Network Functions and S-Domain Analysis
Network Functions and S-Domain Analysis
Network Functions and S-Domain Analysis
-option-
10.1 Complex frequency and generalized impedance
complex frequency, generalized impedance and admittance
10.2 Network function
network functions and impedance analysis
10.4 s-domain analysis
Poles and zeros, forced response, natural response, stability
10.5 Network scaling
impedance scaling, frequency scaling
10-1 電路學講義第10章
10.1 Complex frequency and generalized impedance
Basics
1. Complex-frequency waveform representation of a real signal
x(t ) = X meσt cos(wt + φ x ) = Re[ X me jφ x e(σ + jw)t ] ≡ Re[ X e st ]
s ≡ σ + jw : complex frequency,X = X me jφ x
2. For a two-terminal element, if
i(t ) = I meσt cos(wt + φi ) = Re[Ie st ], v(t ) = Vmeσt cos(wt + φv ) = Re[V e st ]
V I
then Z (s) ≡ : generlizedimpedance,Y (s) ≡ : generlizedadmittance
I V
3.
resistor capacitor inductor
Z(s) R 1/sC sL
Y(s) 1/R sC 1/sL
10-2 電路學講義第10章
4. s-domain diagram
10-3 電路學講義第10章
Discussion
−2 t
1. Ex. 10.2 find i(t), L=1H, R=5Ω, C=0.1F , v(t ) = 20e cos(4t + 90 )
o
1 R RLCs 2 + Ls + R
Z (s) = sL + = sL + =
1
+ sC RCs + 1 RCs + 1
R
V
I= , V = 20∠90o , s = −2 + j 4, Z (−2 + j 4) = 2.5∠143.1°
Z (s)
20∠90o
I= = 8∠ − 53.1°, i (t ) = 8e − 2t cos(4t − 53.1°)
2.5∠143.1°
10-4 電路學講義第10章
2. Ex.10.3 Miller-effect capacitance
(V in − V out )
I
1 = = sC (V in − V out )
1
sC
V out = − AV in
→ I 1 = sC (V in + AV in ) = sC (1 + A)V in
I1
= sC (1 + A) = sC1
V in
→
I − I1 1
2 = = sC (1 + ) = sC2
V out − AV in A
10-5 電路學講義第10章
3. Ex. 10.4 Generalized impedance converter, find Zin(s)
Z1
for the 1st op V in = V a ...(1)
Z1 + Z 2
V in = V b ...(2)
V −V c
for the 2nd op I in = in ...(3)
Z 5
V − Vc
I 3 = b
...(4)
Z 4
Z1 + Z 2
V in − V in
V a − V b (1) Z1 Z
I3 = = = 2 V in ...(5)
Z3 Z3 Z1Z 3
Z2Z4
( 2 ),( 5 )
V in − (V in − V in )
Z2Z4 Z1Z 3 Z2Z4
(4) → V c = V b − Z 4 I 3 = V in − V in , (3) → I in = = V in
Z1Z 3 Z5 Z1 Z 3 Z 5
Z 4 , Z1 , Z 3 , Z 5 resistive elements, Z 2 capacitor → Z in inductor
Z1Z 3 Z 5
⇒ Z in = , 1 1 −1
Z 2 Z 4 Z 3 = Z 5 = , Z1 = Z 2 = Z 4 = R → Z in = 2 2 , Z in ( jw) = 2 2 = Req ( w)
sC sC R wC R
frequency - dependent negative resistor
10-6 電路學講義第10章
10.2 Network functions
Basics
1. t-domain approach: for a second-order circuit, its input signal x(t) and
response y(t) are related by
x(t ) = Re[ X e st ], y (t ) = Re[Y e st ]
a2 y"+ a1 y '+ ao y = b2 x"+b1 x'+bo x → (a2 s 2 + a1s + ao )Y = (b2 s 2 + b1s + bo ) X
Y b2 s 2 + b1s + bo
⇒ = ≡ H ( s) : network function, driving - point function,
X a2 s + a1s + ao
2
transfer function
2. s-domain approach: much easy!!
• use s-domain impedance representation
• use methods of series-parallel reduction, source conversion,
proportionality, node and mesh equation, voltage and current
dividers…to obtain response
• Y
H ( s) =
X 10-7 電路學講義第10章
Discussion IL VL
1. Ex.10.5 use t-domain approach to find H 1 ( s ) = , H 2 ( s ) =
V s Vs
1
KVL → Li L' + Ri L +
C ∫ i L dt = v s ...(1)
d
dt R ' 1 1
(1) → i L" + iL + i L = v s'
L LC L
1
s
R 1 1 IL L
→ [s + s +
2
] I L = sV s , H 1 ( s ) = =
L LC L V s s2 + R s + 1
L LC
d2
dt 2 R ' 1
(1) → ' v L" + vL + v L = v s"
v L = Li L L LC
R 1 VL s2
→ [s + s +
2
]V L = s V s , H 2 ( s ) =
2
=
L LC V s s2 + R s + 1
L LC
10-8 電路學講義第10章
2. Repeat Ex.10.5 using s-domain impedance analysis
1
s
Vs IL sC L
IL = → H1 (s) = = 2 =
sL + R +
1 V s s LC + sRC + 1 s 2 + R s + 1
sC L LC
sL VL s 2 LC s2
VL = V s → H 2 (s) = = 2 =
sL + R +
1 V s s LC + sRC + 1 s 2 + R s + 1
sC L LC
10-9 電路學講義第10章
3. Ex.10.6 obtain the network functions and differential equations to
relate iL, vC to vS
1
VS VS 1 R
IL = = + sC
1 R
sL + sL + V C =V S R = V S 1 + sRC
1
+ sC 1 + sRC 1 R
sL + sL +
R 1
+ sC 1 + sRC
I 1 + RCs
→ H1 ( s) = L = R
V S RLCs 2 + Ls + R V R
→ H 2 (s) = C =
→ RLCiL" + LiL' + RiL = RCvs' + vs V S RLCs 2 + Ls + R
→ RLCvC" + LvC' + RvC = Rvs
10-10 電路學講義第10章
4. Ex.10.7 obtain the network functions to relate v1, v2, v3 to vin
RF + Rµ ( µ − 1) Rµ + Rµ
vout = v3 = v3 = µv3
Rµ Rµ
~ ~
node matrix analysis Y- Y [V ] = I s
2 sC + 2G 0 − sC
[Y ] = 0 2sC + 2G − G
− sC −G sC + G
~
constraint equation V out = µV S , [I S ] = I S + [G ][I ]
sCV in + 2GµV 3 sCV in 0 0 2Gµ V 1
[I S ] = GV in = GV + 0 0
in 0 V
2
0 0 0 0 0 V 3
2( sC + G ) 0 − ( sC + µG ) V 1 sCV in
0 2( sC + G ) −G V = GV
2 in
− sC −G sC + G V 3 0
V V V
⇒ 1, 2, 3
V in V in V in 10-11 電路學講義第10章
10.4 s-domain analysis
Basics
1.Properties of network function H(s)
• It’s much easier to obtain in the s-domain than in t-domain
• It simplifies the calculation of forced response
• One can determine the natural response
2. It is equivalent to
• impulse response or system response in the courses of linear
system, communication theory, signal processing
• Green’s function in electromagnetics
3. Y bm s m + bm −1s m −1 + ...b1s + bo ( s − z1 )( s − z2 )...(s − z m )
H ( s) = = = K
X an s n + bn −1s n −1 + ...a1s + ao ( s − p1 )( s − p2 )...( s − pn )
m ≤ n, n equals to the number of energy - storage elements
ai , bi ∈ R, pi : pole, zi : zero, K : gain factor
K ∈ R, pi , zi ∈ R, or in complex cojugate pairs eg. ( s − pi )( s − pi∗ )
10-12 電路學講義第10章
4. A network function is completely defined by gain factor, zeros and
poles.
5. Pole and zero are displayed in s-plane s=σ+jw as pole-zero
pattern
s ( s + 10)
H ( s) = K
[ s − (−4 + j 7)][s − (−4 − j 7)]
10-13 電路學講義第10章
6. Forced response calculation
• given H(s)
• if x ( t ) = Re[ X e s o t ]
• then
( so − z1 )( so − z 2 )...
y (t ) = Re[ Y e so t ] = Re[ H ( so ) X e so t ] = Re[ K X e so t ]
( so − p1 )( so − p 2 )...
H ( so ) = H ( so ) ∠ H ( so )
so − z1 so − z 2 ...
H ( so ) = K
so − p1 so − p 2 ...
∠ H ( so ) = ∠ K + [ ∠ ( so − z1 ) + ∠ ( so − z 2 )...]
− [ ∠ ( so − p1 ) + ∠ ( so − p 2 )...]
10-14 電路學講義第10章
7. Natural response calculation
• given
Y ( s − z1 )( s − z 2 ) s 2 − ( z1 + z 2 ) s + z1 z 2 Z (s)
H (s) = =K =K 2 =K
X ( s − p1 )( s − p 2 ) s − ( p1 + p 2 ) s + p1 p 2 P(s)
• then
y"−( p1 + p2 ) y '+ p1 p2 y = K [ x"−( z1 + z 2 ) x '+ z1 z 2 x ]
→ y "N − ( p1 + p2 ) y N' + p1 p2 y N = 0
→ characteristic polynomial P ( s ) = s 2 − ( p1 + p2 ) s + p1 p2
→ poles are eigen values, or characteristic values of circuit' s natural response
→ e.g. pi = σ i + jwi , wi ≠ 0, then pi +1 = σ i − jwi ,
y N (t ) = Ai eσ i t + jwi t + Ai∗eσ i t − jwi t + ... = 2 Ai eσ i t cos( wi t + ∠Ai ) + ...
2 Ai eσ i t cos( wi t + ∠Ai ) : one " mode" of the natural response
The term mode is widely used in electromagnetics, e.g. TEM mode.
10-15 電路學講義第10章
8. Modes associated with distinct poles
underdamped
• For repeated poles, the mode
p = −α + j w2 − α 2
1 o waveform is multiplied by t,
p2 = −α − j wo − α
2 2
or power of t.
e.g. p1 = p2 = ... = pr ,
oscillation
then y N (t ) = ( A1 + A2t + ... Ar t r +1 )e pt
α=0
stable unstable
LHP • 2nd-order circuit
RHP
y"+2αy '+ wo2 y = 0
overdamped
p = −α + α 2 − w2
1 o
p2 = −α − α − wo
2 2
critically damped
p1 = p2 = −α 10-16 電路學講義第10章
9. A network is stable when all of its poles are in the left half of
the s-plane, irrespective of pole-zero cancellation.
pi = σ i + jwi in the LHP → σ i < 0 ⇒ y N (t ) = Ai eσ it + jwit → 0, as t → 0
10-17 電路學講義第10章
Discussion
1. Ex. 10.8 find poles and zeros
s 4 + 16s 3 + 164 s 2
H ( s ) = −5
( s + 32)( s 2 + 36)( s 2 + 40 s + 400)
s 2 [ s − (−8 + j10)][ s − (−8 − j10)]
= −5
[ s − (−32)][s − ( j 6)][s − (− j 6)][s − (−20)][s − (−20)]
10-18 電路學講義第10章
2. Ex. 10.9 given H(s) and x(t), find forced response
− 6s s
H (s) = = − 6
s 2 + 12s + 45 [ s − (−6 + j 3)][s − (−6 − j 3)]
x(t ) = 10e − 4t cos 3t = Re[ X e sot ] = Re[10e ( −4+ j 3) t ]
so
H ( s o ) = −6 = 2.37∠ − 108.5°
[ so − (−6 + j 3)][so − (−6 − j 3)]
→ Y = H ( so ) X = 23.7∠ − 108.5°
10-19 電路學講義第10章
3. Ex. 10.10 analyze H2(s) in Ex.10.7
V2 a[ s 2 + ( 2 − µ )as + a 2 ]
H 2 (s) = =
V in ( s + a )[ s 2 + 2(2 − µ )as + a 2 ]
1
a= , if a = 5
RC
P ( s ) = ( s + 5)[ s 2 + 2(10 − 5µ ) s + 25]
p1 = −5, p2 , p3 = −(10 − 5µ ) ± (10 − 5µ ) 2 − 25
1 < µ < 2 → underdamped response
µ = 2 → oscillation response
µ > 2 → unstable response
10-20 電路學講義第10章
10.5 Network scaling sˆ sˆ
Basics − z1 )...( − z m )
(
k kf
1. Network scaling Hˆ ( s) = K f
sˆ sˆ
( − p1 )...( − pn )
impedance scaling Zˆ = k m Z , k m > 0 kf kf
Hˆ (s) 1, H ( s) = V / I
H(s)
Zˆ = k m Z → Kˆ = k mq k nf −m K , q = − 1, H ( s ) = I / V
R, L, C,w Rˆ, Lˆ, Cˆ , wˆ 0, otherwise
Zˆ = k m Z → Rˆ = k m R
( s − z1 )...( s − zm )
H (s) = K k sˆL k L
( s − p1 )...( s − pn ) Zˆ = k m Z = k m sL = m ≡ sˆLˆ → Lˆ = m
kf kf
km km k f 1 C
Z = km Z =
ˆ = ≡ → Cˆ =
sC sˆC sˆCˆ km k f
10-21 電路學講義第10章
Discussion
1. Ex. 10.11 find H(s) using network scaling
Vs
IL = let k m = 5 × 10 −3 , k f = 2 × 10 −4
1
sL +
1 k C
+ sC Rˆ = k m × R = 10, Lˆ = m L = 1, Cˆ = = 0.025
R kf km k f
H (s) =
IL
=
1 1 sˆ + 4 sˆ − zˆ1
Hˆ ( s ) = = =
V s 0.04 s + 1
sˆ +
1 sˆ 2 + 4 sˆ + 40 ( sˆ − pˆ 1 )( sˆ − pˆ 2 )
0.5 ×10 −3 + 25 ×10 −9 s sˆ
+
1
s-z1 40 10
=K
( s-p1 )( s-p2 ) zˆ1 = −4, pˆ 1 , pˆ 2 = −2 ± j 6, Kˆ = 1
kf
zˆi = k f zi , pˆ i = k f pi , Kˆ = K
km
→ zi = −2 ×10 4 , p1 , p2 = (−1 ± j 3) × 103 , K = 25
10-22 電路學講義第10章