FL43CH11-Cattafesta ARI 19 November 2010 16:20

ANNUAL
REVIEWS Further
Actuators for Active
Click here for quick links to
Annual Reviews content online, Flow Control
including:
• Other articles in this volume
• Top cited articles Louis N. Cattafesta III and Mark Sheplak
• Top downloaded articles
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

• Our comprehensive search Interdisciplinary Microsystems Group, Department of Mechanical and Aerospace Engineering,
University of Florida, Gainesville, Florida 32611; email: cattafes@ufl.edu, sheplak@ufl.edu
by Brown University on 09/05/12. For personal use only.

Annu. Rev. Fluid Mech. 2011. 43:247–72 Keywords

First published online as a Review in Advance on aerodynamic control, transducers, electromechanical, zero-net mass flux,
August 24, 2010
fluidic, plasma
The Annual Review of Fluid Mechanics is online at
fluid.annualreviews.org Abstract
This article’s doi: Actuators are transducers that convert an electrical signal to a desired phys-
10.1146/annurev-fluid-122109-160634
ical quantity. Active flow control actuators modify a flow by providing an
Copyright  c 2011 by Annual Reviews. electronically controllable disturbance. The field of active flow control has
All rights reserved
witnessed explosive growth in the variety of actuators, which is a testament
0066-4189/11/0115-0247$20.00 to both the importance and challenges associated with actuator design. This
review provides a framework for the discussion of actuator specifications,
characteristics, selection, design, and classification for aeronautical appli-
cations. Actuator fundamentals are discussed, and various popular actuator
types used in low-to-moderate speed flows are then described, including flu-
idic, moving object/surface, and plasma actuators. We attempt to highlight
the strengths and inevitable drawbacks of each and highlight potential future
research directions.

247
 FL43CH11-Cattafesta ARI 19 November 2010 16:20

1. OVERVIEW
Active flow control necessarily requires actuators to interact with the flow. This field has witnessed
Bandwidth: range of explosive growth recently because of the ubiquitous nature of fluids in engineering systems, the
frequencies over which community’s improved understanding of fluid mechanics, and the potential to dramatically im-
the output of the prove system performance using effective control strategies. However, the number of instances in
actuator is deemed
which active flow control has successfully transitioned from a laboratory prototype to a real-world
acceptable
aeronautical application is small (e.g., Pugliese & Englar 1979, Kibens et al. 1999, Nagib et al.
Control authority:
2004, Greska et al. 2005, Seiner et al. 2005, Shaw et al. 2006). This is a testament to the difficulties
application-specific
term to indicate metric associated with designing active flow control systems. This review focuses on one component
of actuator specifically: actuators.
performance; examples Given the interest in and importance of the topic, it is timely to assess the characteristics
include velocity, of the most common actuators within the context of a general design problem. Owing to space
momentum, vorticity
constraints, it is not possible to summarize all interesting research in this area. The primary
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

flux, and body force

objective is to provide a framework for assessing any flow control actuator. In an effort to achieve
Transducer: any
this, we discuss the fundamentals of actuator design and inevitable trade-offs. Dimensional analysis
device that converts a
signal from one energy is used to identify key dimensionless performance metrics. Potential classifications are summarized,
domain to another as different applications inevitably impose certain restrictions. Various technologies are compared
by Brown University on 09/05/12. For personal use only.

Feedback: process in to highlight their strengths and limitations. Finally, new design trends and possible future research
which sensor output directions are identified and discussed.
from a system provides
information used to
modify the system via
a control law and an 2. INTRODUCTION
actuator
2.1. Historical Perspective
Gad-el Hak (2000) and Williams & MacMynowski (2009) provide a brief history of flow control
and acknowledge that the modern era began with Prandtl’s early experiments on boundary layers.
Generally speaking, active control implies energy addition via an actuator (MacMynowski &
Williams 2009). At a minimum, the actuator can be turned on or off, and some operational
characteristics can be adjusted. Active flow control varies from brute-force techniques, in which
the amplitude is sufficient to modify the mean flow structure, to more efficient approaches that seek
to leverage flow instabilities using small-amplitude perturbations. As our collective understanding
of fluid mechanics has improved, the community has migrated toward small-amplitude forcing in
which it is feasible to reduce actuator power, size, and mass, for example (Theofilis 2009). But in
many cases, such as in high-speed flows, effective small-amplitude forcing remains an elusive goal
because actuators generally lack sufficient bandwidth and/or control authority. These and related
terms are defined and discussed in Section 3.1.
Active flow control is a multidisciplinary field with increased participation from researchers in
seemingly unrelated areas. Beyond fluid mechanicians, the field has attracted researchers in fields
such as applied mathematics, physics, transducers, and controls. The disparate skill sets between
researchers in these fields have historically led to communication barriers that impede progress. For
example, feedback flow control has gradually made inroads in the active flow control community
(Rowley & Batten 2009). Beyond sensor issues, feedback control imposes additional dynamic
requirements on actuators that are often ignored or underappreciated by many experimentalists
with experience in open-loop control (i.e., no feedback). Conversely, theoretical control specialists
often ignore the practical limitations of current hardware. Finally, transducer designers rarely
consider some unique aspects of flow actuation, such as actuator placement and the physics of the
fluid/actuator interaction.

248 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

2.2. Actuator Types

There are various types of actuators used in flow control applications, and these can be classified
in numerous ways. One useful classification is illustrated in Figure 1, in which the actuators are Time response:
organized based on function. The most common type is fluidic, which uses fluid injection or a measure of the time
suction. Within this classification, there are zero-net mass-flux (ZNMF) or synthetic jet actuators constant of a system to
a transient input, such
(Glezer & Amitay 2002). ZNMF devices, by definition, alternately ingest and expel fluid in an
as a step or impulse
oscillatory manner through an orifice/slot using only the working fluid with no external mass
source/sink. Nonzero mass-flux devices, on the other hand, require a fluid source/sink and can
employ steady and/or unsteady (e.g., pulsed) forcing using some combination of a valve (Seifert
et al. 1993, 1996; Bachar 2001), a natural fluidic oscillator (Viets 1975), and/or combustion-driven
device (Crittenden et al. 2001). They also can range in scale from conventional macroscale to
submillimeter-scale microjets (Alvi et al. 2003).
Another class involves a moving body inside or on the domain boundary. Although fluidic
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

devices usually involve some moving part, with the notable exception of most fluidic oscillators,
their primary function is fluidic injection or suction. Alternatively, the purpose of the moving
object/surface is to induce local fluid motion. An example is the electrodynamic ribbon oscillator
used in the classic flat-plate experiments of Schubauer & Skramstad (1948) on laminar boundary
by Brown University on 09/05/12. For personal use only.

layer transition. Other examples listed in Figure 1 include vibrating flaps (Katz et al. 1989,
Cattafesta et al. 1997, Seifert et al. 1998), time-periodic motion of a surface-mounted diaphragm
(Kim et al. 2003), an oscillating wire (Bar-Sever 1989), rotating surface elements (Viets et al. 1981),
and morphing surfaces (Thill et al. 2008).
The final class considered in this review is plasma actuators, which have gained popularity in
recent years because of their solid-state nature and fast time response. Moreau (2007) has reviewed
plasma actuators and their applications. The most popular variant is the single dielectric barrier

Flow control
actuators

Other (e.g.,
Moving
Fluidic Plasma electromagnetic,
object/surface magnetohydrodynamic)

Vibrating
Zero-net ribbon Corona
mass flux or discharge
synthetic jets
Vibrating
flap Dielectric
barrier
Nonzero discharge
mass flux Oscillating
wire
Local arc
Steady Rotating filament
surface

Unsteady Sparkjet
Morphing
surface
Oscillators Valves

Combustion

Figure 1
A type classification of flow control actuators.

www.annualreviews.org • Actuators for Active Flow Control 249

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

discharge (SDBD) plasma actuator, which has been the subject of detailed reviews explaining the
physics, design, and applications in Corke et al. (2007) and more recently in Corke et al. (2010).
Other plasma devices potentially suitable for high-speed flows include local arc filament (Samimy
Gain: the magnitude
of the complex ratio of et al. 2004) and sparkjet (Cybyk et al. 2004, Narayanaswamy et al. 2010) actuators. The devices
system output to listed in the last column of Figure 1 are beyond the scope of this review.
input; an example is
the magnitude of the
frequency response of
3. ACTUATOR FUNDAMENTALS
a system
3.1. Actuators Versus Sensors
Sensitivity: ratio of a
sensor’s electrical This section summarizes important characteristics of an actuator. Figure 2 compares the char-
output to engineering acteristics and design issues associated with actuators, which are intrusive by definition, versus
input sensors, which are ideally nonintrusive. First, the input and output are reversed. The input to an
unsteady sensor is a flow disturbance (e.g., velocity fluctuation), and the output is an electrical
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

quantity (e.g., voltage). For an actuator, an input electrical signal produces an output flow distur-
bance. So the actuator gain is essentially the inverse of sensor sensitivity. For example, the output
gain of a synthetic jet can be quantified using output velocity per unit voltage [(m s−1 )/V], whereas
the sensitivity of a velocity sensor is defined as the output voltage per unit velocity [V/(m s−1 )].
by Brown University on 09/05/12. For personal use only.

Although an ideal sensor only responds to the quantity of interest, it may be sensitive to other
input flow variables. For example, a sensor may exhibit sensitivity to temperature, electromagnetic

Actuators Sensors
• Must be intrusive • Ideally nonintrusive

Input = electrical signal Input = flow disturbance

Output = flow disturbance • Pressure
• Physical output: • Velocity
• Displacement • Wall shear stress
• Mass, momentum, energy • Temperature
• Sound Output = electrical signal
• Body force • Voltage
• Desired output: • Current
• Velocity, pressure, vorticity pert.

Design Issues Design Issues

• Quasi-static response • Static response
• Max output or free displacement • Max input
• Max or blocked force • Minimum detectable signal (MDS)
• Gain = [eng output]/[elec input] • Sensitivity = [elec output]/[eng input]
• Linearity • Linearity
• Dynamic response • Dynamic response
• Frequency response • Frequency response (mag & phase)
(mag & phase) • Bandwidth
• Bandwidth • Time response
• Time response • Other
• Other • Electrical power requirement
• Electrical power requirement • Electrical efficiency
• Electrical efficiency • Robustness
• Robustness • Size, weight, cost
• Size, weight, cost • May respond to other inputs:
• May produce unwanted outputs: EMI, vibration, temperature, etc.
EMI, sound, heat, etc. • Trade-offs
• Trade-offs • Sensitivity versus bandwidth
• Gain versus bandwidth • Max input versus sensitivity
• Displacement versus force

Figure 2
A comparison between actuators and sensors for flow control. Abbreviations: EMI, electromagnetic
interference.

250 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

interference, or vibration, etc. An actuator, alternatively, may produce parasitic perturbations, such
as electromagnetic interference, sound, or heat. A unique aspect of actuators is the actuator/fluid
interaction, which dictates how the physical output of actuators, such as surface motion or momen-
Frequency response:
tum via fluidic injection, couples with the flow in either a linear or nonlinear manner. For example, the steady-state
small perturbations can excite flow instabilities, whereas large amplitudes can significantly modify response of a (stable)
the mean flow and its corresponding stability characteristics (Theofilis 2009). linear system to a
Many of the terms used to characterize sensors are also relevant for actuators. In terms of sinusoidal input
its static response, a sensor is specified by its maximum input level and its minimum detectable
signal, and their ratio provides the dynamic range of the sensor. Similarly, the quasi-static (i.e.,
low-frequency) actuator response is characterized by its maximum output when the actuator is
free or unloaded. At the other extreme, an actuator produces its maximum force when its direct
or induced velocity is zero or blocked. Implicit in the use of any quantitative sensor is the local
linear nature of the device, which is required to maintain high fidelity in spectral and correlation
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

analysis. On the other hand, nonlinearities are often exploited in actuator design, which may
increase performance at the expense of additional modeling complexity. For example, a linear
electromechanical actuator (e.g., a piezo-composite vibrating beam) can be nonlinear owing to
fluid/structure interaction (Cattafesta et al. 2001).
by Brown University on 09/05/12. For personal use only.

In terms of dynamic response, both sensors and actuators are characterized by frequency re-
sponse (magnitude and phase) and bandwidth, but there are subtle yet important differences. The
frequency response describes how the gain (for an actuator), sensitivity (for a sensor), and phase
(for both) vary versus frequency for a linear system (Rowley & Batten 2009). Nonlinear actuators
produce a different gain versus frequency plot for different excitation levels. Many of the fluidic
actuators and moving object/surface actuators mentioned in Figure 1 leverage a resonant phe-
nomenon to produce a large gain at the expense of π phase shifts and nonlinear behavior. The
bandwidth of an actuator refers to the range of frequencies over which the output of the device is
acceptably large. This contrasts with a high-quality sensor that has a flat frequency response over
the desired bandwidth of interest (e.g., a constant-temperature hot-wire anemometer).
Another often misunderstood aspect of actuator design is the importance of time response,
which is the characteristic time delay to a particular input (e.g., step). The time constant and
phase lag can be related in a linear system, but this is not straightforward in a nonlinear system.
High-bandwidth actuators are required in active flow control systems in which the excitation
frequency must be tuned to adjust to changing flow conditions. However, in feedback control
systems (Figure 3) sensors are used to feed back (usually) random signals prevalent in turbulent
flows. Assuming the sensor possesses a flat frequency response with negligible phase lag, then

Forward path Disturbance Noise

d n

r e u x
F Σ C Σ P Σ y

–
H

Feedback path

Figure 3
Schematic of a feedback control block diagram.

www.annualreviews.org • Actuators for Active Flow Control 251

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Cover
Motor
Nozzles
Flexible coupling

Bearing End plug

Bearing

Air
inlet

Valve body

Figure 4
Schematic of a rotary valve. Figure reproduced from McManus & Magill (1996). Reprinted with permission
of the American Institute of Aeronautics and Astronautics.

H = 1. As a result, the error signal that is fed to the controller C is random and consists of multiple
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

frequency components, and the situation is similar for the actuator signal u. The actuator, which
is part of the plant P, must be able to respond to such a complex waveform. This means that
the actuator must have sufficient frequency and time response to adjust its amplitude, frequency,
by Brown University on 09/05/12. For personal use only.

and phase content to track u. In other words, it must be able to provide, at any instant in time, a
time-varying excitation signal containing multiple frequency components.
As an example, a rotary valve (Figure 4) can meet the bandwidth requirement by changing its
rotational speed but, at any given time, contains only a single frequency and its harmonics due to
the finite inertia of the device (McManus & Magill 1996). Similar arguments apply to standard
fluidic oscillators (Viets 1975), which do not provide independent control of the amplitude
and frequency and also preclude the actuator signal from phase locking to u. Counterexamples
include piezoelectric- or electrodynamic-driven synthetic jets or plasmas, which can meet these
requirements.
Next we consider the trade-offs inherent in actuator design. The most important actuator
design characteristic is arguably its control authority or maximum gain in some cases, whereas
bandwidth may be more important in other applications, such as closed-loop control (Rowley &
Batten 2009). Unfortunately, these two requirements are usually at odds with each other, much
like the well-known gain-bandwidth product in electronic amplifiers. For a given type, an actuator
that produces higher gain has a lower bandwidth. For example, Papila et al. (2008) provide a scaling
law that quantifies the trade-off between maximum volume displacement and bandwidth in the
design of clamped circular piezoceramic composite diaphragms used in synthetic jets. The design
goal is that the actuator should have sufficient gain over a prescribed range of frequencies, such
that the actuator will have the desired control effect.
Another trade-off is generalized displacement versus generalized force. Again, the sensor anal-
ogy provides some insight. A sensor is chosen with a full-scale linear range that matches the
expected maximum input. To avoid nonlinearities, the device must be stiffer than one with a
smaller full-scale range and will thus possess lower sensitivity. Similarly, an actuator will pro-
vide its maximum output displacement when the load seen by the actuator is minimized, and the
actuator is free. The output of an actuator often becomes nonlinear as its maximum output is ap-
proached. Alternatively, the actuator will produce its maximum force when the load is maximized
and the actuator is blocked. To illustrate this trade-off, we consider the linear relation between
the volume displaced Vol by a piezoceramic disc loaded by a differential pressure P across the
diaphragm and excited by an applied voltage Vac :

Vol = Ca D P + d a Vac , (1)

252 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

where Ca D = (Vol/P )Vac =0 is the short-circuit acoustic compliance of the composite diaphragm
and d a = (Vol/Vac ) P =0 is the effective acoustic piezoelectric coefficient or the maximum gain
(Prasad et al. 2006). When P = 0, the maximum or free displacement is Vol free = d a Vac , where
Vac = Emax t p is limited by the product of the coercive electric field and the piezoceramic thickness.
Similarly, when Vol = 0, the blocked pressure is Pblocked = −d a Vac /Ca D . The operating point of
the actuator lies along the line given by Equation 1 somewhere between these two intercepts. The
design goal then is to increase Vol free and Pblocked subject to application-specific design constraints.
Other desirable traits include low power, high efficiency, reasonable cost, and robustness. The
actuator should satisfy size and weight requirements for the specific application, and, in general,
smaller size and weight are desirable. In addition, an accurate model for design, scaling, and overall
control-system design is also desirable.
Seifert (2007) discusses potential dimensionless metrics. One is called the overall figure of
merit:
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Fa2 U p
OFM ≡ , (2)
W a Power
where Fa is the thrust, Up is the peak velocity, Wa is the weight of the actuator system, and Power
is the power consumption. The overall figure of merit, or metrics similar to it, defines an objective
by Brown University on 09/05/12. For personal use only.

function that can be used for design optimization. However, it is difficult to define a universal
performance metric suitable for all actuators and applications. For example, nanosecond surface
discharge plasma actuators appear to be quite effective but produce little discernible fluid velocity
(Roupassov et al. 2009).

3.2. Dimensional Analysis

This section briefly addresses dimensional analysis and its relevance to flow control actuators.
Dimensional analysis can be used to properly describe the effects of flow control actuators in wind-
tunnel experiments and then scale the results to full-scale applications. The lack of dimensional
analysis often hinders the assessment of an actuator for a potential application.
The simplified example of separation control using ZNMF actuators illustrates some of the
issues. First let us consider the actuator operating in ambient conditions driven by a sinusoid
with angular frequency ω = 2π f (Holman 2006). An assumed incompressible ZNMF actuator
(Figure 5a) that displaces a cavity volume Vol synthesizes a jet from a two-dimensional slot of
height h and width/diameter d with a Reynolds number that is a function of three other dimen-
sionless parameters:
 
h Vol
Re U j = f un , 3 ,S , (3)
d d
where Uj is a characteristic jet root-mean-square (rms) velocity (Cater & Soria 2002), and S =

ωd 2 /ν is the Stokes number. Note that Re U j /S2 = 1/St, so only two of the three parameters
are unique. As an alternative, the dimensionless stroke length (Smith & Glezer 1998) is often used
instead of St, but these parameters are related (Holman 2006):

1 L̄/d
= . (4)
St 2π
The uncontrolled or baseline lift CL and drag CD coefficients of an airfoil in incompressible flow
are functions of the chord Reynolds number Re c = ρU ∞ c /μ and angle of attack α. With ZNMF
control, the lift and drag coefficients become functions of the actuator parameters, airfoil geometry,
and the flow Reynolds number, which dictates the local boundary layer scales in the vicinity of

www.annualreviews.org • Actuators for Active Flow Control 253

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

a Piezoceramic diaphragm b Electrodynamic diaphragm

Orifice (or slot)

Primary cavity

Coil Diaphragm

d
h Vol Soft Hard
magnet magnet
assembly
Δ Vol(ω)
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org
by Brown University on 09/05/12. For personal use only.

c Piezoceramic equivalent circuit d Electrodynamic equivalent circuit

Electroacoustic Blc /AD
coupling 1:φ CaD RaD MaD RaLin MaO Qout ReC LeC

i i
Q Q
RaNL
QCav
Vac P Vac P
CeB
CaCav MaRad

Electrical Acoustic-fluidic Electrical Acoustic

domain domain domain domain

Figure 5
Schematics and equivalent circuits of the most popular versions of zero-net mass-flux (ZNMF) actuators. The right-hand side of the
equivalent circuit in panel d (not shown) is the same as in panel c.

the actuator. Hence,

 
h δ∗ δ∗ U j f c
CL = CL , , , , . (5)
c c θ U∞ U j
Experience has shown that some of the above parameters can often be combined in the form of
the rms jet momentum coefficient:
ρ j U j2 h
Cμ = . (6)
1
2
ρU ∞
2 c

If compressibility effects are added, either to the jet or the external flow, the list of dimensionless
parameters grows. Furthermore, Uj is affected by the presence of the grazing flow (Raju et al.
2009), so Cμ , which is typically determined via measurements in quiescent conditions, is often
only estimated. If the excitation signal is not sinusoidal, then additional waveform parameters
become important. Finally, we note that different investigators use different velocity scales for

254 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Uj (Smith & Glezer 1998, Cater & Soria 2002) and different length scales (e.g., the length of
the separation region instead of c). All these issues and incomplete parameter specifications make
it difficult to compare results. This example emphasizes the need for dimensional analysis and
careful documentation of flow control experiments.

4. ACTUATOR COMPARISON
Various actuator classifications are possible. These include (a) open loop versus closed loop,
(b) application, and (c) flow regime (i.e., low-speed, transonic). Each has its merits, but the type
classification in Figure 1 is adopted here for ease of access. Table 1 summarizes key advantages
and disadvantages of common actuators.

Table 1 Summary of common unsteady flow control actuators

Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Type Subtype Advantages Disadvantages

Fluidic
ZNMF Requires no external fluid source Peak velocities typically limited to low to
by Brown University on 09/05/12. For personal use only.

moderate subsonic speeds

Amenable to various types of drivers and sizes Resonant devices
Suitable for feedback control
Unsteady Capable of high velocities with either fast time May not be amenable to feedback control
valves response or high bandwidth but generally not both
Requires an external flow source
Oscillators Capable of producing large disturbances Standard versions not suitable for
feedback control
Amenable to a range of sizes and hence frequencies Requires an external flow source
Potential extensions possible to enable independent
control of frequency and velocity
Combustion Capable of producing large perturbations in Currently limited to relatively low
high-speed flows frequencies (a few hundred hertz)
Requires combustion
Moving surface
Piezoelectric Simple design amenable to different frequency ranges Has constant product of max deflection
flaps of interest and bandwidth
Can produce spanwise or streamwise vorticity Susceptible to fluid loading
Suitable for feedback control Resonant devices
Active dimples Potentially suitable for feedback control of turbulent Further development needed to achieve
wall-bounded flows required size and frequency response
Plasma
SDBD Easily installed on models Limited velocity output
Low mass Requires high voltage (kV)
Fast time response
No moving parts
Sparkjet All solid-state device capable of producing large Potential issues associated with EMI,
perturbations in high-speed flows acoustic level, and high temperature

Abbreviations: EMI, electromagnetic interference; SDBD, single dielectric barrier discharge; ZNMF, zero-net mass flux.

www.annualreviews.org • Actuators for Active Flow Control 255

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

4.1. Fluidic
ZNMF or synthetic jet actuators have been reviewed by Glezer & Amitay (2002). Here, we
emphasize recent developments associated with their modeling and design.

4.1.1. ZNMF actuators. Figures 5a,b show schematics of the standard piezoelectric and elec-
trodynamic configurations. The diaphragm oscillates about its equilibrium position, alternately
expelling/ingesting fluid from/into the cavity through an orifice or slot. In Figure 5a, a piezoce-
ramic composite diaphragm is clamped or pinned at its boundary. When excited by an ac voltage,
the diaphragm bends because of the reverse piezoelectric effect and displaces a fluid volume Vol.
In Figure 5b, a diaphragm or piston is driven by a voice coil, which uses electrodynamic transduc-
tion (McCormick 2000, Nagib et al. 2004, Agashe et al. 2009). The magnet assembly generates
a permanent magnetic field with magnetic flux density B, and an ac current i passing through a
wound coil of length produces an ac electromagnetic force F = B i that causes the diaphragm
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

to oscillate.
Other transduction schemes are possible and have their own advantages and disadvantages.
For example, capacitive schemes are easily amenable to microfabrication techniques, whereas
microfabricated piezoelectric and electrodynamic devices require more sophisticated approaches
by Brown University on 09/05/12. For personal use only.

(Coe et al. 1994, Madou 1997). Capacitive devices are highly nonlinear for large deflections
required to obtain high velocities, whereas piezoelectric and electrodynamic transducers are often
adequately described by linear models. All these drivers are capable of producing multifrequency
waveforms suitable for feedback control.
The primary advantages of piezoelectric transduction are low-power requirements, high band-
width, and broadband output from dc to the several kilo-Hertz range. Electrodynamic devices are
attractive for low-frequency applications because of their large displacement capability, but their
increased weight (due to the magnet assembly) and heat transfer (due to heating in the resistive
coil) present design challenges (McCormick 2000, Nagib et al. 2004, Agashe et al. 2009). The
primary disadvantages of conventional ZNMF devices include the following: (a) They typically
achieve a maximum velocity below approximately 100 m s−1 , and (b) they must be operated at
or near resonance for satisfactory performance. Operating near mechanical resonance can lead
to mechanical failure of the device (Tesar et al. 2006). In addition, 180◦ phase shifts associated
with any resonance can lead to instability in feedback control systems (Rowley & Batten 2009).
Non-negligible acoustic levels are likely at high frequencies due to the noncompact acoustic
sources associated with the oscillatory fluid motion near the slot and/or the vibrating diaphragm
(Devenport & Glegg 2009).
An attractive feature is that no external fluid source is required. Orifices or slots are necessarily
required (and similarly for all fluidic devices) to interact with the external flow. These orifices are
potentially subject to fouling. Because no external fluid source is used, any mass expelled during a
cycle must be balanced by that ingested to satisfy mass conservation. From the Reynolds transport
theorem, however, the net momentum flux from the orifice/slot along the jet axis during a period
T is
 T
ρu V · nd A > 0 (7)
0
during both expulsion and ingestion. However, because of the similarity between ZNMF and con-
ventional vortex-ring formation (Glezer 1988), researchers have historically used a characteristic
jet velocity based on a slug-flow model (Smith & Glezer 1998),
 T /2
U 0 = f L0 = f u 0 (t)d t, (8)
0

256 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

in which U0 is determined using only the ejection portion of the cycle from 0 to T/2. Cater &
Soria (2002) later proposed a momentum velocity that represents a spatial- and time-averaged
velocity over the entire slot area A and period
   T 1/2
1 1
Uj = [u (A, t)]2 d td A . (9)
AT A 0
If the amplitude of the diaphragm motion is sufficient, one or more vortex rings are formed
(Ingard & Labate 1950, Smith & Glezer 1998, Holman et al. 2005). These vortex rings synthesize
a time-averaged jet due to entrainment from the surrounding fluid. Holman et al. (2005) use
simple scaling and data from several sources to determine and validate a jet formation criterion,
Re j /S2 > K , where K is a constant.
Substantial progress has been made in the development of low-order models of sufficient fidelity
for performance prediction and design optimization (Rathnasingham & Breuer 1997, Gallas et al.
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

2003, Lockerby & Carpenter 2004, Prasad et al. 2006, Sharma 2007, Agashe et al. 2009, Tang
& Zhong 2009). In the simplest case, lumped-element modeling (LEM) is used (Fischer 1955,
Merhaut 1981, Rossi 1988). LEM provides estimates of the dynamic response of the actuator
(with errors on the order of 10%). The main assumption is that the characteristic length scales
by Brown University on 09/05/12. For personal use only.

of the governing physical phenomena (e.g., acoustic wavelength) are much larger than the largest
dimension. When valid, the governing partial differential equations of the distributed system
are lumped into a set of ordinary differential equations, and an equivalent electrical circuit is
constructed using conjugate power variables (i.e., power = effort × flow, such as voltage × current,
force × velocity, and pressure × volume flow rate). The lumped elements in the different energy
domains are represented by their respective complex-valued impedances, Z = effort/flow.
For piezoelectric-driven ZNMF actuators as in Figure 5c, an ac voltage Vac across the composite
piezoceramic diaphragm possessing a blocked electrical capacitance, CeB , induces an oscillatory
volume velocity by the diaphragm Q = j ωVol and an effective pressure P = φVac , which is
modeled using an electrical transformer with a turns ratio φ = −d a /Ca D (see Equation 1). As
the objective of the diaphragm is to displace fluid, the frequency range of interest is from dc to
beyond its natural frequency but less than the second resonant frequency. The diaphragm is thus
modeled by an acoustic compliance CaD , the inverse of a spring constant, an acoustic damping
resistance RaD , and an acoustic mass MaD . With the exception of RaD , these lumped elements
are obtained analytically using linear composite plate theory and are functions of geometry and
material properties (Prasad et al. 2006).
The motion of the diaphragm either compresses/expands the fluid in the cavity or expels/ingests
fluid through the orifice/slot to produce an output volume flow rate Q o ut = Ao ut ū. At low frequen-
cies, the cavity acts like a spring and is thus modeled by an acoustic compliance CaCav = Vol/(ρc 02 ),
where Vol is the cavity volume, and c0 is the isentropic sound speed. Changes in fluid temperature
and pressure thus affect the cavity compliance. The orifice is represented by a linear resistor RaLin
(i.e., major losses) and a nonlinear resistor RaNL = (K D 0.5 ρ ū 2 )/Q o ut (i.e., minor losses). A resis-
tor can be added to account for acoustic radiation. The kinetic energy of the fluid in the orifice
is modeled by MaO , whereas the radiation mass M aRad accounts for the oscillatory fluid mass or
end effects. The lumped elements are analytical functions of geometry and fluid properties (with
the exception of the empirical minor-loss coefficient KD and the damping of the diaphragm RaD ).
Equations are provided in Gallas et al. (2003) and the references therein. McCormick (2000) and
Agashe et al. (2009) provide a similar framework for electrodynamic ZNMF actuators shown in
Figure 5b.
A key conclusion from the lumped-element analysis, which is confirmed by experiments, is that
a ZNMF actuator is a coupled, nonlinear, two-degree-of-freedom oscillator with two characteristic

www.annualreviews.org • Actuators for Active Flow Control 257

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

 √
frequencies. One is the natural frequency  of the driver fd = 1/ 2π M a D Ca D , and the other is
the Helmholtz frequency f H = 1/(2π (M a O + M aRad )CaCav ) of the cavity/orifice. A common
misconception is that the two resonant frequencies of the coupled oscillator, denoted as f1 < f2 , are
identical to fd or fH . But this may not be the case and, in fact, depends on the coupling between the
diaphragm and the cavity. This phenomenon is discussed for linear, undamped, coupled oscillators
in Fischer (1955, pp. 20–40) and Gallas et al. (2003) and can be used to an advantage when designing
ZNMF actuators to obtain devices with single or dual resonant frequencies. Examples of both types
are presented in Gallas et al. (2003).
There are some inherent limitations with LEM. First, the lumped assumption is valid only for
devices that are sufficiently compact. Kim et al. (2005) extended the lumped-element technique to
consider the relevant case of resonance and delays associated with long lengths of tubing between
the driver and the orifice/slot. LEM can be extended to distributed systems using transfer matrix
analysis (Chaudhry 1979, Sherman & Butler 2007). A second deficiency is that the loss parameter,
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

KD ∼ 1, is an empirical constant. For steady flow, it is a function of geometry and Reynolds number.
For oscillatory flow, however, as explained in Section 3.2, the list of dependent parameters expands
to include the Strouhal (or Stokes) number. Lockerby & Carpenter (2004) have developed a model
that solves the one-dimensional Navier-Stokes equations in the orifice, which can be incorporated
by Brown University on 09/05/12. For personal use only.

in the lumped-element model. Raju et al. (2007) also incorporate additional physics into the
lumped-element model using results from high-fidelity numerical simulations.
For higher-fidelity models, more sophisticated approaches short of full numerical simulations
are possible (Yamaleev et al. 2005, Yamaleev & Carpenter 2006). Yamaleev et al. (2005) solve
the time-dependent compressible quasi-one-dimensional Euler equations, while the diaphragm is
modeled as a moving boundary. The inviscid approach satisfies conservation of mass, momentum,
and energy, but it is unclear whether this approach captures the physics of the unsteady vena
contracta in the orifice (Raju et al. 2007).
Rumsey et al. (2006) summarize the results of a workshop titled CFD Validation of Synthetic
Jets and Turbulent Separation Control. Three test cases of varying complexity were considered
by numerous contributors: (a) a ZNMF jet in a quiescent medium, (b) a ZNMF jet in a cross-
flow, and (c) ZNMF jet control of flow over a hump model. For the isolated jet, case (a), several
issues were noted. These included the difficulty of modeling transitional flow behavior and three-
dimensional effects using Reynolds-averaged Navier-Stokes approaches. Kotapati et al. (2007)
showed that it is essential to match the dimensionless jet parameters in Equation 3 using time-
accurate, three-dimensional simulations via large-eddy simulation or direct numerical simulation.
Despite significant simplifications in the computational fluid dynamics (CFD) model of the actu-
ator geometry and diaphragm motion, good agreement between experiments and the simulations
was obtained.
The hierarchy of modeling fidelity thus varies from a simple lumped-element model to high-
fidelity numerical simulations. A CFD simulation that resolves both the small-scale details of the
actuator and the large-scale behavior of the controlled flow field is impractical for routine design
calculations or optimization. Alternatively, although sufficient for prediction and design purposes,
LEM lacks the fidelity required to provide an appropriate time-dependent boundary condition
for CFD. Thus, an important current research topic is the development of reduced-order models
of a ZNMF actuator(s) suitable for use in a controlled flow field, and recent efforts in this regard
are described in Raju et al. (2009).

4.1.2. Pulsed jets. Pulsed jets have been used for many years in open-loop applications with
good success. It is well established that unsteady forcing has significant advantages versus steady

258 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

forcing for many flow control applications (Greenblatt & Wygnanski 2000). By definition, a steady
jet has negligible temporal or ac variations. A pulsed jet is an unsteady jet that is ideally either
on or off and, in the periodic case, can be characterized by a square wave with a duty cycle that
indicates the percentage of time that the jet is on. Hence, pulsed jets have both a time-averaged
or dc component and an unsteady or ac component. Unlike ZNMF devices, pulsed jets require
an external fluid source.
A pulsed jet can be generated using, for example, a fast-acting solenoid valve (Bons et al. 2002),
a high-speed rotating siren valve (Williams et al. 2007), or the rotating orifice/slot assembly like
that shown in Figure 4 (McManus & Magill 1996). A dc motor controls the rotational speed
of the valve, and when the holes from the rotating inner body align with the fixed outer holes,
pressurized air from the inlet is expelled. Although the bandwidth of a rotating valve may be
sufficient to operate at any one of several characteristic frequencies of the flow, it is not possible to
phase lock or synchronize the latter two types with a bandpass-filtered reference signal in the flow
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

as the phase of the valve cannot be controlled. However, if the time and frequency response of
the solenoid valve are sufficient, then synchronization is possible, although the type of permissible
waveforms is usually restricted to variable duty-cycle square waves.
One of the primary disadvantages of a pulsed jet is the required flow source. The required
by Brown University on 09/05/12. For personal use only.

average mass flow rate is usually quantified using Cμ based on the mean jet velocity and/or a
dimensionless mass flow coefficient

Cq = ρ j U j A j / (ρ∞ U ∞ Ar ) , (10)

where Ar is an appropriate reference area. If the mass flow rate is reduced in some manner while
maintaining control authority, then the system cost of a pulsed jet may be acceptable.
One promising concept is a pulsed microjet, which is an extension of steady microjets used to
control separation (Kumar & Alvi 2006), supersonic impinging jets (Alvi et al. 2003, Lou et al.
2006), high-speed jet noise (Arakeri et al. 2003), and flow-induced cavity oscillations (Zhuang
et al. 2006). The idea is to reduce ṁ j by decreasing the orifice diameter while maintaining high
jet momentum. By pulsing the jet, the mean velocity component is reduced relative to that of
a steady jet, whereas the unsteady momentum coefficient based on the rms jet velocity (using
Equations 6 and 9) remains large. Pulsed microjets can be obtained using the same approaches
as for their macroscale counterparts. For example, Ibrahim et al. (2002) used a high-speed motor
capable of achieving maximum rotational speeds of 12,000 rpm to enable unsteady forcing at
6,300 Hz of compressible jets for mixing enhancement. Choi et al. (2006) devised a scheme to
pulse the microjets using a rotating-cap assembly. Although none of these actuators is suitable
for closed-loop control, the following two variations are. Hogue et al. (2010) and Oates & Liu
(2009) used a piezoelectric stack actuator and hydraulic amplifier circuit to produce oscillatory
flow through a 400-μm-diameter microjet at frequencies up to 800 Hz. Hogue et al. (2009) also
demonstrated a deformable converging-diverging micronozzle using a compact piezoelectric stack
actuator that varies the nominal 1.5-Mach number by ∼25%.
Vortex generator jets are similar to microjets but have been used much more extensively
(Compton & Johnston 1992, Selby et al. 1992, Khan 2000, Bons et al. 2002, Johnston & Nishi
2002, Zhang 2003, Godard & Stanislas 2006, Warsop et al. 2007). The concept is to mimic
conventional vortex generators in a fluidic manner by skewing and pitching the jet axis with
respect to the free-stream flow direction (see Figure 6). Conventional vortex generators that are
always present add parasitic drag, whereas vortex generator jets can be deployed when required. An
effective periodically spaced vortex-generator-jet array produces corotating streamwise vortices
that stay within the boundary layer and inhibit separation via momentum exchange. Pulsing the jet
reduces the required mass flow rates without jeopardizing performance benefits (Bons et al. 2002).

www.annualreviews.org • Actuators for Active Flow Control 259

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

U∞

VGJ

Pitch
Skew
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Surface
by Brown University on 09/05/12. For personal use only.

Figure 6
Schematic of a vortex generator jet (VGJ).

Warsop et al. (2007) have developed a microfabricated piezoelectric version of this device suitable
for feedback control that produces exit velocities greater than 300 m s−1 through a 200-μm-
diameter orifice at 500 Hz using only 50 mW.

4.1.3. Powered resonance tube. Pulsed-jet actuators are usually unable to force high-frequency
instabilities due to bandwidth or time-response limitations. A different type of fluidic actuator
with high-frequency forcing capability is the powered resonance tube, which leverages acoustic
resonance. These actuators are adapted from the Hartmann tube first discovered by Hartmann &
Trolle (1927) and have been subsequently adapted by several researchers for active flow control
applications (Kastner & Samimy 2002; Stanek et al. 2002a,b; Raman et al. 2004; Sarpotdar et al.
2005).
The basic concept, illustrated in Figure 7, impinges either an underexpanded sonic jet or an
ideally expanded supersonic jet from a nozzle into a closed tube of length L. A strong acoustic
resonance, with fluctuating near-field pressure levels as high as 160 dB re 20 μPa, can be established

ΔX L

Figure 7
Schematic of a powered resonance tube.

260 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

in which incident compression waves travel down the tube and reflect from its closed end as
compression waves. These waves then reflect as expansion waves at the tube entrance, which
subsequently travel back down the tube length and reflect from the end as expansion waves.
When these reach the tube entrance, they reflect as compression waves, and the cycle begins
anew.
Raman et al. (2004) describe how a simple quarter-wave resonator model predicts the resonant
frequency for long tubes, fr = c 0 /4L. However, the accuracy of this formula degrades for shorter
tubes, and the finite reactance of the fluid mass in the gap (i.e., end effects) must be accounted for
to improve the prediction (Kerschen et al. 2004). The resonant frequency can be adjusted over
a wide range, from 1.6 to 15 kHz in Raman et al. (2004), by varying L. In practice, Raman et al.
(2004) implemented both a look-up table and feedback control schemes to automatically adjust L
and/or the gap x to achieve a desired resonant frequency.
The main attraction of these actuators is their ability to generate large high-frequency exci-
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

tations potentially suitable for control of high-speed flows—hence the name HIFEX. Although
theories have been proposed about the mechanism responsible for control (Stanek et al. 2002a,b,
2003), it is important to emphasize that these devices impart momentum to the flow and a nonzero
ṁ with Cq ≈ 1% based on the cavity opening area. These disturbances thus affect not only the
by Brown University on 09/05/12. For personal use only.

targeted high-frequency instabilities but also the mean flow (Ukeiley et al. 2004). Because the
disturbances are large, this has two possible effects. First, the mean flow instability characteristics
are modified. Second, the large disturbances may produce nonlinear interactions between multiple
frequency scales.
Another point worth noting is that, although these actuators possess high bandwidth and
tunable resonant frequencies, they do not have fast time response. Hence, the excitation signal
frequency cannot be changed rapidly. As with rotary-valve-based devices, these actuators cannot
be phase locked to a measured signal in the flow. The amplitude of the disturbances cannot be
precisely controlled either. Hence, this actuator is not suitable for feedback control.

4.1.4. Combustion actuators. The combustion-driven actuator is a pulsed jet that is produced
by the ignition of a mixture of gaseous fuel and oxidizer in a small (cm3 scale) combustion chamber
(Crittenden et al. 2001, Crittenden 2003, Glezer & Crittenden 2003, Cutler et al. 2005). The cycle
begins with the injection of premixed fuel and oxidizer into the combustion chamber, displacing the
remaining combustion products from the previous cycle. An integrated small-scale spark ignites
the mixture, and a combustion process ensues that typically lasts several milliseconds (depending
on the type of fuel, mixture ratio, and physical sizes of the combustor and orifice). The combustion
results in a rapid pressure rise in the chamber and the ejection of a pulsed high-speed jet through
one or more orifices. The operating frequency can be varied by controlling the flow rate of the
fuel/oxidizer and the ignition frequency. Frequencies greater than 150 Hz have been achieved
with chamber pressures of up to 5 atm, and these devices are capable of producing sonic velocities
at the jet orifice. Significant jet penetration into a cross-flow at Mach numbers up to 0.7 has been
demonstrated (Crittenden et al. 2001).
The primary attraction of this device is its high velocity output. However, because of the finite
time duration associated with the combustion cycle, the device is limited in its current form to
relatively low frequencies. In addition, although similar in many respects to a synthetic jet, this
device is not a ZNMF device due to the requirement for small but nonzero reactant flow. The
effects of high temperature must also be considered. Finally, it should be noted that the device
is not easily amenable to feedback control applications because of its discrete pulse behavior,
although the ignition can be synchronized to specific flow conditions. The frequency range of this
device has been extended using a fluidic oscillator (Crittenden & Raghu 2009).

www.annualreviews.org • Actuators for Active Flow Control 261

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

a Feedback b Feedback-free

Feedback path
Interaction
region

Interaction
Supply Outlet region
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Figure 8
Schematics of two variants of fluidic actuators. Figure used with permission of Bowles Fluidics Corporation, U.S. Patents 4,463,904,
by Brown University on 09/05/12. For personal use only.

4,645,126, 4,508,267, and 6,253,782.

4.1.5. Fluidic oscillators. Fluidic oscillators are devices that generate a pulsed jet when supplied
by a pressurized fluid. The standard configurations described below are potentially attractive for
flow control because they have no moving parts and produce disturbances with controllable fre-
quency. These actuators exploit fluidics technology (Kirshner 1966, Morris 1973), usually via the
mechanisms of wall attachment, called the Coanda effect (Coanda 1936), and/or fluid interac-
tions (Raghu 2001), as depicted in Figure 8. In the bistable fluidic nozzle shown in Figure 8a,
the primary flow attaches itself to either side due to the Coanda effect. The Euler-N equation,
ρv 2 /r = ∂ p/∂r, dictates that the pressure increases in the feedback loop. This increase in pressure
pushes the jet to the other side, and this process repeats in a cyclic fashion. A pulsed jet is thus
obtained that sweeps from one side of the exit nozzle to the other. The flow rate of the supply jet
and the length of the feedback loop are the primary parameters that establish the frequency. In
Figure 8b, the two jets impinge, and this leads to an unstable interaction and vortical patterns in
which the jets cyclically change their position. The result is a sweeping jet pattern at the nozzle
exit.
Historically, flow control applications used the subsonic flip-flop nozzle devised by Viets (1975),
which was extended to supersonic speeds by Raman et al. (1993). Raman and colleagues subse-
quently evaluated these devices in a variety of applications (Raman & Cornelius 1995, Raman et al.
1994). Macroscale Viets devices are bulky and are typically restricted to frequencies <500 Hz.
Over time, miniature integrated fluidic devices have been developed based on patented designs
by Bowles Fluidics Corporation (Raghu 2001). These miniature devices have been used with
good success in cavity oscillations (Raman et al. 1999), combustion control (Guyot et al. 2009),
thrust vectoring (Raman et al. 2005), and separation control (Seele et al. 2009). Other promising
miniature fluidic oscillators based on (a) the combination of an ejector and a bistable switching
valve and (b) microimpinging jets are described in Arwatz et al. (2008) and Soloman et al. (2008),
respectively.
Although the principles of operation are understood, the detailed flow physics of these actu-
ators are not. (Interested readers are referred to a review on bifurcating jets in Reynolds et al.
2003.) Gregory and colleagues (2004, 2007b) have used a variety of diagnostic tools, including

262 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

fast-response pressure-sensitive paint, to explore the unsteady behavior of these actuators. In ad-
dition, Gokoglu et al. (2009) have used numerical simulations to shed additional light on this
matter. It is important that design tools are developed for these devices that accurately predict
and optimize their performance.
Although these actuators are promising, a major drawback is that the oscillation frequency is
directly dependent on the flow rate through the device. It is highly desirable to decouple these two
parameters. As seen in Figure 8a, the feedback tubes can be eliminated and replaced with control
ports. If a sufficient perturbation can be introduced at one or both of the control ports, then the
jet oscillation frequency and phase can be controlled independent of the flow rate. The device is
then suitable for feedback control applications. Various methods have recently been explored with
mixed success, including a piezoelectric bender used as a diverter (Gregory et al. 2009), a solenoid
(Culley 2006), and an SDBD actuator in one of the control ports (Gregory et al. 2007a).
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

4.2. Moving Object/Surface

Moving surface actuators can take various forms, but the most common are the piezoelectric
composite flaps and electroactive dimples. These are summarized below.
by Brown University on 09/05/12. For personal use only.

4.2.1. Piezoelectric flaps and active dimples. The piezoelectric flap actuator has been used
successfully in a variety of applications, including control of separation (Seifert et al. 1998), tur-
bulent boundary layer streaks ( Jacobson & Reynolds 1998, Jeon & Blackwelder 2000), free shear
flows (Wiltse & Glezer 1993, 1998; Cattafesta et al. 2001), and flow-induced cavity oscillations
(Cattafesta et al. 1997; Raman & Cain 2002; Kegerise et al. 2007a,b). A cantilever composite beam
configuration is commonly used. The actuator can introduce spanwise or streamwise vortical per-
turbations into the flow depending on the geometry and orientation of the vibrating tip with
respect to the local free-stream flow (Cattafesta & Sheplak 2009). Application of an ac voltage
across the piezoceramic causes the beam to vibrate, which then interacts with the flow. Composite
beam modeling is treated in Cattafesta et al. (2001) and an extension to a bimorph is presented in
Mathew et al. (2006).
Typical free-tip displacements range from O(10–100 μm) for a device with a resonant frequency
in excess of 1–2 kHz (Kegerise et al. 2007a) to O(1 mm) when the resonant frequency is reduced
to a few hundred hertz or less. Mathew et al. (2006) describe the gain-bandwidth trade-off and
design-optimization methods. A comprehensive discussion of the performance trade-offs inherent
in the design and selection of microelectromechanical-systems actuators (and sensors), such as
cantilevers, is presented in Bell et al. (2005). Fluid loading via two-way coupling is considered in
Kudar & Carpenter (2007). Cattafesta et al. (2001) discuss a simple model for the quasi-static case
when fluid loading is not significant, f δ/U ∞ 1.
Recent progress has been made in the microfabrication of electroactive polymer dimples for the
control of turbulent boundary layers (Arthur et al. 2006, Dearing et al. 2007). A dimple consists of
an elastomer sandwiched between compliant electrodes. Upon application of a voltage, an effective
mechanical pressure is produced that is proportional to E 2f ield . The induced strain in the thickness
direction produces lateral expansion via the Poisson effect. The lateral strain is constrained at
the boundary, which leads to out-of-plane buckling. The dimples thus produce unsteady surface
depressions that interact with the near-wall turbulent structures. Research is focusing on under-
standing and predicting the behavior of the microfabricated dimples and ultimately designing
devices with the desired size, gain, and bandwidth requirements.

www.annualreviews.org • Actuators for Active Flow Control 263

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

a Corona
Plasma Dielectric

+dc –dc

b DBD
ac
Plasma Dielectric

c Sliding discharge
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

ac + dc
Plasma Dielectric
by Brown University on 09/05/12. For personal use only.

Figure 9
Schematics of three common plasma actuators: (a) corona, (b) dielectric barrier discharge (DBD), and
(c) sliding discharge.

4.3. Plasma
SDBD actuators are becoming increasingly popular because they have no moving parts and have
rapid time response and low mass, for example. The interested reader is referred to recent extensive
reviews on these devices and their applications (Kunhardt 2000; Corke et al. 2007, 2010; Moreau
2007) and a dedicated edition on plasma actuators for flow control in the Journal of Physics D:
Applied Physics (Volume 40, Number 3, February 2007). Here we focus on design-optimization
trends and research issues. Some modeling approaches and issues are discussed in detail in Corke
et al. (2010), Singh & Roy (2007), and Font (2006).

4.3.1. Single dielectric barrier discharge actuators. As shown in Figure 9b, an SDBD actuator
consists of an asymmetric pair of electrodes separated by a dielectric material. A high-voltage ac
waveform, O(1–30 kV) with frequencies O(50 Hz to 20 kHz), is supplied to the exposed electrode,
which results in an asymmetric electric field that ionizes air molecules to form a cold plasma in
which only a small fraction of air molecules are ionized. The accelerated charged particles transfer
momentum to the surrounding gas adjacent to the surface via collisions with neutral particles.
Enloe et al. (2004) provide a clear description of the plasma morphology for an SDBD actuator.
In addition, they report power dissipation
 T  T
1
P= Vac (t) i (t) d t = f Vac (q ) d q (11)
T 0 0

by the plasma actuator for thin dielectrics as P ∝ Vac3.5 , whereas Pons et al. (2005) and Jolibois
& Moreau (2009) show for thick dielectrics that P ∝ f (Vac − V0 )n , where 2 < n < 3 and V0
is a threshold voltage. Thomas et al. (2009) describe the transition in n as the drive voltage or
dielectric thickness increases.

264 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Noting that the actuator induces a wall jet (in quiescent conditions) with a velocity profile u( y),
the mechanical power per unit length is

Pmec h 1 ∞ 3
= ρu d y, (12)
L 2 0
and the maximum efficiency is η = Pmec h /P. The efficiency increases with dielectric thickness, but
peak values on the order of only 0.1% have been observed ( Jolibois & Moreau 2009).
The primary disadvantage of SDBD actuators is their diminished control authority for increas-
ing U∞ . Hence, considerable effort has gone into maximizing the performance of SDBD actuators
(Corke et al. 2010, Forte et al. 2007, Jolibois & Moreau 2009, Thomas et al. 2009). Researchers
have optimized performance via parametric studies of voltage amplitude, frequency, waveform,
slew rate, dielectric material and thickness, electrode geometry, multiple actuator arrays, and even
the use of a plasma synthetic jet (Santhanakrishnan & Jacob 2007). A general recommendation is
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

to use thicker dielectrics (several millimeters) with low dielectric constant for improved efficiency.
Forte et al. (2007) report U j etmax ≈ 7 m s−1 via optimization of several geometrical and electrical
parameters. Likhanskii et al. (2010) provide consistent estimates of the maximum velocity under
√
simplifying assumptions, U j etmax = E f ieldmax ε0 /ρ ≈ 8 m s−1 . Boeuf et al. (2007) similarly provide
by Brown University on 09/05/12. For personal use only.

estimates of the maximum body force. Jolibois & Moreau (2009) show that a physical limit is in-
deed reached as U j etmax saturates at P/L ≈ 2–3 W cm−1 . A unique capability of SDBD actuators is
their ability to excite the flow over a wide range of time/frequency scales using unsteady actuation
(Benard & Moreau 2010).

4.3.2. Other plasma actuators. Collectively, these efforts suggest that other approaches are
required to significantly increase performance. Options include multiple barrier (Benard et al.
2009b, Durscher & Roy 2010) and sliding discharge actuators (Louste et al. 2005, Moreau et al.
2008, Corke et al. 2010), repetitive nanosecond-scale pulse DBD (Roupassov & Starikovskii 2008,
Roupassov et al. 2009, Starikovskii et al. 2009, Unfer & Boeuf 2009), and local arc filament actuators
(Samimy et al. 2004, 2007; Utkin et al. 2007). Benard et al. (2009b) describe a multiple dielectric
barrier device in which the third electrode acts as a shield between two successive DBDs, resulting
in a near-constant electric wind velocity above the multi-DBD actuator. A sliding discharge also
uses three electrodes (Figure 9c) with a combination of an ac DBD to generate a local barrier
discharge and a dc component to induce the formation of extended stable plasma sheets between
the two electrodes on the same side of the dielectric. An increased body force relative to the
standard SDBD is produced. Short-pulse O(5–50 ns) DBD may substantially improve control
authority in high-speed flows (Roupassov et al. 2009). The rapid discharge promotes rapid gas
heating that creates localized shock waves with little discernible electric wind, suggesting that
these are no longer cold plasmas. Localized arc filaments are similar and provide a rapid, on the
order of microseconds, localized heating source capable of producing significant local pressure
perturbations with independent control of the frequency and phase (Samimy et al. 2004, Utkin
et al. 2007). Narayanaswamy et al. (2010) have designed a pulsed-plasma synthetic or sparkjet
generated via an electrical discharge in a small cavity. Their device generates peak velocities of
approximately 250 m s−1 at frequencies up to 5 kHz. Interested readers are referred to some
additional review articles on the application of plasma to high-speed flow control and combustion
(Bletzinger et al. 2005, Shang et al. 2005, Starikovskaia 2006).
Some comments on issues or potential pitfalls in SDBD actuator characterization tests are
warranted. Forte et al. (2007) note that some larger seed particles used in velocity measurements
may be subject to Coulomb forces, suggesting that smaller or dielectric seed particles may be
required. Jolibois & Moreau (2009) point out that the maximum achievable velocity is limited by the

www.annualreviews.org • Actuators for Active Flow Control 265

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

slew rate of the power amplifier. Thomas et al. (2009) discuss the sources of potential discrepancies
between thrust-based and velocity-based measurements. Note that the induced thrust is often
measured and used as an indicator of the body force, but the measured thrust is dependent on the
size of the plate and includes end effects. A control volume analysis to extract Fbo d y from velocity
measurements on the bounding control surfaces (assuming pressure ∼ const) may offer a consistent
method for evaluation of the body force. In terms of operation in various environments, SDBD
can operate at elevated relative humidity levels although with decreased efficiency (Benard et al.
2009a) and at subatmospheric pressures typical of flight conditions (Abe et al. 2008, Benard et al.
2008).

5. SUMMARY AND FUTURE TRENDS

The future is bright for active flow control, but the search for the perfect actuator continues.
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Of course, no such device exists due to the variety of application-specific design issues. The not-
invented-here syndrome and applied nature of actuators often lead to proprietary designs and the
corresponding lack of detail in publications. Unfortunately, this state of affairs impedes progress.
Above we attempt to address the actuator design problem in general terms and thereby encourage
by Brown University on 09/05/12. For personal use only.

researchers to think more broadly about inherent trade-offs in actuator design and how the actuator
fits into the overall active flow control system.
In terms of future directions and trends, the following three areas are promising: (a) the de-
velopment and incorporation of improved smart materials and fabrication methods, (b) the health
monitoring of actuator performance for robust systems-level integration, and (c) the development
of hierarchical optimized design (performance versus cost) tools. This partial list emphasizes the
need for multidisciplinary collaborative efforts. A variety of modeling and design tools of vary-
ing complexity and quantifiable fidelity must be developed to assess performance over a range
of conditions in applications. Based on our limited experience, we believe the flow control com-
munity must increase its emphasis on a fundamental physics-based design approach with proper
recognition and disclosure of actuator limitations. This is necessary to improve the probability of
successful transition of laboratory-scale devices to full-scale applications.

DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.

LITERATURE CITED
Abe T, Takizawa Y, Sato S, Kimura N. 2008. Experimental study for momentum transfer in a dielectric barrier
discharge plasma actuator. AIAA J. 46:2248–56
Agashe JS, Arnold DP, Cattafesta LN. 2009. Development of compact electrodynamic zero-net mass-flux actuators.
Presented at AIAA Aerosp. Sci. Meet., 47th, Orlando, AIAA Pap. 2009-1308
Alvi FS, Shih C, Elavarasan R, Garg G, Krothapalli A. 2003. Control of supersonic impinging jet flows using
microjets. AIAA J. 41:1347–55
Arakeri VH, Krothapalli A, Siddavaram V, Alkislar MB, Lourenco LM. 2003. On the use of microjets to
suppress turbulence in a Mach 0.9 axisymmetric jet. J. Fluid Mech. 490:75–98
Arthur G, McKeon BJ, Dearing S, Morrison JF, Cui Z. 2006. Manufacture of micro-sensors and actuators for
flow control. Microelectron. Eng. 83:1205–8
Arwatz G, Fono I, Seifert A. 2008. Suction and oscillatory blowing actuator modeling and validation. AIAA
J. 46:1107–17

266 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Bachar T. 2001. Generating dynamically controllable oscillatory fluid flow. U.S. Patent No. 6,186,412
Bar-Sever A. 1989. Separation control on an airfoil by periodic forcing. AIAA J. 27:820–21
Bell DJ, Lu TJ, Fleck NA, Spearing SM. 2005. MEMS actuators and sensors: observations on their performance
and selection for purpose. J. Micromech. Microeng. 15:S153–64
Benard N, Balcon N, Moreau E. 2008. Electric wind produced by a surface dielectric barrier discharge
operating in air at different pressures: aeronautical control insights. J. Phys. D Appl. Phys. 41:042002
Benard N, Balcon N, Moreau E. 2009a. Electric wind produced by a surface dielectric barrier discharge operating over
a wide range of relative humidity. Presented at AIAA Aerosp. Sci. Meet., 47th, Orlando, AIAA Pap. 2009-488
Benard N, Mizuno A, Moreau E. 2009b. A large-scale multiple dielectric barrier discharge actuator based on
an innovative three-electrode design. J. Phys. D Appl. Phys. 42:235204
Benard N, Moreau E. 2010. Capabilities of dielectric barrier discharge plasma actuator for multi-frequency
excitations. J. Phys. D Appl. Phys. 43:145201
Bletzinger P, Ganguly BN, Wie DV, Garscadden A. 2005. Plasmas in high speed aerodynamics. J. Phys. D
Appl. Phys. 38:R33–57
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Boeuf JP, Lagmich Y, Unfer T, Callegari T, Pitchford LC. 2007. Electrohydrodynamic force in dielectric
barrier discharge plasma actuators. J. Phys. D Appl. Phys. 40:652–62
Bons JP, Rivir RB, Sondergaard R. 2002. The fluid dynamics of LPT blade separation control using pulsed
jets. J. Turbomach. 124:77–85
Cater JE, Soria J. 2002. The evolution of round zero-net-mass-flux jets. J. Fluid Mech. 472:167–200
by Brown University on 09/05/12. For personal use only.

Cattafesta L, Garg S, Choudhari M, Li F. 1997. Active control of flow-induced cavity resonance. Presented at AIAA
Fluid Dyn. Conf., 28th, Snowmass, Colo., AIAA Pap. 1997-1804
Cattafesta L, Garg S, Shukla D. 2001. Development of piezoelectric actuators for active flow control. AIAA
J. 39:1562–68
Cattafesta L, Sheplak M. 2009. Actuators and sensors. See Joslin & Miller 2009, pp. 149–76
Chaudhry M. 1979. Applied Hydraulic Transients. New York: Van Nostrand Reinhold
Choi JJ, Annaswamy AM, Alvi FS, Lou H. 2006. Active control of supersonic impingement tones using steady
and pulsed microjets. Exp. Fluids 41:841–55
Coanda H. 1936. Device for deflecting a stream of elastic fluid projected into an elastic fluid. US Patent No.
2,052,869
Coe D, Allen M, Trautman M, Glezer A. 1994. Micromachined jets for manipulation of macro flows. Presented at
Solid-State Sensor Actuator Workshop
Compton DA, Johnston JP. 1992. Streamwise vortex production by pitched and skewed jets in a turbulent
boundary layer. AIAA J. 30:640–47
Corke T, Post M, Orlov D. 2007. SDBD plasma enhanced aerodynamics: concepts, optimization and appli-
cations. Prog. Aerosp. Sci. 43:193–217
Corke TC, Enloe CL, Wilkinson SP. 2010. Dielectric barrier discharge plasma actuators for flow control.
Annu. Rev. Fluid Mech. 42:505–29
Crittenden TM. 2003. Fluidic actuators for high speed flow control. PhD thesis. Georgia Inst. Technol., Mech.
Eng.
Crittenden TM, Glezer A, Funk R, Parekh D. 2001. Combustion-driven jet actuators for flow control. Presented
at AIAA Fluid Dyn. Conf., 31st, Anaheim, Calif., AIAA Pap. 2001-2768
Crittenden TM, Raghu S. 2009. Combustion powered actuator with integrated high frequency oscillator. Int.
J. Flow Control 1:87–97
Culley DE. 2006. Variable frequency diverter actuation for flow control. Presented at AIAA Flow Control Conf.,
3rd, San Francisco, AIAA Pap. 2006-3034
Cutler AD, Beck BT, Wilkes JA, Drummond JP, Alderfer DW, Danehy PM. 2005. Development of a pulsed
combustion actuator for high-speed flow control. Presented at AIAA Aerosp. Sci. Meet. Exhib., 43rd, Reno,
AIAA Pap. 2005-1084
Cybyk B, Grossman K, Wilkerson J. 2004. Performance characteristics of the sparkjet flow control actuator. Presented
at AIAA Flow Control Conf., 2nd, Portland, AIAA Pap. 2004-2131
Dearing S, Lambert S, Morrison J. 2007. Flow control with active dimples. Aeronaut. J. 111:705–14
Devenport W, Glegg S. 2009. Aeroacoustics of flow control. See Joslin & Miller 2009, pp. 353–72

www.annualreviews.org • Actuators for Active Flow Control 267

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Durscher R, Roy S. 2010. Novel multi-barrier plasma actuators for increased thrust. Presented at AIAA Aerosp.
Sci. Meet., Orlando, 48th, AIAA Pap. 2010-965
Enloe C, Corke T, Jumper E, Kachner K, McLaughlin T, Van Dyken R. 2004. Mechanisms and responses of
a single dielectric barrier plasma actuator: plasma morphology. AIAA J. 42:589–94
Fischer FA. 1955. Fundamentals of Electroacoustics. New York: Interscience
Font G. 2006. Boundary layer control with atmospheric plasma discharges. AIAA J. 44:1572–78
Forte M, Jolibois J, Pons J, Moreau E, Touchard G, Cazalens M. 2007. Optimization of a dielectric barrier dis-
charge actuator by stationary and non-stationary measurements of the induced flow velocity: application
to airflow control. Exp. Fluids 43:917–28
Gad-el Hak M. 2000. Flow Control: Passive, Active, and Reactive Flow Management. Cambridge, UK: Cambridge
Univ. Press
Gallas Q, Holman R, Nishida T, Carroll B, Sheplak M, Cattafesta L. 2003. Lumped element modeling of
piezoelectric-driven synthetic jet actuators. AIAA J. 41:240–47
Glezer A. 1988. The formation of vortex rings. Phys. Fluids 31:3532–42
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Glezer A, Amitay M. 2002. Synthetic jets. Annu. Rev. Fluid Mech. 34:503–29
Glezer A, Crittenden TM. 2003. Combustion-driven jet actuator. U.S. Patent No. 6,554,607
Godard G, Stanislas M. 2006. Control of a decelerating boundary layer. Part 3: optimization of round jets
vortex generators. Aerosp. Sci. Technol. 10:455–64
Gokoglu SA, Kuczmarski MA, Culley DE. 2009. Numerical studies of a fluidic diverter for flow control. Presented
by Brown University on 09/05/12. For personal use only.

at AIAA Fluid Dyn. Conf., San Antonio, 39th, AIAA Pap. 2009-4012
Greenblatt D, Wygnanski I. 2000. The control of flow separation by periodic excitation. Prog. Aerosp. Sci.
36:487–545
Gregory JW, Gnanamanickam EP, Sullivan JP, Raghu S. 2009. Variable-frequency fluidic oscillator driven
by a piezoelectric bender. AIAA J. 47:2717–25
Gregory JW, Ruotolo JC, Byerley AR, McLaughlin TE. 2007a. Switching behavior of a plasma-fluidic actuator.
Presented at AIAA Aerosp. Sci. Meet. Exhib., 45th, Reno, AIAA Pap. 2007-785
Gregory JW, Sullivan JP, Raghu S. 2004. Visualization of internal jet mixing in a fluidic oscillator. Presented at
Int. Symp. Flow Vis., 11th, Notre Dame, Indiana
Gregory JW, Sullivan JP, Lafayette W, Raman G, Raghu S. 2007b. Characterization of the microfluidic
oscillator. AIAA J. 45:568–76
Greska B, Krothapalli A, Seiner J, Jansen B, Ukeiley L. 2005. The effects of microjet injection on an F404 jet
engine. Presented at AIAA/CEAS Aeroacoust. Conf., 11th, Monterey, AIAA Pap. 2005-3047
Guyot D, Paschereit CO, Raghu S. 2009. Active combustion control using a fluidic oscillator for asymmetric
fuel flow modulation. Int. J. Flow Control 1:155–66
Hartmann J, Trolle B. 1927. A new acoustic generator. J. Sci. Instrum. 4:101–11
Hogue J, Brosche M, Oates W, Clark J. 2009. Development of a piezoelectric supersonic microactuator for
broadband flow control. In Proc. FCAAP: Florida Center Adv. Aero Propul. Conf., Orlando. Tallahassee, FL:
FCAAP
Hogue J, Solomon J, Hays M, Alvi F, Oates W. 2010. Broadband pulsed flow using piezoelectric microjets.
Proc. SPIE 7643:76431V
Holman R. 2006. An experimental investigation of flows from zero-net mass-flux actuators. PhD thesis. Univ.
Florida, Mech. Aerosp. Eng.
Holman R, Cattafesta L, Mittal R, Smith B, Utturkar Y. 2005. Formation criterion for synthetic jets. AIAA J.
43:2110–16
Ibrahim MK, Kunimura R, Nakamura Y. 2002. Mixing enhancement of compressible jets by using unsteady
microjets as actuators. AIAA J. 40:681–88
Ingard U, Labate S. 1950. Acoustic circulation effects and the nonlinear impedance of orifices. J. Acoust. Soc.
Am. 22:211–18
Jacobson SA, Reynolds WC. 1998. Active control of streamwise vortices and streaks in boundary layers.
J. Fluid Mech. 360:179–211
Jeon WP, Blackwelder RF. 2000. Perturbations in the wall region using flush mounted piezoceramic actuators.
Exp. Fluids 28:485–96

268 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Johnston JP, Nishi M. 2002. Vortex generator jets: means for flow separation control. AIAA J. 28:989–94
Jolibois J, Moreau E. 2009. Enhancement of the electromechanical performances of a single dielectric barrier
discharge actuator. IEEE Trans. Dielectr. Electr. Insul. 16:758–67
Joslin RD, Miller DN, eds. 2009. Fundamentals and Applications of Modern Flow Control. Prog. Aeronaut.
Astronaut. Vol. 231. Reston, VA: AIAA
Kastner J, Samimy M. 2002. Development and characterization of Hartmann tube fluidic actuators for high-
speed flow control. AIAA J. 40:1926–34
Katz Y, Nishri B, Wygnanski IJ. 1989. The delay of turbulent boundary layer separation by oscillatory active
control. Phys. Fluids A 1:179–81
Kegerise M, Cabell R, Cattafesta L. 2007a. Real-time feedback control of flow-induced cavity tones, Part 1:
Fixed-gain control. J. Sound Vib. 307:906–23
Kegerise M, Cabell R, Cattafesta L. 2007b. Real-time feedback control of flow-induced cavity tones, Part 2:
Adaptive control. J. Sound Vib. 307:924–40
Kerschen EJ, Cain AB, Raman G. 2004. Analytical modeling of Helmholtz resonator based powered resonance tubes.
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Presented at AIAA Flow Control Conf., 2nd, Portland, AIAA Pap. 2004-2691
Khan Z. 2000. On vortex generating jets. Int. J. Heat Fluid Flow 21:506–11
Kibens V, Dorris J III, Smith D, Mossman M. 1999. Active flow control technology transition—the Boeing ACE
program. Presented at AIAA Fluid Dyn. Conf., 30th, Norfolk, AIAA Pap. 1999-3507
Kim BH, Acharya M, Emo S, Williams D. 2005. Modeling pulsed-blowing systems for flow control. AIAA J.
by Brown University on 09/05/12. For personal use only.

43:314–25
Kim C, Jeon W-P, Park J, Choi H. 2003. Effect of a localized time-periodic wall motion on a turbulent
boundary layer flow. Phys. Fluids 15:265–68
Kirshner JM, ed. 1966. Fluid Amplifiers. New York: McGraw-Hill
Kotapati RB, Mittal R, Cattafesta L. 2007. Numerical study of a transitional synthetic jet in quiescent external
flow. J. Fluid Mech. 581:287–321
Kudar KL, Carpenter PW. 2007. Numerical investigation and feasibility study of a PZT-driven micro-valve
pulsed-jet actuator. Flow Turbul. Combust. 78:223–54
Kumar V, Alvi FS. 2006. Use of high-speed microjets for active separation control in diffusers. AIAA J.
44:273–81
Kunhardt E. 2000. Generation of large-volume, atmospheric-pressure, nonequilibrium plasmas. IEEE Trans.
Plasma Sci. 28:189–200
Likhanskii AV, Shneider MN, Opaits DF, Miles RB, Macheret SO. 2010. Limitations of the DBD effects on the
external flow. Presented at AIAA Aerosp. Sci. Meet., 48th, Orlando, AIAA Pap. 2010-470
Lockerby DA, Carpenter PW. 2004. Modeling and design of microjet actuators. AIAA J. 42:220–27
Lou H, Alvi FS, Shih C. 2006. Active and passive control of supersonic impinging jets. AIAA J. 44:58–66
Louste C, Artana G, Moreau E, Touchard G. 2005. Sliding discharge in air at atmospheric pressure: electrical
properties. J. Electrost. 63:615–20
MacMynowski DG, Williams DR. 2009. Flow control terminology. See Joslin & Miller 2009, pp. 59–72
Madou MJ. 1997. Fundamentals of Micromachining. New York: CRC
Mathew J, Song Q, Sankar B, Sheplak M, Cattafesta L. 2006. Optimized design of piezoelectric flap actuators
for active flow control. AIAA J. 44:2919–28
McCormick DC. 2000. Boundary layer separation control with directed synthetic jets. Presented at Aerosp. Sci.
Meet. Exhib., 38th, Reno, AIAA Pap. 2000-0519
McManus K, Magill J. 1996. Separation control in incompressible and compressible flows using pulsed jets. Presented
at AIAA Fluid Dyn. Conf., 27th, New Orleans, AIAA Pap. 1996-1948
Merhaut J. 1981. Theory of Electroacoustics. New York: McGraw-Hill
Moreau E. 2007. Airflow control by non-thermal plasma actuators. J. Phys. D Appl. Phys. 40:605–36
Moreau E, Sosa R, Artana G. 2008. Electric wind produced by surface plasma actuators: a new dielectric
barrier discharge based on a three-electrode geometry. J. Phys. D Appl. Phys. 41:115204
Morris NM. 1973. An Introduction to Fluid Logic. London: McGraw-Hill
Nagib H, Kiedaisch J, Wygnanski I, Stalker A, Wood T, McVeigh M. 2004. First-in-flight full-scale application
of active flow control: the XV-15 tiltrotor download reduction. Tech. Rep., Ill. Inst. Technol.

www.annualreviews.org • Actuators for Active Flow Control 269

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Narayanaswamy V, Raja LL, Clemens NT. 2010. Characterization of a high-frequency pulsed-plasma jet
actuator for supersonic flow control. AIAA J. 48:297–305
Oates WS, Liu F. 2009. Piezohydraulic actuator development for microjet flow control. ASME J. Mech. Des.
131:14–18
Papila M, Sheplak M, Cattafesta L. 2008. Optimization of clamped circular piezoelectric composite actuators.
Sens. Actuators A Phys. 147:310–23
Pons J, Moreau E, Touchard G. 2005. Asymmetric surface dielectric barrier discharge in air at atmospheric
pressure: electrical properties and induced airflow characteristics. J. Phys. D Appl. Phys. 38:3635–42
Prasad SAN, Gallas Q, Horowitz S, Homeijer B, Sankar BV, et al. 2006. An analytical electroacoustic model
of a piezoelectric composite circular plate. AIAA J. 44:2311–18
Pugliese AJ, Englar RJ. 1979. Flight testing the circulation control wing. Presented at Aircr. Syst. Technol. Meet.,
New York, AIAA Pap. 1979-1791
Raghu S. 2001. Feedback-free fluidic oscillator and method. U.S. Patent No. 6,253,782
Raju R, Aram E, Mittal R, Cattafesta L. 2009. Simple models of zero-net mass-flux jets for flow control
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

simulations. Int. J. Flow Control 1:179–97

Raju R, Gallas Q, Mittal R, Cattafesta L. 2007. Scaling of pressure drop for oscillatory flow through a slot.
Phys. Fluids 19:078107
Raman G, Cain AB. 2002. Innovative actuators for active flow and noise control. Proc. Inst. Mech. Eng. G. J.
Aerosp. Eng. 216:303–24
by Brown University on 09/05/12. For personal use only.

Raman G, Cornelius D. 1995. Jet mixing control using excitation from miniature oscillating jets. AIAA J.
33:365–68
Raman G, Hailye M, Rice E. 1993. Flip-flop jet nozzle extended to supersonic flows. AIAA J. 31:1028–35
Raman G, Khanafseh S, Cain AB, Kerchen E. 2004. Development of high bandwidth powered resonance tube
actuators with feedback control. J. Sound Vib. 269:1031–62
Raman G, Packiarajan S, Papadopoulos G, Weissman C, Raghu S. 2005. Jet thrust vectoring using a miniature
fluidic oscillator. Aeronaut. J. R. Aeronaut. Soc. 109:129–38
Raman G, Raghu S, Bencic T. 1999. Cavity resonance suppression using miniature fluidic oscillators. Presented at
AIAA Aeroacoust. Conf., 5th, Seattle, AIAA Pap. 1999-1900
Raman G, Rice EJ, Cornelius DM. 1994. Evaluation of flip-flop jet nozzles for use as practical excitation
devices. J. Fluids Eng. 116:508–15
Rathnasingham R, Breuer KS. 1997. Coupled fluid-structural characteristics of actuators for flow control.
AIAA J. 35:832–37
Reynolds WC, Parekh DE, Juvet PJD, Lee MJD. 2003. Bifurcating and blooming jets. Annu. Rev. Fluid Mech.
35:295–315
Rossi M. 1988. Acoustics and Electroacoustics. Norwood, MA: Artech House
Roupassov DV, Nikipelov AA, Nudnova MM, Starikovskii AY. 2009. Flow separation control by plasma
actuator with nanosecond pulsed-periodic discharge. AIAA J. 47:168–85
Roupassov DV, Starikovskii AY. 2008. Development of nanosecond surface discharge in actuator geometry.
IEEE Trans. Plasma Sci. 36:1312–13
Rowley CW, Batten BA. 2009. Dynamic and closed-loop control. See Joslin & Miller 2009, pp. 115–48
Rumsey C, Gatski T, Sellers W, Vatsa V, Viken S. 2006. Summary of the 2004 Computational Fluid Dynamics
Validation Workshop on synthetic jets. AIAA J. 44:194–207
Samimy M, Adamovich L, Webb B, Kastner J, Hileman J, et al. 2004. Development and characterization of
plasma actuators for high-speed jet control. Exp. Fluids 37:577–88
Samimy M, Kim JH, Kastner J, Adamovich I, Utkin Y. 2007. Active control of high-speed and high-Reynolds-
number jets using plasma actuators. J. Fluid Mech. 578:305–30
Santhanakrishnan A, Jacob JD. 2007. Flow control with plasma synthetic jet actuators. J. Phys. D Appl. Phys.
40:637–51
Sarpotdar S, Raman G, Cain AB. 2005. Powered resonance tubes: resonance characteristics and actuation
signal directivity. Exp. Fluids 39:1084–95
Schubauer GB, Skramstad HK. 1948. Laminar-boundary-layer oscillations and transition on a flat plate. Tech.
Rep. 909, NACA

270 Cattafesta · Sheplak

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Seele R, Tewes P, Woszidlo R, McVeigh MA, Lucas NJ, Wygnanski IJ. 2009. Discrete sweeping jets as tools
for improving the performance of the V-22. J. Aircr. 46:2098–106
Seifert A. 2007. Closed-loop active flow control systems: actuators. In Active Flow Control, ed. R King, pp. 85–
102. Notes Numer. Fluid Mech. Multidiscip. Des. Vol. 95. Berlin: Springer
Seifert A, Bachar T, Koss D, Shepshelovich M, Wygnanskil I. 1993. Oscillatory blowing: a tool to delay
boundary-layer separation. AIAA J. 31:2052–60
Seifert A, Darabi A, Wygnanski IJ. 1996. Delay of airfoil stall by periodic excitation. J. Aircr. 33:691–98
Seifert A, Eliahu S, Greenblatt D, Wygnanski IJ. 1998. Use of piezoelectric actuators for airfoil separation
control. AIAA J. 36:1535–37
Seiner J, Ukeiley L, Jansen B. 2005. Aero-performance efficient noise reduction for the F404-400 engine. Presented
at AIAA/CEAS Aeroacoust. Conf., 11th, Monterey, AIAA Pap. 2005-3048
Selby GV, Lin JC, Howard FG. 1992. Control of low-speed turbulent separated flow using jet vortex gener-
ators. Exp. Fluids 12:394–400
Shang J, Surzhikov S, Kimmel R, Gaitonde D, Menart J, Hayes J. 2005. Mechanisms of plasma actuators for
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

hypersonic flow control. Prog. Aerosp. Sci. 41:642–68

Sharma RN. 2007. Fluid-dynamics-based analytical model for synthetic jet actuation. AIAA J. 45:1841–47
Shaw LL, Smith BR, Saddoughi S. 2006. Full-scale flight demonstration of active control of a pod wake. Presented
at AIAA Flow Control Conf., 3rd, San Francisco, AIAA Pap. 2006-3185
Sherman C, Butler J. 2007. Transducers and Arrays for Underwater Sound. New York: Springer
by Brown University on 09/05/12. For personal use only.

Singh KP, Roy S. 2007. Modeling plasma actuators with air chemistry for effective flow control. J. Appl. Phys.
101:123308
Smith B, Glezer A. 1998. The formation and evolution of synthetic jets. Phys. Fluids 10:2281–97
Soloman JT, Kumar R, Alvi FS. 2008. Development and characterization of high bandwidth micro-actuator.
Proc. FEDSM2008: 2008 ASME Fluids Eng. Div. Summer Conf., Jacksonville, Florida, FEDSM2008-55032.
New York: ASME
Stanek MJ, Raman G, Ross JA, Odedra J, Peto J, et al. 2002a. High frequency acoustic suppression—the role of
mass flow, the notion of superposition, and the role of inviscid instability—a new model (Part II ). Presented at
AIAA/CEAS Aeroacoust. Conf. Exhib., 8th, Breckenridge, CO, AIAA Pap. 2002-2404
Stanek MJ, Raman G, Ross JA, Odedra J, Peto J, et al. 2003. High frequency acoustic suppression—the mystery of
the rod-in-crossflow revealed. Presented at Aerosp. Sci. Meet. Exhib., 41st, Reno, AIAA Pap. 2003-0007
Stanek MJ, Sinha R, Seiner J, Pierce B, Jones M. 2002b. High frequency flow control—suppression of aero-optics
in tactical directed energy beam propagation and the birth of a new model (Part I ). Presented at AIAA/CEAS
Aeroacoust. Conf., 8th, Breckenridge, CO, AIAA Pap. 2002-2272
Starikovskaia SM. 2006. Plasma assisted ignition and combustion. J. Phys. D Appl. Phys. 39:R265–99
Starikovskii AY, Nikipelov AA, Nudnova MM, Roupassov DV. 2009. SDBD plasma actuator with nanosecond
pulse-periodic discharge. Plasma Sources Sci. Technol. 18:034015
Tang H, Zhong S. 2009. Lumped element modelling of synthetic jet actuators. Aerosp. Sci. Technol. 13:331–39
Tesar V, Hung C, Zimmerman W. 2006. No-moving-part hybrid-synthetic jet actuator. Sensors Actuators A
Phys. 125:159–69
Theofilis V. 2009. Role of instability theory in flow control. See Joslin & Miller 2009, pp. 73–114
Thill C, Etches J, Bond I, Potter K, Weaver P. 2008. Morphing skins. Aeronaut. J. 112:117–39
Thomas FO, Corke TC, Iqbal M, Kozlov A, Schatzman D. 2009. Optimization of dielectric barrier discharge
plasma actuators for active aerodynamic flow control. AIAA J. 47:2169–78
Ukeiley LS, Ponton MK, Seiner JS, Jansen B. 2004. Suppression of pressure loads in cavity flows. AIAA J.
42:70–79
Unfer T, Boeuf JP. 2009. Modelling of a nanosecond surface discharge actuator. J. Phys. D Appl. Phys.
42:194017
Utkin YG, Keshav S, Kim JH, Kastner J, Adamovich IV, Samimy M. 2007. Development and use of localized
arc filament plasma actuators for high-speed flow control. J. Phys. D Appl. Phys. 40:685–94
Viets H. 1975. Flip-flop jet nozzle. AIAA J. 13:1375–79
Viets H, Piatt M, Ball M. 1981. Boundary layer control by unsteady vortex generation. J. Wind Eng. Ind.
Aerodyn. 7:135–44

www.annualreviews.org • Actuators for Active Flow Control 271

 FL43CH11-Cattafesta ARI 19 November 2010 16:20

Warsop C, Hucker M, Press AJ, Dawson P. 2007. Pulsed air-jet actuators for flow separation control. Flow
Turbul. Combust. 78:255–81
Williams D, Cornelius D, Rowley CW. 2007. Supersonic cavity response to open-loop forcing. In Active Flow
Control, ed. R King, pp. 230–43. Notes Numer. Fluid Mech. Multidiscip. Des. Vol. 95. Berlin: Springer
Williams DR, MacMynowski DG. 2009. Brief history of flow control. See Joslin & Miller 2009, pp. 1–20
Wiltse JM, Glezer A. 1993. Manipulation of free shear flows using piezoelectric actuators. J. Fluid Mech.
249:261–85
Wiltse JM, Glezer A. 1998. Direct excitation of small-scale motions in free shear flows. Phys. Fluids 10:2026–36
Yamaleev N, Carpenter M. 2006. Quasi-one-dimensional model for realistic three-dimensional synthetic jet
actuators. AIAA J. 44:208–16
Yamaleev NK, Carpenter MH, Ferguson F. 2005. Reduced-order model for efficient simulation of synthetic
jet actuators. AIAA J. 43:357–69
Zhang X. 2003. The evolution of co-rotating vortices in a canonical boundary layer with inclined jets. Phys.
Fluids 15:3693–702
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Zhuang N, Alvi FS, Alkislar MB, Shih C. 2006. Supersonic cavity flows and their control. AIAA J. 44:2118–28

RELATED RESOURCES
by Brown University on 09/05/12. For personal use only.

Choi H, Jeon W-P, Kim J. 2008. Control of flow over a bluff body. Annu. Rev. Fluid Mech.
40:113–39
Dowling AP, Morgans AS. 2005. Feedback control of combustion oscillations. Annu. Rev. Fluid
Mech. 37:151–82
Joslin RD. 1998. Aircraft laminar flow control. Annu. Rev. Fluid Mech. 30:1–29
Kasagi N, Suzuki Y, Fukagata K. 2009. Microelectromechanical systems–based feedback control
of turbulence for skin friction reduction. Annu. Rev. Fluid Mech. 41:231–51
Rowley CW, Williams DR. 2006. Dynamics and control of high-Reynolds-number flow over
open cavities. Annu. Rev. Fluid Mech. 38:251–76
Paduano JD, Greitzer EM, Epstein AH. 2001. Compression system stability and active control.
Annu. Rev. Fluid Mech. 33:491–517

272 Cattafesta · Sheplak

 FL43-FrontMater ARI 15 November 2010 11:55

Annual Review of

Contents Fluid Mechanics

Volume 43, 2011

Experimental Studies of Transition to Turbulence in a Pipe

T. Mullin p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Fish Swimming and Bird/Insect Flight
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Theodore Yaotsu Wu p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p25

Wave Turbulence
Alan C. Newell and Benno Rumpf p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p59
by Brown University on 09/05/12. For personal use only.

Transition and Stability of High-Speed Boundary Layers

Alexander Fedorov p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p79
Fluctuations and Instability in Sedimentation
Élisabeth Guazzelli and John Hinch p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p97
Shock-Bubble Interactions
Devesh Ranjan, Jason Oakley, and Riccardo Bonazza p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 117
Fluid-Structure Interaction in Internal Physiological Flows
Matthias Heil and Andrew L. Hazel p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 141
Numerical Methods for High-Speed Flows
Sergio Pirozzoli p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 163
Fluid Mechanics of Papermaking
Fredrik Lundell, L. Daniel Söderberg, and P. Henrik Alfredsson p p p p p p p p p p p p p p p p p p p p p p p 195
Lagrangian Dynamics and Models of the Velocity Gradient Tensor
in Turbulent Flows
Charles Meneveau p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 219
Actuators for Active Flow Control
Louis N. Cattafesta III and Mark Sheplak p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 247
Fluid Dynamics of Dissolved Polymer Molecules
in Confined Geometries
Michael D. Graham p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 273
Discrete Conservation Properties of Unstructured Mesh Schemes
J. Blair Perot p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 299
Global Linear Instability
Vassilios Theofilis p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 319

v
 FL43-FrontMater ARI 15 November 2010 11:55

High–Reynolds Number Wall Turbulence

Alexander J. Smits, Beverley J. McKeon, and Ivan Marusic p p p p p p p p p p p p p p p p p p p p p p p p p p p p 353
Scale Interactions in Magnetohydrodynamic Turbulence
Pablo D. Mininni p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 377
Optical Particle Characterization in Flows
Cameron Tropea p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 399
Aerodynamic Aspects of Wind Energy Conversion
Jens Nørkær Sørensen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 427
Flapping and Bending Bodies Interacting with Fluid Flows
Michael J. Shelley and Jun Zhang p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 449
Annu. Rev. Fluid Mech. 2011.43:247-272. Downloaded from www.annualreviews.org

Pulse Wave Propagation in the Arterial Tree

Frans N. van de Vosse and Nikos Stergiopulos p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 467
by Brown University on 09/05/12. For personal use only.

Mammalian Sperm Motility: Observation and Theory

E.A. Gaffney, H. Gadêlha, D.J. Smith, J.R. Blake, and J.C. Kirkman-Brown p p p p p p p 501
Shear-Layer Instabilities: Particle Image Velocimetry Measurements
and Implications for Acoustics
Scott C. Morris p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 529
Rip Currents
Robert A. Dalrymple, Jamie H. MacMahan, Ad J.H.M. Reniers,
and Varjola Nelko p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 551
Planetary Magnetic Fields and Fluid Dynamos
Chris A. Jones p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 583
Surfactant Effects on Bubble Motion and Bubbly Flows
Shu Takagi and Yoichiro Matsumoto p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 615
Collective Hydrodynamics of Swimming Microorganisms: Living Fluids
Donald L. Koch and Ganesh Subramanian p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 637
Aerobreakup of Newtonian and Viscoelastic Liquids
T.G. Theofanous p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 661

Indexes

Cumulative Index of Contributing Authors, Volumes 1–43 p p p p p p p p p p p p p p p p p p p p p p p p p p p p 691

Cumulative Index of Chapter Titles, Volumes 1–43 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 699

Errata

An online log of corrections to Annual Review of Fluid Mechanics articles may be found
at http://fluid.annualreviews.org/errata.shtml

vi Contents