AC 2007-987: INNOVATIVE FLUID MECHANICS EXPERIMENTS FOR MODERN

MECHANICAL ENGINEERING PROGRAM

Afshin Goharzadeh, The Petroleum Institute

AFSHIN GOHARZADEH, Ph.D., is an assistant professor of mechanical engineering at the
petroleum Institute. An expert in experimental fluid mechanics, Dr. Goharzadeh obtained his
Ph.D. at the University of Le Havre in France (2001). After his Ph.D. he joined the prestigious
Max Planck Institute for Marine Microbiology in Bremen (Germany) as scientific researcher. He
characterized experimentally the flow at water-sediment interface. Using the Particle Image
Velocimetry (PIV) and Refractive Index Matching (RIM), he measured the velocity of fluid
particles at fluid-porous interface and estimated the thickness of the transition layer (Brinkman
Layer Thickness) inside the transparent porous media.

Peter Rodgers, The Petroleum Institute

PETER RODGERS, Ph.D., is an associate professor of mechanical engineering at the petroleum
Institute. He was formally with the CALCE Electronic Products and Systems Center, University
of Maryland; Nokia Research Center, Finland; and Electronics Thermal Management, Ltd.,
Ireland. In these positions he has been actively involved in electronics cooling and reliability. He
has authored or co-authored over 50 journal and conference publications on a broad range of
topics in these areas. He has been an invited lecturer, keynote speaker, panelist, and session chair
at international conferences. Dr. Rodgers received the 1999 Harvey Rosten Memorial Award for
his publications on the application of computational fluid dynamics analysis to electronics
thermal design. He is a member of several international conference program committees, and is
program co-chair for EuroSimE 2007.

Christian Mandel, The Petroleum Institute

CHRISTIAN MANDEL received a bachelor’s degree in mechanical engineering from the
University of Applied Sciences Mannheim, Germany, in April 2006. He is presently working as a
Graduate Research Assistant at the Petroleum Institute, with his master’s thesis work focused on
the experimental characterization of multiphase flow phenomena.

Page 12.895.1

© American Society for Engineering Education, 2007

 Innovative Fluid Mechanics Experiments for Modern
Mechanical Engineering Program

Abstract
One of the primary objectives of the Petroleum Institute (PI) is to prepare future
mechanical engineers to assume successful career paths in the Oil and Gas industry.
With this in mind, a state of the art core measurement laboratory represents an important
facility for teaching use within a Mechanical Engineering Program. Such a facility has
been developed at the PI, and is utilized at undergraduate level through a one semester
dedicated course entitled Core Measurements. As part of this course, two innovative
hands-on experiments related to experimental fluid dynamics are given, with the
objective of familiarizing the students, through simple projects, on how to characterize
fundamental fluid flow phenomena. A hands-on project consists of design, fabrication,
data acquisition and validation of a simple experiment.
In this paper an overview of two hands-on fluid mechanics based experiments are
presented, with the specific educational objectives sought given. The first experiment
investigates the dynamics of a jet flow inside a cylindrical enclosure, with the second
experiment characterizing a single phase flow over a backward facing step. State of art
measurement is used with flow visualization undertaken using Particle Image
Velocimetry (PIV). Sample results are presented including the teaching strategy
employed.

1.0 Introduction
The Petroleum Institute’s (PI) undergraduate mechanical engineering program offers
specializations in automation and control, energy and thermofluids, and mechanics and
materials, all focusing on their applications to the oil and gas industries. Upon
graduation, PI students work for the Institute’s sponsoring oil and gas companies as field
test engineers, laboratory test engineers, design engineers, development engineers,
project management engineers, and research engineers. To prepare the students with the
ability to undertake such a variety of engineering tasks, laboratory and hands-on training
has been found as an effective tool in helping to develop the necessary skills.
The Core Measurements course [1], which is offered in the spring semester of the
junior year, focuses on the principle subjects of mechanical engineering such as fluid
mechanics, heat transfer, thermodynamics and combustion, mechanics and materials, and
automation and control. The prerequisite for the course is the Basic measurements course
[1], which focuses on basic measurement and instrumentation techniques and is taught in
sophomore year. Core measurements is a two–credit course, having four contact hours
per week. Students can expect to do eight standardized experiments per semester, with
one to two experiments in each of the four specialty areas of fluid mechanics, heat
transfer, mechanics, and automation and control. Report preparation and presentation
skill requirements are emphasized throughout course. Approximately one hour per week
is devoted to lecturing and three hours for conducting laboratory experiments. Students
are provided with a laboratory manual that provides basic information about the
experiments and the relevant theory involved. In addition to these experiments, students
Page 12.895.2

are assigned a hands-on experiment. The objective of the hands-on experiments is to

prepare the students to design and fabricate a simple experiment to permit investigation
of a fundamental engineering phenomenon. More importantly, this hands-on assignment
instills into them the needs for time-management and efficiency, integration of
fundamental theories with the hands-on experience in the laboratory and effective project
management skills to ensure successful completion of the project. Each group of students
(three to four) are assigned a hands-on experiment to design, build and test over five
weeks period. On completion a report is submitted documenting the work undertaken
and an oral presentation is given. The hands-on project represents 30% of the final
course grade. Each hands-on project has a combined capital/consumables budget of
$500.
In the oil and gas industries, the characterization of fluid flows is of importance at
every stage of hydrocarbon production, from the extraction of crude oil to the fabrication
of derivative products. The PI core measurement laboratory is equipped with modern
measurement sensors to permit such characterization, and includes pressure transducers,
magnetic flow meters, viscometers, high speed CCD and infrared cameras. The
laboratory also has the capability to undertake experimental characterization of fluid flow
using non destructive laser diagnostics [2]. Flow measurement using a laser based
system allows engineers to obtain significant information about the dynamics of the flow
in complex geometries. Different types of laser diagnostics, such as laser doppler
anemometry (LDA), particle imaging velocimetry (PIV), and planar laser-induced
fluorescence (PLIF), are used to determine local velocity, 2D velocity, and concentration
measurements, respectively [3-5]. The primary advantage of particle image velocimetry
(PIV) is its ability to characterize fluid flow without complicated analysis. It is a modern
and powerful method for investigating fluid flow. Using a standard PIV system, two
dimensional flow velocity data can be obtained for a specified cross section inside the
system.
The objective of this paper is to present two fluid mechanics experiments
implemented in the PI Core Measurements course that provided students with hands-on
experience on the characterization of fundamental fluid flow phenomena. After a brief
review of the basic principles of PIV, the actual hands-on experiments are presented. The
first one deals with the jet flow characterization inside a cylindrical enclosure, with the
concept of a Reynolds number for a jet flow and the vortex flow presented. The second
experiment is on the visualization of fluid flow over a classical backward facing step.
The concept of separated flow region and recirculation vortex flow is studied. Both
hands-on experiments are undertaken for fundamental geometries, which enables well-
defined boundary conditions for corresponding predictive analysis. Each experiment
contains a section focused the teaching strategy employed by the instructor.

2.0 Particle Image Velocimetry Measurement Principles

Particle image velocimetry (PIV) is based on the average velocity measurement of the
trajectory of the particle tracers [3-5], with displacement measurement made during a
short period of time. The acquisition method relies on image analysis. To obtain spatial
information about the displacement of particle tracers, a synchronizer simultaneously
controls the laser pulse time and the recording separation time of the CCD camera. The
laser source (diode laser n=532 nm) generates a pulsed sheet, which illuminates a cross
Page 12.895.3

section of the fluid flow. The working fluid is water and contains seeded tracers which
diffuse the laser light in all directions. White polymer powder available in the cosmetic
industry (for instance Polyamide tracers) is used as particle tracers. The shape of polyamide
tracers is spherical, and the diameter is on order 20 micrometers. The CCD camera with
1024 x 1024 pixels, placed perpendicularly to the laser sheet, records one image. This
process is repeated a second time. The separation time t is defined as the time between two
recorded images for the CCD camera and the two laser pulses. The exposure laser pulse
time, ts, represents the lifetime of one laser pulse. Finally, the synchronizer controls the
triggering time in order to record several pairs of images. In figure 1 a sketch is presented
illustrating the typical measurement set-up of a PIV system.

Figure 1. Typical measurement set-up of a Particle Image Velocimetry (PIV) system.

Once images are recorded, post-processing PIV software is used to calculate the
statistical displacement of particle tracers between a pair of recorded images. Further
details are given in [2] on the operation of PIV system used in this study, which is the
FlowMaster model from Lavision GmbH, Germany.
Due to PI safety requirements, students were not expected to set-up the PIV system.
Instead prior to the measurement session, the PIV system was set-up by the instructor.
However students undertook the PIV measurement process themselves by focusing the
CCD camera on the illuminated plane and performing data acquisition. The students
were taught the fundamental principles of PIV prior to the measurement session.

3.0 Jet flow Experiment

When a fluid flows from a pipe into a tank, a head loss is produced, with the kinetic
energy of the fluid from the pipe dissipated through viscous effects as the stream mixes
with the fluid of the tank. The interaction of such turbulent flows occurs in many fluid
mechanic based problems. This experiment was designed for the characterization of flow
fields generated using a round water jet that impinges with a quiescent water reservoir
confined within a cylindrical enclosure.
Page 12.895.4
3.1 Experimental set-up
A simple schematic representation of the experimental set-up is shown in figure 2.
The flow is generated in a circular pipe from which a round jet emanates. The working
fluid is water and using a pump, the water is sucked from a small clearance of width
0.005 m and re-injected through the circular pipe. The jet diameter, D, is 0.0065 m and
the cylinder inner diameter is 0.09 m. The cylinder and circular plates are made of
transparent acrylic material (Plexiglass). Height, H, represents the height between the
circular plate and the base of the circular cylinder.
Flow separation disc
Jet diameter

Flow
meter

Jet Flow clearance

H
Pump

a) Side View b) Top View

Figure 2. Schematic diagram for vertical jet flow characterization.

The entire experiment is placed in a rectangular Plexiglass tank, figure 3, with the
external enclosure filled with Dow Corning silicon oil (DC 550) in order to avoid
curvature issues during the flow visualization. The fluid flow is obtained using a laser
sheet adjusted in vertical direction and CCD camera perpendicular to laser light, figure 3.
A field of view of 0.5 x 0.4 m2 is obtained.

Flow meter
External square
CCD camera enclosure

Silicone oil

Laser Internal cylindrical

enclosure (water)

Figure 3. PIV experimental set-up for vertical jet flow characterization.

PIV images and its corresponding 2D velocity field are shown in figure 4, as a
function of enclosure disk height, H. As can be seen in figure 4(c), the fluid exits the
pipe and initially forms a jet-type structure as it enters the enclosure. The flow rate inside
the pipe is measured using a flowmeter, Q = 2 l/min, which corresponds to a mean jet
Page 12.895.5

velocity, Ujet, of 1 m/s.

 Jet details:

Jet Qjet = 3.33x10-5 m3/s

H Ujet = 1 m/s
Reynolds number = 6612
Measurement field

(a) Flow field region under analysis.

0 0.1 m/s

-10

y (mm)
-20

-30

-40

0 20 40
x (mm)

(b) Typical flow field CCD image, H = 40 mm. (c) Measured velocity field, H = 40 mm.

0 0
0.1 m/s
0.1 m/s

-20

-20
y (mm)

y (mm)

-40

-40
-60

-80
-60

0 20 40 -40 -20 0 20
x (mm) x (mm)
(d) Measured velocity field, H = 60 mm. (e) Measured velocity field, H = 80 mm.
Figure 4. PIV velocity flow field measurements as function of enclosure disk height, H.
Page 12.895.6
 Due to the small height of the enclosure, H=0.4 m, the jet splits up at the enclosure
base and creates a two symmetric vertical vortex. The flow field is steady and the
velocity field corresponds to an average of 100 images recorded over 20 second period.
Using dimensionless analysis the corresponding Reynolds number, Re, is calculated
using the mean inlet jet velocity, Ujet:
U jet D jet
Re ? ? 6612
p water
where Djet = jet diameter and p = kinematic viscosity.

3.2 Teaching strategy

Figure 4 represents the final measurement results obtained by the students on
completion of the hands-on experiment after five weeks work. The instructor uses this
data for teaching several key concepts of the fluid mechanics observed. Students were
asked to measure the location of the vortex center and study the dynamic of this center.
As the flow was steady the center of the vortex remains stable for the given Reynolds
number Re = 6612 and system heights of H = 0.4 m, 0.6 m, and 0.8 m.
For higher jet Reynolds number, the vortex was found to become unstable and a
transitional flow could be shown for teaching the characteristics for the onset of turbulent
flow.
The height of the system influences the structure of the vortex observed. Figure 4(c)
represents the fluid flow structure for H = 0.4 m. The trajectories of particles are
represented using streamlines which show one stable vortex in the system. For higher
height H = 0.6 m, figure 4(d) shows a new fluid structure particularly near horizontal disc
at the top of the image. In this region the trajectories of particles are in opposite direction
than the trajectories of particles inside the main vortex flow. Once the height is increased
to 0.8 m, figure 4(e), two contra-rotating vortices develop. The main vortex is stable at
the container base and the second vortex with smaller diameter is rotating in opposite
direction near the top part of the container. All figures show a stable and symmetrical fluid
flow for the corresponding Reynolds number.
In terms of teaching important features of the PIV system, measurement set-up was
assessed. Once the velocity is calculated, students were requested to focus on the
physical meaning of their results. In practice, every pair of recorded images gives a
velocity field, and the students have to explain if their results were physically correct.
For instance the detection of the particles depends on the separation time ( t) between
images. If t is very small the camera cannot detect a displacement of the particle tracers
between the two images; conversely, if t is very large the camera loses the trajectory of
particles. Thus it is very important to adjust t according to the actual velocity of the
particles inside the flow. The jet velocity at the entrance of the tank (Ujet = 1m/s) is 100
times higher than velocity of the particles inside the vortex (Umax = 0.01 m/s). Therefore
it is difficult to obtain both velocities with the same separation time t. As observed in
figure 4, the velocity at the entrance of the jet appears smaller than inside the tank. This
measurement error in the image was considered by students when they analyzed the flow
Page 12.895.7

fields in their report.

4.0 Backward Facing Step Experiment
Good benchmark tests for assessing computational fluid dynamics (CFD) predictive
accuracy should exhibit complex thermofluid phenomena that can be accurately
characterized by experimental analysis. A number of fundamental benchmark test cases
exist for which CFD codes, or more specifically turbulence models and numerical
methods, are commonly evaluated against. Typical examples include flow over a blunt
plate or around an obstacle, backward facing step, and channels flows [6-11].
The flow over a backward facing step has long served as a test case for such CFD
predictive accuracy assessment, for both steady and unsteady separated flows [7, 12-16].
The characteristic separated shear layer is subjected to a sharp transition and large
velocity gradients and is an indicative test for turbulence closure models therefore. In
addition, the phenomenon of flow separation is a problem of great importance for
fundamental and industrial reasons. For instance it often corresponds to drastic losses in
aerodynamic performances of airfoils or automotive vehicles. The backward facing step
is one of the simplest geometries to study this phenomenon.
A backward facing step is a flow in a straight plane channel with a sudden expansion
at a step edge. In figure 5 a typical flow over a backward facing step geometry is
sketched. The boundary layer built up on the upstream channel wall separates at the step
edge and becomes a free shear layer. The downstream location where the shear layer
reattaches to the channel wall is the reattachment point. The distance from the step edge
to this point is called the reattachment length, LR, and it is a characteristic parameter for
this type of flow. A recirculation zone exists in the region of the flow confined by the
separated shear layer and the lower channel wall. The reattachment length depends on
the expansion ratio, which is the ratio of down stream to upstream channel heights, the
Reynolds number and the upstream flow conditions amongst others [7]. As found from
numerous experimental and numerical studies [7, 12-16], the separated shear layer gives
rise to instabilities when the flow is in the transitional regime.

u Flow direction

H
Recirculating
h zone

LR

Figure 5. Schematic illustration of typical flow features over a backward facing step.
Page 12.895.8
4.2 Experimental set-up
A simple schematic representation of the experimental set-up used is shown in figure
6. It functions as a closed loop water tunnel, comprising of a backward facing step test
section, pump, flow meter and a flow valve for controlling flow rate, hence channel
Reynolds number. The characteristics Reynolds number for the backward facing step
flow is defined as:
U inlet H
Re H ?
p water
where Uinlet = the average velocity in the upstream step section, H = the downstream
channel height, and p = kinematic viscosity.

Test Section:
Backward facing step

Pump
Flow meter Flow valve

(a) Schematic representation of the experimental set-up is shown

40
Flow direction 60
20
80
300 800

Note: All dimensions are in mm.

(b) Test section: backward facing step geometry.
Figure 6. Schematic diagram for backward facing step test facility.

The working section for backward facing test facility is made of transparent acrylic
material, with the working fluid being water. Figure 7 shows the actual test facility
constructed, which included a flow straightener section at the test facility entrance to
produce uniform flow velocity profile at inlet.
The flow field was measured using PIV, with the laser sheet adjusted in the vertical
central plane of the test facility and CCD camera perpendicular to laser light as shown in
figure 3. The actual measurement set-up is shown in Figure 8, with a field of view of
0.065 x 0.05 m2 obtained. In figure 9 PIV measured flow fields are presented as
streamline flow plots.
Page 12.895.9
 Flow direction

Flow straightener Step

(a) Test section: backward facing step.

Porous media

(i) Stream wise direction – inlet flow. (ii) Perforated plate detail.
(b) Flow straightener construction in (a).
Figure 7. Backward facing step test section construction.

CCD Camera

Laser sheet
Test facility
flow

Laser
Page 12.895.10

Figure 8. PIV experimental set-up for flow characterization over a backward facing step.
 30

20

10

y (mm)
Measurement field
0

Step -10 Step

-20
-10 0 10 20 30 40 50
x (mm)
(a) Flow field region under analysis. (b) Typical flow field CCD image.
30 30 0.1 m/s
0.1 m/s

20 20

10 10
y (mm)

0 y (mm) 0

-10
Step -10 Step

-20 -20
-10 0 10 20 30 40 50 -10 0 10 20 30 40 50
x (mm) x (mm)
(c) Measured velocity field, ReH = 500. (d) Measured velocity field, ReH = 1800.
30 30
0.1 m/s 0.1 m/s

20 20

10 10
y (mm)
y (mm)

0 0

-10 Step -10 Step

-20 -20
-10 0 10 20 30 40 50 0 10 20 30 40 50
x (mm) x (mm)
(e) Measured velocity field, ReH = 3500 (f) Measured velocity field, ReH = 6000.
Figure 9. PIV velocity flow field measurements over a backward facing step having an
expansion ratio of 3/2.

4.2 Teaching strategy

The students reviewed the literature for publications relating to backward facing step,
from which one geometry was selected for construction, namely; expansion ratio of 3/2.
The PIV flow field measurements presented in figure 9 are for the Reynolds number
range 500 to 6000, which encompass laminar to transitional turbulent flow regimes.
Page 12.895.11
 With increasing Reynolds number it is observed that the mean reattachment location,
LR, of the free shear layer emanating from the step increases. At a Reynolds number of
500, figure 9(c) the laminar flow characteristics observed are similar in nature to that
previously described for Figure 5. The flow field is steady and the velocity field
corresponds to an average of 100 images recorded over 20 second period. At the higher
Reynolds numbers, the flow was observed to be quasi steady-state, with velocity fields
presented in figures 9(d) to (f) corresponding to an average of five images recorded over
a one second period. Distinct flow characteristics such as both the formation of down
stream recirculation zones past the reattachment point and the detachment of upstream
secondary recirculation zones behind the reattachment point are observed in figures 9(d)
and 9(e) respectively. These features compare qualitatively with the findings commonly
reported by other researchers [7, 12-16] and are discussed by the students in their project
report. As previously outlined in section 3.2, the importance of the PIV system
measurement set-up was also assessed by the students for this test case.

5.0 Conclusions
Two innovative hands-on experiments related to experimental fluid dynamics were
presented, with the objective of teaching students on how to characterize fundamental
fluid flow phenomena through easy to construct experiments, and the use of state of the
art fluid flow measurement techniques. Sample results were presented, with the teaching
strategy employed discussed. By participating in such projects, students learned how to
integrate fluid mechanic fundamentals with hands-on experience in the laboratory, as
well as acquiring project management skills. Future work, through a CFD course, will
permit the students to utilize the measurement presented in this paper as data for
assessing the predictive accuracy of corresponding numerical predictions.

References
[1] Ohadi, M., Sheu, M., and Molki, A., 2005, “An Undergraduate Instructional Laboratory Model for a
Modern Mechanical Engineering Program,” The American Society for Engineering Education (ASEE)
Portland, Oregon, USA.
[2] Goharzadeh, A., Molki, A., Ohadi, M., 2006, “A Proposed Particle Image Velocimetry (PIV) System
for Instructional Purposes in a Modern Mechanical Engineering Undergraduate Laboratory Program,”
The American Society for Engineering Education (ASEE) Chicago, Illinois, USA.
[3]. Adrian, R. J., Durao, D.F.G., Heitor, M.V., Tropea, C., Maeda, M. and Whitelaw, J.H., 2000, Laser
Techniques for Fluid Mechanics. Springer Verlag.
[4] Albrecht H., Tropea C., Borys M. and Damaschke N., 2003, Laser Doppler and Phase Doppler
Measurement Techniques. Springer Verlag.
[5] Stanislas, M., Westerweel, J., Kompenhans, J., 2003, Particle Image Velocimetry: Recent
Improvements, Springer.
[6] Ota, T., and Kon, N., 1980, “Turbulent Heat Transfer of Momentum and Heat in a Separated and
Reattached Flow over a Blunt Flat Plate”, Transactions of the ASME, Journal of Heat Transfer, Vol.
102, pp. 749-754.
[7] Fokkelman, W.F., Meinders, E.R., and Hanjalic, K. 1997, “The Backward-Facing Step Flow in the
Transition Regime,” Proceeding of EUROTHERM Seminar 58: Thermal Management of Electronic
Systems III, Editors, Baradon, J.P., Beyne, E., Saulnier, J.B., Elsevier, Paris, pp 37-43.
[8] Abe, K., Kondoh, T., and Nagano, Y., 1994, “A New Turbulence Model for Predicting Fluid Flow and
Page 12.895.12

Heat Transfer in Separating and Reattaching Flows – I. Flow Field Calculations,” International
Journal of Heat and Mass Transfer, Vol. 37, No.1, pp. 139-151.
 [9] Abe, K., Kondoh, T., and Nagano, Y., 1995, “A New Turbulence Model for Predicting Fluid Flow and
Heat Transfer in Separating and Reattaching Flows – II. Thermal Field Calculations,” International
Journal of Heat and Mass Transfer, Vol. 38, No.8, pp. 1467-1481.
[10] Freitas, C.J., 1995, “Selected Benchmarking from Commercial CFD codes,” Transaction of ASME
Journal of Fluids Engineering, Vol. 117, pp. 208-218.
[11] Cokljat, D., Kim, S.E., Iaccarino, G., and Durbin, P.A., 2003, “A Comperative Assessment of the
V2F Model for Recirculating Flows,” AIAA-2003-0765.
[12] Armaly, B.F., Durst, F., Pereira, J.C.F., Schönung B., 1983, “Experimental and theoretical investigation of
backward-facing step flow,” J. Fluid. Mech. Vol. 127, pp. 473-496.
[13] Kaiktsis, L., Karniadakis, G.E., Orszag, S.A., 1991, “Onset of the three-dimensionality, equilibria,
and early transition in flow over a backwardfacing step,” J. Fluid. Mech. Vol. 231, pp. 501–528.
[14] Kaiktsis, L., Karniadakis, G.E., Orszag, S.A., 1996, “Unsteadiness and convective instabilities in
two-dimensional flow over a backward-facing step,” J. Fluid. Mech. Vol. 321, pp. 157–187.
[15] Kim, J., Moin, P., 1985, “Application of a fractional-step method to incompressible Navier–Stokes
equations,” J. Comput. Phys. Vol. 59, pp. 308–323.
[16] Lesieur, M., Begou, P., Briand, E., Danet, A., Delcayre, F., Aider, J.L., 2003, “Coherent vortex
dynamics in large-eddy simulations of turbulence,” J. Turbulence Vol. 4, pp. 016.

Page 12.895.13