OTC 14017

Technical and Economical Evaluation of Composite Riser Systems

P.Odru and Y.Poirette, Institut Franais du Ptrole, Y.Stassen, Bouygues Offshore, J.F.Saint-Marcoux, Paragon Litwin,
L.Abergel, Technip-Coflexip
Copyright 2002, Offshore Technology Conference
This paper was prepared for presentation at the 2002 Offshore Technology Conference held in
Houston, Texas U.S.A., 69 May 2002.
This paper was selected for presentation by the OTC Program Committee following review of
information contained in an abstract 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 or its officers. 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 an abstract of not more than 300 words; illustrations may not be copied. The
abstract must contain conspicuous acknowledgment of where and by whom the paper was
presented.

ABSTRACT:
Use of composite materials in the deep sea oil production riser
systems may allow a dramatic decrease in weight, as well as
improved fatigue resistance to loads induced by environmental
conditions. Many concepts have been developed by the
industry and could be available in the next years. However, the
cost of composite components will always be higher than the
one of steel components, and only significant advantages for
particular applications will justify their use by the industry.
Up to now, cost comparisons have been made essentially for
TLP or SPAR riser systems. This paper presents a study that
has been carried out to compare steel and composite riser
solutions for catenary risers, submerged export lines, and
hybrid riser towers in ultra deep water. Specifications were
first proposed, following which steel and composite solutions
were designed and compared. This was done both from the
feasibility point of view as well as from consideration of the
operational advantages resulting from the lightweight and the
fatigue resistance of the composite. Acceptable costs of
composite risers were then deduced.
The main conclusions are that in the mild conditions of the
Gulf of Guinea or of Brazil, both steel and composite solutions
are technically feasible, although steel solutions come close to
their limits. Composite riser joints fabricated in moderately
long lengths seem to be the most interesting solution for
transportation and laying purposes. Large cost advantages may
be obtained, particularly during the laying phase, which can
justify using the composite solution.

INTRODUCTION:
During the nineties, the oil industry proceeded to exploit
offshore fields in 1000 metres water depths and beyond. This
has been done generally by simple extrapolation of existing
architectural floating concepts, such as TLPs and FPSOs, or
by using new ones such as SPAR systems. A significant new
development has been the introduction of Steel Catenary
Risers. Also the exploitation of new lightweight materials has
begun with the introduction of the first mooring systems made
out of polyester ropes. New problems have also arisen such as
the necessity for significant thermal insulation of pipes.
With increasing water depth, and particularly in Ultra Deep
Water (between 2000 - 3000 m), the need to decrease the
weight of the risers and mooring lines will become stronger.
The introduction of lightweight materials may become the best
economic option, or possibly the only technically available
option. Among such materials, high strength composites have
been the subject of intensive industrial research and may
become commercially available for operational purposes in the
present decade.
Up to now attention has been generally focused on weight
sensitive floating systems, such as TLPs, on which the balance
of weight and the advantages induced can be easily deduced.
The objectives of the present study were to evaluate the
technical and economic interest of other pipe systems, such as
catenary risers, export lines, or hybrid towers, where the
balance of weight is only part of the interest, and where steel
solutions have run into barriers including fatigue behaviour,
stiffness, installation loads, thermal insulation, etc. The study
was carried out by IFP on behalf of the CLAROM association.
IFP dimensioned the composite pipes and calculated their
behaviour, while the three offshore contractors calculated the
equivalent behaviour of the steel alternatives. They also
evaluated the installation methods, and the cost difference by
which the composite solutions will have to be justified.
CATENARY WATER INJECTION PIPE:
Catenary risers would appear to be good candidates for
composite applications since composite materials have
excellent qualities of lightness and fatigue resistance. A water

P.ODRU, Y.POIRETTE, Y.STASSEN, J.F.SAINT MARCOUX, L.ABERGEL

injection pipe requiring no insulation in very deep water was

chosen for the study.
Specifications:
The specifications used for both composite and steel pipes
were an internal diameter of 11 1/2 and a top end service
injection pressure of 1.5 MPa. The water depth considered was
2500 m. In service, the risers are always full of water. To
reduce weight the steel pipe is laid empty and has to resist
collapse pressure, but the composite pipe can be laid full of
water. The characteristics of the spread moored FPSO are
identical to the one proposed by Bouygues Offshore for the
BONGA field. Two cases were considered: one in the Gulf of
Guinea, and the other for Brazil, using corresponding
environmental data.
Static design:
The static design for both steel and composite cases is carried
out considering a pipe that withstands the internal and external
pressure requirements. The initial profile of the catenary riser
is determined by the shape of a riser of a given length installed
between the FPSO in its mean position and a fixed point on
the seabed representing the extremity of the flowline. Around
this mean position, offsets are superimposed, accounting for
low frequency motions of the FPSO induced by wind, wave
and current. They induce near and far positions (figure 1), for
which corresponding stresses are calculated. If they are found
unacceptable, a new mean position is chosen, and the
calculations are repeated until acceptable results are obtained.

OTC 14017

Composite pipe:
The design of the composite pipe is more complex than the
steel one, due to the anisotropy of the basic material.
However, the empty pipe being naturally buoyant, and its
weight full of water remaining very low, it can be laid full of
water. So there is no requirement for the composite pipe to
resist collapse pressure. Because of its low apparent weight,
the composite pipe is only subject to very small loads, which
leads theoretically to a very thin wall thickness requirement.
Hence the pipe has to be designed on robustness criteria, in
order to be able to withstand shocks and handling loads. To
make the pipe robust two versions were studied with overall
wall thickness of 10 mm and 20 mm respectively.
Four designs were proposed and analyzed (see Table 1), in
which carbon fibers are used to resist structural loads, and
glass is used to increase wall thickness to make the pipe
robust. These designs have been chosen for the following
reasons. Two of them (C10 and C20) have the minimum
thickness of carbon fibers required to withstand the loads, the
remaining thickness being made up of glass fibers in order to
give total thickness of 10 and 20 mm respectively. The two
others (C10 and C20) are made mainly out of carbon fibers,
leading to higher axial stiffness.
The four structures were analyzed and the stresses induced in
the three cases (near, mean, and far positions) were found fully
acceptable.
Dynamic and fatigue verifications:

Steel pipe:

Brazil case:

A particular specification for the steel pipe was an additional

corrosion thickness of 10 mm, which is probably severe. The
pipe was designed to withstand the collapse pressure (without
the corrosion thickness), the effects of weight, and the touch
down point bending curvature, in near and far positions,
corresponding to a drift of 8% of the total water depth.
The current was imposed in the direction with the least
favorable effect.

The dynamic study was carried out considering a Jonswap

spectrum representing hundred year conditions with a
significant wave height of 7.8 m and a peak period of 15.35 s.
The drift was taken to be 8 % of the total water depth (200 m).
The RAOs were obtained using the DIODORE@ software.
Two cases were considered: one with the waves perpendicular
to the vessel, and one with the waves in line with the vessel.

Two materials were considered: X70 and P110 grades. The

global analysis was made using the DEEPLINES@ software.
A 13.8 mm wall thickness was found necessary, to which 10
mm of corrosion thickness was added. The X70 steel was
found slightly insufficient to withstand the induced stresses
with the required safety coefficients.
The steel pipe presented the following characteristics:
Wall thickness (mm):
without additional corrosion thickness:
Apparent weight, full of water (N/m):
Apparent weight, empty (N/m):
Axial stiffness (MN):

23.8 mm
13.8 mm
1500 N / m
900 N / m
4500 MN

With waves perpendicular to the vessel, important roll

movements are induced, which generate high tensile and even
compressive loads in the catenary risers. They were found
unacceptable for both steel riser and composite case.
With waves in line with the vessel and hence perpendicular to
the plane of the risers, both composite and steel designs were
found acceptable. The level of stresses induced is however far
more favorable in the composite version. The case with waves
at 45 to the vessel was found to be acceptable for the
composite pipe.

OTC 14017

TECHNICAL AND ECONOMICAL EVALUATION OF COMPOSITE RISER SYSTEMS

Gulf of Guinea case:

Environmental conditions in the Gulf of Guinea are highly
unidirectional with narrow frequency-band waves. The
Jonswap spectrum with significant wave height of 3.7 m and
peak period of 13.9 s was used to represent hundred year
conditions, which are less severe than the extreme conditions
for Brazil. Analyses of the risers with such waves
perpendicular to the vessel or in line with it, showed that both
the steel and the composite designs were satisfactory.
Although the sea states are mild, they are persistent, and their
effects on fatigue must be studied. The fatigue analysis was
performed using a wave scatter diagram, using both frequency
and time domain methods. The method to assess the fatigue
life of steel pipes is well known. A safety coefficient of ten
has to be applied to the damage calculated using Miners rule.
This method does not apply for composites. Carbon fibers
however present excellent fatigue properties and their ability
to withstand fatigue loading can be reasonably estimated.
The principal result of these studies was that the fatigue life in
the environmental conditions of the Gulf of Guinea was
acceptable for both steel and composite solutions.
Design conclusions:
Both steel and composite designs have been found acceptable
as 12 OD water injection catenary risers in the deep
offshore Gulf of Guinea conditions. However, stress levels in
the steel riser are very high requiring the use of high-grade
steel such as P110. In contrast the stress levels in the
composite are low and such risers only have to be designed for
robustness. Their self-weight is low.
Difficulties appear when trying to design both pipes for the
case of hundred year waves perpendicular to the vessel in the
Brazilian environment. Problems result from the roll
movements of the vessel, which is not adapted to these
environmental conditions, and can be improved by using a
FPSO with turret or flexible pipes. This was beyond the scope
of the study.
Although both materials can be used to achieve satisfactory
designs, they lead to major differences in self-weight of the
two pipes suspended from the vessel. This results in a 420 T
top tension for the steel pipe, and around 28 T for the
composite one.
Estimate of economic benefits:
The estimation of economic benefits resulting from the use of
composite materials is a difficult exercise. The methodology
used was to estimate the economic influence of the following:

the weight of the pipe on the floating support;

the accessories (stress-joints, ball joints, others);

the installation procedure.

The sum of these influences gives the differences between the

steel and the composite concepts, as well as giving indications
on the way the composite has to be manufactured. It will then
be possible to deduce the acceptable cost for the composite,
and in what conditions this can be matched.
The influence of the weight of the pipe is not negligible
(around 30 T for the composite instead of 420 for the steel) but
does not seem to have a significant influence on the direct
design of the FPSO structure, which is capable of supporting
considerable weight.
However, the lightness of the composite pipe, combined with
its flexibility, has an important influence on the requirements
of the topside connection (stress joint or ball joint) which
represents an important part of the global cost. In some cases it
governs the feasibility of the catenary riser solution itself.
The most important impact seems to be on the installation
method. In the steel case a classical procedure has to be
followed. The flowline is raised from the seabed to the surface
and the riser elements are then vertically welded. The vessel
moves towards the FPSO while the elements are added and
laid using the J-lay method. The head of the riser is then
transferred to the FPSO and finally installed. This requires the
use of large laying vessels, such as SAIBOS FDS or the
Seaway POLARIS.
The laying of the composite version may be completely
different. A significant point is that the composite riser joints
need to be equipped with steel connectors which are heavy and
expensive. Three types of composite solutions have been
considered, characterized by the lengths of the riser joints:

Short lengths in the range of 10 to 25 meters can be easily

manufactured and handled, but the cost induced by the
number of steel connectors does not favor this solution;
Intermediate lengths (100 to 300 m) can probably be
manufactured, transported and assembled close to the site
at optimal cost;
Long lengths (300 m to 4000 m) may be considered, but
they would probably be manufactured in a dedicated site
far from the final destination. The required long tow could
be very hazardous. Reeled pipe technique could be
envisaged, but composites having no plasticity domain,
this would induce too large stresses for such types
of diameters.

The second option appears as the more interesting, and was

chosen as the reference case for the whole study. The
composite riser having been assembled and filled with water
could then be towed at the mudline (as for the Girassol
bundles) by two tugs. Before installation of the FPSO, the riser
can be laid directly in the alignment of the flowline and

P.ODRU, Y.POIRETTE, Y.STASSEN, J.F.SAINT MARCOUX, L.ABERGEL

abandoned. Because of the length of the catenary, the

abandonment head would lay on the opposite side of the
FPSO. It can then be picked-up when the FPSO is fully
moored, raised by a tug that transfers the riser head to a chain
operated from the vessel, and installed at its place on the other
side of the vessel [8].
If the influence of the difference of weight on the FPSOs
structure seems negligible, the economic consequences on the
two other items (accessories and installation) have been
evaluated in Table 2. The cost for the composite pipe does not
include the transportation of the intermediate length joints
from the manufacturing plant and their on site assembly, as
well as the mobilization cost of the heavy vessel in the
steel case.
Conclusions:
The balance is of around $1.35 million, on which the excess
cost of the composite version versus the steel one will have to
be justified. It is difficult to estimate the cost of such a product
today. It seems however that the C10 version, which includes
the least material, could be a candidate for such an application,
if manufacturing costs and associated investments are not
too high.
Estimates have been made for both the Gulf of Guinea and the
Brazilian cases. If both steel and composite cases have been
found technically feasible, the former requires high-grade
steel, where as the composite solution leads to very low
stresses. These results suggest that in more severe
environments the steel case will reach its limit. On the other
hand the composite solution should be more competitive, due
to the low stress levels induced by self-weight, and the high
compliance and fatigue resistance.
APPLICATION TO CATENARY
PRODUCTION RISERS:
The use of catenary production risers made out of steel or
composite for direct production purposes in very deep water
will raise the problem of thermal insulation. This was not
addressed in this study. The second problem will be that in a
production case the riser may be empty, or full of gas. The
composite riser will then become buoyant, which is
unacceptable. For such an application we propose to equip the
composite riser with an internal steel carcass, as in flexible
pipes. This will secure the internal liner and allow the weight
to be adjusted. The specification for the study comprised an
internal diameter of 8, a top service pressure of 10 MPa with
a possible maximum at 35 MPa, and a water depth of 2500 m.
The characteristics of the FPSO and the environments studied
were the same as for the water injection riser.
The steel riser was designed following API 2RD, which gave a
9.5 mm thickness on which a 5 mm corrosion thickness

OTC 14017

(probably pessimistic for most of usage) had to be added. The

linear weight in water was then of 675 N/m.
The composite riser was designed to withstand the tensile and
pressure specifications. Its apparent weight full of gas was
adjusted to be zero through the use of an internal steel carcass.
Carbon fibers orientation:
Carbon composite wall thickness:
Glass fibers orientation:
Glass composite wall thickness:
Pipe axial Modulus:
Apparent weight with end fittings:
Apparent weight with internal carcass:

60/20
6.5 mm
55
13.5 mm
11 GPa
160 N/m
440 N/m

Both steel and composite risers were found able to work in

static conditions for a given mean position and a drift of 8%.
Problems were found when the empty pipes were subjected to
a current perpendicular to the axis of the FPSO. The steel pipe
showed a corresponding displacement close to 70 m, which is
probably acceptable in 2500 m water depth. However the
composite riser in the same conditions showed a displacement
of 600 m (figure 2), which is certainly unacceptable, due to the
risks of contact with the other risers. Further analysis indicated
that this was due to the very low weight of the empty
composite pipe (40 N/m). This displacement was found
dependant only on the weight and not on the stiffness.
It can be concluded from this approach that weight may be
necessary to withstand these extreme conditions.
EXPORT LINE:
The second composite application studied was the case of an
export line (Figure 3). The export line of the Girassol field is
made out of two 16 steel pipes, shaped in a W form by the
use of syntactic foams [ref 9], and called for this reason lazy
W. Export lines with larger diameters such as 22 cannot
meet the required fatigue safety coefficient of ten due to the
motions of the Surface Buoy. The interest of a composite
version was its good fatigue resistance to surface
buoy excitation.
Specifications:
Specifications for the line were following:
Length (steel Lazy W):
Diameter:
Service pressure:

2100 m
22
2 MPa

The line is always filled with water or treated oil (density 0.8);
no corrosion thickness was required.

OTC 14017

TECHNICAL AND ECONOMICAL EVALUATION OF COMPOSITE RISER SYSTEMS

Environmental specifications:
The FPSO is considered as fixed. The buoy is located at a
nominal distance of 1850 m of the FPSO, making an angle of
64 with the axis of the vessel. The movements of the buoy are
described by RAOs. The environmental data (centenary wave
and sea state scatter diagram) correspond to the Gulf
of Guinea.
Steel pipe design and analysis:
The steel pipe design was following:
Nominal diameter:
Steel grade:
Yield limit:
Wall thickness:
Apparent weight full of water:
Apparent weight full of oil:

22, 559 mm
API 5L X 65
450 MPa
25 mm
2850 N / m
2450 N / m

The extreme conditions analysis was carried out with the

ORCAFLEX@ software, considering a hundred-year wave in
the direction of the vessel with a significant wave height of
3.62 m and a peak period of 15.9 s. A wind wave in the
direction of the line with significant height of 1.9 m and a
peak period of 6.3 s was added. No current was considered.
The fatigue simulation was carried out considering a
representative scatter diagram and using a spectral method.
A statistical variation of the stress in the pipe can be deduced
from each wave period. S-N curves are then used to evaluate
the damage associated with the number of cycles and the
global fatigue life is estimated with a safety coefficient of ten.
Only first order movements are taken in count; damages due to
installation, second order movements and VIV were ignored.
Global calculation showed that the maximum damage zones
were located close to the buoy, and that the pipe could not
fulfill the safety coefficient of ten. Two solutions can be
then proposed:

use of three smaller diameter lines, like 16 lines

for instance;
use a subsurface buoy, in order to decrease the
movements induced by the environment; this solution was
evaluated during the study, and found acceptable,
maximum damage zone being then in the middle of
the pipe.

Composite pipe design and analysis:

The mechanical and environmental specifications of the lines
being mild, the use of carbon fibres does not seem to be
necessary. A composite material made out of glass fibres in a
thermoset matrix was proposed, for cost purpose. Two cases
were considered, differing by the orientation of the fibres and

the corresponding axial and bending stiffness. The design

criterion was in fact the external pressure. The wall thickness
found seems to be able to insure sufficient robustness of the
pipe (see Table 3).
The lazy W form, as the analysis showed, is required to avoid
the apparition of negative effective tension or real
compression loads in the pipe. It is obtained through the use of
syntactic foams blocks adequately disposed. With the
composite design, the difference of apparent weight between
the case of the pipe full of oil or of water becomes very
significant although weight is very low, and this induces very
different and incompatible behaviours (the Lazy W form
cannot be kept). It was so proposed to use the composite pipe
as a double catenary line, which will have for interest not to
use the expensive syntactic foams blocks.
Results of the dynamic analysis showed that the stresses
induced in the composite for both designs were fully
acceptable, even if the radius of curvature in extreme
conditions could be quite small. However, important values of
negative effective tension or even compressive loads are found
in the proximity of the buoy and of the vessel. They do not
seem to be a problem for the pipe, for they apply only for very
small time (2 to 3 s), reasonably avoiding buckling risks. But
some real compressive loads may be transmitted to the flexjoints, which are not designed for. If the flex-joints are
suppressed, the stresses in the direction of the fibres become
too important, needing the use of an adequate stress joint.
Economical evaluation:
The economic interest of the composite solution will be
evaluated through the comparison with the cost of the steel
version, including transport and laying. The economic
advantages induced then by the composite solution are
quantified, and the acceptable cost of the composite pipe can
be deduced.
The steel version is made of two flex-joints with their supports
and of steel lines assembled by welds with their coatings,
anodes and buoyancy modules. The scenario for the
installation of the steel pipes would be the following. The
pipes are manufactured by 40 lengths and transported to the
site. They are then welded two by two and transported on the
laying barge. The floaters are installed during this operation.
When the line is laid, hydrostatic tests are performed and the
connection to the floating supports is achieved (FPSO
and buoy).
The scenario for the composite line has some differences. As
previously stated for the composite catenary riser, middle long
lengths (100 m or more) are manufactured in a distant
manufacturing plant, and are then transported to the yard,
where the pipes are assembled and the hydrostatic tests
performed. The two lines are then towed to the site by a tug at

P.ODRU, Y.POIRETTE, Y.STASSEN, J.F.SAINT MARCOUX, L.ABERGEL

each extremitiy. On site, the lines are installed and connected

to the floating supports.

OTC 14017

Core pipe:
The core pipe has to be designed for a 1000 T tension.

Admissible cost for the composite solution:

Composite riser design and analysis:
Considering the simplification of the offshore operations, the
admissible cost for the following furniture:

composite material pipes with their connection systems,

necessary flex-joints and connecting systems,
buoyancy if needed,
transport from the manufacturing plant to the yard,

would be of around $ 6 millions. The advantage may be

possibly more, for the need for additional steel lines or for a
subsurface buoy was not quoted. But the composite solution
will probably have other specific problems, as the necessity to
include adequate stress joints at each long length connecting
system, which was not studied in details.
Here too the authors of the paper are not composite
manufacturer and cannot answer definitely, but such a cost
seems possible to reach, if conditions are favourable.
HYBRID TOWER:
Hybrid tower is an innovative concept developed for the giant
Girassol field in the Gulf of Guinea, and installed during
summer 2001. It consists (in this particular case) in three
vertical towers made of a core pipe sustained by a subsurface
buoy, along which the risers and facility systems are
assembled (see figure 4). Syntactic foams disposed around the
pipes insure both buoyancy and thermal insulation (a joint
Stolt Offshore Doris Engineering patent). The connection to
the 300 000 T FPSO is made through a subsurface flexible
pipes system.
The objectives of this study were not to evaluate directly the
economic advantages able to be brought by a composite
version, for the design of such a system is quite complex. But
to evaluate the advantages in terms of weight saving and
possible thermal insulation, as well as the problems that could
arise in a composite version.

One of the problems of the design of such a system is that the

syntactic foams are used for both buoyancy and thermal
insulation. But when the pipes expand under thermal or end
pressure effects, gaps appear between the blocks of syntactic
foams, inducing thermal convection and a loss of insulation
efficiency. Composite pipes however can be tailored to reach
specifications impossible with isotropic materials, such as a
Poissons ratio close to .5 (which would considerably reduce
elongation under end pressure effect), and a thermal expansion
close to zero. In this case however, the initial axial expansion
under the pure effect of the weight will stay important, and
needs the use of carbon fibre composite to be reduced.
The second problem arising with the composite solution will
be that, when full of gas or empty, the pipe will be buoyant,
creating an axial compressive force able to induce buckle. This
problem is well known for instance for the peripheral lines of
drilling risers, and the problem can be solved by increasing the
axial stiffness of the pipe and displaying lateral guides along
it, calculated in order that the critical buckle length cannot be
reached. An other solution could be to increase the weight by
displaying an internal or external steel carcass able to protect
the internal and / or external liner, and adjust it so that the
weight is never negative. The design of the carbon composite
pipe could be the following, without additive weight:

Comparison with steel pipe:

Specifications:
The water depth considered is of 2500 m.
Production risers:
Internal diameter:
Maximum service pressure, sea floor :
Normal fluid density:
Service temperature:

103/4, 273 mm
35 MPa
0.8 T/m3
40C

24 / 76
19 mm
0.55
180 N/m
- 290 N/m
1 030 MN
8.3 MN.m2
41 MPa
-0.05.10-6
10.7 m

Fibres orientation:
Wall thickness:
Poissons ratio:
Apparent weight, full of oil:
Apparent weight, empty:
Tensile stiffness:
Bending stiffness:
Collapse pressure:
Thermal elongation:
Critical buckling length:

Wall thickness:
Apparent weight, full of oil:
Apparent weight, empty:
Total weight, full of oil:

Composite pipe

Steel pipe

19 mm
180 N/m
- 290 N/m
45 T

15 mm
758 N/m
290 N/m
190 T

The suspended weight of the pipe full of oil is reduced from

190 T for the steel pipe to 45 T for the composite pipe.

OTC 14017

TECHNICAL AND ECONOMICAL EVALUATION OF COMPOSITE RISER SYSTEMS

Thermal aspect:

Catenary water injection risers have been found to be

technically feasible in 2500 m water depth for the
environments of Brazil and Gulf of Guinea for both
composite and steel versions. The steel solution however
requires the use of a high grade material, whereas the
composite one can be dimensioned without difficulty. The
economic evaluation shows then that the very large saving
in riser weight has little effect on the FPSO itself. The
major economic benefit comes from the simplification of
the laying method, which allows small vessels to be used
for the composite solution. The excess allowable cost of
the composite version versus the steel one was estimated
to be $1.3 million. It seems possible to the composite
solution to reach this objective. The fabrication of
intermediate lengths (100 m) would appear feasible and
reasonably well suited for transportation and on site
assembling, but the cost involved is difficult to establish.

The technical feasibility of a composite production

catenary riser has also been evaluated. The weight of the
composite has to be adjusted in order to avoid it being
buoyant when full of gas. The resulting lightness will lead
to very large lateral deflections under the effect
of currents.

The feasibility of the steel version of the Lazy W export

line system is limited to diameter of 16. The composite
version on the other hand has been found to work as a
catenary with the required diameter of 22. It will require
stress joints at extremity instead of flexjoints because of
the low tension. The economic advantages induced have
been quoted in the paper at around $ 6 millions and a
composite version may be able to reach the objective,
particularly because low cost glass fibres could be used.
Here too, this will depend on the availability of long
lengths of composite pipes and their costs.

The case of the hybrid tower has only been studied from a
technical point of view. Composite production risers lead
to large weight savings. However their negative apparent
weight, when full of gas, will induce compressive loads,
needing stiffness or additional weight to keep from
longitudinal buckling. The thermal properties of the
composite materials have not been found to be a
significant advantage, because of their low thermal
capacity. An interesting application of composites could
be to replace the central core pipe by carbon tendons in
very deep water.

The oil temperature of the Girassol field is quite low (60C)

and an important insulation is necessary to keep the fluid from
reaching wax or hydrates formation points during
transportation. The specification was a thermal exchange
coefficient of 1.5 W/m2.K.
Table 4 presents the thermal properties of the different
materials. Calculations of required insulation material for steel
and composite solution give 210 mm in the first case and 200
mm in the second one. The difference is quite negligible.
The second aspect is the relaxation of temperature when the
production is stopped. The calculations show that in the steel
solution the required time for the temperature to fall from
40C to 20C will be of 20 hours, although it is only of 15
hours for the composite version, due to the lower thermal
capacity of the material.
As a conclusion, using composite in these conditions will
bring a marginal advantage from the insulating point of view,
and a disadvantage from the thermal capacity point of view.
Core pipe:
The specification of the core pipe is a tensile capacity of 1000
tonnes. It can be designed as an empty steel pipe, and so be
neutrally buoyant. But this becomes impossible in very deep
water, for the collapse design will necessitate thick walled and
heavy pipes. The solution seems to be here the use of
pultruded carbon fibre rods.
A tendon of 1000 T capacity could be made of 300 rods of 6
mm diameter, for a linear apparent weight of 42 N/m without
connections, which leads to a total apparent weight of 13
tonnes in 3000 m water depth. Such a product is to day very
close to be industrially available.
Conclusion hybrid tower:
From a technical point of view, use of composite could allow a
very important mass reduction of hybrid towers, as well for
the riser system as for the central tension member in Ultra
Deep Water. Problems should arise however as the pipes full
of gas are buoyant, needing to take care of possible buckling.
The use of composite material does not seem to bring thermal
properties advantages, although the composite is more
insulating than the steel.
CONCLUSIONS:
The general conclusions relative to each of the pipe systems
that have been studied can be summarised as follows:

As an overall conclusion composites may find applications of

great interest for these types of offshore systems, but their use
will depend strongly on the availability of pipes fabricated in
long lengths at low cost. These advantages could be widely
increased in more severe environmental conditions.

P.ODRU, Y.POIRETTE, Y.STASSEN, J.F.SAINT MARCOUX, L.ABERGEL

OTC 14017

6.

D.B.Johnson, Rigid Composite Risers for Deepwater Oil

Production, Intertech Second International Conference
on the Global Outlook for Carbon Fibre, San Diego,
California, November 1999.
7. B. Melve, T.Meland, P.A.Bergh, K.Klovfjell, Continuous
Composite Lines for Methanol Injection on the Asgard
Field Experiences from Production and Installation,
OMAE2000/MAT-2422.
8. X. Wang, X. Zhou and A.J. Ginnard, J.T. Sprot,
'Installation method evaluation of export SCRs', OTC
12969, Houston 2001.
9. Legras J.L, Traube D, New Concept of Export Line for
Deepwater Fields, ISOPE 99.
10. Hanna S.Y, Salama M.M, Hannus H. New Tendon ans
Riser Technologies Improve TLP Competitiveness in
Ultra Deep Water, Paper 12963, SPE, 2001.

Acknowledgements:
The authors are grateful to all those who participated in the
study and in particular C.P.Sparks from IFP, E.Coche from
Bouygues Offshore, and T.de.Kerdanet from Paragon Litwin.
REFERENCES:
1.

C.P. Sparks, P.Odru, H.Bono, G.Metivaud, Mechanical

Testing of High Performance Composite Tubes for TLP
Production Risers, OTC 5797, Houston 1988.
2. P.J.C. Tamarelle, C.P. Sparks, High Performance
Composite Tubes for Offshore Applications OTC 5384,
Houston 1987.
3. K.H Lo, J.G Williams, M. Karayaka, M.M Salama,
Progress, Challenges and Opportunities in the
Application of Composite Offshore, CMOO3,
Houston 2000.
4. T.M.Hsu, J.Skogsberg, M.Karayaka, Composites
Utilization on a SPAR Platform Potential Economic
Impact and Technical Gaps, CMOO3, Houston 2000.
5. Composite Riser Workshop, Statoil Research Centre,
Trondheim, 14-15 June 1999.

TABLES
DESIGN:
Carbon fibers (, th):
Glass fibers (, th):
Axial stiffness (MN):
Apparent weight (N/m):

C10

C10

C20

C20

90/10, 1.9 mm
55, 8.1 mm
175
95

90/10, 8 mm
10, 2 mm
1020
63

60/20, 2.6 mm
55, 17.4 mm
175
187

90/10, 16 mm
10, 4 mm
2040
118

Table 1: Catenary water injection composite versions design.

STEEL SOLUTION

COMPOSITE SOLUTION

Flowline connector
Flexjoint

Flowline connector
Flexjoint

20%
Raising of flowline
Riser to flowline connection
Riser laying
Riser transfer to the FPSO

10%
Towing
Riser to flowline connection
Riser transfer to the FPSO

80%

20%

Connections:
Cost (*) :
Offshore operations:

Cost (*):
(*): Cost given as a percentage of the steel solution

Table 2: catenary water injection steel and composite version laying operation costs.

OTC 14017

TECHNICAL AND ECONOMICAL EVALUATION OF COMPOSITE RISER SYSTEMS

Fibres orientation:
Glass composite wall thickness:
Apparent weight full of water:
Apparent weight full of oil:
Axial modulus:

Design 1

Design 2

+/- 55
34 mm
580 N/m
160 N/m
58 MN(*)

+/- 10, +/- 90

30 mm
510 N/m
80 N/m
1450 MN(*)

(*): considering a degraded matrix.

Table 3: composite design of 22 export line.

Thermal capacity (Cp):

Thermal conductivity :
Density:

Steel

Composite

Foam

470 J / kg. K
45 W / m. K
7850 kg / m3

900 J / kg. K
0.5 W / m. K
1500 kg / m3

1400 J / kg. K
0.17 W / m. K
850 kg / m3

Table 4: thermal properties of steel, composite, and foams.

FIGURES

Figure 1 : Catenary water injection riser in far, mean and near

positions.

Figure 2 : Shape of composite catenary production riser,

subjected to current, weight having been adjusted to be just
positive when full of gas.

10

P.ODRU, Y.POIRETTE, Y.STASSEN, J.F.SAINT MARCOUX, L.ABERGEL

Figure 3 : Lazy W export line.

Figure 4 : Hybrid Tower concept.

OTC 14017