Hydraulic and Pneumatic Actuators 1 PDF

Hydraulic & Pneumatic
Actuators
• References
• Introduction
• Fluid System Fundamentals
• Electrohydraulic Valve-Controlled
Servomechanism Case Study
• Pneumatic System Closed-Loop, Computer-
Controlled Positioning Experiment and Case
Study
Mechatronics: K. Craig
Hydraulic and Pneumatic Actuators 1
References
• Control of Fluid Power, D. McCloy & H.R.
Martin, 2nd Edition, John Wiley, New York, 1980.
• Hydraulic Control Systems, H. Merritt, John
Wiley, New York, 1967.
• System Dynamics, E. Doebelin, Marcel Dekker,
New York, 1998.
• Modeling and Simulation of Dynamic Systems, R.
Woods & K. Lawrence, Prentice Hall, New
Jersey, 1997.
• Introduction to Fluid Mechanics, R. Fox & A.
McDonald, John Wiley, New York, 1985.
• Control System Principles & Design, E. Doebelin,
John Wiley, New York, 1995.
• Fluid Power Control, J.R. Blackburn, G. Reethof,
and J.L. Shearer, The MIT Press, Cambridge, MA,
1960.
• The Analysis and Design of Pneumatic Systems, B.
Andersen, John Wiley, New York, 1967.
Introduction
• Applications of Hydraulic and Pneumatic
Actuators
• Hydraulic / Pneumatic Systems vs.
Electromechanical Systems
• Hydraulic Systems vs. Pneumatic Systems
Applications of Hydraulic & Pneumatic
Actuators
• Hydraulic and Pneumatic Control System components
include pumps, pressure regulators, control valves,
actuators, and servocontrols.
• Industrial Applications include automation, logic and
sequence control, holding fixtures, and high-power motion
control.
• Automotive Applications include power steering, power
brakes, hydraulic brakes, and ventilation controls.
• Aerospace Applications include flight-control systems,
steering-control systems, air conditioning, and brake-
control systems.
Hydraulic / Pneumatic Systems vs.
Electromechanical Systems
• Power Density Capability
– Electromagnetic motors, generators, and actuators are
limited by magnetic field saturation and can produce up
to about 200 pounds per square inch of actuator.
– In hydraulic systems, 3000 to 8000 pounds per square
inch of actuator is common in aircraft applications and
1000 pounds per square inch is common in industrial
applications.
– Hydraulic systems, actuators and generators, benefit
from a greater ratio of force per unit volume.
• Responsiveness and Bandwidth of Operation
– Electromagnetic actuators have a large inertia
associated with their motion, so they cannot accelerate
quickly.
– Hydraulic and pneumatic systems are more responsive
and have a greater bandwidth of operation at the same
power output levels.
• Heat Dissipation
– Fluid circulating to and from an actuator removes heat
generated by the actuator that is doing work.
– Electromechanical actuators and motors have limited
ability to dissipate heat generated inside the device and
rely on free or forced convection to the surrounding
environment.
– Heat is the predominant damaging mechanism in
electric and electronic systems.
– Reliability of electromagnetic devices is limited
compared to that of hydraulic and pneumatic devices.
• Modeling and Simulation
– Hydraulic and pneumatic systems generally have more
significant nonlinearities than do electric or mechanical
systems.
• Miscellaneous
– Electric power is more readily available, cleaner and
quieter, and easier to transmit, but may create electrical
interference with low-level data signals and can cause
overheating problems at low speeds.
– Hydraulic drives tend to be stiffer with respect to load
disturbances; intermittent and stalled operation without
damage is possible; but oil leakage, flammability, and
fluid contamination may pose problems.
When both
high speed
and high
power are
required,
hydraulic
systems my
be
mandatory.
Hydraulic Systems vs. Pneumatic Systems
• Compressibility of Fluid
– Speed of response: hydraulic systems have a rapid
initial response while pneumatic systems exhibit a time
delay.
– Stiffness to external load disturbances: pneumatic
systems exhibit a lack of stiffness, especially to external
load disturbances.
• Efficiency
– The efficiency of pneumatic systems is much lower than
that of hydraulic systems. Losses are caused by:
• Cooling of the fluid which dissipates energy in the form of heat
• Use of positive-displacement actuators causes loss of energy
by expansion.
• Leakage – it is difficult to achieve satisfactory sealing. This
also puts a limit on system pressure.
• Lubrication – The poor lubricating properties of gases
compared to oils also leads to energy losses.
• High-Temperature Applications
– The pneumatic system can work above 500°C where
both electrical and hydraulic systems would fail.
• Mathematical Model Complexity
– The basic equations of pneumatic systems are nonlinear
and tend to be more complex than those of
incompressible flow systems.
Fluid System Fundamentals
• Classification of Materials
• Fundamental Concepts & Properties of Fluids
• Basic Equations of Fluid Dynamics
• Lumped-Parameter Approach
• Passive Elements
– Fluid Resistance
– Fluid Capacitance
– Fluid Inertance
• Lumped vs. Distributed Fluid System Models
• Fluid Impedance
• Fluid Sources, Pressure, and Flow Rate
Classification of Materials
• Materials may be classified rheologically with

respect to their shear stress – shear strain behavior
in simple shear.
• Rheology is the science of the deformation and
flow of matter.
• In general, a fluid will undergo a continuous
deformation without rupture when subjected to a
constant anisotropic stress, where as a solid will
generally assume a static equilibrium
configuration under such conditions.
• An isotropic quantity is the same in any direction
from a given point in a system. Anisotropy
implies a dependence on direction or orientation.
• This type of behavior is relative and depends upon
the characteristic time required for the material to
respond to a change in stress or strain relative to
the time scale of observation, as well as the
magnitude of the stress or strain.
• What is silly putty? What is granite?
– Silly putty will fracture cleanly as a solid if subjected to
a large suddenly-applied stress, while it will flow freely
as a liquid when subjected to a constant stress of low or
moderate magnitude.
– Granite, normally considered a solid, will flow
measurably in large formations over a period of
geologic time under the influence of gravity alone.
• Basic axiom of rheology is that “everything
flows.”
• Classification of Materials: Idealized Models
– Spectrum of Material Classification in Simple Shear
Deformations
• Rigid Solid γ =0
• Linear Elastic Solid (Hookean) τ = Gγ (G = constant)
S
• Nonlinear Elastic Solid τ = G ( λ ) λ
• Viscoelastic τ = f ( γ, γ& ,t,... )
F • Nonlinear Viscous Fluid (Non-Newtonian) τ = η ( &γ ) γ&
• Linear Viscous Fluid (Newtonian) τ = µγ& (µ = constant)
• Inviscid Fluid τ = 0 γ = shear strain
γ& = shear rate
τ = shear stress
Fundamental Concepts & Properties of Fluids
• Basic Concepts
• Definition of a Fluid: Liquids and Gases
• Density
• Equation of State: Liquids and Gases
• Viscosity
• Propagation Speed
• Thermal Properties
• Reynolds Number Effects
• Classification of Fluid Motions
Basic Concepts
• Continuum
– Fluid is a continuum, an infinitely-divisible substance.
As a consequence, each fluid property is assumed to
have a definite value at each point in space. Fluid
properties are considered to be functions of position
and time, e.g., density scalar field ρ = ρ (x, y, z, t) and
r r
velocity vector field v = v ( x,y,z,t) .
• Velocity Field vr = vr ( x,y,z,t)
– Steady flow – all properties remain constant with time
at each point.
∂η
= 0 η is any fluid property
∂t
– One-, two-, three-dimensional flows – depends on the
number of space coordinates required to specify the
velocity field.
– Uniform flow – velocity is constant across any cross
section normal to the flow. Other properties may be
assumed uniform at a section.
– Timelines, pathlines, streaklines, and streamlines
provide visual representation of a flow. In steady flow,
pathlines, streaklines, and streamlines coincide.
– All fluids satisfying the continuum assumption must
have zero relative velocity at a solid surface (no-slip
condition) and so most flows are inherently two or
three dimensional. For many problems in engineering,
a one-dimensional analysis is adequate to provide
approximate solutions of engineering accuracy.
• Stress Field
– Surface forces: all forces acting on the boundaries of a
medium through direct contact.
– Body forces: forces developed without physical contact
and distributed over the volume of the fluid, e.g.,
electromagnetic and gravitational forces.
– Stresses in a medium result from forces acting on some
portion of the medium. The concept of stress provides
a convenient means to describe the manner in which
forces acting on the boundaries of the medium are
transmitted through a medium.
– State of stress at a point can be described completely by
specifying the stresses acting on three mutually
perpendicular planes through the point.
– A stress component is considered positive when the
direction of the stress component and the plane on
which it acts are both positive or both negative.
1st subscript: plane on which stress acts

2nd subscript: direction in which stress acts
σxx τxy τxz 
 
 τ yx σ yy τ yz 
 τzx τ zy σ zz 
Notation for Stress

Stress at a Point
Definition of a Fluid
• A fluid is a substance (a continuum) that deforms

continuously under the application of a shear
stress no matter how small the shear stress may be
and includes both liquids and gases. A fluid
cannot sustain a shear stress when at rest.
• Liquids are nearly incompressible.
• Gases are highly compressible.
• Liquids are distinguished from gases by orders of
magnitude differences in their density, absolute
viscosity, and bulk modulus.
Density
• The density ρ of a fluid is defined as the mass m

per unit volume V under specified conditions of
pressure P0 and temperature T0:
m
ρ=
V P0 ,T0
Equation of State: Liquids
• An equation of state is used to relate the density,

pressure, and temperature of a fluid.
• A relation derived from a Taylor series expansion
is valid over limited ranges of pressure and
temperature:
∂ρ ∂ρ
ρ(P,T) = ρ ( P0 ,T0 ) + ( P − P0 ) + ( T − T0 )
∂P P0 ,T0 ∂T P0 ,T0
 1 
= ρ0 1 + ( P − P0 ) − α ( T − T0 )
 β 
• β ⇒ Bulk Modulus (inverse of compressibility)
∂P ∂P
β = ρ0 =
∂ρ P0 ,T0 ∂ρ / ρ0 P0 ,T0
– The bulk modulus can be measured by noting the

change in pressure with a fractional change in volume
of a fixed mass of fluid: ∂P
β=−
∂V / V0
– Isothermal Bulk Modulus (or merely bulk modulus) β
can be used when the pressure changes occur at slow
enough rates during heat transfer to maintain constant
temperature.
– Adiabatic Bulk Modulus βa can be used when the rate of
pressure change is rapid enough to prevent significant heat
transfer.
– Cp/Cv, the ratio of specific heats, is only slightly greater
than 1.0 for liquids. ∂P Cp
β=− βa = β
∂V / V0 Cv
• α ⇒ Thermal Expansion Coefficient 1 ∂ρ
α=−
ρ0 ∂T P0 ,T0
– Thermal Expansion Coefficient α relates the incremental
change in volume with changes in temperature and can be
expressed in terms of volumes for a fixed mass of fluid:
∂V / V0
α= α = 0.5 x 10-3/°F for most liquids
∂T P0 ,T0
• The compressibility of power-transmitting fluids
is a major factor in determining systems resonant
frequency. This in turn puts limitations on the
speed of response of the output.
• The actual value of the bulk modulus of oil is
strongly dependent on the amount of air present in
the form of bubbles. Some level of entrained air is
impossible to avoid even in the best circuit
designs.
• An additional factor which will further reduce the
effective bulk modulus will be the type of
pipework used to contain the oil.
Equation of State: Gases
• The equation of state for an ideal gas is

P
ρ=
RT
• P and T are in absolute terms and R is the gas
constant. Most gases follow this ideal behavior
for considerable ranges of pressures and
temperatures.
• A gas undergoing a polytropic process follows the
relationship: P
= C = constant
ρ n
– n = 1.0 for an isothermal or very slow process

– n = k (ratio of specific heats) for an adiabatic or very
fast process
– n = 0.0 for an isobaric process
– n = ∞ for an isovolumetric process
• Bulk Modulus for a gas is:
∂P Cρ n
β = ρ0 = ρ0  nCρn −1  = ρ0 n = nP0
∂ρ P0 ,T0 P0 ,T0 ρ P0 ,T0
– The bulk modulus of a gas is thus related to the absolute
pressure of the gas.
– The bulk modulus of a liquid is on the order of 5000 to
15000 bar (1 bar = 1.0E5 N/m2) compared to 1 to 10 bar
for a gas.
Viscosity
• Absolute Viscosity (or dynamic viscosity) µ
(dimensions Ft/L2 or M/Lt) of a fluid represents
the ability of the fluid to support a shear stress τ
between a relative velocity u of the fluid and a
solid boundary.
τ shear stress
µ= =
∂u / ∂y shear rate
• A Newtonian fluid is a fluid in which shear stress
is directly proportional to shear rate. This is not
true for a non-Newtonian fluid.
• Kinematic Viscosity ν = µ/ρ (dimensions L2/t)
• Absolute viscosity of a liquid decreases markedly
−λ ( T −T )
with temperature: µ = µ 0e L 0
– µ0 and T0 are values at the reference conditions.

– λL is a constant that depends upon the liquid.
– In a liquid, resistance to deformation is primarily
controlled by cohesive forces among molecules.
• Absolute viscosity of a gas increases with
temperature µ = µ 0 + λ G ( T − T0 )
– µ0 and T0 are values at the reference conditions.
– λG is a constant that depends upon the liquid.
– In a gas, resistance to deformation is primarily due to
the transfer of molecular momentum.
• The absolute viscosity of most gases is almost
independent of pressure (from 1 to 30 bar).
• The viscosities of most liquids are not affected by
moderate pressures, but large increases have been
found at very high pressures.
Shear stress τ and time-independent
apparent viscosity behavior
η as a function of shear thinning n < 1
deformation rate
shear thickening n > 1
for one-
dimensional flow
of various non-
Newtonian fluids
Power-Law Model:
n n −1
Other Categories:
 du  du du du time-dependent Thixotropic
τyx = k   = k =η behavior
 dy  dy dy dy Rheopectic
Viscoelastic
Bingham-Plastic Model:
Under a constant applied shear stress:
du Thixotropic – decrease in η with time
τyx = τy + µ p
dy Rheopectic – increase in η with time
Propagation Speed
• Speed of propagation, c0, of a pressure signal in a
fluid depends on the bulk modulus and the
density:
β
c0 =
ρ
• The bulk modulus of a gas being perturbed at a

high speed is the ratio of specific heats, k, times
the absolute pressure of the gas, i.e., β = kP. Thus
the ratio β/ρ reduces to
c0 = kRT
• The speed of sound in air at 25°C is 347 m/s and
in a typical hydraulic fluid at 25°C is 1370 m/s.
Thermal Properties
• Specific heat of a fluid is the amount of heat
required to raise the temperature of a unit mass of
the fluid by 1 degree.
• The specific heat at constant pressure is Cp.
• The specific heat at constant volume is Cv.
• The specific heat ratio k = Cp/Cv.
• For some liquids, the specific heat ratio is
approximately 1.04. For gases, k is larger, e.g., k
= 1.4 for air.
Reynolds Number Effects
• Inertial flow forces ∝ ρAv2 F = ma = ρVa ∝ ρd 3a
ds dv dv
v= and a = ⇒ a=v
dt dt ds
ρd 3 v 2
F∝ = ρd 2 v 2 = ρAv 2
d
• Viscous flow forces ∝ µdv du v
τA = µ A ∝ µ d 2 = µvd
– v is the velocity of the fluid dy d
– d is a characteristic dimension associated with the physical

situation
• Flow conditions are often defined by means of a
dimensionless ratio called the Reynolds Number, the
ratio of inertial to viscous forces: N = ρd v = vd
2 2
µdv ν
r
• Laminar Flow - smooth, streamlined flow where
viscous forces dominate over inertial forces; no
macroscopic mixing of adjacent fluid layers.
• Turbulent Flow - swirling flow where inertial
forces dominate over viscous forces; macroscopic
mixing of adjacent fluid layers.
• At small Nr (< 1400) flow is laminar.
• At high Nr (> 3000) flow is turbulent and has
energy loss due to fluid collisions and mixing.
• For 1400 < N r < 3000, there is a transition from
laminar to turbulent flow. Here flow depends on
the local conditions and on the previous flow
history.
• Laminar flow: pressure loss due to friction has a
first-order relationship with the flow (analogous to
electrical resistance in which voltage is linear with
current).
• Turbulent flow: pressure loss becomes proportional
to the square of the flow. There is no universal
relationship between the stress field and the mean
velocity field; one must rely heavily on semi-
empirical theories and on experimental data.
• Systems with laminar flow are far simpler than
those with turbulent flow; however, laminar flow is
generally impractical for most systems because of
the low pressures and small dimensions it requires.
Classification of Fluid Motions
Basis of observable physical characteristics of flow fields
Continuum
Fluid Mechanics
Inviscid
Viscous
µ =0
Laminar Turbulent
Compressible Incompressible Internal External
– Incompressible Flow – density variations are
negligible. Most liquid flows are essentially
incompressible. However water hammer and cavitation
are examples of the importance of compressibility
effects in liquid flows. Gas flows may be considered
incompressible provided the flow speeds are small
relative to the speed of sound, e.g., < 30% of the speed
of sound, which corresponds to 100 m/s for air at
standard conditions.
– Compressible Flow – density variations are not
negligible.
Basic Equations of Fluid Dynamics
• The basis laws governing fluid motion are:

– Conservation of Mass
– Newton’s 2nd Law Not all are
– Moment of Momentum required to solve
– 1st Law of Thermodynamics any one problem.
– 2nd Law of Thermodynamics
• In addition one needs the equations of state or
constitutive equations that describe the behavior
of physical properties of fluids under given
conditions.
• Methods of Analysis
– A system or control volume must be defined before
applying the basic equations in the analysis of a
problem.
– A system is defined as a fixed, identifiable quantity of
mass; system boundaries separate the system from the
surroundings and they may be fixed or movable,
however, there is no mass transfer across the system
boundaries. Heat and work may cross the system
boundaries.
– A control volume is an arbitrary volume in space
through which fluid flows. The geometric boundary of
the control volume is called the control surface, which
may be real or imaginary, at rest or in motion, but it
must be clearly defined before beginning analysis.
– The basic laws can be formulated in terms of
infinitesimal systems or control volumes, in which case
the resulting equations are differential equations, whose
solutions provide detailed, point-by-point behavior of
the flow.
– When we are interested in the gross behavior of a
device, we use finite systems or control volumes. The
integral formulation of the basic laws is more
appropriate.
– The method of analysis depends on the problem!
– The method of description that follows a fluid particle
is called the Lagrangian method of description.
– The field, or Eulerian, method of description focuses
attention on the properties of the flow at a given point
in space as a function of time. The properties of the
flow field are described as functions of space
coordinates and time. This method of description is a
logical outgrowth of the fluid-as-a-continuum
assumption.
– Two Basic Reasons for using the control volume
formulation rather then the system formulation:
• Since fluid media are capable of continuous deformation, it is
often extremely difficult to identify and follow the same mass
of fluid at all times, as must be done in applying the system
formulation.
• We are often interested not in the motion of a given mass of
fluid, but rather in the effect of the overall fluid motion on
some device or structure.
• Conservation of Mass
– Mass of the system is constant.
 dM 
 dt  = 0
 sys
M sys = ∫ dm = ∫ ρdV
• Newton’s 2nd Law sys − mass sys − vol
– For a system moving relative to an inertial reference

frame, the sum of all external forces acting on the
system is equal to the time rate of change of the linear
momentum of the system.
ur
r  dP  uuur ur ur
( )
F =  Psys = ∫ Vdm = ∫ VρdV
 dt sys
sys
sys − mass sys −vol
• 1st Law of Thermodynamics: Conservation of
Energy
– Rate at which heat is added to the system plus the rate
at which work is done on the system by the
surroundings is equal to the rate of change of the total
energy of the system:
& =  dE 
& +W
Q  dt 
 sys
E sys = ∫ (e ) dm = ∫ (eρ)dV
sys − mass sys − vol
v2
e = u + + gz
2
• Each system equation, written on a rate basis, involves the
time derivative of an extensive property of the system. Let N =
any arbitrary extensive property of the system. Let η = the
corresponding intensive property (extensive property per unit
mass).
Nsys = ∫ ηdm = ∫ ηρdV
sys − mass sys − vol
• The equation relating the rate of change of the arbitrary

extensive property, N, for a system to the time variations of
this property associated with a control volume at the instant
when the system and control volume coincide is:
 dN  ∂ r uur
  = ∫
 dt sys ∂t CV
ηρdV + ∫ ηρv gdA
CS
– This equation says: The time rate of change of the
arbitrary extensive property of the system equals the
time rate of change of the arbitrary extensive property
within the control volume plus the net rate of efflux
(rate of outflow minus the rate of inflow) of the
arbitrary extensive property through the control surface.
– Note:
• Velocity is measured relative to the control volume.
• The time rate of change of the arbitrary extensive property
within the control volume must be evaluated by an observer
fixed in the control volume.
• Let’s apply this equation to the three basic laws:
– Conservation of Mass
– Newton’s 2nd Law
– Conservation of Energy
• Conservation of Mass
– The net rate of mass efflux through the control surface
plus the rate of change of mass inside the control
volume equals zero.
∂ r uur
0= ∫ ρdV + ∫ ρvgdA
∂t CV CS
– Velocity is measured relative to the control volume.

r uur
– ρvg dA is a scalar: it is positive where flow is out
through the control surface, negative where flow is in
through the control surface, and zero where flow is
tangent to the control surface.
• Newton’s 2nd Law for an Inertial Control Volume
– The sum of all forces acting on a non-accelerating
control volume equals the rate of change of momentum
inside the control volume plus the net rate of efflux of
momentum through the control surface.
r r r ∂ r r r uur
F = FS + FB = ∫ vρdV + ∫ v ρv gdA
∂t CV CS
– All velocities are measured relative to the control

volume.
– All time derivatives are measured relative to the control
volume.
• 1st Law of Thermodynamics: Conservation of
Energy uur
& & ∂ r
Q+ W = ∫ eρdV + ∫ eρvgdA
∂t CV CS
v2
e = u + + gz
2
& is positive when work is done on the control volume
–W
by the surroundings and this can take place only at the
control surface.
& =W
& & & &
W shaft + Wnormal + Wshear + Wother
Lumped-Parameter Approach
• Lumped-Parameter Approach
– Some areas have used this approach more than others,
e.g., electrical systems are originally conceived as a
combination of R, L, C, op-amps, and other integrated-
circuit modules, while mechanical systems are rarely
initially conceived in terms of a combination of K, B,
and M, but rather as combination of mechanisms,
actuators, sensors, and controllers.
– Fluid and Thermal systems follow a similar pattern in
that system dynamics may receive relatively light
conscious emphasis during the early conceptual phases.
– In addition, due to the generally less-well-defined shapes
of bodies of fluid and the fact that heat flow rarely is
confined to such simple and obvious paths as current,
these type systems may appear less well suited to the
lumped-parameter viewpoint.
– System dynamics can preserve the identity of individual
components while comprehending the entire system, and
thus often gives insights into needed design changes at
both the component and system levels.
– So, we consider system dynamics methods for fluid and
thermal machines and processes, even though they
initially seem less well suited to these more amorphous
systems.
– There are many practical examples of actual hardware
which have been successfully studied with this approach.
Passive Fluid Elements
• Resistance
• Capacitance
• Inductance
Resistance
• Like mechanical friction and electrical resistance,

this element performs the energy-dissipation
function.
• The dissipation of fluid energy into heat occurs in
all fluid devices to some extent.
• The simplest example is that of a fluid pipe.
– Consider the flow of a fluid in a constant-area, rigid-
walled conduit (as shown).
– Variables of primary interest are: average fluid pressure
p and the volume flow rate q.
– The average flow velocity v is defined as q/A, where A
is the conduit cross-sectional area.
– The product of p and q is fluid power, just as
mechanical power is fv and electrical power is ei.
– While the actual fluid pressure and velocity vary from
point to point over the flow cross section in a real fluid
flow, we assume a one-dimensional flow model in
which the velocity and pressure are uniform over the
area.
– Thus, the average velocity and average pressure
correspond numerically with the values at any point in
the cross section. These averages are spacewise
averages, not tmewise averages.
Lumped Model of a Fluid Pipeline
– In a lumped-parameter dynamic analysis, the pipeline is
broken into segments. Within each segment or lump,
the pressure and velocity may vary arbitrarily with
time, but are assumed uniform over the volume of the
lump. By considering the behavior of one typical lump
(the nth lump) we are led to definitions of all three basic
fluid elements.
– When a fluid is forced through a pipe at a constant
flow rate, a pressure difference related to that flow rate
must be exerted to maintain the flow. It takes a larger
pressure difference to cause a larger flow rate.
– In general, the relation between pressure drop and flow
rate is nonlinear; however, some situations give a
nearly linear effect.
– The pure and ideal fluid friction or fluid resistance
element is defined as:
p1 − p 2  psi 
Rf @
q  in 3 / sec 
– For nonlinear fluid resistances, we can define linearized
values in the neighborhood of an operating point.
– Now consider Conservation of Mass applied to the nth
lump over an infinitesimal time interval dt. During the
time interval dt, the difference between mass flow into
and out of the lump must equal the additional mass
stored in the lump. Mass enters the lump from the left
at a rate Avn-1ρ and leaves at the right at a rate Avnρ,
where ρ is the fluid mass density, which we treat as
being constant, corresponding to a constant operating-
point pressure and temperature.
– For constant density, conservation of mass is the same
as conservation of volume:
V Al
( Avn −1 − Av n ) dt = dV = dp n = dp n
β β
Al
( n −1 n )
q − q dt = dp n
β
1
pn = ∫ ( qn −1 − qn ) dt
Cf
Al
Cf @ @ fluid compliance (capacitance)
β
– We can easily see the electric-fluid analogy: electric
current analogous to net volume flow rate, pressure
analogous to voltage drop, and electrical capacitance
analogous to fluid capacitance.
– Newton’s 2nd Law states that the force (pressure times
area) difference between the left and right ends of a
lump must equal the lump mass times its acceleration.
For the nth lump the result is:
dv n dq n
Ap n −1 − Ap n − AR f q n = ρAl = ρl
dt dt
ρl dq n
( p n −1 − p n ) − R f q n =
A dt
– Since this equation contains both resistance (friction)
and inertance (inertia) effects, we consider each (in
turn) negligible, to separate them. With zero fluid
density we have:
( p n −1 − p n ) = R f q n
– If the resistance were zero, we have:
ρl dq n dq n
( p n − p n −1 ) = = If
A dt dt
ρl
If @ @ fluid inertance
A
– Again we see the electric-fluid analogy: pressure drop
analogous to voltage drop, volume flow rate analogous
to current, and fluid inertance analogous to electric
inductance.
– We will return to the fluid compliance and inertance
elements in more detail, since they occur in more
general contexts, not just pipelines. They were
introduced here to illustrate that the three elements are
always present whenever a body of fluid exists.
– Whether all three will be included in a specific system
model depends on the application and the judgment of
the modeler.
– Let’s consider fluid resistance in a more general way.
When a one-dimensional fluid flow is steady, the
inertance and compliance cannot manifest themselves
(even though they exist), and only the resistance effect
remains. We can thus experimentally determine fluid
resistance by steady-flow measurements of volume
flow rate and pressure drop, or if we attempt to
calculate fluid resistance from theory, we must analyze
a steady-flow situation and find the relation between
pressure drop and volume flow rate. If a nonlinear
resistance operates near a steady flow q0, we can define
a linearized resistance, good for small flow and
pressure excursions from that operating point.
– Just as in electrical resistors, a fluid resistor dissipates
into heat all the fluid power supplied to it. Fluid power
at a flow cross section is the rate at which work is done
by the pressure force at that cross section:
Power = pAv = pq
– For our assumed one-dimensional incompressible flow,
the power dissipated into heat is:
∆p 2
q ( p1 − p 2 ) = q∆p = q 2 R f =
Rf
– While we can often determine flow resistance values by
experimental steady-flow calibration, it is desirable to
be able to calculate from theory, before a device has
been built, what its resistance will be. For certain
simple configurations and flow conditions, this can be
done with fairly good accuracy.
– Refer to the previous discussion on Laminar and
Turbulent flow and the Reynolds Number.
– Laminar flow conditions produce the most nearly linear
flow resistances and these can be calculated from
theory, for passages of simple geometrical shape.
– For example, a long, thin flow passage called a
capillary tube of circular cross section has a fluid
resistance: πD4
q= ∆p
128µL
∆p 128µL
Rf @ =
q πD 4
– This is valid for laminar flow with end effects
neglected.
– Note that in the above calculations and experimental
measurements of flow resistances, the approach has
been to use formulas relating flow rate and pressure for
steady flows as if they held for general (unsteady)
flows. This approach is widely used, and usually of
sufficient accuracy; however, it should be recognized as
an approximation.
– Now consider orifices, where resistance is concentrated
in a short distance. The pressure drop across an orifice
is basically due to a conversion from the form of fluid
power, pressure times volume flow rate, to the power of
kinetic energy. From conservation of energy, for a
level flow of a frictionless incompressible fluid:
pressure/flow power + kinetic energy power = constant
Characteristics
of
Orifice Flow
– Considering any two locations 1 and 2, we may write:
p1q + (KE per unit time)1 = p2q + (KE per unit time)2
ρv12 q ρv 22 q
p1q + = p2 q +
2 2
ρ 2
p1 − p 2 = ∆p = ( v 2 − v12 )
2
– Since q = A1v1 = A2v2,
1 − ( A2 / A1 )
2 2 2
2∆p  q   q 
=  −  = q2
ρ  A 2   A1  A22
Basic
A2 2∆p pressure/flow
q=
1 − ( A 2 / A1 )
2 ρ relation
for an orifice
– Because of turbulence in the flow, viscosity of real
fluids, difficulties in measuring the areas of the fluid
stream, the practical formula used to predict the
pressure/flow relation for orifices in pipelines uses
experimental data, defines the pressure drop as that
between stations 1 and 3, and uses only the cross-
sectional area of the hole in the metal orifice:
2∆p
q = Cd Ao
ρ
– The value of the orifice discharge coefficient Cd

depends mainly on the Reynolds number and the area
ratio Apipe / Aorifice.
– The fluid resistances discussed so far are all intended to
be essentially constant in numerical form. Many fluid
systems require adjustable resistances, and often these
take the form of some kind of valve used to smoothly
modulate the flow rate. The vast majority of valves
have a square-root type of pressure/flow relation and
usually require experimental calibration if we want a
reasonably accurate flow model. The flow modulation
is achieved by somehow varying the “orifice” area Ao.
A complete calibration would give a family of flow rate
versus pressure drop curves, one for each flow area.
Compliance
• We have already seen that a fluid itself, whether a
liquid or gas, exhibits compliance due to its
compressibility.
• Certain devices may also introduce compliance into
a fluid system, even if the fluid were absolutely
incompressible.
– Rubber hoses will expand when fluid pressure increases,
allowing an increase in volume of liquid stored.
– Accumulators use spring-loaded cylinders or rubber air
bags to provide intentionally large amounts of
compliance.
– A simple open tank exhibits compliance, since an
increase in volume of contained liquid results in a
pressure increase due to gravity.
• In general, the compliance of a device is found by
forcing into it a quantity of fluid and noting the
corresponding rise in pressure. For liquids, the
input quantity is a volume of fluid V, and the ideal
compliance is:
V ∫ qdt
Fluid Compliance @ Cf @ @
p p
– For nonlinear compliances, the actual p-V curve can be
implemented in a computer lookup table, or the local
slope can be used to define an incremental compliance.
– We previously calculated the compliance of a section of
hydraulic line, due to bulk modulus of the liquid itself,
as: Al
β
– Additional contributions to compliance which may be
significant here are due to entrained air bubbles and the
flexibility of the tubing.
– Accumulators are devices intentionally designed to
exhibit fluid compliance, e.g., spring-loaded piston and
cylinder; flexible metal bellows; nitrogen-filled rubber
bag. Some devices can store large amounts of fluid
energy and are widely used for short-term power
supplies, pulsation smoothing, and to reduce pump size
in systems with intermittent flow requirements.
Accumulators
– Consider a vertical cylindrical tank of cross-sectional
area A supplied with a volume flow rate q; the pressure
at the tank inlet is p, liquid height is h. The vertical
motion of the liquid in such tanks is usually slow
enough that the velocity and acceleration have
negligible effects on the pressure p and it is simply
given by p = γh, where γ is the specific weight of the
liquid.
– Think of the tank as a large-diameter vertical pipe and
Rf (velocity effect) and If (acceleration effect) are
negligible relative to the height effect.
– If we add a volume V of liquid to the tank, the level h
goes up an amount V/A and the pressure rises an
amount Vγ/A.
Liquid and Gas Tanks as Fluid Compliances
– The tank compliance is thus:
volume change V A  in 3 
Cf = = = 
pressure change Vγ / A γ  psi 
– For non-cylindrical tanks the compliance effect is

nonlinear, but can be linearized in the usual way if
desired.
– Consider a rigid tank of volume V containing a gas at
pressure p. For slow (isothermal) pressure changes in
which fluid density is nearly constant, we may write
pV = MRT. If we force a mass dM = ρdV of gas into
the tank we cause a pressure change dp given by:
RT RT RT p
dp = dM = ρdV = dV
V V V RT
 dV  V  in 3 
Cf @   =  
  p= p0
dp p 0  psi 
– This is a linearized compliance useful for small changes
near an operating point p0. For rapid (adiabatic) but
still small pressure changes, analysis shows the
compliance is:
V
Cf = where k is the ratio of specific heats
kp 0
• Fluid Capacitance relates how fluid energy can be
stored by virtue of pressure.
• Law of Conservation of Mass (Continuity
Equation) for a control volume:
d d
& net =
m ( Mcv ) = (ρcv Vcv )
dt dt
& net
m & + V ρ&
= ρQnet = ρcv Vcv cv cv
• If all densities of the system are equal to ρ, then:
Qnet & V
= V + ρ&
ρ
• This assumption is justified for incompressible fluids
and is quite accurate for compressible fluids if
pressure variations are not too large and the
temperature of flow into the control volume is almost
equal to the temperature of flow out of the control
volume.
∂ρ ∂P ρ0 & ∂P
• Now ρ = ρ(P,T) ⇒ ρ& = = P β = ρ0
∂ P ∂t β ∂ρ P0 ,T0
& V&
Q = V + Pcv
β
V = 0 ⇒ Q = Pcv = Cf P& cv
& V&
• If the container is rigid, then β
• Cf is the fluid capacitance. Any large volume of a
compressible fluid becomes a capacitance.
• The continuity equation gives the differential
equation for the pressure inside the control
volume. The following equation should be used in
systems analysis (for systems with an inlet flow
with the same density as the control volume),
without repeating its derivation:
&+
Q=V
V
P& cv
β
β
P&CV = ( Q − V
&)
V
• The term V & can take several forms depending

upon the exact configuration of the system
(control volume).
• For Example:
– A spring-loaded piston of stiffness k and Area A can be

used as a capacitor.
– By applying a force balance equation, neglecting
inertial and frictional effects, we obtain:
A ( Pcv − Patm ) − kx = AδP − kx = 0
A &
x& = δP
k
– The volume of the cylinder is: V = A(x + x0 )
& = Ax&
V
– x0 is the minimum stroke representing the dead volume
space in the actuator and fittings.
2
– Substitution yields: & =
V
A
δP&
k
– The continuity equation can now be written as:
Q=V & + V P& cv
β
 V A2  & &
Q= +  δ P = C δ P
β
f
 k 
– The total capacitance is a combination of the fluid
compressibility effects and the mechanical compliance
of the container.
– Thus, fluid capacitance can be either a compliant
container or a volume of fluid.
– Increasing A or decreasing k increases the mechanical
capacitance term, and increasing V or decreasing β
increases the fluid compressibility capacitance.
– Capacitors for liquids are called accumulators. Typical
accumulators are spring-loaded pistons, bellows, and
gas-filled bladders for hydraulic systems.
– Since β is large for incompressible fluids, mechanical
types of capacitors are used. To obtain significant
compressibility capacitance with a liquid, V would
have to be very large.
– Capacitance of gaseous systems can be that of
mechanical capacitors or volume-type capacitance,
since β is low for compressible fluids.
• The effect of fluid capacitance must be considered
relative to the rest of the system. For example, the
resistance connected to the capacitor and the
bandwidth of interest determine how significant
the capacitive effects are.
Inertance (Fluid Inductance)
• While devices for introducing resistances
(orifices, capillaries) and compliance (tanks,
accumulators) are often intentionally designed into
fluid systems, the inertia effect is more often than
not a parasitic one.
• The inertance of pipes is perhaps most commonly
encountered and we will emphasize it.
– Any flowing fluid has stored kinetic energy because of
its density (mass) and velocity. The inertance of a
finite-size lump of fluid represents a summing up of
this kinetic energy over the volume of the lump.
– The simplest assumption possible here is that of one-
dimensional flow where all the fluid particles have
identical velocities at any instant of time. Since every
fluid particle has the same velocity, a lump of fluid can
be treated as if it were a rigid body of mass M = ρAL.
– A pressure drop ∆p across a pure inertance element will
cause a fluid acceleration according to Newton’s Law:
dv  1 dq 
A∆p = ρAL = ρAL  
dt  A dt 
ρL dq dq
∆p = @ If
A dt dt
ρL
If @
A
– This is analogous to e = L(di/dt) for inductance in
electrical systems and f = M(dv/dt) for mass in
mechanical systems.
Lumped vs. Distributed Fluid System Models
• Lumped-parameter models are always approximations to

the more-correct distributed models.
• The comparison of different models for a system is often
best done in terms of frequency response.
• Any practical dynamic system will experience input
signals whose maximum frequency will be limited to a
definite value.
• In lumped modeling, the variation of pressure and velocity
along the length of a pipeline was assumed a stepwise one.
Within a given lump there was no variation, but pressure
and velocity did change when we went to a neighboring
lump.
• All distributed models allow a smooth, not stepwise,
variation, which of course is more correct.
• It is clear that as a lumped model uses more and smaller
lumps, the stepwise variation more nearly approximates
the true smooth variation.
• How many lumps are needed to get accurate results with a
lumped model?
• Experience with many kinds of systems shows that if we
choose 10 lumps per wavelength of the highest frequency,
we usually get good results. That is, a stepwise variation is
an acceptable approximation to a sine wave if there are 10
steps per wavelength. If we decide that our lumped model
needs to be good for excitations of higher frequency than
we originally planned, the lumps must get smaller and
there must be more of them.
• Once we have a formula for the velocity of propagation
(the speed with which a disturbance propagates through
the medium) and choose the highest frequency of interest,
we can pick a size and number of lumps which will give
good accuracy up to that frequency, using the “10 lumps
per wavelength” guideline, where the wavelength λ is the
velocity of propagation c divided by the frequency f.
Higher operating frequencies require more and smaller
lumps.
• Remember why we use lumped-parameter models. They
can be solved easily for the time response to any form of
input and they also allow easy simulation with standard
software.
Fluid Impedance
• Most fluid systems do not really require the separation of
pressure/flow relations into their resistive, compliant, and
inertial components; this separation is mainly one of
analytical convenience.
• For complex fluid systems where experimental
measurements may be a necessity, the measurement of
overall pressure/flow characteristics has become a useful
tool.
• The term fluid impedance is directly analogous to
mechanical and electrical impedance and is defined as the
transfer function relating pressure drop (or pressure), as
output, to flow rate, as input.
• Fluid Resistance ∆p
(D) = Rf
q
∆p
( iω ) = R f
q
• Fluid Compliance ∆p 1
(D) =
q Cf D
∆p 1
( iω ) = ∠ − 90°
q ωCf
∆p
• Fluid Inertance ( D ) = If D
q
∆p
( iω) = ωIf ∠ + 90°
q
Fluid Impedance of the Three Basic Elements
• If the fluid impedance is known as an operational transfer
function, one can calculate the response to any given input
by solving the corresponding differential equation.
• If the fluid impedance is measured by the frequency
response technique, we then do not have a transfer
function in equation form, we only have curves. The
response to sinusoidal inputs is of course easily calculated
from such curves.
• Suppose, however, that we want to find the response to an
input which is not a sine wave but rather has a time
variation of arbitrary shape. Two methods are available:
– Curve-fit the measured frequency-response curves with analytical
functions, trying different forms and numerical values until an
acceptable fit is achieved. Software to expedite such curve-fitting
is available. Having a formula for the transfer function is the same
as having the system differential equation.
– Use the measured frequency-response curves directly, without any
curve fitting. One must compute the Fourier transform of the
desired time-varying input signal to get its representation in the
frequency domain. This operation converts the time function into
its corresponding frequency function, which will be a complex
number which varies with frequency. This complex number is
multiplied, one frequency at a time, with the complex number,
whose magnitude and phase can be graphed versus frequency.
This new complex number is the frequency representation of the
output of the system. The final step is the inverse Fourier
transform to convert this function back into the time domain, to
give the system output as a specific, plottable, function of time.
• This discussion applies to any linear, time-invariant,
dynamic system, not just fluid systems. That is, if we can
measure the frequency response, we can get the response
to any form of input, not just sine waves.
Fluid Sources, Pressure and Flow Rate
• An ideal pressure source produces a specified pressure at

some point in a fluid system, no matter what flow rate
might be required to maintain this pressure.
• An ideal flow source produces a specified flow rate,
irrespective of the pressure required to produce this flow.
• In fluid systems, the most common source of fluid power
is a pump or compressor of some sort.
• A positive-displacement liquid pump draws in, and then
expels, a fixed amount of liquid for each revolution of the
pump shaft. When driven at a constant speed, such a pump
closely approximates an ideal constant-flow source over a
considerable pressure range.
• Its main departure from ideal behavior is a decrease in
flow rate, due to leakage through clearance spaces, as load
pressure increases. This leakage flow is proportional to
pressure; thus one can represent a real pump as a parallel
combination of an ideal flow source and a linear (and
large) flow resistance Rfl.
• If the inlet flow impedance of the load is low relative to
Rfl, most of the flow goes into the load rather than the
pump leakage path, and the pump acts nearly as an ideal
flow source.
Positive-Displacement
Pump
as a
Flow Source
• When we need to manipulate a flow rate as function of
time, several possibilities exist.
– A fixed-displacement pump may be driven at a time-varying
speed. An electric motor drives the pump shaft, a flow sensor
measures the flow, and a feedback controller compares the desired
flow with the measured flow and commands the motor to change
speed so as to keep the actual flow close to the desired at all times.
– Instead of a fixed-displacement pump, a variable-displacement
pump could be used. Here the pump shaft speed is constant, but
pump output per revolution can be varied, while the pump is
running. A stroking mechanism allows flow rate to be varied
smoothly and quickly from full flow in one direction, through zero
flow, to full flow in the reverse direction. The stroking mechanism
could be driven directly or we could again use a feedback scheme.
• By combining a positive-displacement pump with a relief
valve, one can achieve a practical constant-pressure
source. This real source will not have the perfect
characteristic of an ideal pressure source, but can be
modeled as a combination of an ideal source with a flow
resistance.
• A relief valve is a spring-loaded valve which remains shut
until the set pressure is reached. At this point it opens
partially, adjusting its opening so that the pump flow splits
between the demand of the load and the necessary return
flow to the tank. To achieve this partial opening against
the spring, the pressure must change slightly; thus we do
not get an exactly constant pressure.
• This real source can be modeled as a series combination of
an ideal pressure source with a small flow resistance.
• These examples do not exhaust the possibilities with
regard to power sources in fluid systems, but they should
give some idea of how real sources may be modeled in
terms of ideal sources and passive elements.
• Other fluid power sources encountered in practice include
centrifugal pumps, accumulators (used for short-term
power supplies), elevated tanks or reservoirs (gravity is the
energy source), etc.

Hydraulic and Pneumatic Actuators 1 PDF

Uploaded by

Copyright:

Available Formats

Hydraulic and Pneumatic Actuators 1 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hydraulic and Pneumatic Actuators 1 PDF

Uploaded by

Copyright:

Available Formats

Hydraulic & Pneumatic

• Materials may be classified rheologically with

1st subscript: plane on which stress acts

Notation for Stress

• A fluid is a substance (a continuum) that deforms

• The density ρ of a fluid is defined as the mass m

• An equation of state is used to relate the density,

– The bulk modulus can be measured by noting the

• The equation of state for an ideal gas is

– n = 1.0 for an isothermal or very slow process

– µ0 and T0 are values at the reference conditions.

η as a function of shear thinning n < 1

• The bulk modulus of a gas being perturbed at a

– d is a characteristic dimension associated with the physical

Compressible Incompressible Internal External

• The basis laws governing fluid motion are:

– For a system moving relative to an inertial reference

• The equation relating the rate of change of the arbitrary

– Velocity is measured relative to the control volume.

– All velocities are measured relative to the control

• Like mechanical friction and electrical resistance,

– The value of the orifice discharge coefficient Cd

– For non-cylindrical tanks the compliance effect is

• If all densities of the system are equal to ρ, then:

• The term V & can take several forms depending

– A spring-loaded piston of stiffness k and Area A can be

• Lumped-parameter models are always approximations to

• An ideal pressure source produces a specified pressure at

You might also like