Putting all things together:

tp 1 2
tp = = y 1 e 1 e
d n 1 2 max

Delay time :
1 2
Overshoot : Mp e 0.8 0.9 1.4
td
n n
1 2
percentage : e 100%

4.5( 0.2) 1.5 2.5 2

Rise time : tr
n n n

Settling time:
ts

ln tol 1 2 ln(tol ) 3, or 4, or 5

2) When = 1, d = 0
n2 n2
H s 2 2
s 2 n s n s 2n s n2
2

n2

s n 2
n2 1
Y s H s U s
s n s
2

1 1 n

s s n s n 2
y t u s t e nt u s t nte nt u s t
u s t e 1 n t u s t
t n
The tracking error:
e t e t 1 nt us t
n

ess lim e t 0 No steady - state error

t

y ss 1 ess 1

Note : e t is monotonically decreasing to zero

y t 1 e t is monotonically increasing to 1

No oscillation
No overshoot
3) Over damped: > 1 b b 2 4ac
n2 p1, 2
H s 2 2a
s 2 n s n2 n


s n n 2 1 s n n 2 1
1
Y s H s
s
1 n 1 n 1

s 2 2 1 p1 s p1 2 2 1 p2 s p2

n e p2t e p1t
y t us t u s t
2 1 p2 p1

p1 n 2 1n , p1 p2 0
p2
n 2 1 n
 n e p1t e p2t
e t 1 y
2 2 1 p1 p2

ess lim e t 0 No steady-state error

t

y ss 1

Furthermore, e t 0 t

y t never over shoots

 Prototype 2nd order system:
n2
H s 2
s 2 n s n2

Useful case :
Under damped :
0 1 target

tp = = tp 1 2
d n 1 2 ymax 1 e 1 e

1 2
percentage overshoot: e 100%
1.5 ~ 2.5 2
Rise time: tr
n n
ln(tol 1 ) 3, 4,5
2
Settling time: ts

n
From and n :

d 1 2
n
2 2
n d

pole: p j The other way :

d

2 d2
Example: Given 0.6, n 5, estimate specs.
sol: n 0.6 5 3
d 1 2 n 4

peak time: tp 0.8sec.
d

1 2
Mp e 0.96 10%
5 5
settling time for 1% : ts 1.67 sec.
3
4
2% tol : t s 1.33sec.
3
1.8
rise time: tr 0.36sec.
n
Try to remember these:
3 4 5
ts or or

5% 2% 1%

M p 5% or 10% or 16% or 25%

0.7 0.6 0.5 0.4

2
tr
n
 k
Example: +

-
s s 1

When given unit step input, the output looks like:

i.e. M p 25%
t p 3sec.
Q: estimate k and .
sol: from M p 25% 0.4

t p 3sec. d
tp 3
1 2 n

d
n 1.14
1 2 3 1 0.42
From block diagram:
k k
H s 2
s s k s2 1 s k

n2
Match this against 2
s 2n s n2
 k n2 1.142

1
2n 2 0.4 1.14

Solve for and k , get:

1.09

k 1.42
 Effects of additional zeros
Suppose we originally have:
1 t H0 s y0 t i.e. step response
Now introduce a zero at s = -z
s
H s 1 H 0 s
z
The new step response:
1 s 1
Y s H s 1 H 0 s
s z s
s
1 Y0 s
z
1 d
y t 1 y0 t
z dt
 1
y t y 0 t y0 t
z
Typically, y0 t is rising at beginning.
1
1) If z 0 , y 0 t 0
z
y t is rising faster than y0 t , overshoots
before y0 t , and has larger overshoot.
1
2) If z 0 , y 0 t 0
z
y t rises slower than y0 t
smaller overshoot ?
may start off in wrong direction
i.e. y t 0 initially
 e.g. z 0 :

Effects: Increased speed, i.e. t r , t d , tp

Larger overshoot, M p
Might increase ts
When z < 0, the zero s = -z is > 0,
is in the right half plane.

Such a zero is called a nonminimum phase zero.

A system with nonminimum phase zeros is called a

nonminimum phase system.

Nonminimum phase zero should be

avoided in design.
i.e. Do not introduce such a zero in your controller.
 Effects of additional pole
Suppose, instead of a zero, we introduce
a pole at s = -p, i.e.
1
H s s H0 s p 0
p
1
1 1 1
Y s H s s H0 s
s p 1 s
p
Y0 s
s p
p
is a first order low pass filter
s p
with corner frequency at p.
 L.P.F. has smoothing effect, or
averaging effect

Effects: Slower, i.e. t r , t d , t p

Reduced overshoot, M p
May increase or decrease ts