Chapter 3: Standard Controllers: P, PI, PD, PID
Asso. Prof. Dr. H. Benyezza
Eng. 3 Ind. Elec.
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 1 / 28
Outline
1 Introduction
2 Elementary Corrective Actions
3 PID Controller Structures
4 Implementations
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 2 / 28
Introduction
In industrial control systems, maintaining stability, accuracy, and fast response is essential
for ensuring optimal performance of processes.
Standard controllers—namely Proportional (P), Integral (I), Derivative (D), and their
combinations (PI, PD, PID)—are the most widely used due to their simplicity, robustness,
and effectiveness.
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 3 / 28
Elementary Corrective Actions
Three basic corrective actions are used in control systems:
Proportional (P)
Integral (I)
Derivative (D)
Each action addresses specific performance criteria, but may degrade others.
They are suitable for simple systems. For higher performance requirements, these actions are
combined (e.g., PI, PD, PID controllers).
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 4 / 28
U ( p)
C ( p) = = Kp.
ε( p)
Proportional Corrector (P) # VX 1z2i
7ϔόЛЁζ eYlY Ģ +Raa3,jCRN TaRURajCRNN3II3
MRnc qRNc qn IRac 03 Iȕājn03 03 I Uaā,CA
Ks · N ( p)
cCRN \n3 Iȕ3aa3na CN0C,C3II3 ε i 0ȕnN cwcjěL3 N3 URccā0Nj Uc 0ȕCNjā<ajCRN T ( p) = Vq3, α = 0W
pα D ( p)
3cj 03 I 8RaL3 -
EO
εi = q3, K = K p · Ks I3 <CN 03 I 7i#QY
1+K
Transfer Function:
0#
X( s) ( s) 10 Y( s) 40
+ Kp
− 1 + s + 0,2 · s2 30
c(t) = Kp ⇒ C (s) = Kp 20
10
y(t) 0 ⊗ ⊗
SY9
−10
Effects: −20 K=
−30
K=
↑ Kp : Reduces response time S
K=5 −40
10−1
◦
100 ω0dB101ω0dB 102
a
K=1 0
↑ Kp : Improves steady-state zY9
−30
−60
−90
accuracy −120
−150 Pm
−180
Pm
t
↑ Kp : Increases overshoot z
z S l −210
10−1 100 101 102
a
↑ Kp : Reduces phase margin 7ϔόЛЁζ eYkY Ģ BNn3N,3 0ȕnN ,Raa3,j3na UaRURajCRNN3I
HRac\n3 I3 <CN n<L3Nj3
Iȕ3aa3na CN0C,C3II3 ε i 0CLCNn3 d
ġ ġ I3 0C<aLL3 0ȕLUICjn03 3cj 0ā,Iā q3ac I3
ġ
I cRajC3 3cj 03 UInc 3N UInc Rc,CIINj3 d @nj VI U@c3 a3cj3 CN,@N<ā3WY
I3 0āUcc3L3Nj d n<L3Nj3.
ġ ġ I $N03 UccNj3 03 I 7i#Q n<L3Nj3. I3
ġ
I3 j3LUc 03 LRNjā3 0CLCNn3 d j3LUc 03 LRNjā3 3cj 0RN, aā0nCj 0ȕnjNj V+8Y
I Uc3n0RĢUāaCR03 Tp 0CLCNn3 Vω p n<A
ġ aUC0Cjā 03c cwcjěL3cWY
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter L3Nj3W
3 d 5 / 28
1
Ci ( p) = .
Ti · p
Integral Corrector (I) +3 jwU3 03 ,Raa3,j3na U3nj ćja3 aāICcā3 q3, nN3 $RNN3 UUaRuCLjCRN Ua nN LRNj<3 CNjā<aj3na ©
$c3 0ȕLUIC~,j3nac RUāajCRNN3IcY
VX 1z2ib /m +Q``2+i2m` AMiû;`H
Transfer Function:
Z t
0#
40
X( s) ( s) 1 10 Y( s) 30
1 1 +
− Ti · s 1 + s + 0,2 · s2 20
dt ⇒
10
c(t) = C (s) = 0 ⊗ ⊗ ⊗
Ti 0 Ti s y(t) Ti = 1,9 s
−10
−20
−30
−40
−50
No correctio
−60 a0gc
Advantages: 10−1
0
◦
100
ω0dB ω0dB ω0dB
101 102
Ti = 1
−30
Ti = 8
Ti =8s
Eliminates steady-state errors S
No correction−90
−60
−120
Enhances system accuracy −150
−180
Pm > 0 Pm
−210 Pm < 0
t
Good for step inputs z
z S
−240
−270
10−1 100 101 102
a0gc
Disadvantages: 7ϔόЛЁζ eY:Y Ģ 2{3jc 0ȕnN ,Raa3,j3na CNjā<aI
Phase lag: −90◦
H3 ,Raa3,j3na UURaj3 nN3 CNjā<ajCRN 0Nc I ,@ţN3 03 ,RLLN03Y +3jj3 CNjā<ajCRN 0Nc I 7i#Q
U3aL3j 0ȕNNnI3a Iȕ3aa3na cjjC\n3 URna nN3 3Njaā3 3N ā,@3IRN. IȕCNjāaćj UaCN,CUI 03 ,3 jwU3 03 ,Raa3,j3na
3cj 0RN, 0ȕLāICRa3a I Uaā,CcCRNY BI CNjaR0nCj LI@3na3nc3L3Nj nN 0āU@c<3 03 −90ï 3j aCc\n3 03 0RN,
a3N0a3 I3 cwcjěL3 CNcj$I3 Ua 0CLCNnjCRN 03 I La<3 03 U@c3Y QN a3jaRnq3 ,3c 3{3jc. cna Iȕ3u3LUI3 03
Reduces phase margin I ~<na3 eY:. I3 cwjěL3 NRN ,RaaC<ā. Nȕ3cj Uc Uaā,Cc. IȕCNjā<ajCRN a3N0 I3 cwcjěL3 Uaā,Cc LCc U3nj nccC I3
a3N0a3 CNcj$I3Y QN ,RNcjj3 nccC cna I aāURNc3 j3LURa3II3 \n3 IȕCNjā<aj3na 0CLCNn3 I3 j3LUc 03 aāURNc3
Slows response time 0n cwcjěL3 V0CLCNnjCRN 03 I $N03 UccNj3 03 I 7i#Q ⇒ 0CLCNnjCRN 03 I $N03 UccNj3 03 I 7i#7
⇒ n<L3NjjCRN 0n j3LUc 03 LRNjā3 ⇒ n<L3NjjCRN 0n j3LUc 03 aāURNc3WY
+3 ,Raa3,j3na 3cj aa3L3Nj njCICcā c3nIY
Rarely used alone
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 6 / 28
Derivative Corrector (D)
Transfer Function: Disadvantages:
Reduces accuracy
d Amplifies high-frequency noise
c(t) = Td ε(t) ⇒ C (s) = Td s
dt Rarely used alone
Advantages: Typically combined with P
Phase lead: +90◦
Improves stability
Increases phase margin
Faster response
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 7 / 28
Effect of PID Controller Parameters
Parameter Rise Time Overshoot Settling Time Steady-State Error
Kp ↑ Decrease Increase Small Change Decrease
Ki = T1i ↑ Decrease Increase Increase Eliminate
Kd = Td ↑ Small Change Decrease Decrease Small Change
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 8 / 28
Proportional-Integral (PI) Controller
Advantages: Disadvantages:
Combines fast response of P May introduce overshoot
Eliminates steady-state error (I) Can cause integral windup
Widely used in industry (accumulation during saturation)
Tuning requires compromise
Slower than P alone
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 9 / 28
Proportional-Derivative (PD) Controller
Advantages: Disadvantages:
Phase lead: improves stability Amplifies high-frequency noise
Increases phase margin Does not eliminate steady-state error
Faster transient response Less effective for slow processes
Reduces overshoot
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 10 / 28
Proportional-Integral-Derivative (PID) Controller
The features of each of the PI and PD controllers are utilized.
Improves both steady-state errors as well as transient response specifications.
The PID controller algorithm involves three separate constant parameters that need to be
tuned.
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 11 / 28
1
C ( p) = K p 1 + (1 + Td · p) ;
PID Controller Structures Ti · p
SA. KBti2 , V~<Y eYSe$W
1
C ( p) = K p 1 + + Td · p .
Ti · p
Three Common Structures:
1 Parallel PID: Kp
1 ( s) 1 + U( s) ( s) 1 + U( s)
+ Kp +
C (s) = Kp + + Td s Ti · s + Ti · s +
Ti s
Td · s Td · s
2 Serial PID: VW Parallel TB/ V$W Mixed TB/
1 ( s) 1 + U( s)
C (s) = Kp 1+ (1 + Td s) Kp
Ti · s
+ Td · s ++
Ti s V,W Serial TB/
7ϔόЛЁζ eYSeY Ģ bjan,jna3 03c ,Raa3,j3nac TB/
3 Mixed PID: `3La\n3 - RN ,RNcjj3 \n3 I3 03<aā 0n NnLāaj3na 0ȕnN ,Raa3,j3na TB/ 3cj cnUāaC3na n 0
NRLCNj3na. Ra ,RLL3 URna I3 ,Raa3,j3na UaRURajCRNN3I 0āaCqj3na VT/W C0āI. CI Nȕ3uCcj3 Uc 0
1 TB/ C0āIY mN ,Raa3,j3na TB/ UUaR,@ā aāICc$I3 3cj 03 I 8RaL3 -
C (s) = Kp 1 + + Td s 1 T ·p
Ti s C ( p) = K p + + d q3, Td τ.
Ti · p 1 + τ · p
KI<aā jRnj. URna Iȕājn03 \nC cnCj NRnc N3 ,RNcC0āaRNc \n3 I3c ,Raa3,j3nac C0ānuY
+3c jaRCc 8RaL3c N3 cRNj Uc jRjI3L3Nj ā\nCqI3Nj3c. 3N 3{3j cȕCI 3cj jRnERnac URccC$I3 03
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 LCuj3 © nN TB/ UaIIěI3
TB/ 12 / 28
PID Implementations
Electronic Implementation
Analog: op-amps (P, I, D circuits)
Digital (Discrete): microcontrollers, PLCs
High accuracy and fast response
Pneumatic Implementation
Based on air pressure signals (3–15 psi)
Used in industrial environments
Robust but slower response
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 13 / 28
Analog: op-amps (P, I, D circuits)
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 14 / 28
Pneumatic PID Controllers
Definition:
A pneumatic controller uses air pressure signals (typically 3–15 psi).
It compares:
Process Variable (PV)
Setpoint (SP)
Generates an output pressure to control a final element (valve, actuator).
Key principle: Mechanical force or motion balance.
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 15 / 28
Proportional Control Action: Force-Balance Principle
Pneumatic controllers use a force-balance
mechanism
Input pressures (PV and SP) generate forces
via bellows
Output pressure generates an opposing force
Objective:
Maintain equilibrium by balancing forces
Keep the beam in a stable position
Detection:
Beam movement detected by a nozzle–baffle
system
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 16 / 28
Direct Action Mechanism
Operation:
PV ↑ pushes the beam upward
Baffle moves closer to the nozzle
Nozzle backpressure ↑
Output pressure ↑
Feedback:
Output bellows expands, pushes beam back
Beam returns to near-initial position
Result
Direct action: PV ↑ → Output ↑
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 17 / 28
Reverse Action
Principle:
Controller action can be reversed
Method:
Swap pneumatic connections:
PV applied to upper bellows
SP applied to lower bellows
Result
Reverse action: PV ↑ → Output ↓
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 18 / 28
Proportional Control Action: Motion-Balance Principle
Alternative Design:
Some controllers use motion-balance
Lever is free to tilt (not force-restrained)
System maintains constant nozzle gap
Operation:
PV ↑ → lever moves away from nozzle
Nozzle backpressure ↓
Output bellows collapses
Nozzle gap returns to constant
Result
Reverse action: PV ↑ → Output ↓
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 19 / 28
Automatic and Manual Modes
Bumpless Transfer:
Balance indicator shows pressure
Allows mode switching without process difference between output bellows
disturbance and manual regulator
Prevents sudden output changes
Auto → Manual:
1 Adjust output regulator until balance indicator
shows zero
2 Switch transfer valve to ”manual”
Manual → Auto:
1 Adjust setpoint regulator until balance
indicator shows zero
2 Switch transfer valve to ”auto”
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 20 / 28
PD Control Action
How to Add Derivative:
Place a restrictor valve between nozzle and
output feedback bellows
Causes bellows to fill/empty slowly
Effect on Sudden Change:
PV or SP changes suddenly
Output pressure spikes before bellows equalizes
Creates derivative action
Effect on Ramping Input:
Output shifts more than proportional alone
Adds offset pressure
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 21 / 28
PI Control Action
How to Add Integral:
Add a reset bellows opposite the feedback
bellows
Add a restrictor valve (integral adjustment)
How It Works:
Reset bellows creates delayed opposing force
Output bellows must stay ahead of reset
bellows
Creates constant pressure difference
Key Characteristic:
Larger error = faster output change
Error determines velocity of output
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 22 / 28
PI Controller: Thought Experiment
Initial Condition:
PV and SP differ by 3 PSI
Output = 3 PSI to feedback bellows
Reset (integral) valve = closed
Reset bellows = 0 PSI (no integral action)
Reset Valve Opens Slightly:
Output (3 PSI) slowly fills reset bellows
Reset bellows pushes down on beam
Baffle moves closer to nozzle
Output pressure rises
Result:
Output must stay ahead of reset bellows
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 23 / 28
How Integral Action Works
Key Principle:
Output pressure = 3 PSI + rising
Reset bellows pressure = rising (but slower)
Difference = constant (3 PSI)
Why Constant?
Output must always be 3 PSI ahead of reset
bellows
Creates constant flow through restrictor
Output rises at constant rate
Error Effect:
Larger error = larger pressure drop
Faster flow = faster output change
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 24 / 28
Integral Action: Key Takeaways
What It Does:
Output continues to change as long as error
exists
Eliminates steady-state error
How to Adjust:
Change restrictor valve opening
Add capacity tanks to reset bellows
The ”Race”:
Feedback bellows = instantaneous
Reset bellows = delayed
Output must always lead reset bellows
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 25 / 28
Foxboro Pneumatic Controllers
Figure: Foxboro Model 43AP Figure: Foxboro Model 130
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 26 / 28
Comparison of PID Controller Implementations
Type Technology Advantages Disadvantages
Pneumatic Air pressure (3–15 ✓Robust ×Slow response
psi) ✓Safe (explosive ar- ×Low precision
eas)
Analog Elec- Op-amps, analog ✓Fast response ×Noise sensitive
tronic circuits ✓Continuous signal ×Component drift
Digital Microcontroller, ✓High precision ×Sampling delay
PLC ✓Flexible tuning ×Programming
needed
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 27 / 28
Thank You Questions?
Asso. Prof. Dr. H. Benyezza (Eng. 3 Ind. Elec.) Chapter 3 28 / 28