An Investigation into the Premature Failure of the Wire Ropes
Guiding ‘Mallard’ Across Lake Windermere.
Dr. Bob Cattley October 2009.
Summary:
It is shown in this report that the wire ropes fitted to Mallard in spring 2009 are made of inferior material, compared with the original, and are of the wrong construction. Also the fairlead pulleys were incorrectly assembled. It is recommended that a return to the original rope design and pulley assembly be made at the first opportunity.
Contents:IntroductionPossible failure scenariosVisit 6/4/2009Characteristics of the failed wiresConstruction of the wire ropeChemical analysis of the failed ropeMicroscopical analysis of the failed wire ropeStress analysis of the wire ropeMechanical tests on the wire rope samplesDiscussionConclusionAppendix 1, load / extension and stress / strain
data for the unworn section of the broken ropeAppendix 2 wire rope productionAppendix 3 original wire rope specificationReferences
Introduction
The Mallard is a wire guided ferry boat, shown in figure 1, plying between Ferry Nab on the East Bank of Lake Windermere and Hawkeshead on the West Bank a distance of about 500m.
Figure 1, Mallard moored at Ferry Nab
Mallard is driven by two capstans mounted on the hull which engage with two parallel wire ropes fastened at the ends to Ferry Nab and the Hawkeshead landings. The capstans (drive rims) are driven by two internal diesel engines, main and stand by. These haul the boat across the lake. The ropes are guided onto and off the capstans by fairleads. Figure 2 shows the arrangement. Internally, the working diesel engine drives a shaft-mounted gearbox of normal design mounted on a shaft driving both capstans.
Figure 2, showing a drive capstan and the number 4 fairlead pulley
Orientation of the ferry means that fore – aft, port – starboard have no meaning in the conventional shipboard sense. If one stands on the ferry deck facing the Hawkeshead bank and assuming one is facing forward, the south, port or left side is known as the ‘wheel-house side’ and the north, starboard or right side is known as the ‘bulkhead side’. The corners of the ferry are numbered one to four counting clockwise in plan from the corner closest to the Ferry Nab bank on the wheel-house side. Figure 3 shows the arrangement.
Figure 3, orientation of the Windermere Ferry Mallard
The wire ropes are normally changed at twelve monthly intervals or when the rope stretches unevenly. Uneven stretch in the ropes is detected by observing the boat ‘crabbing’ as it crosses the lake.
Anecdotally, individual wire ropes normally do not fail during service in this application. The original wire ropes were made by a reputable UK manufacturer. A change was made to a lower cost rope of Indian manufacture (Usha Martin). It is understood that the low cost rope failed by stretching. The low cost rope endured about three months in service before becoming unfit for service. The failed rope was replaced by a new one from the same manufacturer and met with a similar fate, lasting three months of service. Again this was replaced by a rope from the same source this survived a similar length of time before failure. A forth rope was installed over the Christmas period 2008 – 9. This rope failed 7/4/2009. Accurate records of rope changes have been lost.
To avoid danger to other lake users the ferry cables lie on the lake bed and rise only as the ferry passes over.
The ropes are anchored at the Hawkeshead end and are tensioned by a simple winch and ratchet arrangement at Ferry Nab. The winches serve to equalise the length of rope on installation and can be used to account for uneven stretch in service.
The capstans or drive rims are made of cast steel with a nominal diameter of 1130 mm, Mr. J. Peart, private communication.
Costello gives minimum dimensions for wire rope sheaths, Costello 1985, page 37.5. Mallard’s drive rims are much larger than the minimum value quoted.
Possible failure scenarios
There are several possible causes of failure in wire rope leading to the failure of individual wires and the possible failure of the rope:
1. improper material
2. improper processing
3. overloading
4. fatigue
5. corrosion
6. abrasion
Improper material may be detected by chemical analysis and by examination of the material’s microstructure.
Improper processing is detected by tensile testing of individual wires, the composite rope and by examination of the material’s microstructure. The tensile strength of an individual Wire should be of the order of 1700 MN/m2. The microstructure should be fine pearlite, appendix 2.
Overloading of individual wires is usually caused by bending the wire over a small radius pulley or capstan. Detection of overloading would be by observation of the fracture surface and calculation of the maximum stress in the wire.
Fatigue is caused by repeated loading and unloading. This may be caused by the overloading regime described above. Detection of fatigue failure would be by observation of the fracture surface where characteristic ‘beach marks’ should be found and calculation of the maximum stress in the wire and comparing the number of loaded cycles with the expected fatigue life of the material.
Corrosion of individual wires by lake water or improper storage conditions will lead to a reduction in area of the wire this will cause high stress in the individual wire leading to failure. Detection of corrosion would be by observation of the fracture surface and the appearance of other wires in the rope.
Abrasion is most likely caused by incorrect lay of the rope for the service conditions, an incorrect profile on the capstan causing rubbing of wires against each other or slipping of the rope on the capstan. Detection of abrasion would be by observation of the damaged surface and adjacent sites in the wire. Abrasion should be observed in intact wires also.
Visit 6/4/2009
On 6/4/2009 the writer visited the ferry and was able to observe it in motion. Both wires had significantly stretched. The wire at the bulkhead side had visibly stretched more than the wheel-house side wire. The stretch was most easily observed as the ferry neared the slipway as the wire’s catenary at each side had a different profile and the bulkhead side one had the greatest deflection. Both wires were showing signs of serious wear. The south or wheel-house side showed least wear, figures 6 and 7. Each bundle of wires was worn at the outer surface figures 4 and 5 show magnified views of the worn ropes. The capstans appeared to have worn to a semicircular profile similar to the worn wire. There was a significant amount of slip (two to three full revolutions) of the capstans on the rope before they took up the drive when starting to cross the lake. The ferry proceeded across the lake at an estimated angle of between five and ten degrees. Noise from the capstans as the rope fed on and off was high.
Figure 4, magnified view of a short length of the worn wheel-house side rope
Figure 5, magnified view of a short length of the worn bulkhead side rope
Figure 6, worn wheel-house side rope
Figure 7, worn bulkhead side rope
Characteristics of the failed wires
On 7/4/2009 the rope on the bulkhead side failed. The number three or Hawkeshead north fairlead was found to be clogged with sisal fibres and the rope retaining bolt showed signs of damage caused by wire rope jamming in the fairlead pulley. The wire rope damage showed significant signs that it had been trapped, figure 8 and the rope had been unusually twisted, figure 9. It is concluded that the slack rope was able to leave the control of the fairlead pulley and that it jammed between the pulley and the retaining bolt. The rope subsequently was crushed from the normal circular shape, extruding the fibre core, and sliding in the available space. The wire spun sufficiently to plastically deform it, figure 10. Discussions with Mr. J. Peart revealed that the fairlead pulleys had been incorrectly assembled on the last reconditioning cycle. A spacer had been omitted from the assembly that was intended to retain the rope in the pulley groove.
Figure 8, failed wire rope showing evidence of trapping in the fairlead assembly at the start of the failure
It will be observed in figure 8 that the rope is circular in section at the far right. The section is deformed at about the 420 mm mark this is where the rope was first trapped between the retaining bolt and the pulley. The sisal core was extruded by the crushed rope and had parted by the 200 mm mark. At about the 100 mm mark a wire in one of the strands fractured. This can be clearly seen at the 150 mm mark where the writer pulled it out to make its presence clear. The fracture was at a site where the wire had very nearly worn through. The sisal core was wound onto the pulley spindle as the ferry continued on its way. The fractured wire was pushed along the rope by the ferry. The wire tried to uncoil its self from the wire but was prevented by the lay of the rope. The wire bunched up as the ferry travelled along. The bunch built up on the fairlead unable to pass through. The increasing bunch dimensions caused increasing resistance to the ferry and increased the load on the rope between the drive rim and the fairlead. Eventually the stress in the already weakened rope increased to the point where the rope parted. Drive was lost. The ferry was disabled in the middle of Lake Windermere.
As the rope was trapped in the fairlead and the ferry proceeded on its way the rope was forced to spin. It uncoiled in the direction of Ferry Nab and coiled tighter in the direction of the Hawkeshead landing. The stress involved was sufficient to exceed the elastic limit of the rope material as can be seen on the left hand side of figure 10.
Figure 9, failed wire rope showing the build up of the fractured wire at the Hawkeshead end of the failure which directly caused the failure by jamming the rope in the fairlead assembly
Visual examination of the fractured wires shows most wires failed in tension at stress raisers caused by the wear of the rope. A few wires failed in torsion as the rope finally parted.
Between the initiation of failure and the rope parting the ferry had travelled about 15 metres. There was nothing the ferry crew could have done to prevent the failure once the rope had left the fairlead.
The crew could not see the rope and did not know anything was going wrong until they had lost drive.
Figure 10, failed wire rope (Ferry Nab end in the middle of the picture) showing damage to the rope construction caused by passing it through the narrow gap between fairlead pulley and rope retaining bolt, note the plastically deformed (highly twisted) rope on the left of the picture, this is the Hawkeshead end of the failed rope
Construction of the wire rope
The failed wire is constructed conventionally with Lang’s lay consisting of six bundles of seven wires of 2.0 mm steel wire with a core of three 1.6 mm wires formed on to a sisal core. There is some grease used during construction of the rope present in all samples examined. The grease is fairly ‘heavy’ and does not seem to have washed out of the wire during service. A diagrammatic section of the rope is shown in figure 11.
Figure 11, diagrammatic section through the wire rope
Chemical analysis of the failed rope
A single 2mm wire of the failed rope was chosen at random for chemical analysis. To avoid contamination from wear debris derived from the drive rim the sample for analysis was taken from the unworn end of the rope anchored at Ferry Nab. The results of the analysis are given in table 1.
| C | Mn | Si | Ni | Cr | Mo | Cu | P | S |
| 0.48 | 0.75 | 0.21 | 0.06 | 0.01 | 0.02 | 0.02 | 0.01 | 0.007 |
Table 1, wet chemical analysis of the failed rope
The analysis corresponds to ordinary medium carbon steel. It is typical of BS 970:1956 En 43 or En 49A or B spring steel wire or to any of several American steels in the 104x and 105x range as listed in table 2. The levels of alloying elements; nickel, chromium, molybdenum and copper are low and are probably tramp from melting scrap rather than intended alloying elements. The values of phosphorous and sulphur are low indicating good steelmaking practice.
| Specification | C | Mn | Si | P | S | |
| En 43 | min | 0.45 | 0.60 | 0.10 | ||
| En 43 | max | 0.60 | 0.80 | 0.40 | 0.05 | 0.05 |
| En 49A | min | 0.40 | ||||
| En 49A | max | 0.85 | 1.00 | 0.3 | 0.05 | 0.05 |
| En 49B | min | 0.45 | ||||
| En 49B | max | 0.55 | 1.00 | 0.3 | 0.05 | 0.05 |
| SAE 1044 | min | 0.43 | 0.30 | 0.15 | ||
| SAE 1044 | max | 0.50 | 0.60 | 0.35 | 0.04 | 0.05 |
| SAE 1045 | min | 0.43 | 0.60 | 0.15 | ||
| SAE 1045 | max | 0.50 | 0.90 | 0.35 | 0.04 | 0.05 |
| SAE 1045 H | min | 0.42 | 0.50 | 0.15 | ||
| SAE 1045 H | max | 0.51 | 1.00 | 0.35 | 0.04 | 0.05 |
| SAE 1046 | min | 0.43 | 0.70 | 0.15 | ||
| SAE 1046 | max | 0.50 | 1.00 | 0.35 | 0.04 | 0.05 |
| SAE 1049 | min | 0.46 | 0.60 | 0.15 | ||
| SAE 1049 | max | 0.53 | 0.90 | 0.35 | 0.04 | 0.05 |
| SAE 1050 | min | 0.48 | 0.60 | 0.15 | ||
| SAE 1050 | max | 0.55 | 0.90 | 0.35 | 0.04 | 0.05 |
| SAE 1053 | min | 0.48 | 0.70 | 0.15 | ||
| SAE 1053 | max | 0.55 | 1.00 | 0.35 | 0.04 | 0.05 |
Table 2, limits on chemical composition for common similar alloys to the sample wire, references: Chandler 1995, Timken 2006 and Woolman and Mottram 1969
| Wire sample; calculated heat treatment parameters: | |
| CE = 0.528%C | |
| Ae1 = 716.4°C | Ae3 = 766.1°C |
| Ac1 = 720.2°C | Ac3 = 778.5°C |
| Martensite formation | |
| Ms = 307.1°C | |
| M10 = 297.1°C | |
| M50 = 259.9°C | |
| M90 = 204.1°C | |
| Mf = 92.1°C | |
| Bainite formation | |
| Bs = 628.3°C | |
| B50 = 568.3°C | |
| Bf = 508.3°C | |
Table 3, calculated heat treatment parameters for the wire sample, reference: Woolman and Mottram 1969
The carbon content is very low for steel suitable for the manufacture of high strength wire rope.
Microscopical analysis of the failed wire rope
A serving of 18 SWG tinned copper wire was applied to a section of the failed wire rope and was secured by soft soldering. The served section was cut out of the rope with a metallurgical sectioning saw. Three samples were taken from the served section and mounted using phenolic resin in the usual way. The samples were ground through increasingly finer grades of silicon carbide paper in a semi-automatic metallurgical polishing and grinding machine with tap water as lubricant. The sections were polished on ‘Leco Metcloth’ using 5, 1 and ¼ micron diamond paste. After cleaning the sections were etched with picral. They were briefly inspected under a microscope and re-polished and etched with picral.
At low magnification, figure 12, wear on the outside of the wires was obvious. Most of the outer wires had worn halfway through. Damage of the inside of the rope is also visible. Some of the outer wires have material from a passing wire or the drive rim friction welded to them. There is much wear debris in the wire interior.
At high magnifications, figure 13, the wire is seen to be fine grained ferrite (light etching) and pearlite (dark etching). Figure 13 is a longitudinal section of the wire. The structure represents heavily cold worked steel. It is unlikely that the material was subject to patenting during manufacture as one would expect a much finer structure than observed. The proportion of ferrite is large compared with what would be expected in a high strength wire rope.
Figure 12, section through failed wire rope showing macroscopic features (8x)
Figure 13, longitudinal section of a wire from the failed rope showing ferrite (light etching component) and pearlite (dark etching), etched with picral (2000 x)
Stress analysis of the wire rope
Bending a wire rope around a pulley or capstan induces a tensile stress on the outer wires in addition to any service load.
The stress due to bending is found by:
Where:
σ = stress due to bending
Er = Elastic modulus of the rope 90 x 109 [N/m2] assumed as this data is not known, this value is typical of wire rope
δ = diameter of the element under consideration 0.002 [m]
D = diameter of the drive rim 1.13 [m]
From appendix 1 the ultimate tensile strength of the wires was found to be 1400 [N/mm2]
Factor of safety (wire as found)
From appendix 3 the ultimate tensile strength of the original wires is quoted to be 1700 [N/mm2]
Factor of safety (wire specified) 
The stress due to bending is small but not insignificant compared with the ultimate tensile strength of the wire.
Mechanical tests on the wire rope samples
Seven samples of wire rope were supplied to the University these terminated in thimbles held with ‘talurit’ sleeves swaged on to the wire. There were four samples of worn and corroded wire and three samples of apparently new rope. There was an identification number stamped on each sample, these are recorded in table 4. The samples were tensile tested in the Avery 100 tonne universal tester with the following results:
Normal 0 false false false MicrosoftInternetExplorer4
| Sample No. | OM091339 | OM091340 | OM091341 | OM091342 | OM091343 | OM091344 | OM091349 | |
| condition | corroded | corroded | corroded | corroded | new | new | new | |
| diameter | 18.11 | 18.51 | 18.10 | 17.92 | 20.00 | 20.00 | 20.00 | mm |
| ultimate load | 158.0 | 174.0 | 154.0 | 159.0 | 202.0 | 204.0 | 195.0 | kN |
Table 4, tensile tests of the wire rope
Failure in all cases appeared to initiate within or at the exit of the swaged joint. It is concluded that failure of the test specimens was due to a tri-axial stress pattern caused by the combination of residual stress from the residual radial compression from fitting the talurit sleeve or the local stress of wire crossing under compression from the talurit sleeve as shown in figure 18 and the tensile stress induced by the testing machine.
It is significant that the corroded specimens failed in the same way as the new ones. It might have been expected that the corrosion and service wear could have produced stress raisers (notches) that could have initiated failure in the wire under test.
The wires sent for tensile testing were constructed differently from the wire rope that failed on 7/4/2009. Instead of the three core wires to each strand as shown in figure 11 a metal triangular core was found. The material that the core was made from is unknown. It is understood that all samples tested have come from the same batch of wire rope. The core as found in the new samples is quite soft and ductile. The triangular core from the corroded samples was very brittle. When a corroded sample was tensile tested fragments of the core flew about the laboratory at high speed. It is understood that the test specimens came from the same batch. If this is the case, the material of the cores displays some unusual properties.
The material is ferromagnetic and is difficult to etch with the usual reagents.
Figure 14, diagrammatic construction of the strands of rope supplied for tensile testing, compare with figure 11
Figure 15, micrograph of the unknown core material removed from one of the corroded wires sent for test, notice the crack in the upper half of the image, etched with alcoholic ferric chloride (800 x)
Discussion
The wires tested appear to be of 1400 N/mm2 grade material (appendix 1). No test certification appears to be available for the failed wire. No identifications could be found on the talurit sleeves or on the manufacturer’s tracer. Consequently, it is impossible to trace the wire back to source. Repeated emails to the maker’s offices in India and their UK representative requesting data for the rope have received no response. Mr. Peart indicated that 1700 N/mm2 was the grade of wire previously used (appendix 3). There does not appear to be a purchasing specification for the rope. The low strength steel would have a lower wear resistance than the original material. Also it would be more likely to noticeably stretch before tensile failure. The low grade wires stretched unevenly and these probably lead to the observed failure on 7/4/2009 as the slack rope lost the control of the fairlead pulley and jammed.
Figure 16, a section through the failed rope showing damage to individual wires
The failed rope was very badly worn as can be seen in figure 15. There is also evidence in figure 16 that the rope had also twisted significantly under load. The wear surfaces are not all at the rope periphery. Also evidence is present of wear taking place within the rope.
The failed rope and the samples supplied for testing were constructed using Lang’s lay, figure 17. From discussions with the ferry operators it appears that the ropes originally were constructed using normal lay, figure 18. When a wire rope is manufactured a bundle of wires (usually seven) is laid up into a strand. The wires in the strand are locked together by twisting. Several strands, six in this case, are laid up around a fibre core into the finished rope. It is possible to twist the rope to lock the strands together in either the same way as the strands (Lang’s lay) or in the opposite way (normal lay). Rope constructed using Lang’s lay has superior abrasion resistance as it allows mineral particles to embed between wires and these provide a sacrificial wear surface. Normal lay ropes have minimum twist when they are loaded.
Figure 17, Lang’s Lay wire ropes as found on Mallard
Figure 18, Regular Lay wire rope
As the rope is fed on and off the capstan the ropes rotate significantly on the drive rim and the fairlead pulleys. Wires in the outside of ropes made with Lang’s lay on the capstan are not aligned at the crossing and they abrade each other, figure 19. Outer wires of regular lay ropes at the crossing on the capstan are aligned and minimum wear takes place, figure 22.
Figure 19, wires crossing in Lang’s Lay wire ropes as found on Mallard note the way the wires in adjacent ropes make only point contact
Figure 20, wires crossing in Regular Lay wire rope, note the wires in adjacent ropes make line contact reducing the stress compared to figure 19 above
The main cause of wear of the ropes and the drive rim is caused by rotation of the wire due to the construction of the rope. Any slippage of the wire on the drive rim will result in wire constructed using Lang’s lay to abrade the rim much more severely than one of normal lay. For the rim to drive the ferry it must stretch the rope elastically. This is because the tension changes over the periphery of the drive rim from that required to pull the ferry against any resistance, plus the product of its mass and acceleration, where the rope initially engages the rim to close to zero where the rope leaves the rim. In addition, rotation of the rope as the tension changes around the rim would cause it to rotate and wear the rim around its periphery.
There is much anecdotal evidence from the ferry operators that the wire is stretching under load. This indicates that the stress in the wire during operation exceeds the elastic limit of the rope material. This is a most dangerous situation as wire ropes return suddenly to the coiled form from manufacture when the load is suddenly removed. The presence of the damping effect of the lake water may mitigate some of the effect of rope breakage. This should not be relied upon. The presence of members of the public on the ferry indicates that the highest possible standards of safety must be applied.
Conclusion
A thorough examination of the wire rope fitted to Mallard has been made and the following conclusions can be drawn:
The failed wires are made form an inferior material, compared with the specification.
The rope material is entirely unsuitable for this application.
The failed wires are constructed using the wrong lay.
A return to rope of original specification and correct assembly of the fairleads is recommended.
It is worth while to investigate the installation of an equalizer at the Hawkeshead end of the ferry run to share any subsequent uneven stretch in the ferry wires. This would prevent ‘crabbing’ of the ferry as the wires stretched.
The failure of the rope on 7/4/2009 was directly a consequence of its stretching and leaving the control of the fairlead pulley. The operators were aware of the problem. The operators could have prevented it if they had been allowed to correct the problem. It is the writer’s opinion that the ferry operators should be trained and authorised to tension the cables correctly during operation of the ferry, say, at the start of a shift. A method of reporting cable tensioning operations to the maintenance organisation should then be instituted. A closer relationship between the operators and the maintenance organization should be developed and maintained.
Appendix 1, load / extension and stress / strain data for the unworn section of the broken rope
Tensile testing of wire rope elements is difficult because the wires are of relatively small diameter and are hard. The small size and hardness make it difficult to securely grip the wires in the tensile test machine. The difficulty can be seen in figure A1, 1 where core 2 is seen to slip about 2 mm in the grips at a load of around 450 N. The following tests were conducted using a Hounsfield Tensometer with a capacity of 20 kN.
| Core 1 | Core 2 | Core 3 | Total load | ||||||
| load | N | 1742 | 1882 | 1591 | 5215 | ||||
| stress | N/mm2 | 866 | 936 | 791 | |||||
| Wire 1 | Wire 2 | Wire 3 | Wire 4 | Wire 5 | Wire 6 | Wire 7 | |||
| load | N | 4907 | 3772 | 3572 | 4675 | 3883 | 4833 | 5110 | 30752 |
| stress | N/mm2 | 1562 | 1175 | 1137 | 1488 | 1236 | 1539 | 1627 |
Table 5, breaking load and equivalent stress
Table 5 records the observed breaking loads and stresses for each element of the rope that failed. Specimens were taken from an unworn section of the rope. The average breaking load for the core wires is 1,738 N and 4,393 N for the main Wires.
The average ultimate tensile strength of the core wires is: 864 [N/mm2]
The average ultimate tensile strength of the outer wires is: 1400 [N/mm2]
When the elements are assembled into a rope each is twisted about its own bundle and also about the rope axis. As the rope is loaded in tension secondary stresses are induced in the rope from the pre-formed twists as they try to align themselves with the load direction. From table 5 it can be deduced that the breaking load for the rope is somewhat less (allowing for secondary stresses) than 6 x (5,215 + 30,752) = 6 x 35,967 = 215,802 N.
The tests recorded in table 4 indicate the failure load for the complete cable is about 200 kN.
Figure A1, 1, load / extension of the core wires
Figure A1, 2, stress strain graph for core wires
Figure A1, 3, load extension graphs of the main wires
Figure A1, 4, stress / strain curves for the wires
Appendix 2 wire rope production
Wire rope is manufactured from wire made from alloy, medium and high carbon steel (stainless steel and non-ferrous metals are also used but are not pertinent to this discussion). The selected steel is rolled to a suitable size for drawing to final size by conventional rolling mill equipment. Conventionally, the wire rod ‘patented’. The process can be viewed as a form of annealing or more correctly as ‘Austempering’. The wire rod is heated to a temperature above Ac3 (in the region of 970°C) and held there until large crystals of Austenite form. It is then quenched into a bath of lead or similar material held at about 510°C. The wire rod is held there for a sufficiently long time to ensure that the structure is 100% pearlite. To achieve 100% pearlite, there must be enough carbon in the steel to permit this (0.87%, Rollason 1973, page 161). In the rope examined in the body of this report there is insufficient carbon (0.48%) to achieve 100% pearlite. The wire rod is cooled to room temperature. Next the wire rod is drawn through successive dies to the desired diameter. The severe mechanical work done during the reduction in diameter as the wire is drawn causes the tensile strength and hardness to increase and the elongation to decrease. The structure of the wire consists of very fine pearlite.
Patenting is an expensive process. In an effort to reduce costs, wire for rope production is also made by ‘controlled cooling’. The wire rod leaving the rolling mill at about 950°C is cooled to 510°C by water sprays or other means and then allowed to cool to ambient temperature. The structure is ferrite and pearlite. When the rod leaves the rolling mill its structure is small elongated grains of Austenite. It is not held at a suitable temperature long enough for recrystallization to occur or to allow the new Austenite grains to grow to a sufficient size.
Appendix 3 original wire rope specification
Test certificates at Amey for the ferry wires.
Strength 1700 N/mm2
The first certificate states:-
Actual Breaking Load-249KN.
Min. Breaking Load- 231KN.
The second certificate states:-
Actual Breaking Load- 247KN.
Min. Breaking Load- 231KN.
Source; Mr. J. Peart, private communication.
These are significantly above the breaking load of the samples sent to the university recorded in table 4. The average breaking load of the new samples was 200 kN.
References
AISI, 1979, Wire Rope User’s Manual, American Iron and Steel Institute,
Baldo, A, F., 1987, Machine Elements in Avallone and Baumeister, Mark’s Standard Handbook for Mechanical Engineers, Ninth Edition, McGraw Hill, New York
Bramfit, B. L., 1991, Annealing of Steel, in, ASM Handbook Volume 4, Heat treating, ASM International, Metals Park, Ohio
Chandler, H, 1995, Heat Treater’s Guide, Practices and Procedures for Irons and Steels, Second Edition, ASM International, Metals Park, Ohio
Costello, G. A., 1985, Wire Rope in Mechanical Designs and Systems Handbook, Rothbart, H. A. (Editor), Second Edition, McGraw Hill, New York
Rollason, E. C., 1973, Metallurgy for Engineers, Edward Arnold, London
Shigley, J. E., 1986, Mechanical Engineering Design, First Metric Edition, McGraw Hill, New York
Timken, 2006, Practical Data for Metallurgists, Fifteenth Edition, Timken Ltd, USA
Woolman, J. and Mottram, R. A., 1969, the Mechanical and Physical Properties of the British Standard En Steels, Volume 3, En40 – En363, British Iron and Steel Research Association, Pergamon Press, London