EE 42/100

Lecture 6: Network Theorems

ELECTRONICS
Rev C 3/1/2012 (8:30PM)
Prof. Ali M. Niknejad

University of California, Berkeley

Copyright
c 2012 by Ali M. Niknejad

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 1/?? – p.

 Superposition
• If a circuit is linear, then by the principal of superposition, we can analyze the
circuit one source at a time. The total response is the sum of the outputs due to
the individual sources.
• This is clear if we re-write the matrix equation as follows

Ax = b = b1 + b2 + · · ·

• Note that we have partitioned the source terms so that each bk only contains a
single source. Clearly, the solution is given by

x = A−1 b1 + A−1 b2 + · · · = x1 + x2 + · · ·

• where xk is the solution with source k turned on and all other sources set to zero.
That means that other voltage sources are short circuited (zero voltage) and other
current sources are open circuited (zero current).

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 2/?? – p.

 Example: Analyzing a Circuit with Superposition

v1
is2
R1 R2

vs1 R3
is1

• In this example there are three independent sources. When we analyze the circuit
source by source, the circuit is often simple enough that we can solve the
equations directly by inspection.
• First turn of is1 and is2 . Zero current means that we replace these sources with
open circuits. The node voltage v1 is therefore by inspection

R2
v1vs1 = vs1
R1 + R2

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 3/?? – p.

 Superposition Example (2)

v1
is2
R1 R2

vs1 R3
is1

• Next turn off vs1 (short circuit) and is2 (open circuit). The current is1 will therefore
divide between r2 and r1 and establish a voltage at node v1 (equivalently, it see’s
a parallel combination of R1 and R2

R2 R1
v1is1 = is1
R1 + R2

• Finally, we turn off all sources except is2 . Now only R1 and R2 remain (R3 is
dangling)
R2 R1
v1is2 = is1
R1 + R2

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 4/?? – p.

 Superposition Example (3)
• By superposition, the node voltage v1 is the sum of the three node voltages due to
each source

R2
v1 = v1is1 + v1is2 + v1vs1 = (vs1 + R1 (is1 + is2 ))
R1 + R2

• We can verify the solution by performing KCL directly at node 1

(v1 − vs1 )G1 + v1 G2 − is1 − is2 = 0

v1 (G1 + G2 ) = vs1 G1 + is1 + is2

• The answer here is just as fast but we don’t have any intuition about the operation
of the circuit. We’re perhaps more likely to make an algebraic error if it’s all math
without any thinking.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 5/?? – p.

 Thevenin Equivalent Circuit

Rth
+
vth +
+

“Black Box”

• A powerful theorem in circuit analysis is the Thevenin equivalent theorem, which

let’s us replace a very complex circuit with a simple equivalent circuit model.
• In the black box there can be countless resistors, voltage sources (independent
and dependent), current sources (independent and dependent), and yet the
terminal behavior of the circuit is captured by two elements.
• How can this be? Well, there is a big assumption in that all the resistors are linear
(follow Ohm’s Law) and all dependent sources are also linear.
• The equivalent circuit representation is often called a “black box", since the details
of the circuitry are hidden.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 6/?? – p.

 Thevenin Derivation
Rth
+ + +
voc vth + vth
− −
+

“Black Box”

• Since a circuit is linear, then no matter how complicated it is, it’s response to a
stimulus at some terminal pair must be linear. It can therefore be represented by a
linear equivalent resistor and a a fixed constant source voltage due to the
presence of independent sources in the circuit.
• To find the equivalent source value, called the Thevenin voltage source vth , simply
observe that the open-circuit voltage of both the “black box" and the original circuit
must equal, which means
vth = voc
• In other words, open-circuit the original circuit, find its equivalent output voltage at
the terminals of interest, and that’s vth

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 7/?? – p.

 Thevenin Source Resistance
Rth
+
vth
isc vth +
Rth
+

“Black Box”

• To find equivalent Thevenin source resistance Rth , notice that in order for the
terminal behavior of the two circuits to match, the current flow into a load resistor
has to be the same for any load value. In particular, take the load as a short circuit.
• The output current of the Thevenin equivalent under a short circuit is given by

vth
Rt

• Equating this to the short-circuit current of the original circuitry, we have

vth
isc =
Rt

or equivalently
vth voc
Rth = =
isc isc
A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 8/?? – p.
 Thevenin Equivalent Example

vs R1 v2
R2
R3
v3
R4

• In the above circuit, we will calculate the Thevenin equivalent circuit.

• We begin by finding the open-circuit voltage. In this case, it’s a simple application
of the voltage divider.
R4
v3 = v2
R3 + R4

R2 ||(R3 + R4 )
v2 = vs
R1 + R2 ||(R3 + R4 )

R4 R2 ||(R3 + R4 )
voc = v3 = vs
R3 + R4 R1 + R2 ||(R3 + R4 )

R4 R2 (R3 + R4 )
=
A. M. Niknejad
(R3 + R4 )(R1University
(R2 of+California,
R3 + R4 ) + R2 (R3 + R4 ))
Berkeley EE 100 / 42 Lecture 6 p. 9/?? – p.
 Thevenin Example (2)

vs R1 v2
R2
R3

R4 isc

• Next we find the short-circuit current in the original circuit. The resistance loading
the source under this condition is given by

R2
isc = is
R2 + R3

vs vs (R2 + R3 )
is = =
R1 + (R2 ||R3 ) (R2 + R3 )R1 + R2 R3

R2 vs
isc =
R1 R2 + R1 R3 + R2 R3
voc
Rth =
isc
A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 10/?? –p
 Non-Linear Components and Sources
• A non-linear resistor has a non-linear I-V relation. For example

V = r1 I + r 2 I 2 + r3 I 3

or
V = cos(I · Rx )
• A non-linear dependent source is a non-linear function of one or more independent
currents/voltages in the circuit. Some examples of non-linear dependent sources
are:

vk = Ki2j A squarer

ik = Ki2j + M ij A quadratic function

vk = Kvj · vm A multiplier

vk = Aij + B Isn’t this linear?

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 11/?? –p

 Linear Relations
• By definition, a linear relation means that if two inputs are applied, then the output
of the sum is the sum of the individual outputs. Suppose

y = Kx

then y1 = Kx1 y2 = Kx2

y = K(x1 + x2 ) = Kx1 + Kx2 = y1 + y2

• Now suppose y = Kx + z. Note that

y1 = Kx1 + z

y2 = Kx2 + z

y = K(x1 + x2 ) + z = Kx1 + Kx2 + z 6= y1 + y2

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 12/?? –p

 Calculating Rth Directly

ix
+ + +
Rth + v
x
−
+ +

“Black Box” “Black Box”

No “Deps”

• If there are no dependent sources in the “black box”, then you can calculate the
Thevenin equivalent circuit directly by setting the independent source values to
zero. As before, zeroing out sources means shorting voltage sources (zero volts)
and open circuiting current sources (zero current). Now just “inspect" to find the
equivalent Thevenin resistance.
• For the general linear circuit, another approach to find Rth is to probe the circuit
with an independent voltage/ current source while zeroing out all internal sources.
The current / voltage is monitored and the ratio of the test voltage to test current is
the equivalent Rth
vx
Rth =
ix

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 13/?? –p

 Example: By Direct Method

R1

R2
R3

R4

• Let’s redo the same example but now we zero out the voltage source and redraw
the circuit. Now we can readily find Rth by simply observing that the resistors in of
the original circuit are in series/parallel:

Rth = R4 ||(R3 + R1 ||R2 )

R4 R3 + R4 R1 ||R2
=
R4 + R3 + R1 ||R2

R4 R3 (R1 + R2 ) + R4 R1 R2
=
(R4 + R3 )(R1 + R2 ) + R1 R2

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 14/?? –p

 Thevenin Example (2)

R1 v1 R3 ix

vs1
R2 Gv 1 + +
vx
−

• Consider the above circuit with two sources. We find the equivalent open circuit
voltage by writing KCL at the intermediate node.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 15/?? –p

 Norton Equivalent Circuit

+
iN RN
+

“Black Box”

• In a similar vein, we can also replace a complex circuit with a Norton equivalent
circuit, which contains a current source and a shunt source resistance.
• To find the equivalent source value, we find the short circuit current for both the
model and the original circuit and note that

in = isc

• Similarly, to find the Norton resistance, we note that if we open-circuit the model,
the output voltage is given by

voc = in Rn or voc
Rn =
isc
• This is the same exact equation as before.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 16/?? –p

 Source Transformations
Rth

vth + iN RN

• A trivial application of Norton or Thevenin’s Theorem shows us that we can

transform from one representation to the other. For instance, starting from the
Thevenin, let’s find the Norton. Short circuit the Thevenin to find

Vth
in = isc =
Rth

and it’s trivial to see that Rn = Rth .

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 17/?? –p

 Norton Equivalent Example

is1

R3

+ R1
vs + R2
−

• For the above circuit, find the Norton equivalent circuit. We first start out by finding
the short-circuit current.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 18/?? –p

 Norton Example (2)

is1

R3

+ R1
vs + R2
−

• Next we find the open-circuit voltage.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 19/?? –p

 Norton Example (3)

is1

R3

+ R1
vs + R2
−

• The same calculation can be done using superposition.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 20/?? –p

 Voltage Amplifier Example

R1 R3

+ +
vs vx R2 + Kvx
− −

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 21/?? –p

 Transconductance Amplifier Example

R1 R3

+ +
vs vx R2 Gvx
− −

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 22/?? –p

 Maximum Power Transfer
Rth

vth + RL

• An important question arises in many electrical circuits when we wish to interface

one component to another while maximizing the power transfer to the second
component.
• A good example is a battery with internal resistance and a motor. What’s the best
“load" resistance to choose in order to maximize the power transfer?
• Interestingly, if we maximize the current or voltage transfer, the power transfer is
exactly zero.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 23/?? –p

 Maximum Power Transfer (cont)

Rth

vth + RL

• No matter how complicated the black box source, we can represent it as a

Thevenin equivalent circuit, vth and Rth .
• The general procedure is to find the power through the load and then to find the
optimal load value. We can take the derivative of the load power with respect to
the load resistance and set it equal to zero (occurs at only a maximum or
minimum). A second derivative test confirms that it’s a peak.
„ «2
2 vth
PL = IL RL = RL
RL + Rth
„ «2 „ «3
dPL vth 2 1
= − 2RL vth =0
dRL RL + Rth RL + Rth

(RL + Rth ) = 2RL → RL = Rth

The optimum load resistance is equal to the Thevenin Equivalent Value, or it’s Matched.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 24/?? –p

 Example of Maximum Power Transfer

Rth

Amp vth + RL

RL

• A common example occurs when designing amplifiers in RF applications. For a

receiver, we wish to extract the maximum possible power from the antenna (since
the received signal can be very weak). We can represent the antenna by its
Thevenin equivalent source resistance Rth .
• The load resistance, which is the input of the amplifier in this case, should present
a value that is matched to the antenna impedance.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 25/?? –p

 The Wheatstone Bridge

R1 R3
VG

+
R2 Rx

• The Wheatstone Bridge (originally invented by Samuel Hunter Christie in 1833 and
then popularized by Sir Charles Wheatstone in 1843) is used to measure an
unknown resistance. It is highly accurate and only requires an adjustable resistor
(or set of well known calibrated resistors) and a method of measuring zero current,
such as a galvanometer.
• Since we only need to measure if the current is zero, we can do this very precisely
with a galvanometer.
• The Wheatstone bridge is often used with strain gauges, thermocouples, and other
transducers.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 26/?? –p

 Wheatstone (2)

R1 R3
VG

+
R2 Rx

• The operation of the Wheatstone Bridge is as follows. One leg of the bridge
contains an unknown resistance which we would like to find. The other leg
contains an adjustable resistor R2 (of known value). The goal is to adjust the
resistor R2 until the circuit is “balanced", in other words until no current flows
through the galvanometer.
• Under the balanced condition, there is no current Ig , so the current in R1 and R2
is the same, say I1 , and the current through R3 and Rx is also the same, I3 . By
KVL, under the balanced condition Vg = 0, we have

I3 R3 = I1 R1

I3 Rx = I1 R2

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 27/?? –p

 Wheatstone (3)
• Taking the ratio of these two currents, we have

R3 R1
=
Rx R2

• Which means that the unknown resistance if found

R3
Rx = R2
R1

• In practice, we vary R2 until we achieve balance. In some commercial units, the

scale factor R3 /R1 can be changed as well.

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 28/?? –p

 Wheatstone Example
Suppose that R1 = 100Ω, R2 can be adjusted from 1Ω steps from 0 to 100Ω, and R3
can be selected to be 100Ω, 1 kΩ, 10 kΩ, or 100 kΩ. (a) If the bridge is balanced with
R2 = 36Ω and R3 = 10 kΩ, find Rx . (b) What’s the largest Rx that can be measured?
(c) For R3 = 10 kΩ, what is the accuracy of the measurement?
• (a) This is a trivial application of the equations

R3 10, 000
Rx = R2 = 36 = 3600Ω
R1 100

• (b) The largest Rx is given by

R3,max 100, 000

Rx,max = R2,max = 100 = 100 kΩ
R1 100

• (c) The accuracy of the measurement is set by the scale factor. As R2 changes by
R3
1Ω, the value of Rx changes by R , which is equal to 100Ω.
1

A. M. Niknejad University of California, Berkeley EE 100 / 42 Lecture 6 p. 29/?? –p