Ac PDF

Realization of Electronic Proportional (P) controller
e P Controller m
Error Controller
output

e Kp m
R1 R3
R2
e _ -m R3 _
m
+ +
Inverter
The controller output may be expressed as:

m = Kp .e
where Kp = R1/R2
and P = % proportional band = 100 / Kp
Realization of Electronic PID controller
(P, PI, & PD are special cases of PID)
The controller output may be expressed as:
 1 t de 
m = K p  e + 0 edt + TD 
 Ti dt 
where Ti = integral time constant
TD = derivative time constant
Kp de
m = K p e +  edt + K pTD
t
Or,
0
Ti dt
Taking Laplace transform with zero initial conditions,
K p E (s )
M (s ) = K p E (s ) + + sK pTD E (s )
sTi
M (s ) Kp
= Kp + + sK pTD
E (s )
Or,
sTi
M (s )  1 
= K p 1 + + sTD 
E (s )
Or,
 sTi 
the transfer function of PID controller
Parallel Realization of PID Controller
Kp
proportional term
E(s) Kp / sTi + M(s)
integral term
s Kp TD
derivative term
Circuit realization
R1
R2
_ v1 R3
+ R3
proportional term
CI
_
e m
RI +
_ v2 R3
+
integral term
Inverting adder
RD
CD
_ v3 R3
+
derivative term
M (s ) = −(V1 (s ) + V2 (s ) + V3 (s ))
 R1 1 
Or, M (s ) = E (s ) + + sRD CD 
 R2 sRI CI 
Circuit realization
Therefore, transfer function of PID controller becomes
M (s )  R1 1 
= + + sRD CD 
E (s )  R2 sRI CI 
Here,
R1
Kp =
R2
Kp 1
=
Ti RI C I
K pTD = RD C D
Circuit realization
Proportional and integral terms may be combined as:
R1 CI
R3
R2
e _ v1 R3 _
0V + m
+
From Derivative R3
ckt.
E (s ) V1 (s )
+ =0
R2 1
R1 +
sC I
E (s ) V1 (s ) V1 (s )sC I
Or, =− =−
R2 R1 +
1 1 + sR1C I
sC I
Circuit realization
V1 (s ) 1 + sR1C I
=−
E (s )
Or,
sR2 C I
 R1 1 
= −
 R + sR C  
 2 2 I 
R1  1 
=− 
 1+ 
R2  sR1C I 
 1 
= −K p 
1 + sT  
 i 
Where
R1
Kp = and Ti = R1CI
R2
A simple PID Controller with two op-amps
e +
_ m
e e +
R1 _ m
equivalent
e
CI circuit CI
RD _ RD
m
CD m +
CD +
RI Unity R2 RI m
_
gain
buffer
R2
Here, =
RI + R2
A simple PID Controller with two op-amps
e +
_ m
e e +
R1 _ m
equivalent
e
CI circuit CI
RD _ RD
m
CD m +
CD +
RI Unity R2 RI m
_
gain
buffer
R2
Here, =
RI + R2
The unity gain buffer amplifier is required to avoid the loading effect of
the feedback network
Circuit solution
 1 
M (s ) = I1  RI +
e +
e _ m Loop equations:  − I 2 RI
RD CI  sC I 
CD
RI
+
_ m  1 
I2 I1 0 = − I1 RI + I 2  RI + RD + 
 sC D 
In matrix form:
 1 
R +
 I sC − RI   I1    M ( s ) 
 I
    
 1   = 
 − RI RI + RD +   I 2   0 
 sC D
 
Circuit solution
1
e + RI +  M (s)
e _ m By Cramer’s rule: sCI
CI − RI 0
RD
I2 =
 +1 
CD + R
 sC − R 
_ m
RI I I
I2 I1
 I

 1 
−
 I R R + R +
sCD 
I D

 M ( s ) RI
Or, I2 =
 1  1 
 I
R +  I
R + RD +  − RI
2
 sC I  sC D 
 M ( s ) RI
 E (s) =
I2
=
( sCD )
sCD  1  1  2
 I
R +  I
R + RD +  − RI
 sCI  sCD 
Circuit solution
Then, the T.F. becomes:
e +
_ m
e
  1  1  2  
RD CI
  RI +  RI + RD +  − RI  sCD 
M ( s ) 1   sCI  sCD   
CD + =
RI _ m E (s)   RI 
 
I2 I1
 
1 1 + s (T1 + T2 + RI CD ) + s TT 
2
Simplifying, =  1 2

  1  1 sT 1  2  
  RI +  RI + RD +  − RI  sCD 
M ( s ) 1 1  T1 + TsC+I RI CD  1 sCD   
= =   2
+
 R + sT2 
E (s)    T  sT
 1 I 1  
 
where T1 = RICI and T2 = RDCD
Circuit solution
e + T1 + T2 + RI C D
e _ m By substituting A=
RD CI T1
CD +
RI _ m CD T2
I2 I1 Or, A = 1+ +
CI T1
This expression of A presents the problem of interaction and hence,

the controller developed is called an interacting controller.
Circuit solution
e + T1 + T2 + RI C D
e _ m By substituting A=
RD CI T1
CD +
RI _ m CD T2
I2 I1 Or, A = 1+ +
CI T1
The T.F. becomes:

A simplified PID Controller with one op-amp
e +
_ m
e e +
R1 _ m
equivalent
e
CI circuit CI
RD _ RD
Original circuit with two op-amps
m
CD m +
CD +
RI Unity R2 RI m
_
gain
buffer
The unity gain buffer amplifier may be omitted if the resistances are chosen such that
(R I
|| RD )  (R1 || R2 )
e +
_ m
CI R1 + R 2
TD
adjust Final circuit with one op-amp
CD RD m adjust
RI
Ti
adjust
Desirable features of a Process Controller
➢ Provision for providing a bias term
➢ Provision for anti derivative kick
➢ Provision for anti-integral wind-up
➢ Output must be limited (saturation)
➢ Provision for Auto/Manual modes of control
➢ Provision for Local/Remote modes of operation

Provision for providing a bias term
 1 t de 
The controller output is m = K p  e + 0 edt + TD  + bias
 Ti dt 
Bias may be properly adjusted for zero steady state error with P and PD
controllers, when load is constant. Proper bias can also ensure efficient
operation under start-up condition.
Non-zero steady state error
P-control without bias

Provision for providing a bias term
 1 t de 
The controller output is m = K p  e + 0 edt + TD  + bias
 Ti dt 
Bias may be properly adjusted for zero steady state error with P and PD
controllers, when load is constant. Proper bias can also ensure efficient
operation under start-up condition.
P-control without bias P-control with bias

Provision for anti-derivative kick
Derivative action may produce an unwanted kick at the controller output

when there is a step change in the set-point. This effect may be
eliminated if the derivative term is computed from the measured variable,
instead of the error (anti-derivative kick feature).
PD-control with derivative kick PD-control with anti-derivative kick

Provision for anti-integral wind-up
Integral action in PI and PID controllers may produce a large integral

error term when a non-zero error persists for a long time. Integration
action may be switched off to combat this situation, otherwise, a long
time may be required to come back to normal working range.
In some controllers, when the integral term is present, the integral

amount during Ti is summed with each pass through the calculation and
becomes the controller bias. This technique is known as automatic reset
and this is done to avoid long duration operation of the integral action –
thus error due to drift etc. during integration may be avoided.
Output must be limited (saturation)
It is desirable to limit the controller output (say between 0% and 100%) so

that control valves or other final control elements may operate safely
within their working limits.
Provision for Auto/Manual modes of control
A change-over switch is normally provided for configuring the controller

as an automatic controller (AUTO) for closed loop operation or as a
manual controller (MAN) for open-loop operation.
Provision for Auto/Manual modes of control
A change-over switch is normally provided for configuring the controller

as an automatic controller (AUTO) for closed loop operation or as a
manual controller (MAN) for open-loop operation.
Manual command
MAN
Final
Error Controller control Actuating signal
AUTO element
Change-over
switch
Provision for Local/Remote modes of operation
A provision is made for the set point input, such that, the set point may
be changed either locally or from a remote link.
Electronic Process Controllers
Advantages
➢ No transmission lag (as compared to pneumatic systems)
➢ Absence of dead zone due to friction or hysteresis in moving parts
➢ Capable of operating at low temperatures
➢ Compatibility with other electrical components

Electronic Process Controllers
Disadvantages
➢ Prone to electrical hazards (in inflammable atmosphere)
➢ Analog integrators are not very reliable
➢ Conversion equipments are necessary to interface pneumatic and hydraulic devices

Pneumatic Baffle-Nozzle or Flapper-Nozzle amplifier
Fulcrum
+ a nozzle Restriction (orifice)

Set point Constant pressure
x air supply (P s) (20 psi)
r
displacement b Baffle m
Error (or flapper) Output pressure (To control valve)
e (Nozzle back pressure)
+ +
Cm (joint)
Measured
variable
e = (r - C m)/2
Power source
Set point +
Output
Measured
variable
-
M (Nozzle back pressure)

Fulcrum
Ps
Set point Constant pressure Typical 15 psi approx. linear working
x air supply (P s) (20 psi) values range
r
displacement b Baffle 3 psi
m Pa
0 Baffle-Nozzle
e (Nozzle back pressure) X
0.001" 0.002" separation
+ +
Cm (joint) Typical
Measured values
variable A typical curve relating nozzle back pressure M to Baffle-Nozzle separation X
e = ( r - C m)/2
x: change in baffle-nozzle separation.

m: change in back pressure.
Pa : the lowest possible pressure ( = ambient pressure).
M (Nozzle back pressure)
Ps
Typical 15 psi approx. linear working

values range
3 psi
Pa
0 Baffle-Nozzle
0.001" 0.002"
X
separation
Typical
values
A typical curve relating nozzle back pressure M to Baffle-Nozzle separation X
Fulcrum

Set point Constant pressure
x air supply (P s) (20 psi)
r
displacement b Baffle m
e (Nozzle back pressure)
+ +
Cm (joint)
Measured
variable
e = (r - C m)/2
Under steady state condition, change in output pressure may be expressed as:
m = − K n  x , where –Kn is the nozzle gain

 a 
Here, −x= e
a +b
 a   a 
Thus, m = K n  e  = K e , K is called the amplifier gain and K = Kn  
 a + b   a + b 
Arrangement for mechanical set-point
and pneumatic measured variable
set point
(Screw)
e
Bellows
pneumatic
measured
variable C m
Bellows
Relay valve or pilot valve
The output pressure from the pneumatic amplifier is not suitable for driving the final
control element due to the presence of restriction in the air supply.
For this reason, a buffer stage, known as relay valve or pilot valve, is added at the
output to allow sufficient air flow at pressure ‘m’.
A relay valve may be direct-acting (positive gain) or reverse-acting (negative gain).

Direct-acting relay valve with a Baffle-Nozzle amplifier
x Restriction
m Air supply (P s)
Vent
Relay valve (gain = 1)

e
Bellows Relay output
Relay
m
m = - K n.x
As the nozzle back pressure m increases, the relay output pressure also increases
Direct-acting relays
Bleed type relay

In all positions of the valve, except at the
position to shut off the air supply, air
continues to bleed into the atmosphere.
Reverse-acting relay
As the nozzle back pressure increases, the ball valve is forced towards the lower seat, thereby
decreasing the output pressure
Pneumatic Proportional Controller
Feedback bellows
m m (Controller
output pressure)
(3 - 15 psi)
a
air supply (P s)
b Restriction (20 psi)
x
e
Error
The Baffle-Nozzle separation may be expressed as:
 a   b 
− ( x) =  e −  Kb m
 a+b  a+b
where Kb is the bellows stiffness factor and m is the change in output pressure.
Feedback bellows
 a   b 
m m (Controller
− ( x) =  e −  Kb m
 a+b  a+b
output pressure)
(3 - 15 psi)
a
air supply (P s)
Now m = – Kn .x, where Kn is the nozzle
b
x Restriction (20 psi) gain.
e
Error
Block diagram of the controller
a ++ -X(s)
E(s) Kn M(s)
a+b -
b
Kb
a+b
Feedback bellows
 a   b 
m m (Controller
− ( x) =  e −  Kb m
 a+b  a+b
output pressure)
(3 - 15 psi)
a
air supply (P s)
Now m = – Kn .x, where Kn is the nozzle
b
x Restriction (20 psi) gain.
e
Error
Thus,  a   b 
m = e−
K n  a + b   a + b  b
Km
1  b    a 
or m +  Kb  =  e
 Kn  a +b    a +b 
1
if Kn is very high, then  0,
Kn
Feedback bellows
m m (Controller
output pressure)
(3 - 15 psi)
a
air supply (P s)
b Restriction (20 psi)
x
e
Error
This gives
 a 
   a 
 a+b
m= e=  e = K p .e
 b   bK b 
  Kb
 a+b
a
where Kp = proportional gain =
bK b
Feedback bellows
 a 
m m (Controller
   a 
 + 
output pressure)
a b
m= e=  e = K p .e
(3 - 15 psi)
a
 b   bKb 
air supply (P s)
  b
K
b
x Restriction (20 psi)
 a+b
e a
Error where Kp = proportional gain =
bK b
Simplified block diagram of the controller
E(s) Kp M(s)
Pneumatic proportional controller with a direct-acting relay
Feedback bellows
Controller output
m m
(3 - 15 psi)
a
m Relay valve (gain = 1)
Air supply (Ps)

b x Restriction (20 psi)
e
Error
Pneumatic Proportional-Derivative Controller
Derivative time
Feedback bellows
Restriction R D (Needle valve)
CD m
(output pressure)
a Baffle
Nozzle Restriction (orifice)

air supply (P s)
b R
x
e
Error
The Baffle-Nozzle separation may be expressed as:
 a   b   Kb 
−X (s) =   E (s) −    M (s) (assuming RD >> R)
 a+b   a + b   1 + sTD 
where Kb = Bellows stiffness factor,
TD = Derivative time
= RDCD, (assuming RD >> R),
CD = Capacity of the bellows.
Pneumatic Proportional-Derivative Controller
Derivative time
Feedback bellows
Restriction R D (Needle valve)
 a   b   Kb 
CD m −X (s) =   ( ) 
E s −   M (s)
 a+b   a + b   1 + sTD 
(output pressure)
a Baffle
M (s ) = − K n  X (s )
Restriction (orifice)
Nozzle
air supply (P s)
Now
b R
e
x
where Kn is the nozzle gain.
Error
M (s)  a   b   Kb 
Thus =  E (s) −    M (s)
Kn  a + b   a + b  1 + sTD 
 1  b   Kb    a  1
or, M (s)  +   =   E (s) Now, K n  1,  0,
 K n  a + b   1 + sTD    a + b  Kn
M (s)  a 
   (1 + sTD )
E (s)  bK b 
a
= K p (1 + sTD ) where Kp = proportional gain =
bK b
Pneumatic Proportional-Integral Controller
Integral time
Restriction (Needle valve)
RI proportional
bellows
(output
CI m m
pressure)
a
Integral negative
positive bellows feedback
feedback Restriction (orifice)
air supply (P s)
e b x R
Error
The Baffle-Nozzle separation may be expressed as: (assuming RI >> R and

same stiffness for both bellows)
 a   b   b   Kb 
−X (s) =   E ( s ) −   b
K M ( s ) +    M (s)
 a + b   a + b   a + b  1 + sTi 
where Ti = Integral time

= RICI
and CI = Capacity of integral bellows
Kb = Bellows stiffness factor.
Integral time
RI proportional
bellows
 a   b   b   Kb 
−X (s) =  ( ) ( )  M (s)
(output
CI m m
pressure)
 E s −   K M s +  
+ + + +
b
Integral
a
negative
 a b   a b   a b  1 sTi 
feedback Restriction (orifice)
air supply (P s)
e b x R
Error
Hence,
 a   1  b 
−X (s) =   ( ) 
E s − 1 −   Kb M ( s )
 a + b   1 + sTi  a + b 
 
 a   1  b 
=  E (s) −    Kb M ( s )
 a+b  1+ 1  a + b 
 sT 
 i 
Now, M ( s ) = − Kn . X ( s )
 1  b 
Integral time
 a 
−X (s) =   ( )   Kb M ( s )
E s − 1 − 
RI
 a + b   1 + sTi  a + b 
proportional
 
bellows
(output
CI m m
pressure)
 a   1  b 
=  E (s) −    Kb M ( s )
a
Integral negative
Restriction (orifice)  a+b  1+ 1  a + b 
 sT 
feedback
air supply (P s)
e b x R  i 
Error
and M ( s ) = − Kn . X ( s )
1
Thus, assuming Kn >> 1, i.e.  0,
Kn
M (s) a  1   1  a
= 1 +  = K p 1 +  , where Kp =
E ( s ) bKb  sTi   sTi  bK b
Pneumatic Proportional-Integral-Derivative Controller
Integral time
Restriction
RI
1
CI CD
Derivative time
a RD Restriction
(output pressure)
m
x
Air supply (P s)
b R
e Restriction
Error
The Baffle-Nozzle separation may be expressed as: (assuming RI>>RD>> R)

 a   b  K b 
− X (s ) =   E (s ) −    M (s )
 a + b   a + b  1 + sTD 
 b  Kb 
 M (s )
+ 
 a + b  (1 + sTi )(1 + sTD ) 
 a   bK b  1 1 
=  E ( s ) −   −  M (s )
a+b  a + b  (1 + sTD ) (1 + sTi )(1 + sTD ) 
Integral time
Restriction
 a   b  K b 
− X (s ) =   E (s ) −   M (s )
RI
1

a +b  a + b  1 + sTD 
CI CD
 b  Kb 
 M (s )
+ 
 a + b  (1 + sTi )(1 + sTD ) 
Derivative time
RD
 bK b  
a Restriction
 a  1 1
 E (s ) −   M (s )
(output pressure)
=  −
 a + b  (1 + sTD ) (1 + sTi )(1 + sTD ) 
m
x a+b
Air supply (P s)
b R
e Restriction
Error
Now, 1 1 1 + sTi − 1
− =
1 + sTD (1 + sTi )(1 + sTD ) (1 + sTi )(1 + sTD )
sTi
=
1 + sTi + sTD + s 2TiTD
1
=
1 TD
1+ + + sTD
sTi Ti
Integral time
Restriction
 a   b  K b 
− X (s ) =   E (s ) −   M (s )
RI
1

a +b  a + b  1 + sTD 
CI CD
 b  Kb 
 M (s )
+ 
 a + b  (1 + sTi )(1 + sTD ) 
Derivative time
RD
 bK b  
a Restriction
 a  1 1
 E (s ) −   M (s )
(output pressure)
=  −
 a + b  (1 + sTD ) (1 + sTi )(1 + sTD ) 
m
x a+b
Air supply (P s)
b R
e Restriction
Error
Therefore,
 
 
 a   bK b  1  M (s )
− X (s ) =   E ( s ) −  
a+b  a + b  1 + 1 + TD + sT 
 sT T D 
 i i 
M (s ) = − K n  X (s )
Now,
Integral time
Restriction
RI
1
 
 
 a   bK b 
CD
− X (s ) =   E (s ) − 
1  M (s )
CI

a+b 
 a + b  1+ 1 T 
Derivative time
 + D
+ sTD 
 
RD
a Restriction
(output pressure) sTi Ti
m
x
M (s ) = − K n  X (s )
Air supply (P s)
b R
Restriction
and
e
Error
Therefore,
 
 
M (s )  a   bK b  1  M (s )
=  E (s ) −  
Kn a+b  a + b  1 + TD + 1 + sT 
 D 
 Ti sTi 
  
 1  bK   
 1   a 
or, M (s ) +  b
  =   E (s )

 Kn  a + b  1 + D +
T 1   a+b
 T sT + sTD 
  i i 
Integral time
Restriction
RI
1
  
 1  bK  
CI CD
M (s ) + b

1  =  a  E (s )
a RD
Derivative time
 K n  a + b  1 + TD + 1 + sT   a + b 
 D 
Restriction
(output pressure)
m   Ti sTi 
x
Air supply (P s)
R 1
K n  1, 0
b
e Restriction
as
Error Kn
Therefore,
M (s )  a   TD 1   TD 1 
=   1 + + + sTD  = K p 1 + + + sTD 
E (s )  bKb   Ti sTi   Ti sTi 
a
where Kp =
bK b
Integral time
Restriction
RI
1
CI CD
Derivative time
a RD Restriction
(output pressure)
m
x
Air supply (P s)
b R
e Restriction
Error
M (s )  a   TD 1   TD 1  a
=   1 + + + sTD  = K p 1 + + + sTD  where Kp =
E (s )  bKb   Ti sTi   Ti sTi  bK b
Here,
 TD 
  imposes an interaction between integral and derivative operations of the controller. If
 Ti  we choose Ti >> TD, the interaction reduces and the transfer function becomes
M (s )  1 
 K p 1 + + sTD 
E (s )
the ideal relation of a PID controller.
 sTi 
Integral time
Restriction
RI
1
CI CD
Derivative time
a RD Restriction
(output pressure)
m
x
Air supply (P s)
b R
e Restriction
Error
The controller gain becomes infinite when Ti = TD, if derivative time

restriction is placed in position (1).
Digital Controller in a Process Control Loop
All the desirable features of a process controller (e.g. anti-integral wind-up,

auto/manual modes of operation with bump-less transfer etc.) may be easily
incorporated while maintaining the high accuracy and precision of digital
systems.
No drift problem is encountered in a digital system.
In a processor based digital controller, rapid switching from one algorithm

to another (e.g. a P controller to a PID controller) and automatic tuning of
controller parameters are possible.
+ e en mn m Final V C
r Digital
+ ADC Controller
DAC Control Process
Set point _ Error
Error controller Element
sequence output Controller
sequence output
ADC equivalent DAC equivalent

e en mn m
 C (hold
Sampler capacitor)
Sample & hold

e mn
Controller
Error output
 sequence
time n
0 0 1 2 3
en  
DAC m
Error
output
sequence
m
n time
0 1 2 3 0  
  Sampling interval   (ADC conversion time + computation time)

The error computation may be performed digitally if the set-point is available in

digital form (say, from digital keyboard) and the measured variable is digitized with
the ADC.
Digital System
en mn m Final V C
rn Digital
+ Controller
DAC Control Process
+ Error
Set point _ Element
sequence sequence
(digital)
Cn
output
ADC
sequence
Proper selection of the sampling interval ‘’ is necessary for satisfactory operation of
the process control loop.
A large ‘’ may lead to unstable operation of the loop (because of the extra lag
introduced in the loop), whereas a very small ‘’ requires a high speed digital hardware
(hence high cost) to implement the controller.
Realization of Digital Controllers
(through discrete approximation of analog controllers)
Proportional (P) Controller
The analog controller output is
m = Kp e + b
The controller output at the nth instant is given by
mn  K p en  bn  mn  bn
where, Kp = proportional gain,
bn = fixed bias.
Realization of the P controller:
bn
m'n
en Kp + mn
Proportional-Integral (PI) Controller
 1 t 
m  K p e  o e.dt   b
 Ti 
 1 t 
= m + b, where Ti is the integral time and m  K p e  o e.dt 
 Ti 
I  o e.dt
t
Let the integration term be
n
The integration term at the nth instant is I n  o e.dt where  is the sampling interval
n  n
 o e.dt  n  e.dt for n = 1,2,3,..
n   n1
Now, o
e.dt  o e.dt  I n1
Then,
Pl Controller
The second term of the above relation represents the area

under the curve ‘e’ for (n – 1)  t n.
This area may be approximated by the shaded rectangle (called the method of
rectangular integration) as
e t n
e t n1
t
 n  1 n

Pl Controller
e tn
e tn1
t
 n  1 n

n
Thus,
  n 1 e.dt   e t n
Using the notation en for e t n , I n  I n1   en

Pl Controller
Now, output m at the nth instant may be expressed as
where I n  I n1   en
Similarly the output at the (n – 1)th instant may be expressed as
The difference between these two outputs is
 1 
mn  mn 1  K p en  en1   I n  I n1 
 Ti 
Pl Controller
 1 
mn  mn 1  K p en  en1   I n  I n1 
 Ti 
Substituting the value of (In – In-1 ) for rectangular integration, [(In – In-1 ) =  en ]
  en 
mn  mn 1  K p en  en1  
 Ti 
 
or, mn  en K p 1    K p en1  mn 1
 Ti 
 ao en  a1en1  mn 1 , say
 
where, ao  K p 1   and a1   K p
 Ti 
Realization of the PI Controller
Controller output at the nth instant:
mn = m’n + bn
m’n= a0en + a1en-1 + m’n-1
en-1
a1
bn (bias)
delay 
m'n
en ao + + mn
error Controller
sequence
 delay output
m'n-1 sequence
Problem: Develop a digital PI Controller using Trapezoidal rule for integration.

Velocity or incremental form of PI controller
The controller output is proportional to the derivative of a standard PI

controller and it may be expressed as (without bias):
mn  mn  mn1  mn  mn 1  aoen  a1en1
Velocity form of controller is useful when the actuator is some kind of

adder (integral action), like a stepping motor.
Software Realization of the PI Controller
e1 a1
b

mp m
e a0 + +

# include < studio.h>

void main (void)
{
float e = 0, e1, m, mp = 0;
float a0, a1, b;
float adc (void) ; // digitized error
void dac (float m ) ; // analog output
a0 = - - - - - - - - ; // Kp [1 + /Ti ]
a1 = - - - - -- - - ; // -Kp
b = - - - - - - - - - ; // bias
e1 a1
b

mp m
e a0 + +

for (;;) // continuous loop

{ // loop time is the sampling interval 
e1 = e;
e = adc ();
mp = mp + a0*e + a1*e1;
m = mp + b;
// provision for saturation
if (m < 0) m = 0;
if (m > 100) m = 100;
dac (m);
}
}
e1 a1
b

mp m
e a0 + +

float adc (void) // Analog-to-digital conversion

{
float v;
scanf (%f, &v); // for (keyboard) simulation
return v; // (to be replaced for actual
} // realization)
void dac (float m) // Digital-to-analog conversion

{
printf (%f\n, m); // for (VDU) simulation
} // (to be replaced for actual realization)
Proportional-Derivative (PD) Controller
 de 
m  K p e  Td   b where Td is the derivative time.
 dt 
de
Let the derivative of error ‘e’ at the nth instant be Dn 
dt t n
Dn may be approximated using the backward difference algorithm as
e slope = D n
en  en 1 en
Dn 

en-1 approximate slope

time
(n-1)  n
PD Controller
Thus the controller output at the nth instant is
  e  en 1  
mn  K p  en  Td  n    bn
   
 Td   K pTd 
or mn  K p 1   en     en 1  bn
    
 ao en  a1en 1  bn
 Td  K pTd
where ao  K p 1   and a1  
   
Realization of the PD Controller
en-1
a1
delay  bn (bias)
en ao + + mn
(error) Controller
output
Problem: Develop a ‘c’ program for software realization of the PD Controller.

PD Controller
To avoid derivative action from a sudden change in set-point, the derivative action
is generally derived from the measured output.
Now, e = r – c
de dr dc
then,   ,
dt dt dt
dc
 , assuming set-point ‘r’ is constant.
dt
Thus, the controller output may be expressed as,
 dc 
m  K p  e  Td b
 dt 
PD Controller
 dc 
m  K p  e  Td b
 dt 
using backward difference algorithm, the controller output at the nth instant is
  cn  cn 1  
mn  K p  en  Td     bn
   
K pTd K pTd
or mn  K p en  cn  cn 1  bn ao  K p
 
 ao en  po cn  p1cn 1  bn where K pTd
po  

K pTd
p1    po

Realization of PD Controller with anti-derivative kick
en
ao
error bn (bias)
cn po + + mn
(process (Controller output)
output)  delay
cn-1 p1
Problem: Develop a ‘c’ program for software realization of the PD Controller

with anti-derivative kick
Proportional-Integral-Derivative (PID) Controller
 1 t de 
m  K p e  o edt  Td  b
 Ti dt 
 m ' b (say)
 1 t de 
where m '  K p e  o edt  Td 
 Ti dt 
The controller output (without bias) at the nth instant, using backward difference algorithm, is
 In  en  en 1  
mn  K p en   Td  
 Ti   
PID Controller
The controller output (without bias) at the nth instant is
 In  en  en 1  
mn  K p en   Td  
 Ti   
The controller output at the (n-1)th instant is
 I n1  en1  en2  

mn 1  K p en1   Td  
 T i   
Subtracting,
 I n  I n1 Td 
mn  mn 1  K p en  en1    en  en2  2en1 
 Ti  
PID Controller
 I n  I n1 Td 
mn  mn 1  K p en  en1    en  en2  2en1 
 Ti  
Using rectangular integration algorithm, I n  I n1   en
  Td 
then, mn  mn 1  K p en  en1  en   en  2en1  en2 
 Ti  
  Td   2Td   K pTd 
or, mn  en 1    K p  en1   1 K p  en2    mn 1
 Ti        
or, mn  ao en  a1en1  a2en2  mn 1

PID Controller
mn  ao en  a1en1  a2en2  mn 1
  Td 
where, ao  K p 1   
 Ti  
 2Td 
a1   K p   1
  
K pTd
and a2 

Realization of the PID Controller
mn  ao en  a1en1  a2en2  mn 1

and mn = m’n + bn
en-2 a2

en-1 a1 bn (bias)

m'n
en ao + + mn

m'n-1
PID Controller
Problems:
1. Develop a digital PID controller using trapezoidal rule for integration
2. Develop a program in ‘C’ for software realization of the PID controller
3. Modify the above controller to provide anti-derivative kick feature

Techniques used for anti-integral windup
 By saturating or limiting the integral value
 By resetting the integral value to zero
 By omitting the integral term
 By adaptive adjustment of controller parameters

Some anti-integral windup schemes
Stop integration when PI/PID controller internal output (prior to the

saturation block) exceeds the saturation limits
Saturation
r e 100 n C
+
+ Kp + 0 100
Process
_
Multiplier
_
Kp +
X sTi
+
hard
switching Scheme
1
1/0
Stop integration if 0
saturated
Scheme for PI Controller

Performance of anti-integral windup scheme
Saturation
r e 100 n C
+
+ Kp + 0 100
Process
_
Multiplier
_
Kp +
X sTi
+
hard
switching Scheme
1
1/0
Stop integration if 0
saturated
Scheme for PI Controller
PI control without anti-integral windup PI control with anti-integral windup

Reduce integration gradually as PI/PID controller internal output

exceeds the saturation limits
Saturation
r e 100 C
+
+ Kp + 0 100
Process
_
Kp _
1 +
+ +
Ti s soft
switching Scheme
G : a constant
Clegg integrator – the integrator is set to zero (reset) when the error
crosses zero
reset control
reset
SW
R
error _ integrator output
+
Automatic/Manual modes of Operations
 Automatic mode – means automatic closed loop operation
 Manual mode – means open loop manual control
Manual input or commnd
Man
e m Final v
Controller Process
error Auto Control element
Man
e m Final v
Controller Process
If there is any difference between the controller output and the manual command, a
bump occurs in the process output when the switch position is altered.
To provide ‘bump-less transfer’ from auto-to-manual change over, special

arrangements may be made for ‘set-point initialization’.
The manual command is driven to equal the controller output when the loop is in
AUTO mode.
Man
e m Final v
Controller Process
When the loop is in MANual mode, if there is a steady error existing due to any
difference between the set-point of the controller and the process output (under
manual control), integral term, in case of PI and PID controllers, may wind-up to a
large value, and consequently anti-integral wind-up is necessary for such
situations.
To provide bump-less transfer for all the operating modes, incremental or velocity
from of controller is used with an additional integrator.
Scheme for bump-less transfer
Incremental manual command
Man bn (bias)
Digital m'n m'n mn

en incremental Integrator +
controller Auto
The incremental controller output (without bias) at the nth instant may be expressed as
mn  mn  mn 1
  Td   2Td  K pTd
 en K p 1     en 1K p   1  en  2 for a PID controller
 Ti      
 ao en  a1en1  a2en2
Man bn (bias)

controller Auto
The integrator may be represented as
m'n Integrator m'n
The integrator output may be represented as m'n

m'n +
mn  mn  mn 1  mn 1

 mn  mn 1 m'n-1
Man bn (bias)

controller Auto
The presence of integrator at the output ensures a smooth output variation even
when the actual manual command is different from the actual controller output
under closed-loop control.
Realization of the incremental type PID Controller
en-2 a2

Incremental manual
Command
en-1 a1
bn
MAN

m'n m'n
mn
en ao + + +
AUTO
Incremental or
Velocity 
Controller m'n-1
Integrator
Automatic tuning of PID Controllers – the Relay autotuner
Suitable for processes with non-zero dead-time.
This is based on a special technique for determining the critical gain K c and critical
time period Tc of the process loop.
Kc is the gain margin of the process loop.
Controller parameters Kp, Ti and Td are calculated according to Ziegler – Nichols

(Z-N) rule for a stable time response.
Z-N settings
Controller Kp Ti Td
gain margin: 2
P – Controller 0.5 K c
PI – Controller 0.45 K c Tc/1.2
PID – Controller 0.6 K c Tc/2 Tc/8

The critical gain Kc and critical time period Tc are determined from an experiment
with relay (switching element) feedback.
D
set point + e v
r=0 + Process c
_ 0
-D
Relay
The relay control provides ON / OFF control of the process.
The input ‘r’ is set to zero.
The output ‘c’ oscillates around a mean value of zero (limit-cycle oscillations).
The process is driven by a square wave of amplitude ‘D’.

D
set point + e v
r=0 + Process c
_ 0
-D
Relay
Assuming the process to be a low-pass system, the process output ‘c’ contains
mainly the fundamental component.
Thus the error signal ‘e’ becomes sinusoidal,
e = A sin t
D
set point + e v
r=0 + Process c
_ 0
-D
e = A sin t
Relay
The relay output ‘v’ may be found out as follows:

v v
D
D
e time
0  0  2
  
-D
-D
time
2



e
0 A
-A


v
D D
set point + e v
r=0 + Process c
_ 0
-D time
  2
  
Relay
-D
The Fourier series of the relay output (v) may be expressed as:
4D  1 1 
v      
  
sin t sin 3 t sin 5 t .......
3 5 
The process practically attenuates all higher harmonics other than the fundamental.
Then the process gain at frequency ‘’ becomes
output amplitude A A
G     
input amplitude 4 D 4D

D
set point + e v
c
Process gain at frequency ‘’:
r=0 + Process
0
A
_
output amplitude A
G   
-D
 
Relay input amplitude 4D 4D

Now, to maintain steady oscillations at  = c, the loop gain is 1   (considering

negative feedback).
Thus the gain of the relay controller (i.e. the critical gain) at  = c is
1 4D
Kc    as K c . G c   1
G c  A
Also G (c) = , as relay phase shift is zero.

.
D
set point + e v
r=0 + Process c
_ 0
-D
Relay
Thus by knowing the relay amplitude ‘D’ and by measuring the amplitude ‘A’ of the
 4D 
process output ‘c’, critical gain Kc may be determined K 
 c  A  .
 2 
Tc may be estimated by measuring the frequency of the output oscillation Tc 
c 
.

Block diagram of the Relay autotuner
(The Satt Control Autotuner by Satt Control, Sweden)
Manual
command
M
r + e PID A v
+ Controller
Process c
_
T
Mode
D switch
0
D
Relay
M  Manual position
A  Auto position
T  Tune position
At first, the process is brought to equilibrium state ( zero error), by setting a

constant control signal in manual mode.
The tuning is then activated by pushing the mode switch to tune position.
Modified relay characteristic for a non-zero set-point
v v
Controller
D = 0.1v m ax output for
zero error
e t
0 0
Manual
setting
The relay amplitude ‘D’ is initially set to 10% of the controller output-range.
This amplitude is adjusted after one and a half period to give oscillation of 2% of the mean output.
This ensures minimum disturbance at the process output due to tuning.
This adjustment is done by measuring the change in output during the first one and a half period.
The modified relay amplitude is stored for the next tuning operation.
Satt Control Autotuner
Manual
command
M
r + e PID A v
+ Controller
Process c
_
T
Mode
D switch
0
D
Relay
M  Manual position
A  Auto position
T  Tune position
The system is automatically switched to Auto mode after estimating the critical gain Kc
and critical time period Tc during first 5½ period of oscillation.
The parameters of PID controller (viz. Kp, Ti and Td ) are determined from Kc and Tc
according to Z-N rule.
Final Control Element
m Final Control v
Element manipulated
controller
output variable
 (actuating signal)
m displacement v
Control
Actuator
or position 'x' valve
 Actuator provides an output position proportional to the

input signal
 Control Valve adjusts the value of the manipulated
variable
Final Control Element
m Final Control v
Element manipulated
controller
output variable
 (actuating signal)
m displacement v
Control
Actuator
or position 'x' valve
 The actuator accepts a signal from the control system and, in

response, moves the valve to a fully-open or fully-closed position, or
a more open or a more closed position (depending on whether
'on/off' or 'continuous' control action is used).
Actuators
spring –
Pneumatic diaphragm
actuators piston
motor
Electro-pneumatic
actuators
Actuators Motorized-
rotary or linear
Electric
actuators Solenoid
operated
Hydraulic
actuators
Pneumatic Actuators
Spring – Diaphragm type Actuators
input spring
pressure
m
(3-15 psi)
diaphragm
backing plate output position

x
stem constant
thrust force
 Diaphragm actuators have compressed air applied to a flexible

membrane called the diaphragm.
Pneumatic Actuators
Spring – Diaphragm type Actuators
 At equilibrium, (assuming no
input spring
pressure
m
change in thrust force on the
(3-15 psi)
stem): mA = Kx
diaphragm
backing plate output position m = the change in input pressure

stem
x
constant
thrust force
A = the effective area of the diaphragm,
K = the spring constant (including
diaphragm),
x = the change in output position (stem)
Note: The actual value of ‘x’ (stroke length) is limited within

¼ and 3 depending on the size of the actuator.
Spring – Diaphragm
type Actuators
Another Variation:
Direct-Acting and Reverse-Acting
Pneumatic Spring-Diaphragm Actuators
Direct-Acting and Reverse-Acting
Pneumatic Spring-Diaphragm Actuators
Direct-acting Actuator: Reverse-acting Actuator:

(spring-to-retract) (spring-to-extend)
This actuator is designed with The diaphragm is pushed

the spring below the diaphragm, upwards, pulling the spindle up,
having air supplied to the space and if the spindle is connected to a
above the diaphragm. The direct acting valve, the plug is
result, with increasing air opened. With a specific change of
pressure, is spindle movement in air pressure, the spindle will move
the opposite direction to the sufficiently to move the valve
reverse acting actuator. through its complete stroke.
Pneumatic Actuators
Positioners
Why Positioners are provided with Actuators ?
 To overcome high static friction forces in the actuators.

 To improve response time.
 To improve linearity and to reduce hysteresis.
 To reduce loading on controller output.
In case of using actuators, we have non-linearities due to diaphragm

area and the spring constant. So the change in position due to change
in controller output is non-linear. With the use of positioners we can
decrease the effect of non-linearity.
Pneumatic Actuators
Positioners
Features
 A positioner ensures that there is a linear relationship between the signal

input pressure from the control system and the position of the control valve.
 A positioner may be used as a signal amplifier or booster.
 Some positioners incorporate an electropneumatic converter so that an
electrical input (typically 4 - 20 mA) can be used to control a pneumatic valve.
 Some positioners can also act as basic controllers, accepting input from
sensors.
Pneumatic Actuators
Spring – Diaphragm Actuator with Positioner
n
gain = 1
n
Relay
restriction
spring
Air supply
Bellow
Bellows
(Ps) Kbm
c Feedback
lever
m n 0
input pressure c y a
b
b x
n  nozzle back pressure x
a
stem
K m b x
 y  b  
2 a2
Pneumatic Actuators
n
 At equilibrium, Baffle-Nozzle
gain = 1
Relay
n
separation is:
restriction
bx K m
y     b
spring
Air supply
a2
Bellow
(Ps) Bellows Kbm
Feedback
2
c
lever
m n 0
input pressure c y
b
a
Kb is the bellows stiffness factor
b x
n  nozzle back pressure
Nozzle back pressure: n = – Kn. y
x
a
stem
Kbm  b  x
y  
2 a2
Kn is the nozzle gain
The change in output pressure is related to the back pressure as:

Kx = nA, K is the spring constant and A is the effective area of diaphragm.
Kx n  nA  Kx Kb m  b  K 
n y          x    
A K n  n 
K A K n A 2   2 a  K n 
A
b
Now, K n A  K ,
K
 0, and y  0. Kb m  x  
Kn A a
Pneumatic Actuators
n
gain = 1
n
Relay
restriction
spring
Air supply
Bellow
(Ps) Bellows Kbm
c Feedback
lever
m n 0
input pressure c y a
b
b x
n  nozzle back pressure x
a
stem
Kbm  b  x
y  
2 a2
Conclusion:
 Thus, change in output position is related to change in input pressure with
only feedback lever ratio and the bellows stiffness factor, and it is not dependent
(if KnA >> K) on the spring-diaphragm non-linearities.
 As a >> b, large position change is obtained with a small change in bellows

position, thus ensuring linear characteristic of the bellows.
Pneumatic Actuators
Piston type Pneumatic Actuator
m
piston
spring
 They are generally used where the stroke of a diaphragm actuator would
be too short or the thrust is too small.
 They are used for long strokes.
Pneumatic Actuators
Motor type Pneumatic Actuator
m
+
Air
Motor
spring
 They are used for large thrust forces. Large torques are generated
from motor-gear arrangements to balance large thrusts.
Electro-pneumatic Actuators
 When the controller output is electrical and a suitable air supply is available,
using an electro-pneumatic actuator, a large output power may be obtained from
a low power control signal.
cascading an electro-pneumatic
converter and a pneumatic actuator
Realization
an electro-pneumatic actuator
Electro-pneumatic Converter
Electro- pneumatic
'i' electrical output 'm'
pneumatic
signal (3-15 psi)
input Converter
(current
4-20mA)
input
current
i former (support system)
pneumatic
output (m)
S
Nozzle
air supply
Voice m
(Ps)
Coil N Restriction
Motor
a
S
voice coil motor

balance beam - linear motor
m
Feedback bellows
input
current
i
voice used in
former (support system)
pneumatic coil loud speakers
output (m)
S
Nozzle
air supply
Voice m
(Ps)
Coil N
Motor
Restriction
S
S
a
+ Force
voice coil motor
balance beam - linear motor
m
b
force
N
Feedback bellows = BlNi
B flux density
Effective force l mean length / turn
experienced by the N no. of turns
S
former is a linear one.
force
input
current
i former (support system)
At equilibrium, for a change in input current
‘i’, the change in output pressure ‘m’ is:
pneumatic
output (m)
S
Nozzle
a.B ln i  bmAb 
air supply
Voice m
(Ps)
Coil N Restriction
Motor
a
S
voice coil motor

balance beam m
- linear motor Ab = the active area of the bellows assuming
b
a small Baffle-Nozzle separation.
Feedback bellows
 aBlN 
m i
 bAb 
Electro-pneumatic Actuator
gain = 1
Relay diaphragm
Air supply
(Ps)
Restriction m
input
i
current
spring
Voice S
coil
motor Nozzle
N
a
S
0
x output
b Feedback position
c lever
balance Feedback
beam d spring
(Ks)
Electro-pneumatic Actuator
Relay
gain = 1
diaphragm
At equilibrium,
Air supply
 b  
(Ps)
cB ln i   d .  x.K s 

Restriction m
input
i
 a  
current
spring
Voice S
coil
motor
N
Nozzle
Ks = spring constant of the
a
feedback spring (for a small
S
0
c b Feedback
x output
position
Baffle-Nozzle separation)
lever
balance Feedback
beam d spring
(Ks)
Conclusion: output ‘x’ is independent of the

characteristics of diaphragm and spring.
Electric Actuators
Motorized Rotary Actuator
input Error Position

amplifier sensor
output
position
Low inertia
servo motor low inertia servo
gear train motor - fast response
to increase torque
Electric Actuators
Motorized Linear Actuator
Rack and pinion

output position
+
Servo Motor
Electric Actuators
Solenoid Actuator (for On/Off Operation)
Armature
AC or DC
Coil
Spring Output position
 A spring return type electric solenoid is used to actuate an iron cored armature.
 A shading band type armature is used with AC supply to create unidirectional

pull.
Hydraulic Actuators
They use incompressible fluid (oil). These are

used for high power and high speed applications.
Control Valves
 A control valve in a pipeline acts as a variable restriction. An
actuator controls the ‘lift’ of a control valve to alter restriction.
 All control valves have an inherent flow characteristic that

defines the relationship between valve opening and flowrate
under constant pressure conditions.
 The three main types of control valves available are:
 quick/fast opening
 linear
 equal percentage
Control Valves
 The physical shape of the plug and seat arrangement, sometimes referred to
as the valve 'trim', causes the difference in valve opening between these valves.
Typical trim shapes for spindle operated globe valves

Flow-Lift Characteristic of Control Valves
% flow   v 100 
 vmax 
Manipulated
Position x variable v
(or lift) (or flow)
Control Valve
From laws of fluid dynamics,

v = Kh x
K  overall coefficient
h  difference in head across
valve
% lift   x 100 
 xmax 
Quick/Fast Opening Control Valve
 Here, the valve sensitivity dv dx at any flow decreases with increasing
flow. The fast opening characteristic valve plug will give a large change in flow
rate for a small valve lift from the closed position.
 Fast opening valves tend to be electrically or pneumatically actuated and

used for 'on / off' control.
Linear Control Valve
 The linear characteristic valve plug is shaped so that the flow rate is directly
proportional to the valve lift, at a constant differential pressure. Here the valve
sensitivity is (approximately) constant.
Flow-lift characteristic
for a linear valve
Equal Percentage Valve
 These valves have a valve plug shaped so that each increment in valve lift
increases the flow rate by a certain percentage of the previous flow. The
relationship between valve lift and orifice size (and therefore flow rate) is not
linear but logarithmic.
 For these valves, sensitivity increases with flow. As the valve sensitivity at
any given flow rate is a constant percentage of the given flow rate, the term
equal percentage is used. dv
dv
Valve sensitivity,  Kv where K is a constant. Or, dx  K
dx v
i.e. sensitivity expressed as percentage of flow (= 100K %) is constant.
 dv dx 
v
0 vm in vm ax
working range
Rangeability of a Control Valve
maximum controllable flow  v max 

R=
minimum controllable flow  v min 
 The minimum controllable flow of a control valve depends on

its construction. The range of R is usually between 20 and 50
under constant pressure drop across the valve (equal to constant
head). It is typically 50 for a globe type control valve.
Relation between % flow, % lift and rangeability under
constant pressure drop (or head) across the valve
Linear Valve
 vmax  vmin 
v v  vmin  x  
 xmax 
vm ax
v vmin  vmax  vmin  x
vm in   
x vmax vmax  vmax  xmax
0 xm ax
v 1  1 x 1 x 
  1    1   R  1 
vmax R  R  xmax R  xmax 
Valve
v 1 x  dv  R  1  vmax
 1   R  1    Sensitivity
vmax R  xmax  dx  R  xmax constant
 x   x 
v  1
Differentiating,  1   dv   1
1
R  xmax 
    R
 xmax 
.ln R.
vmax  vmax   dx  xmax
v 1
dv  .ln R.
vmax xmax
dx   ln R  ,
Equal Percentage
  characteristic
v  xmax 
dv  ln R 
Valve Sensitivity is proportional to ‘v’   v,
dx  xmax 
V = volumetric flow through the valve at lift H,

x
e  x  ln R H ,

V  Vmax R = valve rangeability,
R H = valve lift (0 = closed, 1 = fully open),
Vmax = maximum volumetric flow through the valve.
x
According to our previous notation: V  v and H
x max
Therefore, ex
V  Vmax can be written as,
R
 ln R  x x   ln R  x x   ln R  x x 

e  e   v  e max

v ln    ln 
max max
v  v max
or   or 
R v max R  max 
v

R

     
   
 v     v   x 
or ln    ln R 
x
  ln R or ln    ln R   1
 vmax   xmax   vmax   max
x 
 x   x 
 v   1   1 
  ln R   R 
 x max 
v  x max 
or ln  or
 vmax  v max
Classification of Control Valves
Control Valves
Sliding Stem Valves Rotating Shaft Valves
Single -Seat Double-Seat Lifting Gate

Plug Valves Plug Valves Valves
Single-seat Sliding-stem Control Valves
Stem open
x
Packing
gland position
close
v
(fluid flow)
plug
 Here, depending on the plug shape, flow-lift characteristic will

vary.
Stem open
x
Packing
gland position
close
Plug Types
v
(fluid flow)
plug
V-port equal percentage characteristic
plug Parabolic linear characteristic
Poppet quick opening characteristic

Single-seat
Sliding-stem
Control Valves
Another
variation
 A  P   Friction allowance  F
A= Valve seating area (m2)

· P = Differential pressure
(kPa)
F = Closing force required (kN)
Stem open
x
Features
Packing
gland position
close
v
(fluid flow)
plug
 Can be shut off completely to provide zero flow – an advantage.

 Requires large thrust force, thus a powerful actuator is necessary
– a disadvantage.
Solution ?
 To overcome the disadvantage, double-seat arrangements are
used, which require a small thrust force to operate.
Double-seat Sliding-stem Control Valves
x
position
v
v
Double-seat
Sliding-stem
Control Valves
Another
variation
Double-seat Sliding-stem Control Valves
x
position
Disadvantages
v
v
 It
cannot be shut off completely because of the differential
temperature expansion of plug and body (when hot fluid is flowing).
 If plug expandsmore than body, it may cause breakage. If body
expand more, there will be significant leakage or offset in the
system.
Double-seat valves are used where it becomes impractical to provide

sufficient force to close a conventional single seat valve.
Lifting-Gate Control Valves
'x' position open
close
gate
v v
Single-seat Lifting-Gate Valve

 These are used to control flow of fluids containing solid matters
(e.g. paper pulp).
 Here no change in direction of fluid flow takes place, due to the
control action.
Rotating-Shaft Control Valves
Rotating - shaft control valves
Rotary plug Butterfly Louver
used to control used to control use to control

liquid flow air,gas and liquid air flow at
(e.g. oil to burner) at low pressure low pressure
difference (draft control)
 Rotary type valves, often called quarter-turn valves, include plug

valves, ball valves, butterfly valves etc. All require a rotary motion to
open and close, and can easily be fitted with actuators.
Rotary plug type Control Valves
rotation
x
stem
(circular opening)
v v plug
plug (shown open)
 When the pipe opening and plug opening completely align,

complete flow takes place. If they do not align at all, no flow takes
place.
Rotary plug type Control Valves
rotation % flow
x
stem 100
(circular opening)
rectangular opening
v v plug
0  x 
% rotation   100 
plug (shown open) 0 100  xmax 
Circular
for different
plug opening V-shaped flow-rotation
(or part) characteristics
Rectangular
 Here 100% rotation corresponds to an angular rotation or

movement of 90o for x.
Ball Valves
Features
 It consists of a spherical ball

located between two sealing rings
in a simple body form.
 Rotating the ball through 90°

opens and closes the flow
passage.
 Ball valves are an economic

means of providing control with
tight shut-off for many fluids.
Butterfly Valves
vane
circular vane
rotation
x x
v v
Circular
Vane
Rectangular
% flow
100  Here, 100%

rotation corresponds
Circular vane
to an angular
rotation of 90o .
% rotation
0 100
Butterfly
Valves
Another
variation
Louver Valves
rectangular vane
linkage
v v x
x
rectangular duct
 Louvers cannot provide tight shut off due to long length of seating surfaces.
Hence considerable leakage takes place.
Louver Valves
rectangular vane
linkage
v v x
x
rectangular duct
% flow
100
% rotation
0 100
Three-port Valves
 Three-port valves can be used

for either mixing or diverting
service depending upon the plug
and seat arrangement inside the
valve.
Methods of Fluid Control
Methods of fluid control
Series Bypass Variable delivery

throttling (variable source)
requires no separate
control valve
Series Throttling
Head
inflow tank
Command
head
outflow
to process
control
valve
 For fluctuating or intermittent inflow, closed-loop control is necessary to

maintain constant outflow. The closed loop control will generate necessary
command for the control valve to maintain constant outflow .
Bypass
Head
tank
head
Process
restriction
by pass
Control valve
command
 Bypass is required when we cannot shut down the source, then the extra water is
bypassed. the control valve employed will have inverse gain.
 By pass is not economical, as a considerable portion of fluid is wasted.
Variable Delivery
Speed command
Process
Storage
tank
Variable speed
pump
Variable Delivery from a Variable Speed Pump

Electrical Signal Transmission Systems
for Process Control Applications
© Anjan Rakshit and Amitava Chatterjee

Single-Loop Controllers
Analog controller
Digital controller
4 – 20 mA 4 – 20 mA
Signal Controller
Sensor ADC DAC or
conditioner or 1 – 5 V software
1- 5 V
Measuring element
Final Control
Process
element
A single-loop controller
 Two popular modes of electrical signal transmission are:
 Voltage transmission used for short distance

 Current transmission used for long distance
Multi-Loop Controllers
 When the computer shares different control loops (for
economic reasons), multiplexers are used for data collections.
Digital Controller
Voltage 1
Virtual controller # 1 1
from sensor # 1 DAC
or current 2
from sensor # 2 Virtual controller # 2 2 n-channel
3
DAC
ADC controller output
(voltage or current)
n
from sensor # n Virtual controller # n DAC
n
MUX
(Multiplexer) Software
Problems:
 Sensor errors (e.g., offset, gain, non-linearity etc.) should be
taken into account by the software.
 High cabling cost due to star connection/ junction at the
controller.
Alternate Solution
 Utilize smart sensors, by incorporating a processor with requisite
interface along with a traditional sensor.
Smart Sensor
Digital Controller
Software
Voltage or Current
Sensor ADC Processor DAC ADC DAC
From
Transmission To final
process
control
element
Problems:
 Interface inaccuracies (e.g. ADC and DAC errors).
An Improved version of the
Previous Solution
 Interface inaccuracies can be minimized with a serial digital
transmission between a smart sensor and digital controller.
Serial Serial data

Sensor ADC Processor
Interface Transmission
Smart actuator with

serial interface Digital Controller
Final control Serial Serial

Software
element Serial Interface Interface
link
Adaptation for Multi-Loop Control
Digital
Actuator Sensor Sensor
Controller
#1 #1 #n
Serial bus
Sensor Actuator Actuator
#2 #2 #n
Features:
 A multiplexed serial data transmission system (bus) is used to
minimize the cabling cost.
 Proper communication protocol is necessary for selective
transmission of data from sensor-to-controller and controller-to-
final control element .
Serial Interface Standards
Standard Origin Type Max. Max. Number of
Length Data Drivers
Rate and
Receivers
RS – 232C EIA Single ended 50 ft 20 kb/s 1 – driver
(unbalanced)
(V24) (CCITT) (at 20 kb/s) 1 – receiver
RS – 423 EIA Single ended 4000 ft 100 kb/s 1 – driver

(unbalanced) (< 1 kb/s) (upto 30 10 –
ft) receivers
RS – 422 EIA Differential 4000 ft 10 Mb/s 1 – driver
(balanced) (< 100 (upto 40 10 –
kb/s) ft) receivers
RS – 485 EIA Differential 4000 ft 10 Mb/s 32 – drivers
(balanced) (< 100 (upto 40 32 –
kb/s) ft) receivers
EIA : Electronics Industries Association (USA)
CCITT: International Telephone and Telegraph
Consultative Committee
RS-232C
Driver Receiver
(or transmitter)
data data
in out
RS - 232C
interface
Connector
RS-423
Driver Receiver (differential)

data #1
+ data
in
RS - 423 out
_
Vg
Receiver #2
+ data
_ out
RS-422
Driver (differential) Receiver (differential)

#1
data + data
RS 422 out
in
_
Optional
terminating resistors (100 
Receiver #2
+ data
out
_
RS-485
R D
T
Rt Rt
120 120
T R
D
R
Rt : terminating resistor; R: receiver; D: driver; T: trans-receiver

(driver + receiver)
 In RS – 485, drivers can withstand bus faults and bus contention
Serial Data Transmission in presence of Noise
 In differential or balanced data transmission systems, common mode
noise voltages (with respect to ground) are nullified.
 For both balanced and unbalanced transmission systems, receivers are
provided with adequate hysteresis to reject interference.
Provision of Hysterisis with Asymmetry
output
A failsafe feature
he re o utpu t w ill be in
high state under open
input condition
input
O
A System Without Hysterisis
output
output
signal
Receiver characteristic
without hyetersis
O
input O time
input signal (noisy) false data
time
A System With Hysterisis
output
Receiver characteristic output

with hystersis signal
band
O input O
time
input signal
(noisy)
hysteresis band > peak to peak noise

time
Electrical Specifications of Serial Interface Standards
RS – 232C RS – 423 RS – 422 RS – 485
Driver output maximum voltage  25V  6V - 0.25V to 7V to

+ 6V + 12V
Driver output loaded  5V  3.6V  2V  1.5V

 15V  6V  5V  5V
Signal level unloaded
Driver load impedance 3K to 7K 450 min 100  54 
Slew rate 30V/S*
Receiver input voltage range  15V  12V  7V  7V to

+ 12V
Receiver input sensitivity (min  3V  200 mV  200 mV  200 mV

voltage)
Receiver input resistance 3K to 7K 4K min 4K min 12K min
*As the line is open ended, effects of line reflections are minimized with limited slew rate
Asynchronous Serial Data Communication
transmits data in one

simplex direction
transmits data in both

serial
half-duplex directions, but not
communication simultaneously
transmits data in both

full-duplex directions simultaneously
 In asynchronous transmission, inter character spacing

is not fixed. Each data character starts with a start bit
with one or two stop bits at the end. Each character is
identified by its start and stop bits – thus it can be sent at
any time (i.e. in an asynchronous manner).
Data Format for Asynchronous Transmission
Character length (10/11 - bits)

one
O 1 1
zero
Do D1 D2 D3 D4 D5 D6 parity
or D7
start
LSB 1/2 stop
bits
one bit bit
time 7 or 8 bit data
Parity Bit - set as odd or even – to identify one-bit error per

character during transmission.
An Example:
character length (11 bits)
O 1 O 1 O O O 1 O 1 1
start Do D1 D2 D3 D4 D5 D6 parity
bit 2 stop
bits
number of ‘1’s 3 odd
7 bit ASCII character ‘E’ with odd parity and two stop bits
‘E’: 69 d = 45 h
overall parity of 7-bit character ‘E’

= total number of ones including parity bit = 3  odd.
Another Example:
character length (11 bits)
O 1 O 1 O O O 1 1 1 1
start Do D1 D2 D3 D4 D5 D6 parity
bit 2 stop
bits
number of ‘1’s 4 even
7 bit ASCII character ‘E’ with even parity and two stop bits
‘E’: 69 d = 45 h
overall parity of 7-bit character ‘E’

= total number of ones including parity bit = 4  even.
RS-232C based Data Transmission Systems
Data Transmission Using MODEMs
MODEM stands for (Modulator + Demodulator)
Modulation Techniques:
 AM (Amplitude Modulation)
 FSK (Frequency Shift Keying)
 PSK (Phase Shift Keying)
 ….
Computer terminal Modem Modem Computer terminal
TxD TxD
RxD RxD
RTS RTS
Telephone
CTS line CTS
(Voice band
DTR carrier DTR
300Hz - 3kHz)
DSR DSR
GND GND
DTE DCE DCE DTE
RS 232C link RS 232C link

RS 232C – Pin Connections (DTE) (Male)
Computer terminal Modem Modem Computer terminal
Name DB 25 pin no. DB9 pin no.
TxD TxD
RxD RxD
RTS RTS
TxD 2 3
Telephone
CTS line CTS RxD 3 2
(Voice band
DTR carrier DTR GND 4 7
300Hz - 3kHz)
DSR DSR RTS 5 8
GND GND CTS 6 6
DTE DSR 20 4
DTE DCE DCE
7 5
RS 232C link RS 232C link DTR
DTE connector : 25 pin Male
DCE connector : 25 pin Female
DTE: Data Terminal Equipment; DCE: Data Communication Equipment
TxD: Transmitter data [ zero : + 12V, one : – 12V (typical value)]  for all signal lines
RxD: Receiver data [--do--] RTS :Request to send [ --do--]
CTS :Clear-to-send [ --do--] DTR :Data terminal ready [--do--]
DSR :Data-set-ready [ --do--]
Data Transmission Without MODEMs (Null Modem)
Computer terminal Computer terminal
TxD TxD
RxD RxD
RTS RTS
CTS CTS
DTR DTR
DSR DSR
GND GND
DTE DTE
Null modem
RS232C link
Data Transmission Without MODEMs and
Without Handshaking
Computer terminal Computer terminal
TxD TxD
RxD RxD
RTS RTS
CTS CTS
DTR DTR
DSR DSR
GND GND
DTE DTE
Null modem
link without handshake lines
 If hardware handshaking is not available, software based handshaking is normally

used through a suitable half-duplex communication protocol.
Transmission Rate
It indicates the rate at which serial data is
Baud Rate being transferred. It is defined as the
number of unit time intervals per second in
the transmission.
An Example:
If a maximum of ten characters with a data format using 7 bits per
character, parity bit and 2 stop bits (i.e. 11 bit-times per character,
considering the start bit) are transmitted each second,
then baud rate:
11 bit-times per character  10 characters per second
= 110 baud
i.e. baud rate = 1/ bit-time
Note: Baud rate and Bit rate (i.e. frequency) are not same.
Transmission Rate
An Example:
Consider the following character with even parity and one stop bit
Character length (= 10 bits)
O 1 O 1 O 1 O 1 O 1
Start Do D1 D2 D3 D4 D5 D6 even stop
bit parity bit
Bit-time
Time
period
Baud Rate = 1/bit-time Here, Time Period = 2  bit-time,

Bit Rate (frequency) Hence, Bit Rate = ½  Baud Rate
= (1/Time Period)
Transmission Rate
Standard Baud Rates

 110 Baud
 300 Baud
 600 Baud
 1200 Baud
 2400 Baud
 4800 Baud
 9600 Baud
 19.2 K baud
 ……
Microprocessor Interface
Serial Interface Driver
TxD TxD
RTS RTS
DTR DTR RS232C
P Interface
Data, RxD RxD
Address,
and Control CTS CTS
DSR DSR
Receiver

Ac PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Ac PDF

Hochgeladen von

Copyright:

Verfügbare Formate

Realization of Electronic Proportional (P) controller

The controller output may be expressed as:

E(s) Kp / sTi + M(s)

Therefore, transfer function of PID controller becomes

Proportional and integral terms may be combined as:

This expression of A presents the problem of interaction and hence,

The T.F. becomes:

➢ Provision for providing a bias term

➢ Provision for anti derivative kick

➢ Provision for anti-integral wind-up

➢ Output must be limited (saturation)

➢ Provision for Auto/Manual modes of control

➢ Provision for Local/Remote modes of operation

Non-zero steady state error

P-control without bias

P-control without bias P-control with bias

Derivative action may produce an unwanted kick at the controller output

PD-control with derivative kick PD-control with anti-derivative kick

Integral action in PI and PID controllers may produce a large integral

In some controllers, when the integral term is present, the integral

It is desirable to limit the controller output (say between 0% and 100%) so

A change-over switch is normally provided for configuring the controller

A change-over switch is normally provided for configuring the controller

➢ No transmission lag (as compared to pneumatic systems)

➢ Absence of dead zone due to friction or hysteresis in moving parts

➢ Capable of operating at low temperatures

➢ Compatibility with other electrical components

➢ Prone to electrical hazards (in inflammable atmosphere)

➢ Analog integrators are not very reliable

➢ Conversion equipments are necessary to interface pneumatic and hydraulic devices

+ a nozzle Restriction (orifice)

M (Nozzle back pressure)

x: change in baffle-nozzle separation.

M (Nozzle back pressure)

Typical 15 psi approx. linear working

+ a nozzle Restriction (orifice)

m = − K n  x , where –Kn is the nozzle gain

A relay valve may be direct-acting (positive gain) or reverse-acting (negative gain).

Relay valve (gain = 1)

Bleed type relay

The Baffle-Nozzle separation may be expressed as:

Block diagram of the controller

Simplified block diagram of the controller

Air supply (Ps)

Nozzle Restriction (orifice)

The Baffle-Nozzle separation may be expressed as:

The Baffle-Nozzle separation may be expressed as: (assuming RI >> R and

where Ti = Integral time

The Baffle-Nozzle separation may be expressed as: (assuming RI>>RD>> R)

The controller gain becomes infinite when Ti = TD, if derivative time

All the desirable features of a process controller (e.g. anti-integral wind-up,

No drift problem is encountered in a digital system.

In a processor based digital controller, rapid switching from one algorithm

Sample & hold

  Sampling interval   (ADC conversion time + computation time)

The error computation may be performed digitally if the set-point is available in

The analog controller output is

Realization of the P controller:

The analog controller output is

The second term of the above relation represents the area

Using the notation en for e t n , I n  I n1   en

Now, output m at the nth instant may be expressed as

Similarly the output at the (n – 1)th instant may be expressed as