Vertical caving terminology and methods > General hardware
Where offcuts of rope are subjected to a series of shock loading tests. A drop test is different from a pull test, since pull tests slowly increase the force applied to a rope to work out what weight it can hold before it snaps. There is no shock loading in a pull test. Static/semi-static rope, when it is certified in Europe, is tested in accordance with CEN standard EN 1891. The standard differentiates between "type A" rope which is designed for caving and used for SRT, or an intentionally lower quality "type B" which is not designed for caving. In standard tests, the rope is loaded statically with 1529 kg or 15 kN (1223 kg or 12 kN for "type B") using figure of 8 on a bight knots for three minutes. A separate sample is loaded with a 100 kg weight (80 kg for "type B") using figure of 8 on a bight knots for a minute to allow the rope to stretch, then subjected to a fall factor 0.3 shock load without exceeding 6 kN impulse force (meaning that the rope should still have the right amount of elasticity), followed by 5 shock loads of fall factor 1 without snapping. With a separate sample, the extra amount that the rope stretches when a 50 kg load is increased to 150 kg must not exceed 5%, showing that it is static rope. (Static ropes certified in the USA use a completely different standard, such as NFPA 1983. This allows them to have almost no stretch - anywhere between 1% and 10% stretch at 10% of their specified minimum breaking strength - and almost no shock absorbing qualities. A popular American rope might stretch just 3% with a 400 kg load. The shock loading produced by drop testing or falling on these ropes can be extreme. American cavers may also use ropes certified to the voluntary standard UIAA 107, which is based on the European standards, or ASTM F2116-01, which is not based on anyone else's standard.)
The standard tests are carried out at a specific temperature and humidity. The ropes are tested when they are dry. When following standard tests with a new rope that is wet, it loses around 70% of its shock absorbing performance, and might only survive 2 or 3 fall factor 1 shock loads before snapping, when it would have survived 7 to 10 if it were dry. The wet rope will stretch around 10% more when absorbing a shock load, but can produce a shock load 10% higher than the same rope would have done when dry on the first shock load, increasing with each shock load to reach about 25% higher than the same dry rope by the fifth shock load. Wet ropes also bounce back far harder, and bouncing around after the initial catch of a fall can produce forces 65% higher than the same rope would have done while dry. If a rope is shock loaded while it is wet, it will be permanently damaged, so that even after drying, it will not return to its dry performance. Ropes that have been treated with a chemical treatment that stops the fibres absorbing water (known as "dry" ropes), might initially cope with more, but this benefit is lost if the rope is soaked for several hours (which is why there is no point in using "dry" treated ropes for caving). After slowly drying a rope which has not been shock loaded while wet, the rope regains its dry performance. Testing with ropes at high temperatures, such as 70°C or 80°C, reduces the rope's performance in a similar way to being wet.
Cavers who are testing their ropes often ignore the static load tests, and might use 80 kg for drop testing (presumed to be a representative weight for a person, but this excludes clothing and equipment). They then typically use the same set of drop tests as EN 1891 expected, and if the rope survives, they are then likely to continue with repeated tests at fall factor 1, or sometimes even increasing to fall factor 2, to see how many the rope will cope with before it snaps. It is common for a rope to survive numerous fall factor 1 tests when new, far exceeding 5 and sometimes approaching 40, and perhaps even a couple of fall factor 2 tests. However, an old rope that would still be considered usable might only survive 2 of the fall factor 1 tests. This testing can help to decide when to discard an old rope. The offcut that was tested must always be discarded, even if it did not snap during testing, as it might now be at the limit of its endurance.
It is worth noting that very few testing rigs owned by cavers follow the actual standard. For the test to be comparable with the standard tests, the drop distance is critical; while it might seem that the fall factor should be the same no matter how far it falls, the ratio of knotted rope to unknotted rope changes significantly at shorter lengths. Most caver setups are much smaller than they should be. For example, the British Caving Association's mobile testing rig can only test ropes at 80 cm long, instead of the standard 2 metres, and there is no facility for getting the environmental conditions right. Traditionally, it used wet rope (at whatever temperature it is outside), and an increasing fall factor instead of just fall factor 1, and this will dramatically alter how many tests the rope will survive. It does use the standard 100 kg weight. The results are still useful, but not directly comparable with the official standard.
Dynamic rope has the CEN EN-892 standard instead, or the international UIAA 101 equivalent. This is different for single ropes (normal rope used as a single strand), half ropes (also known as double ropes, thinner ropes used in pairs which do not share connections to anchors) and twin ropes (thinner ropes used in pairs which do share anchors). Single ropes are allowed a maximum of 10% static stretch with an 80 kg load, then must survive 5 fall factor 1.75 falls with an 80 kg load tied in figure of 8 on a bight with rope passing over a standardised lip, not exceeding 12 kN on the first fall, and a maximum 40% stretch on the first fall. Half ropes are allowed a maximum of 12% static stretch with an 80 kg load, then must survive 5 fall factor 1.75 falls with a 55 kg load tied in figure of 8 on a bight with rope passing over a standardised lip, not exceeding 8 kN on the first fall, and a maximum 40% stretch on the first fall. Twin ropes are allowed a maximum of 10% static stretch with an 80 kg load for a single rope, then a pair of ropes together must survive 12 fall factor 1.75 falls with an 80 kg load tied in figure of 8 on a bight with rope passing over a standardised lip, not exceeding 12 kN on the first fall, and a maximum 40% stretch on the first fall. I am not aware of any caver-owned rigs capable of testing dynamic ropes. In places where a half rope is used, one rope will be used primarily to catch a fall, while the other rope assists later during the catch, their dynamics are different from twin ropes, which are expected to catch a fall at the same time. Half ropes are not designed to be used as twin ropes, and will produce a higher shock load if used that way.
When ropes are stressed so much that they fail and break or snap, they almost always do so at the mouth of a knot, which often gets bluntly stated as "knots weaken a rope". That is not very helpful as a statement, because knots are essential to our use of ropes. Ropes are over-engineered to allow use of knots without the strength being reduced too far, and knots are carefully selected for their strength.
Nevertheless, over-stressed ropes normally break at the mouth of a knot. This is often said to be because of the tight bends that the knot passes around. It is true that bends do increase the stress, since one side is longer than the other, and the tensile stress from the load is concentrated on the outside of the bend, while in the unknotted parts of the rope, it is shared by the entire cross section of the rope. However, the stress on a bend is linear, along the length of the rope, the direction where it is strongest. This is why a rope rarely breaks on the big obvious bends of a knot. Instead, it normally breaks between two opposing small bends that have a relatively tight radius. Up then down, or left then right, which often happens just inside the mouth of a knot, such as a figure of 8 knot.
At that point, two things happen. Firstly, ropes are not a single unit, they are a braid of multiple fibres. The fibres that are loaded on one side of a rope around a bend, need to pass the load onto fibres that will now be loaded on the other side of the rope at the next bend. This happens due to friction of the fibres against each other, which relies on the braiding and compression from the sheath to bind them against each other, and this normally takes some distance. When the opposing bends happen in a very short distance, the load needs to pass from the fibres on one side of the rope to the other side very quickly, so the load has to pass across the rope via the sheath in addition to the cores (which are relatively linear). Secondly, the opposing bends appear to create a point on the rope where there is shear stress in addition to the linear stress created by the bends themselves. Shear stress acts across the rope, rather than along it, in the direction where it is weakest, similar to how scissors cut across paper. This happens at the same point as where the load passes from one side to the other. This makes it much more likely that the rope will start to split at that point. Because there is tensile stress at the same location on the outside of a bend, the splitting is likely to start on the outside edge of the bend just where it starts to transition into that bend from a location where it was under tensile or shear stress, and the load was switching sides of the rope. Once it starts to split, the tensile stress from the bend causes the split to rapidly propagate across that strand of rope, so the rope splits completely.
In some knots, there might be a two dimensional change in direction, which would create two different directions of shear stress, or a greater amount of shear stress, or torsional stress at the same point. These knots are therefore likely to be weaker than others. Knots where the bend direction remains very consistent throughout the knot while holding a relatively large radius, such as a double fisherman's knot, are likely to be much stronger, since they are dominated by bend stress, and shear stress is not a factor. They will still eventually split when the tensile stress on the outside of a bend is enough to split the rope on its own. Once the rope has passed some distance into the intertwined parts of the knot (after an initial bend), a lot of the tensile stress has been passed on to the other parts of the knot by friction between the rope strands. The individual strands are no longer under so much tensile stress, and any shear stress will also be reduced as a result, so there is less chance that it will break further into the knot. Therefore the rope is likely to break at the first tight bend or pair of opposing bends within the knot.
In knots tied on a bight, where the loop takes the load and one of the ropes emerging from the knot takes the load, the break almost always happens on the single rope's side, since it has the full amount of stress. The end with the loop also has stress, but shares it between two strands, so it is effectively halved. In drop testing, this means that unless there is a localised weak spot in the rope, the eventual failure of the rope will almost always happen just inside the mouth of the figure of 8 on a bight knot, on the side of the knot where a single strand carries the load. If the figure of 8 on a bight is tied in the way that makes it harder to untie after loading, with the loaded rope passing around the collar of the knot at the loop end, rather than the shoulders of the knot, the stress point appears a little closer to that bend of the knot, so the break might appear to be on the bend itself. However, the actual break normally starts at the point of shear stress immediately before the bend, rather than on the bend itself. Whichever way the knot is tied, once the rope snaps, the rope on the bend, which had been under a huge amount of tension, finally has the tension released, so it shortens and retracts part way back around the bend, so it might appear to have broken on the bend itself, but in actuality, it broke while that part of the rope was further around the knot, at the point of maximum shear stress immediately before the bend. See the stress diagrams in localization of breakage points in knotted strings, Piotr Pieranski et al 2001 New J. Phys. 3 10, to see where the breakage is most likely to happen. (I would appreciate confirmation of the shear stress presumption.)
<< Fall factor | Cable, wire, wire cable, wire rope >>
This page is not intended to be viewed this way, please load the entire article. This version exists only to make it easier for search engines to understand the contents.