Parafil Rope Terminations.

Click here for a 3D pdf of the frictional stress along the inside of the barrel in a parafil termination.

Click here (pdf 2.4MB) for the whole PhD as a pdf.

Click here (pdf 0.3MB) for a pdf of The friction and wear of Kevlar 49 sliding against aluminium at low velocity under high contact pressures, Journal of Wear.

Background.

Ropes have been used for millennia but until a few decades ago very little quantitative study had been done on their properties. With the advent of new polymers much progress has been made in improving the performance of ropes, in particular the development of parallel-lay ropes which utilise the full stiffness and strength of the individual yarns. The best method found for terminating parallel-lay ropes is to clamp the rope between a spike and a barrel, this has the advantage that as the rope is loaded the spike is drawn into the barrel thereby clamping the rope even more tightly.

For static loading, spike-and-barrel terminations utilise 100% of the capacity of the individual yarns, however problems can arise if the termination is not assembled correctly. When the rope is subjected to a cyclic loading abrasion occurs between the rope and the termination which ultimately leads to failure, there is also a curiosity whereby the spike can suddenly jump when the load is increased after a period of cyclic loading. To date there has been no detailed analysis of spike-and barrel terminations, their design has evolved via a build-it-and-see approach, however this approach is not feasible for the design of large terminations - fabrication of ropes with a breaking load of 10,000 tonnes is possible.

Finite element analysis predictions for the frictional stress (N/mm²) on a spike in a 60 tonne Parafil G rope for pre-load, cyclic loading, and loading to failure (the base is at 50mm, the nose at 200mm). There is also a larger more detailed

As the rope is loaded, it draws the spike into the barrel, generating a radial stress which grips all fibres uniformly. This is different from systems with external wedges where only the outer fibres are gripped effectively. The slightly inclined radial stress, and the associated friction forces transmit the axial load from the rope to the barrel. The angles of the surfaces inside the termination are critical for the whole package to function correctly. Too great an angle, and the spike shoots out backwards!

Aim of this project.

The main aim of this study therefore is to provide an insight into the workings of a parallel-lay spike-and-barrel termination. To achieve this a finite element model was developed to follow the stresses, strains and displacements within a termination subjected to a cyclic loading history. This is compared with strains and displacements gathered from tests performed here on an actual 60 tonne Parafil G rope. (Parafil G is the trade name for a Kevlar 49 parallel-lay rope made by Linear Composites Ltd.) Comparisons are then drawn up between long term cyclic fatigue tests performed elsewhere and those extrapolated from yarn-on-capstan tests using the finite element model. To formulate the finite element model it was necessary to determine the bulk transverse properties of Kevlar 49, and the frictional properties of Kevlar 49 on aluminium.

Animated gif of a termination under cyclic loading

Outline of the dissertation.

Chapter 2 contains a review of parallel-lay ropes, indicating the advantages they have over the two other main types; twisted and braided. A brief synopsis of the current applications for parallel-lay ropes is included. A discussion of alternative termination methods is presented. The advantages and problems associated with a spike-and-barrel design are explained.

A brief literature review of friction in polymeric materials is presented in Chapter 3, this is then extended to propose a new formulation relating the frictional and contact stresses (). for inclusion in the finite element modelling. Experimental results of Kevlar 49-on-aluminium performed here are presented.

As the spike moves further into the termination on loading, the rope is progressively compressed. In Chapter 4, experimental stress-strain curves for the transverse compression of pads of 1,000 yarns of Kevlar 49 are presented. A literature review of the other properties of Kevlar 49 is discussed along with the rearrangement for inclusion in the finite element model.

Chapter 5 outlines the formulation of the finite model to use the Abaqus package and its friction and material subroutine options. The analytical results for a loading history including two load cycles is presented in detail.

The finite element results are compared with the tests performed here on actual 60 tonne Parafil G ropes in Chapter 6.

Chapter 7 outlines the factors affecting the abrasion of polymeric materials. Tests performed here of Kevlar 49 yarns on aluminium capstans are presented, these are used to predict an equation relating the lifetime to the contact stress and the amplitude of abrasion. This equation is then combined with the predicted contact stresses and amplitudes from the finite element model to predict the lifetimes of actual ropes under cyclic loading. These predictions are then compared with experimental data.

Finally, a summary of the factors involved in analysing spike-and-barrel terminations is presented in Chapter 8, along with suggested avenues that may optimise the performance and predictability of parallel-lay ropes.

Photograph of strain gauged 60 tonne spike and barrel

Cross section through a termination

Spike bed-down.

The bed-downs of the spikes for three tests carried out here and for the Abaqus predictions are remarkably similar. The most notable feature for the Abaqus results is how the stiffnesses match the test results. There are five stiffnesses present, which correlate with the stiffness changes for the transverse moduli of the Kevlar. The stiffness is increased at 180kN and at 360kN for the loading phase, and increased substantially for the unloading-reloading phase, and again for the final unloading.

The one feature that would be very hard to model by finite elements is the sudden jump of the spike backwards in test 3. This is a common feature of Parafil terminations that have been cycled then loaded further. It may be due to the limiting friction being lowered by the abrasion from cycling, this could allow the spike to slip backwards. Due to the non-linear nature of the friction, there comes a point at which the limiting friction holding the spike in can no longer match the contact pressure pushing it out, and it jumps.

Lifetime under cyclic loading.

The above figure shows the lifetimes of 6 and 60 tonne Parafil G ropes; both the experimental ones and those predicted independently in this thesis. There is a very good correlation between them, given the scatter normally associated with fatigue tests and the assumptions made here.