SPE-174416-MS

Use of Flow Control Devices (FCDs) to Enforce Conformance in Steam

Assisted Gravity Drainage (SAGD) Completions
G. P. Vachon, ConocoPhillips; W. Klaczek, P. J. Erickson, D. C. Langer, D. Booy, and A. Baugh, C-FER
Technologies

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Canada Heavy Oil Technical Conference held in Calgary, Alberta, Canada, 9 –11 June 2015.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers 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 SPE copyright.

Abstract
Flow Control Devices (FCDs) have demonstrated significant potential for improving recovery in Steam
Assisted Gravity Drainage (SAGD) production wells. One initial hypothesis was that steam breakthrough
was delayed because the FCDs better homogenized injection and production by equalizing flow and
compensating for pressure changes along the wellbore. However, in many cases, the field results were far
greater than such an approach would have justified. The actual physics for this process are unclear, and
not demonstrated in literature. Upon review of field data published by ConocoPhillips, the possibility of
a steam blocking effect was proposed (Stalder, 2012), although the physical basis for this effect was not
explored. This paper proposes an updated hypothesis to explain this effect, presents preliminary data to
support the assumption, and introduces a new apparatus and methodology to characterize FCDs for SAGD
applications.
The traditional approach to steam control states that steam flashing at the producer should be avoided,
as it will eventually lead to a completion failure. Alternatively, the proposed hypothesis contemplates
using steam flashing at the producer to regulate flow in various segments of the completion, thus better
enforcing conformance. The physics of this process will primarily be described analytically; however, this
effect was also observed qualitatively in a small-scale experiment where water was flashed across an
orifice.
In order to design SAGD completions that leverage FCDs (and this effect), it was necessary to
accurately characterize different FCDs under these challenging multiphase flow conditions. Since vendors
use a variety of approaches when designing their FCDs, a protocol was developed to create a character-
ization procedure which was independent of the underlying FCD design and architecture, resulting in a
direct comparison of the overall performance of each FCD. Part of this protocol required the construction
of a new, high temperature multiphase flow loop capable of subjecting FCDs to representative SAGD
operating conditions. Through fine control of the relevant test parameters, accurate performance mea-
surements can be obtained for each FCD. This paper will present some information regarding the design
and specifications of this new flow loop, as well as impart some of the lessons learned from its
commissioning and initial operation.
2 SPE-174416-MS

Background Information
This section provides a brief explanation of the motivation for investigating the use of FCDs in SAGD
applications, where the use of this technology is not well understood.
SAGD Production
The SAGD process (Butler, 1991) has been extensively used for heavy oil production, and detailed
descriptions are abundant in the literature. As the process has continued to evolve, several significant
operational issues have been investigated and modifications have been developed to improve its econom-
ics and efficiency. Despite these advances, the control of steam placement and reservoir conformance has
continued to be an issue that requires further advancement, as the process economics are heavily driven
by the cost of generating steam and effectively reaching as much of the oil in the reservoir as possible
(Banerjee et al., 2013a).
Many methods have been developed in an attempt to increase conformance, such as the installation of
tubing strings in the injection or production wells, use of PID controllers, and the use of FCDs on the
injector and/or producer wells (Stone et al., 2013). Some of the wells in the ConocoPhillips Surmont Field
use tubing strings in both the producer and injector wells to provide improved steam distribution control
in the reservoir and maximize the efficiency of the well pairs. However, the use of these additional tubing
strings results in larger wellbore requirements and increased capital costs when compared to conventional
SAGD wells (Stalder, 2012), while still providing only limited control of the flow distribution. FCDs
installed in both the injection and production wells have the potential to simplify these well completions
while providing better control of conformance in the reservoir, thus improving overall production
efficiency.
Operating Principles of FCDs
FCDs are throttling devices used to control the inflow (or outflow) through various sections of horizontal
wells, typically in an attempt to improve the flow uniformity across the well. While this paper focuses
primarily on Inflow Control Devices (ICDs) which are used to control flow into production wells, the
general operational theories and methodologies presented can likely be expanded to other forms of FCDs,
such as those used in injection wells.
FCDs are installed to improve production through a combination of: avoiding non-uniformities in the
production well caused by frictional pressure drops which lead to preferential production from the heel
(especially for long horizontals or laterals), altering the flow distribution along the well to avoid excess
production near gas or water zones to prevent coning or early breakthrough, and throttling the flow of gas
(or water) that may enter the well due to breakthrough in order to allow for continued oil production
(Ratterman et al., 2005; Birchenko et al. 2010; Garcia et al., 2009). The flow control provided by FCDs
can be used to benefit oil production, as well as improve the effectiveness and longevity of sand control
devices by minimizing localized sand failures caused by rapid drawdown which leads to increased
plugging or erosion of the sand control elements (McIntyre et al., 2006).
FCDs perform the functions outlined above by modifying the pressure drop across the well inlet
sections at various locations in the horizontal (or in the lateral) sections. There are typically three primary
mechanisms which FCDs use to control this flow (Garcia et al., 2009):
● Flow Restrictions, which use changes in flow area to create instantaneous pressure drops in the
device (typically through an orifice or a nozzle);
● Frictional Methods, which direct flow through helical channels or long tortuous pathways; and
● Hybrid Designs, which use a combination of these restriction and frictional pressure drop methods.

While some work has been presented to characterize the behavior of different styles of FCDs operating
under representative field conditions (Garcia et al., 2009; Least et al., 2012, Lauritzen and Martiniussen,
2011; Jones et al., 2009, Lee et al., 2013), few direct comparisons of these devices operating under
SPE-174416-MS 3

controlled conditions are available. In addition, no information was found that considered the effects of
operating with fluids at or near saturation conditions, which would be expected during steam breakthrough
and in SAGD applications. Further testing and evaluation of commercially available FCDs is required to
allow for proper specification of these devices in SAGD wells.

Previous Use of FCDs in SAGD

FCDs have become commonplace in conventional (non-thermal) applications over the past several years,
and have been found to provide an effective means of well control. Several examples are available which
show that FCDs have been effectively used to control the flow distribution in conventional wells and
enhance well life and performance (McIntyre et al., 2006, Lyngra et al., 2007, Lorenz et al., 2006,
Henriksen et al., 2006, Vela et al., 2011, Regulacion et al., 2013).
The effectiveness of FCDs in these conventional applications has led to more recent interest in the use
of these devices in SAGD, where steam breakthrough may present some of the same issues as water or
gas breakthrough in non-thermal applications. Banerjee et al. (2013a) provides a summary of the
anticipated effects of FCDs on SAGD production, emphasizing the potential improvements to steam
chamber development, steam-trap control, Steam-to-Oil Ratio (SOR), and recovery factors. Simulations
(Stone et al., 2013) showed that FCDs provided improved pressure stability and control, allowed for
improved conformance in the reservoir, and enabled more consistent production across the wellbore.
Kyanpour and Chen (2013) provided a workflow for designing FCD completions for SAGD operations
using reservoir and wellbore flow modeling, and their results showed that properly designed FCD wells
could significantly benefit SAGD production. Banerjee et al. (2013b) showed that FCDs are useful tools
that can help to improve portions of the SAGD process (such as steam-trap control, and the SOR);
however, reservoir geology and heterogeneity will continue to play a significant role in steam confor-
mance and thus the overall production in SAGD wells. This emphasizes the need for coupled wellbore/
reservoir models for the design of SAGD wells with FCDs.
While minimal field data regarding FCDs in SAGD applications is available, Stalder (2012) observed
that FCDs allowed for simplified well completions (removing the requirement for a toe tubing string) and
resulted in improved steam chamber development and overall performance. While Stalder emphasized
that further operational experience is required, these initial field observations show the potential for FCDs
to provide significant improvements to SAGD production. Due to this limited operational experience,
further investigation of FCD behavior and a better understanding of how FCDs will affect reservoir
performance are still required to allow for their effective use in SAGD wells.

Updated Hypothesis on the Use of FCDs in SAGD

This section explains the basis for an updated hypothesis on the operation of FCDs under conditions that
could be expected in a SAGD well, specifically where steam flashing can be anticipated across an FCD.
This updated hypothesis builds on the observed field results and on an analytical examination of steam
formation within an FCD that was later confirmed with various laboratory tests.

Field Experience
Although little information is available concerning the performance (or potential benefits) of FCDs under
SAGD representative conditions, ConocoPhillips previously released information from their Surmont
Asset about the optimization of SAGD well pair 102-06 which was completed with FCDs (Stalder, 2012).
As part of this past field evaluation, a 4-D seismic interpretation of the pad and several of the wells was
released, as shown in Fig. 1.
4 SPE-174416-MS

Figure 1—4-D Seismic Interpretation of Steam Affected Regions for Pad 102N (Stalder, 2012)

In Fig. 1, it is evident that the steam chamber showed improved conformance for the P06 well pair after
comparing it to both the P04 or the P05 well pairs which did not contain FCDs (after allowing for
approximately the same amount of time following the initial start-up). The performance of the P06 well
pair was ultimately better than expected and it was noted that the key benefits of the FCDs included a
better SOR, Water-to-Oil Ratio (WOR), and improved ultimate oil recovery. The combination of premium
sand screens and FCDs should also allow the subcool to be reduced to approximately zero without sand
production and loss of the well, as the added pressure drop (⌬P) through the FCDs should better distribute
the flow along the liner and reduce the chance of steam breakthrough.

Analytical Examination of FCD Behavior

An FCD can fundamentally be considered as a throttling device that imparts a pressure loss on the fluid
stream entering the production well. In a SAGD environment, the throttling process is likely adiabatic and
if the fluid begins at zero quality, the ⌬P across the FCD may result in a phase-change caused by the
formation of steam, depending on how close the pressure of the fluid stream was to the pressure saturation
curve (i.e. the level of subcool). A graphical representation of this behavior is shown below in Fig. 2.
SPE-174416-MS 5

Figure 2—Graphical Representation of the Updated Steam Blocking Hypothesis

As shown in Fig. 2, the formation of steam should ultimately result in an increased ⌬P across an FCD.
This is likely due to the fact that as steam forms, the volume of the fluid increases and thus the volumetric
flowrate of the fluid mixture also increases (as the flow area remains the same). This increase in
volumetric flowrate drives the increase in ⌬P across the FCD. Likewise, the density of the mixture will
decrease with steam formation. However, basic fluid mechanics indicates that the ⌬P generated across the
FCD should be related to viscosity and density but more strongly affected by an increase in the flow
velocity. This ultimately means that a relatively small change in the fluid composition (due to steam
flashing in the FCD) may significantly increase the ⌬P across the FCD.

Qualitative Testing: Confirmation of the Hypothesis

Different vendors have suggested various relationships to quantify the ⌬P that can be expected for their
devices, where these relationships are generally dependent on parameters such as: the FCD architecture
(e.g. orifice, tortuous path, etc.), the internal geometry and size, and/or the internal resistance, which is
sometimes referred to as a Flow Resistance Rating (FRR). A common method of characterizing FCD
performance is to provide the ⌬P in terms of a dimensionless pressure loss coefficient (e.g. K) versus
Reynolds Number (Re) of the inlet fluid.
If the proposed hypothesis holds true, it is expected that as steam begins to flash across the FCD (in
any amount) an increased ⌬P (or K) would be observed while operating at approximately the same Re at
the inlet of the FCD. This means that under SAGD representative conditions the ⌬P through the FCD
should diverge from the published vendor correlations, which are typically generated under either
single-phase or multiphase flow conditions using liquid and/or gas but do not consider a phase change (as
would be expected in this case). An example of this behavior is shown in Fig. 3.
6 SPE-174416-MS

Figure 3—Example of How Steam Flashing may Affect the ⌬P Across an FCD

Prior to developing a methodology for characterizing the performance of FCDs under SAGD repre-
sentative conditions, ConocoPhillips first conducted several laboratory tests to qualitatively confirm the
validity of their updated hypothesis. The experimental results from these initial laboratory tests were
somewhat coarse, and could not (for example) quantify the exact amount of steam entering the orifice;
however, they did provide a relative and qualitative measure of flashing. Some data from these initial tests
is shown in Fig. 4, where it can be observed that although the mass flowrate was held relatively constant,
the ⌬P across the orifice increased significantly as the amount of steam flashing increased.

Figure 4 —Initial Laboratory Testing of Steam Forming and Flashing Across an Orifice

FCD Characterization Protocol

Although the proposed hypothesis provides some insight into the operation of FCDs under SAGD
representative conditions (specifically when steam forms across the FCD), further work was required to
provide meaningful assistance to ConocoPhillips’ field operations. As previously noted, different vendors
use various approaches to design their FCDs, but the characterization protocol needed to be independent
of these underlying designs in order to establish the performance of all FCDs. Due to the difficulty in
predicting how a phase-change may influence FCD performance under these multiphase conditions, it was
SPE-174416-MS 7

determined that the best approach was to experimentally characterize varying types and sizes of FCDs
from different vendors in two phases:
● Phase 1: experimentally determine the performance of various FCDs over a range of representative
SAGD conditions using a consistent methodology (e.g. tightly controlled test variables, a consis-
tent test matrix, a universal apparatus that can accommodate all styles/types of FCDs, etc.).
● Phase 2: build a predictive model that allows the performance of the FCDs to be estimated based
on the relevant fluid characteristics and operating conditions (using this experimental data). The
ultimate goal would then be to incorporate this predictive model into existing production and
reservoir modeling capabilities.

Phase 1: Building a Multiphase FCD Characterization Flow Loop

ConocoPhillips subsequently worked with C-FER Technologies (Edmonton, Canada) to design and
construct a novel multiphase flow loop to achieve the first phase of the FCD Characterization Protocol
outlined above in Section 4.0. The following section summarizes some of the flow loop capabilities; the
key controlled and measured variables during operation, a typical test matrix to characterize an FCD, and
some of the initial commissioning results and observations.
General Flow Loop Capabilities
Due to the complexity and scope outlined in the FCD characterization protocol, a wide range of
operational capabilities were required as part of this flow loop design, including:
●Operating Envelope: operating at fluid temperatures between 25°C (77°F) and 300°C (572°F),
fluid pressures up to 10.3 MPa (1,500 psig), and controlled liquid mass flowrates at the FCD inlet
between 2 and 22 m³/day (0.4 and 4.0 GPM) Cold Water Equivalent (CWE).
● Liquid Properties: operating with a range of test fluids with varying density, viscosity (0.01 to 300
cP), and lubricity that includes 100% water, 100% oil (of varying types and characteristics), and
(perhaps most importantly) stable oil and water emulsions of various percentages of Water Cut
(WC) to replicate the specific fluid behavior experienced in a SAGD production well, including
operation around the emulsion inversion point.
● Steam or Gas Mixtures: operating with controlled and measured multiphase mixtures using either
(or both) steam and inert high pressure gas to evaluate the effects of flashing across the FCD and/or
to allow for the investigation of partial pressure effects which may occur with multiple gas phases.
This required operation with injected steam quality and/or inert gas flowrates of up to 5% (by
mass) while operating at fluid inlet temperatures up to 270°C (518°F).
● Operating System: fully automated operation, with an inherently safe design to minimize opera-
tional risk and allow for unmanned 24-hr testing on a wide variety of FCDs (or downhole tool
designs).
Key Variables and Measurements
In order to meet these requirements, a novel measurement and control system was developed to allow for
a comprehensive examination of the phenomenon under investigation. In support of the FCD Character-
ization Protocol, a large number of controlled, measured and calculated variables are required during flow
loop operation. Some of the most critical variables include: inlet liquid flowrate and density, inlet gas
flowrate, inlet and discharge temperature, inlet and discharge pressure, FCD ⌬P, and a series of calculated
channels related to fluid properties (e.g. Re, enthalpy, etc.)
Stability and Experimental Error it was necessary for this experimental apparatus to operate as a
stable closed-loop system at several predefined test conditions. In order to meet this requirement, criteria
were developed to outline an acceptable amount of “noise” in each of the controlled variables. An error
8 SPE-174416-MS

analysis was then completed to determine the test duration that would allow for the total experimental
error of each key variable to be under an acceptable value (e.g. less than 1%). The minimum experimental
error is also heavily dependent on the instrumentation: thus high accuracy instruments and regular
recalibration to traceable standards were required. Varying control strategies were also used to optimize
the test procedure.
The measurement and control system was assembled using a variety of National Instruments hardware
platforms including PXI, SCXI and CompactRIO form factors. In addition, an internally developed
LabVIEW application was developed for acquisition, control and data management, including utilities for
data analysis and reduction. Acquisition requirements were set for all measured and derived variables at
a minimum frequency of 4 samples per second with an ability to increase this rate to capture high speed
transient events of interest.

High-level Flow Loop Design

The flow loop was designed to be operated as a closed-loop liquid and open-loop gas (or steam) system
that allows each fluid phase to be independently metered and controlled with a high degree of precision.
Thus, the apparatus can accurately control the fluid properties and conditions at the inlet of the FCD
during testing, while monitoring the performance of the FCD. A simplified schematic of the experimental
flow loop showing the basic piping and key components is shown in Fig. 5 while some images are
provided in Fig. 6.

Figure 5—High-level Schematic of the FCD Flow Loop

SPE-174416-MS 9

Figure 6 —High-level Design and Image of the Flow Loop (photo from above)

Some of the key components in this flow loop include:

● Injection Pump: a positive displacement metering pump is used to control the liquid flowrate
through the FCD. As a piston style pump was ultimately chosen, the piping downstream of the
pump was designed to minimize any flowrate or pressure oscillations and improve stability at the
FCD inlet during testing.
● Heater and Heat Exchangers: an electric immersion circulation heater was installed near the
discharge of the injection pump, while a closed-loop water-cooled heat exchanger system is
situated at the discharge of the separator. Both are used to control fluid temperature during testing.
● Separator: is used to remove injected gas (or steam) as part of an open-loop system while retaining
fluid, and also to directly measure the steam quality or the gas mass fraction of the fluid leaving
the FCD (assuming minimal heat losses and pressure changes between the FCD and the separator).
● Flow Meters: were installed at multiple locations to measure the mass flowrate and fluid density
of single phase mixtures (i.e. gas, steam, water, or an emulsion) to allow for the calculation of the
fluid composition entering and exiting the FCD, and the properties thereof.
● Instrumentation: the flow loop is also instrumented with pressure and temperature transmitters to
determine the fluid properties at various locations, high accuracy level sensors on each vessel to
ensure that the flow loop is operating at equilibrium conditions, a differential pressure measure-
ment across the FCD to determine the FCD behavior, and a viscosity measurement section to better
characterize the fluids entering the FCD.
● Universal Test Fixture (UTF): was designed to allow different FCD designs to be installed into the
flow loop and tested under identical inlet conditions. This fixture can accommodate FCDs with a
maximum external diameter of approximately 235 mm (9.25 inches) and allows for a maximum
FCD length of approximately 0.81 m (32 inches). The goal when designing the UTF was to: 1)
minimize flow losses to and from the FCD, 2) provide a consistent testing environment for each
variety of FCD, and if possible, 3) do it in a manner that was somewhat consistent with the
completion of a SAGD production well.
A schematic and image of the UTF, which allows for a variety of FCDs to be tested in a controllable
manner is shown in Fig. 7.
10 SPE-174416-MS

Figure 7—Universal Test Fixture (UTF) Drawing and Photograph

As previously mentioned, it is possible to test a variety of fluid compositions within the flow loop
(including fluid mixtures such as pure water, any oil/water emulsion, and pure oil) with a total operable
viscosity range of approximately 0.01 to 300 cP. The liquid flowrate during testing is primarily controlled
using the injection pump, while the open-loop gaseous phase can be composed of either pure inert gas
(which is injected immediately upstream of the FCD), pure steam (which is controlled by flashing across
the inlet control valve upstream of the FCD), or combinations thereof entering the UTF.
Steam Quality steam quality is a key control variable during testing. By assuming that the process
across the throttling valves and the UTF is isenthalpic, it is possible to calculate the steam quality at both
the inlet and discharge of the FCD. In addition to this, the steam quality is independently measured at the
discharge of the FCD using the separator and compared to the calculated steam quality. Due to the control
capabilities of the flow loop, it is possible to conduct testing where steam begins to form at the inlet of
the FCD (at specific steam quality targets), but it is also possible to monitor the ⌬P across the FCD when
steam first begins to form at the outlet or at another point within the FCD.
Steam is created using a throttling valve (i.e. a series of control valves in parallel, with overlapping
operating ranges). After passing through the FCD this steam is vented from the separator, thus the amount
of water in the flow loop also slowly decreases during the course of steam testing. For this reason, a high
pressure water injection system is used to replenish the liquid water in the separator to account for the
water vapor vented from the flow loop during steam testing. Alternatively, this high pressure water (or oil)
injection system can also be used to modify the water cut during emulsion testing. The remaining FCD
inlet conditions (namely the inlet temperature and pressure) are typically controlled by varying the flow
loop pressure, and the heater input power. The individual control strategies for each piece of equipment
are quite complex, but ultimately provide the operators with total control of the fluid composition and
operating conditions at the inlet to the FCD.
Safety Systems an extensive safety program was implemented for this flow loop, including layers of
redundant safety systems. Computer and electronic systems are supplemented with mechanical fail safes
(e.g. pressure relief valves, emergency venting systems, containment systems designed to account for
shrapnel and high temperature and pressure steam release, etc.), safe operating procedures, and Personal
Protective Equipment (PPE) to protect personnel. Safety is the most critical consideration during all
testing, and remains a key consideration for ConocoPhillips and C-FER Technologies.

Typical Test Matrix for Characterizing an FCD

Detailed results of FCD behavior have not been presented in this paper; however, a high-level test matrix
showing how an FCD may be characterized using this flow loop is shown below in Table 1. This is not
meant to represent all testing that may be conducted on a single FCD; however, it does outline a number
of baseline tests.
SPE-174416-MS 11

Table 1—Example of a Typical Test Matrix

FCD FCD Inlet FCD Inlet
Inlet FCD Inlet FCD FCD Inlet Liquid Gas Steam
Fluid Properties Pressure Temperature Orientation Flowrate Fraction Fraction

FCD Variant 100% (Water) Variable Ambient 0° and 10 Rates per 0.0% 0.0%
90° Orientation
3 Rates per 5
Orientation Fractions
per
Liquid
Flowrate
0% WC (Pure Oil 5 Targets 0° 10 Rates per 0.0%
with varying ␮,␳) Temperature
3 Targets (Water, and 3 Targets 0° 10 Rates per
2 Emulsions, with Temperature
varying ␮,␳) 3 Rates per 5 Fractions
Temperature per Liquid
Flowrate

Flow Loop Commissioning

During commissioning, all measurement, control, and safety systems were verified using external
standards and found to be operating within acceptable margins of error. It was confirmed that the flow
loop was capable of reaching all design conditions and the performance was as expected. Tuning of the
control loops and optimization of the DAS and test procedure will not be discussed in depth; however, a
few notable observations were made following the commissioning process:
● Influence of measurement locations and their impact on process conditions need to be well
understood when attempting to determine the validity of test results. Filtering effects, pressure
losses, and fluid head should all be taken into account during instrument selection and placement.
Energy transfer at the boundary layer between the process and transducers is also a key design
consideration, especially in high accuracy thermal measurement applications.
● Redundant derived channels which generate predicted results based on first principles (and known
fluid properties) had a great deal of value when validating the measured results, or anticipated
behaviors.
● Repeat tests, executed periodically throughout the program, are an excellent way of determining
that equipment performance has remained unchanged and that performance results are valid.
Phase 2: Modeling FCD Behavior
As previously discussed, the purpose of this ongoing experimental work was to better understand and
characterize the effects of steam flashing through FCDs under SAGD representative conditions, in a
manner that allowed for a predictive model to be constructed. FCDs can have very complex behavior
(more so under multiphase flow conditions) and various operating principles, but at a fundamental level
their purpose is to generate ⌬P. This ⌬P can improve production by delaying or inhibiting undesired
outcomes like steam breakthrough, although it can also inhibit production if not applied correctly.
Some of the variables that can affect the ⌬P through an FCD include:
● Viscosity and density of the produced fluid;
● Flowrate (mass or volume);
● Steam fraction (and/or gas fraction, if present);
● Emulsion properties (water continuous, oil continuous or near the inversion point, where substan-
tial changes in behavior are expected when emulsions change from water-in-oil to oil-in-water);
and
12 SPE-174416-MS

● Flow regime, where it is generally assumed that FCDs operate in the turbulent flow regime while
operating in the field (i.e. Re is greater than 2300). However, testing was conducted under both
turbulent and laminar conditions (laminar conditions may be present, for instance, in some cases
with a very high fluid viscosity) to check the influence of flow regime.
Modeling Strategy
Traditional tools to estimate ⌬P assume that it is a function of Re (which is influenced by flowrate,
viscosity and density), and most reservoir simulators rely on this assumption in their computations.
Unfortunately, this assumption does not hold true when there are phase transitions in the fluids, such as
the formation of steam. In order to accommodate the effects of phase transitions, it may be possible to
estimate the performance of the FCD using models of orifices, and applying steam flashing calculations.
⌬P Estimation for Flow through Orifices (Turbulent Flow) Beginning with the classical equation of
flow through an orifice:

where:
● ⌬P is the pressure drop across an orifice (Pa)
● K is a dimensionless friction factor which is a function of Re and will be determined empirically
● ␳ is the fluid mass density (kg/m3)
● V is the fluid velocity (m/s)
● is the fluid mass flowrate (kg/s)
● A is the conduit’s cross-sectional area (m2)
Also including the classical definition of Reynolds Number (Re):

where:
● d is the diameter (mm)
● V is the fluid velocity (m/s)
● ␳ is the fluid density (kg/m3)
● ␮ is the dynamic viscosity in centipoise (cP)
The formula to fit K as a function of Reynolds Number is typically determined empirically. As
previously mentioned, there are numerous fitting equations for FCDs referenced in the literature. By way
of an example, one such equation that has been used to fit both tortous path and hybrid style FCDs under
single-phase flow conditions (Garcia et al. 2009) is as follows:

where:
●
●
● a1, a2, b1, b2, c, d, and t are all empirical factors, based on flow testing
SPE-174416-MS 13

Enthalpy Steam Flashing Calculations As previously discussed, the ⌬P through the FCD may result in
some amount of water flashing to vapor, where the mass fraction (␹) of saturated liquid at zero quality
that will be converted to vapor is then:

where:
● hf1 is the specific enthalpy of the saturated liquid at the higher pressure (kJ/kg)
● hf2 is the specific enthalpy of the saturated liquid at the lower pressure (kJ/kg)
● hfg is the latent heat of evaporation of the fluid at the lower pressure (kJ/kg)
The volume of fluid will increase as the vapor phase occupies more volume than the liquid phase,
which will cause the velocity of the fluid to increase, as the greater volume will need to pass through the
same area in the next slot. This change would be taken into account in the ⌬P computation of the
succeeding slot and so on.
Coupled Wellbore and Reservoir Simulation
The results of the FCD simulation will ultimately be incorporated into a coupled wellbore and reservoir
simulation model to allow for a better understanding and prediction of the impacts of FCDs on the
reservoir behavior. This work is currently underway and will require further investigation of FCD
behavior (using the flow loop discussed earlier) before being implemented into reservoir models. The
current predictive model can be used with various coupled reservoir wellbore models where each has its
own mechanism for accounting for the behavior of the FCD (which the aforementioned model can
furnish).
Preliminary Results from Subsequent Testing
Further experimental results are provided for an FCD removed directly from a joint of casing to
demonstrate some of the interesting results that have been gained. For this particular FCD (architecture
and size), the results demonstrated a strong steam blocking effect, where ⌬P goes up significantly as the
steam quality increases. This is caused by the volume flow increasing significantly with steam quality
even as mass flow is held constant. Plotting ⌬P as a function of steam quality reveals other interesting
patterns, as shown in Fig. 8.

Figure 8 —Measurements of ⌬P as a Function of Steam Quality Illustrating the Steam Blocking Effect
14 SPE-174416-MS

Fig. 8 also shows that measurements were made when there was no steam flashing, at the onset of
flashing within the FCD, at the onset of flashing at the inlet of the FCD, as well as at various steam
qualities at the inlet of the FCD. There is a very strong effect as flashing starts across the FCD and a
continuing effect as steam quality increases. Other subsequent tests (not presented in this paper) revealed
that the initial ⌬P is largely independent of inlet pressure. However, during tests with steam at the FCD
inlet, the ⌬P varied more significantly with inlet pressure due to the steam blocking effect.

Conclusions
In summary, the design and implementation of a flow loop capable of testing FCDs under SAGD
representative conditions has been a significant achievement. The principal result of the testing has been
a confirmation of the underlying hypothesis of how FCDs operate in a SAGD environment. This
substantiates that the results observed in the field at ConocoPhillips’ well number 102-06 (Stalder, 2012)
were a consequence of how FCDs operate in SAGD environments.
The flow loop enables a process for characterizing FCDs under representative SAGD applications. The
current test procedure and matrix has also been demonstrated to establish this characterization. The
experimental results have confirmed the existence of a steam blocking effect, and provided a mechanism
to quantify it. A model was developed that accurately predicts the behavior of FCDs in response to the
key variables. This characterization and model will ultimately facilitate a more accurate coupled wellbore/
reservoir simulation for the use of FCDs in SAGD. This capability ultimately allows a workflow for
designing and optimizing SAGD equipped completions in the field under these challenging conditions.
Future tests will enable comparing various FCD architectures on attributes such as the magnitude of this
steam blocking effect and the sensitivity to fluid properties (such as viscosity).

Go Forward Plans
The main objective of the 2013 program was to develop the experimental capabilities required to complete
these types of tests. Now that this has been demonstrated, the next step will be to characterize the behavior
of different FCD architectures (e.g. tortuous paths, orifices, nozzles, autonomous FCDs) and determine
how their behavior scales. Further testing is planned to provide this representative data, and ultimately to
increase the scope and functionality of the modeling efforts.

Symbols and Acronyms

CWE Cold Water Equivalent
FCD Flow Control Device
FRR Flow Resistance Rating
ICD Inflow Control Device
PPE Personal Protective Equipment
SAGD Steam Assisted Gravity Drainage
SOR Steam-to-Oil Ratio
UTF Universal Test Fixture
WC Water Cut
WOR Water-to-Oil Ratio
⌬P Pressure Drop across FCD (psi)
Re Reynolds Number
K Dimensionless Friction Factor
␳ Fluid Mass Density (kg/m3)
V Fluid Velocity (m/s)
Fluid Mass Flowrate (kg/s or kg/m)
SPE-174416-MS 15

A Cross-Sectional Area (m2)

␹ Mass Fraction
d Diameter (mm)
␮ Dynamic Viscosity in Centipoise (cP)
hf1 Specific Enthalpy of the Saturated Liquid at the Higher Pressure (kJ/kg)
hf2 Specific Enthalpy of the Saturated Liquid at the Lower Pressure (kJ/kg)
hfg Latent Heat of Evaporation of the Fluid at the Lower Pressure (kJ/kg)
Sk Dimensionless Scaling Factor

References
Banerjee, S., Abdelfattah, T., Nguyen, H. 2013. Benefits of Passive Inflow Control Devices in a
SAGD Completion. SPE Heavy Oil Conference Canada, Calgary, Alberta, Canada, June 11-13,
SPE 165478.
Banerjee, S., Jobling, R., Abdelfattah, T., Nguyen, H. 2013b. The Role of Autonomous Flow Control
in SAGD Well Design. SPE Annual Technical Conference and Exhibition, New Orleans, Loui-
siana, USA, 30 September-2 October, SPE 166266.
Birchenko, V.M., Muradov, K.M., Davies, D.R. 2010. Reduction of the Horizontal Well’s Heel-Toe
Effect with Inflow Control Devices. Journal of Petroleum Science and Engineering, Volume 75,
Pages 244 –250.
Butler, R.M. 1991. Thermal Recovery of Oil and Bitumen, Chapter 7: Steam-Assisted Gravity
Drainage. Prentice Hall.
Garcia, L.A., Coronado, M.P., Russell, R.D., Garcia, G.A., Peterson, E.R. 2009. The First Passive
Inflow Control Device that Maximizes Productivity during Every Phase of a Well’s Life.
International Petroleum Technology Conference, Doha, Qatar, December 7-9, IPTC 13863.
Henriksen, K.H., Gule, E.I., Augustine, J. 2006. Case Study: The Application of Inflow Control
Devices in the Troll Oil Field. SPE Europec/EAGE Annual Conference and Exhibition, Vienna,
Austria, June 12-15, SPE 100308.
Jones, C., Morgan, Q.P., Beare, S.P., Awid, A.E., Parry, K. Design, Testing, Qualification and
Application of Orifice Based Inflow Control Devices. International Petroleum Technology Con-
ference, Doha, Qatar, December 7-9, IPTC-13292-MS.
Kyanpour, M., Chen, Z. 2013. A New Approach for Designing Steam Splitters and Inflow Control
Devices in Steam Assisted Gravity Drainage. SPE Heavy Oil Conference Canada, Calgary,
Alberta, Canada, June 11-13, SPE 165487.
Lauritzen, J.E., Martiniussen, I. B. 2011. Single and Multiphase Flow Loop Testing Results for
Industry Standard Inflow Control Devices. Offshore Europe, Aberdeen, UK, September 6-8, SPE
146347-MS.
Least, B., Greci, S., Burkey, R., Ufford, A., Wileman, A. 2012. Autonomous ICD Single Phase
Testing. SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, October
8-10, SPE 160165.
Lee, B.O., Rabeh, M., Garcia, G.A., Gavioli, P., Vicario, R. 2013. Multiphase (Oil-Water) Loop Flow
Test for Helical and Hybrid Passive Inflow Control Devices. International Petroleum Technology
Conference, Beijing, China, March 26-28, IPTC-17125-MS.
Lorenz, M., Ratterman, G., Augustine, J. 2006. Uniform Inflow Completion System Extends Eco-
nomic Field Life: A Field Case Study and Technology Overview. SPE Annual Technical
Conference and Exhibition, San Antonio Texas, USA, September 24-27, SPE 101895.
Lyngra, S., Hembling, D., Al-Zahrani, T.M., Al-Otaibi, U.F., Al-Marhoun, H.S., Anderson, A.B.
16 SPE-174416-MS

2007. A Case Study of the Application of Slimhole Passive Inflow-Control Devices to Revive
Wells with Tubular Limitaitons in a Mature Field. SPE Middle East Oil and Gas Show and
Conference, Bahrain, March 11-14, SPE 105624.
McIntyre, A., Adam, R., Augustine, J., Laidlaw, D. 2006. Increasing Oil Recovery by Preventing
Early Water and Gas Breakthrough in a West Brae Horizontal Well: a Case History. SPE/DOE
Symposium on Improved Oil Recovery, Tulsa, Oklahoma, USA, April 22-26, SPE 99718.
McKenzie, T., Wenk, A., Kumar, S., Khan, P., Garcia, G. 2011. The Design and Application of Latest
Generation Inflow Control Devices in Non-Horizontal Wells in the Bhagyam Field, India. SPE
Annual Technical Conference and Exhibition, Denver, Colorado, USA, October 30-November 2,
SPE-146623-PP.
Ratterman, E.E., Voll, B.A., Augustine, J.R. 2005. New Technology to Increase Oil Recovery by
Creating Uniform Flow Profiles in Horizontal Wells: Case Studies and Technology Overview.
International Petroleum Technology Conference, Doha, Qatar, November 21-23, IPTC 10177.
Regulacion, R.E., Sampedro, T.F., Least, B., Penaranda, R.P., Gutierrez, F.P.C., Bonner, A.J. 2013.
Autonomous ICD Installation Success in Ecuador Heavy Oil: A Case Study. SPE Annual
Technical Conference and Exhibition, New Orleans, Louisiana, USA, 30 September-2 October,
SPE-166495-MS.
Stalder, J. 2012. Test of SAGD Flow-Distribution-Control Liner System in the Surmont Field, Alberta,
Canada. Journal of Canadian Petroleum Technology, Volume 52, Issue 02, Pages 95–100, SPE
153706-PA.
Stone, T.W., Law, D.H., Bailey, W.J. 2013. Control of Reservoir Heterogeneity in SAGD Bitumen
Processes. SPE Heavy Oil Conference Canada, Calgary, Alberta, Canada, June 11-13, SPE
165388.
Vela, I., Viloria-gomez, L.A., Caicedo, R., Porturas, F. 2011. Well Production Enhancement Results
with Inflow Control Device (ICD) Completions in Horizontal Wells in Ecuador. SPE EUROPEC/
EAGE Annual Conference and Exhibition, Vienna, Austria, May 23-26, SPE-143431-MS.