Copyright 2005, Offshore Technology Conference

This paper was prepared for presentation at the 2005 Offshore Technology Conference held in
Houston, TX, U.S.A., 25 May 2005.

This paper was selected for presentation by an OTC Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Offshore Technology Conference and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Offshore Technology Conference, its officers, or members. Papers presented at
OTC are subject to publication review by Sponsor Society Committees of the Offshore
Technology Conference. Electronic reproduction, distribution, or storage of any part of this
paper for commercial purposes without the written consent of the Offshore Technology
Conference is prohibited. Permission to reproduce in print is restricted to a proposal of not
more than 300 words; illustrations may not be copied. The proposal must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract

A high cycle fatigue test programme up to 40 million
cycles has been conducted for the first time on 10, 150 and
250 tonne polyester ropes and a new design TN curve is
proposed. The fatigue test duration achieved has surpassed any
other work to date and represents the most comprehensive data
set of its kind. Polyester rope shows a very significant increase
in fatigue perfomance over steel wire rope.

A new fatigue curve has been established for polyester
deepwater mooring rope. The establishment of the
fundamental fatigue mechanisms will enable proper planning
of through life integrity management systems.

In addition to generating fatigue data covering endurances
to around 20 million cycles, two ropes (10 tonne and 250
tonne) were each tested to 40 million cycles without failure.
Post test examination and textile yarn residual strength tests
revealed that no scale effect was apparent in terms of internal
wear (or any other fatigue mechanism).

The study has conclusively established the prime fatigue
mechanisms that may operate over platform lifetime. The
magnitude of any strength loss, if any, is expected to be very
small.

Strength loss due to internal abrasion, under low mean and
high load range, resulted from the slip between strands within
the subropes as the helix angle changes due to changing load.

Note: There was no wear between parallel laid subropes.

Even in the highest loaded fatigue tests, peak 65% of
actual break load, no strength loss occurred on the inner yarns,
which confirms no creep rupture strength loss had been
sustained. This study has confirmed that creep-rupture induced
strength loss is not a dominant mechanism in these high load
fatigue tests (short duration compared to platform lifetime)
and unlikely to be in mooring lines in service over long term
cycling.

Introduction

A laboratory test program investigated the durability of
polyester deepwater mooring ropes. Durability topics include;
high cycle fatigue, fatigue degradation mechanisms,
cumulative damage, strength loss and short term and long term
property behaviour. The work presented in this paper
concentrated on high cycle fatigue, degradation mechanisms
and strength loss (specific to low minimum load fatigue) on
10, 150 and 250 tonne break load rope samples. For the
degradation work, run out samples were carefully removed
from midspan, eye and splice and extensively examined to
establish residual strength and any fatigue mechanisms.

This paper provides the basis of a new fatigue curve for
polyester parallel strand ropes that has been proposed in the
draft ISO code
(1)
. Fatigue mechanisms have been qualitatively
and quantitatively documented which is a significant addition
to the industry technical knowledge.


Samples and Test Methods

Rope samples were provided by three rope makers, with
each supplier being left to offer its preferred construction,
provided they were of zero or low torque. One rope maker
supplied ropes of parallel strand construction (PSC), another
supplied ropes of parallel braid construction (PBC) and the
third rope maker supplied ropes in wire rope construction
(WRC). Note that the PBC and PSC rope samples were made
from several small subropes laid parallel and covered with a
jacket. In contrast, the WRC was a 9/3 conventional wire lay
construction and also covered with a jacket. Photographs of
these constructions are detailed in Appendix A.





OTC 17510
Durability of Polyester Deepwater Mooring Rope
S.J . Banfield, TTI; N.F. Casey, TUVNEL; and R. Nataraja, NDE
Figure 1 Spliced strop with bushes in each eye

The test samples (included bushes) were supplied by each
of the three rope makers. All of the samples were provided
2 OTC 17510
with spliced terminations and soft eyes as shown in Figure 1.
The eyes were protected with Dyneema cloth to essentially
eliminate wear at the back of the eye and at the tangent points
during fatigue testing.

All three manufacturers supplied nominal 10 and 150
tonne samples. One rope maker supplied a 250 tonne rope in
PSC for the high cycle fatigue test. These are are all nominal
strength, refer to the respective tables for actual measured
strength.

Note: The different sizes of ropes from the same
manufacturer were of the same nominal construction so the
effects of scale could be investigated.

Each rope sample was fitted into the appropriate test
machine and a static load equal to the mean fatigue load was
then applied. The fatigue loads were based on the appropriate
average actual breaking strength (ABS).

Note: The break tests were conducted in air but prior to
testing the ropes were immersed in tap water for
around 24 hours, so it can be assumed that the break
test results apply to the fully wetted condition. The
tests were carried out in general accordance with the
OCIMF procedure, that is 10 bedding-in cycles to
50% of the rated minimum breaking load (MBL) and
then break. For the 10 tonne ropes, six break tests
were conducted for each of the three constructions.
Whereas, for the mid strength and 250 tonne rope,
between three and five break strength tests were
conducted. The exception to this however was the
WRC mid strength rope where there was only
sufficient rope to produce one break test sample. The
results of the individual break tests has not been
included.

The ropes were fatigue tested fully immersed in tap water.
Fresh water was used at the start of each test and the water
was changed during testing on an as-required basis.

In order to complete the fatigue test programme in a
reasonable time scale the loading frequency was selected on a
per test basis. Frequencies in the range 0.3 to 4 Hz were used,
depending upon the rope size and load range. Experience of
the appropriate frequency for a given fatigue load, without
inducing heating, was gained during the NEL funded study
(2)

which preceded the Durability J IP
(3)
.

Testing was continued until the rope could no longer
support the fatigue test load. Upon completion of each test the
rope was removed from the test machine and the position of
failure confirmed.

10 Tonne Fatigue Test Results

The purpose of these tests was to generate high cycle T-N
plots for the 10 tonne ropes supplied by the three rope makers.
In order to achieve failures within a reasonable time scale and
still generate high cycle fatigue data, the fatigue tests were
centred around a mean load of 40% ABS. For each rope type
one test was also conducted at 60% mean to investigate the
effects of high mean with high load range. The fatigue test
programme was also designed to look at repeatability, albeit in
a fairly limited way.

Given in Tables 1 to 3 are the results that were generated
during the 10 tonne fatigue study. Included in the tables are
the test mark, mean load and load range (expressed as a
percentage of ABS) and cycles to failure.

The results in Tables 1 to 3 have been used to generate the
T-N plots in Figure 2 (one for each rope type). In the figure
three separate lines have been drawn through the data, one line
per rope type.

Tests conducted under the high mean load condition of
6017.5% ABS showed significant reductions in life
compared to the 4017.5% ABS tests (refer to Tables 1 to 3).
Increasing mean load from 40 to 60% ABS would appear to
have a detrimental effect upon endurance (perhaps by as much
as a factor of 3). If there is a mean load effect below 40% it is
likely to increase fatigue life rather than decrease it and it
might never be possible to quantify this because of the long
testing time and associated costs involved.

With regards to reproducibility, a limited number of repeat
tests were carried out. Three repeat tests were each conducted
on the WRC samples at 4025% ABS and 4020% ABS
(refer to Table 1). The average endurance of each set of three
tests was 626,330 cycles for the 4025% ABS tests and
2,267,977 cycles for the 4020% ABS tests. The resulting
coefficients of variation were 6.2% and 39.3% respectively.
Because the PSC ropes lasted much longer in fatigue only two
tests (Table 2) were conducted at a load range of 4025%
ABS with another two at 4022.5% ABS. The average
endurance of the two tests at 4025% ABS was 2,659,300
cycles with a coefficient of variation of 32.6%. At 4022.5%
ABS the average of the two endurances was 3,761,390 with a
coefficient of variation of only 5.1%. For the PBC samples
two tests were each conducted at 4025%, 4022.5% and
4020% ABS (refer to Table 3). At 4025% ABS the average
endurance was 1,883,230 cycles with a coefficient of variation
of 18.6%. At 4022.5% ABS the average endurance was
2,311,815 cycles with a coefficient of variation of 17.3%.
Finally, at 4020% ABS the average endurance was
4,530,395 cycles with a coefficient of variation of 11.1%. In
terms of fatigue, all of these coefficients of variation are very
low and reflects very well on both sample preparation and the
quality of the testing procedures and equipment.










OTC 17510 3

Table 3
Fatigue Test Results: 10 Tonne PBC Samples

Average Breaking Strength (ABS) =113.88 kN
2.1kN 1 SD (CoV 1.8%)


Test mark Mean load
(%ABS)
Load range
(%ABS)
Endurance

(Cycles)
MQAN 9 40 50 1,858,440
MQAN 11 40 40 4,887,260
MQAN 12 40 45 2,029,750
MQAN 13 40 50 1,908,020
MQAN 16 40 40 4,173,530
MQAN 17 40 45 2,593,880
MQAN 18 40 35 12,807,610
MQAN 21 60 35 3,834,130
MQAN 22 40 31 15,200,440





10
100
100000 1000000 10000000 100000000
Endurance (Cycles)
L
o
a
d

R
a
n
g
e

(
%
A
B
S
)
WRC 40%Mean
PSC 40%Mean
PBC 40%Mean
WRC 60%Mean
PBC 60%Mean
PSC 60%Mean
Power (WRC 40%Mean)
Power (PSC 40%Mean)
Power (PBC 40%Mean)
60% Mean Load Tests


Figure 2 10 Tonne T-N Plot

Very High Cycle Fatigue Tests

The purpose of these tests was to investigate the effect of
scale between the 10 and 250 tonne PSC ropes using yarn
testing and examination to investigate internal wear. The test
load of 303% ABS was chosen because it represents a harsh
test in that 6% load range is equivalent to severe weather
conditions. The majority of cycles experienced on a platform
mooring would normally be far less than the 6% load range
selected. Testing was stopped after each rope had reached just
over 40 million cycles, as indicated in Tables 4 and 5.



Table 4
Fatigue Test Result: 250 Tonne PSC Sample

Average Breaking Strength (ABS) =2,546 kN
51kN 1 SD (CoV 2%)


Test mark Mean load
(%ABS)
Load range
(%ABS)
Cycles
MPWM35* 30 6 40,000,660

Table 5
Fatigue Test Result: 10 Tonne PSC Sample

Average Breaking Strength (ABS) =101.67 kN
2.51kN 1 SD (CoV 2.1%)

Test mark Mean load
(%ABS)
Load range
(%ABS)
Cycles
MPWM 19 30 6 40,000,660

Low-Tension Fatigue Test Results

The investigation into the effect of low-tension fatigue
(low minimum load with low mean) was carried out using
both the 10 tonne and mid strength samples. This allowed the
effect of scale between the 10 tonne and mid strength samples
under low-tension fatigue conditions to be investigated. The
tests were conducted at 109% ABS for one and five million
cycles in the case of the 10 tonne ropes and one million cycles
for the mid strength ropes. The ropes were then either
subjected to a retained break strength test or to examination of
internal wear and retained break strength estimation through
textile yarn break testing.

It should be noted that the load values used (low mean
with relatively high load range) for the low-tension fatigue
tests are believed to represent a harsh load regime. The results
of the retained break strengths provided in this section,
therefore, can be considered as very severe case condition.

10 Tonne Test Results

Six 10 tonne tests were conducted, two per rope maker. In
each case one sample was tested to one million cycles at
109% ABS with the other tested at the same load range but to
five million cycles. Both the PSC samples together with the
WRC sample that was tested to one million cycles was
subjected to internal examination and yarn testing (Table 6).
The rest of the samples were subjected to retained break
strength tests.

The results of these tests are summarised in Table 6.
Using the retained break strength values and comparing them
with the average breaking strength in the as-new condition, it
can be seen that the WRC sample (MPVQ 18) lost 8.1% break
strength after five million cycles at 109% ABS. The two
PBC samples MQAN 20 and MQAN 19 lost 4.9% and 7.2%
break strength after one million and five million cycles
4 OTC 17510
respectively The remainder of the 10 tonne rope samples
were subject to internal examination and retained break
strength estimation.

Table 6
Low Tension Fatigue Test Results: 10 Tonne Ropes

WRC As-new ABS =116.682.51kN
PSC As-new ABS =101.671.11 kN
PBC As-newABS =113.882.10 kN

Test mark Rope type Cycles Retained strength
(kN) (%)
MPVQ 18 WRC 5,000,000 107.2 kN (92)
MPVQ 19 WRC 1,000,000 Examined
MPWM 10 PSC 5,000,000 Examined
MPWM 16 PSC 1,000,000 Examined
MQAN 19 PBC 5,000,000 105.7 kN (95)
MQAN 20 PBC 1,000,000 108.3 kN (93)

Mid Strength Test Results

In total five tests were conducted (two WRC samples, two
PSC samples and one PBC sample) to investigate the effects
of low-tension fatigue upon rope strength. The samples were
tested at 109% ABS to one million cycles. Three samples
(one WRC, one PSC and one PBC) were each subjected to
retained break strength tests. The remaining WRC and PSC
samples were subjected to internal examination and retained
break strength estimation. A summary of the results is
provided in Table 7.

As with the 10 tonne ropes the percentage strength losses
given are based on the average of the as-new break tests. The
exception being the WRC rope, where only one break test was
conducted due to insufficient rope. Using the retained break
strength values in Table 7 and comparing them with the as-
new values it can be seen that the WRC sample MQMK 2 lost
22.1% break strength after one million cycles at 109% ABS.
This is significantly more than the 10 tonne sample that lost
only 7.6% strength after five million cycles. The PSC sample
MPWM 39 lost 14.8% strength. The PBC sample MQNL 4
lost 1.2% break strength but the result fits in well with the
three break tests conducted in the as-new condition.

Table 7
Low Tension Fatigue Test Results: Mid Strength Ropes

WRC ABS =1,314 kN
PSC ABS =1,85719 kN 1 SD (CoV 1%)
PBC ABS =1,86542 kN 1 SD (CoV 2.3%)

Test mark Rope type Cycles Retained strength
(kN) (%)
MQMK 1 WRC 1,000,000 Examined
MQMK 2 WRC 1,000,000 1,024 (78)
MPWM 38 PSC 1,000,000 Examined
MPWM 39 PSC 1,000,000 1,583 (85)
MQNL 4 PBC 1,000,000 1,843 (99)
The results presented in this paper do seem to indicate the
presence of a scale effect on retained strength, for both the
WRC and PSC samples between 10 tonne and mid strength.
However, this does not appear to be the case with the PBC
rope. Clearly, these findings result from a limited programme
of work and further work with regard to scale effect upon both
fatigue life and retained strength is currently being further
studied
(5)
.


Yarn Tests and Examination of 10, 150 and 250 tonne
Ropes

The 10 tonne ropes were separated into inner and outer
layers of subropes. One inner and one outer subrope were
selected for testing, from which 20 textile yarns were removed
from each for residual strength break tests. There was one
ropeyarn in each strand ie only 1 layer. In the 150 tonne PSC
there were 2 layers of ropeyarns in the strand, from which
ropeyarns were removed from each layer.


Low Tension Fatigue

10 tonne PSC 1 and 5 Million Cycles Unfailed 109% ABS

The SEM examination of the inter-strand contact region is
shown in Figure 3 after 1 million cycles where a low level of
filament wear damage was found affecting the surface layer of
filaments only. After 5 million cycles a low level of filament
damage was again found affecting the surface layer of
filaments only as shown in Figure 4, but more deeper than
after 1 million cycles. The spacing of damage along the length
of one filament was consistent with a filament on filament
contact. Note that the under layers of filaments were not
affected.



Figure 4 10 tonne PSC, Inter-strand contact region
with a low level of filament damage affecting the
surface layer of filaments only, after 5 million cycles
10% mean 18% range






















OTC 17510 5
Figure 5 shows a close up of one filament (20 micron
diameter) where it can be clearly seen that a filament has worn
a smooth groove, removing some of the cross section.
A summary of the average strengths of twenty textile yarns
is shown in Table 8. Residual textile yarn strength in the outer
subropes was 93% compared to the inner subropes at 96%
after 1 million cycles. The rope residual strength calculated
from textile yarn strength from these two layers of subropes
was 94%.
Table 8
10 tonne PSC Yarn Residual Strength Test Results and
Calculated Rope Residual Strength


% residual strength
(average 20)
CYCLES (millions) 1 5
OUTSIDE SUBROPE
TY
93 75
CORE SUBROPE
TY
96 69
% RESIDUAL ROPE
STRENGTH
(calculated from textile
yarn strength)
94 73

This rope lost around 29% strength for 5 million cycles
compared with 6% for 1 million cycles, which shows a trend
of increasing strength loss with cycles. This indicates that
abrasive wear is reducing material cross section, confirmed by
the SEM examination. There is no significant difference
between inner and outer subropes, which would be expected
since they are all laid parallel (thus, no radial forces) and also
shows that the jacket has had not influenced the strength loss
(can provide some radial force).

150 tonne PSC 1 Million Cycles Unfailed 109% ABS

The effect of scale can be seen where the abrasion groove
is much longer in Figure 6, due to the higher slip between
strands. This is confirmed by the higher strength loss in the
textile yarns and the rope shown in Table 9, compared to
Table 8 for the 10 tonne rope at 1 million cycles.

Again, there is no significant difference between inner and
outer subropes, showing that there is no scale effect from the
jacket (jacket radial pressure in some designs may influence
wear rates).


Table 9
150 tonne PSC Yarn Residual Strength Test Results
and Calculated Rope Residual Strength for 1 Million
Cycles


% residual strength
(average 20)
OUTER SUB-ROPE
OUTSIDE TY 88
INNER TY 87
CORE SUB-ROPE
OUTSIDE TY 77
INNER TY 79
% ROPE RESIDUAL
STRENGTH (calculated from
textile yarn strength)
84


10 and 150 tonne WRC 1 Million Cycles Unfailed
10%9% ABS

SEM examination of the inter-strand contact region of the 10
tonne rope is shown in Figure 7 which is similar to that found
in the 10 tonne PSC rope.

In Figures 8 and 9 the wear damage is much higher in the
150 tonne rope, due to the scale effect of increased slip
between strands.

















Figure 7 10t WRC Inter-strand contact region
where a low level of filament damage was found
affecting the surface layer of filaments only after 1
million cycles 10% mean 18% range














Figure 8 150t WRC The amount of abrasion
damage had reduced the filament diameter by more
than half

Figure 6 150t PSC, Inter-strand contact region
with a low level of filament damage affecting the
surface layer of filaments only, after 1 million
cycles 10% mean 18% range

6 OTC 17510
A summary of the average strengths are shown in Table 10 for
the 10 tonne and 150 tonne ropes. Again the scale effect is
significant with much lower residual strengths for the
150 tonne rope compared to the 10 tonne rope.

Table 10
10 tonne and 150 tonne WRC - Yarn Residual Strength
Test Results and Calculated Rope Residual Strength


% residual strength
(average 20)
Rope nominal break load 10t 150t
OUTER STRAND
OUTSIDE TY 90 56
INNER TY 109 81
CORE STRAND
OUTSIDE TY 81 57
INNER TY 110 78
% RESIDUAL STRENGTH
(calculated from textile yarn
strength)
91 67
















The mechanism of the abrasion damage is different in the
WRC rope compared to the PSC ropes. The stresses induced
by the slip between the strands in the subrope are cyclic and
reverse in direction. These stresses cause a small crack to open
and then a fibril will peel off from the surface. This
mechanism continues in successive layers as reported by
Hearle 1998
(4)
for polyester (and other materials) fibres. The
sequence of the peeling mechanism is shown in Figure 10.














Although three 10 tonne samples were cycled to 1% minimum
load for 1 million cycles, three 10 tonne to 5 million cycles
and two 150 tonne samples for 1 million cycles, not a single
filament from all the extensive optical and SEM examinations
showed any axial compression. Likewise, there was no
additional strength loss in regions where some minor wavy
buckling occurred. This means that the API axial compression
limits can be considered for lifting or removing entirely for
polyester. This was subject to further study in the 3
rd
year
extension to the Durability J IP
(5)
.

Very High Cycle Fatigue Tests (303% ABS)

10 tonne and 250 tonne PSC 40 Million Cycles Unfailed
303% ABS

SEM examination of the inter-strand contact region for the 10
tonne rope is shown in Figure 11 where no wear was found at
all, as well as the 250 tonne rope.


















Figure 11 SEM examination Mid-span between splices,
inter-strand contact region showing no damage on 10
tonne rope (250 tonne rope identical appearance)
Figure 9 150t WRC There were many examples of
fibrils peeling away from the surface of the filament
Residual strength of the textile yarns was around 100% for all
the tests from both the 10 tonne and 250 tonne ropes after 40
million cycles as shown in Tables 11 and 12. Thus, no scale
effect was found at the 303% ABS.

Table 11
10 tonne PSC rope, Yarn Residual Strength Test Results
and Calculated Rope Break Load After 40 Million Cycles
303% ABS


% residual strength
(average 20)
1
st
test Repeat repeat
OUTSIDE SR
TY

101 101 102
CORE SR
TY
99 98 102
% RESIDUAL ROPE
STRENGTH
100 100 102

Figure 10 Mechanism for surface peeling abrasion
damage at high slip



OTC 17510 7

Table 12
250 tonne PSC Rope Yarn Residual Strength Test Results
And Calculated Rope Break Load After, 40 Million Cycles
303%ABS


% residual strength
(average 20)
Outer sub-rope
OUTSIDE TY 101
INNER TY 101
Inner sub rope
OUTSIDE TY 101
INNER TY 104
% Residual rope strength 100


High Mean and Very High Load Range Fatigue Test

10 tonne PSC Failed at 3,272,080 Cycles 4025% ABS

Unlike all the other samples which were run-outs, this
sample failed in fatigue. The rope completely broke 50mm
from the end of splice. At the other end of the test strop the
sample was well preserved with no apparent visual damage
from the failure, so it was deemed that something may be
learned from this sample.

The rope residual strength calculated from textile yarn
strength for the end of splice buckled region was 64% as
shown in Table 13. The peak cyclic load was 65%, which is
the same as the rope residual strength and explains why the
rope failed. Clearly, the strength loss was very localised and
correlates with the observed wear damage in that region. This
damage was localised around end of splice over 50100 mm
length, but immediately adjacent in the splice and along the
rest of the mid-span between splices, less than a few hundred
mm away, was in relatively good condition with as high as
89% residual textile yarn strength and 84% calculated rope
strength. The reason for the lower strength at the end of
splice is not fully understood and is believed to be associated
with a termination effect. This is not a negative finding since
the fatigue life was so high, but highlights the fact that there
may be room for further improvement in fatigue life.















Table 13
Yarn Residual Strength Test Results and Calculated Rope
Residuals Strength


% residual strength
(average 20)
Position End of splice mid-span
OUTSIDE SUB-ROPE
TY
66 82
CORE SUB-ROPE
TY
59 89
% RESIDUAL ROPE
STRENGTH
(calculated from textile yarn
strength)
64 84


Textile yarn break elongation curves are shown for the end of
splice region in Figure 12, there are two distinct groups of
behaviour. Around 4 textile yarns gave as new properties,
around 100% residual strength and break elongation. The
others had widely varying residual strength and break
elongation. This was due to abrasion damage affecting only
the surface of the ropeyarn, whereas the inner textile yarns
were protected and did not suffer abrasion. Creep rupture at
very high loads will bring down the strength of the textile
yarn, in a similar manner to the abrasion. However, on those
inner yarns, there is no evidence at all of creep rupture.





The identical trend is seen in the mid-span textile yarns shown
in Figure 13, except that the abrasion damaged textile yarns
are in a much tighter cluster, since they have not been further
damaged by termination effects.

As new
Varying
degrees
abrasion

Figure 12 Textile yarn load elongation curves from end
of splice region


8 OTC 17510
Proposed T-N Design Curve

A design curve has been developed from a conventional
statistical analysis of the 10 tonne rope test results given in
Tables 1-3. It can been seen from the tables that the mean load
was 40% ABS in a total of twenty five tests and 60% in the
three tests MPVQ 20, MPWM 17 and MQAN 21. The
regression analyses included each type WRC, PSC and PBC
considered separately and all the three types with 40% ABS
mean load data lumped together. The coefficients of
determination for each type was in excess of 0.95, while for
the combined data this reduces to 0.74. This may be attributed
to the influence of the rope type on fatigue performance. This
possibility was investigated through confidence limits. Ninety-
five percent confidence limits were plotted for rope types in
pairs and all together. It was concluded that there is sufficient
evidence to combine three sets of test data.

The proposed "mean minus two standard deviation" T-N
design curve is based on the all three types lumped together
and with a standard deviation of 0.234, the curve is given by

Log (N) =4.13 5.09 Log (T)

This curve is illustrated in Figure 14 (for clarity, a larger
version is shown in Appendix B Figure B-1) and compared
with the 10 tonne test data, the mean curve, the present
RP2SM fibre rope curve
(8)
and the RP2SK steel rope curve. It
is clear from the figure that the slope -5.09 of the proposed
polyester rope curve is comparable to -5.05 of the steel rope
curve, but is significantly different from 9.0 of the RP2SM
polyester rope curve. It is also clear from the figure that the
polyester rope exhibits a factor of about fifty superior fatigue
performance compared with the steel rope.

The three test results with 60% ABS assist in modelling
the effect of mean load T
m
(based only on 40 and 60% mean)
.

The proposed T-N curve including the effect of mean load is
given by

Log (N) =4.71 1.44 T
m
5.09 Log (T)

As new
Abrasion
damaged

Figure 13 Textile yarn load elongation from midpsan
spliced strop
API RP2SM
(8)
recommends tension-tension fatigue
qualification tests to be carried out for a minimum of 380,000
cycles at 30%15% ABS tension. It can be seen from
Figure 14 that this point is well below the latest mean line and
outside the 95% confidence limit line. It is recommended that
API consider the implications of this important finding.
0.1
1.0
1.E+05 1.E+06 1.E+07 1.E+08
Endurance Cycles (N)
T
e
n
s
io
n

R
a
n
g
e
/
A
c
t
u
a
l
B
r
e
a
k
in
g

S
t
r
e
n
g
t
h

Mean Line
Design Line
SM Fibre Rope
SK Steel Rope
API Test Qual.
10 Ton Data
API Test
Qualification
380,000 Cycles,
30% mean
30% range
Figure 14 Comparison of proposed design T-N curve for
fibre ropes and RP2SM fibre rope and RP2SK steel curves
The T-N curve developed here forms the basis of a new
fatigue curve for polyester parallel strand ropes that has been
proposed in the draft ISO code
(1)
.

Conclusions

Polyester rope exhibits a factor of about fifty superior
fatigue performance compared with steel rope.

With good manufacturing and splicing procedures it is
possible to obtain coefficient of variation values of less than
3% on break strength for up to 250 tonne ropes.

Samples (including spools) that have been properly
prepared for fatigue testing and in conjunction with well
defined and appropriate test procedures, achieve extremely
high fatigue lives (well beyond that previously achieved)
under constant amplitude sinusoidal loading.

The results of the 10 tonne T-N fatigue programme
demonstrate that with good sample preparation and test
procedures, low values of coefficient of variation can be
obtained from reproducible fatigue tests using samples from
the same production run.

40 million cycle run-out data points have been achieved
for 10 and 250 tonne Marlow PSC samples at 303% ABS.
Based on a 10 second period this represents about 13 years
cyclic loading.

Under low tension fatigue loading, reductions in retained
rope break strength did occur with the primary form of
degradation being internal wear.

OTC 17510 9
Under low tension fatigue loading, there are indications of
a scale effect on retained break strength between 10 and ~150
tonne rope, in both WRC and PSC.

Wavy buckling was not damaging, even after a severe
fatigue test. The filaments were buckling in an elastic mode
which was not damaging to polyester. No axial compression
was found in the low minimum load (or other higher load
fatigue) tests.

Based on textile yarn examination and tests, at low
minimum load and high range, internal abrasion was found at
the interface between strands in the subrope. The strength loss
increases with size for the PSC and WRC. Based on rope
break strength tests, the PBC does not appear to have any
effect on scale.

Rope residual strength calculated from yarn strength tests
was 100% for the 10 tonne and 250 tonne 40 million cycle
tests for 30% mean and 6% range. There was no evidence of
any scaling in textile yarn or rope residual strength properties
between 10 tonne and 250 tonne ropes at this test condition.

Localised damage at the end of splices has highlighted the
possibility of a further improvement in fatigue life. Splice
design should pay attention to torque/twist response to reduce
effects of localised damage near the termination. PVC tape
wrappings localise damage at the end of splice and should not
be used.

Of the six fatigue modes identified in polyester ropes
(6,7)

five can be ruled out as follows. No axial compression, creep,
hysteresis, structural or external wear has been found in any of
the ropes. The remaining only mode found was internal
abrasion. This includes a reasonable span of mean loads and
load ranges that would be expected in service.

The latest test data suggests that the design T-N curve for
polyester ropes has slope similar to the steel rope curve, but
polyester rope has a fifty-fold superior fatigue perfomance.


Acknowledgements

This J oint Industry Project into the 'The Durability of
Polyester Ropes' was co-promoted by NEL and TTI Ltd and
supported by the following :-

ScanRope
Marlow Ropes
Quintas and Quintas
Chevron
BP-Amoco
Norsk Hydro
Statoil
Shell
Texaco
Total Fina Elf
Health and Safety Executive (Offshore Safety
Division)
US Minerals and Management Service
Lloyds Register
American Bureau of Shipping
Bureau Veritas

Noble Denton in return for participation in the study has
provided free issue work and use of their premises in London
for meetings.

The technical input and support provided by the company
representatives during the five steering committee meetings
has proved invaluable and contributed in no small way to the
successful outcome of the study. Their input is greatly
appreciated by the project promoters. The project promoters
also acknowledge the considerable effort from the ropemakers
in providing well designed and well made samples that has
lead to an unsurpassed set of fatigue data.


Nomenclature

ABBREVIATIONS

SEM Scanning Electron Microscopy
WRC Wire rope construction
PSC Parallel strand construction, made from 3 strand sub-
ropes
PBC Parallel strand construction, made from braided sub-
ropes
SR Subrope (could be WRC, 3 strand or braided
construction)
TY Textile yarn

DEFINITIONS

Filament Smallest fibre component around 20 m
diameter
Textile yarn Parallel assembly of filaments, typically
100-200
Rope yarn Twisted assembly of textile yarns
Strand Twisted assembly of rope yarns
T-N Tension-tension fatigue line

References


1. Fibre Rope for Offshore station Keeping Polyester,
Draft ISO TC 38/SC N, 13/10/2004
2. Fatigue Evaluation of 6 Tonne Nominal Strength
Polyester Sub-rope. NEL Report NEL/PS1-1 Issue 3.
March 2000
3. The Durability of Polyester Ropes, J IP co-promoted
and managed by NEL and TTI, 1999-2002 ref
MOG020, October 1988
4. Atlas of fibre fracture and damage to textiles, second
Edition 1998, J WS Hearle, B. Lomas, W.D. Cooke,
The Textile Institute, ISBN 1 855733196
5. Durability Polyester Ropes, J IP outline proposal to
extend the study Investigating minimum load effects
and the slip induced internal wear fatigue
10 OTC 17510
mechanism J IP Reference No: DPR001, NEL
Reference No: FCO/1309 Issue No: 3, Date: 23 May
2003
Figure A-4 150t PSC with jacket removed
showing parallel laid 3 strand sub-ropes.
The end of the PVC tape can be seen, where
wavy buckling is concentrated.
6. Hearle, J.W.S; Parsey, M.R; Overington, M.S; and
Banfield, S.J ; 'Modelling the long term fatigue
performance of fibre ropes' ISOPE, Singapore. July
1993.
7. Parsey, M.R, Hearle, J.W.S, and Banfield, S.J ., 'Life of
Polymeric Ropes in the Marine Environment',
Polymers in a Marine Environment Conference,
London Sept 1988.
8. Recommended Practice for Design, Manufacture,
Installation, and Maintenance of Synthetic Fibre
Ropes for Offshore Mooring, API RP 2SM, Edition,
API, March 2001



Appendix A. Rope Constructions




























































Figure A-1 150t WRC with jacket removed
showing 9 outer strands

Figure A-2 150 tonne WRC Outer 9 strands
unlayed showing 3 strand core

Figure A-3 10t PSC with jacket removed
showing parallel laid 3 strand sub-ropes
OTC 17510 11



Appendix B. Proposed Polyester Design T-N Curve

































































0.1
1.0
1.E+05 1.E+06 1.E+07 1.E+08
Endurance Cycles (N)
T
e
n
s
i
o
n

R
a
n
g
e
/
A
c
t
u
a
l

B
r
e
a
k
i
n
g

S
t
r
e
n
g
t
h

Mean Line
Design Line
SM Fibre Rope
SK Steel Rope
API Test Qual.
10 Ton Data
API Test
Qualification
380,000 Cycles,
30% mean
30% range

Figure B-1 Comparison of proposed design T-N curve for fibre ropes and RP2SM fibre rope and RP2SK steel
curves