Proceedings of the ASME 2011 30th International Conference on Ocean, Offshore and Arctic Engineering

OMAE2011
June 19-24, 2011, Rotterdam, The Netherlands

OMAE2011-49

QUALIFICATION OF MULTI-COMPOSITE HOSES FOR STS LNG TRANSFER

Gerard van der Weijde, Niels Mallon

TNO - Centre of Mechanical and Maritime Structures;
Van Mourik Broekmanweg 6, Delft, The Netherlands.
Gerard.vanderWeijde@tno.nl

ABSTRACT
Reliable transfer systems are a key element in
developing floating LNG technology. Multi-composite hoses
may prove to be a reliable and cost effective solution. TNO, the
Dutch contract research organisation, has executed an extensive
test program on the multi-composite hose of Gutteling B.V. for
Ship-to-Ship (STS) LNG transfers. It has resulted in
qualification of the hose in accordance to EN1474-part 2 [2].
This hose is the first product that has been qualified in
accordance with this new internationally accepted standard. The
test program was performed in close cooperation with Gutteling
B.V., EXMAR and DNV. It focused on the mechanical and flow
behaviour at ambient and cryogenic operating conditions. In
excess of the EN1474-requirements, a multi-level test
programme is performed, more samples are tested and more
extreme load combinations are applied. The purpose was to
provide a data set that enables transfer system qualification in Figure 1, Ship to Ship LNG transfer © Exmar – Gutteling B.V.
accordance with EN1474-part 3.
INTRODUCTION
The hose appears to have a complex mechanical behaviour: Transfer systems are a crucial enabler for floating
elastic non-linear, coupling of deformation modes, hysteric LNG [4], [5]. To offload LNG offshore, various aerial and
behaviour and large damping. Damage tolerance tests, such as floating transfer systems are currently under development [6],
impact and crushing, show that the composite hose performs [7]. Some are based on loading arm technology, others on metal
exceptionally. This paper summarises the test program and flexible pipe or multi-composite hoses. The multi-composite
describes some of the tests performed in excess of the hose consists of helical wires, thin liquid tight films of
requirements. In particular, testing for off-spec operation polymeric materials and layers of fabric like polymeric
conditions, fatigue and creep is addressed. Absences of materials. For floating hoses the composite hose is covered with
engineering models to predict fatigue and creep performance insulation and a polymeric outer hose. Similar composite hoses
are future obstacles for realising the interesting proposition of are frequently used in the oil and chemical industry for transfer.
multi-composite hoses. As such, the composite hose technology is not new. However,
reliability and safety requirements and the cryogenic

1 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 temperature during operation makes the LNG hose a class on its A hose may be applied in various system configurations. In
own. qualifying a hose it is worthwhile to consider those various
configurations as it provides input to the qualification
In 2007 the Belgian based shipping company Exmar, and US requirements.
based gas trader Excelerate, developed a near-shore transfer
system utilising 8” multi-composite hose of Gutteling, see Part 2 provides guidance. It specifies various tests to be
Figure 1. Through trials the performance of the system was performed. The test methods themselves and the requirements
proved and subsequently the system was used for commercial to be fulfilled are open for interpretation. The standard does
transfers. In parallel, the hose was extensively tested by the specify the maximum load requirements, such as proof pressure,
Dutch contract research institute TNO. For the qualification a burst pressure, maximum working loads. On LNG vessels the
newly developed international standard EN1474-part 2 was typical operating pressure is 3 [barg] and the systems are
used as reference. The Gutteling hose was successfully qualified designed for a maximum working pressure of 10.5 [barg]. This
by DNV in 2010 as the first qualified product under this results in proof pressure requirements of 15.8 [barg] and a
standard. Qualification of the complete transfer system requires minimum burst pressure of 53 [barg]. Loads due weight and
a review of all system components and operating procedures. system constraints need to be multiplied by a factor to derive
Ideally the hose qualification program provides sufficient maximum working loads. For each test required, it is sufficient
information to feed into the system qualification. In this respect to test one sample. That is, no multiple samples are required and
EN1474-2 seems to contain some flaws. This paper briefly scatter is not considered. Further, the standard allows using a
reviews the qualification program and addresses some new sample for each test. There is no requirement to execute
peculiarities of the composite hose performance. In excess of various tests subsequently on the same sample. Both testing at
the qualification program, other testing was performed to ambient temperature and cryogenic temperature is required.
provide more input to the qualification of transfer systems. This Liquid nitrogen can be used for tests at cryogenic temperature.
paper highlights some of these tests. Liquid nitrogen is at ambient pressure colder than LNG (-196
[°C] versus -168 [°C]). The programme heavily concentrates on
One of the main areas lacking is the substantiation of the fatigue testing of the prototype hose. Very limited testing on the
and creep performance of multi-composite hoses. A further material level is required. Rather than having very specific
analysis of this issue identifies that commonly accepted requirements, it should be demonstrated that the materials are
engineering tools to predict hose performance do not exist. This suitable for their intended purpose.
paper highlights an approach to fill this gap. It is recognized by the standardization committee that this new
standard needs some time to become mature.

EN1474 OVERVIEW HOSE QUALIFICATION PROGRAM

At the end of 2009, the European standard EN1474 This section provides a brief overview of the qualification
entitled “Installation and equipment for liquefied natural gas - program for hoses and addresses some of the peculiarities of
Design and testing of marine transfer systems” [2] was these hoses.
approved. This standard gives general guidelines for the design, The stiffness of the hose has to be characterised under expected
material selection, qualification, certification, and testing details and maximum operation conditions. For the maximum
for Liquefied Natural Gas (LNG) transfer hoses for offshore operating conditions the strength also needs to be proven. The
transfer or on coastal weather-exposed facilities for aerial, sample does not have to be tested until failure. Figure 2 shows
floating and submerged configurations or a combination of the test set-up used for tension, twist and bending performance
these. The standard, recognized by the offshore LNG of the hose.
community, consists of 3 parts [1]- [3]. Part 3 covers the safety
of the complete system, which includes for example ERS,
QCDC and mooring systems, and handling and operation
procedures. This part employs an overall safety philosophy
without specifying particular acceptance criteria. Part 3
references parts 1 and 2 that cover the design, qualification and
certification of loading arms and transfer hoses, respectively. As
such the qualification program of part 2 provides input to the
qualification of the system according to part 3. Qualification of
the transfer system is to be substantiated by the system user. Figure 2, test setup for tensile, bending, twist and bend radius performance ©
Qualification of the hose can be substantiated by the hose Gutteling B.V..
manufacturer. Note that qualification of the hose includes its
end fittings.

2 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 In tension the hose appears to have a highly non-linear and Hoses returned from service are more flexible than new hoses.
hysteric behaviour. Due to the hysteric behaviour, the hose This effect was also observed in testing. The stiffness changes
stiffness depends on its load history. In addition, the hose length after the sample has experienced a certain load history. This is
at zero pressure and external force shows a moderate variation. attributed to some forms of (temporary and/or permanent) creep
EN1474-part 2 does not provide guidelines on how to deal with of the hose ply construction. In several tests also the dynamic
the history dependency. For example, for fibre ropes and behaviour of the hose is characterised in order to be able to
mooring lines, which exhibit a similar hysteric and nonlinear assess the dynamic system behaviour. The hose exhibits a
behaviour under tension, much more details are provided in the damping ratio that is an order of magnitude higher than metal
guidelines about how to deal with length and stiffness structure [10]. It is concluded that determining the stiffness of
determination [8], [9]. Possibly, these guidelines may be the hose is much more complex than EN1474-part 2
adopted to obtain a better definition of composite hose stiffness anticipated. Considering the system configurations in which the
and length. composite hose are used, knowing the tensile, torsion and bend
stiffness at minimum bend radius has limited value.
Figure 3 shows a typical example of a tensile force-deflection
curve at ambient and cryogenic condition. Note that
temperature has a considerable effect on the stiffness. When the straight bending stiffness ambient hose
14000
pressure is increased, or a tensile load is applied, the hose not
measured
only elongates, but also twists. When twist is applied, the hose 12000 linear
also elongates or becomes shorter. Obviously this behaviour
should be taken into account when designing tests. As one may 10000
expect, the hose stiffness is very much influenced by the y = 9.6e+002*x + 7.8e+002
internal pressure. EN1474-part 2 requires quantifying the hose EI [Nm2] 8000

stiffness at its extreme temperatures. Effects of pressure and the

m

6000
coupling of elongation and twist are not addressed by the
standard.
4000

2000

0
0 2 4 6 8 10 12
P [barg]

Figure 4, bending stiffness as a function of pressure © Gutteling B.V..

A series of pressure testing is to be performed. It shall be

proven that the hose is leak-tight at proof pressure. Also hoses
with some light forms of impact and crush damage need to be
leak-tight. Unlike for loading arms in part 1, EN1474-part 2
does not specify the acceptable leak rate and the amount of
damage to be applied is subject for debate. Figure 5 shows
examples of the light impact and crush damage applied to the
hose. To further substantiate the hose strength, pressure testing
up until burst has been performed at ambient and cryogenic
temperatures.

Figure 3, Typical force-deflection curve in an ambient and cryogenic tensile Flow testing up to the maximum flow rate is required to
stiffness test © Gutteling B.V.
substantiate the hose integrity under this condition. Though only
In bending and torsion the hose exhibits a non-linear and ambient flow test is required, also a full scale LNG flow test is
hysteric behaviour as well. The hysteric behaviour should be performed for providing additional justification. This LNG flow
considered in designing test setups and procedures for test provided valuable input with respect to flow resistance,
executing the test [10]. The hose bending stiffness is also hose dynamics and leak tightness. Figure 6 shows the ambient
dependent on the internal pressure. Figure 4 shows the bending and cryogenic flow test set-up.
stiffness of the hose for large bend radii as a function of the
internal pressure (varying between 0 and MAWP).

3 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 Figure 7, cryogenic bending fatigue test setup © Gutteling B.V..

The execution of such a test requires 6 weeks continuous

cryogenic operation.
The 8” Gutteling hose successfully passed all EN 1474-part 2
requirements. As explained above, hose qualification data is to
Figure 5, light impact and crush damage applied to the hose © Gutteling B.V.. be used to substantiate the transfer systems. In Gutteling’s
program higher maximum loads were applied to cover potential
future systems. More load conditions were tested to further
characterise the hose performance. For several tests more than
one sample per test was tested. Also reliability and safety
requirements in EN1474-part 3 were carefully considered. This
resulted in a significant extension of the test program. The next
two sections address some of the extensions to the qualification
test program.

PERFORMANCE UNDER OFF SPEC CONDITIONS

EN1474-part 2 requires testing leak tightness for light
damage only. In the extended program, extreme impact damage
and crushing were applied at both ambient and cryogenic
conditions. Series of damages were applied to determine what
Figure 6, flow test setup ambient (left) and cryogenic LNG (right) © Gutteling damage could be detected in service. With respect to damage
B.V.. tolerance, there appeared not to be much difference between
testing under ambient or cryogenic conditions. Figure 8 shows
With the so-called thermal fatigue test it is proved that repeated the extreme damage inflicted to the parts at ambient conditions.
cool down - heat up cycles do not result in leakage. Finally a Similar damage has been applied in cryogenic operating
bending fatigue test is to be performed. Under operating conditions. Even after this extreme damage no leakage
conditions (cryogenic and pressure) a minimum of 400,000 occurred.
bending cycles shall not result in leakage or damage. EN1474 –
part 2 specifies neither the load cycle nor what should be
considered to be damage or leakage. For the Gutteling hose, a
large test set-up was built to simulate the transfer system
accurately. The dominant bending fatigue in the system is
induced by the roll of the LNG vessels. Figure 7 shows the hose
bending fatigue set-up. The similarity with the system
configuration in figure 1 is obvious. In the test set-up this
bending and straightening is simulated by cyclic moving the
right-end upwards and downwards.

Figure 8, impact and crush loads applied to the hose © Gutteling B.V..

4 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 The damaged hose samples were subsequently tested until they
burst. The surprising result was that even with extreme damage,
the hose burst pressure still well exceeded the requirement. A
further positive result was that in cryogenic and ambient burst
testing no jetting of liquid occurred. A kind of diffuse leaking
process was observed instead. Jetting of LNG is a risk since the
cold fluid can potentially reach the steel ship structure. That
structure subsequently can become extremely brittle and fail.
Obviously countermeasures are required to prevent such an
event. With this respect the multi-composite hose has the
preferred failure mode.

It was assumed that the outer stainless steel wire has an

important role on the pressure carrying ability of the hose.
Thus, the effect of failing of the outer wire was assessed, see
Figure 9. Also in this case no large spill of LNG is anticipated
under its normal operating conditions. It is believed that this Figure 10, test setup for hose emergency release.
excellent damage tolerance performance is inherent to the
multi-composite hose technology. To that end, a system may be installed that releases the hose and
couplings in a controlled manner. TNO performed tests on such
system at ambient operating condition and at elevated pressures
to simulate the higher stiffness at cryogenic condition. Part of
the test program was also a release without such system in
place, see Figure 10. In this test it was proven that the
pressurised hose will very likely not fail and leak. This test
program was supported by numerical dynamic analyses using
the hose stiffness characteristics determined in qualification
testing. The hose dynamics during the release event appeared to
be predictable.

In a cryogenic axial tensile test it was proven that the hose axial
strength exceeds the 600 [kN]. All these additional tests for off
spec operation conditions provide valuable input to the risk and
safety assessment for qualification of the system. The risk of a
large spill due to the direct failure of the 8” Gutteling hose is
considered to be very small. This is a major result since up until
these tests the hoses were pictured as suitable for emergency
transfer only.

VALIDATING HOSE FATIGUE PERFORMANCE

Particularly for continuous operating transfer systems,
Figure 9, simulation of failure of the hose outer wire © Gutteling B.V.. fatigue performance is one of the important parameters of the
hose performance. EN1474-part 2 requires performing a
Transfer systems have to be fitted with a system that allows cryogenic bending fatigue test and to prove that 400,000 cycles
instant release of the hoses and separation of the vessels in case can be passed without leakage. EN1474-part 3 supposes that an
of emergency. To that end transfer systems are fitted with an engineering model is available allowing to translate load spectra
emergency release coupling (ERC). That coupling first closes into a damage accumulation model to assess fatigue. That load
both sides of the system and subsequently separates. The rapid spectra may consist of high frequency, flow induced vibration
closing operation can result in significant pressure peaks in the resulting in small cyclic stresses and low frequency high
transfer and vessel system. The high pressure capability of the bending stresses due to the relative vessel motions. Clearly the
Gutteling hose (>150 [barg]) proven in the burst testing covers part 2 testing (one fatigue test only) will provide insufficient
this effect adequately. In releasing the hoses, in emergency data to provide input to these part 3 requirements. In addition,
situations, both personnel and the vessel shall be protected unlike for metal designs, no engineering models exist for multi-
adequately to prevent further damage. composite hoses in the public domain that translates global hose

5 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 deformations to strains in film and fabric plies. The Gutteling qualification program on a multi-composite hose for
consequence is that the hose fatigue performance can only be aerial transfer systems, and assuming that hose qualification
tested on the full-scale level. This is, particularly for large programs should provide sufficient information for system
diameter hoses, extremely expensive and time consuming qualification, the required program seems not to be balanced.
considering the full spectra of loadings. In addition, when a The purpose of characterising the hose stiffness is in particular
small change in materials or the production process is made, limited considering the non-linearity, hysteric behaviour and
such an extensive program needs at least partly to be repeated. creep. Too little emphasis is placed on damage tolerance and
Further, the load spectra of hoses in service will deviate from fatigue performance. These properties are essential for the
that of the test program. reliability and safety over service life. Depending on the system
creep would also need to be reviewed in more depth. Multi-
Without an engineering model one cannot estimate the effect of composite hoses pose some very interesting performance
the deviating service load spectra on the service life. As such characteristics such as light-weight, high flexibility and a gentle
the safety and reliability of the transfer system over service life failure mode which certainly adds to the system safety. The
cannot be substantiated adequately without such a model. In the LNG industry will benefit if engineering models are available in
Gutteling qualification program this is partly addressed by order to predict the fatigue and creep similarly as for metal
doing additional fatigue tests. Such an engineering model based structure.
consists of two parts. The first part links the global hose
deformation to local deformations of the films and plies. The ACKNOWLEDGMENTS
second part involves the material fatigue failure model. The The authors thank Gutteling B.V., Exmar and Exellerate for
material fatigue failure model can likely be developed on the allowing TNO to use the information from the 8” multi
material level. Figure 11 depicts the anticipated process of composite hose qualification program.
predicting the fatigue performance of hoses.
REFERENCES
For the multi-layer composite hoses such a failure model will be [1] EN1474-1 (December 2008)- Installation and equipment
complex since a local small failure in one film will not result in for liquefied natural gas – Design and testing of marine
leakage or gross degradation of the hose. A damage transfer systems – Part 1: Design and testing of transfer
accumulation model has to be developed in addition. It is arms.
believed that the industry will benefit from such a model. Hose [2] EN1474-2 (December 2008)- Installation and equipment
manufactures can validate their products and alter their products for liquefied natural gas – Design and testing of marine
at much lower costs and shorter lead times. End users benefit transfer systems – Part 2: Design and testing of transfer
from being able to predict the service life dependent on the hoses.
application and assess the safety over the service life. The class [3] EN1474-3 (December 2008)- Installation and equipment
organisations have a tool to judge the reliability and safety of a for liquefied natural gas – Design and testing of marine
hose in a particular system. Therefore TNO took the lead to set transfer systems – Part 3: Offshore transfer systems.
up a Joint Industry Project. [4] Foss, M.M., 2006, “Offshore LNG receiving terminals”,
Center for Energy Economics.
[5] McDonalds, D., Chiu, CH, Adkins, D, 2004,
CREEP PHENOMENA “Comprehensive evaluations of LNG transfer technology
As mentioned in the previous chapter, some form of for offshore LNG development”, Proc. of LNG14.
creep or relaxation makes the hose more flexible over its [6] Rombout G., Peigne, A, Loisel, P., Le Cloirec, A..,
service life. EN1474-part 2 mentions that for a pressure test the Machouat, F. Maocec, D., 2008, “LNG trials of a new 16”
hoses have to be stabilized for some hours. Other than that it flexible hose based LNG transfer system”, OTC 19405.
does not require any testing for creep. [7] Witz, J.A., Ridolfi, M, Hall, G., 2004, “Offshore LNG
When hoses are used in a system that exposes the hose transfer – an flexible cryogenic hose for dynamic service”,
continuously to a load, it is anticipated that creep performance OTC 16270.
becomes a significant parameter for safety and reliability. Creep [8] Offshore mooring fibre ropes, offshore standard DNV-OS-
is a function of load level and time. As for fatigue, this can best E303, October 2008.
be treated by an engineering model. That model consists of [9] Cordage Institute standard CI-1500, Test Methods for Fiber
similar steps as the fatigue model. Rope, May 2006.
[10] Mallon, N. and van der Weijde, G., 2011. “Influence of
CONCLUSIONS hysteresis on the dynamics of cryogenic LNG composite
The recently released EN1474 standard provides hoses”, In Proc. of ASME-OMAE 2011, OMAE2011-
guidance to qualifying hoses, loading arms and transfer systems. 49188.
Aiming to cover all kinds of transfer systems using different
kind of technologies, it is very general. When reviewing the

6 Copyright © 2011 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 Mechanical- thermal Degradation/aeging of
Degradation/aeging of plies
hose model hose

Probability of failure:
Film properties evolution of properties
Hose production
Hose properties
process
of risk

7
Fabric properties Consequences of
Evolution

failure: failure mode

Strains, thermal load and Production
aeging in plies parameters Mechanical load spectrum
(frequency/time, strain
Condition monitoring
levels)
Strains, thermal load and Mechanical- thermal
Thermal loads, Aeging
aeging in stacks hose model

predicting fatigue performance through an engineering model.

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

Copyright © 2011 by ASME