ECEN 457 (ESS)

Fall 2015

NODAL EQUATIONS BY INSPECTION

Knowing how to write systematically network analysis equations is
primordial to good circuit design. Here we will review one popular
analysis technique partially used in SPICE.
Consider the following example

a) Using conventional KCL

Z1 Vx Z2
Vo

Vin Z3 Z4

Vin  Vx Vx Vx  Vo
  0 (1)
Z1 Z3 Z2

1
Re - writing (1) becomes

 1 1 1  Vo Vin
Vx       (1 )
 1
Z Z 3 Z 2 Z 2 Z1

Conventional ,

Vx  Vo Vo
 (2)
Z2 Z4

Re - writting (2) becomes

Vx  1 1 
-  Vo     0 (2)
Z2  Z4 Z2 

2
 How can we write directly by inspection Eqs (1) and (2)?

We can do it using the indefinite admittance matrix which must be expressed as

YV  I
Where Y is the indefinite admittance matrix, V is the nodal voltage vector and I is the
independent current source vector. In the example here presented we have two
independent nodal voltage Vx and Vo. Thus we can write directly the equations (1) and
(2).
 1 1 1  V V
Vx      o  in  0
 Z1 Z 2 Z3  Z 2 Z1
or equivalent
 1 1 1  V V
Vx      o  in (1 )
 Z1 Z 2 Z3  Z 2 Z1

Note that for the node under analysis we multiply Vx by the sum of all admittances
connected to node X minus the contributions of the neighbor connected nodes, in
this case Vin and Vo. These (negative) contributions consist of the neighbor
voltages multiplied by the sum of the admittances connected between Vx and Vin
and Vo.
3
Let us practice with node at Vo , thus

 1 1  Vx
Vo     0 2
 Z2 Z4  Z2

Consider next a more general example.

Let us assume a Norton transformation and that the ground node becomes different
from zero and equal to Vy i.e.

Vx Z2
Vo
where
V in
Vin
Z1 Z1 Z3 Z4 I in 
Z1
Vy

4
Now we will have 3 nodal equations

 1 1 1   1 1  V
Vx      Vy     o  I in (3)
 Z1 Z 2 Z3   Z1 Z3  Z 2

 1 1   1 1 1  Vo
- Vx     Vy        I in (4)
 Z1 Z3   Z1 Z3 Z 4  Z 4

Vx Vy  1 1 
-   Vo     0 (5)
Z2 Z4  Z2 Z4 

Notice that if you sum all the coefficients of Vx , Vy or Vo the sum is zero, which
is a consequence of KCL.

5
 Observe that you can write equations (3), (4), and (5) in a matrix form.

1 1 1 1 1 1  
Z        Vx  I n 
 1
Z 2 Z3 Z1 Z3 Z2
   
    
 1    
 
1 1

1

1

1   Vy    I 
 Z1 Z3 Z1 Z3 Z 4 Z4   n
     
    
 1 1 1 1   
 Z    V  0 
 2 Z4 Z2 Z4   o 

Here you can verify that the sum of any column or row of Y is zero. Also observe the
symmetry of the matrix Y due to the fact that we do not have depended sources.

Next let us consider another example to illustrate how the dependent sources can be
incorporated in the analysis.

6
 Vx
Vo
Z2
V in
Z1 Z1 Z3 Z4
ID
Vy
where
ID=Gm (Vx-Vy)
We can go directly to our previous matrix and incorporate ID or writing the 3 nodal equations.

 1 1 1   1 1  Vo Vin
Vx      Vy     
 1
Z Z 2 Z 3   1
Z Z 3  Z 2 Z1

 1 1   1 1  Vo
 Vx     Vy  
1
  
V
  in  G m Vx  Vy 
 Z1 Z 3   Z1 Z 3 Z 4  Z 4 Z1

 1 1 

Vx
 Vy
1
 Vo   
  G m Vx  Vy 
Z2 Z4  2
Z Z 4 

Note that we must have only independent sources on the right side of the
equations. Thus we will move the dependent sources into the left side of the
equations. 7
  1 1 1   1 1  Vo Vin

Vx    
 
 Vy     
 Z1 Z 2 Z 3   Z1 Z 3  Z 2 Z1

 1 1   1 1 1  Vo Vin

 Vx    
 G m   Vy    
 Gm   
 Z1 Z 3   Z1 Z 3 Z 4  Z4 Z1

 1   1   1 1 
 Vx   G m   Vy   G m   Vo     0
 Z2   Z4   Z2 Z4 

Observe that for all floating nodes the sum of the coefficients of all nodal voltages is zero.

8
 EXAMPLES TO ILLUSTRATE NODAL ANALYSIS

EXAMPLE 1
VDD Vo
+
RL Vin gm Vin  Vs 
RL
Vo -
Vs
Vin Rs
Vs
Rs
Small Signal Circuit
- Two Nodal Equations for Vo and Vs.
Vo
  g m Vin  Vs 
RL

Vs
 g m Vin  Vs 
Rs
Solving these two equations yields
Vo  g m R L

Vin 1  g m R s
9
 Example 2
R1 V y R2

V1 A1 Vo
RG
R1 R2 VZ R3
Vx
V2
A2

Assume first ideal op amps. (Note Vx = Vy )

 1 1  V1
Vx     0 (1a )
 R1 R 2  R1

 1 1  Vz V2
Vx      (1b)
 R1 R 2  R 2 R1

10
Also note that

R3
Vz   Vo (1c)
RG

(1b) - (1a) yields

Vz V1 V2  R V V V
-      3  o  1  2
R 2 R1 R1  R G  R 2 R1 R1

V2  V1  R G R 2 R 2 R G 
Vo   V2  V1 
R1 R3 R1R 3

R 2R G
Vo  V2  V1 
R1R 3

11
Example 3
Z1 Vx ZF

V1 A Vo

V2

If A is not ideal we can write

Vo  AV2  Vx 

or

Vo
Vx  V2 
A
12
Thus we can solve the following equivalent circuit replacing Vx .

 V  Vo 
Z1  2  ZF
 A
Vo Using Nodal Equations

Vo  V  1 1  V1
A    V2  o     0
V1 ZF  A  Z1 Z F  Z1

Vo Vo  1 1  V1  1 1 
V2        V2   
Z F A  Z1 Z F  Z1  Z1 Z F 

Thus (multiplying both sides by ZF)

ZF  ZF 
 V1 1   V2
  1
Z1 Z
Vo 
1 Z  1 Z 
1  1  F  1  1  F 
A Z1  A Z1 

13
Example 4 R01

C1 RQ

C2
RK
Vin -VB R02
Vn1 A1 -1
VB Vn2 A2 Vo

 1 1  V V
Vn1    sC1   sC1VB  o  in (1)
 R K R o1  R o1 R K

 1 1  VB  1 
  sC2  
Vn 2 
R R   R  Vo  sC2  R   0 (2)
 o2 Q o2  Q

14
 Note that

0 for ideal A1,2


Vn1,2   VB,O
 for finite A1,2
 A

Let us assume the ideal case then eliminating VB from (1) and (2) yields.

Vin Vo
 VB  
sR K C1 sR o1C1

1

Vo R K R o 2C1C2

Vin s2  s 1 1

R QC2 R o1R o 2C1C2

VB V
Reader. Solve for Vn1   and Vn 2   o
A1 A2

15
 SUMMARY OF NODE METHOD

YV  Is
Where Y is a n-1 x n-1 array, V is a vector, representing the unknown node voltages,
of dimension n-1. Is is also a vector of n-1 dimension of the know current sources.

How to generate systematically the admittance matrix Y:

i j

16
 For voltage sources we can consider two illustrative cases:

GOUNDED VOLTAGE SOURCES

R5  G1  G 3  G 2  G 3   V1   Vs 
    R 
R1 1 R3       V1 
Vs 2     s
R2 R4
  G1 G 3  G 4  G 5   V2   R 5 

1 G1 v 1   i v

FLOATING VOLTGE SOURCES 2 G 2  G 3 V2  G 3V3  i v

iv Vs 2 R3
1 3 3  G 3V2  G 3  G 4 V3  0

R1
R2 R4 but
V1  V2  Vs
or V1  Vs  V2
G1  G 2  G 3 V2  G 3V3  G1Vs By adding equations 1 and 2 and substituting V1, we can obtain

 G 3V2  G 3  G 4 V3  0

17