You are on page 1of 13

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

OTC 14182
Optimized Design of Pipe-in-Pipe Systems

M. Hausner, M. Dixon / DeepSea Engineering & Management Ltd

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
Deepwater subsea developments must address the flow
assurance issues and increasingly these are forming a more
critical part of the design. Pipe-in-pipe systems are one of the
options available in the designers toolbox for overcoming these
problems and are recognized as a thermally efficient, reliable
and proven technology for insulated subsea transportation of
well bore fluids.
Although extremely low U-values are
achievable pipe-in-pipe systems come at a cost and have their
increased weight as a penalty for use in deepwater
developments.
By applying an inside-out optimization process for the
design of pipe-in-pipe systems the top tension loading on the
surface vessel (installation or production) can be significantly
reduced whilst also minimizing procurement expenditure on raw
materials.
Specifically the design optimization of each
component reduces steel volumes and the overall outer diameter
of the system.
This paper presents the methodology for optimized design of
pipe-in-pipe systems and illustrates the potential cost savings in
terms of raw materials and installation cost through a case study
for a typical large West African field. In addition commercial
savings relating to surface platform hull costs are presented for a
case where the development employs pipe-in-pipe risers.
Introduction
At present the pipe-in-pipe (PIP) market is dynamic with
numerous projects requiring pipe-in-pipe solutions and many
more examining pipe-in-pipe as a development option.
The objective of this paper is to present an optimization

design process for pipe-in-pipe systems for deepwater


applications, specifically 1000m or deeper. The focus is on
establishing the actual required pipe diameters for flowline and
carrier, rather than employing API standard sizes, by performing
the thermal and mechanical design in an integrated manner. In
this way the design meets the project requirements for
production rate and steady state thermal performance whilst
minimizing as-installed system cost. Cooldown considerations
have not been included in the designs generated here.
The inside-out design methodology is presented along with
the as-installed costing, which has been used as the ultimate
comparison condition.
The following parameters are
investigated with pipe-in-pipe designs and costs generated for
each variable combination:

U-values of 1.0, 1.5 and 2.0 W/m2K


Flowline lengths of 5, 10, 20, 40 & 60 km
Water depths of 1000, 1500, 2000, 2500 & 3000m
2 types of insulation material polyuethane foam
(PUF) and microporous material (MP)

In addition to presentation of the results for the parametric


matrix detailed above, a project with typical characteristics for a
large West Arican development are discussed including the cost
and top tension implications on the host platform when
employing pipe-in-pipe steel catenary risers (SCRs).
With an increasing number of pipe-in-pipe systems on offer
it is increasingly difficult to rapidly evaluate the options to
determine or identify the most appropriate options on a technical
and economic basis.
What follows is a brief definition and classification of pipein-pipe systems. There are two specific criteria that can be used
to describe any particular pipe-in-pipe system:

Insulation type (material dependent)


Structural compliance (configuration dependent)

Associated with each of the above are compatible types of


field joints and installation methods.
Table 1 is the overall compatibility matrix showing the

M. HAUSNER, M. DIXON

possible combinations of insulation material type, field joint and


installation method for the main structural categories. The table
is arranged in this manner, as it is the structural compliance that
drives the choice of insulation and installation method, the latter
of these heavily influencing selection of the field joint.

OTC 14182

are limited and may not justify the additional design and
procurement complexity. For deepwater project significant cost
benefits are obtainable.
Flow diagram 1 below represents the design methodology for
designing a pipe-in-pipe system using standard API pipe sizes.

Table 1: Compatibility Matrix

Insulation
Group

Compatible
Field Joints

Installation
Methods
System
Usage

Injected PUF
Sprayed PUF
Granular
material (e.g.
microspheres)
Microporous
material
Vacuum (full or
assisted)
Phase change
material
Half shells
Carrier to carrier
butt weld
Threaded
Flowline weld
only (wet field
joint)
Reel
S-lay
J-lay
Flowlines
Risers

Sliding

Fixed

Restrained

!
"

"
"

"
"

"

"

"

"

"

"

"

"

"

"

"

(")

"

"

"

"
"
"
"
"

!
"
"
"
!

(")
"
"
"
!

Input Data

Calculate flowline diameter from


reservoir hydraulics calculations

Select next Std. API pipe diameter flowline

Calculate flowline wall thickness

Select initial estimate


standard API carrier
pipe diameter

Select next standard


API pipe diameter
for carrier

Calculate carrier wall thickness

Calculate U-value

Pipe-in-pipe systems are more installation vessel-dependent


than conventional pipe and this dependency is more pronounced
with:

Increasing water depth

Extreme requirements such as HPHT (high pressure,


high temperature) applications

Use of pipe-in-pipe for catenary risers


The Table 2 contained at the back of this paper presents the
pros and cons for the three structural classifications.
The definitions for strucutral classification and a description
of the insulation material types are contined in Appendix 1.
Optimised Pipe-in-Pipe Design The Inside-Out Method
The inside-out process for designing focuses on optimization
of each layer from the flowline internal bore outwards in order
that thickness is minimized. In this way cost and weight of the
final system are also minimized. This requires the use of nonAPI standard size pipe for the carrier. The use of non-API size
for the flowline is also advocated as this leads to additional cost
and weight savings for deepwater, in particular for pipe-in-pipe
SCRs.
For shallow water applications the benefits of this approach

Project
U-value
met?

No

Yes
Pipe-in-pipe design

Flow Diagram 1
Flow diagram 2 below represents the design methodology for an
optimized pipe-in-pipe system. There is a certain amount of
iteration required in this design process as the contribution of the
carrier pipe wall thickness to the overall heat transfer coefficient
changes with variation in its diameter and wall thickness. This
change in contribution affects the amount of insulation required
to achieve the desired U-value, which then necessitates
recalculation of the carrier diameter and wall thickness. The
iteration continues until the optimized combination is achieved.

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

the heat transfer area is given as:


Input Data

Ao = " ! Do ! L
(2)
where :
Do = Overall outside diameter of the pipe including

Calculate exact flowline diameter from


reservoir hydraulics calculations

all layers (ft or m)


L = Length of pipe section (ft or m)

Calculate flowline wall thickness

If pipe length L is relatively small then Q essentially defines


the local heat transfer rate.
A reasonable estimate of Uo can generally be obtained from
the relationship,

Calculate insulation thickness to


satisfy required U-value

Uo =

Calculate required carrier pipe internal diameter

OD " RF + ! RLayers + RS

(3)

where :
OD = Outermost diameter of the pipe (ft or m)

Calculate carrier wall thickness

RF = Fluid resistance to heat transfer (ft.F.hr/BTU or m.K/W)


Re-Calculate U-value

Project
U-value
met?

Iterative
Loop

!R

= Sum of all layers' resistance to heat transfer

(ft.F.hr/BTU or m.K/W)
R S = Surroundings resistance to heat transfer

No

(ft.F.hr/BTU or m.K/W)
The resistance to heat transfer of the product and
surroundings is not considered in the design calculations
performed in this paper as their contribution to the overall heat
transfer coefficient are small.

Yes
Result of optimised
pipe-in-pipe design

Resistance to Heat Transfer of the Insulation Layer and Pipe


Walls. The resistance to heat transfer of the pipe layers is
calculated using the following equation:

Flow Diagram 2
Explanation of Heat Transfer Theory
The average rate at which heat is lost from the fluid flowing
through a section of pipe, due to steady state heat transfer
between the fluid and the pipe surroundings, is generally
calculated using the following equation:

Q = U o " Ao " (TF ! TS )

Layers

(1)

where :
Q = Average heat transfer rate (BTU/hr or KJ/hr)
U o = Overall Heat Transfer Coefficient based on the
area A o (BTU/hr.ft?.F or W/m?.K)
A0 = Area of heat transfer surface (ft? or m?)
T F = Temperature of fluid flowing in pipe (F or C)
TS = Temperature of pipe surroundings (F or C)
Equation 1 and all subsequent theory are based on onedimensional conduction only. For a particular length of pipe, L,

RLayer =

& ODLayer
1
' ln$
2 ' k $% IDLayer

#
!
!
"

(4)

where :
k = Thermal conductivity of specific layer material
(BTU/hr.ft.F or W/m.K)
OD = Outer diamater of specific layer (ft or m)
ID = Inner diameter of specific layer (ft or m)
Generally the thermal conductivity of insulating materials
increases with increasing temperature and this dependence on
temperature must be included in the evaluation of equation (4).
Calculating the OHTC Referenced to the Inner or Outer
Diameter. As part of the supplied project information the
required OHTC will be specified with reference to either the
inner or outer diameter of the pipe. As a rule of thumb Gulf of

M. HAUSNER, M. DIXON

Mexico projects reference the outer diameter and the rest of the
world reference the OHTC to the inner diameter. Equation 3
calculates the OHTC at the outer diameter of the pipe i.e.:

U0 =

The above equation calculates the OHTC with reference to


the outer diameter of the pipe. To calculate the OHTC
referenced to the inner diameter of the pipe the following
equation is used:
U o ! OD = U i ! ID
(5)

where :
ID = Inside diameter of the pipe (ft or m)
Rearranged, equation 5 provides the OHTC referenced to the
inner diameter:

U o ! OD
ID

(6)

Calculation of Flowline Wall Thickness (Internal Pressure).


The following equation can be used to calculate the flowline
wall thickness to withstand internal pressure (ref.1):

Pb = 0.90 " (# y + # U )"

t
D!t

(7)

where :
! y = Yield strength of pipe (MPa)

! U = Ultimate tensile strength of pipe (MPa)


t = Wall thickness (mm)
D = Outside diameter (mm)

Pb =

Py ! Pe
2

(9c)

where :
PC = Collapse Pressure (MPa)
Py = Plastic collapse pressure (MPa)
Pe = Elastic collapse pressure (MPa)

PC =

Po
0.7

(10)

and

Po = " ! g ! h

(11)

where :
Po = design hydrostatic pressure (MPa)
h = water depth (m)
Parametric Study
Case descriptions. Five cases have been selected to illustrate
the cost benefits that the Inside-out design optimization
methodology offers. It has been assumed that the hydraulic
design calculations indicate the flowline ID must be no less than
8.825 inch. This results in selection of a 10.75 inch OD flowline
from the API range for an unoptimised flowline case. Table 3
provides the main dimensions and components of the five cases
illustrating those with API and non-API flowline and carrier
pipe:

(8)

Calculation of Carrier Pipe Wall Thickness (External


Pressure). The following equations can be used to determine
the carrier pipe wall thickness to withstand external hydrostatic
pressure(ref.1):

Py + Pe

Table 3: Summary of Main Parameters for Cases

Pd
0.72

where :
Pd = Design pressure (MPa)

PC =

(9b)

E = Young' s Modulus
! = Poisson' s ratio

U i = OHTC at inner diameter (BTU/hr.ft?.F or W/m?.K)

Ui =

&t #
Py = 2 ' ( y ' $ !
%D"
(t %
& #
'D$
Pe = 2 ) E )
1 "! 2

1
OD " R F + ! R Layers + R S

OTC 14182

(9a)
2

Case Number
1

Flowline OD (inch)

10.75

Flowline wall thickness (mm)

14.11

Insulation material
Carrier size

10.75

9.84

9.84

10.75

14.11

12.92

12.92

14.11

PUF

PUF

PUF

MP

MP

API

NonAPI

NonAPI

NonAPI

NonAPI

The five cases have combinations of standard API and


optimized pipe sizes for flowline and carrier along with variation
in insulation material.
Case1
Case 1 is considered the base case. It comprises standard API
pipe sizes for both the flowline and carrier pipe with PUF

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

insulation material.
Case 2
Case 2 comprises standard API pipe sizes for the flowline
with an optimized carrier pipe. The insulation material for Case
2 is PUF.
Case 3
Case 3 comprises an optimized flowline and an optimized carrier
pipe with PUF insulation material.
Case 4
Case 4 comprises an optimized flowline and an optimized
carrier pipe with MP insulation material.
Case 5
Case 5 is similar to Case 2 comprising a standard API pipe
size for the flowline and an optimized carrier pipe. The main
difference is that the insulation material used is MP. Table 4
summarises the relevant comparison cases:
Table 4: Comparison of Cases
Comparison
Reference

Compared
Cases

Difference between cases

A
B

1 vs. 2
1 vs. 3

1 vs. 4

D
E
F

1 vs. 5
2 vs. 5
3 vs. 4

Non-API carrier, both PUF


Non-API flowline & carrier, both
PUF
Non-API flowline & carrier, Case
4 MP
Non-API carrier, Case 4 MP
PUF vs. MP, both non-API carrier
PUF vs. MP, both non-API
flowline & carrier

Design Parameters. As well as comparing different levels of


pipe optimization each case was run for all combinations of the
following parameters to demonstrate the influence of each of
these parameters on the system cost:

U-values of 1.0, 1.5 and 2.0 W/m2K


Flowline lengths of 5, 10, 20, 40 & 60 km
Water depths of 1000, 1500, 2000, 2500 & 3000m
2 types of insulation material polyurethane foam
(PUF) and microporous material (MP)

For calculating the flowline wall thickness the internal


design pressure is taken as 300 bar (30 MPa).
Insulation Material Selection. Of the five cases three have
polyurethane foam (PUF) insulation material and the other two
have microporous (MP) insulation material.
The two insulation materials where chosen because their
thermal conductivity and costs are from either end of the
insulation material spectrum.
The PUF system considered is an injected system whereby
the entire annular gap between the flowline outer diameter and

the carrier inner diameter is filled with foam leaving no air gaps.
The microporous insulation (MP) is wrapped onto the
flowline and the system contains a minimum radial gap of 10mm
between the outside of the insulation and the inner diameter of
the carrier pipe. This is to account for tolerances during
insertion of the flowline into the carrier.
Cost Breakdown. A cost model has been developed that
includes the procurement, onshore fabrication and installation of
the pipe-in-pipe systems for all combinations. Some of the key
cost assumptions are shown in Table 5.
Table 5: Summary of Costing Assumptions Employed
Cost Centre
Linepipe
Re-tooling for nonAPI size pipe
Insulation
PUF
MP
J-lay vessel day rate
Host hull cost

Value
1000
100,000

Units
$/tonne
$ per size

150
1700
250,000
3

$/m3
$/m3
$/day
$/kg buoyancy

For the situation where pipe-in-pipe SCRs are to be


employed the cost associated with the top tension required at the
vessel has been estimated from the hull cost per kg of uplift.
The riser lengths for a given water depth are calculated using
simple catenary equations and are shown in Table 6.
Table 6: SCR Length for Respective Water Depth
Water Depth (m)

SCR length (m)

1000

1500

2000

2500

3000

1234

1857

2479

3102

3725

Results of Parametric Study


The data generated from the five cases and the various design
parameters are evaluated in two sections. Firstly the results
relating to a typical West African (WA) development are
considered and secondly all the results are considered and
analyzed to draw conclusions relating to the cost savings
involved with pipe-in-pipe optimization.
Typical West African Development. The West African
development considered here has the following field parameters:

1500m water depth


FPSO host
U-value of 1.5 W/m!K
6 producing wells located in pairs

M. HAUSNER, M. DIXON

Dual pipe-in-pipe tied back to FPSO via SCRs


(pigable loops)
40 km total flowline length

Figure 1 represents the cost comparison of the five cases


showing the breakdown and contribution of the following cost
centres:
Linepipe
Insulation
Re-tooling (if applicable)
Onshore fabrication
Installation
Host hull cost for risers
The onshore fabrication and installation costs for each case
are the same as the variation in pipe diameter, and therefore the
increased welding time, is small in comparison to the overall
time and cost of these operations and is therefore neglected.
The host cost refers to hullform cost due to the risers. The
water depth, and therefore the riser line length, is constant for
this WA development, the cost variation being dependent on the
submerged weight of the riser.
Table 7 presents the total costs for each case (without riser
costs) as well as the percentage saving over the Base Case (Case
1).
Table 7: Total Cost Comparison
Case Number
1

Total Cost ($M)

18.7

17.4

16.3

17.0

18.2

Difference ($M)

1.3

2.5

1.8

0.5

% difference

14

11

The results clearly show that Case 3, with the optimized


flowline and carrier utilising PUF insulation, offers the most cost
effective solution at $16.3M. Case 4, which uses the more
expensive MP insulation material but also has an optimized
flowline and carrier, is the closest solution to this at $17.0M, the
difference being 3% between these two cases.
By comparing the steel costs of these two cases (Figure 2),
there is a greater saving by using the MP insulation (circa 5.3%)
however the increased cost of the insulation material makes the
total onshore, and therefore the overall cost, higher than Case 3
(Figures 3 and 4 respectively).
Case 2 and Case 5, each with the standard flowline and
optimized carrier but with PUF and MP insulation respectively,
represent the third and forth most cost effective systems
respectively. The cost difference between these two cases is in
the region of 3.7%.
Case 1 (Base case) represents the least cost effective option

OTC 14182

with a total cost of circa $18.7M. The overall cost saving


achieved by the full optimization of the system (i.e. optimizing
flowline and carrier) is 14%, a saving of $2.5M.
By only partially optimizing the system (i.e. optimizing the
carrier only) a significant saving can still be achieved. When
compared to Case 1, Cases 2 and 5 offer cost savings of $1.3M
(7%) and $0.5M (3%) respectively.
By referring to Figures 2, 3 and 4 graphical comparisons for
steel cost, onshore cost and total cost can be seen.
Pipe-in-Pipe SCR Costs
The total costs shown in Figure 1 for the host hullform costs
are difficult to rationalize as specific values due to their complex
dependence on host type (FPSO, Semi, TLP or SPAR) and
project requirements for topsides processing, storage etc. They
are presented to give a feel for their magnitude.
Figures 5 and 6 are more useful in that the top tension saving
over the Base Case is clearly identified for the optimized
designs. Case 4 has a reduction in top tension of 325 Te and
Case 3 this value is 316 Te, allowing greater host deck load.
This is particularly important for top tension-sensitive host types
such as TLPs, Semi-submersibles and SPARs.
Discussion of General Trends
The following evaluation of the five cases is for the range of
water depths, U-values and pipe lengths as described previously
in this paper.
Case 1 is the Base Case with API sized flowline and carrier,
utilizing PUF insulation.
Cases 2 and 5 are comparable as they have optimized carrier
pipe and employ PUF and MP insulation respectively.
Case 3 and 4 are comparable as they have optimized flowline
and carrier pipes and employ PUF and MP insulation
respectively.
U-Value Trends
For a U-value of 1.0 W/m!K use of the more expensive MP
insulation (Cases 4 and 5) is consistently the cheapest option.
The cost differential increases with increasing water depth and
line length. Figures 7 and 8 show the total installed cost for U =
1.0 W/m!K with line lengths of 5,000 and 60,000 m
respectively.
As U-value increases there is a switchover and the PUF
insulation costs drive the cheapest solution (Cases 2 and 3). The
cost differential between the PUF and MP solutions decreses
with water depth but increases for greater line length. The
influence of water depth decreases ae U-value increases fro 1.5
to 2.0 W/m!K. Figures 9 and 10 show the total installed cost for
U = 2.0 W/m!K with line lengths of 5,000 and 60,000 m
respectively.
Re-Tooling Costs
The re-tooling cost has greatest influence on the solutions

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

where the water depth is 2000m or less. After this depth the
Base Case requires a jump to the next API size for the carrier.
This reinforces the benefits of optimizing pipe diameter.
Steel Costs
For U = 1.0 W/m2K there is a cross-over from the Base Case
steel cost being cheaper to the carrier optimized cases (Cases 2
and 5) being less sexpensive. This happens at a water depth of
2000m or for flowline lengths greater than 10 km.
For all U-values greater than this steel costs are lower for
Cases 2 and 5.
Water Depth
Generally speaking as water depth increases so the cost
savings achieved from pipe-in-pipe optimized design increase
over the Base Case. Depending on the various parameters
(water depth, U-value, lie length, insulation type) these savings
range from zero to 35% at the extremes.
Conclusions
(1)
(2)

(3)

As U-value decreases to around 1 W/m2K and lower the


more expensive insulation (microporous material) provides
the most economic and the lightest systems
To offset the additonal cost for re-tooling to non-API size
pipe for the carrier only then either of the follwoing
conditions are required:
Flowline length must be greater than 10 km
Water depth must be greater than 2000m
For pipe-in-pipe SCRs top tension savings of over 20% are
achievable through full design optimization.

References
1.

API RP 1111, Design, Construction, Operation and Maintenance


of Offshore Hydrocarbon Pipeline (Limit State Design), 3rd Edn.,
July 1999

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

PIP
Classification

Pros

Cons

Sliding

Butt weld on carrier pipe


Greater fatigue life for riser applications
Small reduction in thermal efficiency between main body & field joint
Compatible with all installation methods
Compatible with sprayed PUF, microporous blanket & aerogel insulation
Possible use of mechanical connectors on carrier pipe instead of welding
Suitable for single, double or quad joints

Requirement for pup pieces


Need for some flexibility in welding position (S-lay only)
Intermittent or no axial connections between flow line and carrier
Requires spacers
Not compatible with granular material, vacuums or phase change
material insulation
Offshore sequencing is highly installation vessel dependent

Fixed

Regular axial constraint o flow line and carrier


Compatible with all types of insulation
Exceptional insulation capability can be achieved
Flow line only weld at field joint in conjunction with a wet field joint
Good track record
Almost entirely independent of installation vessel setup
Most suitable for double joints

Requires either welding to outer surface of flow line (swages and


tulips) or additional forged components (bulkheads & tulips)
Cold spots where flow line and carrier are welded together
Localized cool down time driven by cold spots
Cannot be reeled
Good insulation required at field joint to counter cold spots

Restrained

Compatible with most installation methods


Good track record
Relatively independent of installation vessel setup
Most suitable for double joints

Not compatible with vacuum or phase change material insulation


Cannot be reeled
Limited axial compliance between flow line and carrier
Requires spacers and stoppers such as rubber bulkheads

Table 2: Advantages and disadvantages of the structural classifications

Figure 1: Cost Comparison for Typical WA development


25000

20000

15000
Host hullform

Cost ($ k)

Installation
Onshore Fab
Re-tooling
Insulation
Linepipe
10000

5000

0
Case 1

Case 2

Case 3

Case 4

Case 5

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

Figure 2: Steel Cost vs Water depth


(U = 1.5 W/m_K, Pipe length = 40000m)
14000
13500
13000
12500
12000

Steel Cost ($ k)

11500
11000

Case 1

10500

Case 2
Case 3

10000

Case 4

9500

Case 5

9000
8500
8000
7500
7000
6500
1000

1500

2000

2500

3000

Water Depth (m)

Figure 3: Onshore Cost vs Water depth


(U = 1.5 W/m_K, Pipe length = 40000m)
16500

15500

Onshore Cost ($ k)

14500

13500

Case 1
Case 2

12500

Case 3
Case 4
Case 5

11500

10500

9500

8500
1000

1500

2000

2500

3000

Water Depth (m)

Figure 4: Total Cost vs Water depth


(U = 1.5 W/m_K, Pipe length = 40000m)
28000
27000
26000
25000

Total Cost ($ k)

24000
23000

Case 1

22000

Case 2
Case 3

21000

Case 4

20000

Case 5

19000
18000
17000
16000
15000
1000

1500

2000

Water Depth (m)

2500

3000

10

M. HAUSNER, M. DIXON

OTC 14182

Figure 5: Top Tension for 6 Risers


(U=1.5 Wm 2K, water depth 1500m)
1600
1400

Top Tension (Te)

1200
1000
800
600
400
Diference to Case 1

200

Total Top Tension

0
Case 1

Case 2

Case 3

Case 4

Case 5

Figure 6: Top Tension Percentage Comparision for 6 Risers


(U=1.5 Wm 2K, water depth 1500m)
100%

Top Tension (% of Case 1)

80%

60%

40%

Diference to Case 1

20%

Total Top Tension


0%
Case 1

Case 2

Case 3

Case 4

Case 5

Figure 7: Total Cost vs Water Depth


(U = 1.0 W/m_K, Pipe length = 5,000m)
7000
6500
6000

Total Cost ($ k)

5500
5000
4500
4000

Case 1
3500

Case 2
Case 3

3000

Case 4
Case 5

2500
2000
1000

1500

2000

Water Depth (m)

2500

3000

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

11

Figure 8: Total Cost vs Water Depth


(U = 1.0 W/m_K, Pipe length = 60,000m)
45000
43000
41000
39000

Total Cost ($ k)

37000
35000
33000
31000

Case 1

29000

Case 2
Case 3

27000

Case 4
25000

Case 5

23000
1000

1500

2000

2500

3000

Water Depth (m)

Figure 9: Total Cost vs Water Depth


(U = 2.0 W/m_K, Pipe length = 5,000m)

6000

5500

5000

Total Cost ($ k)

4500

4000

3500

Case 1
Case 2

3000

Case 3
Case 4

2500

Case 5

2000
1000

1500

2000

2500

3000

Water Depth (m)

Figure 10: Total Cost vs Water Depth


(U = 2.0 W/m_K, Pipe length = 60,000m)
41000
39000
37000

Total Cost ($ k)

35000
33000
31000
29000

Case 1

27000

Case 2
25000

Case 3
Case 4

23000

Case 5

21000
1000

1500

2000

Water Depth (m)

2500

3000

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

Appendix 1 Definitions for Pipe-in-Pipe


Classification
Structural Compliance Definitions
Sliding
The sliding configuration is so named for the capability of
the carrier pipe to slide over the flow line and insulation such
that consecutive joints of carrier can be butt-welded.
Sliding systems will usually have spacers for alignment and
centralisation of the two pipes but the system can work without
spacers.
An additional requirement is a temporary clamp (or similar
device) to axially hold the flowline and carrier during
transportation, handling and loading into the firing line. This
device is removed prior to welding and may be located at either
end of the joint depending on the installation vessel setup.
Typically the field jointing sequence is as follows:

Weld and inspect flow line

Corrosion coat flowline weld region

Insert half shell insulation (or equivalent) this is


often the same as the main body insulation

Position weld backing strip

Slide carrier over field joint and perform butt weld

Inspect and coat carrier weld


Suitable vessels for sliding systems include J-lay vessels,
such as the Saipem FDS, and S-lay vessels, e.g. the Allseas
Solitaire, but adaptation of other vessels is possible. Fabrication
of the pipe-in-pipe strings for reeling can be preformed using the
sliding approach.
The sliding configuration does not have any regular axial
connection between flow line and carrier but can easily
accommodate bulkheads to lock the pipes together axially and
to provide compartmentalisation of insulation against carrier
breach.
Fixed
The fixed type of pipe-in-pipe systems use either swaged
connectors, forged bulkheads or forged tulips that are welded to
both the carrier and the flowline. This fixes the flow line and
carrier both axially and laterally at the end of each joint length,
hence the term Fixed system.
The swaged connectors and forged tulips are initially welded
to the carrier pipe and then welded to the outer surface of the
flowline. This is done at both ends of the joint. With a forged
bulkhead arrangement the following sequence is followed for
onshore fabrication of double, treble or quad joints:
1. First bulkhead butt welded to the flow line
2. Flowline inserted into the carrier pipe (non-bulkhead
end first)
3. Carrier butt welded to the bulkhead upstand
4. Second bulkhead butt welded to the end of the flowline

12

protruding from carrier


Insulation placed over the flowline weld region
Gap between carrier and upstand on the second bulkhead
is closed with steel half shells
Offshore construction involves either:

Butt welding of flowline ends together and then


application of insulation on the outside of the joint, i.e.
insulation is exposed to the seawater

Steps 4 to 6 (as described above) are repeated


This structural category of pipe-in-pipe system can be
installed by almost any S-lay and J-lay vessel but cannot be
reeled due to the structural discontinuity at the field joint
location.
One major advantage of this type of construction is the
automatic compartmentalization of the system every double,
treble or quad joint. The primary drawback is the additional heat
loss throught the steel connection between the flowline and the
carrier
5.
6.

Restrained
The restrained structural category of pipe-in-pipe systems
consists of the flowline concentrically located inside the carrier
by spacers in the main body of the joint and by some form of
non-metallic bulkhead at either end, typically of rubber/EDPM
material. The term Restraied is used as the bulkheads provide
an amount of axial and lateral compliance between the flowline
and carrier, prinaily during installation, but differenctial axial
movement of the flowline and carrier may occur during
operation.
The insulation material is either pre-attached to the flowline
prior to insertion into the carrier or the annulus is filled once the
flow line has been inserted.
The purpose of the bulkheads is as follows:
1. To prevent relative axial movement of the flowline
and carrier during fabrication, transportation and
handling
2. To contain the insulation when filling the assembled
joint (flowline already inserted in carrier)
3. Concentric alignment of the flowline in the carrier
The field joint arrangement offshore usually utilises steel
half shells to close the gap between consecutive joints of carrier
pipe, requiring two circumferential and two longitudinal welds.
The field joint insulation may be of foam half shells or
rockwool.
Installation of restrined pipe-in-pipe systems is relatively
independent of S-lay and J-lay vessel setup and current wok is
ongoing to prove this type of system for reeling.
Insulation Material Categories
Polyurethane Foam (PUF)
A common and cheap material that can be sprayed onto the
flowline prior to insertion in the carrier pipe or injected into the

OTC 14182

OPTIMIZED DESIGN OF PIPE-IN-PIPE SYSTEMS

annular space between the pre-assembled flowline and carrier


pipes. These two methods of insulation are distinctly different
and determine the type of structural system that can be used.
For example the injected version cannot be used for sliding
PIP systems as the PUF bonds to both carrier and flowline,
thereby eliminating the axial movement required to achieve butt
welding of the carrier pipe. Sprayed PUF, however, permits the
sliding option.
Thermal conductivity of PUF material is in the range of 0.03
to 0.04 W/m2K depending on cell size and foam blowing agent.
Granular Materials (e.g. microspheres)
This category consists of granular material that is poured into
the annular space between the flowline and carrier pipe. The
granules are usually alumina-silicate microspheres, also known
as fly ash, which is a waste product from coal-fired power
stations. The microspheres range in diameter from 10 to 150
"m, are almost perfectly spherical and are completely inert.
Thermal conductivity of microspheres is between 0.09 and
0.11 W/m2K.
The pipe assembly can consist of single or, more usually,
double joints and is normally inclined to a specific angle and
vibrated at a specific frequency to ensure compaction and
optimum filling. In order to fill the annulus a stopper is
required at the bottom end of the PIP section being filled to
contain the granular material and to hold the flow line and
carrier pipe concentric and axially aligned during filling.
Microporous Materials (MP)
Microporous material is formed from spherical particles of
fumed silica that are bonded together at their point of contact
with one another, minimizing the heat conduction through the
solid. The interstitial voids between the particles trap air
molecules and prevent heat transmission through convection.
The panels are prepared for application to the pipe with parallel
saw cuts and then sealed in either a polyethylene-polyamide film
or aluminium foil packet. This packet/film may be filler with
air, inert gas (e.g. argon) or a drawn vacuum. The insulation
needs to remain dry.
Thermal conductivity of microporous materials range from
0.025 to 0.015 W/m2K for air-filled systems, 0.015 to 0.01
W/m2K for gas-filled and down to 0.006 W/m2K for a vacuum
system.
Vacuum (full or assisted)
A perfect vacuum provides the best insulation possible but
creating and maintaining a near-perfect vacuum is difficult as
diffusion of gases through steel (mainly hydrogen) create partial
pressures. These significantly affect the overall heat transfer
coefficient and require inclusion of special getters, types of
material that absorb the diffused gases. Getters are typically
granules or tablets and are added during the fabrication process.

13

The vacuum insulated method leveraged on the vacuum


insulated tubing technology, which has been used extensively
since the early 1980s.
Phase Change Materials
This is a new class of material under consideration for use in
subsea pipelines. Essentially the insulation material stores heat
that is released during shutdown as the material crystallizes.
This technology is currently being investigated for use in other
areas of the industry.