Basic Of PID Control

&
Tuning

BASIC ON/OFF CONTROL

Inflow

START

Qin
L2

Solenoi
d

L1
S
P

P
Qou
t

Solenoid
power
supply

In On/Off control- control signal is either 0% or 100%

Control at set point not achievable, a dead band must
be incorporated
Useful for large , sluggish system particularly those
incorporating electric heaters
Examples- Large tank level control
- Lube oil tank temperature control by
heaters

BASIC PROPORTIONAL
CONTROL
V1
Qin

50%
60%

Qo=50
Air to Close

SP
50kPa
40kPa
L
T

Qout

50t/h
60t/h

20-100 kPa

Qo=60

 P.C provides a control signal, proportional to the

magnitude and direction of the error signal.
After a disturbance, P.C will provide only a new mass
balance situation. A change in control signal requires
a change in error signal, therefore offset will occur.
P.C stabilizes an error; it does not remove it

TERMINOLOGY
M=Measure signal

k= gain

SP=set point

b=bias

e(error)= SP - M
m= controller signal output
m=ke + b
PB=proportional band
Gain(k)=

100%
PB

Proportional band is defined as that input signal span change, in

percent, which will cause a hundred percent change in output

PRACTICAL PROPORTIONAL
CONTROL
V1

Qin

50%
60%
60
%

LI
C

50kPa
L
T

40kPa
45kPa
48kPa
Qout

60t/h
50t/h

Flow
Change
Kp=
Air to Open
1
Kp=
2
Kp=
5
Kp 1
5
2
=

20-100 kPa

Loss In
Volume

Higher PB

Lower PB

Outflow
Inflow
t0

t
t0

Level originally SP

t1

New Level S

Offset
Offset
t1

t1

New Level SP

STEP
CHANGE
TIME
WIDE PB
OFFSET

MODERATE PB

NARROW PB

SUMMARY & OPTIMUM PB

Highly stable but sluggish system

Fast acting system with large offset
Unstable on/off system
Require bias to avoid undesirable situation because m=
ke,
so m= ke + b
Optimum setting for PB should result in the process
decaying in a decay mode

A/4

A/16

RESET OF INTEGRAL
ACTION
To restore the process to the set point after disturbance
then only proportional is insufficient
The additional inflow must replace the lost volume

Initial mass balance

Reset action

Loss in volume

Outflow

Inflow

Set point
Offset removed

Additional Control signal restores

process to set point

PHENOMENA OF RESET
ACTION
Reset action is the Integration of the error signal to zero
After time say t reset action has repeated original proportional
response , it is called repeat time
R.A is defined as either reset rate in repeats per minute (RPM) or
reset time in minutes per repeat (MPR) . MPR= 1/RPM
Reset action will cause a ramping of the output signal to provide
the necessary extra control action.

st
Fa

se
e
R

r
No

t
ese
R
l
ma
t
Slow rese

Ke
Proportiona
l
Response

Exam
ple

A direct acting controller has a proportional band of 50% is subjected to

a sustained error. The set point is 50% and the measurement 55%.
After 4 minutes the total output signal from the controller has increased
by 30%. What is the reset rate setting in RPM and MPR?

Soluti
on

PB = 50%

Since k will be negative

gain = 100% = 2
50%

Proportional Signal = -2 x error = -2 x

-5%
+10%
Total=signal
after 4 minutes = +30%
=P+I
Integral Signal = +20%
i.e., integral action has repeated original proportional signal twice in 4
minutes, 1 repeats per 2 minutes or 0.5 repeats per minute.
Reset rate = 0.5 RPM or 2 MPR

SUMMAR
Y
Mathematical expression for integral action

m = control signal
e = error signal (e = SP . M) (+ or -)
k = controller gain ( = ) ( = +)
TR = reset time (MPR)
b = bias signal

Reset action removes offset

If reset action is faster than the process can respond, Reset
Windup can occur
Reset Action makes a control loop less stable
Do not subject process loops with reset control to sustained
errors . the control signal will be ramped to the extreme
value . reset windup will occur.

DERIVATIVE ACTION
The proportional mode considers the present state of the
process error
The integral mode looks at the past history of the error
The derivative mode anticipates the future values of the
error and acts on that prediction
Derivative is related to the rate of change of the error
signal and an anticipatory control, which provides a large
initial control signal to limit the final deviation
inpu
t

Proportiona
l action

outp
ut
Derivati
ve
action

PHENOME
NA
Mathematical equation for PD controller

m = controller signal
k = controller gain
TD = derivative time
e = error
b = bias signal

It should help reduce the time required to stabilize an error ,

derivative action ceases when the error stops
changing.
Its use, in practice, is also limited to slow acting processes.

EXAMPL
E
Consider a simple flow control system
B

Proce
ss

Contr
ol
signa
l

A
t
Proportional
0
action
A-B
Rate
action
A-B

C
t
1

t
2

Tim
e
Rate action due to end of
Rate increase in e
action
B-C
Control signal at end of
excursion
Rate action due to end of
increase in e

Proportional
action
B-C

SUMMA
RY
Derivative or rate action is anticipatory and will
usually reduce, but not eliminate, offset.
Its units are minutes (advance of proportional action).
It tends to reduce lag in a control loop.
Its use is generally limited to slow acting processes.

PID
Response

PID Tuning
The term tuning is used to describe methods used to select
the best controller setting to obtain a particular form of
performance.
There are three methods that widely used for tuning
1. Process reaction
method

This method uses certain measurements made from

testing the system with the control loop open so that no
control action occurs.
A test input signal is applied to the correction unit and the
response of the controlled variable determined.

Give step input as a test signal

The graph of controlled variable is plotted against time

Final
value
Maximum
gradient
line

Measur
ed
Variabl
e

Original
value

Response of
Controlled
M
variable
Percentage
change of the
variable
per
minute

Start of test
signal

Tim
e

 Criteria given by Ziegler and Nichols

Type of
controll
er

Kp

P/ML

PI

0.9/ML

Ti

Td

3.3L

The basis
these criteria
is 0.5L
to give a closed-loop
PID behind
1.2P/M
2L
response for theLsystem which exhibits a quarter amplitude
decay
EXAMPLE
8

L = 5 min

0
0

10

15

M = 8/10 = 0.8 % / min

Kp = 1.2P / ML = 1.2 X 10/0.8 X 5
=3
Ti = 2L = 10 min
Td = 0.5L = 2.5 min

2. Ultimate cycle
method
1. Set the controller to manual operation and the plant near
to its normal operating conditions.
2. Turn off all control modes but proportional.
3. Set Kp to a low value, i.e. the proportional band to a wide
value.
4. Switch the controller to automatic mode and then
introduce a small set-point change, e.g. 5 to 10% and
observe the response.
5. Set Kp to a slightly higher value, i.e. make the proportional
band narrower.
6. Introduce a small set-point change, e.g. 5 to 10% and
observe the response.
7. Keep on repeating 6 and 7 until the response shows
sustained oscillations which neither grow nor decay.

Note the value of Kp giving this condition (Kpu) and the period
(Tu) of the oscillation.

tu

The Ziegler and Nichols criteria controller settings to have

quarter amplitude decay is given by Table
Type of
controll
er

Kp

Ti

0.5 Kpu

PI

0.45
Kpu

Tu/1.2

PID

0.6 Kpu

Tu/2

Td

Tu/8

2. Quarter
Amplitude Decay

Control
led
variabl
e

The controller is set to proportional only

With a step input to the control system,
the output is monitored and amplitude
decay is determined
If the amplitude decay is greater than a
quarter the proportional gain is increased
If the amplitude decay is less than a
quarter the proportional gain is
decreased
By method of trial & error the test input
is repeated until a quarter wave
amplitude decay is obtained
Note this value of proportional gain
The integral time constant is set to be
T/1.5
The derivative time constant is T/6

Amplitude
reduced by a
quarter

Tim
e

Thank You!