Circuits, Devices, Networks, and Microelectronics

CHAPTER 2. BASIC CONCEPTS AND RULES OF ENGAGEMENT

2.1 CONDUCTANCE and RESISTANCE
Although it may have seemed preposterous to earlier civilizations that anything could flow through a
solid rod of metal, it probably would have had more meaning if they touched the rod to something
electrically hot. The concept of there being electrical charge stored at a high potential energy may have
been something of a shock to your ancestors, but is now accepted as a natural fact. And the concept of
charge flow is also accepted and formulated by the rule known as Ohms law.
Ohms law is a law of fluid flow. It says that under a pressure difference across a conduit containing
charge, charge will flow. The flow rate of the charge will be proportional to the pressure difference
(voltage difference) between the ends of the conduit. This concept is mostly just common sense, but is
stated as
I GV

(2.1-1a)

where I is the charge flow in Coul/sec, V is the voltage (electrical pressure difference) in Volts and G is
the conductance of the conduit. Ohms law is also stated as
V IR

(2.1-1b)

Where R = 1/G is defined as the resistance. The usage of I for electrical current, V for voltage, G for
conductance and R for resistance is the conventional electrical nomenclature.
The distinction of an electronic fluid is that it is a flow of charges (quantized by q = 1.6 10-19C = single
electron charge). It is also called an electron gas since it can be compressed, although we will not do so
until later.
Some conduits are more conductive to the flow of electrical charge than are others. The conductance
depends on (1) geometry and (2) conductive properties of the material (otherwise known as conductivity).
The reality of the property of conduction is that all materials will conduct electrons, some more than
others. Metals are very conductive. Oxides are not. Electrical conduction properties are typically
identified by tables, either of conductivities or of resistivities.
Figure 2.1 shows a representation of a uniform geometry for a conductance (or resistance) with crosssection of area A and length L. Within an electronic circuit, strips of material not unlike that of the
geometry shown form the conductive paths for flow paths of the electronic fluid.

Circuits, Devices, Networks, and Microelectronics

Figure 2.1-1. Representation of electrical conductance/resistance (uniform geometry)

The figure tells all. The conductance is
G

A
L

(2.1-2a)

and/or the resistance is

R

L
A

(2.1-2b)

The coefficients and are the conductivity and resistivity, respectively. These symbols are the ones
used in the table referenced by the URL and are the conventional notation.
From the point of view of a flow of charge fluid it should make sense that a conduit with more crosssectional area A will proportionally allow more current to flow. If A2 is twice as large as A1 then twice
as much fluid will flow. And a conductance with twice the length L would be half as conductive.
Equations (2.1-2a) and (2.12b) are statements thereto.
In circuit context the conductance path is more likely formulated in terms of its resistance to the flow of
current. Resistance carries the unit of choice. The measure is in the ohm (symbol: ) and is defined by
Ohms law as (1 = 1 Volt/Amp). So the coefficient of resistivity (= ) must then be in terms of (.m)
(in semiconductor context it is more often given in .cm). Conductivity is given in terms of (1/.m) or
more correctly, as Siemans/meter (S/m). With equations (2.1-2a) and (2.1-2b) that confirms resistance in
and conductance in mho (= ohm spelled backwards) (symbol:
. The symbol is difficult to find in
print but is more commonly used for conductance measure than is Siemans.
Since many orders of magnitude are possible, prefixes are common and should be used with relative
fluency and abandon. Later in the practice of semiconductor electronics the resistances will likely default
to k and current to mA. This is sometimes called the k mA convention since milliamperes (mA)
result when potentials in volts are applied across k.
Consider the following example

10

Circuits, Devices, Networks, and Microelectronics

EXAMPLE 2.1-1(a): An integrated circuit interconnect strip of polysilicon is printed on an insulated

substrate. What is its resistance?

SOLUTION: From equation (2.1-2b) the mathematics is

R

L
20 m
.025..cm
A
2.0 m 1.0 m

.025

20
10 4
2.0 1.0

= 2.5k

Even though this mathematics is complete and sufficient it has some extra overhead in the prefixes. Note
the factor of 104. It is the units conversion from cm to m. The relative relationships for unit prefixes are
an essential part of your knowledge base. Pausing over conversion details is only for the unpracticed.
Also notice that the answer was not in but in k. The result must be in a sensible form. Scientific
notation is not invalid but is not used when dealing resistance measure.

EXAMPLE 2.1-1(b): It appears that the stripe of example 2.1-1(a) is actually two strips in parallel.
What is the resistance of one of these strips?

SOLUTION:
R

20
L
20 m
.025
10 4
.025 ..cm
1 .0 1 .0
A
1 . 0 m 1 . 0 m

= 5k

This was an unnecessary calculation. It should be fairly evident that a single stripe is half as conductive
as two in parallel. And if the conductance is half then the resistance is twice.

11

Circuits, Devices, Networks, and Microelectronics

These two examples point out a context fact that should be added to your knowledge base as a simple rule
of resistance topologies, namely that conductances in parallel will add, i.e.

Geq G1 G2 G3
Which is the same as

(2.1-3a)

1
1
1
1

Req R1 R 2 R3

(2.1-3b)

The rule is stated in terms of resistances since resistance are the circuit component, usually given in terms
of ohms (or k). Consider the exercise set of Example 2.1-2.

EXAMPLE 2.1-2: What is the equivalent resistance of each of the parallel resistance options shown?

SOLUTIONS: By inspection, listed in parentheses.

The symbol for resistance (shown) looks like a crinkled channel even though it most likely is a thin film
or a surface strip. Carbon films are a favored choice for discrete resistance components.
The absence of math calculations in Example 2.1-2 reflects the suggestion that simple topologies should
be done by inspection, even if approximate (as was done for parts (d), (e) and (f) of the above example).
Example 2.1-2 represents the old school form, as if a circuit were assembled from discrete parts. But
reality is that circuits are now more likely to begin their existence as a printed form for which resistances
are either connection paths or designated stripes.
The simplicity of calculation is even more evident for resistances in series. Consider Example 2.1-1(b) if
the length is assessed as a sum of sections.
12

Circuits, Devices, Networks, and Microelectronics

EXAMPLE 2.1-3: What are the resistances of the two segments? Assume polysilicon.

SOLUTION:
R1

L1
8 m
.025 ..cm
A
1 . 0 m 1 . 0 m

.025

R2

L2
12 m
.025 ..cm
A
1 . 0 m 1 . 0 m

.025

8
10 4
1 . 0 1 .0

= 2.0k

12
10 4
1 .0 1 .0

= 3.0k

This was another example of doing more math than necessary. It should be evident that a strip of 8m
length is (= 2/5) times that of a 20m length. Then R1 = (2/5) 5.0k k, by inspection.
Using the answer from Example 2.1-2(b)

Req = 5 k = 2k + 3 k.

Example 2.1-3 emphasizes the rule of resistances in series which is

Req R1 R2 R3

(2.1-4)

EXAMPLE 2.1-3: What is the equivalent resistance of each of the resistance topologies shown?

SOLUTIONS: By inspection, answers listed in parentheses.

13

Circuits, Devices, Networks, and Microelectronics

As reflected by the example it is worth noting that the subsets of resistances, whether series or parallel,
can be done by inspection. In these examples the numbers are designed for simplicity. But even if they
were not, a fairly reasonable result can still be roughed out. For example part (b) should acknowledge
that 3k||6k = 2k by inspection. And then Req will be 2kk4k
Similar situations exist with the other topologies. For example the right end of part (e) has 24k||8k=
6kby inspection. And from there 6k + 6k = 12k, which is in parallel with 6k, which gives
4kby inspection. The 4k is in series with 8 k and that is in parallel with the 12k at the front end to
give an Req of 6k. All is done by inspection. A symbolic representation of the decomposition process
for part (e) is shown by figure 2.1-2.

Figure 2.1-2 Symbolic decomposition of equivalent resistance (part (e) of example 2.1-3.)
**Part (f) of Example 2.1-3 is a special topology known as a binary ladder , also called an R-2R ladder.
The result Req = 2R is derived by a right-to-left, step by step process like that of figure 2.1-2 and is the
outcome no matter how long the ladder may be. The R-2R ladder should be consigned to your special
circuits collection since it has considerable utility.
The rest of the story is that the resistance topologies serve to sort and divide electrical current and voltage
according to Ohms law. The topology therefore begins to take the form of a network, consisting of
components in the form of branch elements and junctions between components in the form of nodes.
The simplest of the passive components are resistances, but as the reach of circuit topologies is extended,
some of the branches will be sources and others will be various forms of control and reactive elements.
Since a network will distribute electrical quantities along and through the network of nodes and branches
it is in order that sources of electromotive energy be included in the mix. The simplest source is that of
the electrochemical cell, a.k.a. battery and the symbol (two equivalent forms) is shown by Figure 2.1-3.
The figure also shows a common usage of resistances in series with a voltage source and serves to form a
particularly useful topology called a voltage divider.

14

Circuits, Devices, Networks, and Microelectronics

Figure 2.1-3. Voltage dividers and symbols for voltage sources. (Vx values by inspection.)
Voltage dividers are a means of parsing the source voltage into lesser fractions. This context also gives
rise to a special slide-wire resistance component called a potentiometer, represented by figure 2.1-4. The
potentiometer has a wiper arm that rides upon a track. Many circuits that have a fine-tuning requirement
may include one or more pots with a screw slot. Precision circuit applications often include a
potentiometer with a helical slide track.

Figure 2.1-4. Potentiometer = resistance track with wiper arm to select VP.

2.2 NODE VOLTAGES and BRANCH CURRENTS

As represented by the resistance ladders, a circuit is a set of paths and junctions. It may be extended in all
directions and be interconnected as an array of branches and nodes. The lace of interconnections is
assumed to be ideal with zero resistance (unless the interconnect material happens to be a poor conductor
as in the case of refractory metals). As a mathematical necessity all circuit topologies should have a
reference node, usually identified as ground. Circuit simulation software always requires a ground
node. All of the other node voltages may then be identified relative to the ground (GND) reference.
The ground symbol sneaked into the discussion on voltage dividers because they were an application
oriented toward voltages. Relative to the reference point (i.e. GND) the network will be characterized by
a set of node voltages (all relative to GND) and a set of branch currents. And even though we may not

15

Circuits, Devices, Networks, and Microelectronics

necessarily need to know each and every one of the electrical facts of the network, the rest of the story is
in the analysis methods by which the circuit can be decomposed and its electrical facts identified.
If the topology is a ladder of series and parallel resistance combinations, its electrical facts can be
decomposed in the same way the resistance network was decomposed, but also by use of the rules of
voltage division represented by equation (2.2-1).

VB

R2
VA
R1 R2

(2.2-1)

for which VA is the voltage value of a parent node (as illustrated by the figure) relative to ground and the
outcome VB is also relative to ground.
Decomposition of the node voltages for a ladder network is illustrated by figure 2.2-1. Figure 2.2-1 is
the same as that of figure 2.1-2 except that a GND and a source voltage are included. Consequently the
node voltages can be extracted by voltage division from left to right. Branch currents can subsequently be
obtained from the node voltages and resistances if desired.

Figure 2.2-1. Electrical facts by decomposition. Node voltages are labeled.

VA = VS = 24V
VB = [4/(8 + 4)] VA
VC = [6/(6 + 6)] VB
Ix = Vx/8
IS = VS/Req

= (4/12) 24
= (6/12) 8
= 4.0/8
= 24/6

= 8.0V
= 4.0V
= 0.5mA
= 4.0mA

= Vx

Ladder topologies of the form represented by figure 2.2-1 are primarily concept exercises. In contrast, the
R-2R ladder is of functional significance since no matter how many stages it may have Req = 2R. And
16

Circuits, Devices, Networks, and Microelectronics

therefore each stage will consist of an R in series with a next stage Rnext = R , which then forms a divideby-two voltage divider as illustrated by figure 2.2-2.

Figure 2.2-2. R-2R ladder. Decomposition = divide-by-2 for each stage. VN = VS/2N
Since each stage is a divide-by-2 then the node voltages along the ladder may be ascertained by a divideby two count from the source node, as shown by the figure. And since the node voltages are a divide-bytwo, the branch currents are also a divide-by-two. As well as being electrically simple and direct it should
be evident why the R-2R ladder belongs in the special collection of circuit toplogies.
Most circuit topologies do not decompose as readily as do the the ladder forms. So general rules
associated with the nodes and branches of the network are of necessity. There are two rules for
networks, identified in 1845 by a gentleman named Gustav Kirchoff, and known as Kirchoffs laws. The
first is a law of nodes and is given by

(2.2-2)

node

It is a law of conservation of current. Equation (2.2-2) is called Kirchoffs current law (KCL). In effect
the statement is that current out = current in. By convention the current out of a node is designated as
positive and the current into the node is designated as negative. Sometimes direction may be indicated by
an arrow, in which case the polarity of the discovered value will determine whether the arrow was a good
choice of flow direction.
The second of the Kirchoff laws is a statement about the potential drops around a loop, and is given by

(2.2-3)

loop

Equation (2.2-3) is called Kirchoffs voltage law (KVL) and recognizes that the sum of voltage
increments and decrements around a loop must add to zero. It is an electrical statement of the
conservation of potential energy. Around a loop the circuitous return to its starting place will bring the
17

Circuits, Devices, Networks, and Microelectronics

potential energy back to its starting potential. The convention is that the CW (clockwise) direction
around a node will be positive and that resistance branch components will each have a positive voltage
drop across them due to Ohms law as stated by equation (2.1-1b).
These two network laws offer a simple and straightforward means for decomposition of the network into
electrical facts. For the simplest case in which the branches are of the form of resistances, Ohms law
equations (2.1-1a) and (2.1-1b) produce a set of (linear) equations amenable to linear computational
algorithms. And whether the process is done by back-of-the-envelope or by simulation software, node
voltages and branch currents (a.k.a the electrical facts of the topology) can be efficiently resolved.
All of the electrical facts within a network are a consequence of the energy sources embedded within the
network structure. Courtesy of the two forms for electrical energy, electrical sources will fall into two
possibilities: (1) voltage sources and (2) current sources. The ideal forms and symbols of these source
options are represented by figure 2.2-3, along with their electrical characteristics.

Figure 2.2-3. Electrical source types. (Note that negative current can flow into a positive source
and negative voltage can fall across a positive current source).
Source characteristics are not just defined by the level of electrical energy but also by the slopes of their
electrical (I vs V) specifications. In figure (a) the slope of infinity corresponds to a slope conductance of
infinity. Therefore an ideal voltage sources will have zero resistance and can provide any level of current
at the voltage specified, whether to warm up a resistance or to displace a small planet. A slope of zero is
a conductance of zero and so ideal current sources have infinite resistance and provide any level of
voltage at the current specified, no matter what magnitude or polarity. Ideal sources are of computational
simplicity but are also a little excessive since they can promise anything, no matter how unrealistic.
Current sources may sound like an artifact but there is a component (discussed later) which will act very
much like an ideal current source.
The identification of the current source as an energy source also lends itself to identification of another
simple and direct topology, called a current divider, shown by figure 2.2-4.

18

Circuits, Devices, Networks, and Microelectronics

Figure 2.2-4. Current divider. For part (b) I1 = 1.0mA, I2 = 1.5mA and I3 = 0.5mA
Current division results for path resistances in parallel. Parallel paths parse the available current
according to their conductances, i.e.
I3

G3
I1
G 2 G3

(2.2-4)

where I1 is the value of the current of parent branch. Equation (2.2-4) might be compared to that of the
voltage divider (equation (2.2-1). The current divider is a common subcircuit topology.
Whatever the form in which electrical energy may exist, the circuit will have the effect of parsing it into
sums and fractions. So the rest of the story is in the network topology. Small subsets in the form of
parallel and series combinations can be subdued and asserted as simplifications. But in most part the
network will need to be resolved by KCL and KVL.
The more general analysis acknowledges a circuit topology as a network of interconnected active and
passive branches. As a first cut it will suffice to let the passive branches be of the form of resistances
until more entertaining components are addressed. Nodes are not unlike solder junctions even though
the solder blob analogy suffers if the circuit is an integrated circuit rather than a joined assembly. Nodes
may be stretched or spread, but in most part will be an identifiable junction point.
A node will selectively link to other nodes by its set of connecting branches. Then KCL and Ohms law
equation (2.1-1a) I = GV will give a sum of conducting links from any one node to the other nodes at the
ends of its set of branches. There will be as many equations as there are nodes. The result is a set of
simultaneous equations of the form:
G11V1 G12V 2 G1nV n I S 1

(2.2-5)

G 21V1 G 22V 2 G 2 nV n I S 2

G n1V1 G n 2V 2 G nnV n I Sn

where the Gij are sums of conductances. Since a node is not usually connected to each and every one of
the other nodes many of the Gij terms will be zero (a.k.a. sparse). This method is called nodal analysis.
19

Circuits, Devices, Networks, and Microelectronics

Consider the network of figure 2.2-5, which is chosen because it contains both types of sources and a
tractable set of resistance branches.

Figure 2.2-5. Typical resistance network. Rendition (a) = pspice. Rendition (b) = emphasis on
the nodal nature of the network.
The convention is that positive flow is out of the node. Then nodal analysis at node VC will be
I X VC V A R 4 VC V B R3 0

(2.2-5)

If this is reorganized into an equation in nodal form then

1
1
1
1

VC

VB
VA I S
R
R
R
R
4
3
4
3

(2.2-6)

The form of equation (2.2-6) is less clumsy if we use conductances instead of resistances, i.e.

G3 G 4 VC

(2.2-7)

G 3V B G 4V A I S

As a shortcut, use the concept that each node voltage will push a (+) current out to the other nodes and
the other nodes will push back(-). If we apply this principle to node VB then

G1 G 2 G3 V B

(2.2-8)

G1V A G 3VC 0

Equation (2.2-8) avoids the extra steps used to acquire equation (2.2-7). Take note that the GND node
cannot push back since it has value zero.
And (for this example) node VA needs no equation since it is constrained to be the same as the source VS.
So we might let values be added to figure 2.2-5. The network of figure 2.2-5 with values is represented
by example 2.2-1.

20

Circuits, Devices, Networks, and Microelectronics

EXAMPLE 2.2-1: For the network shown execute a nodal analysis to determine node voltages and
branch currents.
SOLUTION:

Node VA = 5.0

(= constraint)

Node VB (using equation (2.2-8) gives

(0.5 + 1.0 + 2.0)VB 1.0 VC 0.5 5.0 = 0
Node VC gives

-1.0 VB + (1.0 + 1.0)VB 1 5.0 = 2.0

So we are left with

3.5VB VC = 2.5
- VB + 2VC = 7.0

eliminating VC (Gaussian elimination: multiply top equation 2 and add to the bottom equation) gives
(7.0 1) VB = (5 + 12)

or

VB = 2.0V

Back substitution of VB = 2.0 into the first equation gives VC = 4.5V

Knowledge of all of the node voltages gives

I1
I2
I3
I4

= (5 2)/2.0 = 1.5mA
= (2 0)/0.5 = 4.0mA
= (2 4.5)/1.0 = -2.5mA
= (5 4.5)/1.0 = 0.5mA

Be aware that it would have been much simpler to let pspice do

the grit and grind of the numerical work (as shown).

Nodal analysis is usually the simplest and most direct method by which the electrical facts of a network
may be ascertained and is the method favored by circuit simulation utilities.
The other network analysis method makes use of KVL and equation (2.1-1b) and is based on circuit
loops. As a convention the set of loops are interior loops, although that is not a hard and fast
requirement. The loop mathematics is defined by a set of unknown loop currents with conceptual flow
direction being clockwise. So the application of KVL will give a set of simultaneous equations of the
form
21

Circuits, Devices, Networks, and Microelectronics

R11 I 1 R12 I 2 R1n I n V S 1

R 21 I 1 R 22 I 2 R 2 n I n V S 2

(2.2-9)

R n1 I 1 R n 2 I 2 R nn I n V Sn

where the Rjk are sums of resistances and the Ij are the loop currents. Loop analysis also goes by the name
mesh analysis and each loop is defined by a loop current. The solution of equation (2.2-9) gives the loop
currents and then the currents in the various branches of the circuit are then determined from the loop
currents. The default flow direction of a loop current is clockwise.
The concepts are represented by figure 2.2-6 which is a 3-loop topology.

Figure 2.2-6. Rendition (a) = pspice. Rendition (b) = default (= interior) analysis loops of the network.
Applying KVL to loop IA of figure 2.2-6b and an I R drop across each resistance gives
V S I A I B R1 I A I C R 2 0

(2.2-10)

Take note that voltage drops across resistances that are shared with an adjacent loop will link IA to
neighbor loop currents. If this equation is reorganized to a more convenient form then

R1 R2 I A R1 I B

R3 I C V S

(2.2-11)

As a shortcut it we might note that the terms associated with the IA loop are positive and the contribution
by those for the adjacent loops are negative. Then by inspection, loop IB will be

R1 I A R1 R4 R3 I B R3 I C 0

(2.2-12)

The current in loop IC is defined by the current source IS so that IC is constrained to be

I C I S

(2.2-13)
22

Circuits, Devices, Networks, and Microelectronics

If values are applied to figure 2.2-6 and elected to be the same as used for Example 2.2-1 then analysis
and value comparisons may be made, as follows:

EXAMPLE 2.2-2: For the network shown execute a loop analysis to determine branch currents and
node voltages.
SOLUTION:

Loop current IC = -2mA

(constraint)

Loop IA (using equation (2.2-11)) gives

(2.0 + 0.5)IA 2.0 IB 0.5 (-2) = 5
Loop IB gives
-2.0 IA + (2.0 + 1.0 + 1.0) IB 1.0 (-2) = 0
So we are left with

eliminating IB

2.5IA 2IB = 4
-2IA + 4IB = -2

(Gaussian elimination: multiply top equation 2 and add to the bottom equation):

(5.0 2) IA = (8 2)

IA = 2.0mA

And back substitution of IA = 2mA in the second equation gives IB = +0.5mA

Knowledge of all of the voltage nodes gives

I1
I2
I3
I4

= (IA IB) = (2 0.5) = 1.5mA

=(IA IC) = (2 (-2)) = 4.0mA
= (IB IC) = (0.5 (-2)) = 2.5mA
= IB = 0.5mA

Node voltages are VA = VS = 5.0V (by constraint)

VB = 0.5 I2 = 0.5 4.0 = 2.0V

VC = VA I4 R4 = 5 0.5 1.0 = 4.5V
and the answers are the same as Example 2.2-1. Once again we might acknowledge that it would have
been much simpler to let pspice do the grit and grind numerical work as shown.

23

Circuits, Devices, Networks, and Microelectronics

As illustrated by the two examples it should be evident that the nodal analysis (KCL) method and the loop
analysis (KVL) method are equals in the world of circuit analysis. Simulation software prefers nodal
analysis since the node count defines the number of simultaneous equations and the size of the
conductance matrix as represented by the linear equation form

(2.2-14)

where Gjk are conductance sums. The diagonal terms will always be sums of positives and the offdiagonal terms will always be sums of negatives. Source terms ISK should be expected to be relatively
sparse as reflected by example 2.2-1. Off-diagonal Gjk should also be relatively sparse.
Likewise the mesh analysis will be a linear set of equations of the form

(2.2-15)

where Rjk are resistance sums. The diagonal terms will be sums of positives and those that are offdiagonal will be sums of negatives. The source terms VSK should be expected to be relatively sparse and
off-diagonal Rjk terms are usually also sparse.
Mesh analysis may have difficulty in selection of the best set of interior loops since the circuit simulation
software is blind. It also may have difficulty if the network is not planar.
The primary application for the general KCL and KVL nodal and mesh methods is in circuit simulation
platforms since linear algebra is a quick and simple inversion. Otherwise these methods afford a general
means to the user to derive predictive algebraic forms if the topology is fortunate enough to be relatively
small and self-contained.

2.3 THEVENIN and NORTON EQUIVALENT CIRCUIT THEOREMS and MAXIMUM

POWER THEOREM
Electrical networks are always a part of system and consequently will be vested with one or more I/O
ports. The most basic black-box formulation will be that of the two-port network, with an input port and
an output port. Multi-terminal ports are diagnosed as a superset of two-terminal ports.
And key to the network function is the output port. It has two terminals one of which is GND. The port
therefore will appear to be a source limited by an internal resistance, not unlike that of non-ideal voltage
source. As an equivalent voltage source its internal resistance is the factor that limits the current. This

24

Circuits, Devices, Networks, and Microelectronics

perspective and reduced equivalent network is known as Thevenins theorem. The output port is
otherwise expected to drive an external load RL. The Thevenin theorem is represented by figure 2.3-1.
Figure 2.3-1(a) represents a circuit which otherwise is a mix of active branches and passive branches.
Figure 2.3-1(b) represents its Thevenin equivalent, which is a voltage source Vth and a limiting resistance
Rth , also called the Thevenin voltage and the Thevenin resistance, respectively.

Figure 2.3-1. Thevenin theorem representation. Figure (a) is the (representative) network and
Figure (b) is the Thevenin equivalent.
A dual form of Thevenins theorem is shown by figure 2.3-2 and is called Nortons theorem which restates Thevenins theorem in terms of an equivalent current source In shunted by resistance Rn.

Figure 2.3-2. Representation of Nortons theorem. Figure (a) is the (representative) network and
Figure (b) is the Norton equivalent.
Since the two interpretations must have the same electrical effect then
I n Vth Rth

(2.3-1a)

Vth I n R n

(2.3-1b)

R n Rth

(2.3-1c)

25

Circuits, Devices, Networks, and Microelectronics

and sometimes it is convenient to recognize this equivalence of the two interpretations by the schematic
equivalent representation as shown by figure 2.3-3.

Figure 2.3-3. Source equivalence: The two circuit forms are equivalent and are the schematic
representation of equations (2.3-1a), (2.3-1b) and (2.3-1c).
The realization of Thevenin equivalent circuits is readily achieved courtesy of the reduction methods and
network analysis techniques used heretofore. When there is no load, Vth is merely the value of the node
voltage at the output node. And when all of the electromotive sources are turned off, then there is nothing
left but the resistance network, which can be collapsed to a single Rth ( = Rn) value. Thevenin resistance
Rth is also the output resistance and may also be acknowledged as Rout.
Electromotive forces may be due to sources of different types. A voltage source with VS = 0 is a short
circuit since an ideal voltage source has zero resistance. A current source with IS = 0 is an open circuit
since an ideal current source has infinite resistance. When these criteria are applied to the Thevenin and
Norton source equivalents of figure 2.3-3 then the resistance into the output port = Rth = Rn (= Rout).
Consider the following example (Example 2.3-1)

EXAMPLE 2.3-1: Determine the Thevenin equivalent

resistance Rth for the circuit of Example 2.2.1 if output is at VC.

SOLUTION: Let all sources be zero. The outcome is then

represented by figure E2.3-1(b) for which VS becomes a shortcircuit and IS becomes an open circuit.

Figure E2.3-1(a) Example network with Vout

The topology of figure 2.3-1(b) can also be mentally reoriented to the form shown by figure 2.3-1(c).
This reorganization is not essential and mostly just for conceptual convenience. For reference it might be
noted each of the resistances are between the same nodes as they were in figure 2.3-1(b).

26

Circuits, Devices, Networks, and Microelectronics

Figure E2.3-1(b) Same as (a) with sources zeroed.

Figure E2.3-1(c) Same as (b) but reoriented.

The last figure lets us ply our skills at assessment of a resistance network by inspection, for which we
should take note that

R1||R2 = 2.0||0.5

= 0.4k

R3 + 0.4k = 1.0 + 0.4 = 1.4k

R4||1/4k = 1.0||1.4

= 0.58k (approximately)

or Rth = 0.58k

The rest of the story is that either In or Vth must be identified. Since Example 2.3-1 is the same as
Example 2.3-1 then Vout = VC = Vth courtesy of nodal analysis.

Vth = VC = 4.5V

(for which)

i.e.

Vth = 4.5V

Since the Thevenin equivalent form is equivalent to a non-ideal source, then with load RL it will form a
voltage divider with voltage across the load of
VL

RL
Vth
Rth RL

RL
Vth
Rout RL

(2.3-2)

Equation (2.3-2) also acknowledges that Rth is another name for output resistance Rout.
If an electrical source is applied to a resistance RL, a flow of energy results and heats up the load,
sometimes to incandescence. Electrical energy is of the form of charge stored under voltage, i.e.
w QV

(2.3-3)

where w = energy (work energy equivalent) and Q = quantity of charge (w and Q are the conventional
symbols) and V is the voltage relative to the GND reference.
27

Circuits, Devices, Networks, and Microelectronics

For the flow of charge at potential V, the power = rate of energy transferred is
P

dw
dQ
V
dt
dt

V I

(2.3-4)

A less likely option is one in which charge does not flow but is raised or lowered to a different potential.
But in circuits for which there is current-flow, equation (2.3-4) prevails and is an axiom of circuits.
In the case of a power being dissipated in a resistance, Ohms law states that current is proportional to
voltage and therefore
P V I V

or (alternatively)

V
R

V2
R

(2.3-5a)

I 2R

(2.3-5b)

P V I IR I

Energy dissipated by a resistance results in heat. One of the earliest applications of electrical power was
to heat a (vacuum-shielded) filament to incandescence. Consequently, electric power found immediate
utility as a smoke-free form of illumination, thereby economizing on a lot of fossilized hydrocarbons and
also considerably reducing the risk of residential fires.

EXAMPLE 2.3-2: (a) Determine the Thevenin equivalent voltage

and resistance Rth for the circuit shown.
(b) Determine the voltage VL if the load resistance is
(1) RL = 6.0k
(2) RL = 1.0k
(c) Determine the power dissipated in each of the load resistances
of part (b)
SOLUTION: (a) With the load omitted (i.e. no load) this circuit is a simple voltage divider and may be
resolved by inspection. In which case Vout = Vth = 5.0V.
And with the source VS1 = 0 the upper rail is shorted to GND and Rth = 12k||6k = 4k. (by inspection)
(b) (1) For RL = 6.0k
And for (2) RL = 1.0k

6
5.0
46
1
VL
5.0
4 1
VL

= 3.0V
= 1.0V

28

Circuits, Devices, Networks, and Microelectronics

(c) The power PL that dissipated in RL for each case will then be
2

PL VL
RL

(1) RL = 6.0k :
(2) RL = 1.0k :

3.0 2
= 1.5mW
6k

1 .0 2
1k

= 1.0mW

Example 2.3-2 might raise the question: For what resistance RL is the power transfer to the load =
maximum?
If we undertake a little fun algebra then
2

PL

VL2 RL
1
Vth

RL Rth RL
R
L

which is a maximum when

R LVth2

(2.3-6)

Rth R L 2

PL
0 . So after a little fun calculus we get
R L

R R L 2 R L 2
PL
0 th
V
R L
Rth RL 3 th

Rth R L

Rth RL

Vth2

which shows that the maximum PL occurs when RL = Rth and will be
PL

1 Vth2
1
I nVth
4
4 RL

at RL = Rth

(2.3-7)

Equation 2.3-7 is called the maximum power transfer theorem.

EXAMPLE 2.3-3: (a) What is the maximum power dissipated by the circuit of example 2.3-2?
(b) And compare to pspice

SOLUTION:

PL

1 Vth2
4 RL

1 5.0 2

4 4k

= 1.563mW

In pspice PL vs RL is shown as a plot, with maximum at PL = 1.56mW.

29

Circuits, Devices, Networks, and Microelectronics

Figure E2.3-3(a) PL vs RL with answers to

exercise 2.3-2marked by cursors.

Figure E2.3-3(b) Same as (a) but with decade

resistance scale and peak marked by cursor.

Maximum efficiency is a collateral issue but is not the same as maximum power transfer. Efficiency is
given by

PL PS

(2.3-8)

Where PS is the power developed by the source = IL x Vsource = IL x Vth . So equation (2.3-8) becomes

I L VL VL

I L Vth Vth

RL
Rth R L

(2.3-9)

since the Thevenin (source) resistance Rth and the load resistance RL form a voltage divider. If equation
(2.3-9) is rewritten as

1
1 Rth R L

(2.3-10)

From equation (2.3-10) it should be evident that maximum efficiency (100%) occurs when either R L
or when Rth = 0.
For the case of maximum power transfer (for which RL = Rth ) the efficiency = 50%.

30

Circuits, Devices, Networks, and Microelectronics

2.4 SOURCES and SUPERPOSITION

Courtesy of the network laws series paths relate to a sum of voltages. Parallel paths relate to a sum of
currents. This concept extends to sources. Voltage sources in series add. Current sources in parallel add.
On the other hand, voltage sources may not be deployed in parallel and current sources may not be
deployed in series. If done so it creates a contention. Circuit simulation software will reject circuit
topologies that have these faults and make snide error messages about the intelligence of the perpetrator.
But by Thevenins theorem and Nortons theorem the outcome of any single source within the circuit will
result in a Thevenin equivalent Vth1 or an equivalent Norton In1. So if there is more than one source then
their individual contributions will add, i.e.

or

Vth = Vth1 + Vth2

(2.4-1a)

In = In1 + In2

(2.4-1b)

This is called the principle of superposition. Superposition is a property of linear systems. And it aso
serves as a simplification technique for circuit topologies with more than one embedded source or I/O
networks with more than one input.
Consider example E2.3-1 (redeployed as Example E2.4-1.)

EXAMPLE 2.4-1: Determine the Thevenin equivalents Rth

and Vth for the circuit of Example 2.2-1 if the output is at VC.
SOLUTION: Let all sources be zero. The outcome is then
the same as example 2.3-1 for which VS becomes a short circuit
and IS becomes an open circuit. Then by inspection

Rth = 1.4k||1.0k = 0.58 k

Figure E2.4-1(a) Example network
(1) Let VS = 0 and I S 0 . The network will then collapse
to that of figure E2.4-1(b)
The resistance inside the dashed line box is the same as Rth,
which means that figure E2.4-1(b) simplifies to a Norton
equivalent. Then the Thevenin voltage for figure E2.4-1(b)
will be
Figure E2.4-1(b) Example network collapsed by VS = 0.
31

Circuits, Devices, Networks, and Microelectronics

Vth1 = IS Rn

= 2mA 0.58k

= 1.167V

(2) Let IS = 0 and VS 0 . Then the network will collapse to the form shown by figure E2.4-1(c)
The resistance inside the dashed line box is
= 0.5k + 2k||2k = 1.5k
And therefore V B

0.5k
1
5.0 5.0
1.5k
3

And so Vth 2 V B

1
5.0 V B
2

2
5.0
3

= 3.333 V

Figure E2.4-1(b) Example network collapsed by IS = 0.

And then

Vth = Vth1 + Vth2

= 1.167 + 3.333

= 4.5V

= same as that determined by nodal analysis via Example 2.2-1.

Other techniques associated with network simplification are source equivalence and source absorption
using the Thevenin-Norton equivalence given by figure 2.3-3.
Consider the following example

EXAMPLE 2.4-2: Find:

(a) the Thevenin equivalent for the circuit shown using
the source transformation and absorption technique
and (b) determine the maximum power transfer
possible for this circuit.
Figure E2.4-2(a) Example multi-source circuit

32

Circuits, Devices, Networks, and Microelectronics

SOLUTION: (part (a) is represented by Figure E2.4-2(b)

Figure E2.4-2(a) Example multi-source circuit and transformation/absorption steps

(1) The source to the left can be converted to Norton form with In1 = 20/4k = 5.0mA (= transformation)
(2) The two current sources are in parallel and add to give In2 = In1+ (-2mA) = 3.0mA (= absorption)
1/ 8
= 1.0mA
(3) The short-circuit current at the output In
3.0mA
1/ 8 1/ 4
(4) The Norton-Thevenin conversion give Vth = 1.0mA 3k = 3.0V
And the Rth = 4k||(8k + 4k)

= 3.0k

for all sources = 0.

(b) The maximum power to RL is when RL = 3.0k and is PL = 0.25 (1.0mA 3.0 V) = 0.75mW

As a snake check this example can also be done by superposition (Example 2.4-3).

EXAMPLE 2.4-3: Find the Thevenin equivalent for

the circuit shown using superposition.

Figure E2.4-3(a) Example circuit

33

Circuits, Devices, Networks, and Microelectronics

SOLUTION:
Current source = 0, equivalent to open circuit.
4k
Vth1
= 5.0V
20
12k 4k

Voltage source = 0, equivalent to short circuit.

Current through 4k resistance at output under no
load at output
1 / 12
I2 =
= -0.5mA
2mA
1 / 4 1 / 12
= -2V
Vth2 = -0.5mA 4k
So Vth = Vth1 + Vth2
And for all sources = 0.

= 5.0 + -2.0

= 3.0V
= 3.0k

Rth = 4k||(8k + 4k)

The reality of having at least two methods to achieve the same result is always a benefit. It gives a snake
check. If different results occur then both (or all) assessments should be revisited
The two examples also show the two instances in which the current source may be parsed. For example
2.4-2, the Norton equivalent analysis is accomplished with the output shorted (as is required for Norton
analysis). But for example 2.4-3, the contribution of the current source was assessed with the output
open consistent with the Thevenin equivalent analysis that Vth is the open-circuit output voltage.

2.5 TWO-PORT NETWORKS AND DEPENDENT SOURCES

The Thevenin and Norton theorems are oriented toward a black-box equivalent circuit. These theorems
focus on the output as a source equivalent, either of the form of its open-circuit voltage (= Vth) or its shortcircuit current (= In). The source equivalents are otherwise limited by resistance Rth = Rn and obedient to
the relationship Rth = Vth / In.
However, the more general form of a black-box equivalent is one in which the box has both an input
port and an output port, i.e. a two-port network form. Inputs are usually of the form of signals that may
be (1) from a sensor or (2) from a previous black-box stage.
If the black box has more than two inputs or more than two outputs it can be constructed from individual
two-port networks in parallel, as indicated by figure 2.5-1.

34

Circuits, Devices, Networks, and Microelectronics

Figure 2.5-1. Multiport options with two-port networks

The transfer relationship between input and output indicates the need to identify a transfer component,
a.k.a dependent source. A dependent source is like a Thevenin/Norton Vth or In source except that it is
controlled from elsewhere in the circuit and is zero if the electrical measure at the controlling site is zero.
The control site is like an input port, and probably is. Dependent sources therefore are defined as twoport type components and fall into the category of one of four types, as represented by figure 2.5-2.

Figure 2.5-2. Dependent sources

Where:
VCT = voltage-to-current transducer
VVT = voltage-to-voltage transducer
CCT = current-to-current transducer
CVT = current-to-voltage transducer
Figure 2.5-1 represents ideal transfer components with output properties that are not unlike the
independent sources referenced so far. But if framed in terms of input and output ports they are expected
to be consistent with the Thevenin and Norton theorems. The output port assume a Thevenin/Norton
source designated either as vout or as iout with output resistance Rout = Rth = Rn. And as defined by the
theorems the output resistance is defined under the criterion that that all sources = 0 to include the input.

35

Circuits, Devices, Networks, and Microelectronics

It is also not unreasonable to assume that the two-port I/O network is unidirectional, or nearly so, in
which case the input port will have no source component hiding within and its only feature will be input
resistance Rin. The transfer characteristics of the two-port I/O forms are shown by figure 2.5-3 and
reflect the context of an Rin and an Rout.
It is also not uncommon for the generic types of two-port networks to have a common ground for both
input and output.

Figure 2.5-3. Generic two-port network forms

Figures 2.5-2 and 2.5-3 and the text description also emphasizes that the two-port form is oriented more
toward signals (represented by lower case) rather than level values (as represented by upper case). In this
respect the two-port network is a signal transfer element.
The essential two port network characteristics are the port resistances Rin, and Rout and a (signal) transfer
function. The most common transfer function is vout/vin in which case the two-port network may be
designated as a voltage amplifier (regardless of whether it amplifiers or not). The next most common
transfer function is iout/iin , i.e. current amplifier. All transfer function types are related to one another
through input and output resistances Rin and Rout .
For example the current transfer function and the voltage transfer function are related by
iout vout Rout

vin Rin
iin

Rin vout

Rout vin

(2.5-1)

The four types of dependent sources given by figure 2.5-2 also reflect the four different types of transfer
functions even though the internal transducer may not be the same type as its two-port outcome.
Input resistance and output resistance are not alike, however. They only look like it. They will both form
voltage dividers with the external resistances. And they will both serve to translate one type source to
another ype source. But input resistance Rin is a load resistance whereas output resistance is a source
resistance.

Rout fulfills the role that it does since it is defined in terms of open circuit voltage vout and short-circuit
current iout , respectively, as represented by equation (2.5-2).
36

Circuits, Devices, Networks, and Microelectronics

Rout

vo
v(output _ open)

i SS i(output _ short circuit)

(2.5-2)

Equation (2.5-2) is nothing other than a repeat of the Thevenin/Norton theorems with an emphasis on the
character of Rout.
Load, source, and I/O (input and output) resistance will each form voltage dividers with their source/load
associates. By way of example, when a load RL is connected to the output port, then Rout and RL form a
voltage divider which has (signal) transfer ratio
vL
RL

vout Rout RL

(2.5-3)

The role of the I/O resistances and signal transfer ratios are of considerable significance in the transfer
relationship of power from input to load as represented by figure 2.5-4 and equation (2.5-4)

Figure 2.5-4. Power transfer from source to load via two-port interface
The power transfer ratio is
pL vL iL

pin vin iin

vL iL

vin iin

v L v L RL

vin vin Rin

v
L
vin

where

v
vL
RL

out
vin Rout RL vin

37

Rin

RL

(2.5-4)

Circuits, Devices, Networks, and Microelectronics

EXAMPLE 2.5-1: An echolocation sensor is

characterized by signal voltage vS = 10mV or
short-circuit current iSS = 0.05A. If connected to
an amplifier stage with input resistance Rin =
50k, output resistance Rout = 2.5k, and opencircuit signal gain AV = 25V/V, and drives a load
of resistance RL = 10k determine
Figure E2.5-1. Example transfer circuit
(a) transfer ratio vL/vS
(b) signal voltage vL at the output and
(c) power gain pL/pin.
SOLUTION:
Then

Source resistance of the echolocation sensor is

vin
Rin
50

vS
RS Rin 200 50

Using signal ratios

vo
RL

vin Rout RL

25

= 0.2 20

Then (a) vL/vS

And (b) vL

= 20 2.0mV

vS
10mV

i SS 0.05A

(for which vin = 0.2 10.0

= 0.2V/V

v L vo v L

vin vin vo

RS

10
2.5 10

= 200k
= 2.0mV)

= 20 V/V

= 4.0 V/V
( also vL = 10 4.0

= 40mV

= 40mV )

p L v L Rin
50
2
20

pin vin RL
2.5

Therefore (c) the power gain is

= 8000W/W

Notice that the two-port network also serves as a means to match compatible resistances. Had the 2-port
network been omitted and the signal vS attenuated by the voltage divider made by RS and RL then

vL2

10
10mV .05mV
200 10

with insertion gain

vL1/vL2 =

40mV/.05mV

= 800 V/V

The two-port network emphasizes the relationships of transfer ratios and the fact that the network may
have capability to add or lose power as the signal is transferred across the network. Sometimes it is many
orders of magnitude, as represented by example 2.5-1. And that suggests that the signal gain or loss

38

Circuits, Devices, Networks, and Microelectronics

should also be identified in order-of-magnitude measure. For electrical circuits and the history associated
with applications thereto this measure is in terms of a unit called the dB (decibel).
dB measure is a identified first and foremost in terms of the impact on the application, which is invariably
the (energy-effect) relationship of power. So one order-of-magnitude is 10dB, two orders-of- magnitude
is 20 dB, three orders of magnitude is 30 dB, etc. So order-of-magnitude measure in dB for a two-port
network is
(dB measure) = 10 log (pout/pin)

(2.5-5)

So for Example 2.5-1 the power gain would be

(db measure)

= 10 log (8000)

= 39dB

Since factors of two are after important we often single them out in dB measure, i.e
10 log (2)

= 3dB

(2.5-6)

To add to the context of dB measure, take note that the power ratio has a quadratic relationship with the
amplitude ratio as represented by equation (2.5-4). So a factor-of-10 increase in the amplitude ratio
corresponds to a factor increase in the power ratio of 100, or 20 dB.
So dB measure of amplitude ratios then has an effect of
(db measure)

= 20 log (vout/vin)

(2.5-7)

Often the amplitude ratios are identified on an equal footing as the power ratio factors, and so equation
(2.5-7) may prevail. And if so then
3dB (for amplitude)

= 20 log 2

(2.5-8)

Amplitude factors of 2 are denoted as a 3dB reference level.

39

Circuits, Devices, Networks, and Microelectronics

PORTFOLIO and SUMMARY

I GV
V IR

Ohms law

A
L

L
A

1
1
1
1

Req R1 R 2 R3

Req R1 R2 R3

Voltage divider:
VB

R2
VA
R1 R 2

R-2R ladder

Nodal analysis

Loop (mesh) analysis

node

V
loop

Sources:

40

Voltage source

Rout = 0

Current source

Rout =

Circuits, Devices, Networks, and Microelectronics

P IV

Power dissipated in a resistance

V2
R

I 2R

Thevenin and Norton theorems:

I n Vth Rth
Vth I n R n
R n Rth

Maximum power transfer:

PL

1 Vth2
4 RL

1
I nVth
4

Maximum efficiency for

at RL = Rth

Rth << RL

Figure A-1.1 pspice rendition, maximum power transfer

Superposition (Thevenin/Norton outputs):

Vth = Vth1 + Vth2

In = In1 + In2

Dependent sources options

Two-port transfer circuits

Rin

vin
iin

Rout

vout
iout

RL

vL
iL

v
vL
RL
RL

out

vin Rout RL vin

Rout RL
pL
v i
L L
pin vin iin

dB measure:

(dB measure) = 10 log (pout/pin)

3dB (for amplitude)

= 20 log 2

corresponding to

41

vL iL

vin iin

= 20 log (vout/vin)
3dB (for power) = 10 log (2)

v R
L in
vin RL

Circuits, Devices, Networks, and Microelectronics

42