E2.

2 Analogue Electronics

• Instructor : Christos Papavassiliou

• Office, email : EE 915, [email protected]
• Lectures : Monday 2pm, room 408 (weeks 2-11)
Thursday 3pm, room 509 (weeks 4-11)
• Problem, Quizzes: Thursday 4pm, room 408 (weeks 3-4, 7-11)

• Office hours: Tuesday 11am

Thursday 10 am
• Office hours start week 3 (week of 1910/09)
• Course website: on blackboard
(or on my home page)

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 1

 What analogue electronics is
• Engineering, i.e. the analysis (study and reverse engineering) and
synthesis (“design”) of circuits:
– Amplifiers, Filters, Oscillators
– Radio
– Multipliers (Modulators – Demodulators)
– Analogue signal processing (e.g. rectifiers, logarithmic amplifiers)
– Fast Digital gates (!)
• Applications:
Communications, Signal Processing, Control, Instrumentation, …
• Areas of human activity:
Industrial, Consumer, Biomedical, …

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 2

 analogue electronics is not only
• CMOS integrated circuits
• Transistor circuits
• Op-amp circuits
• Audio electronics
• Mobile phone circuits
• Radars
• Printed circuit board design
• Integrated circuit design
• Television set repair

In fact, it is all of the above, and much more. Restricting our point of
view to one of these will make understanding the topic more difficult.

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 3

 Course Aims
Learn to analyse electronic circuits
•Analysis is prerequisite to:
– DESIGN (including Integrated Circuit design)
– APPLICATION (including repair)

•Analysis is performed through modelling:

– ABSTRACTION: replace groups of components with one symbol
– SIMPLIFICATION but NOT oversimplification
– MATHEMATICS (is a language, not a torture device!)

•The inverse of ANALYSIS is SYNTHESIS (i.e. DESIGN!)

•The course is closely connected to two other 2nd year courses:

– Control Theory
– Signal processing
because circuits are used to implement control and signal processing
AND because without their methods circuit analysis is not possible (except in
trivial cases)

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 4

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 5
L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 6
 Prerequisites
• Analysis of circuits:
– Kirchhoff's laws,Nodal Analysis,Thevenin and Norton Theorems,
Phasors, Frequency and step Response of RC circuits, DC op-
amp circuits: Inverting and Non-Inverting Amplifiers.

• Analogue electronics:
– Static I-V characteristics of BJT and FET transistors
– Small signal models of BJT and FET transistors
– Single Stage Transistor amplifiers with BJT, FET

• Maths:
– Differentiation, Integration
– Second order Ordinary differential Equations
– Complex numbers
– Fourier transforms

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 7

 References

• These slides

• Main Course text:

– Sergio Franco: Design with Operational Amplifiers and Analog
Integrated Circuits, McGraw Hill

• Any book in analogue electronics you like. There are many! e.g.
– Sedra and Smith, Microelectronic Circuits, Oxford U. Press

• The best general reference in electronics (but not the best to learn from):
– Horowitz and Hill, The Art of Electronics, Cambridge U. Press.

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 8

 Workload

Direct workload:
• 17 lectures
• 8 problem sessions starting on week 3
• 1 problem sheet/week

Assessment:
• final exam : 2 hours long (100% of the marks)

Indirectly related lab work

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 9

 The “tools of the trade”

• Kirchhoff's Current Law (KCL) Æ “nodal analysis”

• Kirchhoff's Voltage Law (KVL) Æ “mesh analysis” (difficult!)
• Ohm’s Law Æ Idealised Resistors and conductors
• Phasor analysis (i.e. Fourier transform): Capacitors, Inductors
• Modeling:
– large signal analysis
– small signal analysis
• Approximations:
– Thevenin and Norton “Theorems”
• Bode Plots
• All of the above, and more, is implemented in simulators: eg. SPICE
• Mathematics!

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 10

 Circuit Components

• Independent Sources
– Voltage, Current
• Dependent Sources
– Voltage Controlled (Voltage, Current) Source: VCVS and VCCS
– Current Controlled (Voltage, Current) Source: CCVS and CCCS
• Resistors
• Capacitors, Inductors, transformers
• Diodes,
• Transistors
– Bipolar Junction (BJT): NPN, PNP
– Field Effect (FET): n-channel, p-channel, JFET, MOSFET…
• Operational amplifiers

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 11

 Computer simulation: SPICE
• Written at the University of California – Berkeley
• User inputs netlist, ie a list of components connecting nodes
– The netlist is a way of writing the admittance matrix !
– Schematic capture is a GUI producing netlists
• User specifies one or more types of analysis to be performed:
– DC bias
– AC small signal
– Transient
– Component sensitivity
• Commercial versions: very expensive, confusingly complicated!
• Free packages (with schematic, on the web; grab one!):
– 5Spice: http://www.5spice.com (schematic capture, includes WinSpice)
– SIMetrix: www.catena.uk.com/ (complete package)
– LT Spice : http://www.linear.com/designtools/software/switchercad.jsp

History of SPICE: http://www.ecircuitcenter.com/SpiceTopics/History.htm

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 12

 Resistors, Conductors, Capacitors, Inductors
• Ohm’s law can be written in two ways: V = IR , I = GV
• Clearly R=1/G
• This is useful in simplifying circuits:
– Resistors in series: Rtotal = R1 + R2
R1 R2
– Resistors in parallel: Gtotal = G1 + G2 NOT: Rtotal =
• Some standard terminology:
R1 + R2
– Impedance Z: Z = R + jX (R: Resistance, X: Reactance)
– Admittance Y: Y = G + jB (G: Conductance, B: Susceptance)
– Immitance: Impedance and Admittace considered together
• Note that:
– Capacitor: BC = ωC; The capacitor is naturally a susceptor
– Inductor: X L = ω L; The inductor is naturally a reactor
• Nodal analysis is easier if we treat components as admittances
ÆNodal equations and the admittance matrix
L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 13
 The Thevenin Theorem
Any part of a circuit with 2 external terminals can be replaced by a Thevenin
equivalent circuit.

• Entire networks with 2 terminals can considered to be a single component

• The Thevenin voltage source may be a:

• Fixed voltage source

• Voltage controlled voltage source
• Current controlled voltage source

• Note that VL =VT ZL / (ZT + ZL )

• In real life, ZT is never zero !

•The Thevenin theorem is a Taylor expansion of V(I) about the operating

point V0 , I0 (if the expansion exists!)
L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 14
 The Norton Theorem
Any part of a circuit with 2 external terminals can be replaced by a Norton
equivalent circuit.

• Entire networks with 2 terminals can considered to be a single component

• The Norton current source may be a:

• Fixed current source

• Voltage controlled current source
• Current controlled current source

• Note that IL =IN YN / (YN + YL )

• In real life, YN is never zero !

•The Norton theorem is a Taylor expansion of I(V) about the operating point
V0 , I0 (if the expansion exists!)
L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 15
 Thevenin and Norton special cases
A Thevenin equivalent circuit with:
• zero Thevenin impedance is an ideal voltage source
• infinite Thevenin impedance is an ideal current source

Î If the magnitude of the Thevenin impedance is larger than the

magnitude of the load impedance connected to it we usually prefer to
model the component using a Norton equivalent circuit

•A Norton equivalent circuit with:

• zero Norton admittance is an ideal current source
• infinite Norton admittance is an ideal voltage source

Î If the magnitude of the Norton admittance is larger than the

magnitude of the load admittance connected to it we usually prefer to
model the component using a Thevenin equivalent circuit

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 16

 The Thevenin equivalent of a voltage divider
• Think of the series element as an impedance Z(s)
• Think of the shunt element as an admittance T(s)
• Both Z and Y are arbitrary functions of frequency s=jw
• Thevenin voltage is the value of the voltage divider:

VIN VT ZT
VIN
VT =
Z(s)
Y(s) 1+ Z ( s)Y ( s)
• Thevenin impedance is the parallel combination of Z an Y:

Voc ⎛ V ⎞ V Z (s)
ZT = =⎜ ⎟ =
I sc ⎝ 1 + Z ( s ) Y ( s ) ⎠ Z ( s ) 1+ Z ( s )Y ( s )
• These forms are easy to remember!

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 17

 Ladder networks

First Cauer form Second Cauer form

L1 L2 L3 Ln
C1 C2 C3 Cn
L2
C1 C2 Cn L1 Ln

(a) (b)

When they are made of L and C they are called “canonical filters” or
“Cauer forms”. They are the most efficient filter implementations.

•Start from left

• do a sequence of Thevenin and Norton transformations
• NEVER do nodal analysis
• if components are RLC both the Thevenin voltage and Thevenin
impedance are ratios of polynomials in frequency

L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 18

 Small Signal and Large Signal models

I I
Large Signal
VOC=VT Small Signal: VOC=VT

V ISC=IN V
GN=1/RT
ISC=IN
GN=1/RT

Model is secant on the IV curve Model is tangent on the IV curve

The following relations are always valid:
RT = 1/ GN
VT = I N RT
• Most circuits (except of ideal sources) have both Thevenin and Norton equivalents
• Unless we say otherwise, we are only concerned with small signal equiv. circuits.
L1 Autumn 2009 E2.2 Analogue Electronics Imperial College London – EEE 19