th

Proceedings of the 5 International Offshore Pipeline Forum

IOPF 2010
October 20-21, 2010, Houston, Texas, USA

IOPF2010-4004

INNOVATIONS IN TESTING OF DEEP WATER INSULATION

Marcus Heydrich Bo Xu
ShawCor ShawCor
Toronto, Ontario, Canada Toronto, Ontario, Canada

Raphael Moscarello Sanjay Shah Peter Jackson Stephen Edmondson

Bredero Shaw ShawCor ShawCor ShawCor
Toronto, Ontario, Canada Toronto, Ontario, Toronto, Ontario, Toronto, Ontario,
Canada Canada Canada

ABSTRACT surrounding these systems while controlling the temperature

Insulation is commonly applied to offshore pipelines to inside the pipe. By measuring the heat flow, thermal
ensure the flow of hydrocarbons at elevated temperatures. The
thermal properties of the insulation can be readily modeled;
however, the performance of the insulation needs to be verified
under conditions similar to those encountered in deepwater
service. Autoclave testing of individual materials can be
conducted but this is not representative of the conditions that
the materials see in service. The insulation is typically present
in layers, and not every layer is exposed to the same
environment. Simulated service testing, where a full size pipe is
exposed to a water pressure equivalent to that it will see in
deepwater service load, is typically used to verify the
performance of the insulation.

In this paper, recent developments in simulated service

testing, including the design of a more advanced Simulated
Service Vessel (SSV) will be presented. The verification and
prediction of the performance of insulation systems will also be
described. conductivity and compressive creep of the insulating material,
both the thermal efficiency and depth rating capabilities of the
INTRODUCTION insulation can be confirmed. This data will also be used to
During the development and qualification of insulation verify design of the system.
systems for subsea oil and gas pipelines it is important to
understand and quantify the behavior of the insulation under
service conditions experienced in subsea environments.
ShawCor’s Simulated Service Vessel (SSV) is a key part of a
new state of the art facility designed to do just this.

This pressure vessel will simulate service conditions

experienced by insulated subsea systems such as linepipe, field
joints and, flexible pipe by applying pressure to water

1 Copyright © 2010 by ASME

NOMENCLATURE NEW SHAWCOR SSV
α – Thermal expansion coefficient The new ShawCor SSV allows for an accurate assessment
cp - Heat capacity of the performance capability of an insulated pipe system under
E – Young’s modulus precisely controlled conditions of temperature and pressure.
h - Convection heat transfer coefficient The SSV is fundamentally a large cylindrical autoclave which
k – Thermal conductivity can accommodate a 6 meter (20’) length of the insulated pipe or
ε - Strain structure to be evaluated.
ν - Poisson's Ratio
σ – Stress A thermally insulated sample is mounted in the vessel
r - Radius which is instrumented with the following: displacement
ri – Inner radius transducers to measure diametrical change under hydrostatic
ro – Outer radius load; thermal sensors for temperature measurement; and, heat
R - Radius flux sensors to determine the heat loss from the system. Once
t – time thermal equilibrium has been established, the vessel is
T – Temperature pressurized and held at the required pressure for the duration of
Top – Operation temperature the test. The pressure is increased in specific increments to
Tcondition – Temperature condition assess the immediate and long term response of insulation
u – Radial displacement properties to changes in pressure at a specific operating
U – U value temperature which is set by internally heating the test pipe.
w – Axial displacement
r, θ, z - coordinates Changes in compressive creep and heat flow are measured
over time. U-value of the system can be calculated and
BACKGROUND compared with the design assumptions for the given pressure
From both performance and safety aspects, the need for a and temperature field.
precise understanding and determination of the service
capability of insulated oil and gas subsea pipelines is essential. INSTRUMENTATION SYSTEM
The increasingly high pressure and temperature performance To obtain the desired measurements, the measurement and
demanded by the industry due to deeper drilling and control system in the new SSV facility are built to have features
deployment makes it ever more important that the performance as shown in the following schematic. The steel pipe is heated
limitations of insulated subsea pipelines are fully known. with a seven-zone electrical heater, controlled by thermocouples
in each zone. The following parameters are monitored:
EXISTING TECHNOLOGY  water temperature
The existing ShawCor SSV installed in Calgary, Canada
was designed based on similar systems used in the late 1990s.  temperature of the insulation coating surface
This oil heated system utilizes a closed loop internal cooling  heat flux in the radial direction at different zones
coil to control outside water temperature. The system uses two  radial displacement using LVDT sensors
heat flux (HF) belts circumferentially mounted to measure heat
 power consumption in each zone
flow through the insulation system with each belt containing
three HF sensors and four temperature sensors at the surface of
the insulation. Six Linear Variable Differential Transformer
(LVDTs) are strategically located throughout the vessel to
measure compressive creep during the simulated service test.

The current SSV is limited to a pipe length of 1.83 m (6’),

with outside pipe diameter including insulation of 356 mm
(14”) and minimum inside diameter of 125 mm (4.9”). The test
pipe can be subjected to hydrostatic pressure of 280 bar with a
maximum internal temperature of 160°C (320°F). Due to the
increasing demand for higher temperature, large diameter
offshore pipelines, ShawCor has decided to develop a larger,
state of the art, 80 tonne SSV, which will be installed in
Toronto, Canada. The existing Calgary vessel will then be used
for long term material characterizations.

2 Copyright © 2010 by ASME

 The system continuously monitors and records the As well, the SSV has the capability to compare the performance
measurements during the test period. The multi-zone controls of three insulation coatings simultaneously, with multiple
and multi-point measurements of this system provide integrity measurements for each coating.
testing of the insulation coating performance under the defined
testing conditions. Furthermore, the SSV can test for cool down time and
provide accurate, real time measurements of creep to evaluate
UNIQUE TESTING CAPABILITIES long term reliability of the insulation. The heating system
To accommodate market requirements trending towards provides precise temperature control to ensure accuracy and a
deep and ultra deep development opportunities and higher oil high capacity chiller provides a water temperature of 4 – 6°C
transportation temperatures, the new ShawCor SSV has been (40 - 43°F).
designed to test to an equivalent water depth up to 3,000 m
(9,842’) at an internal pipe temperature of up to 180°C (356°F). The testing schedule can be adjusted to accommodate
customer requirements. Qualified technical staff will be on-site
and third party inspectors will validate test results for clients.
Capability/Property Specification
Minimum Test Pressure 25 bar (± 2) DATA FROM SSV TESTING
Maximum Test Pressure 300 bar (± 5) The figures below illustrate typical responses of an
Chilled Water Temperature 4°C (40°F) (± 2) insulated pipe system during simulated service testing. The
Inside Pressure Vessel system included a 273 mm (10”) OD steel pipe of wall
Internal Temperature 20°C – 180°C (68°F – 356°F) thickness 16 mm (0.63”) coated with 50 mm (2”) of ShawCor’s
new Thermotite® ULTRATM insulation coating.
Sample Length 6 m (18’) max
Vessel ID 1.2m (48’") 300 3
Number of Pipe Samples one pipe Pressure Creep

Pipe Inside Diameter 95 mm - 660 mm (4” - 26”) 250 2.5

Pipe Outside Diameter 145 mm - 810 mm (6” - 32”)

(includes insulation) 200 2
Pressure (bar)

Coating Thermal 0.1- 0.3 W/m K

Creep (%)
150 1.5
Conductivity (0.06 – 0.17 BTU / ft hr F)
Overall Heat Transfer 1.5 - 6 W/m2 K
100 1
Coefficient (U) (0.3 – 1.1 BTU / ft2 hr F)
50 0.5

0 0

11 15 19 23 27 /1 /5 /9 /1
3
/1
7
/2
1
/2
5
/2
9 /2 /6 /1
0
9/ 9/ 9/ 9/ 9/ 10 10 10 10 10 10 10 10 11 11 11
Time

300
Pressure k-factor
0.4
250

200 0.3

k-factor (W/mK)
Pressure (bar)

150
0.2

100

0.1
50

Testing is designed to accommodate the exact dimensions

of the pipe (up to 26" ID) that will be put into service so that 0 0

clients will accurately know the true representative thermal 9/

11
9/
15
9/
19
9/
23
9/
27
10
/1
10
/5
10
/9 /1
3
/1
7
/2
1
/2
5
/2
9
11
/2
11
/6 /1
0
10 10 10 10 10 11
performance of their installed pipe. The vessel also has the Time

ability to test both the field joint and the main line insulation.

3 Copyright © 2010 by ASME

 175


Temperature k-factor
0.4 T
150
 c p   k 2 T
 t

T (0, r )  T0 ( ri  r  ro )
125
0.3

Temperature (oC)

k-factor (W/mK)
T
T r  r  Top ,
100
 h (T  Tcondition ) / k
0.2  i
r r  ro
75 (2)
50
0.1
The strain and stress distribution due to hydraulic pressure
25
and the thermal stress when the pipe is tested can be simulated
using the governing equations:
0 0

11 15 19 23 27
  r  rz  r
/1 /5 /9 13 17 21 25 29 /2 /6 10
9/ 9/ 9/ 9/ 9/ 10 10 10 10
/
10
/
10
/
10
/
10
/ 11 11 11
/

Time
 r  z  r  0

  rz   z   rz  0
The compressive creep curves generated over time are
important for computer modeling and thermal design of solid,  r z r
foamed and syntactic insulation systems for various water 
depths and temperatures. The increase in pressure (blue line) is   u ,   u ,   w ,   u  w
planned in steps to show the elastic (immediate) and the  r r  r z z rz z r
inelastic (long term) responses of the system to the pressure  1
 i  ( i   j   k )  T ,
changes. This allows for construction of the design model. The  E (3)
typical response for foam is that the material compresses over
time. For a solid polymer, the material shows an immediate
Creep is mechanical deformation resulting from the
response that reaches a plateau after about one week.
composite effects of time, temperature and mechanical load.
Initial performance and creep response are significantly affected
EVALUATING CREEP AND COMPRESSION
by initial compositional properties such as crystallinity in
SSV testing is performed to evaluate the long term creep
crystalline polymer and foam structure where applicable. These
properties of thermal insulation systems and to estimate the
properties are heavily dependant on initial processing
service life of the coating. The initial performance is defined by
conditions.
the thickness of the material, the compressive modulus and the
thermal conductivity. These values are readily determined in
Creep can be measured for homogenous materials in a
laboratory tests, and the initial U Value is calculated for any
triaxial creep test, which determines the specific material
given geometry and material composition, as follows:
dependant deformation as a function of time, temperature and
pressure; however, these samples are not representative of
1
U n
insulation comprising multiple materials and layers.
ln( Ri 1 / Ri )
A R  
Measurements are not simply additive since their relative
thermal and mechanical properties are interdependent.
i 1 ki (1)
The creep characteristics are determined by the creep test
Under operating conditions, the coating is placed under a and described as:
thermo/mechanical load which alters the coating’s thickness,
resistance to compression, and therefore its thermal  c  A m t n (4)
conductivity. For thick coatings (>25.4 mm or 1”), there is a
substantial temperature gradient through the coating thickness
(hottest on the inside and coldest on the outside) which The above relationship will be further verified by the SSV
produces a corresponding gradient of compressive strengths and test and applied to the FEA modeling.
thermal conductivities. These are best modeled using laboratory
data for compressive strength and thermal conductivity as a PREDICTING PERFORMANCE
function of temperature, and resolved with Finite Element Creep performance of a full scale, insulated pipe cannot be
Analysis (FEA) techniques. accurately predicted from FEA models alone. The best estimate
of creep behaviour is the result of comparing FEA models with
In FEA modeling, the temperature profile in the system is the actual performance of a full scale installed pipe from an
described using the following heat transfer equations: SSV test. The difference between the FEA model and SSV

4 Copyright © 2010 by ASME

estimates provides a measure of uncertainty (variance), which In the short term, the combination of SSV test data,
can then be used to generate predictive confidence limits around laboratory material testing and the analytical methodology
the service life of the pipe. described above will allow ShawCor to establish the most
reliable predictors of long term performance available in the
From laboratory test data, an FEA model can be generated market. This is particularly important for finding novel material
that predicts thermal and mechanical properties of the insulation and design solutions for challenging deep sea environments,
based on coating thickness. Triaxial creep tests can be used to and demanding performance validations from our customers.
predict changes in thickness as a function of time, temperature
and pressure. The combination of the triaxial data with the FEA The long term objective is to discover and exploit synergies
model will then generate a second FEA model which can between material and design properties which will significantly
predict long term changes in thermal and mechanical properties improve the performance of subsea products.
due to creep. Confidence limits around these predictions are
caused by variances in:

 laboratory data
 triaxial creep tests
 SSV test data

The FEA model will then be compared to the SSV test data
collected on commercial size insulated pipe. Consideration of
the pipe to pipe variance will be established using process
capability studies and quality control (QC) data gathered during
product development. This pipe to pipe variance, combined
with the deviations between FEA predicted results and SSV test
data will establish confidence intervals around the performance
of all pipes manufactured under the specified production
conditions.

The determination of confidence intervals should lead to a ACKNOWLEDGMENTS

more accurate prediction of the long term performance of the The authors would like to thank the management of
coating. Furthermore, this methodology should yield far more ShawCor for permission to publish this paper.
reliable predictions for untested conditions than predictions
using only an FEA model because estimates of uncertainty REFERENCES
establish confidence thresholds which a mathematical model 1. J. Lienhard IV and J. Lienhard V, A Heat Transfer Text
alone cannot. Book, 3rd Ed., Cambridge, MA, Phlogiston Press, 2006
2. Benjamin T.A. Chang, Han Jiang, Hung-jue Sue,
ShawCor is uniquely positioned to apply this analytical Dennis Wong, Al Kehr, Meghan Mallozzi,
process with the commissioning of a new SSV facility. The Disbondment Mechanism of 3LPE Pipeline Coatings,
precise measurement capabilities and capacity of the vessel 17th International Conference on Pipeline Protection,
allow detailed characterization of initial performance and Edinburgh, UK: 17-19 October 2007
promise accurate real time measures of continuous creep and 3. J.Betten, Creep Mechanics, 2nd Edition, Garman,
thermal conductivity available for full scale testing. In addition, Springer, 2004
the close proximity of the SSV facility to ShawCor’s research
and development laboratories provides direct access to
laboratory testing according to specific requirements. Finally,
access to global production records and QC data permits much
broader interpretation of results, especially as SSV data is
collected over time.

5 Copyright © 2010 by ASME