what is breaking strength of wire rope free sample

Some of our calculators and applications let you save application data to your local computer. These applications will - due to browser restrictions - send data between your browser and our server. We don"t save this data.
Google use cookies for serving our ads and handling visitor statistics. Please read Google Privacy & Terms for more information about how you can control adserving and the information collected.

Wire ropes are essential for safety purposes on construction sites and industrial workplaces. They are used to secure and transport extremely heavy pieces of equipment – so they must be strong enough to withstand substantial loads. This is why the wire rope safety factor is crucial.
You may have heard that it is always recommended to use wire ropes or slings with a higher breaking strength than the actual load. For instance, say that you need to move 50,000 lbs. with an overhead crane. You should generally use equipment with a working load limit that is rated for weight at least five times higher – or 250,000 lbs. in this case.
This recommendation is all thanks to the wire rope safety factor. This calculation is designed to help you determine important numbers, such as the minimum breaking strength and the working load limit of a wire rope.
The safety factor is a measurement of how strong of a force a wire rope can withstand before it breaks. It is commonly stated as a ratio, such as 5:1. This means that the wire rope can hold five times their Safe Work Load (SWL) before it will break.
So, if a 5:1 wire rope’s SWL is 10,000 lbs., the safety factor is 50,000 lbs. However, you would never want to place a load near 50,000 lbs. for wire rope safety reasons.
The safety factor rating of a wire rope is the calculation of the Minimum Break Strength (MBS) or the Minimum Breaking Load (MBL) compared to the highest absolute maximum load limit. It is crucial to use a wire rope with a high ratio to account for factors that could influence the weight of the load.
The Safe Working Load (SWL) is a measurement that is required by law to be clearly marked on all lifting devices – including hoists, lifting machines, and tackles. However, this is not visibly listed on wire ropes, so it is important to understand what this term means and how to calculate it.
The safe working load will change depending on the diameter of the wire rope and its weight per foot. Of course, the smaller the wire rope is, the lower its SWL will be. The SWL also changes depending on the safety factor ratio.
The margin of safety for wire ropes accounts for any unexpected extra loads to ensure the utmost safety for everyone involved. Every year there aredue to overhead crane accidents. Many of these deaths occur when a heavy load is dropped because the weight load limit was not properly calculated and the wire rope broke or slipped.
The margin of safety is a hazard control calculation that essentially accounts for worst-case scenarios. For instance, what if a strong gust of wind were to blow while a crane was lifting a load? Or what if the brakes slipped and the load dropped several feet unexpectedly? This is certainly a wire rope safety factor that must be considered.
Themargin of safety(also referred to as the factor of safety) measures the ultimate load or stress divided by theallowablestress. This helps to account for the applied tensile forces and stress thatcouldbe applied to the rope, causing it to inch closer to the breaking strength limit.
A proof test must be conducted on a wire rope or any other piece of rigging equipment before it is used for the first time.that a sample of a wire rope must be tested to ensure that it can safely hold one-fifth of the breaking load limit. The proof test ensures that the wire rope is not defective and can withstand the minimum weight load limit.
First, the wire rope and other lifting accessories (such as hooks or slings) are set up as needed for the particular task. Then weight or force is slowly added until it reaches the maximum allowable working load limit.
Some wire rope distributors will conduct proof loading tests before you purchase them. Be sure to investigate the criteria of these tests before purchasing, as some testing factors may need to be changed depending on your requirements.
When purchasing wire ropes for overhead lifting or other heavy-duty applications, understanding the safety dynamics and limits is critical. These terms can get confusing, but all of thesefactors serve an important purpose.
Our company has served as a wire rope distributor and industrial hardware supplier for many years. We know all there is to know about safety factors. We will help you find the exact wire ropes that will meet your requirements, no matter what project you have in mind.

Wire rope is also known by many other names, such as: wire, multi-strand wire, flexible wire, cable, cord, steelcord, etc. but it is essentially a collection of small filaments wound around each other in a manner that largely retains its shape when bent, crushed and/or tensioned.
It is a system for significantly increasing the strength and flexibility of steel wire and is used in almost every important application we see around us. For example: suspension bridges, tyres, brake and accelerator cables (in cars), high-pressure flexible pipes, lifting and rigging cables, electrical conductors, etc. and it comes in many different forms. Fig 2 shows just a very small sample of available designs.
With minor variations, the generally accepted method for designating a wire rope construction in the industry is by describing it numerically. For example:
"0.43+6x0.37+6x(0.37+6x0.33) HT" refers to a seven strand construction: a single central strand (one central filament diameter 0.43mm and 6 planetary filaments of diameter 0.37mm) and 6 planetary strands (one central filament of diameter 0.37mm and 6 planetary filaments of diameter 0.33mm) all manufactured from high-tensile steel"
Whilst "IWRC" wire ropes offer a slightly greater tensile capacity (≈7%) than those with fabric or polymer fillers, the additional strength does not come from the tensile capacity of the core filaments but from improved dimensional stability under load. And whilst they are also much more resistant to crushing, they are stiffer than fibre core ropes and therefore not recommended for applications where tension occurs under bending.
Warrington (Fig 1) is a parallel lay construction with an outer layer comprising wires of alternating large and small diameters, each outer layer having twice the number of wires as the layer immediately beneath. The benefit of this design is to increase packing and therefore strength density, however, unless the different diameter filaments are of the same strength (unlikely), this construction is limited by the strength of the weakest filaments.
Seale (Figs 1 & 2 6x36) is also a parallel lay construction but with the same number of wires in each wire layer. All the wires in any layer are the same diameter. This is an alternative to the Warrington construction, with similar benefits and disadvantages.
Tyrecord generally comprises a single strand of less than 1.5mm in diameter and normally contains about 12 filaments all of the same diameter between 0.15mm and 0.25mm, but designs and configurations can vary considerably dependent upon manufacturer and tyre design requirements. This design tends to be the most flexible of all constructions.
OTR is more or less a complicated tyrecord construction (see above) up to 4.5mm in diameter containing around 100 filaments of a similar size to tyrecord, albeit towards the larger end of the size range (0.2mm to 0.25mm).
Regular lay constructions are used much more widely (than Lang lay) because they have excellent structural stability and less tendency to unwrap under tension (see Rotating vs Non-Rotating below). However, because it has a knobbly (undulating) surface it will wear both itself and any surface over which it is run much more quickly than Lang lay wire rope.
Lang lay constructions have a flatter surface than regular lay constructions giving them better resistance to wear and bending fatigue, especially when made from flattened (elliptical) filaments. They are, however, much less structurally stable and subject to birdcaging if the wire rope is over-bent or twisted against its wrapped direction.
"Regular Lay", multi-strand constructions are normally subject to slightly less rotation under tension (than Lang lay) due to the opposite helical direction of the filaments (within the strands) and the strands (within the rope), however, you can improve their rotation characteristics still further by;
Whilst there are very distinct non-rotating constructions such as 19x7 and rotation-resistant designs such as 19x19, new ideas are evolving all the time and each manufacturer will have its own design preferences.
"Lang lay" and single strand (e.g. Fig 2 1x7) constructions will always try to straighten (unwrap) under tension. There are a number of things that could be done to minimise this problem, such as a) to c) above and/or;
Fillers (Fig 2) may be fabric, polymer or even smaller diameter filaments (e.g. 6x36). Whilst they contribute little to the tensile strength of wire rope, they can significantly; improve performance under bending (fabric and polymer cores only), reduce axial growth, reduce rotation in rotation-resistant constructions, improve structural stability and increase fatigue life.
There is little point in having a central core manufactured from the same material as the filaments as it will be the first to break. If you need a metal core, this should be of a material with lower axial stiffness than the strand that surrounds it.
This filler material should not be included in strength (tensile capacity) calculations, but must be included in those for axial stiffness (extension). If it is ignored, your calculations will reveal excessive extension as the wire rope collapses.
Suspension bridges tend to be constructed from densely packed, single strand plain "Wire Rope" constructions using large diameter galvanised filaments. Little heed is paid to rotational resistance as strength is paramount and once tensioned, they should remain in that loading condition for their design life.
Lifting & winching normally require wire ropes of good flexibility and fatigue resistance. Therefore they tend to be similar to 6x36 but with fibre core instead of the IWRC in Fig 2
Hosecord is suitable for HPHT flexible pipes as lateral flexibility is generally considered less important than minimal longitudinal growth or maximum tensile strength (per unit cross-sectional area).
Remote operating cables such as hand-brakes and accelerators on cars normally only work in tension so they need to be strong but not necessarily stiff (as they are fully contained in reinforced outer sheaths). These tend to be manufactured from large diameter "TyreCord" or small diameter single-strand "Wire Rope".
Axial stiffness is the linear relationship between axial strain and force that allows us to predict the condition of any material or structure when exposed to a specified tensile force. However, it works only with materials and structures that obey Hooke"s law.
Wire rope does not obey Hooke"s law. Therefore, you cannot accurately predict how much it will stretch for any specified force. This unpredictability applies to any section removed from the same manufactured length of cord and even between cords produced to the same specification but by different manufacturers.
CalQlata has decided that the accuracy of axial stiffness (EA) of wire rope falls outside its own levels of acceptability and therefore does not include it in the wire rope calculator. The extension calculated in the Wire Rope calculator (δLᵀ) is based upon the effect of axial tension on packing density. It is therefore important that core material is not ignored when using the calculator to evaluate this characteristic.
Torsional stiffness is the linear relationship that allows us to predict the rotation of any material or structure when exposed to a torque. However, it works only with materials and structures that obey Hooke"s law.
Wire rope does not obey Hooke"s law. Therefore, you cannot accurately predict how much it will twist for any specified torque. This unpredictability applies to any section removed from the same manufactured length of cord and even between cords produced to the same specification but by different manufacturers.
CalQlata has decided that the accuracy of torsional stiffness (GJ) of wire rope falls outside its own levels of acceptability and therefore does not include it in the wire rope calculator.
1) No wire rope calculator, whether dedicated or generic, will accurately predict the properties of any single construction under a wide range of loading conditions
2) No wire rope calculator, whether dedicated or generic, will accurately predict any single property for a range of constructions under a wide range of loading conditions
3) Unless additional heat treatment or material modification is performed during the manufacturing (drawing) process, the smaller the filament diameter the greater will be its SMYS
The only wire rope that can be reliably analysed is that which is used for suspension bridges, because; it comprises a single strand, is very densely packed, has negligible twist, contains filaments of only one diameter, is never subjected to minimum bending and every filament is individually tensioned.
There is a very good reason why manufacturers do not present calculated performance data for construction or design proposals, because even they cannot accurately predict such properties and quite rightly rely on, and publish, test data.
During his time working in the industry, the wire rope calculator"s creator has seen, created and abandoned numerous mathematical models both simple and complex. He has gradually developed his own simplified calculation principle based upon his own experience that still provides him with consistently reliable results of reasonable accuracy.
The purpose of CalQlata"s wire rope calculator is to provide its user with the ability to obtain a reasonable approximation for a generic construction, after which, accurate test data should be sought from the manufacturer for the user"s preferred construction.
The calculation principle in the wire rope calculator is based upon changes in the properties of the wire rope that occur with variations in packing density under tension
Bearing in mind the above limitations CalQlata can provide the following assistance when generating (manipulating) the wire rope calculator"s input data and interpreting its output
Alternatively, for wire rope with multiple filament diameters, you need to find an equivalent diameter with the following proviso; you must enter the minimum filament yield stress (SMYS)
It is expected that apart from fillers, all the material in the wire rope will be identical and therefore have the same density, i.e. using different materials will result in less than "best" performance. However, if such a construction is proposed, you can calculate an equivalent density as follows:
It is expected that apart from fillers, all the material in the wire rope will be identical and therefore have the same tensile modulus, i.e. using different materials will result in less than "best" performance. However, if such a construction is proposed, you should enter the highest tensile modulus.
The wire rope calculator simply adds together the total area of all the filaments and multiplies them by the SMYS entered, which represents a theoretical maximum breaking load that would exist if this load is equally shared across all of the filaments and the lay angles have been arranged to eliminate localised (point) loads between adjacent filaments.
If the wire rope has been properly constructed it is likely that its actual break load will be greater than 80% of this theoretical value. However, given the vagaries of wire rope construction, the actual break load can vary considerably dependent upon a number of factors. CalQlata suggest that the following factors may be used to define the anticipated break load of any given construction:
Accuracy is expected to be within ±0.1% of the calculated value for good quality manufacturing but variations in manufactured filament diameter can, in extreme cases, reduce this to ±1%
The axial stiffness and strain under load will be affected by this value, hence the reason why the most reliable (predictable) constructions tend to be minimum [number of] strands and single filament diameter. The Warrington and Seale constructions and combinations thereof tend to provide the highest packing density (but lowest flexibility) and there is little to be gained from using these constructions in more than single stranded wire rope as the benefit of high-packing density will be lost with no gain in flexibility.
The anticipated second moment of area of the wire rope at tension "T" due to deformation but insignificant flattening as it is assumed the wire rope will be bent over a formed (shaped) sheave or roller.
The anticipated tensile modulus of the wire rope at tension "T" due to deformation but insignificant flattening as it is assumed the wire rope will be bent over a formed (shaped) sheave or roller.
It is not advisable to induce this bend radius in operation due to uncertainties associated with wire rope construction, especially for dynamic applications. CalQlata suggests that a similar approach to that used for the break load (Fb) above also be applied here, i.e.:
A change in diameter will occur in all wire rope, irrespective of construction, until packing density has reached a limiting value. The value provided in the wire rope calculator is that which would be expected if the construction remains intact at the applied tension "T"
Unreliability of this value increases with complexity in wire rope due to its longitudinal variability and the increased likelihood of premature failure.
The accuracy of this data will range from about ±1% for wire rope with a single strand and a single filament diameter, up to about ±15% for constructions of similar complexity to OTR cord
A change in length of any wire rope will occur due to the fact that the packing density increases with tension. This is not, however, a linear relationship.
This can be an unreliable value as illustrated by tests carried out (by the author) on two pieces of wire rope supplied by the same well-known manufacturer both of which were cut from the same length, varied in tensile capacity by only 1.5%, but the tensile modulus (and strain at break) varied by 34%. Whilst this was an extreme case, significant variations have been seen in wire rope manufactured by a number of manufacturers.
Whilst the wire rope calculator does not calculate axial stiffness (see Calculation Limitations 9) above), CalQlata can suggest the following rule-of-thumb that will provide reasonable results for most constructions at the applied tension "T":
Where: θ = the "absolute" sum of the average filament lay angle and the average strand lay angle⁽²⁾. Note; the angle of twist (θ) will reduce as tension approaches break load.
Whilst the wire rope calculator does not calculate bending stiffness (see Calculation Limitations 8) above), CalQlata can suggest the following rule-of-thumb that will provide reasonable results for most constructions at the applied tension "T":
Low complexity means single strand and single wire diameter. Medium complexity means multi-strand and single wire diameter. High complexity means multi-strand and multiple wire diameters.

Rope strength is a misunderstood metric. One boater will talk about tensile strength, while the other will talk about working load. Both of these are important measurements, and it’s worth learning how to measure and understand them. Each of these measurements has different uses, and here we’re going to give a brief overview of what’s what. Here’s all you need to know about rope strength.
Each type of line, natural fiber, synthetic and wire rope, have different breaking strengths and safe working loads. Natural breaking strength of manila line is the standard against which other lines are compared. Synthetic lines have been assigned “comparison factors” against which they are compared to manila line. The basic breaking strength factor for manila line is found by multiplying the square of the circumference of the line by 900 lbs.
When you purchase line you will buy it by its diameter. However, for purposes of the USCG license exams, all lines must be measured by circumference. To convert use the following formula.
As an example, if you had a piece of ½” manila line and wanted to find the breaking strength, you would first calculate the circumference. (.5 X 3.14 = 1.57) Then using the formula above:
To calculate the breaking strength of synthetic lines you need to add one more factor. As mentioned above, a comparison factor has been developed to compare the breaking strength of synthetics over manila. Since synthetics are stronger than manila an additional multiplication step is added to the formula above.
Using the example above, letÂ’s find the breaking strength of a piece of ½” nylon line. First, convert the diameter to the circumference as we did above and then write the formula including the extra comparison factor step.
Knots and splices will reduce the breaking strength of a line by as much as 50 to 60 percent. The weakest point in the line is the knot or slice. However, a splice is stronger than a knot.
Just being able to calculate breaking strength doesn’t give one a safety margin. The breaking strength formula was developed on the average breaking strength of a new line under laboratory conditions. Without straining the line until it parts, you don’t know if that particular piece of line was above average or below average. For more information, we have discussed the safe working load of ropes made of different materials in this article here.
It’s very important to understand the fundamental differences between the tensile strength of a rope, and a rope’s working load. Both terms refer to rope strength but they’re not the same measurement.
A rope’s tensile strength is the measure of a brand-new rope’s breaking point tested under strict laboratory-controlled conditions. These tests are done by incrementally increasing the load that a rope is expected to carry, until the rope breaks. Rather than adding weight to a line, the test is performed by wrapping the rope around two capstans that slowly turn the rope, adding increasing tension until the rope fails. This test will be repeated on numerous ropes, and an average will be taken. Note that all of these tests will use the ASTM test method D-6268.
The average number will be quoted as the rope’s tensile strength. However, a manufacturer may also test a rope’s minimum tensile strength. This number is often used instead. A rope’s minimum tensile strength is calculated in the same way, but it takes the average strength rating and reduces it by 20%.
A rope’s working load is a different measurement altogether. It’s determined by taking the tensile strength rating and dividing it accordingly, making a figure that’s more in-line with an appropriate maximum load, taking factors such as construction, weave, and rope longevity into the mix as well. A large number of variables will determine the maximum working load of a rope, including the age and condition of the rope too. It’s a complicated equation (as demonstrated above) and if math isn’t your strong point, it’s best left to professionals.
However, if you want to make an educated guess at the recommended working load of a rope, it usually falls between 15% and 25% of the line’s tensile strength rating. It’s a lotlower than you’d think. There are some exceptions, and different construction methods yield different results. For example, a Nylon rope braided with certain fibers may have a stronger working load than a rope twisted out of natural fibers.
For safety purposes, always refer to the information issued by your rope’s manufacturer, and pay close attention to the working load and don’t exceed it. Safety first! Always.
If you’re a regular sailor, climber, or arborist, or just have a keen interest in knot-tying, be warned! Every knot that you tie will reduce your rope’s overall tensile strength. Some knots aren’t particularly damaging, while others can be devastating. A good rule of thumb is to accept the fact that a tied knot will reduce your rope’s tensile strength by around 50%. That’s an extreme figure, sure, but when it comes to hauling critical loads, why take chances?
Knots are unavoidable: they’re useful, practical, and strong. Splices are the same. They both degrade a rope’s strength. They do this because a slight distortion of a rope will cause certain parts of the rope (namely the outer strands) to carry more weight than others (the inner strand). In some cases, the outer strands end up carrying all the weight while the inner strands carry none of it! This isn’t ideal, as you can imagine.
Some knots cause certain fibers to become compressed, and others stretched. When combined together, all of these issues can have a substantial effect on a rope’s ability to carry loads.
Naturally, it’s not always as drastic as strength loss of 50% or more. Some knots aren’t that damaging, some loads aren’t significant enough to cause stress, and some rope materials, such as polypropylene, Dyneema, and other modern fibers, are more resilient than others. Just keep in mind that any knots or splices will reduce your rope’s operations life span. And that’s before we talk about other factors such as the weather or your rope care regime…

The tensile strength is the load at which a new rope, tested under laboratory conditions, can be expected to break. Rope strength is the approximate average for new rope tested under ASTM test method D-6268. To estimate the minimum tensile strength of a new rope, reduce the approximate average by 20%. Age, use and the type of termination used, such as knots, will lower tensile strengths significantly.
One area of misunderstanding that needs to be brought to the surface is the proper interpretation of rope strength, appropriate usage and care. Let"s start by defining two important terms: "tensile strength" and "working load". Tensile strength is the average strength ofnewrope under laboratory conditions. This is determined by wrapping the rope around two large diameter capstans and slowly adding tension to the line until it breaks. The manufacturer"s recommended working load is determined by taking the tensile strength and dividing it by a factor that more accurately reflects the maximum load that should be applied to a given rope to assure a comfortable safety margin and longevity of the line. Of course that factor varies with the type of fiber and the weaving construction. There are however always exceptions, most notably the fact that rope is susceptible to degradation and damage in numerous ways that are not controllable by the manufacturer.
It may surprise you to find out that the working load for most kinds of rope is between 15% and 25% of the tensile strength. Now consider the fact that any time you tie a knot in a rope you effectively cut the tensile strength in half. The knot when tensioned cuts the line. While certain kinds of knots damage the line less than others, the 50% loss of tensile strength is a good general rule to live by. Research has shown that the figure 8 knot reduces the tensile strength by approximately 35% instead of 50% for other common knots tested.
At Ravenox, we use a third-party mechanical services company to test the failure point or break strength of our ropes. There are two common types of breaks: the sharp break and the percentage break. The sharp break is referred to the measurement when load or force drops by 5% from its peak load measurement. A percentage break is another form of break and is generally determined by the sample material and its relationship to load degradation from a peak load measurement. We measure the percentage break.

As a wire rope is used, the outer wires wear through abrasion and so the rope suffers loss of cross-sectional area – this obviously reduces the breaking strength of the wire rope. Resistance to abrasive wear is therefore an important property of a wire rope.
Abrasion resistance is directly related to the design of the rope, in particular the design of the strands of the rope. In general, ropes with fewer larger wires will be more abrasion resistant than a similar rope made up of smaller wires – a 6 x 19 rope will therefore be more abrasion resistant than a 6 x 36 rope.
In a later article in this technical series we will discuss fatigue resistance in great detail but for the purposes of this discussion, we must recognise that a cyclic stress reversal occurs when a body is subjected to alternating tensile and compressive loads.
In a drilling operation, wire ropes run through sheaves constantly and so are constantly subjected to alternating tensile and compressive loads – i.e. cyclic stress reversals.
Figure 1 illustrates a wire rope bending over a sheave. It is clear that the outer parts of the rope running over the sheave are in tension and the inner parts of the rope running over the sheave are in compression and as the rope moves over the sheave these stresses reverse.
It should be obvious that the smaller the diameter of the sheave the greater the magnitude (amplitude) of the stress reversal and so the more rapidly fatigue will occur in the rope.
Wire ropes experience external forces that will tend to alter or distort the shape of the rope. Crushing prevents wires and strands moving easily over one another during normal operation and this can lead to accelerated wear and reduced rope life.

The manufacturers of a new type of wire rope terminal conducted a destruction test. This test resulted in the sample of wire rope breaking at exactly 10,000 lbs. The wire broke right next to one of the terminals. The rated strength of the wire rope was also 10,000 lbs.
This is an obscure one, so here is what I hope is a large hint: wire rope must meet, within very narrow tolerances, a minimum breaking strength, or it cannot be sold. Bonus points if you can tell what detail in my description of the test might have made the technician skeptical.
The solution lies in the term “rated strength.” By international standards, average strengths, as tested to destruction, must be not more than 2.5% weaker than the published rated strength (more on this later). Since break strengths can vary significantly from batch to batch, manufacturers will see to it that their wires are significantly stronger than the official requirement, sometimes by as much as 20%.
So when the tests on the new terminal were right at the official strength, the technician was suspicious; if the terminal really was 100% efficient, one would expect the samples to break well above the rated strength.
The puzzle’s bonus points answer was the fact that the wires broke right at the terminal. This usually only happens when the terminal creates a stress riser, weakening the wire so that it breaks adjacent to that stress riser – the terminal itself. With a truly efficient terminal, the wire might be expected to break at least 2.5” away from the terminal, or “in the clear.” Where the wire broke confirmed the technician’s suspicions.
This puzzle has real-world significance. For instance, Norseman terminals were for years touted as being at or near 100% efficient, based on tests just like the above. But when the actual strength of the wire was known, they struggled to achieve, you guessed it, 80% efficiency. Happily, those terminals are no longer being made.
If you would like to see some destruction test results for sailboat terminals, based on actual wire strength, I can recommend the tests carried out by yours truly for Practical Sailor. The article is for subscribers only, but it is worth the subscription fee, and not just for that one article. Sign up here: https://www.practical-sailor.com/
For those of you who would like a more technical description of terminal efficiency, you might enjoy this information-dense entry by longtime reader and entrant Mike Madden:
If the manufacturer breaks x samples of his new wire rope, the result must surely be x different breaking strength values normally distributed (probably) about a mean . However, the minimum accepted strength can only be 2 1/2% lower than the published nominal breaking strength*. That is 2 standard deviations (SD) below the mean breaking strength of the wire rope. Since 6 SDs includes 99.9% of values, that suggests that the highest breaking strength found during testing would be about 5 SD above the published minimum. I couldn’t find any wire rope breaking strength data that included Sds, but in a thread in a Sailing Anarchy forum, a test of Dyneema break strength test found that the SD of the sample breaks was 3%. Using that value, 3% of 10000 llb is 300 lb, so 5 SD above 10000 lb would be 11500 lb. Whew!
Anyway, given that series of guesstimates, 10000 lb is 15% lower than the tested maximum breaking strength of the wire, which is what the testing machine said. Since break tests for wire rope are only valid if the break occurs 2 or more inches from from the ends, and this break occurred at the terminal, the testing operator suspects the terminal as the cause of the failure.
Ain’t statistics grand? Mike also included a link to a website describing the formal standards behind rated wire rope strengths: http://vernoncorp.com/products/black-diamond-wire-rope/design-factors-breaking-strengths/.
We got some splendid entries to this week’s puzzle, and most of you got the basic answer and the bonus question. But there can only be one winner, chosen at random from the pool of correct entries, and this week that winner was … Tim Omlor! Congratulations, Tim, you have just won my new E-Book, Falling, a series of essays on the topic of, well, falling. We are in the final stages of editing, but I hope it will be ready to send in less than a month. Be in touch to remind me, please.
Meanwhile, if you enjoyed this puzzle, you might enjoy all the other ones we’ve posted, and maybe even the now-considerable number of essays that have accumulated on my blog. To find and binge-read, see: Our Blog
If you are reading this on Wednesday morning, it is because you are a subscriber to this blog, and you get to see it a day ahead of everyone else. If you are not yet a subscriber, consider scrolling to the bottom of this page, and signing up. It’s free.

Wire rope and cable are each considered a “machine”. The configuration and method of manufacture combined with the proper selection of material when designed for a specific purpose enables a wire rope or cable to transmit forces, motion and energy in some predetermined manner and to some desired end.
Two or more wires concentrically laid around a center wire is called a strand. It may consist of one or more layers. Typically, the number of wires in a strand is 7, 19 or 37. A group of strands laid around a core would be called a cable or wire rope. In terms of product designation, 7 strands with 19 wires in each strand would be a 7×19 cable: 7 strands with 7 wires in each strand would be a 7×7 cable.
Materials Different applications for wire rope present varying demands for strength, abrasion and corrosion resistance. In order to meet these requirements, wire rope is produced in a number of different materials.
Stainless Steel This is used where corrosion is a prime factor and the cost increase warrants its use. The 18% chromium, 8% nickel alloy known as type 302 is the most common grade accepted due to both corrosion resistance and high strength. Other types frequently used in wire rope are 304, 305, 316 and 321, each having its specific advantage over the other. Type 305 is used where non-magnetic properties are required, however, there is a slight loss of strength.
Galvanized Carbon Steel This is used where strength is a prime factor and corrosion resistance is not great enough to require the use of stainless steel. The lower cost is usually a consideration in the selection of galvanized carbon steel. Wires used in these wire ropes are individually coated with a layer of zinc which offers a good measure of protection from corrosive elements.
Cable Construction The greater the number of wires in a strand or cable of a given diameter, the more flexibility it has. A 1×7 or a 1×19 strand, having 7 and 19 wires respectively, is used principally as a fixed member, as a straight linkage, or where flexing is minimal.
Cables designed with 3×7, 7×7 and 7×19 construction provide for increasing degrees of flexibility but decreased abrasion resistance. These designs would be incorporated where continuous flexing is a requirement.
Selecting Wire Rope When selecting a wire rope to give the best service, there are four requirements which should be given consideration. A proper choice is made by correctly estimating the relative importance of these requirements and selecting a rope which has the qualities best suited to withstand the effects of continued use. The rope should possess:Strength sufficient to take care of the maximum load that may be applied, with a proper safety factor.
Strength Wire rope in service is subjected to several kinds of stresses. The stresses most frequently encountered are direct tension, stress due to acceleration, stress due to sudden or shock loads, stress due to bending, and stress resulting from several forces acting at one time. For the most part, these stresses can be converted into terms of simple tension, and a rope of approximately the correct strength can be chosen. As the strength of a wire rope is determined by its, size, grade and construction, these three factors should be considered.
Safety Factors The safety factor is the ratio of the strength of the rope to the working load. A wire rope with a strength of 10,000 pounds and a total working load of 2,000 pounds would be operating with a safety factor of five.
It is not possible to set safety factors for the various types of wire rope using equipment, as this factor can vary with conditions on individual units of equipment.
The proper safety factor depends not only on the loads applied, but also on the speed of operation, shock load applied, the type of fittings used for securing the rope ends, the acceleration and deceleration, the length of rope, the number, size and location of sheaves and drums, the factors causing abrasion and corrosion and the facilities for inspection.
Fatigue Fatigue failure of the wires in a wire rope is the result of the propagation of small cracks under repeated applications of bending loads. It occurs when ropes operate over comparatively small sheaves or drums. The repeated bending of the individual wires, as the rope bends when passing over the sheaves or drums, and the straightening of the individual wires, as the rope leaves the sheaves or drums, causing fatigue. The effect of fatigue on wires is illustrated by bending a wire repeatedly back and forth until it breaks.
The best means of preventing early fatigue of wire ropes is to use sheaves and drums of adequate size. To increase the resistance to fatigue, a rope of more flexible construction should be used, as increased flexibility is secured through the use of smaller wires.
Abrasive Wear The ability of a wire rope to withstand abrasion is determined by the size, the carbon and manganese content, the heat treatment of the outer wires and the construction of the rope. The larger outer wires of the less flexible constructions are better able to withstand abrasion than the finer outer wires of the more flexible ropes. The higher carbon and manganese content and the heat treatment used in producing wire for the stronger ropes, make the higher grade ropes better able to withstand abrasive wear than the lower grade ropes.
Effects of Bending All wire ropes, except stationary ropes used as guys or supports, are subjected to bending around sheaves or drums. The service obtained from wire ropes is, to a large extent, dependent upon the proper choice and location of the sheaves and drums about which it operates.
A wire rope may be considered a machine in which the individual elements (wires and strands) slide upon each other when the rope is bent. Therefore, as a prerequisite to the satisfactory operation of wire rope over sheaves and drums, the rope must be properly lubricated.
Loss of strength due to bending is caused by the inability of the individual strands and wires to adjust themselves to their changed position when the rope is bent. Tests made by the National Institute of Standards and Technology show that the rope strength decreases in a marked degree as the sheave diameter grows smaller with respect to the diameter of the rope. The loss of strength due to bending wire ropes over the sheaves found in common use will not exceed 6% and will usually be about 4%.
The bending of a wire rope is accompanied by readjustment in the positions of the strands and wires and results in actual bending of the wires. Repetitive flexing of the wires develops bending loads which, even though well within the elastic limit of the wires, set up points of stress concentration.
The fatigue effect of bending appears in the form of small cracks in the wires at these over-stressed foci. These cracks propagate under repeated stress cycles, until the remaining sound metal is inadequate to withstand the bending load. This results in broken wires showing no apparent contraction of cross section.
Experience has established the fact that from the service view-point, a very definite relationship exists between the size of the individual outer wires of a wire rope and the size of the sheave or drum about which it operates. Sheaves and drums smaller than 200 times the diameter of the outer wires will cause permanent set in a heavily loaded rope. Good practice requires the use of sheaves and drums with diameters 800 times the diameter of the outer wires in the rope for heavily loaded fast-moving ropes.
It is impossible to give a definite minimum size of sheave or drum about which a wire rope will operate with satisfactory results, because of the other factors affecting the useful life of the rope. If the loads are light or the speed slow, smaller sheaves and drums can be used without causing early fatigue of the wires than if the loads are heavy or the speed is fast. Reverse bends, where a rope is bent in one direction and then in the opposite direction, cause excessive fatigue and should be avoided whenever possible. When a reverse bend is necessary larger sheaves are required than would be the case if the rope were bent in one direction only.
Stretch of Wire Rope The stretch of a wire rope under load is the result of two components: the structural stretch and the elastic stretch. Structural stretch of wire rope is caused by the lengthening of the rope lay, compression of the core and adjustment of the wires and strands to the load placed upon the wire rope. The elastic stretch is caused by elongation of the wires.
The structural stretch varies with the size of core, the lengths of lays and the construction of the rope. This stretch also varies with the loads imposed and the amount of bending to which the rope is subjected. For estimating this stretch the value of one-half percent, or .005 times the length of the rope under load, gives an approximate figure. If loads are light, one-quarter percent or .0025 times the rope length may be used. With heavy loads, this stretch may approach one percent, or .01 times the rope length.
The elastic stretch of a wire rope is directly proportional to the load and the length of rope under load, and inversely proportional to the metallic area and modulus of elasticity. This applies only to loads that do not exceed the elastic limit of a wire rope. The elastic limit of stainless steel wire rope is approximately 60% of its breaking strength and for galvanized ropes it is approximately 50%.
Preformed Wire Ropes Preformed ropes differ from the standard, or non-preformed ropes, in that the individual wires in the strands and the strands in the rope are preformed, or pre-shaped to their proper shape before they are assembled in the finished rope.
This, in turn, results in preformed wire ropes having the following characteristics:They can be cut without the seizings necessary to retain the rope structure of non-preformed ropes.
They are substantially free from liveliness and twisting tendencies. This makes installation and handling easier, and lessens the likelihood of damage to the rope from kinking or fouling. Preforming permits the more general use of Lang lay and wire core constructions.
Removal of internal stresses increase resistance to fatigue from bending. This results in increased service where ability to withstand bending is the important requirement. It also permits the use of ropes with larger outer wires, when increased wear resistance is desired.
Outer wires will wear thinner before breaking, and broken wire ends will not protrude from the rope to injure worker’s hands, to nick and distort adjacent wires, or to wear sheaves and drums. Because of the fact that broken wire ends do not porcupine, they are not as noticeable as they are in non-preformed ropes. This necessitates the use of greater care when inspecting worn preformed ropes, to determine their true condition.

Although Westech Rigging Supply strives to manufacture and sell the highest quality rigging and safety gear, use of the gear is dangerous if not used correctly by competent trained professionals. Westech Rigging Supply disclaims any liability resulting from the misuse of its rigging and safety gear. Please take a moment to more thoroughly review our disclaimer.
Westech Rigging Supply rigging and safety gear is only intended to be used by competent trained professionals. Misuse of the rigging and safety gear can result in serious injury up to and including loss of life. As such, Westech Rigging Supply disclaims liability for any misuse or incorrect product selection by our customers.
Rigging and safety gear purchased from Westech Rigging Supply should be used in strict accordance with all industry and OSHA standards. At no time should rigging or safety gear be used beyond its certified load ratings (aka Working Load Limits). Normal wear and tear should be expected with use of rigging and safety gear; therefore, all gear should be thoroughly inspected before each and every use. Worn or unsafe rigging and safety gear should never be used.

Since 2005 when we achieved access to the international market, we have been holding the win-win principle to continuously improve the production process and service level. It is our responsibility to create “security” for users. In every production process, we take “safety” as the concept, perfect production process and quality control system to ensure the strict monitoring of the whole production process. As our production scale grows, the production cost reduces relatively, enabling us to offer crane wire ropes and wire rope assemblies at the most competitive price while maintaining high quality. So far, we have established a strategic partnership with BRIDON and Arcelor Mittal. We created a proprietary brand – “Panda” and owned more than 700 excellent customers covering the US, the UK, French, the Netherlands, and more. Most of our customers are Fortune 500 companies.
A: Our main market is foreign customers. 80% of the products are for exporting. The main countries are Japan, South Korea, South Asian area, Europe, America, Canada, and South America etc.
A: For quality checking, we accept small quantity orders like one coil. For normal cooperation, we prefer FCL container quantity to ensure the safety of goods in transportation.

The 6 x 19 classification of wire ropes includes standard 6 strand, round strand ropes with 16 through 26 wires per strand. The 6 x 36 classification of wire ropes includes standard 6 strand, round strand ropes with 27 through 49 wires per strand. Although their operating characteristics vary, all have the same weight per foot and the same nominal strength, size for size.
While the 6 x 19 ropes give primary emphasis to abrasion resistance in varying degrees, the 6 x 36 ropes are important for their fatigue resistance. This fatigue resistance is made possible by the greater number of small wires per strand.
Although there are exceptions for special applications, the constructions in 6 x 36 classification are primarily designed to be the most efficient for each rope diameter. As the rope size increases, for instance, a large number of wires can be used to achieve required fatigue resistance, and still those wires will be large enough to offer adequate resistance to abrasion.
In this construction, each strand has nine outer wires over nine smaller inner wires over one large center wire. A comparison of cross-sections shows that these outside wires are larger than those of the 6 x 25FW or 6 x 26WS. Therefore, its resistance to abrasion is increased, but its fatigue resistance is decreased. This is a good rope to withstand abrasion or crushing on the drum.
To most wire rope users, 6 x 19 means 6 x 25 filler wire. It is the most common rope in the 6 x 19 classification. This rope has a good balance between both abrasion resistance and fatigue resistance in relation to other ropes.
This construction has better resistance to abrasion than a 6 x 25FW. It also features a compact construction with solid support for the wires; hence, it has a high resistance to crushing. Its number and relative size of the inner wires add to the stability of the strand and gives it a fatigue resistance comparable to a 6 x 25FW.
A standard 6 x 26WS construction provides the best rope for a wide range of applications. In general, we recommend the use of a 6 x 26WS in any application where a 6 x 25FW is used.
In most rope sizes, only one 6 x 36 classification rope is made. These constructions were selected to provide fatigue resistance without having wires that are too small.
The greater number of wires in the 6 x 36 classification makes these ropes more susceptible to crushing. This can be minimized, however, by specifying an Independent Wire Rope Core (IWRC) and by using well-designed sheaves, grooved drums and proper operating techniques.
Rotation-resistant ropes can frequently provide the best and most economical service in specific applications when you choose, handle and use them properly.
Contra-helically laid, rotation-resistant ropes are different from standard ropes because they"re designed to reduce rope torque. Modes of failure and wear for rotation-resistant ropes can differ from those for standard rope constructions. The very nature of these ropes requires special handling, selection and usage not encountered with standard constructions. They are susceptible to kinking, crushing and unbalancing in the form of "core pops" and "birdcages" Use extreme care to avoid operational practices that can possibly lead to these conditions.
Rotation-resistant ropes should not be used with swivels that allow rope rotation -- or in single part lifts where the load can rotate. Rotation will cause a reduction in strength, unequal loading in the rope and possible rope unbalance. If any significant change in diameter is found in a short length of a rotation-resistant rope, the rope needs to be replaced.
These ropes should be replaced when you see two randomly distributed crown wire breaks in six rope diameters -- or four randomly distributed crown wire breaks in 30 rope diameters.
Because rotation-resistant ropes are special, there are separate design, maintenance, inspection and removal criteria established for them by applicable industry regulations and standards.
In an application where a single-part hoist rope is used to lift a free load -- or where rotation-resistant properties are essential for rope performance -- the 19 x 7 can be used. Its rotation-resistant characteristic is achieved by laying six strands around a core strand in one direction, then laying 12 strands around the first operation in the opposite direction. Thus, when the rope is in tension, opposing rotational forces are created between the inner and outer layers.
In addition, frequent and regular inspection for broken wires is critical when using this rope. Due to its design, the 19 x 7 construction has a relatively low reserve strength. This can result in short service life between the point in time when the broken wire removal criteria are met and when actual rope failure occurs.
In a multi-part wire rope system where the blocks have a tendency to twist -- or for a single-part hoist line that doesn"t require the degree of rotation-resistant properties found in a 19 x 7 rope -- the 8 x 25 Resistwist rope has found successful application. The rotation-resistant characteristic is achieved by laying the eight outer strands around an independent wire rope core so these strands are in the opposite direction to the lay of the core. Thus, when the rope is in tension, opposing rotational forces are created between the core and the outer strands.
Though not as rotation-resistant, the 8 x 25 Rotation Resistant rope is more stable than a 19 x 7 rope. It also has increased resistance to bending fatigue and crushing. This is achieved through the use of eight-strand construction with an independent wire rope core.
Like any application where an installation"s rope type is changed, the 8 x 25 Rotation Resistant rope should be substituted only after carefully comparing specifications and strength requirements.
8613371530291