Hydrodynamic Flow Separation Control through Vortex Generators

Jinsong Xu, Jie Xie, Cheng Yu

State Key Laboratory of Ocean Engineering, Shanghai J iao Tong University
Shanghai, P.R.China






ABSTRACT

Flow separation is a major cause of the unexpected drag increase in
marine structures. The experimental study of hydrodynamic flow
separation control through vortex generators (VGs) was conducted in
Shanghai J iao Tong University. Vortex generators have been widely
used in aircraft industry to prevent or delay flow separation over wings.
The function of VGs is to introduce vortices into the flow ahead of the
separation point to energize the boundary layer and thus to delay the
flow separation. The application results on a surface-piercing cylinder
and an AHTS ship model demonstrate the VGs effectiveness in
resistance improvement of marine structures.

KEY WORDS: Flow separation control; vortex generators;
resistance improvement; marine structures.

INTRODUCTION

Flow separation accompanied with the large energy losses is a major
cause of the unexpected drag increase in marine structures. The
effective control of the flow separation remains important for the
resistance improvement of the marine structures.

Various flow control methods have been investigated and studied in the
past decades. Vortex generators (VGs), first introduced by Taylor
(1947), have been widely used in aircraft industry to prevent or delay
flow separation over wings. Conventional VGs consist of a row of
small plates that project normal to the surface and are set an angle of
incidence to free-stream. The function of VGs is to introduce vortices
into the flow ahead of the separation point to energize the boundary
layer and thus to delay the flow separation. The simplicity and the
effectiveness of VGs have attracted a great deal of attention in the
aerospace community. They seem to be a logical choice as a potential
method to control the hydrodynamic flow separation on the marine
structures.

Compared with the airfoil operation in the air, the marine structures
usually operate on the water surface. The existing free surface
complicates the flow separation problem (Chow, 1967; Stern et al.,
1989). Current research mainly focus on the physical mechanism of
wave-induced separation (Metcalf et al., 2006), but very few on flow
control and resistance improvement. The motivation of this research is
to explore the possible application of VGs on the marine structures to
improve the resistance performance.

METHOD FOR FLOW SEPARATION CONTROL

Flow Separation Detection

The effective flow separation control requires the accurate detection of
the separation region. Theoretically, the separation is indicated by the
vanishing of the skin friction on the surface. Achenbach (1968) studied
the skin friction and static pressure distribution of the circular cylinder
in cross-flow. The obtained results shown in Fig. 1 reveal that the zero
skin friction position corresponds to the constant pressure distribution.
The flow separation region is thus indicated by the constant pressure
feature, and the separation line could be determined as the initial
location of the constant pressure region. This detection criterion has
been widely used in the separation study of the airfoils (Bragg and
Gregorek, 1987).



Fig. 1. Circular cylinder: skin friction and pressure distribution. Re=10
5

Proceedings of the Twentieth (2010) International Offshore and Polar Engineering Conference
Beijing, China, J une 2025, 2010
Copyright 2010 by The International Society of Offshore and Polar Engineers (ISOPE)
ISBN 978-1-880653-77-7 (Set); ISSN 1098-6189 (Set); www.isope.org

1075


In some cases, the complete pressure distribution is not available. The
abrupt change of the single-point pressure value is used instead as
indication of the flow separation occurrence (Patel et al., 2002). But the
exact separation line could not be easily detected from single-point
pressure change. Similarly, the tuft method is also helpful for direct
observation of flow separation.

Vortex Generators for Flow Control

Flow separation control, by means of passive devices such as VGs, is
the less expensive and the quickest solution to implement (Godard,
2006). VGs have been used to control from low-speed separated flows
in adverse pressure gradient to transonic shock-induced separation.
From the airfoil application, the most usual and effective devices are
the vane-type counter-rotating micro vortex generators as suggested by
Lin (1991) and illustrated in Fig. 2. These VGs could produce the
longitudinal vortices downstream and induce momentum transfer
between the free-stream and the near wall region.



Fig. 2. Lins counter-rotating vortex generators

The above counter-rotating VGs consist of an array of thin plates fixed
perpendicularly to the surface and skewed at an angle
pd
to the main
flow direction. The geometric design of the height h, the length l, the
distance L and should be based on the boundary layer thickness
at separation location. The value Xvg is the streamwise distance
between the VGs and the separation line. From the comparative studies
in airfoil application, Lin (1999) concluded that VGs of h/=0.2
placed at 5h to 10h upstream of the separation line were the most
effective devices. Obviously, the VGs design and deployment must be
based on the accurate detection of the separation line and the estimation
of the boundary layer thickness at separation location.

EXPERIMENTAL SET-UP

Surface-piercing Cylinder Towing Tests

The twoing tests were conducted in Ship Model Towing Tank at
Shanghai J iao Tong University, which has inside dimensions of 110m
in length, 6m in width, and 3m in depth. The towing carriage over the
tank has a maximum speed of 6m/s.

The test cylinder is a GRP tube with a length of 1m and a diameter of
300mm. It was attached to the towing carriage through the
dynamometer without any motion freedom. The cylinder was set at a
draft of 700mm piercing through the water surface. It was instrumented
with 15 static pressure probes around the cylinder circumference below
the water surface 350mm. The probe distribution starts from the
stagnation point and ends at =168 with an equal step of 12. The
general test arrangement is illustrated in Fig.3.

During the calm water towing tests, the cylinder resistance and the
static pressure distribution were measured electronically. The tuft
method was employed for flow visualization, and the tuft pattern was
recorded using underwater camera.



Fig. 3. Illustration of surface-piercing cylinder

AHTS Ship Model Towing Tests

The offshore vessels possess some special characteristics such as
smaller L/B ratio and larger propeller diameter, which usually result in
the severe flow separation problem. A GRP model of Anchor Handling
Tug / Supply Vessel (AHTS) was used for flow control study.

The test AHTS model has an overall length of 4.08m, a breadth of
0.91m, a draft of 0.34m, and a volume of 0.82m
3
. From the aft-body
lines plan shown in Fig. 4, the ship stern shape shows a very abrupt
change due to the large propeller diameter. Totally 16 static pressure
probes were distributed along the stern bilge in two parallel rows, and
numbered consecutively as shown in Fig. 5.

During the clam water towing tests, the AHTS model was attached to
the towing carriage through the dynamometer. The model resistance
and the static pressure distribution were measured electronically. The
tuft method was employed for flow visualization, and the tuft pattern
was recorded using underwater camera.



Fig. 4. Aft-body lines plan of the test AHTS model

1076



Fig. 5. Distribution of the static pressure probes around stern bilge

RESULTS AND DISCUSSIONS

Surface-piercing Cylinder Flow Control

The clean cylinder without the VGs was first tested to detect the flow
separation region. The towing speed V was set to 1.7m/s, and the
corresponding Re number is 3.610
5
. In Fig. 6, the static pressure
distribution is plotted as coefficient C
p
versus the periphery angle .
The pressure coefficient C
p
is defined as

2
1
2
p
P P
C
V

= (1)

where P is the local static pressure, and P is the ambient static
pressure. From the initial location of the constant pressure region, the
separation line is thus determined as =120. From the recorded tuft
pattern shown in Fig. 7, the fluctuating yarns after =122validates
the above separation detection.



Fig. 6. Static pressure distribution at V =1.7m/s



Fig. 7. Tuft pattern at V =1.7m/s

Based on the separation line location, the boundary layer thickness
and the VGs geometrical dimensions illustrated in Fig. 2 were
determined and listed in Table 1. The fabricated thin plate VGs were
glued on the cylinder surface as shown in Fig. 8.

Table 1. Geometrical dimensions of cylinder VG (mm)

H Xvg l L pd
8.9 2.5 131 5 6.25 15 18




Fig. 8. The deployment of the VGs on cylinder

The resistance values of the clean cylinder and the cylinder with VGs
are plotted together in Fig. 9, and listed in Table 2. At the design speed
V =1.7m/s, the VGs produce 2.5% resistance reduction. At the high-
speed point V =2.1m/s, the resistance reduction is 5.3%. In the low-
speed range 0.7~0.9m/s, the reduction rate is up to 23%. Obviously, the
VGs have achieved the remarkable improvement of the cylinder
resistance.

In Fig. 10~13, the comparison results of static pressure distribution are
plotted at several speed points. The VGs could not prevent the flow
separation, but generally delay the flow separation and increase the
pressure values in the separation region. The flow control objective is
achieved for the surface-piercing cylinder.

1077



Fig. 9. The cylinder resistance curves with and without VGs

Table 2. The cylinder resistance values with and without VGs

V
(m/s)
R (kgf)
no VGs
R (kgf)
with VGs
Reduction
(%)
2.1 29.91 28.32 5.3
2.0 28.72 27.61 3.9
1.9 27.48 26.52 3.5
1.8 25.85 24.96 3.5
1.7 23.02 22.44 2.5
1.6 20.32 19.80 2.6
1.5 17.71 17.16 3.1
1.4 15.01 14.62 2.6
1.3 12.47 12.25 1.8
1.2 10.24 10.29 -0.5
1.1 8.35 8.44 -1.1
1.0 6.93 6.79 1.9
0.9 6.30 5.31 15.7
0.8 5.12 4.08 20.3
0.7 4.02 3.06 23.8



Fig. 10. Comparison of static pressure distribution at V =0.8m/s




Fig. 11. Comparison of static pressure distribution at V =1.2m/s





Fig. 12. Comparison of static pressure distribution at V =1.7m/s



Fig. 13. Comparison of static pressure distribution at V =2.1m/s


1078


AHTS Ship Model Flow Control

The towing tests of the AHTS ship model were conducted in the speed
range 0.98~2.21m/s. From the static pressure curves shown in
Fig.14~15, the pressure values along the stern bilge had an abrupt drop
in the speed range 1.7~2.1m/s. There existed violent splashing and
bubbles around stern surface as seen in Fig.16. It seems that the flow
separation occurred in the speed range 1.7~2.1m/s and the separation
line is close to the stern bilge.





Fig. 14. Static pressure curves along the stern bilge



Fig. 15. Static pressure curves away from the stern bilge



Fig. 16. The violent splashing and bubbles at V =1.84m/s

Based on the separation line location and speed value V =1.72m/s, the
boundary layer thickness and the VGs geometrical dimensions
illustrated in Fig. 2 were determined and listed in Table 3. The
fabricated thin plate VGs were glued on the AHTS model as shown in
Fig. 17.

Table 3. Geometrical dimensions of AHTS VGs (cm)

H Xvg l L pd
4.6 1.7 27.0 3.4 4.3 6.8 18



Fig. 17. The deployment of the VGs on AHTS model

The resistance values of the AHTS model with and without VGs are
plotted together in Fig. 18, and listed in Table 4. At the design speed V
=1.72m/s, the VGs produce 1.2% resistance reduction. At the speed
point V =1.60m/s, the resistance reduction is 6.18%. In the low-speed
range 0.98~1.23m/s, the reduction rate is up to 10%. In other speed
range, the model resistance increased. Obviously, the VGs are effective
for the AHTS model over limited operational ranges. This is similar to
the results of airfoil application (Patel et al., 2002).

1079



Fig. 18. The AHTS resistance curves with and without VGs

Table 4. The AHTS resistance values with and without VGs

V
(m/s)
R (kgf)
no VGs
R (kgf)
with VGs
Reduction
(%)
0.98 1.94 1.78 8.39
1.11 2.57 2.3 10.44
1.23 2.95 2.76 6.4
1.35 3.38 3.47 -2.65
1.48 4.08 4.08 0
1.6 5.29 4.96 6.18
1.72 7.61 7.52 1.21
1.84 10.76 11.03 -2.43
1.97 14.07 14.19 -0.88
2.09 16.65 17.06 -2.48
2.21 18.84 19.58 -3.93


CONCLUSIONS

Flow separation is a major cause of the unexpected drag increase in
marine structures. Flow separation control, by means of passive devices
such as VGs, is the less expensive and the quickest solution to
implement. The application results on a surface-piercing cylinder and
an AHTS ship model demonstrate the VGs effectiveness in flow control
and resistance improvement of marine structures.

The existing free surface complicates the flow separation problem on
marine structures. The optimal VGs dimensions obtained from airfoil
study can no longer apply effectively to marine structures. Some
fundamental research is essential to develop the best VGs design for
marine application.

Flow separation control is strongly associated with the value range of
Re number. Current results on model scale can not be easily
extrapolated to the full scale cases. The actual application in full scale
is necessary to further validate the VGs effectiveness on marine
structures.



REFERENCES

Achenbach, E (1968). Distribution of local pressure and skin friction
around a circular cylinder in cross-flow up to Re =5X10
6
, Journal of
Fluid Mechanics, Vol. 34, part 4, pp. 625-639.
Bragg, MB, Gregorek, GM (1987). Experimental Study of Airfoil
Performance with Passive Vortex Generators, Journal of Aircraft, Vol.
24, No.5, pp. 305-309.
Chow, SK (1967). Free-surface effects on boundary layer separation on
vertical struts, the University of Iowa, Iowa City, IA.
Godard, G, Stanislas, M (2006). Control of a decelerating boundary layer
Part 1: Optimization of Passive Vortex Generators, Aerospace Science
and Technology, Vol. 10, pp. 181191.
Lin, JC, Selby, GV, Howard, FG (1991). Exploratory study of vortex-
generating devices for turbulent flow separation control, 29th
Aerospace Sciences Meeting, Reno, Nevada, AIAA Paper 91-0042.
Lin, JC (1999). Control of turbulent boundary layer separation using micro
vortex generators, AIAA Paper 99-3404,.
Metcalf, B, Longo, J, Ghosh, S, Stern, F (2006). Unsteady Free-surface
wave-induced boundary-layer separation for a surface-piercing NACA
0024 foil: Towing tank experiments, Journal of Fluids and Structures,
Vol. 22, pp. 7798.
Patel, MP, Carver, R, Lisy, FJ et al. (2002). Detection and Control of Flow
Separation Using Pressure Sensors and Micro-Passive Vortex
Generators, AIAA 2002-0268.
Stern, F, Hwang, WS, Jaw, SY (1989). Effects of Waves on the Boundary
Layer of a Surface-Piercing Flat Plate: Experiment and Theory, Journal
of Ship Research, Vol. 33, No.1, pp. 63-80.
Taylor, HD (1947). The Elimination of Diffuser Separation by Vortex
Generators, United Aircraft Corporation Report No. R-4012-3.



1080