN short term and 0.75 kN long term for unarmoured cable, 2.2 kN and 1.1 kN for armoured cable.

Blowing: the same run with an order of magnitude less.

ITU-T L.156, Air-assisted installation of optical fibre cables, approved on 16 March 2018, describes the alternative. High-speed airflow pushes the cable, distributing the force along its whole length: the tensile load is an order of magnitude lower than with pulling techniques, bends matter less, and the cable is left effectively relaxed in the duct. Watch the variant with a piston at the front end, though (clause 6.2.1): there a pulling force does exist, and the cable’s rated tensile strength stays the hard limit.

The operational numbers:

compressor pressure typically 10 to 12 bar, depending on the machine (6.1.4);
above 30 °C ambient an air cooler is needed between compressor and cable insertion machine: heat softens jacket and duct, friction rises, performance falls (6.1.4);
up to 3 km with a single blowing machine, depending on route, cable, duct and machine; for longer runs, machines are placed in series (6.3.1);
liquid lubricant is spread by pushing a sponge along with the airflow (6.3.1).

Appendix I, informative, reports Indian experience with pre-lubricated HDPE ducts: duct bending radius at least ten times its outer diameter, coefficient of friction typically below 0.1. Indicative values, not specification requirements.

Microducts: fill is a number, not an impression.

For microcables and fibre units there is ITU-T L.108, edition 2.0 approved on 16 March 2018. Table I.1 is explicit: the ratio of the microduct cable cross-sectional area to the inner aperture area of the microduct should generally not be more than 60%; Note 3 allows 70%, warning that blowing distance may decline. Appendix I.3 gives 45% to 65% as reasonable ratios and pressures of up to 15 bar for microduct cables, 10 bar for fibre units. And there is the simplest check of all (I.2): an object at least 85% of the nominal inner diameter must pass through the microduct. If that does not go through, the cable will not either.

What to write into the specification.

The rated tensile load LS of the cable offered, in newtons, per clause 6.2.2.1 of ITU-T L.100, capped at 2700 N.
Recording of the pulling force with a dynamometer and a limiter set to LS, peak value entered in the record.
Minimum bending diameter during installation (40 × OD) and roller sizes in the chambers.
Mandatory duct proving, with the outcome recorded: duct clear, not crushed, filling ratio checked against the cables already present.
Installation method stated — pulling, blowing with a piston, blowing without one — because the cable requirements change.
Lubricant compatible with jacket and duct material, data sheet attached.
For microducts, maximum filling ratio and the 85% inner-diameter passage check.

The point.

A cable pulled beyond its limit almost never breaks on the day: it answers later, with a few tenths of a dB that no acceptance test can trace back to that morning on site. Maximum tension is a design figure, like the optical budget, and it is calculated before the cable is ordered: it is part of how we design an installation and how we write the numbers into a specification. Where the duct runs is decided by the trench; how much force the cable can take is stated on its data sheet.

Do you have a long duct run or an installation specification to write? Talk to an engineer: calculating the permitted tension and choosing between pulling and blowing are part of the site survey, free of charge.