The full version of the article can be downloaded in pdf at: https://craa.cz/wp-content/uploads/2021/08/strength-of-damaged-butterfly.pdf

 

Abstract

The article deals with the influence of the rope damage discarded from the rope chain by the alpine butterfly on the strength of the knot and the rest of the rope chain. It presents the results of tests performed on damaged knots on the tensile machine in order to determine the possibility of further use of the knot to secure the rope access technician.

Keywords

Rope, knot, alpine butterfly, strength test

1.    Introduction

Both in occupational and leisure rope access, it is common practice to exclude a damaged section of rope from the entire chain using one of the eccentric knots, most often an alpine butterfly. In such a case, common sense dictates not to use the loop of the bow tie in which the damaged part of the rope is located, even though the loop of the knot is formed by two ropes and is therefore doubled.

With this in mind, some training agencies prohibit the use of a loop of butterfly when securing a knot to secure by inserting a cow’s tail’s connector into the loop and require the use of other means (e.g., ascenders during descent or descender when ascending). On the one hand, this increases the requirements for the amount of equipment carried and used (even today there are still rope access technicians performing work exclusively on the descent who do not have and do not use ascenders) and at the same time it increases the time needed to pass the knot.

The aim of the test was to determine the significance of loop damage in the rope chain, whether the damaged loop is sufficient to secure the rope access technician and what role the location of loop damage plays.

2.    Equipment used

2.1   Rope

The tests used a new, never-used, low-stretch type A rope, certified according to the EN 1891 standard manufactured by the French company Courant, model Truck. The manufacturer states the following information in the manual:

  • diameter: 10.5 mm,
  • material: polyamide,
  • static strength: 30 kN,
  • knot strength (using a figure eight loop): 19 kN,
  • sewing strength: 23 kN
  • sheath strength (of the total strength of the rope): 47 %,
  • sliding of sheath: 0,00 %,
  • elongation 50/150 kg[1]: 3 %,
  • mass per meter of length: 73 g,
  • shrinkage 3.5 %,
  • knottability: 1.

Ropes from two different series were used, one was manufactured in January 2021, the other in June 2021.

The test itself was preceded by verification of the rope diameter according to Subclause 5.3 of EN 1891[2], which showed that the actual rope diameter is 10.69 mm.

2.2   Tensile machine

The tests were performed on a stand-up (vertical) hydraulic tensile machine EU 20 produced by VEB Werkstoffprüfmaschinen in 1987, refurbished in 2001, capable of producing a maximum force of 200 kN. It displays the measured values on an analog load cell, which is part of the shredder’s control panel and at the same time records them via a coordinate paper recorder. The measurement accuracy of this regularly calibrated shredder is 0.5%. The tearing speed was controlled manually so that the traction speed of the traction device complied with the requirements of Subclause 4.1.2.2 of EN 364.

The rope and other synthetic material are attached by clamping cylinders AVH-2 and AVD-2. Due to the relatively short working range of the tensile machine (which is 300 mm), it proved as inappropriate to anchor the rope in the AVH-2 through the originally planned tensionless hitch, but it was necessary to use a knot. Rock Empire Magnum Steel 2T[3] connector with strength of 46 kN in main axis was connected to the AVD-2 to the AVD-2 via a Rock Empire Open Sling Work 20 mm[4] textile lanyard, with strength of 35 kN. During each test, the connector was attached to the loop of the knot as if the rope access technician would have been attached to it, and acted on the loop of the knot in the same direction as the suspended (or falling) rope access technician would impact on the loop.

2.3   Test specimens

In order to determine whether the location of the damaged area on the butterfly’s loop affects the strength of the loop itself, the test specimens were cut with a hot knife cutter:

  • from the outside of the upper part of the loop (test No. 3),
  • from the inside of the upper part of the loop (test No. 4),
  • from the outside of the apex of the loop (test No. 1),
  • from the inside of the apex of the loop (test No. 2),
  • from the outside of the lower part of the loop (test No. 5),
  • from the inside of the lower part of the loop (test No. 6).

In the case that the rope core was not damaged, but only the sheath, other test specimens (without the use of a hot knife cutter) were completely stripped of the sheath:

  • in the upper part of the loop (test No. 8),
  • at the apex of the loop (test No. 7),
  • in the lower part of the loop (test No. 9).

The last of the test specimens was stripped of both the sheath and seven of the nine strands in the core. Only two strands of core and a rope identification tape (the inner marking) remained at the top of the loop (test No. 10).

It is probable that the hot knife cuts did not have the same depth (they were the deepest in test No. 1, i.e. on test specimens 1-1, 1-2 and 1-3, when their depth of the cut was only estimated, in other cases a stop jig was used during cutting so that the depth of cut corresponds to the radius of the rope).

 

3.    Methods

The tests were performed in the following order:

  • Test No. 0 – testing of undamaged knots in order to determine the strength of the rope in the knot,
  • Test No. 1 – testing of specimens damaged from the outside of the apex of the loop,
  • Test No. 2 – testing of specimens damaged from the inside of the apex of the loop,
  • Test No. 3 – testing of specimens damaged from the outside of the upper part of the loop,
  • Test No. 4 – testing of specimens damaged from the inside of the upper part of the loop,
  • Test No. 5 – testing of specimens damaged from the outside of the lower part of the loop,
  • Test No. 5 – testing of specimens damaged from the inside of the lower part of the loop,
  • Test No. 7 – testing of specimens with removed sheath at the top of the loop,
  • Test No. 8 – testing of specimens with removed sheath in the upper part of the loop,
  • Test No. 9 – testing of specimens with removed sheath at the bottom of the loop,
  • Test No. 10 – testing of specimens consisting of two strands of the core and an identification tape at the top of the loop.

Each of the knots was properly dressed and statically loaded with a force of 1.5 kN before starting the test. In the tests themselves, the knots were loaded by the upper end of the rope and the loop of the knot, not eccentrically behind both working ends of the rope.

Each specimen was tripled, so each of the tests was performed three times. A total of 33 knots were tested, of which 30 knots were damaged prior testing. The authors are aware of the fact that the three measurements may not be considered sufficiently conclusive. But it is necessary to point out that the purpose of the test was to determine the significance of rope damage in the loop for securing rope access technician, not a study of the extent of the effect of rope damage on the strength of the rope chain.

When evaluating the test results, the arithmetic mean was calculated, then the sample standard deviation, and the mean squared error (of the arithmetic mean). To get the result the Student’s distribution with a given value of reliability (for three measurements performed with a choice of 68.3% probability) was 1.32[5]. Due to the fact that significantly different values were measured for some tests, the median was also calculated as a matter of interest.

4.    Results

4.1   Summarizing results

Of the thirty alpine butterflies tested with damage, twenty-one of them had ruptured in the knot (damage of the loop did not affect the rupture), and in three tests the other knot (the anchoring knot, not the tested knot nor the damaged loop on the tested knot) ruptured  (in one case it was a rope from the beginning of 2021, in two cases a rope from June 2021).

In six cases only the damaged loop ruptured and the damage had the effect of decreasing the strength of the whole system. In three cases it was a loop consisting of two strands of the core and an identification tape (tests 10-1 to 10-3), in two cases a rope cut from the outside of the apex of the loop, where the depth of the cut was only estimated (tests 1-1 and 1 -3) and in one case a rope cut from the inside of the apex of the loop (test 2-2). However, in no case did the force required to break the rope fall below 9 kN (for tests No. 10 the force was in the range of 9.22 – 9.60 kN, for test 1-1 then 10.62 kN, and for tests 1-3 and 2-2, a force of 18.60 kN was required).

4.2   Results of particular test

4.2.1     Test No. 0 – rope without any damage

The test marked as 0 was not intended as a test, it was used to determine the strength of the knot with undamaged loop so that it was possible to compare the impact of individual damages. After some of the other performed tests completely minimized the impact of damage, this test was included in the results.

The strength of the undamaged alpine butterfly loaded by the loop was 20 ± 1.7 kN. In all cases, broke the knot itself, not the loop of the knot.

Test No. 0
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
0-1 22.18 kN 20.45 kN 20.13 kN 1.70 kN 1.29 kN knot
0-2 20.45 kN knot
0-3 17.76 kN knot

 

4.2.2     Test No. 1 – damage to the outer side of the apex of the butterfly’s loop

Placing rope damage from the outside of the apex of the loop is the most common way to use in rope access technician real life. This is due to the tying itself, the rope access technician instinctively bends the rope under the damaged section and then keep on tying the rest of the knot with it, and also because the damage at the apex of the knot is most noticeable, thus increasing the likelihood that other rope access technicians who don’t know about the damage, can register it on apex of the loop in time. Last but not least, placing the damage at the apex of the loop reduces the risk that the damaged section will run into the knot (move out of the loop) when the rope load changes.

No stopping jig was used in the preparation of the first test specimens, as in the others, and the depth of cut was only estimated. The first test specimen itself (No. 1-1) had the deepest cut of all test specimens, which was also visible visually and showed the lowest strength during the tests. The average strength of alpine butterflies damaged from the outside of the apex of the loop was 16 ± 3.6 kN. In tests No. 1-1 and 1-3, the loop was ripped at the location of damage, in the case of test No. 1-2, the knot itself was broken (regardless of loop damage). It should be recalled again that the cuts damaging the rope were deeper than in all other test specimens.

 

Test No. 1
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
1-1 10.62 kN 18.60 kN 16.10 kN 3.62 kN 2.74 kN loop in the damage
1-2 19.08 kN knot
1-3 18.60 kN loop in the damage

4.2.3     Test No. 2 – damage to the inner side of the apex of the butterfly’s loop

The damaged loop in the location of damage the inside of the apex of the knot loop) was ripped only in the case of test 2-2 at a force of 18.62 kN. In both other cases (2-1 and 2-3), the knot itself was broken regardless of damage to the rope in the loop. Alpine butterflies with a damaged inner side of the loop showed strength 19.9 ± 0.85 kN.

Test No. 2
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
2-1 20.70 kN 20.35 kN 19.89 kN 0.85 kN 0.64 kN knot
2-2 18.62 kN loop in the damage
2-3 20.35 kN knot

 

4.2.4     Test No. 3 – damage to the outer side of the upper part of the butterfly’s loop

The upper part of the butterfly’s loop means that part of the loop which, if unloaded points upwards, away from the ground and does not come into contact with the standing part of the rope hanging down when the loop is loaded.

From the point of loading, the variant in which there is no damage at the apex of the loop appears to be safer, since two parts of the loop, one of which is without damage, are evenly loaded from the undamaged top. In all three tests performed in Test No. 3, the knot ruptured, while the damage to the loop did not affect the strength of the chain. The force required to rupture an alpine butterfly with a loop damaged outside the apex was 20.2 ± 0.65 kN.

Test No. 3
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
3-1 19.23 kN 20.36 kN 20.16 kN 0.65 kN 0.49 kN knot
3-2 20.90 kN knot
3-3 20.36 kN knot

4.2.5     Test No. 4 – damage to the inner side of the upper part of the butterfly’s loop

Even in the case of an alpine butterfly with damage to the inner side of the upper part of the loop, the damaged loop was not broken during the tests, but the knot ruptured. This occurred at a force of 19.9 ± 0.29 kN.

Test No. 4
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
4-1 19.45 kN 19.95 kN 19.87 kN 0.29 kN 0.22 kN knot
4-2 19.95 kN knot
4-3 20.20 kN knot

 

4.2.6     Test No. 5 – damage to the outer side of the lower part of the butterfly’s loop

In the case of an unloaded alpine butterfly, the lower part of the loop means the part of the loop which is closer to the ground and, after loading the loop, approaches (comes into contact) with the standing part of the rope leading down below the knot.

Damage to the outside of the loop itself did not affect the strength in the case of test No. 5, in all three tests the knot broke. The rupture occurred at a force of 21 ± 1.1 kN.

Test No. 5
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
5-1 20.25 kN 20.25 kN 20.55 kN 1.12 kN 0.85 kN knot
5-2 22.15 kN knot
5-3 19.25 kN knot

4.2.7     Test No. 6 – damage to the inner side of the lower part of the butterfly’s loop

As in the case of a loop damaged from the outside, the damage from the inside did not affect the strength of the loop and rupture occurred at the knot at a force of 19.6 ± 0.23 kN.

Test No. 6
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
6-1 19.85 kN 19.65 kN 19.58 kN 0.23 kN 0.18 kN knot
6-2 19.25 kN knot
6-3 19.65 kN knot

 

4.2.8     Test No. 7 – removed sheath at the apex of the butterfly’s loop

The removal of the sheath at the apex of the loop should theoretically result in a 47% reduction[6] in the strength of the loop compared to its original strength. Nevertheless, in tests 7-1 and 7-2, the knot ruptured (loop damage had no effect on chain strength) and in test 7-3, the anchoring knot, which was identical to the tested knot, was broken (in the anchoring knot itself). The knots ruptured at a force of 20 ± 0.8 kN.

It should be noted that several yarns in one strand were inadvertently cut during the removal of the sheath of specimen 7-1.

Test No. 7
test breaking force median Mean s. stand. deviation mean sq. error place of rupture of the rope
7-1 19.50 kN 19.50 kN 19.97 kN 0.80 kN 0.60 kN knot
7-2 21.17 kN knot
7-3 19.25 kN anchoring knot on AVH-2

Note:   Test No. 7 was the last to use a rope made in January 2021, and all subsequent tests continued on a rope made in June 2021

 

4.2.9     Test No. 8 – removed sheath in the upper part of the butterfly’s loop

When testing alpine butterflies with the sheath removed in the upper part of the loop always broke the knot (the removed sheath did not affect the strength of the loop). An interesting finding was that when testing samples 8-1 and 8-3, the rope identification tape was partially damaged (but not completely torn) and several yarns in two strands tore during test 8-2. The knots themselves ruptured when a force of 20.78 ± 0.02 kN was reached.

Test No. 8
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
8-1 20.80 kN 20.80 kN 20.78 kN 0.02 kN 0.02 kN knot
8-2 20.75 kN knot
8-3 20.80 knot

 

4.2.10  Test No. 9 – removed sheath in the bottom part of the butterfly’s loop

Whilst during tests of an alpine butterfly with the sheath removed in the upper part of the loop always ruptured the tested knot, during tests of butterflies with sheath removed in the lower part of the loop in two cases (9-1 and 9-2) ruptured the anchoring knot on the AVH-2 (which was identical to the tested knot) and in one case the tested knot ruptured. In neither cases did the removal of the sheath impact the strength of the whole chain. This was 20.1 ± 0.97 kN.

Test No. 9
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
9-1 18.65 kN 20.80 kN 20.12 kN 0.97 kN 0.73 kN anchoring knot on AVH-2
9-2 20.90 kN knot
9-3 20.80 kN anchoring knot on AVH-2

 

4.2.11  Test No. 10 – the apex of the loop formed by the two strands of the core and the identification tape only

The last of the tests was verifying the worst case variant of damage, i.e. a rope stripped of sheath and most of the core. Only two strands and an identification tape of rope (internal markings) remained at the apex of the loop. In this test, as in the only one, the loop ruptured at the site of damage in all tests, reaching a force of 9.4 ± 0.17 kN. Interesting thing is that the identification tape broke when a force of around 2 kN was applied (this does not mean that the identification tape will hold 2 kN, but that the remaining two springs were able to provide such support to the identification tape that it lasted up to 2 kN !!!).

Test No. 10
test breaking force median mean s. stand. deviation mean sq. error place of rupture of the rope
10-1 9.22 kN 9.23 kN 9.35 kN 0.17 kN 0.13 kN loop in the damage
10-2 9.60 kN loop in the damage
10-3 9.23 kN loop in the damage

 

5.    Discussion

During the tests, 33 alpine butterflies underwent a strength test, 30 of them with various damages. Only six of them ruptured at the damage (i.e. only ⅕) and the force required to rupture never fell below 9 kN. An average force of 18.64 ± 0.89 kN (median 19.75 kN) was required to rupture all damaged knots (regardless of whether they ruptured at the damage, at the test knot, or at the anchoring knot).

So is it possible to use the damaged loops of alpine butterflies to secure oneself when passing a knot? In order to obtain an answer, it is necessary to take into account the following facts.

The minimum force required for the rope breakage tests (test 10-1, where the apex of the test knot loop was comprised only by two strands and an identification tape) was 9.22 kN and the breakage occurred at the point of damage.

A rope access technician weighing 120 kg at rest exerts a force of approximately 1.18 kN towards the ground. In the event of a fall, the impact force exceeds this value, depending on the length of the fall, the lanyard used (type of cow’s tail), the distance (length of the rope) between the anchor point and the knot to be passed, and on the rope on which the alpine butterfly is tied (tightening the butterfly at the moment of fall is reflected in the impact energy), or on the use of a shock energy absorber.

The technical standards determine that the impact force must not exceed 6 kN[7]. It is also worth of remembering that, for example ascenders, it is required to withstand a force 4 kN[8]. In the case of accessory cords with a diameter of 4 mm, the standard is the required tensile strength of 3.2 kN, for a diameter of 5 mm 5 kN, for a diameter of 6 mm 7.2 kN and only for a diameter of 7 mm is the required minimum strength 9.8 kN[9].

Last but not least, when using standardized equipment, the connection of the rope access technician to the working rope is always doubled. Thus, even in the case where the rope access technician is secured by a cow’s tail to the loop of the knot, he does not hang only in the cow’s tail itself, but always has another means on the rope, usually an ascender. Although ascender is not designed to arrest a fall in the event of a ripped loop of the knot, it is also not designed to arrest a fall when the second ascender with which is used, fails, or when the second ascender damages the rope.

On the other hand, it should be recalled that the tests were carried out with new, unused 10.5 mm Courant Truck ropes manufactured in 2021. They do not take into account possible differences due to age or wear of the rope, different rope diameters, different rope models from the same manufacturer, ropes of other manufacturers or other relevant factors. Nor can it take into account the level of damage, which will never be the same. As a result, the damaged loop of the alpine butterfly may not withstand the force with which it withstood the tests performed.

Even so, securing into the damaged loop of an alpine butterfly should not endanger the user’s safety. Not only for a better feeling, but mainly in order to achieve maximum safety, it is appropriate to use a double alpine butterfly and secure oneself in both of its loops. Such securing certainly cannot endanger the rope access technician and in addition will speed up the passing of the knot.

 

6.    Conclusion

The purpose of the tests was to determine whether damage to the rope in the loop of an alpine butterfly would reduce the strength of the knot so that the loop could not be used to secure the rope access technician when passing the knot. Surprisingly, the reduction of strength caused by knotting was in the vast majority of cases more significant than the effect of rope damage. The rupture (loops, examined knots or anchoring knots) occurred in the range of forces 9.22 – 22.18 kN (on average at 18.64 ± 0.89 kN), which is probably sufficient to secure the rope access technician (observing the principle that in addition to securing to the knot another PPE against fall is attached to the rope).

Nevertheless, it is necessary to keep in mind that the values were measured in laboratory conditions on new rope specimens of one manufacturer, a specific type and the same diameter. At the same time, three tests for each of the tests may not be understood as sufficiently conclusive, on the other hand they can be used as a takeoff to perform further tests of a specific test, during which dozens of different ropes in different condition and different manufacturers will be tested.

Damage to the test specimens, with the exception of removal of the sheath (Tests No. 7-9) and removal of the sheath and more than half of the core (Test No. 10), also took the form of cutting the rope to half diameter with a hot knife cutter, so their extent was very similar. Thus, various other methods of mechanical damage deserve further research, whether caused by falling objects, damage of the rope by the edge, abrasion of part of the rope by other PPE against fall or other careless handling.

 

Acknowledgments

The article could not have been created (and the tests performed could not have taken place) without the support of Jaroslav Čech, who was able to respond flexibly to the need to supply new AVH-2 and AVD-2 clamping cylinders, which replaced those damaged in previous tests. Special thanks to the following entities (alphabetically) Courant, Lezectví.cz and Worksafety.cz for providing the test specimens, i.e. ropes, loops and connectors.

 

References

  • BELICA, Ondřej. Work and rescue at heights. Praha: Grada, 2014. ISBN 978-80-247-5055-2.
  • FRANK, Tomáš a Tomáš KUBLÁK. Climbing alphabet. Praha: Epocha, 2007. Horolezecká abeceda. ISBN 978-80-87027-35-6.
  • SMOLEK, Jan. Metrology – Evaluation of the accuracy of a measured quantity, Measurement uncertainties. Brno: SPŠS Brno, 2015.
  • ČSN 83 2610 Knots – Terminology, 2021.
  • ČSN EN 355 Personal protective equipment against falls from a height – Energy absorbers, 2003.
  • ČSN EN 363 Personal fall protection equipment – Personal fall protection systems, 2019.
  • ČSN EN 364 Personal protective equipment against falls from a height. Test methods, 1996.
  • ČSN EN 564 Mountaineering equipment – Accessory cord – Safety requirements and test methods, 2015.
  • ČSN EN 795 Personal fall protection equipment – Anchor devices, 2013.
  • ČSN EN 1891 Personal protective equipment for the prevention of falls from a height – Low stretch kernmantel ropes, 2000.
  • ČSN EN 12841 Personal fall protection equipment – Rope access systems – Rope adjustment devices, 2007.

[1] a test according to Subclause 5.6 of EN 1891, the rope is preloaded with a load of 50 kg for 5 minutes, then the load is increased to 150 kg and the elongation of the rope is measured.

[2] a three-meter rope sample was loaded 1350 mm from the anchor point with a load weighing 10 kg and every 300 mm the diameter was measured on two sides with a distance of 90° using a caliper

[3] Magnum Steel 2T, certified according to EN 362: 2004 B and EN 12275: 2013 B

[4] Open Sling Work 20 mm, certified according to EN 354: 2010 and EN 795: 2012 B

[5] t = t(P, n)

[6] the instruction manual states a 47% sheath strength of the total rope strength, see chapter 2.1 Rope

[7] e.g. EN 363, Subclause 4.3.3; EN 355, Sublause 6.4; EN 795 Clause 7, Paragraph b), and others

[8] EN 12841 Subclause 4.3.3; in the case of prEN 12841 then 4 kN or 1.4 times the maximum rated load, whichever is higher, see Subclause 4.3.3

[9] EN 564, Subclause 4.3

 

Ondřej Belica and Jan Smolek, 10. 8. 2021