SPE 84497

Balancing Act: Gulf of Mexico Sand Control Completions, Peak Rate Versus Risk of
Sand Control Failure
G. K. Wong, SPE, Shell E&P Company, P. S. Fair, SPE, Shell E&P Technology Company, K. F. Bland, SPE, Shell E&P
Company, and R. S. Sherwood, SPE, Shell E&P Company.
Copyright 2003, Society of Petroleum Engineers, Inc.
This paper was prepared for presentation at the 2003 SPE Annual Technical Conference and
Exhibition held in Denver, Colorado, U.S.A., 58 October 2003.
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, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes 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 where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
This paper outlines a general concept and develops a practical
well surveillance method to monitor and operate sand control
completions that optimizes production without introducing
extra risk of completion impairment and failure. The general
concept requires: (a) determining the sand control failure
criteria that the surveillance method should be based on, (b)
establishing a direct link between the identified failure criteria
and pressure transient analysis information, and (c) validating
the surveillance method. The proposed surveillance method
utilizes readily accessible well information without requiring
production log measurements of down-hole velocities within
the completion interval.
The proposed method is fully developed for cased hole,
gravel pack completions assuming the gravel-filled
perforations dominate flow within the completion. The
equations are given and velocity criteria are established. In
this case the two dominant completion failures are screen
erosion and destabilization of annular pack. The maximum
flowing velocities for these failure mechanisms can be
established using field production logs, laboratory screen
erosion data, and mathematical calculations of fluid flow in
the annulus pack. The pressure drop across the gravel packed
perforation tunnel is the dominant completion pressure drop
for the cased-hole gravel pack system. This pressure drop
equation is non-linear or proportional to square of perforation
flowing velocity.
The well surveillance method monitors down-hole flowing
velocity and completion pressure drop to operate the well
without introducing unnecessary completion impairment and
sand control failure risks. The process of ramping up the well
and determining a safe maximum rate goes beyond the strict
adherence to absolute values of acceptable completion

pressure drop and down-hole velocities and is integrated with

prudent surveillance.
The method recommends several
pressure transient analyses be taken, at increasing flow rate,
during the process of ramping up toward the peak rate, to help
assess and diagnose the performance of the completion. This
integrated assessment by direct measurement of completion
pressure drop and corresponding calculation of flowing
velocity provides real time feedback of information on the
performance of the completion during the ramp up process.
Using this direct and simple process, engineers can customize
the operating guidelines for each well so that less impaired
wells can be produced at higher rates and the more impaired
or higher risk wells are identified for remedial operations and
produced less aggressively.
This paper demonstrates the application of the proposed
concept and surveillance procedure, in reaching the peak rate,
for the cased-hole gravel pack completion. Three field
examples are provided to illustrate how engineers are able to
use standard pressure transient analyses and the proposed
simple method to assess the completion response to each
additional choke increase and to optimize well productivity
during the ramp up process. This paper summarizes key
findings on the implementation of this new integrated
surveillance tool.

Introduction
The economics of developing unconsolidated sand and geopressured reservoirs in deepwater in the Gulf of Mexico
(GOM) increasingly demands fewer wells per project. These
wells need sand control completions and require higher
production rate and higher ultimate recovery per well.
Production from each well is more critical which highlights
the importance of maintaining well productivity and
minimizing completion failure risk.
Proper selection, effective design, rigorous quality
assurance and control of equipment and field execution
procedures of sand control completion have been recognized
to be essential for meeting productivity and reliability
requirements.1,2 However, for geo-pressured reservoirs the
higher reservoir pressure also provides the potential of
delivering greater rate with higher production drawdown and
larger tubing. This approach elevates the need for developing
well surveillance guidelines to operate these wells without
impairing and failing the sand control completion.

Production and surveillance engineers are constantly

balancing the act of maximizing the production rate per well
versus the risk of opening the well too far and compromise the
integrity of the sand control completion. Different maximum
rate guidelines, largely based on field experiences, are
available in literature using completion pressure drop ranging
between 200 and 400 psi.3,4 However, systematic definitions
of maximum flow rate, completion pressure drop, and
practical well surveillance guidelines to operate sand control
wells are not available in literature and warrant consideration.
The main objective of this paper is to develop a practical
well surveillance method to monitor and operate sand control
completions that optimizes production without introducing
extra risk of completion impairment and failure. This method
is based on the mechanical skin component of the completion
pressure drop (pskin-mechanical) and down-hole flowing velocity.
To be practical, pskin-mechanical is determined directly from
standard pressure transient analyses (PTA). The down-hole
flowing velocity is calculated using pskin-mechanical and a
realistic completion model with known reservoir and
completion parameters. The surveillance method recommends
that pressure surveys be taken, at increasing flow rates, during
the process of ramping up toward the peak rate. This
approach provides real time feedback of changes in the
completion performance and allows engineers to customize
the operating guidelines for each well. For example, less
impaired wells can be produced at higher rates and the more
impaired or higher risk wells are identified early for possible
remedial operations or to be produced less aggressively.
This paper begins with the description of a generalized
approach in arriving to the proposed surveillance method.
This is followed with application of the approach to casedhole gravel pack completion which includes high rate water
pack (HRWP) and fracpack (FP). Field examples of high rate
wells are presented to illustrate and discuss the practicality
and limitation of this new surveillance method. Finally, main
lessons learned are discussed and conclusions of this study are
summarized.
The reader should remember three key things while
reviewing this strategy:
It is not a draw down limit
Use both pcompletion and flowing velocity
Different completions have different critical failure
mechanisms and maximum velocity criteria
The process of opening up a well involves a wide range of
operations that could impact the well performance. Some of
these operations include: unloading a new well for the first
time, beaning up or rate of choke changes, re-starting a well
following a shut-in, and ramping up the well to reach the
peak-rate. The surveillance method presented in this paper
only addresses the ramp-up - bringing a completion to its
peak rate for the first time. Results from this paper can be
utilized to design and operate sand control completions more
effectively. This paper should be of interest to production,
completion, and reservoir engineers.

A Generalized Approach
This section outlines a generalized procedure to develop a

SPE 84497

surveillance method that maximizes production of sand

control completions without introducing extra risks for
completion impairment and failure. The procedure involves
(a) determining the sand control failure criteria that the
surveillance method should be based on, (b) establishing a
direct link between the identified failure criteria and PTA
information, and (c) validating the surveillance method.
Sand control failure criteria. Knowing the maximum rate to
safely produce a sand control completion is a critical step.
This depends on causes of completion failures and would be
specific to the types of sand control completion, for examples,
cased-hole gravel pack, open-hole gravel pack, screen only
completion, expandable sand screen, etc. Therefore, for a
given type of sand control completion the following should be
considered:
1. Identifying the dominant failure mechanisms. Sand
control failure is usually defined as production of
intolerable amounts of sand that requires choking back or
losing all production. There are numerous possible
causes for completion failures during well production.
These include: (a) screen erosion, (b) screen corrosion,
(c) hot spots from localized flow after screen plugging
by scale, (d) hot spots caused by inadequate gravel
packing in the annulus resulting in localized flow, (e)
screen collapse due to compaction, (f) destabilization of
annular pack because of excessive down-hole flowing
velocity from perforation (at casing ID), (g) screen
collapse due to plugging, etc. The dominant failure
modes would depend on the type of sand control
completion being considered. For cased-hole gravel
pack, the dominant failure mechanisms are screen erosion
and destabilization of the annular pack. For cased-hole
expandable screen the failure mechanism is screen
erosion.
2. Determining the maximum constraints for these
failures. The limiting condition or maximum constraint
of pressure drop or flowing velocity for each failure
mechanism usually is obtained from physical testing,
numerical modeling, and field data (e.g., production logs
and well performance and failure databases). Some
general rules of thumb have also been developed based on
field observations. For example, completion pressure
drop between 500 and 1000 psi has been viewed as
intolerably high for gravel pack impairment and screen
erosion.3
Failure criteria and PTA. This step integrates completion
failure criteria with surveillance data. The principal tool for
understanding the performance of sand control completions is
PTA. The primary reason is practical. Although a full
characterization of the completion failure and performance
requires distributed pressure and flow velocity measurements,
the reality is that performance data is usually limited to
pressure data from permanently installed down-hole pressure
gauges or the occasional test obtained by running slickline
conveyed memory pressure gauges. Production log data
provide the closest and most direct down-hole flow velocity
information. Unfortunately, such logs are not commonly run.
Thus any use of PTA carries all of the inherent uncertainties
associated with how the method quantifies the reservoir and

SPE 84497

completion components.
Although the fundamentals of PTA are well established,5 a
brief discussion is warranted to clarify how PTA results are
intended to be used with this surveillance method. The main
parameter for quantifying the completion is the skin (S). The
fundamental definition of skin has not changed significantly
since the concept was first introduced by van Everdingen and
Hurst.6,7 The pressure drop due to skin (p skin) is determined
from the PTA and skin is obtained by converting this pressure
drop attributed to skin into a dimensionless pressure.8

S=

kh
pskin .. (1)
141.2qB

Generally, p
components:

skin

could

be

separated

into

different

p skin = p skin-mechanical + p skin-geometrical + p skin-other .. (2)

The pskin-mechanical is the main part of the completion pressure
drop that directly impacts the completion failure mechanism in
question. The pskin-geometric is attributed to the geometry of the
well bore. Partial penetrations like limited perforation,
hydraulic fractures in the FP, and hole deviation angle are
some of the possible contributing factors.9-11 They affect the
total completion performance but they are not directly related
to the completion failure mechanism. Finally, pskin-other is the
part of pressure drop that is not directly attributed to the
completion failure mechanism that the engineer can
effectively estimate and remove. The most typical example
for down-hole permanent pressure gauge installation is the
pressure drop between the gauge and the perforation due to
both friction pressure drops in the tubing and the differences
between flowing and static bottom-hole fluid gradients. In
addition, in gas-condensate systems flowing at pressures
below the dew point, one might remove the pressure drop due
to condensate buildup surrounding the well bore.
The completion pressure drop needed to relate and
integrate with the completion failure mechanism is given as:
p skin-mechanical = p skin - p skin-geometrical - p skin-other (3)
The down-hole flowing velocity is then calculated with a
realistic completion model that describes the flow in the
dominant failure mode using pskin-mechanical and known
reservoir and completion parameters as inputs.
Surveillance method. Both down-hole flowing velocity and
p skin-mechanical are monitored during the completion ramp-up
process. Practical surveillance guidelines should not focus
solely on how close a completion is with respect to the
maximum velocity and pressure drop limits. Instead, the
emphasis should be on assessing and monitoring relative
changes in the completion performance. The surveillance
procedure recommends several pressure transient analyses be
taken, at increasing flow rates, during the process of ramping
up toward the completion peak rate. This process helps
diagnose the performance of the completion with real time

feedback and allows for corrective actions, when needed, be

taken before opening to a higher production rate. Engineers
can customize the operating guidelines for each well so that
less impaired wells can be produced at higher rates and the
more impaired or higher risk wells are identified for remedial
operations and produced less aggressively.

Implementation for Cased-hole Gravel Packs

This section develops the surveillance method for cased-hole
gravel packs using the generalized approach outlined above.
Cased-hole gravel packs include HRWP and FP completions.
They cover a large portion of sand control completions
currently deployed in deepwater in GOM. Applications to
other types of sand control completions, using the outlined
approach, are possible but are beyond the scope of this paper.
Sand control failure criteria.
Figure (1) shows the
completion components of cased-hole gravel packs and
depicts the production flow path through the system. This
illustrates the highest flowing velocity is through the
perforations where the flow converges. The average flowing
velocity exiting the perforation at the casing inside diameter
(ID) is labeled as Vc and the flowing velocity on the screen
surface directly across the perforation is labeled as Vs. For
cased-hole gravel packs, the two dominant completion failure
modes which may be avoided by controlling production rate
are: (a) destabilization of annular pack and (b) screen erosion.
Other failure mechanisms such as screen collapse due to
plugging and screen collapse due to casing deformation
induced by reservoir compaction could also be considered.
However, they are generally not viewed as dominant
mechanisms during the completion ramp-up process.
The destabilization of the annular pack is an instability
failure that occurs when the perforation velocity at the casing
ID (Vc) is high enough to fluidize the granular pack in the
annular region around the perforation. Laboratory tests and
field production log data can provide estimates of the
maximum flowing velocity (Vcm) above which this failure
occurs. Using a semi-radial laboratory flow model with a 2
pre-pack outside a 0.5 perforation, Penberthy and Cope12
observed no annular pack instability and impairment up to the
limit of the equipment at 80 bbl/d/perf or 3.8 ft/sec. Figure (2)
shows a production log for a FP completion at 8,000 bpd
where 3,500 bpd of production is being produced over a 3
perforated interval at 8 SPF with 0.7 perforation and a
formation volume factor of 1.5. This well reached a peak rate
of 22,586 bpd and failed the gravel pack shortly thereafter.
Assuming that the production profile remained unchanged
from 8,000 to 22,586 bpd, the estimated down-hole flowing
velocity (Vc) was greater than 14.6 ft/sec. Laboratory tests to
better define the critical or maximum flow velocity that would
destabilize an annular pack are ongoing in Shell. In the
meantime, a conservative maximum velocity limit (Vcm) of 10
ft/sec is recommended until additional laboratory and field
results are available. Equation (4) defines the maximum flow
velocity criterion for the destabilization of annular pack
failure as:
For Failure: Vc Vcm = 10 ft/sec . (4)

SPE 84497

Screen erosion is a progressive failure that depends on

fluid flow velocity, the angle of incidence of the flow, sand
particle size and concentration, duration of the flow, and fluid
properties (such as density and viscosity).13 Laboratory testing
procedures to determine the screen erosion resistance data for
different types of screen and screen erosion failure models are
available.14 For the cased-hole gravel pack, Figure (1) shows
the fluid flow exiting from the perforation (Vc) is directly
impinging on the surface of the screen at a flow velocity of Vs.
The annulus gravel pack protects the screen by bridging and
preventing large and high particle concentration from eroding
the screen. Defining the critical velocity to erode the screen is
a challenging task as down-hole flow dynamics are difficult to
predict. Samples of eroded screen retrieved from the field
usually show screen erosion to be localized. If we define a
wire-wrap screen as eroded or failed when its slot
increases by 0.001, Figure (3) shows the modeled time to
failure as a function of screen flow velocities for different
sand concentrations.14 For a desired screen life of 1 year (at
the peak rate), the screen flow velocities are 0.65 and 1.2
ft/sec for the high sand concentration of 50 lbs per thousand
bbls (pptb) and the medium concentration of 10 pptb,
respectively. Until more reliable field screen erosion failure
data are available for additional model calibrations, a
maximum screen erosion velocity limit (Vsm) of 1 ft/sec is
currently proposed. Equation (5) defines the maximum flow
velocity criterion for the screen erosion failure as:
For Failure: Vs Vsm = 1 ft/sec .(5)
It is important to highlight that laboratory screen erosion data
for gas are not available.14 Extension of the data and erosion
model to down-hole gas flow should be done with caution.
Failure criteria and PTA. After the dominant failure
mechanisms are identified and the corresponding criteria or
maximum flow velocities are established (Equations (4) and
(5)), the next step is to develop solutions to calculate the
flowing velocities Vc and Vs using realistic completion flow
models and PTA data. This step links the surveillance data
and the completion failure criteria to effectively operate the
well without increasing risks of completion impairment and
failure.
For screen erosion, the flow velocity outside the screen, Vs,
can be described using porous media flow equations with the
flow velocity at the casing ID, Vc, being one of the boundary
conditions.15 For Darcys flow the solution of Vs depends on
the thickness of the annular pack (tap), casing inside diameter
(ID), and perforation pattern. The thickness of the annulus
pack helps diffuse the flow exiting the perforation (at the
casing ID) and reduces the flow velocity impacting the screen.
Increasing the thickness of the annular pack reduces Vs for a
given Vc. For example, Vs is 1/16 of Vc after the flow diffuses
over 1 thick annular pack in a 7 ID casing. In the extreme
case when casing ID equals screen OD (i.e., no annular pack),
such as the case of expandable sand screen on perforated
casing, Vc and Vs would be the same. Therefore, in general
once Vc is known, then through Darcys flow, Vs can be
determined.15

For the destabilization of annular pack, the perforation

flowing velocity, Vc, is the main variable. In cased-hole
gravel packs the perforations are the most vulnerable part of
the completion. All flow from the reservoir must pass through
the perforation tunnels (between the casing cement sheath and
the casing ID) packed with gravels. The porous media flow
across this cylindrical perforation tunnel can enter the nonDarcy regime and the flowing pressure drop, p perf, as a
function of Vc, for a nearly incompressible fluid, is given as:3
p perf = b (Vc) + a (Vc) 2 . (6a)
where
b = 1.138E+06 Lp / kg .. (6b)
(6c)
a = 1.799E-05 g Lp
Equation (6a) provides a simple and effective means to
calculate Vc after p perf is known. This pressure drop will be
provided from PTA.
The perforation tunnels packed with gravel are the most
likely choke point within the cased-hole gravel pack
completion where both the flow velocity and pressure drop
would be the highest. For simplicity, the assumption is made
that the entire completion pressure drop from PTA can be
attributed to flow in the perforation tunnels only. Therefore,
Equation (6a) can be re-written as:
p skin-mechanical = p perf = b (Vc) + a (Vc) 2..(7)
The use of Equation (7) would generally result in a higher
value of Vc as other pressure drops in the completion (Figure
(1)), like flow converging into perforations outside the casing
and flow across the annular pack are ignored with the above
assumption. This is conservative but not inappropriate given
the uncertainties of field data and economic consequences of
well failures. The proper application of Equation (7) requires
that the completion pressure drop be extracted consistently as
described in Equation (3).
Besides p skin-mechanical, PTA also provides the production
rate, q, before the pressure buildup. This and the calculated
Vc provide a direct means to calculate the number of flowing
perforations, N, and effective shots per foot, SPF in the
completion as:
N = 9.358E-03 ( B q / (Vc Ap )) .. (8a)
SPF = N / hp .. (8b)
Surveillance method. Equations (6) to (8) and solution of Vs
complete the set of equations needed for the proposed
surveillance method. They require (a) field surveillance
information of q and p skin-mechanical from the PTA, (b)
reservoir properties of B, , and , and (c) completion
information of tap, casing ID, Ap, Lp, kg, and g. All the
reservoir and completion inputs are readily available and can
be bracketed using laboratory data and engineering judgment.
They are also less likely to change and are assumed to be
constant during the ramp-up period in order to arrive to a
practical surveillance method. However, the calculated
flowing velocity (Equation (7)) depends on the gravel

SPE 84497

properties of kg, and g. These inputs are chosen such that the
surveillance method can optimally identify and separate
known failures without being too limiting as calibrated from a
set of available field data.
To assess the feasibility of the maximum velocity criteria
and to select the optimum gravel input properties for the
surveillance method, a set of 45 wells is gathered and
analyzed using the PTA and flow calculations described in
Equations (3) and (7). These wells all have permanent downhole pressure gauges to ensure the most consistent and reliable
PTA interpretations. There are 32 FP completions with 30 oil
and 2 gas wells. The peak production rate ranges between
5,000 and 38,740 bpd for oil wells and 55 and 101.7 MMCF/d
for gas wells. The 13 HRWP completions have 8 oil and 5
gas wells with peak rates ranging between 10,640 and 28,730
bpd and 74.6 and 119.3 MMCF/d for oil and gas wells,
respectively. Although Shell has many more sand controlled
wells, the choice was made to limit the wells included in this
study to recently installed sand control systems with high
quality PTA data such that the method implied by Equation
(3) could be maintained.
Within this dataset, there are 8 known sand control
completion failures: 3 gas and 1oil HRWP and 3 oil and 1 gas
FP wells. They are labeled as Failures (f1) to (f8). Detailed
evaluations of these failures reveal that: Failures (f1) and (f2)
are completion installation induced failures (with burst
screens), Failure (f4) is a high risk completion design with an
annular gravel pack thickness of 0.5, and Failure (f5) is a
late-life failure (over 14MMbbls recovered) attributed to
casing deformation due to reservoir compaction. No obvious
completion design and installation problems have been found
for the remaining Failures (f3), (f6), (f7), and (f8). Only these
last four failures should be used to assess the applicability of
the proposed maximum velocity limits. The other failures
(design, installation, and compaction problems) are included
for completeness and to highlight the importance of how nonrate factors can also greatly impact completion failures.
Figures (4) and (5) show p skin-mechanical vs. Vc and p skinmechanical vs. Vs plots, respectively, for all 45 wells at their peak
rates using a gravel permeability with 65% damage. Sand
control failure points and the appropriate maximum velocity
limit line (Vcm = 10 ft/sec or Vsm = 1 ft/sec) are depicted on
these plots to help assess the applicability of the method.
When sand control wells fail below the maximum velocity
limits, the consequences are costly well failures. Inputting
severely damaged gravel (i.e., lower kg and higher g) in
Equation (7) would predict lower perforation and screen
velocities and increase the risk of failing the completion
before reaching the maximum limits.
For annulus
destabilization, Figure (4) shows Failures (f6) to (f8) are
appropriately placed above the velocity limit of 10 ft/sec but
Failure (f3) is below the velocity limit. For screen erosion,
Figure (5) shows the velocity limit of 1 ft/sec delineates
Failures (f3) and (f7) but not Failures (f6) and (f8). These
results suggest that, with the input of 65% gravel damage,
neither one of the failure mechanisms alone properly defines
all four failures in this dataset. This is not unexpected given
the complexity of the completion. Fortunately, when both
proposed velocity limits are used they predict all 4 failures.

For example, Failure (f3) is below the annulus destabilization

limit in Figure (4) but is correctly placed above the screen
erosion limit in Figure (5). This is encouraging as Failure (f3)
had a small annular thickness of 0.65 where the screen
velocity is expected to be higher. Similarly, Failures (f6) and
(f8) are below the screen erosion limit in Figure (5) but both
failures are already covered in the annulus destabilization
limit in Figure (4).
Therefore, the proposed annular
destabilization and screen erosion velocity limits are
complementary. Both need to be considered in cased-hole
gravel packs.
When sand control wells operate above the maximum
velocity limits without failing, the implication is the velocity
limits are too restrictive or the predicted velocity is too high
such that production potentials may be lost as wells below the
rate limits are being held back unnecessarily. Inputting less
damaged gravel (higher kg and lower g) would accentuate
this effect of predicting higher flowing velocities. Figures (4)
and (5) show that wells with flowing velocities larger than the
maximum velocity limits do not always fail. For example, in
Figure (4) there are 5 non-failure wells and 3 failed wells with
Vc greater than 10 ft/sec. This results in a non-failure rate of
62.5% (5/8) when Vc exceeds the limit of 10 ft/sec. This
conservatism may be partly due to the assumption that all
completion pressure drops are attributed to the pressure drop
in the perforation tunnels. Optimizing the maximum velocity
limits and the input of gravel properties require balancing the
well failure risk against the potential production benefits.
This is still a major engineering challenge inhibited by limited
and uncertain data from production, the completion, and the
reservoir. The selection of gravel properties with 65%
damage has allowed the proposed method to correctly separate
all 4 failures without being too restrictive. The surveillance
method proposes applications of the maximum flowing
velocities in Equations (4) and (5) and the input of gravel
properties with 65% damage.
Relying on a single maximum completion pressure drop
limit to define sand control failure is not effective. Figure (4)
shows Failure (f8) failed at 315 psi while Failure (f7) failed at
3050 psi. However, completion pressure drop is still the most
readily available and directly measurable information that is
useful in identifying impaired wells and wells that are
operated outside the known pressure envelops. Producing a
well with high completion pressure drop for a long extended
period is not recognized to be a good practice. Therefore, any
proper surveillance tool should make use of this completion
pressure drop together with down-hole flowing velocities.
Well surveillance should not focus solely on determining
the precise maximum velocity limits (Vcm and Vsm). Many
completion design and execution factors directly impact
completion failures. Failures ((f1), (f2), (f4), and (f5)) are
good examples. As shown in Figures (4) and (5), these
failures usually occur at flowing velocities below the
maximum velocity limits. An effective well surveillance
method should focus on identifying high risk completions
early and developing an effective diagnostic tool to closely
monitor performance changes so that corrective actions may
be taken early enough to prevent a potential failure. The
process of identifying high-risk completions involves

SPE 84497

evaluations of reservoir and sand control demand, completion

design (material and equipment), and completion execution.
Details on these requirements are beyond the scope of this
paper. Instead, the next section discusses the surveillance
diagnostic tool and monitoring procedures illustrated using
several field examples.
Field Examples
The first step in the process is the determination of the
completion pressure drop as given in Equation (3). In
practice, skin determined from the pressure buildup tests are
resolved by matching the measured pressure response for the
well to completion/reservoir model specific dimensionless
pressure solutions. Extrapolating the model back to the time
of shut in and determining the difference between the flowing
pressure and the model results establishes the pressure drop
due to skin.
Several factors prevent the straightforward calculation of
the pskin. The first factor is one of measurement. Time from
shut in and distance from the well bore are roughly correlated.
The radius of investigation16 (ri) demonstrates this point.

ri =

8(0.0002637)kt
...(9)
ct

The better quality gauges provide data at sampling rates of 1

to 2 seconds. During this time frame, the pressure response is
significantly influenced by flow in the reservoir beyond the
actual completion. The second factor is well bore storage.
Because the well is shut in at a valve at the surface, there is a
period of time where the sand face flow rate decreases as the
pressure rises filling the well bore with production. During
this time, the pressure transient is dominated by well bore
capacity influences rather than flow from the base pipe of the
screen out to the reservoir. It is not uncommon to lose at least
one log cycle of reservoir pressure response from the early
time data for purposes of interpretation.
Finally, the
interpretation models are relatively simple compared to the
actual completion. Rarely do actual pressure transient models
explicitly model details of the completion such as flow
convergence into the individual perforations, the variation in
layer permeabilities, and flow within the sand face to the top
of the completion interval.
One other aspect must be considered. In the near well bore
region, high local velocities may be encountered leading to
non-Darcy flow and rate dependent skins. For purely radial
flow into a cylindrical well bore, the Reynolds numbers are
not sufficiently large to indicate the presence of non-Darcy
flow except in high rate gas wells17,18; however, the sand
controlled completions that are HRWPs and FPs are cased
hole, perforated completions. When the completion acts as if
there is only a limited effective shot density (SPF), the
velocities in the perforation tunnels enter the non-Darcy
region even for many oil wells. The use of Equation (7)
indicates this point. The skin results of the buildup tests
contain both the non-Darcy and Darcy components of skin.
The total skin is the mechanical skin. It is rate dependent.
Altering the completion with different perforations, gravel, or
even stimulation to clean the perforations will lead to different

mechanical skins by changing the flow patterns or the porous

media properties. Running a buildup test after producing at a
higher rate will also lead to higher skins in the absence of a
reduction in impairment affecting both the Darcy and nonDarcy components. Thus mechanical skin determined from
the pressure buildup tests is more than just the Darcy
component of skin.
Well A. The first example, Well A, is a FP oil well. The well
was originally brought on stream at a rate of 2000 bpd to
unload the completion fluids with about 100 psi drawdown.
After oil cuts measured over 95%, the well was switched from
the clean-up system into the production process train. The
well was slowly ramped up to achieve an estimated 500 psi
total drawdown, which corresponded with a rate of 10,800
bpd, then stabilized. A pressure buildup test was performed.
Figure (6) shows a picture of the measured data along with the
model match. Note the derivative exhibits the transition from
linear to pseudo-radial flow which helps quantify the
geometric skin. For many FP wells, no signs of linear flow
are present in the buildup due to the low conductivity of the
fractures and the high diffusivity of the formations. In those
cases, the fracture geometric skin must be estimated from the
design parameters along with the actual installation reports.
The pertinent results along with the gauge corrections are
shown in Table 1.
Figure (7) shows the results of the first two buildup tests
plotted on a standard diagnostic plot. This has rate (q) (in the
Y axis) and both casing and screen velocities (Vc and Vs) (in
the Second Y axis) as a function of the p skin-mechanical in the Xaxis. The pressure drop due to the mechanical skin is defined
by Equation (3). Equations (7) and (8) define the q curve
through Point (1) where the velocity (Equation (7)) is
converted to rate using the average cross-sectional area
(Equation (8a)) and effective number of shots per foot
(Equation (8b)). The curvature is a result of non-Darcy flow
and highlights the sensitivity of q with pskin-mechanical for
cased-hole gravel packs, even for oil wells. Point (2), the
second buildup test result, is also plotted on Figure (7). This
point lies above the curve defined by Point (1). On this plot,
the rate line is the current estimated operating line (based on
the first buildup test). If the next data point falls above or to
the left of this curve, as in this case, the well has cleaned up
with production. If the point lies below or to the right of the
curve, then impairment is developing indicating that the well
should be monitored carefully. Naturally, a well with
increasing impairment also has an increased risk of well
failure if the impairment is completion related, such as caused
by a plugging mechanism. Note that despite the increase in
skin interpreted in the PTA between Points (1) and (2) in
Table (1), Well A cleaned up, yielding a higher target peak
rate, according to this model.
The velocity curves for the secondary Y-axes on Figure
(7) can be used to quickly identify the ranges of p skin-mechanical
corresponding to the maximum velocity limits. By solving
Equation (7) with the first buildup results, the estimated
casing and screen velocities can be obtained for the entire
range of operating pressure based on the method assumptions

SPE 84497

that all reservoir and completion inputs remain constant

during the ramp-up period. From this plot, it is obvious that
neither the casing nor the screen velocity will be limiting this
well (10 and 1 ft/sec, respectively). Instead, the extrapolated
pressure drop of 1,220 psi at a rate of 30,000 bpd would
exceed the standard operating range for pskin-mechanical of 1,000
psi or less. Again, this range is not a limit; however, good
wells are not often operated close to the upper pressure end
because other practical facilities constraints enter in or the
skin can be reduced through stimulation to produce the well
more efficiently. The majority of high drawdown wells are
either geologic disappointments or wells that have been
impaired with production. Therefore, it makes sense to take
an intermediate step (Point (3)) before going all the way to the
peak-rate to ascertain if well clean up would continue.
Table (1) contains the data for the sequence of buildup
tests used to achieve the peak rate for Well A. Figure (8)
depicts all the buildup test points. They show continued clean
up over the entire ramp-up period, which represents 320,000
total cumulative barrels of oil produced. After Point (4) the
risk to reach the peak-rate has been reduced with an
extrapolated pressure drop of less than 1,000 psi. Based on
this and the history of continued clean up, Well A was then
opened up to the peak-rate (Point (5)). Not all of these
buildups were planned. Because permanent down-hole
gauges were installed, surveillance data are also obtained
when routine field operations or upsets shut the well in.
Water production is usually too small to measure during the
ramp-up period but samples of completion brine are obtained
with major rate fluctuations. This supports the contention that
the well is still cleaning up by slowly yielding completion
fluids. Attempts to increase production beyond the last point
were not warranted as the well was fully open into the lowest
pressure system available.
Well B. This is a FP oil well in a 5.5 23 lb/ft casing with a
3.5 screen outside diameter. The annulus gap is only 0.6
thick; thereby, raising pre-production concerns with the
potential for high Vs and screen erosion failure. Figure (9)
depicts the diagnostic plot with the buildup data during the
entire ramp-up period. The q vs. pskin-mechanical curve shown is
based on the last buildup (Point (6)). Both, Vc and pskinmechanical beyond the forecast rate of 11,000 bpd were not the
limiting factors. As anticipated, due to the 0.6 annulus pack,
Vs became the limiting factor. At Point (2), Vs was already
0.81 ft/sec at a production rate of 5,000 bpd. Therefore, a
more cautious approach with additional buildup tests was
warranted. Fortunately, this well continued to show clean up
(Points (1) to (5)). After Point (5), the history of continued
improvement gave us the confidence to open up this well
toward a higher than expected peak-rate of 14,400 bpd at a Vs
of 1.0 ft/sec. This example illustrates how surveillance data
and performance history were used to effectively manage the
risk of pushing this challenging completion (5.5 liner) to a
higher than targeted rate as it approached the maximum
velocity limit.
Well C. This last example is a FP oil well that had a sand
control failure. The well was ramped up before the current
surveillance method was developed. This example is used to

demonstrate how the current surveillance method could have

detected the potential failure risks and discuss possible
surveillance procedures to help prevent the failure.
Figure (10) depicts the diagnostic plot with the buildup
data during the entire ramp-up period. Again, the q vs. pskinmechanical curve was based on the last buildup data of Point (5).
From the first buildup data point both velocity curves (Vc and
Vs) are already available for the entire operating range. Table
(2) details the operating values for desired rates of 15,000 to
20,000 bpd, which highlights the challenge and the potential
risk in operating this completion with high pressure drop
(greater than 1,000 psi) and high casing velocity (exceeding
the limit of 10 ft/sec). Therefore, a more cautious and careful
monitoring for this well would be warranted, as the well is
being ramped-up. The second pressure buildup point, Point
(2), was collected at 14,600 bpd. The well showed clean up
and the actual pskin-mechanical was only 435 psi. However,
between Point (2) and Point (3) the well showed impairment.
Unfortunately, the well was opened up to a peak rate of
22,590 bpd, for a brief period of several days. This resulted in
a predicted pskin-mechanical of 1,240 psi, Vc of 11.5 ft/sec, and
Vs of 0.9 ft/sec based on the buildup Point (3). The well was
brought back to a lower rate of 17,120 bpd (Point (4)). Again,
between Points (4) and (5) the well showed additional
impairment. One more time, in spite of the impairment
information the well was opened up to the last point of 18,300
bpd at a predicted pskin-mechanical of 990 psi, Vc of 10 ft/sec,
and Vs of 0.8 ft/sec. By coincidence, the well started to make
gravel and formation sand shortly after this last point.
It is difficult to conclude that well operations caused this
well to fail. However, the surveillance information as
presented in the diagnostic plot has revealed several high risk
activities that should have been avoided. Common sense
suggests that when a well shows sign of impairment, it is not a
good practice to open the well to higher rate before
understanding the cause of impairment. In this case the risk is
further compounded as the well was already operating close to
the casing velocity limit and standard high pressure drop
envelop.
Summary
A practical well surveillance method to monitor and operate
sand control completions that optimizes production without
introducing extra risk of completion impairment and failure
has been outlined. The feasibility and applicability of this
method for cased-hole gravel pack completions has been
established.

Operating Strategy. A simple, direct, & efficient ramp

up method using available PTA, reservoir & completion
information is developed.
No special data (like
production logs) and large number of calibration factors
are needed. The only calibration factor is the amount of
gravel impairment. In this case, a reasonable 65%
damage was calibrated from the dataset.
Two Complementary Velocity Failure Criteria.
Annulus destabilization and screen erosion were used for
the cased-hole gravel pack failure criteria. The velocity

SPE 84497

criteria or maximum velocity limits for these failure

mechanisms of 10 and 1 ft/sec have been shown to work
reasonably well without being too restrictive. However,
these limits should be updated as additional field and
laboratory data are made available.
Surveillance Tool. A simple but effective diagnostic plot
has been introduced to facilitate data evaluations during
the surveillance period that shows all 4 critical parameters
(pskin-mechanical, q, Vc, and Vs) on the same plot.
o As an outcome of our simple perforation model and
the assumption that reservoir and completion inputs
are unchanged during the ramp-up period, the casing
and screen flow velocities can be calculated for the
entire operation range once we have just a single
pressure buildup data point. This allows early
assessments of different failure mechanisms at the
peak production rate. For example, the dominant
mechanisms for the three field examples were: high
pressure drop across completion for Well (A) , high
screen velocity for Well (B), and high pressure drop
and casing velocity for Well (C).
o The relative positions of successive buildup data
points with respect to q vs. pskin-mechanical curve
provide a quick and simple visual presentation on
how the well is progressing. More importantly, it
tracks the relative changes of productivity during the
ramp-up process.
o The q vs. pskin-mechanical curve predicts changes in
pressure drops for higher production rates more
reliably as it captures the non-Darcy flow effects.
Lessons Learned. Key lessons learned from the 45 wells
were:
o Well surveillance should focus on relative changes or
performance history and not just on how close the
predicted velocities are to the limiting values. When
a well shows continued clean up (Wells (A) and (B)),
then the risk to open up the well toward the velocity
limits or to higher operating pressures may be lower.
However, when a well is showing impairment (Well
(C)), then the relative cost and benefits of opening up
the well should be addressed. Generally, wells with
history of impairment should not be opened up until
the mechanism or cause of impairment is identified
with additional data or diagnostics.
o Knowing precisely the maximum rate limits may not
be the most critical item (as this may be beyond the
capacity of the well system). Instead, it is more
critical to establish a real time feedback approach
through effective surveillance to assess the
performance of the well before reaching peak rate.
Limitations. The method is indeed simple and it is not
perfect. We need to recognize its limitations that it is
based on average velocities and average completion
pressure drop (PTA limitations). Therefore, applications
of this method to multi-layered reservoirs or completions
with high potential of flow localizations should be
performed cautiously. In these cases, more detailed near

well bore modeling would be warranted. Lastly,

important down-hole data for surveillance are the flow
velocity and pressure distribution along the sand control
completion interval. Therefore, developments of direct
and continuous pressure and flow measurements along
the sand control completion are encouraged.
Design Tool. Finally, this model has also been utilized to
affect the design and forecast of completions. For
instance, in a FP gas well planned to have a small annular
clearance, the perforation gun choice was based upon
modeled perforation inner diameter relationships with
screen velocities. Pre-production assumptions must be
made for skin and effective perforation density for the
completion types. Once build-ups are available, the
completion strategy and the peak production forecast
rates can be reviewed and reconciled.

Conclusions
The following conclusions may be drawn based on our study:
A simple surveillance method for sand control
completions can be developed using both completion
pressure drop and down-hole flowing velocities to
monitor well performance.
For cased-hole gravel packs, two dominant, and rate
dependent, failure modes were (a) destabilization of the
annular pack and (b) screen erosion. These two failure
modes were shown to be complementary and should be
used for each completion. The maximum velocities that
define these failures were: Vc = 10 ft/sec and Vs = 1 ft/sec
for annular destabilization and screen erosion failure,
respectively.
Using only pressure buildup data, consistent definition of
completion pressure drop, and the assumption that gravelfilled perforation dominates the flow in cased-hole gravel
packs, the down-hole flowing velocities can be easily
calculated from pskin-mechanical. This relationship is not
linear.
The reliability of the maximum velocity criteria and the
feasibility of the surveillance method were demonstrated
using data from 45 sand control completions.
It is recommended that during the ramp-up period, in
reaching the peak production rate for the first time,
several PTAs be taken, at increasing flow rates, to assess
and diagnose performance changes with the completions.
Wells with a history of impairment, or operating close to
maximum velocity limits or operating with high
completion pressure drop should be monitored carefully
and identified for further evaluations before opening up to
higher rates.
Acknowledgment
The authors wish to thank the management of Shell for their
permission to publish this paper. Special acknowledgment is
extended to D. A. Cole for laying down the foundation that
most of this work is based on, to D. E. Carpenter and A. R.
Melancon for their critical review of this paper and analyses
of the field examples, and to Victor Dunayevsky for
developing solutions of the flow in the annular pack. Finally,

SPE 84497

we would like to thank our colleagues in Shell E&P Company

who have contributed the field data used in this study.
References
1.

2.

3.

4.

5.
6.
7.
8.
9.
10.

11.
12.

13.
14.
15.

16.
17.
18.

Wong, G. K., Fors, R. R., Casassa, J. S., and Hite, R. H.:

Design, Execution, and Evaluation of Frac and Pack (F&P)
Treatments in Unconsolidated Sand Formations in the Gulf of
Mexico, paper SPE 26563 presented at the 68th Annual
Technical Conference and Exhibition, Houston, TX, Oct. 3-6.
McLeod, H. O., Jr. and Minarovic, M. J.: Monitoring and
Analysis of Gravel-Packing Procedures to Explain Well
Performance, paper SPE 27356 presented at the SPE Intl.
Symposium on Formation Damage Control, Lafayette, LA,
Feb. 9-10.
McLeod, H. O., Jr. and Crawford, H. R.: Gravel Packing for
High Rate Completions, paper SPE 11008 presented at the
57th Annual Technical Conference and Exhibition, New
Orleans, LA, Sept. 26-29.
Crouch, E. C. and Pack, K. J.: Systems Analysis Use for the
Design and Evaluation of High-Rte Gas Wells, paper SPE
9424 presented at the 55th Annual Fall Technical Conference,
Dallas, TX, Sept. 21-24.
Matthews, C. S. and Russell, D. G.: Pressure Buildup and
Flow Tests in Wells, Monograph Volume 1, Society of
Petroleum Engineers of AIME, New York, 1967, Dallas.
Van Everdingen, A. F.: The Skin Effect and Its Influence on
the Productive Capacity of a Well, Trans., AIME (1953) 198,
171-176.
Hurst, William: Establishment of the Skin Effect and Its
Impediment to Fluid Flow Into a Well Bore, Pet. Eng. (Oct.
1953) B-6 through B-16.
Earlougher Jr., Robert C.: Advances in Well Test Analysis,
SPE Monograph Volume 5 (1977).
Brons, F. and Marting, V.E.: The Effect of Restricted Fluid
Entry on Well Productivity, J. Pet. Tech. (Feb. 1961) 172174; Trans., AIME, 222.
Cinco, Heber, Samaniego-V., F., and Dominguez-A., N.:
Transient Pressure Behavior for a Well With a FiniteConductivity Vertical Fracture, paper SPE 6014 presented at
the SPE-AIME 51st Annual Fall Technical Conference and
Exhibition, New Orleans, Oct. 3-6, 1976.
Cinco, H., Miller, F. G., and Ramey, H. J.: Unsteady-State
Pressure Distribution Created by a Directionally Drilled Well,
Trans., AIME (1975) Vol. 259, pp. 1392-1400.
Penberthy, W. L., Jr. and Cope, B. J.: Design and
Productivity of Gravel Packed Completions, paper SPE 8428
presented at the 53th Annual Fall Technical Conference, Soc.
Pet Eng. AIME, Las Vegas, NV, Sept. 23-26.
Procyk, A., Whitlock, M. and Ali, S.: Plugging-Induced
Screen Erosion Difficult to Prevent, Oil & Gas Journal, July
20, 1998.
Bennett, C. and Svedeman, S.: Sand Control Screen Erosion
Industry Joint Project, Final Report, SwRI Project 04-8560,
May 1998.
Yildiz, T. and Langlinais, J. P.: Calculation of Pressure
Losses Across Gravel Packs, paper SPE 17167 presented at
the SPE Formation Damage Control Symposium, Bakersfield,
CA, Feb. 8-9, 1988
Hurst, William: Advances in Petroleum Engineering,
PennWell Publishing Company (1981) Tulsa, Oklahoma
Muskat, Morris: Physical Principles of Oil Production,
McGraw-Hill Book Company, Inc. (1949), New York, NY.
Geerstma, J.: Estimating the Coefficient of Inertial Resistance
in Fluid Flow Through Porous Media, Soc. of Pet. Eng. J.,
(Oct. 1974) 14(5):445-450.

Nomenclature
B = Formation volume factor, r-bbl/s-bbl for oil wells, and rbbl/sMcf for gas wells.
k = Formation permeability, md
q = Surface production rate, bpd for oil wells and Mcf/d for
gas wells.
h = True stratigraphic thickness of the reservoir pay, ft.
= Down-hole reservoir fluid viscosity, cP
= Down-hole reservoir fluid density, lb/ft3
g = Kinetic energy coefficient (for nonDarcy flow through
porous media) or beta factor for the gravel, 1/ft.
kg = Gravel permeability in the perforation tunnel, md
Lg = Length of the perforation tunnel, defined as the
difference between well bore radius and radius to casing
ID, inches.
Ap = Average cross-section area of individual perforation, in2
hp = Measured length of perforated interval, ft
t = Time in hours
= Formation porosity, fractions
Ct = Reservoir compressibility, 1/psi
PreBU
Rate
(bpd)

Shutin
Time
(hrs)

PTA
Mech
Skin

Adj
p
Skin
Mech
(psi)

2.7
175
1 10,800 5.60
17,747
0.84
3.9
424
2
21,276
1.14
4.0
505
3
3.5
421
4 21,003 2.48
28,693
1.29
4.0
655
5
Table 1 Well A PTA Results

q (bpd)
pskin-mechanical (psi)
Vc (ft/sec)
Vs (ft/sec)

Vc
(ft/s)

Vs
(ft/s)

Tot
Compl
etin
p
(psi)

Target
Peak
Rate
(bpd)

3.0

0.24

504

26,990

4.8

0.38

948

27,921

5.2

0.42

1145

Outflow

4.8

0.39

1116

Outflow

6.0

0.48

1638

Outflow

15,000
700
8
0.6

Table 2 Well C Operating Values

20,000
1,200
11
0.9

10

SPE 84497

SEPCo GOM - Cased Hole GP at Peak Rate

Cement Sheath

FP-Oil

FP-Gas

HRWP-Oil

HRWP-Gas

SC Failures

3500

Casing

3000

Vc
dP skin-mechanical (psi)

Vs
Annulus
Pack

0.5" Annulus Pack

f7

2500

Compaction
Failure

2000

Installation
Failures

1500

f6
f3

1000

f2

f5

f4

500

f8
f1

Screen

Perforation
Tunnel

10

12

14

16

18

20

22

Vc, Avg Velocity Exiting Perf @ Casing ID (ft/sec)

Fig. 4 Predicted casing velocity at peak rates for 45 wells.

Fig. 1 Completion components and flow schematics for

cased-hole gravel packs

SEPCo GOM - Cased Hole GP at Peak Rate

FP-Oil

FP-Gas

HRWP-Oil

HRWP-Gas

SC Failures

3500

f7
3000

10,000

75

60

6,000

45

4,000

30

Gamma Ray (GAPI)

Cumulative Rate (bpd)

"Hot" Spot
8,000

dP skin - mechanical (psi)

Production Log Data

Perforation: 17,200' to 17340'

0.5" Annulus Pack

2500

Covered by Vc

2000

Installation
Failures

1500

Compaction
Failure

4.7" casing - not

covered by Vc

f6

f3

1000

f5
f2

500

f4
f1
0.0

Shunt Tubes

f8

0
0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Vs, Avg Velocity @ Screen (ft/sec)

2,000

0
17180

15

17200

17220

17240

17260

17280

17300

17320

17340

Fig. 5 Predicted screen velocity at peak rates for 45

wells.

0
17360

Measured Depth (ft)

Fig. 2 Production log used to estimate the maximum

flowing velocity limit.
6 Gauge Wire-Wrap - Erosion Life
(failure = 0.001 inch slot opening)

Life to Failure (year)

1.0E+02

1.0E+01

1.0E+00

1.0E-01

Fig. 6 PTA pressure data match for Point (1) of Well A

1.0E-02
0.1

1.0

10.0

Vs, Screen Velocity (ft/sec)

(lb/1000 bbls) 50

(lb/1000 bbls) 10

(lb/1000 bbls) 1

Fig. 3 Wire-wrap screen erosion modeling results.

SPE 84497

11

Well C

Well A
30000

10.00

27000

9.00

Well is cleaning
up with

Rate (Bopd)

6.00

getting

15000

5.00

impaired with
1

production.

4.00

9000

3.00

6000

2.00

3000

1.00

0.00
1000

100

200

300

400

500

600

700

800

900

Measured Data dP

Casing Vel

Vcasing

20000

Rate (Bopd)

7.00

6.00

15000

5.00

12000

4.00
1

3.00

6000

2.00

3000

1.00

0
100

200

300

400

500

600

700

800

900

0.00
1000

P Skin Mech (psi)

dP Skin Mech

Measured Data dP

Screen Vel

Casing Vel

Vcasing

Vscreen

Fig. 8 Diagnostic plot for Well A

10.00

27000

9.00

24000

8.00

21000

7.00

18000

6.00
6

15000

5.00

4.00
4

9000

3.00

6000
1

2.00

3000

1.00

0
0

100

200

300

400

500

600

700

800

0.00
1000

900

P Skin Mech (psi)

dP Skin Mech

Measured Data DP

Screen Vel

Fig. 9 Diagnostic plot for Well B

Casing Vel

Vc

Vs

Casing & Screen Vel (ft/sec)

Rate (Bopd)

Well B
30000

12000

Well Failed
shortly after

8.0

4.0

200

400

600

800

1000

1200

Fig. 10 Diagnostic plot for Well C

Casing & Screen Vel (ft/sec)

8.00

12.0

9.00

24000

9000

10.00
5

27000

5000

Well A

18000

16.0

5
2

15000

10000

Fig. 7 A proposed diagnostic plot for well surveillance.

21000

Cleaning
up

Vscreen

30000

Vs

20.0

Impairing

Dp Skin Mech (psi)

Screen Vel

Vc

Peak Rate

P Skin Mech (psi)

dP Skin Mech

Casing Vel

1400

1600

1800

0.0
2000

Vc and Vs (ft/sec)

Well is

18000

Screen Vel

24.0

25000

Casing & Screen Vel (ft/sec)

7.00

12000

Measured Data DP

8.00

production.

21000

Rate (Bpd)

24000

dP Skin Mech
30000