Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 15.5.1 - Black Powder Erosion in Sales Gas Pipeline Bends
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00380
 Material Information
Title: 15.5.1 - Black Powder Erosion in Sales Gas Pipeline Bends Particle-Laden Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Elsaadawy, E.
Sherik, A.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: solid particle erosion
black powder
particle-laden flow
dispersed phase model
 Notes
Abstract: Computational Fluid Dynamics (CFD) modeling of dilute solid-gas flow through a 90 degree sales gas pipeline bend is performed and presented in this paper. The solid particles used mimic the size and distribution of those of the Black Powder (BP) found in Saudi Aramco sales gas pipelines. BP is the name given to the solid grayish material that is found inside most, if not all, sales gas pipelines worldwide. Because of its hardness and jagged features, BP can cause a serious threat to the integrity of gas pipelines. An experimental program was developed to determine the erosion resistance of the bend material so the erosion rate can be computed however; the details of that experimental program and the corresponding results are not the subject of the current paper and hence are not presented. The focus of the paper is to present the computational prediction of the erosion rate in sales gas pipeline bends under different gas flow rates and bend turn radii along with flow field variables for both phases namely, the gas and the BP particles. The pipeline bend under study has a pipe diameter of 36 inch with ratios of bend radius of curvature to pipeline diameter of 1.5 and 3.0. The gas flow rate values considered result in pipeline average velocities of 5, 10, and 20 m/s. The software Fluent 6.3 was used to solve the Reynolds Averaged Navier-Stokes (RANS) equations and predict the erosion rate. The continuous phase, gas, was modeled using the Eulerian approach while the dispersed phase, BP particles, was modeled using the Lagrangian approach which is based on the calculation of the trajectory of several individual solid particles through the flow field. The motion of the tracked particles is taken to describe the average behavior of the dispersed phase. To account for the influence of turbulent fluid fluctuations on particle motion, the stochastic tracking Discrete Random Walk model is used. The particles effect on the gas flow is neglected i.e., one-way coupling. The Dispersed Phase Model (DPM) of Fluent is used as the particles loading ratio is very small, typically less than 0.01. Five simultaneous particles injections, representing actual field conditions, were released from the inlet of the pipeline and to be tracked. The particles diameters are 2, 4, 7.5, 12.5, and 20 micron. To resolve the flow turbulence, the standard k-e turbulence model with enhanced wall functions is used. The removal of wall material due to erosion (erosion rate) is calculated using the Finnie model developed for ductile materials and implemented into Fluent. The mathematical model was validated against published experimental data. The validation process showed a good agreement between the model and the experimental data. The profiles of both mean velocity and fluctuations, of the gas phase and the particles along with the corresponding erosion rates are presented. The results show that a larger radius of curvature bend has more uniformly distributed erosion rate over it surface area. However, the results show that BP has negligible erosive effects even in a pipeline bend with a radius of curvature to pipe diameter ratio of 1.5. The current study is the first in the literature to investigate BP erosive effects on pipeline bends.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00380
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1551-Elsaadawy-ICMF2010.pdf

Full Text



7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010



Black Powder Erosion in Sales Gas Pipeline Bends


E. Elsaadawy and A. Sherik


Research & Development Center, ARAMCO Oil Company, Dhahran 31311, KSA



Keywords: solid particle erosion, black powder, particle-laden flow, Dispersed Phase Model


Abstract
Computational Fluid Dynamics (CFD) modeling of dilute solid-gas flow through a 90 degree sales gas pipeline bend is
performed and presented in this paper. The solid particles used mimic the size and distribution of those of the Black Powder (BP)
found in Saudi Aramco sales gas pipelines. BP is the name given to the solid grayish material that is found inside most, if not all,
sales gas pipelines worldwide. Because of its hardness and jagged features, BP can cause a serious threat to the integrity of gas
pipelines. An experimental program was developed to determine the erosion resistance of the bend material so the erosion rate
can be computed however, the details of that experimental program and the corresponding results are not the subject of the
current paper and hence are not presented. The focus of the paper is to present the computational prediction of the erosion rate in
sales gas pipeline bends under different gas flow rates and bend turn radii along with flow field variables for both phases namely,
the gas and the BP particles. The pipeline bend under study has a pipe diameter of 36 inch with ratios of bend radius of curvature
to pipeline diameter of 1.5 and 3.0. The gas flow rate values considered result in pipeline average velocities of 5, 10, and 20 m/s.
The software Fluent 6.3 was used to solve the Reynolds Averaged Navier-Stokes (RANS) equations and predict the erosion rate.
The continuous phase, gas, was modeled using the Eulerian approach while the dispersed phase, BP particles, was modeled using
the Lagrangian approach which is based on the calculation of the trajectory of several individual solid particles through the flow
field. The motion of the tracked particles is taken to describe the average behavior of the dispersed phase. To account for the
influence of turbulent fluid fluctuations on particle motion, the stochastic tracking Discrete Random Walk model is used. The
particles effect on the gas flow is neglected i.e., one-way coupling. The Dispersed Phase Model (DPM) of Fluent is used as the
particles loading ratio is very small, typically less than 0.01. Five simultaneous particles injections, representing actual field
conditions, were released from the inlet of the pipeline and to be tracked. The particles diameters are 2, 4, 7.5, 12.5, and 20
micron. To resolve the flow turbulence, the standard k-e turbulence model with enhanced wall functions is used. The removal of
wall material due to erosion (erosion rate) is calculated using the Finnie model developed for ductile materials and implemented
into Fluent. The mathematical model was validated against published experimental data. The validation process showed a good
agreement between the model and the experimental data. The profiles of both mean velocity and fluctuations, of the gas phase
and the particles along with the corresponding erosion rates are presented. The results show that a larger radius of curvature bend
has more uniformly distributed erosion rate over it surface area. However, the results show that BP has negligible erosive effects
even in a pipeline bend with a radius of curvature to pipe diameter ratio of 1.5. The current study is the first in the literature to
investigate BP erosive effects on pipeline bends.


Introduction
In the Oil & Gas industry, Black Powder (BP) is the catchall name that is used to describe the grayish material found inside most,
if not all, sales gas pipelines worldwide. Black powder can be found in several forms, such as wet with a tar-like appearance or
dry in the form of a very fine powder [1-5]. It is composed of different forms of iron sulfide (FeS), iron oxides (Fe304, FeOOH)
and iron carbonate (FeCO3), mechanically mixed or chemically combined with any number of contaminants, such as salts, sand,
liquid hydrocarbons, metal debris, and even Naturally-Occurring Radioactive Material (NORM) [2].
Black Powder once exists and moving with the flow, can represent a serious threat to the integrity of gas pipelines by eroding
compressor components and pipeline control valves, plugging metering instrumentation and filters, and reducing the accuracy of
the in-line inspection. Also, BP could have major adverse effects on customers by contaminating the customers' sales gas supply
leading to interruptions of the customers' operations and/or poor quality of products in which the sales gas is used as feedstock
[3]. Although BP erosion itself was not the subject of any research work the authors are aware of, many research work was
devoted to the solid particle erosion of pipelines and pipeline bends. The main drive of studying solid particle erosion in pipelines
and pipeline bends was not Black Powder erosion rather, it was pneumatic conveying systems which are applied widely in
modern industries which are processing solid materials. However, there is a couple of works that addressed the BP in a way.
These are the work by Smart [6] and Smart and Winters [7] where they determined the required fluid velocity to entrain, carry
away, black powder in liquid and gas pipelines respectively. In these two studies, it was concluded that the velocity required to
move black powder particles in gas pipelines is independent of particle size and ranges from 10.4 fps to 13.6 fps for 8 inch and 30







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


inch pipelines respectively while, in liquid pipelines, water, velocity required depends on the equivalent particle size up to a size
of about 5 mils after which it depends only on the pipe diameter.
On the other hand, many researchers addressed the features of gas-solid pipe flow affecting transportation have been extensively
studied. These include the fluid turbulence, the lift force due to the shear flow, the Magnus effect on rotating particles, particle
interactions, particle-wall collisions, particle size and shape, pipe roughness and gravitational settling. The effect of the drag
coefficient and inlet conditions (inlet velocity profile) of solid particles on the particles tracks calculations in vertical and
horizontal ducts are studied by Kuan and Schwarz [8] using the commercial CFD package CFX-4.4. They found that the drag
coefficient needs to be reduced by as much as 35% of the standard value to achieve good agreement with the corresponding
experimental data in case of a vertical channel flow. On the other hand, for a horizontal channel flow it needs to be reduced only
20% to achieve similar agreement. Regarding the velocity inlet conditions, Kuan and Schwarz reported that the vertical turbulent
flow seems to be insensitive to the inlet conditions while for a horizontal flow it is found to be strongly dependent on inlet
conditions.
Kuan et al. [9] performed CFD simulations of a dilute particulate turbulent flow in a 90o duct bend with a radius of curvature
equals 1.5 duct hydraulic diameter. As in the previous work of Kuan and Schwarz [8], they performed the simulations using CFX
4.4 where they used the Differential Reynolds Stress Model (DRSM) with fully-developed inlet conditions to solve the turbulent
flow in the bend and also used the same test facility to produce the experimental data used in validating the simulations. In a
following work, Kuan [10] used different solid size distributions rather than a single uniform particle size and also made use of a
modified shear-slip lift force formula which is consistent with experimental data for 0.18 < Rep < 8.0. From these studies, [8 -
8], it was concluded that the DRSM did not capture the correct pressure gradient effects within the bend. Also it was found that
even the finer particles (66 micron) experienced a gas-solid segregation due to the centrifugal effect and this was characterized by
a local drop in particle concentration near the inner wall and it is well reflected in predictions where the averaged velocity
profiles discontinued in the locality. The experimental part of the study [10] is reported in more detail in Yang and Kuan [ll1].
Although the CFD based erosion modeling can be applied to predict erosion in many complex geometries, its accuracy has be
questionable for some applications. To assess the viability and accuracy of the CFD based erosion modeling, Zhang et al.[12]
performed a comparison between computed and measured particle velocities and erosion in both water and air flows in a direct
impact test section. Also, erosion models other than the one that is built into Fluent were used and examined by implementing
them into Fluent using a user defined function (UDF). It was found that for a sand/water flow in a direct impact test, not like
sand/air, the particle impact velocity is much lower than the velocity of the slurry jet and varies in a wide range. Also, they found
that among the erosion models tested, the Erosion-Corrosion Research Center erosion model and Oka et al. erosion model [13,
14] are more accurate within the frame of the work. In FLUENT 6, the near-wall effect is not taken into account when computing
the particle tracking and the particles are handled the same way as those in regions away from the wall. Also, in FLUENT 6, and
some other codes, the particles are assumed of zero volume when deciding whether a particle hits the wall. In an effort to enhance
the current CFD-based erosion modeling, Zhang et al. [1] introduced into the CFD code FLUENT 6 a couple of modifications
using user defined functions. The first is including the standard wall-function effect when a particle is moving in the near-wall
region and the second is rebound the particle at a particle radius from the wall. It was found that FLUENT 6 built-in erosion
model predicts accurately the erosion due to rounded sand particles of large size, 256 pm. On the other hand, for 25 pm sharp
sand particles it over predicts the erosion by a factor of 17-78 which is reduced by a factor of 2 when applying the standard wall-
function effect in the near-wall region particle tracking computations.
In order to evaluate the performance of elbows and plugged tees geometries under erosive service conditions and using the
experimental data of Enayet et al. [15] to validate the simulation results, Edwards et al. [16] developed a procedure to predict
erosion in standard elbows, long radius elbows, and plugged tees. This procedure is implemented into the CFD code CFX 4.2.
Chen et al. [17] also studied, computationally and experimentally, the relative erosion severity between plugged tees and elbows
under dilute gas/1iquid-solid flow conditions. From the study it was shown that the relative erosion severity is greatly affected by
the type of the carrier fluid (liquid or gas) properties as simulations showed that for water-sand flows in plugged tees cause more
erosion than in elbows while for air-sand flows erosion in plugged tees is found to be two orders of magnitude less than erosion
in standard elbows. In another study by Hengshuan and Zhong [18] it was shown that the longer the radius of curvature of the
elbow the less the erosion it experiences due to solid particle impact.
Not only the radius of curvature that affects the erosion but also the elbow (bend) orientation was found to have a large effect on
particle motion and hence erosion rates as shown in the study by Keating and Nesic [19].
Sample of the BP collected from Saudi Aramco sales gas system were analyzed and the characterized. Based on that study, it was
found that Magnetite (Fe304) pOWder is considered a reasonable substitute of the BP for erosion studies as it has similar
mechanical properties. Erosion resistance of different materials under the erosion effects of Magnetite was determined
experimentally. The details and the results of that study is not the subject of the current paper and will be published elsewhere.
However, in the current study, flow and erosion of Magnetite powder in pipeline bends are studied.



Geometry & Flow Conditions Studied
The pipeline bend under study is a standard 90 degree bend that has a 36 inch pipe diameter, D. The ratios studied of the bend
radius of curvature, R, to the pipeline diameter are 1.5 and 3.0. The bend has a 240 inch inlet horizontal pipe an 80 inch vertical
outlet pipe. The gas phase used in the current study is a typical Saudi Aramco natural gas at operating conditions with a mass
























dp (O m) 7ih, (gis)
2 0.143
4 0.372
7.5 1.25
12.5 0.476
20 0.153


Table 2: Reynolds and particle Stokes numbers of the cases studied

d, (Om) Re = py7gD/P~ St, = ppd vB/18pLD
5 m/s 10 m/s 20 m/s 5 m/s 10 m/s 20 m/s
2 2.13E+07 4.26E+07 8.53E+07 4.71E-04 9.41E-04 1.88E-03
4 2.13E+07 4.26E+07 8.53E+07 1.88E-03 3.76E-03 7.53E-03
7.5 2.13E+07 4.26E+07 8.53E+07 6.62E-03 1.32E-02 2.65E-02
12.5 2.13E+07 4.26E+07 8.53E+07 1.84E-02 3.68E-02 7.35E-02
20 2.13E+07 4.26E+07 8.53E+07 4.71E-02 9.41E-02 1.88E-01



Mathematical Model
The commercial CFD software Fluent 6.3 was used to solve the Reynolds Averaged Navier-Stokes (RANS) equations and predict
the erosion rates. The particles volumetric loading ratio is very small, typically less than 10% therefore the flow was dealt with as
dilute gas-solid multiphase flow and hence the Dispersed Phase Model (DPM) of Fluent was used. The continuous phase, gas,
was modeled using the Eulerian approach while the dispersed phase, particles, was modeled using the Lagrangian approach. The
Lagrangian approach is based on the calculation of the trajectory of several individual solid particles through the flow field. The
motion of the tracked particles is used to describe the average behavior of the dispersed phase.
The governing equations of the continuous phase are the incompressible continuity and momentum equations as shown below.


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


density, p, of 61.8 kg/nl3 and a dynamic viscosity, yL, of 1.33xl~ Pa.s. The gas flow rates considered resulted in bulk gas
velocities, v,, of 5, 10, and 20 m/s. The Black Powder (Magnetite) has a mass density, pp, of 5150 kg/m3 and was released at a
rate of 2.4 g/s into the stream. Particles sizes dp, and corresponding mass flow rates are summarized in

Table 1. The corresponding Reynolds numbers and particle Stokes numbers are summarized in Table 2.



Table 1: Particle sizes and the corresponding mass flow rates


a pD)+ V C~)- (00 -p+ V. p +V0 .+ V0+ p


To resolve the flow turbulence, the standard k-e turbulence model with enhanced wall functions is used. The equations of
turbulent kinetic energy, k, and turbulent dissipation, e, are as follows.


+ V.(6 k )


V.i~ p + V + Gk PE







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010



Whee k= -u'2+ U2 r = 2vt e'e, the Reynolds stress tensor e'r ,the turbulent viscosity
k2 ivx+ ux:

At,= pck2,and Gk Pt ~U.VV T_2~V.v(3jrV. v+pk)
The standard values of the (empirical) constants in the k-s model are:
c, = 0.09, 0-k = 1.0 0-, = 1.30 cs, = 1.44 and Ce2 = 1.92

While setting up the Lagrangian tracking and erosion model, the following assumptions have been made.
* Only the influence of turbulent fluid fluctuations on particle motion was considered using the stochastic tracking Discrete
Random Walk model. The particle-particle interactions are neglected and any change of the flow turbulence caused by the
particles is not accounted for i.e., one-way coupling.
* Non-reacting and non-fragmentation particles are considered.
* The geometry modifications, caused by the removal of wall by the solid particles, have been neglected. This means that the
computational model geometry during simulation was invariable.

In order to obtain a reasonable statistical distribution and to reduce scatter in erosion predictions, a large number of particles are
normally required to perform the particle tracking. Each particle is tracked through the flow domain separately, and the particle-
wall interaction information is then recorded and used to calculate the erosion.
The particle trajectory is determined by integrating the force balance on the particle. This force balance equates the particle
inertia with the forces acting on the particle, Newton's second law. This equation can be written as





where mp is the particle mass, Opis the particle velocity vector, Pis an external force acting on the particle. The forces acting on a
particle can be the drag force, the buoyancy (gravitational settling) force, the pressure gradient force, the added mass force,
Brownian diffusion (motion), Saffman lift force, and Basset force, and rotating reference frame force. For small particles with
density ratio pp/p much greater than one only the drag force and the gravitational settling force will impact the trajectory of the
particle while other forces will be negligibly small. Along a particle trajectory, the equation of motion, to be integrated, can be
reduced to the following form.


d o, =I 1 g ( p -~o p)
dt p

The dragl forceP nper unit m~las can be epvressedl as~_-/l( / FD2 /pd ) 3 CDe/24) whre~y Rep is the particle1 Reynolds~
numberRep = (pdp |v, v|)/p and CDis the drag coefficient.
Although equation (6) is linear, the fluid velocity along the particle trajectory must be known in order to solve it. As the velocity
depends on the particle path itself, the general solution in even a simple turbulent flow is not possible.

During particle trajectory calculation, the particle-wall interaction information such as impact speed, impact angle, and impact
location as well as impact intensity is stored. This information is then applied to the appropriate erosion equation(s) to compute
the erosion. In Fluent 6.3, the removal of wall material due to erosion (erosion rate) is calculated using a model based the Finnie
model developed for ductile materials and implemented. The erosion model used is given in the following form [20].


R -= i rhz. C(d, ). g(ax). vbA 7
p=1 Arace

where Rerosion is the erosion rate given in units of mass of target material removed per unit area per unit time, itjl is the particles
mass flow rates, and Aface is the area of the cell face at the wall. The functions C(dp) and g(a) must be specified in consistent
units to build a dimensionless group with the relative particle velocity and its exponent.
In ANSYS Fluent model Eq. (7), g (u) is an empirical polynomial function of the particle impact angle, which allows for
simulation of ductile erosive systems (i.e. particles that impact the surface at a shallow angle will cause a higher erosion rates
than particles that impact the wall at higher angles).
In the current study, five simultaneous particles injections, representing actual field conditions, as given in







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


Table 1, were released from the inlet of the pipeline to be transiently tracked. The particles were released at a velocity equal that
of the gas and with a uniform distribution as the as well.
In order to take into account the effect of turbulence velocity fluctuations on particles trajectories, the stochastic Discrete
Random Walk (DRW) model of Fluent was used.


Validation of the Mathematical Model
The published experimental data of Yang and Kuan 1 1] was used to validate the mathematical model. Yang and Kuan studied a
dilute gas-solid flow through a curved 900 duct bend. The curved bend is squared-section (15 cm x 15 cm) and has a radius of
curvature, R of 1.5 times the duct hydraulic diameter, D, (22.5 cm). The bend has a 3.5 m horizontal inlet duct and a 1.8 m
vertical straight duct. Gas phase, air, measurements were obtained using a Laser Doppler Anemometer (LDA) at a bulk gas
velocity vB, of 10 m/s in the absence of solid phase. The solid phase which is glass spheres with an average diameter of 66 Om
was released into the flow from a fluidized bed. Solids/gas mass loading ratio reached is well below 1% so as to setup a dilute
gas-solid flow regime. For more details about the experimental setup used and the procedures, refer to reference [ll].
The case was solved using Fluent 6.3 and a user defined function (UDF) was created for post processing of particles parameters
specifically, sampling the particles velocities at any section of the domain. A mesh independent solution was achieved on a 9x105
nodal points mesh.
Figure 1 shows the velocity contours of the gas phase on the synmletry plane of the duct while Figure 2 presents the particles
tracks, colored by velocity magnitude, of particles released at the inlet. Mean velocity profiles along the synmletry plane and at
different angles from the inlet plane are shown in Error! Reference source not found.. The model captures the gas velocities
fairly good particularly in the vicinity of the walls where interaction with particles takes place and erosion develops. Only at 750
from the inlet and into the bend where the predicted profile does not capture the right slope, shear stress, at the inner wall of the
bend. At this location, flow is decelerating due to the adverse pressure gradient and flow will start to separate and a secondary
flow is formed at the outlet on the inner wall as can be seen from the velocity contours in Figure 1.
The user defined function (UDF) used for post processing calculates an average velocity magnitude for particles in a domain and
stores this average in a User Defined Memory (UDM) for post processing afterwards. As a particle passes through a cell the UDF
is called and calculates the velocity magnitude and appends it to an average for that cell which the particle is leaving. Therefore,
the more the particles the better and more meaningful the average is and consequently in regions with low particles concentration
the average can be unrealistic. In post processing, the author excluded those values where the particle concentration were very
low as can be seen in Error! Reference source not found.. However, one can still see that the model reasonably captured the
particles velocity magnitudes which are used in erosion predictions.


Results & Discussion
For a pipeline bend of R/D = 1.5, the velocity vectors along the vertical synmletry plane are presented in Figure 5. The left to
right flow of the bend has a gas bulk velocity of 20 m/s. One can see from the velocity contours, Figure 5 that flow accelerates
close to the inner wall as it enters the pipeline bend due to the favorable pressure gradient at that part of the wall while it
decelerates at the outer wall due to the adverse pressure gradient. This behavior starts to reverse after the bend midpoint i.e. at a
45 deg plane. Refer to Figure 6 for the locations of different profiles and contours.
Although the inner wall of the bend encounters higher velocity stream, it is the outer wall of the bend that encounters high
erosion rate due to the impact angles of particles.
The gas phase velocity contours at different planes of a R/D = 3.0 bend are shown in Figure 7 and the corresponding velocity
profiles, at 15, 30, 45, 60, 75 deg., are presented in Figure 8. Similar to the sharper bend, R/D = 1.5, flow accelerates at the
inner wall up to 45 deg. And then starts to decelerate while the opposite takes place at the outer wall. For particles mean velocity
profiles, shown in Figure 9, show similar trend and for a closer inspection, a comparison between gas and particles velocity
profiles at two different cross sections of the pipeline bend (30 and 60 degrees from inlet plane) is presented in Figure 10. The
figure shows that particles follow the gas with negligible slip except at the 60 deg. plane where they deviate noticeably close to
the outer wall of the bend. The gas decelerates at that wall due to the adverse pressure gradient while particles lag in response and
continue to move with higher velocities.
The erosion contours as predicted by Fluent for a pipeline bend of R/D = 1.5 are shown in Figure 11 where it can be seen that
the maximum erosion takes place at the midpoint, along the synmletrIy plane, of the bend the location where velocity profiles
starts reverse behavior. Using a longer radius of curvature pipeline bend redistributes the erosion rates more uniformly over the
outer wall area of the bend as compared to the shorter radius of curvature bend as can be seen from the comparison of the ero sion
rate contours of two bends R/D = 1.5 and R/D = 3.0 as shown in Figure 12. However, the erosion rates for black powder are
found to be negligibly small for the conditions considered in the current study.










7th International Conference on Multiphase Flow,

ICMF 2010, Tampa, FL, May 30 June 4, 2010


1 40e401
1 33e401
1 26e+01
1 19e+01
1 12e+01
1 05e401
9 B2e4OD
9 12e400
B 41e400
7 71e+OD
701e+0D
6 31e+OD
561e400
4 91e+0D
4 21e4OD
351e+0D
2 80e+00
2 10e+0D

D D0e+OD


1 169401


1 12e401
I e
9 B2eDO
9 12ei

771e400



4 91ei
4 21e400


1351ei
2 10eX00

0 0e400


m-


Figure 1: Gas velocity contours in the symmetry
plane.


Figure 2: Particles tracks colored by velocity
magnitude.


0.7



0.6








0.4



0.3



0.2



0.1


0.4



0.3



0.2



0.1


C= 15'

-eMeasured --Predicted


C= 45'

-eMeasured --Predicted


(L= 75'

-e-easured --Predicted


Figure 3: Gas velocity profiles inside the bend.










7th International Conference on Multiphase Flow,

ICMF 2010, Tampa, FL, May 30 June 4, 2010









0.9



0.8



0.7



0.6




S0.5



0.4



0.3



0.2



0.1








S= 75'

-eMeasured --Predicted


uL= 15*

+-Measured --Predicted


L= 45'

-eMeasured --Predicted


Figure 4: Particles velocity profiles inside the bend.


25701





204eal



1729101

162e*01
151001




774e+00
668000
ses1co


| 00'Exi


t


(P
Enhance


Figure 6: Bend cross sectional geometry, vertiacl
symmetry plane


Figure 5: Velocity vectors in the symmetry plane, R/D=1.5







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


Figure 7: Contours of velocity magnitudes at different
planes along the pipeline bend, R/D=3


U/Ub


Figure 8: Gas phase velocity profiles at different planes.


00 8LLL L L~~
0.0 0.2 04 0.6 08


00 0.2 04 0.6 08 10 12i 14
Up/Db


10 1.2 14


Figure 9: Solid phase (particles) velocity profiles at
different


Figure 10: Comparison between gas and particles velocity
profiles at 30 and 60 degrees planes


trJT






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


Figure 11: Erosion contours on the wall of the standard bend


;j


R/D = 3.0


R/D = 1.5


Figure 12: Erosion rate contours for pipeline bends (R/D= 1.5 & 3.0)






Concluding Remarks
Dilute solid-gas flow modeling using the Discrete Phase Model of Fluent 6.3 was performed for pipelined bends of R/D = 1.5
and R/D = 3.0 under typical field operating conditions. The turbulence effects on the trajectories of the particles are accounted
for by means of the stochastic DRW model of Fluent while particle-particle interaction and particle effect on the continuous
phase (gas) are neglected. The standard k-e turbulence model with enhanced wall functions was used. The main focus in the







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


current work is on the mean velocities of both gas phase and solid phase and the erosion rate. From the results one can conclude
that although the erosion rate is negligibly small under the conditions studied, a pipeline bend with longer radius of curvature will
have a more uniformly distributed erosion over the area of the bend which makes it a better candidate for applications with
erosive type of service such as sale gas pipelines loaded with Black Powder.
The mathematical model and erosion treatment in the current study can be extended to predict the erosion in globe control valves
mounted on sales gas pipelines and suffering form BP erosion damage.



Acknowledgment
The authors would like to extend their appreciation to Dr. Benny Kuan of the CSIRO, Australia for providing the experimental
data used in validating the mathematical model used in the current study. The authors would like to acknowledge the Saudi
Arabian Oil Company (Saudi Aramco) for the support and granting permission to present and publish this work.



References
1. Sherik, A.M., Black Powder-1i: Study examines sources, makeup in dry gas systems ,Oil & Gas Joumnal, Vol. 106,
No. 30, August 11, 2008.
2. Sherik, A.M., Black Powder--Conclusion: Management requires multiple approaches ,Oil & Gas Joumnal, Vol. 106,
No. 31, August 18, 2008.
3. Tsochatzidis, N. A., Study addresses black powder' s effects on metering equipment, Oil & Gas Joumnal, Vol. 106, No.
12, March 24, 2008.
4. Tsochatzidis, N. A., and Maroulis, K.E., Methods help remove black powder from gas pipelines, Oil & Gas Joumnal,
Vol. 105, No. 10, March 12, 2007.
5. Baldwin, R.M., Here Are Procedures For Handling Persistent Black-Powder Contamination, Oil & Gas Joumnal, Vol.
96, No. 43, October 26, 1998.
6. Smart, J., Determining the velocity required to keep solids moving in liquid pipelines, The Pipeline Pigging and
Integrity Management Conference, Houston, Texas, February 14-15, 2007.
7. Smart, J., and Winters, R., Black powder migration in gas pipelines and associated problems, The Pipeline Pigging and
Integrity Management Conference, Houston, Texas, February 13-14, 2008.
8. Kuan, B., and M. Schwarz, P., Numerical prediction of dilute particulate flows in horizontal and vertical ducts, Third
International Conference on CFD in the Minerals and Process Industries, CSIRO, Melboume, Australia, 10-12
December 2003. pp. 135-140.
9. Kuan, B., Yang, W., and Solnordal, C., CFD simulation and experimental validation of dilute particulate turbulent
flows in a 900 duct bend, Third Intemnational Conference on CFD in the Minerals and Process Industries, CSIRO,
Melboume, Australia, 10-12 December 2003. pp.531-536.
10. Kuan, B., CFD simulation of dilute gas-solid two-phase flow with different solid size distributions in a curved 90o duct
bend, Third Intemnational Conference on CFD in the Minerals and Process Industries, ANZIAM J., Vol. 46, pp.C744-
C763, 2005.
11. Yang, W., Kuan, B., Experimental investigation of dilute turbulent particulate flow inside a curved 90. bend, J.
Chemical Engineering Science, Vol. 61, pp. 3593-3601, 2006.
12. Zhang, Y., Reuterfors, E.P., McLaury, B.S., Shirazi, S.A., and Rybicki, E.F., Comparison of computed and measured
particle velocities and erosion in water and air flows, Wear, Vol. 263, pp. 230-238, 2007.
13. Oka, Y.I., Okamura, K., and Yoshida, T., Practical estimation of erosion damage caused by solid particle impact Part 1:
Effects of impact parameters on a predictive equation, Wear, Vol. 259, pp. 95-101, 2005.
14. Oka, Y.I., Okamura, K., and Yoshida, T., Practical estimation of erosion damage caused by solid particle impact Part 2:
Mechanical properties of materials directly associated with erosion damage, Wear, Vol. 259, pp. 102-109, 2005.
15. Enayet, M.M., Gibson, M.M., Taylor, A. M. K. P., and Yianneskis, M., Laser-Doppler measurements of laminar and
turbulent flow in a pipe bend, Intemnational Joumnal of Heat and Fluid Flow, Vol. 3, No.4, pp. 213 -219, December 1982.
16. Edwards, J. K., McLaury, B.S., and Shirazi, S.A., Modeling solid particle erosion in elbows and plugged tees, Joumnal
of Energy Resources Technology, Vol. 123, pp 277-284, December 2001.
17. Chen, X., McLaury, B.S., and Shirazi, S.A., Numerical and experimental investigation of the relative erosion severity
between plugged tees and elbows in dilute gas/solid two-phase flow, Wear, Vol. 261, pp. 715-729, 2006.
18. Hengshuan, C., and Zhong, X., Numerical Analysis and experimental investigation of Erosion in Variable Rectangular-
Section Bends by Solid Particles, Chinese J. Mech. Eng., Vol. 3, pp. 111-118, 1990.
19. Keating, A., and Nesic, S., Prediction of two-phase erosion-corrosion in bends, Second Intemnational Conference on
CFD in the Minerals and Process Industries, CSIRO, Melboumne, Australia, 6-8 December 1999. pp. 229-236.
20. ANSYS Fluent 12 User's Manual, Chapter 22, Lebanon, New Hampshire, USA 2009.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs