Actuator, Hydraulic - Equations

The figure below will be used to develop the relationships and models.

In the figure,
pA Chamber A pressure [lbf/in2]
pB Chamber B pressure [lbf/in2]
QA Flow into Chamber A (+ flow into chamber, - flow out of chamber)
[in3/sec]
QB Flow into Chamber B (+ flow into chamber, - flow out of chamber)
[in3/sec]
Ap Actuator Piston Area for Chamber A [in2]
α Piston Area Ratio = AB/AA
QLi Piston Leakage Flow Rate (+ is flow from Chamber A to Chamber B)
[in3/sec]
Ff Friction force [lbf]
mp Mass of the piston and load [lbm]
FLoad External Load Force [lbf]
x , x& , &&
x Piston Position, Velocity & Acceleration, respectively [in, in/sec, in/sec2]
p p p

Incompressible, Steady State Relationships

The following equations can be used to quickly assess actuator flow and force characteristics
based on actuator geometry and inlet flow rates. These equations are approximations as they
assume the hydraulic fluid is incompressible and do not account for dynamic actuator/flow effects
(which can only be done via nonlinear simulation models).
For incompressible, steady state flow into an actuator, which implies a constant velocity of the

p −αp = F
actuator piston, the following relationships are helpful:
/A (1)
A B Load p
 Q = x& A Q = x& α A (2)
A p p B p p

A= p p = 1
Q x& A
x& α A α
(3)
Q
B p p

x& p = = B
Ap α Ap
QA Q
(4)

x& = p Q − α p Q [= ∆pQ if area ratio is 1]

The power relationship for an actuator is
F (5)
Load p A A B B

= p −αp
Also, by definition, load pressure and load flow are defined as
P Load Pressure (6)
Load A B
Q =Q −Q Load Flow (7)
Load A B
Load pressure and load flow are used in the discussion of servoactuators. These parameters

Note α = A A , which is the area ratio. For a symmetric actuator, α = 1.

represent the net amount of pressure and flow applied to the load.

B A
Actuator Dynamic Model
The dynamic equations for an actuator consist of a pressure derivative equation for each actuator
chamber and a 2nd order derivative equation governing piston motion. Hence an actuator model

(∑ Q − ∑ Q )+ V&
is a 4th order system. The equations for actuator chamber pressure is derived from the continuity

β′ β′
of mass equation

p& = (8)
V in out V

β′
( β′ &
)
the equations for the cylinder volumes are

p& = Q −Q −Q −

( )
V
A V A Li LeA V A

β′ β′ &
A A (9)
p& = Q +Q −Q + V
B V B Li LeB V B
B B
where
QLi internal leakage flow (in^3/sec)
QLe external leakage flow (in^3/sec)

V =V + x A
As a function of actuator piston position, the volumes are

V = V − x αA
A A0 p p
(10)
B B0 p p
Derivatives of the volumes are
V& = A x&
A p p
V& = −α A x&
(11)
B p p
where α is ratio of APB / APA. After substituting, the final pressure derivative equations are
β′  Q − A x& − Q − Q 
p& =
A V +A x  A p p Li LeA 

β′
A0 p p (12)
p& =  Q + α A x& + Q − Q 
B V − αA x  B p p Li LeB 

(pA − pB )
B0 p p
The cross chamber leakage is

Q =C (13)
Li Li
where CLi is the internal laminar leakage coefficient.

Usually external leakage is negligible so that QLeA = QLeB = 0. The pressure derivative equations
shown in (12) indicate that the pressure rate builds up proportionally (linearly) with flow into the
actuator chamber and that the pressure rate will be non-zero as long as there is flow into or out of
chamber. In other words, pressure increase/decrease behaves like an integrator, where the right

modulus, β’, is in the numerator and is in the 100,000 to 200,000 psi range, pressure rates are
hand side of the equation must be zero for the pressure to be constant. Also, since the bulk

very high for any reasonable flow rate.

( )
The equation for piston motion is derived from Newton’s 2nd law,

x = p − α p A − F  x&  − F
f  p
m && (14)
t p A B p load

m =m +m +m
where

t p A, fl B, fl
mp mass of the piston and load
mA,fl mass of the fluid in chamber A and pipeline
mB,fl mass of the fluid in chamber B and pipeline

= ρ[V + A x ]
The mass of the fluid in each chamber is
m
A, fl A0 p p

= ρ V − α A x 
(15)
 B0 p p 
m
B, fl
The mass of the fluid can be neglected if small relative to piston mass.

The friction equation is given by

F  x&  = F  x&  + F  x&  + F  x& 

f  p v  p c  p s  p
(16)

which has components of viscous, coulomb and Stribeck (seal) friction. More specifically,
  
  p 
F  x&  = σ x& + sign  x&   F + F exp  −
x&

f  p  p   c0
 c  
(17)

  
p s0

s
where
σ viscous friction coefficient (hydraulic viscosity)
Fc0 coulomb friction (mechanical friction)
Fs0 Stribeck friction (seal friction)
cs Stribeck velocity

A notional friction chart is shown in Figure 8. Actual friction values and coefficients should be
based on manufacturer data or actual hardware measurements.

Friction Force

Total Friction
Seal
Friction Coulomb Friction
Viscous Friction
- Velocity
+ Velocity
Viscous Friction
Coulomb Friction Seal
Friction
Total Friction

Figure 8 Actuator Friction Behaviors

As shown by equation (14), the behavior of an actuator piston is 2nd order (see frequency
response section). This implies that the step response of the piston will generally have some
overshoot and oscillations before settling into a commanded position. An example is shown in
Figure 9, where a actuator piston with a natural frequency of 500 rad/sec and damping ratio of
0.05 is given a 1 inch step command at 1 second.
 Actuator Response to 1" Step Input at t = 1 sec
2
Piston Position
1.8 Step Input

1.6

1.4
Amplitude (inches)

1.2

0.8

0.6

0.4

0.2

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (sec)

Figure 9 Actuator Piston Response to a Step Response

Using the previous equations, a nonlinear state space model for the piston motion is
   0 

( )
 x&1     
x
 = 1    −  1  Fload
2
 
 x&2    pA − α pB A p − F f  x& p     m 
 t   t
(18)
m
y=x
1
where x1 is the piston position and x2 is the piston velocity.

A state space model for the complete actuator is

   0 0 

( )
  
1 
x 0
2
 1      0 
 x&1   m  pA − α pB A p − F f  x& p     m 
t Q 

( )
0
   t  
 x&2   1   1  A 
 A x& − C 
 = p − p  + 0  Q 
 x& 3   Ch  p p A   C  B 
0
 hA  F
19)

( )
  
Li B
    load 
 4   1   
0 
A

x&
 α A x& + C p − p   0
1
 h     
 B  
C p p Li B A C
h
B
where x1 is the piston position and x2 is the piston velocity, x3 is pA and x4 is pB.

Hydraulic Natural Frequency Computation

Published data indicates hydraulic fluids have bulk modulus values of about 200,000 lb/in2;
however, due to entrained air in the fluid a more practical value of β fluid = 100,000 lb/in2 should be
used for any analysis calculations. This bulk modulus value provides a more conservative natural
frequency calculation. The equation for calculating an actuator’s natural frequency is:

β  A2 A2 
ω = =  bore + rod 
K
V 
fluid _ min fluid
(20)
 bore rod 
nat M M V
load load

ω
where

actuator/load natural frequency [rad/sec]

nat
β hydraulic fluid bulk modulus [lbf/in2]
fluid
M actuator load mass [(lbf*sec2)/in ]
load
A actuator effective area on the head side [in2]
bore
A actuator effective area on the rod side [in2]
rod
V fluid volume on rod side of actuator to the servovalve [in3]
rod
V fluid volume on bore side of actuator to the servovalve [in3]
bore
For an equal area actuator the equation can be reduced to:

4β A2
ω =
fluid
(21)
nat M V
load Total
where
A actuator effective area, equal rod and head area [in2]
Vtotal total fluid volume on the bore and rod sides of an actuator [in3]

Hydraulic Fluid Spring Rate

The fluid spring rate for an actuator is calculated as:

β * A2 β * A2
= +
fluid bore fluid rod
K (22)
fluid V V
bore rod

β
where
hydraulic fluid bulk modulus [lbf/in2]
fluid
A actuator effective area on the head side [in2]
bore
A actuator effective area on the rod side [in2]
rod
V fluid volume on rod side of actuator to the servovalve. Includes fluid volume
rod
in the actuator itself and fluid in connecting tubing to the servovalve. [in3]
 V fluid Volume on bore side of actuator. Includes fluid volume in the actuator
bore
itself and fluid in connecting tubing to the servovalve. [in3]

See Figure 10 to see how an unequal area actuator stiffness changes as a function of actuator
stroke. The actuator hydraulic stiffness is much larger when the actuator is near full retract than
when the actuator is near full extend. By reviewing equation (22), the actuator stiffness increases
by the actuator area squared and decreases proportionally to the volume.
Unequal Area Actuator Hydraulic Spring Rate As A Function of
Stroke

450000
400000
Actuator Stiffness (lb/in)

350000
300000
250000
200000
150000
100000
0
Figure 10 1 Spring Rate 2for an Unequal Area
Variation in Fluid 3 Actuator 4
Actuator Stroke (inch)
The following equation can be used to determine the actuator position where the minimum
stiffness will occur.
 
 1  
 3   Arod * S + Vrtube  −
V
btube
 A 
= 
A3
rod bore
+
X (23)
min_ Stiffness 1 1
A A
rod bore
where:
X actuator Position where the minimum stiffness occurs as measured from
min_ Stiffness
full retract [inch]
S total Actuator Stroke [inch]
V fluid volume in the hydraulic tube between the servovalve and the cylinder
rtube
on the actuator rod side [in3]
V fluid volume in the hydraulic tube between the servovalve and the cylinder
btube
on the actuator head side [in3].
A actuator effective area on the head side [in2]
bore
A actuator effective area on the rod side [in2]
rod
Once the actuator position where the minimum stiffness occurs is known, then the unequal
actuator minimum stiffness can be solved given the previous equations. Note if the actuator area
ratio “R” is relatively large and the tube volumes are relatively small compared to the actuator
fluid volumes then the minimum actuator stiffness can be assumed to occur at mid stroke.
See Figure 11 to see how an equal area actuator stiffness changes as a function of actuator
stroke. For an equal area actuator the minimum stiffness occurs when the actuator is at
midstroke, provided the fluid volumes in the connecting tubes are equal. The stiffness for an
equal area actuator is calculated by the following equation.

4*β * A2
=
fluid
K (24)
fluid _ min V
total

β
where
hydraulic fluid bulk modulus [lbf/in2]
fluid
A actuator effective area, equal rod and head area [in2]
V total fluid volume in the actuator [in3]
total

Typical Equal Area Actuator Hydraulic Spring Rate

Characteristic
100

90
(percent of maximum)
Actuator Spring Rate

80

70

60

50

40

30
0 20 40 60 80 100
Actuator Stroke (percent of total )

Figure 11 Variation in Fluid Spring Rate for an Equal Area Actuator

It should be noted that the above equations only calculate the hydraulic stiffness. The above
calculation assumes rigid hydraulic tubing between the servovalve and the actuator where the
length is kept to a minimum. If hydraulic hoses are used between the servovalve and the
actuator this will tend to reduce the hydraulic stiffness since hose are less rigid than steel tubing.
If hoses are used, a general guideline is to use four times the calculated fluid volume in the hose
when calculating the actuator hydraulic stiffness.
Other stiffnesses such as actuator mounting bracket and attachment structure will also tend to
lower the overall stiffness of the actuator installation.