ISBN-3-934584-49-7 The Bosch Yellow Jackets Edition 2001 Technical Instruction Automotive Electrical and Electronic SystemsOrder Number 1 987 722 122 AA/PDI-03.02-En

2002

AutomotiveMicroelectronics

Tech

nica

l Ins

truc

tion

Mic

roel

ectr

onic

s in

Mot

or V

ehic

les

The

Bos

ch Y

ello

w J

acke

ts

Basic principles Components Development

and production

Automotive Technology

The Program Order Number ISBN

Automotive electrics/Automotive electronicsBatteries 1 987 722 153 3-934584-21-7Alternators 1 987 722 156 3-934584-22-5Starting Systems 1 987 722 170 3-934584-23-3Lighting Technology 1 987 722 176 3-934584-24-1Electrical Symbols and Circuit Diagrams 1 987 722 169 3-934584-20-9Automotive Sensors 1 987 722 131 3-934584-50-0Automotive Microelectronics 1 987 722 122 3-934584-49-7

Diesel-Engine ManagementDiesel Fuel-Injection: an Overview 1 987 722 104 3-934584-35-7Electronic Diesel Control EDC 1 987 722 135 3-934584-47-0Diesel Accumulator Fuel-Injection System Common Rail CR 1 987 722 175 3-934584-40-3Diesel Fuel-Injection Systems Unit Injector System/Unit Pump System 1 987 722 179 3-934584-41-1Radial-Piston Distributor Fuel-Injection Pumps Type VR 1 987 722 174 3-934584-39-XDiesel Distributor-Type Fuel-Injection Pumps VE 1 987 722 164 3-934584-38-1Diesel In-Line Fuel-Injection Pumps PE 1 987 722 162 3-934584-36-5Governors for Diesel In-Line Fuel-Injection Pumps 1 987 722 163 3-934584-37-3

Gasoline-Engine ManagementEmission Control (for Gasoline Engines) 1 987 722 102 3-934584-26-8Gasoline Fuel-Injection System K-Jetronic 1 987 722 159 3-934584-27-6Gasoline Fuel-Injection System KE-Jetronic 1 987 722 101 3-934584-28-4Gasoline Fuel-Injection System L-Jetronic 1 987 722 160 3-934584-29-2Gasoline Fuel-Injection System Mono-Jetronic 1 987 722 105 3-934584-30-6Spark Plugs 1 987 722 155 3-934584-32-2Ignition 1 987 722 154 3-934584-31-4M-Motronic Engine Management 1 987 722 161 3-934584-33-0ME-Motronic Engine Management 1 987 722 178 3-934584-34-9Gasoline-Engine Management: Basics and Components 1 987 722 136 3-934584-48-9

Driving and Road-Safety SystemsConventional Braking Systems 1 987 722 157 3-934584-42-XBrake Systems for Passenger Cars 1 987 722 103 3-934584-43-8ESP Electronic Stability Program 1 987 722 177 3-934584-44-6Compressed-Air Systems for Commercial Vehicles (1): Systems and Schematic Diagrams 1 987 722 165 3-934584-45-4Compressed-Air Systems for Commercial Vehicles (2): Equipment 1 987 722 166 3-934584-46-2Safety, Comfort and Convenience Systems 1 987 722 150 3-934584-25-XAudio, navigation, and telematics in the vehicle 1 987 722 132 3-934584-53-5

The up-to-date program is available in Internet under:www.bosch.de/aa/en/fachliteratur/index.htm

Published by: Robert Bosch GmbH, 2002Postfach 30 02 20,D-70442 Stuttgart.Automotive Aftermarket Business Sector,Department AA/PDT5.Product Marketing,Diagnostics, Test Equipment.

Editor-in-chief:Dipl.-Ing. (FH) Horst Bauer.

Editorial team:Dipl.-Ing. (BA) Jrgen Crepin,Dipl.-Ing. Karl-Heinz Dietsche,Dipl.-Holzw. Folkhart Dinkler.

Authors:Dipl.-Ing. Adolf Fritz (Control Unit Development),Dr. rer. nat. Ulrich Goebel (Hybrid Circuit Board Production),Armin Hess (Example Circuit),Karlheinz Isert (Printed Circuit Board Production),Dr. rer. nat. Ulrich Schaefer (Basic Principles, Components, Microcontroller)

and the editorial team in co-operation with the responsible technical departments of Robert Bosch GmbH.

Unless otherwise stated, the above are all em-ployees of Robert Bosch GmbH, Stuttgart.

Reproduction, duplication, and translation of thispublication, including excerpts therefrom, is onlyto ensue with our previous written consent andwith particulars of source.Illustrations, descriptions, schematic diagrams,and other data only serve for explanatory pur-poses, and for presentation of the text. Theycannot be used as the basis for design, installa-tion, and scope of delivery. We undertake no lia-bility for conformity of the contents with nationalor local regulations.Robert Bosch GmbH is exempt from liability,and reserves the right to make changes at anytime.

Printed in Germany.Imprim en Allemagne.

1st Edition, August 2001.English translation of the German edition dated:February 2001.(1.0)

Imprint

Robert Bosch GmbH

Automotive microelectronics

Robert Bosch GmbH

Robert Bosch GmbH

4 Automotive microelectronics4 Overview6 Demands on electronic systems6 History of development

8 Basic principles ofsemiconductor technology

8 Terminology8 Electrical conductivity

12 Electronic components12 Passive components13 Semiconductor components14 Diodes16 Semiconductor resistors17 Transistors20 Monolithic integrated circuits24 Microcomputers27 Semiconductor memories30 Opto-electronic components32 Micromechanical sensors

34 Microcontrollers34 Microcontroller developments36 Microcontroller components37 Microcontroller design and

operating concept41 Microcontroller memories42 Microcontroller peripheral modules44 Microcontroller busses

46 Circuit example 46 Overview48 Design and construction55 Operating concept58 Programming

60 ECU development60 Overview64 Hardware development

68 Manufacture of semiconductorcomponents and circuits

68 Semiconductor components72 Micromechanics73 Conventional printed-circuit

boards78 Film and hybrid circuits

80 Glossary and Tables forautomotive microelectronics

88 Index of technical terms91 Abbreviations

Contents

Robert Bosch GmbH

Motor vehicles are now inconceivable without electronic control units and their associ-

ated sensors and actuators. Such components have revolutionised the automotive

world. All essential vehicle functions are electronically controlled by systems and com-

ponents that occupy only a tiny space. There are systems that control the function of

engine and gearbox, the safety and security systems and a large number of comfort and

convenience systems, and the number of intelligent vehicle systems is growing at a

rapid pace. Nor is there any foreseeable end to this trend.

Electronic control systems open the door to a vast array of possibilities. They im-

prove vehicle safety and ride comfort. At the same time, they make vehicles more

economical and kinder to the environment.

The triumphant advance of electronics has created a vocabulary of terms with

which we are bombarded every day. It is becoming more and more difficult to under-

stand what precisely those terms refer to.

This publication in the Bosch Yellow Jacket Technical Instruction series explains the

specialist terminology of microelectronics in detail.

It starts with a short introduction followed by an explanation of the principles of

semiconductor technology and how microelectronic components interact. A practical

example of a circuit helps to consolidate the theoretical concepts described. This is

then followed by a description of the development and production of control units.

Finally, the glossary provides a quick guide to the most important microelectronics

terms.

The functions of the individual electronic systems (e.g. MED, EDC, ESP) are described

in detail in separate publications in the Yellow Jacket Technical Instruction series.

Robert Bosch GmbH

Microelectronics have revolutionised motorvehicle technology. Initially, mechanicalcomponents were replaced by electronicdevices in order to make systems morereliable. This was the case with the contactbreaker points in the conventional coil igni-tion system, for example. Gradually, how-ever, more and more new vehicle systemswere introduced which simply would nothave been possible without the use of elec-tronics. The impetus for these new develop-ments was provided by increasingly de-manding requirements placed on theexhaust-emission characteristics of theinternal-combustion engine (e.g. emission-control systems), on comfort and con-venience (e.g. climate control and naviga-tion systems) and on safety (e.g. antilockbraking system [ABS] and airbags).

Overview

Cars and commercial vehicles equipped with the latest available technical improvementsare absolutely brimming with electronic sys-tems. Those systems can be subdivided intothe following areas of application: Engine and drivetrain Safety Comfort and convenience Communication and multimedia.

Fig. 1 provides an overview of the electronicsystems that can be found on modern motorvehicles. Many of the systems referred to arenow standard equipment on all new vehi-cles. By contrast, there are others that re-main (as yet) the preserve of the most ex-pensive luxury models.

Electronic systems can be subdivided intothe following functional areas: Sensors and setpoint generators Control units (ECUs) Actuators ECU communication links (networks)

and Electronic diagnosis

Sensors and setpoint generators Sensors detect operating conditions (e.g. en-gine speed, wheel speed, temperature). Theyconvert physical variables into electrical sig-nals. Setpoint generators (e.g. controls oper-ated by the driver) specify desired settings.

Control units (ECUs)Control units process the information re-ceived from the sensors and setpoint gener-ators using specific mathematical calculationsequences (control algorithms). They con-trol the actuators by means of electrical out-put signals. Control units also form the in-terface with other systems and the vehiclediagnostics.

ActuatorsActuators convert the electrical output sig-nals from the control unit into physicalvariables. Examples of actuators are: gasoline-engine fuel injectors diesel-engine fuel injectors electric motors (e.g. for driving the

power-window regulator mechanism or asthrottle-valve positioner on vehicles withETC (Electronic Throttle Control))

fans

NetworksAs the number of electronic systems in vehi-cles grows, so does the amount of intercon-necting wiring needed. The total length ofthe wiring in the wiring harness of a mid-range car is now roughly 1.6 km on averageand incorporates up to 300 connectors witha total of around 2000 connector pins.

Networking of the various systems reducesthe overall length of the wiring required. Ashared data bus consisting of only two wires(e.g. CAN bus) carries data that is read by all bus users. Depending on the specific re-quirements, a vehicle may also have separatebusses for engine and drivetrain, comfortand convenience systems, and communica-tion systems.

4 Automotive microelectronics Overview

Automotive microelectronics

Robert Bosch GmbH

Motor vehicles are now inconceivable without electronic control units and their associ-

ated sensors and actuators. Such components have revolutionised the automotive

world. All essential vehicle functions are electronically controlled by systems and com-

ponents that occupy only a tiny space. There are systems that control the function of

engine and gearbox, the safety and security systems and a large number of comfort and

convenience systems, and the number of intelligent vehicle systems is growing at a

rapid pace. Nor is there any foreseeable end to this trend.

Electronic control systems open the door to a vast array of possibilities. They im-

prove vehicle safety and ride comfort. At the same time, they make vehicles more

economical and kinder to the environment.

The triumphant advance of electronics has created a vocabulary of terms with

which we are bombarded every day. It is becoming more and more difficult to under-

stand what precisely those terms refer to.

This publication in the Bosch Yellow Jacket Technical Instruction series explains the

specialist terminology of microelectronics in detail.

It starts with a short introduction followed by an explanation of the principles of

semiconductor technology and how microelectronic components interact. A practical

example of a circuit helps to consolidate the theoretical concepts described. This is

then followed by a description of the development and production of control units.

Finally, the glossary provides a quick guide to the most important microelectronics

terms.

The functions of the individual electronic systems (e.g. MED, EDC, ESP) are described

in detail in separate publications in the Yellow Jacket Technical Instruction series.

Robert Bosch GmbH

Another advantage of the bus system is thatsensor signals need only be analysed by asingle control unit. For example, the instru-ment cluster can use the individual wheel-speed signals delivered by the ABS controlunit to calculate the vehicles road speed.This complex calculation even takes accountof cornering differences and spinningwheels. The road-speed signal is transferredon the CAN bus to all other users (e.g. theABS control unit which uses it for control-ling brake application, the engine-manage-ment ECU which uses it as the basis forcruise control, or the car radio for the speed-related volume adjustment).

Electronic diagnosisElectronic diagnosis functions on the con-trol unit continuously monitor the opera-tion of the system and its components. Anyfaults that occur (e.g. short circuits in thewiring, sensor failure) are stored in the con-trol units fault memory. Those stored faultscan then be read out in the course of serviceusing a dedicated system tester that is con-nected to the control units diagnosis inter-face. The system tester can also be used toscan sensor signals and send commands toactuators to perform specific operations.Electronic diagnosis thus enables faults to belocated more quickly and easily in the courseof vehicle service.

Automotive microelectronics Overview 5

Elektronik

im Kfz

Drive

train

Comm

unica

tion

Electr

onic

voice

outpu

t

Voice

contr

ol of

functi

ons

(spee

ch re

cogn

ition)

Audio

equip

ment

(radio

, CD,

etc.)

Video

On-bo

ard co

mpute

r

Car p

hone

Navig

ation

New

displa

y tec

hnolo

gies

(head

-up di

splay

)

Intern

et an

d PC

Bus s

ystem

(e.g.

CAN

)

Syste

m dia

gnos

is

Digit

al en

gine m

anag

emen

t:

gaso

line e

ngine

: Motr

onic

Dies

el en

gine:

Electr

onic

diese

l con

trol (E

DC)

with

electr

onica

lly co

ntroll

ed

fuel in

jectio

n,ele

ctron

ic ign

ition

(gaso

line e

ngine

),

Lamb

da ex

haus

t-emi

ssion

contr

ol bo

ost-p

ressu

re

contr

ol, et

c. Ele

ctron

ic tra

nsmi

ssion

-

shift

contr

ol On

-board

diag

nosis

(OBD

)

Contr

oller

Area

Netw

ork

(CAN

)Sy

stem

diagn

osis

Cruis

e con

trol

Adap

tive c

ruise

contr

ol

(rada

r sen

sor)

Heati

ng an

d air-c

ondit

ioning

Seat

adjus

tmen

t with

positi

on m

emory

Centr

al loc

king

Adap

tive s

uspe

nsion

Reve

rsing

assis

tance

Contr

oller

Area

Netw

ork (C

AN)

Syste

m dia

gnos

is

Antilo

ck br

aking

syste

m (A

BS)

Tracti

on co

ntrol

syste

m (T

CS)

Electr

onic

stabili

ty pro

gram

(ESP

)

Head

lamp a

djustm

ent

and c

leanin

g Lit

ronic

Wash

-wipe

contr

ol

Indivi

duali

sed s

ervice

interv

al dis

play

Monit

oring

syste

ms fo

r

cons

umab

les an

d

weari

ng pa

rts

Trigg

ering

syste

ms fo

r

airba

gs, s

eatbe

lt ten

sione

rs

and

roll-o

ver b

ar

Vehic

le se

curity

syste

ms

Tire-p

ressu

re mo

nitori

ng

Contr

oller

Area

Netw

ork (C

AN)

Syste

m dia

gnos

is

bar

Safet

y

Comf

ort an

d con

venie

nce

Electronics in Motor Vehicles (Cars)1

UA

E08

56E

Robert Bosch GmbH

Demands on electronicsystemsElectronic systems in motor vehicles areexposed to extreme stresses (e.g. due toextreme temperature variations, unusualclimatic conditions, poor road surfaces andthe effect of corrosive substances). These aresome of the requirements they must meet inorder to be able to function reliably andwithout faults over long periods: Resistance to temperatures ranging from

40 C ... 125 C EMC (electromagnetic compatibility):

immunity to external interference (e.g.mobile phone signals) and no emission ofelectromagnetic radiation likely to causeinterference on other equipment

Resistance to shocks and vibration Resistance to water and damp Resistance to corrosive fluids (e.g. oils and

salt-water spray) Light weight Economical production costs and Secure and trouble-free mounting

History of development

The amount of electronic equipment inmotor vehicles is continually increasing.Fig. 2 provides an overview of the growth of electronic equipment expressed as a pro-portion of vehicle cost.

Because of their cost, electronic systemswere initially reserved for vehicles at the lux-ury end of the market. This explains why in1980 electronic equipment accounted foronly half a percent of vehicle cost. From thattime onwards, and particularly in the 1990s,that proportion grew rapidly as the price ofelectronic equipment continually droppedwith the result that more and more systemscould be fitted to mid-range and even smallcars.

Gradually, more and more electronic sys-tems were fitted to motor vehicles (Table 1).And the trend continues.

The new science of mechatronics dealswith the interaction between mechanical,electronic and data processing devices.

6 Automotive microelectronics Demands on electronic systems/History of development

Table 1

1980 1990

2000 2010Prognosis

0.5 % 7 %

17 % 24 %

Proportion of cost per vehicle (overall average)

2

U

AE

0855

E

1958 DC generator with variode1962 3-phase alternator with variode1965 Transistorised ignition1967 D-Jetronic gasoline-injection system

(pressure-controlled)1978 Antilock braking system (ABS)1979 Motronic

(combined ignition and fuel-injection system)1982 Electronic ignition system1982 Knock control1986 Electronic diesel control (EDC)1986 Electronic throttle control (ETC)1987 Traction control system (TCS)1989 Electronic transmission-shift control

(stand-alone system Tiptronic)1989 CAN (Controller Area Network)1989 Vehicle navigation system (Travelpilot)1991 Litronic1994 ME-Motronic (integrated ETC)1997 Electronic stability program (ESP)2000 MED-Motronic (gasoline direct injection)2000 Adaptive Cruise Control

Historical development of electronic systems in motor vehicles (examples)

1

Robert Bosch GmbH

MiniaturisationTransistorised ignition was first used ongasoline engines in 1965. It did away withthe negative effect on ignition timing accu-racy of erosion caused by electrical arcingbetween the contact-breaker points. Thetransistor had gained a foothold in the mo-tor vehicle, heralding the start of the elec-tronic age. But is wasnt until electroniccomponentry was miniaturised that the de-cisive step was taken in making electronicsystems in automobiles capable of the levelsof performance that are taken for grantedtoday. Enormous advances in miniaturisa-tion were made in the area of semiconductorcomponents in particular, making it possibleto integrate more and more functions withina component that occupied only a tinyamount of space. Every ECU contains mi-crocontrollers that combine millions oftransistor functions on semiconductor chipsthat take up only a few square millimetres.

It has also been possible to substantiallyreduce the dimensions of power compo-nents such as output stages for controllingactuators. For example, multiple ignitionoutput stages are now combined in a singlecomponent. This means that an external ig-nition output stage is no longer required. Itis now integrated in the engine-management

ECU. Consequently, the external ignitionoutput stage previously used can be dis-pensed with. The associated reduction in thenumber of components also improves thereliability of the system.

The size of discrete components (resistors,capacitors) has similarly been significantlyreduced. SMDs (Surface Mounted Devices)are soldered or bonded to the circuit boardwithout wire connections.

In spite of the continual growth in thenumber and complexity of the functionsperformed, the miniaturisation of electronicequipment has meant that the size of theECUs continues to shrink (Fig. 3).

Memory capacityWhereas a memory capacity of 4 kilobyteswas adequate for the modest requirementsof, for instance, a management system for agasoline engine in the late 1970s, 10 yearslater the figure had reached 30 kilobytes.The incorporation of more and more func-tions in the engine-management ECU led to an explosion in the demand for memorycapacity. By the year 2000, the required ca-pacity had reached 500 KB. Other automo-tive electronic systems have followed a simi-lar pattern of development. And there is noforeseeable end to this trend.

Automotive microelectronics History of development 7

Fig. 3a 1979 Jetronic ECU

with 290 compo-nents and a weightof 1.14 kg

b 1996 Motronichybrid ECU with82 components anda weight of 0.25 kg

a b

Gasoline-engine ECUs3

U

AE

0857

Y

Robert Bosch GmbH

Semiconductors occupy a unique positionbetween conductors and non-conductors.The electrical conductivity of semiconduc-tors is dependent on pressure, temperature,intensity of incident light or the proportionof foreign atoms in the semiconductor ma-terial. Those properties are the basis forsemiconductor technology.

Terminology

ElectronicsAccording to the definition of the IEC (In-ternational Electrotechnical Commission),electronics is the branch of science and tech-nology that deals with the study and utilis-ation of the physical phenomena in gases,solids and vacuums that are related to theflow of electricity.

MicroelectronicsAccording to DIN 41 857, microelectronicsis a branch of technology that deals with theconception, design, technology, manufactureand use of highly miniaturised electroniccircuits.

It is evident from that definition that micro-electronics is concerned with miniature cir-cuits made up of multiple individual com-ponents rather than miniaturised discretecomponents (i.e. components with clearlydefinable individual functions). Only inte-grated film and semiconductor circuits andcomposite microcircuits (hybrid circuits)fall into that category.

Electrical conductivityThe specific suitability of different materialsfor conduction of electricity is determinedby the number and mobility of free chargecarriers that they contain. The electricalconductivity of solids at room temperaturecan vary by 24 powers of ten between differ-ent materials.

Solids are subdivided into three classes ofmaterial according to their electrical con-ductivity (Table 1).

All solids contain around 1022 atoms percubic centimetre which are held together byelectrical forces.

Conductors (Metals)In metals, the number of free charge carriersis very large (one or two free electrons peratom). Their level of mobility is moderate.The electrical conductivity of metals is high.In good conductors it can be as much as106 siemens/cm.

Non-conductors (Insulators)In insulators, the number of free chargecarriers is practically zero and consequentlythe electrical conductivity virtually non-existent. The conductivity of good insulatorsis of the order of 1018 siemens/cm.

SemiconductorsThe electrical conductivity of semiconduc-tors is somewhere between that of conduc-tors and insulators. Under normal conditionsthey also have a very small number of freecharge carriers, but that number can be sub-stantially increased by the application of en-ergy from an external source. Their conduc-tivity, therefore in contrast to that of metals

8 Basic principles of semiconductor technology Terminology/Electrical conductivity

Basic principles of semiconductor technology

Table 1

Classes of material based on conductivity (with examples)

1

Conductors (Metals)

Non-conductors (Insulators)

Semiconductors

Silver Teflon Silicon

Copper Quartz glass Germanium

Aluminum Aluminum oxide Gallium arsenide

Robert Bosch GmbH

and insulators is heavily dependent on: pressure (affects the mobility of charge

carriers), temperature (affects the number and

mobility of charge carriers), exposure to light (affects the number of

charge carriers), and added impurities (affects the number and

type of charge carriers).

The sensitivity of semiconductors to pres-sure, temperature and light makes themsuitable for use as sensors.

The ability to accurately modify andlocalise the conductivity of semiconductorsby the controlled introduction of impuritiesthat affect electrical properties (doping) isthe basis of semiconductor technology. Theelectrical conductivity that can reliably bebrought about in silicon by doping rangesfrom 104 to 102 siemens/cm.

As silicon is by far the most importantsemiconductor material, the explanationsthat follow will restrict themselves exclu-sively to that material. When solid, siliconconsists of a crystal lattice in which each sili-con atom is linked to four equally spaced ad-

jacent atoms. Every silicon atom has fourouter electrons (Fig. 1). Each pair of adjacentatoms is linked together by two shared elec-trons. In such a perfect crystal lattice, there-fore, there are no free charge carriers, whichmeans that the silicon is a non-conductor.This condition is changed fundamentally by the introduction of suitable impurities(doping), or energy from an external source.

n-type dopingThe addition of foreign atoms with fiveouter electrons (e. g. phosphorus) intro-duces free electrons because only four arerequired to bind each atom within the sili-con crystal lattice. Each phosphorus atomintroduced therefore supplies one free, nega-tively charged electron. The silicon becomesnegatively conductive (Fig. 2). It is then re-ferred to as n-type silicon.

p-type dopingThe addition of foreign atoms with threeouter electrons (e. g. boron) creates electrongaps. The boron atom is one electron shortof the number required to properly bind itwithin the silicon crystal lattice. The result-

Basic principles of semiconductor technology Electrical conductivity 9

Si4 +

Si4 +

Si4 +

Si4 +

Si4 +

Structure of a perfect, pure-silicon crystal lattice with four outer electrons (valance electrons )

1

S

AE

0047

Y

102

101

1

10

102

103

1041014 1016 1018 1020 cm3

Doping concentration

T= 300 K

n-type

p-type

Silicon51022 Si atoms/cm3

Ele

ctric

al c

ondu

ctiv

ity

Siemenscm

Electrical conductivity of silicon relative toconcentration of added doping substance

2

S

AE

0048

E

Robert Bosch GmbH

ing gaps are referred to as holes. In silicon,those holes are mobile, and within an electri-cal field they move in the opposite directionto electrons. Holes therefore act like free pos-itive charge carriers. Each boron atom intro-duced therefore supplies one free, positivelycharged hole. The silicon becomes positivelyconductive (Fig. 2) and is therefore referredto as p-type silicon.

The conductivity of n-type silicon is around10 times as high as that of p-type silicon as-suming other factors are equal (e.g. numberof foreign atoms, temperature, pressure) be-cause electrons can move about 10 times asfast as holes.

Intrinsic conductivityThe application of heat or light can generatefree charge carriers even in undoped silicon.They consist of electron-hole pairs and makethe semiconductor intrinsically conductive.Such conductivity is generally low comparedwith that produced by doping. The numberof electron-hole pairs increases exponentiallywith rising temperature and ultimately erasesthe electrical differences between areas ofp-type and n-type silicon created by doping.Consequently there are maximum limits forthe operating temperatures of semiconductorcomponents, as shown in the following table:

An n-type semiconductor always has someholes and a p-type semiconductor alwayshas some free electrons. Such minoritycharge carriers are fundamental to the wayin which almost all semiconductor compo-nents work (refer to section entitled Elec-tronic Components).

p-n junctionThe boundary between the p-type and n-type regions of the same semiconductorcrystal is called the p-n junction. Its proper-ties are fundamental to almost all semicon-ductor components.

p-n junction without external voltageIn the p-type region there are a large numberof holes () and few free electrons. In the n-type region, by contrast, there are ex-tremely few holes and a large number of freeelectrons () (Fig. 3). Due to the concentra-tion differentials, the mobile charge carriersin each region diffuse into the other region ineach case (diffusion currents). As a result, thep-type region is negatively charged and the

10 Basic principles of semiconductor technology Electrical conductivity

Fig. 3a Diffusion of mobile

charge carriersb Formation of

depletion layer(space-chargeregion)

Holes Electrons

Depletion layerDirection of diffusion

Boundary zone

Diffusion voltage

a b

n-type regionp-type regionn-type regionp-type region

+

Boundary zone

p-n junction without external voltage3

S

AE

0858

E

Material Max. operating temperature

Germanium 90 ...100C

Silicon 150 ...200C

Gallium arsenide 300 ...350C

Robert Bosch GmbH

n-type region positively charged. A potentialdifference (diffusion voltage) is thus createdbetween the p-type region and the n-typeregion which counteracts migration of thecharge carriers. This brings the equalisationof holes and electrons to a halt. As a result, aregion that is deficient in mobile charge car-riers, and therefore has poor electrical con-ductivity, is created at the p-n junction. It isreferred to as the depletion layer or space-charge region. Due to the diffusion voltage,the depletion layer has as strong electric field.

p-n junction with external voltageIf an external voltage is applied to a p-n junction, it produces the effectsexplained below (Fig. 4).

Reverse biasIf the negative terminal is connected to thep-type region and the positive terminal tothe n-type region, the width of the space-charge region increases. Consequently, cur-rent flow is largely inhibited apart from avery small residual current (reverse current)produced by minority charge carriers.

Forward biasIf the positive terminal is connected to thep-type region and the negative terminal tothe n-type region, the depletion layer is

broken down. When the diffusion voltage isexceeded, the charge carriers flood the p-n junction and a large current flows inforward direction.

Breakdown voltageThe breakdown voltage is the reverse-direc-tion voltage above which a small increase in voltage brings about a steep rise in thereverse current.

The cause of this effect is the release ofbound electrons from the crystal lattice inthe space-charge region due to the high fieldstrength (Zener breakdown) or due tosurges of accelerated electrons. The acceler-ated electrons strike other electrons, break-ing them free of their bonds and starting anavalanche-like increase in the number ofcharge carriers (avalanche breakdownor first breakdown). Both effects are re-versible. The nature of the cause and thelevel of the breakdown voltage are depen-dent on the doping concentration profile.

A second breakdown occurs if there is lo-calised heating of a semiconductor compo-nent caused by current constriction so thatthe area concerned becomes more conduc-tive. This results in a self-accelerating in-crease in current and leads to the destruc-tion of the semiconductor component.

Basic principles of semiconductor technology Electrical conductivity 11

Fig. 4a Reverse biasb Forward bias

+

+

+

Flow of holesFlow of electrons

Forward voltage

n-type regionp-type region + n-type regionp-type region

Reverse voltage

+ a b

p-n junction with external voltage4

S

AE

0754

E

Robert Bosch GmbH

Electronic devices are made up a large num-ber of components that can be subdividedinto two main categories: passive compo-nents and semiconductor components(active components); the latter can befurther subdivided into subcategories.

Passive components

Ohmic resistors, capacitors and inductancesare classed as passive components.

Ohmic resistorsOhmic resistors are generally made of ma-terials with conductive properties similar to metals e. g. carbon (carbon-film resistors)or special metal alloys (metal-film resistors).They are constructed so as to reduce to theabsolute minimum the effect of voltage,current and temperature on their electricalresistance. The conductor has a very smallcross-section relative to its length, this beingachieved either by the application of thinfilms to insulators or by winding wire into acoil.

In electronic circuits, resistors limit the cur-rent flow I or produce a voltage U propor-tional to the current. The resulting energyloss is converted into heat. An extreme ex-ample of this is a light bulb, in which anultra-fine tungsten filament is heated tosuch a degree that it starts to glow.

The resistance in a circuit is referred to bythe symbol R and its unit of measurement isthe ohm ().

CapacitorsThe most simple type of capacitor consists of two parallel metal plates that are not incontact with one another. The area and sep-aration of the plates as well as the mediumseparating them (dielectric) determine thequantity of charge carriers that can be storedby the capacitor (its capacitance). Using avacuum (or air) as the dielectric offers thelowest capacitance. The capacitance can be

substantially increased by the use of other in-sulating materials. The amplification factoris referred to as the relative permittivity ordielectric constant r. Since, in many cases,very large plates would be required for thefrequently used capacitance levels, capacitorsare normally made by winding long stripsinto a coil (wound capacitors) or packing a large numbers of small plates together(multi-layer capacitors). The electrolyticcapacitor uses a thin layer of oxide as thedielectric. This method makes it possible to manufacture small capacitors with largecapacitances.

When direct current is applied to a capaci-tor, it stores up charge until the limit of itscapacitance is reached; at that point currentcan no longer flow. Thus for a certainperiod at least the capacitor stores electri-cal energy that is then available if the exter-nal power supply fails. This effect is utilisedin the circuit for triggering the airbag, forinstance. Even if the wires connecting theairbag triggering unit to the vehicles batteryare severed in a serious accident, the capaci-tor still holds sufficient electrical energy toensure that the airbag is deployed.

In alternating-current circuits, a capacitorhas a similar effect to a resistor. Its resistanceis dependent on the capacitance and the fre-quency of the alternating current. As the fre-quency decreases, the resistance increases. Ifthe frequency is reduced to its lowest limit,i.e. zero (direct current), the resistance isequal to infinity and, consequently, no cur-rent flow is possible. This relationship isutilised, for example, by frequency filters inspeaker systems in order to filter out thehigh-frequency sounds from the lower-frequency ones (high-pass).

The capacitance in a circuit is referred to bythe symbol C and its unit of measurement isthe farad (F).

12 Electronic components Passive components

Electronic components

Robert Bosch GmbH

InductancesWhen an electric current flows through acoil, a magnetic field is created. The strengthof the magnetic field depends on thestrength of the current, the number of wind-ings in the coil and the properties of the coilcore (generally a ferrite or iron core). Induc-tance is the characteristic of a coil which in-dicates the amount of magnetic energy thecoil can store for a given current.

If the current or the magnetic flux changes,a voltage is induced in the coil which coun-teracts the generation of the magnetic field.Once the magnetic field of a coil carryingdirect current has reached its maximumstrength, current flow is no longer restricted.Thus inductance does not represent a lastinghindrance to the flow of direct current.

In an alternating-current circuit, because ofthe constant generation and collapse of themagnetic field (and the energy containedwithin it), an inductance acts as a frequency-dependent resistor, the characteristics ofwhich, however, are exactly the opposite ofthose of a capacitor. In this case, the higherthe frequency, the greater is the resistance.This means that in frequency filters such asthose referred to above in the description ofcapacitors, inductors can be used to filterout the low-frequency sounds from thehigh-frequency ones (low-pass).

The inductance in a circuit is referred to bythe symbol L and its unit of measurement isthe henry (H).

Semiconductor componentsSemiconductor components are active com-ponents that are generally subdivided intofour categories (Table 1). These categoriesare further subdivided according to the wayin which the components are manufacturedand their function. The first two classicalcategories are discrete semiconductor com-ponents and monolithic integrated circuits(see Table 1 for examples).

Discrete semiconductor components are self-contained, individually distinguishable com-ponents (discrete being derived from discre-tus, the past participle of the Latin verb dis-cernere meaning to distinguish). Accord-ing to the currently accepted definition, semi-conductor components with fewer than 100transistor functions are referred to as discrete.

Monolithic integrated circuitsICs (Integrated Circuits) are active compo-nents which incorporate more than 100 in-dividual functions on a single chip (mono-

Electronic components Passive components/Semiconductor components 13

Table 1

Semiconductor components (Examples)1Discrete active components

Integrated circuits (ICs)

Opto-electronic components

Micromechanical sensors (Examples)

Diodes Semiconductor resistors Transistors Thyristors

Analog circuits Digital circuits Mixed-signal circuits

Photoresistor Photodiode Photovoltaic cell Laser diode Phototransistor Charge-coupled device

Pressure sensors Acceleration sensors Yaw rate/angle sensors Flow sensors Temperature sensors Position/angle sensors (Hall-effect sensors) Gas sensors

Robert Bosch GmbH

lithic literally meaning made from a singlestone from the Greek monolithos mean-ing single stone).

Optoelectronic components form the thirdcategory. They are so distinctly differentfrom the classical active components interms of their method of production andusage, that classification as a separate cate-gory makes sense.

Micromechanical sensors using MST (Micro-System Technology) or MEMS (MicroElectroMechanical Systems) have more re-cently come into being as a fourth category.

In this case too, the methods of productionand the type of use differ substantially fromconventional active components.

DiodesDiodes are semiconductor components witha p-n junction and two connections one tothe p-type region and one to the n-type re-gion (diode means two ways). A diodeutilises the characteristics of the p-n junc-tion. The pattern of doping impurity con-centration within the crystal determines thespecific characteristics of diodes.

Diodes designed for a forward current ofmore than 1 A are referred to as power diodes.

14 Electronic components Semiconductor components

Fig. 1a Characteristic of

diodes, e.g. rectifierdiode, Zener diodeand Schottky diode

b Section of tunnel-diode characteristic

Fig. 2a Half-wave rectifierb Bridge rectifierUF Diffusion voltage

(approx. 0.7 V)U~ AC voltageU1 Transformed

AC voltageUg Rectified voltageRL Load resistanceU1M Amplitude of U1UgM Amplitude of Ugt Time

Voltage in re-verse direction

Break-down zone

Reverse-bias zone

in forward direction

reve

rse

dire

ctio

nC

urre

nt in fo

rwar

d di

rect

ion

0

0

Forward voltage

Forw

ard

curr

ent

0

1

2

3

4

5

mAFor-ward-bias zone

0.7 V

a b

0 100 200 300 mV

Diode characteristics1

U

AE

0886

E

a b

+

U UUgU1

Ug

U

UFU1M

UgM

t

U1

U1

UF

RL

+

-

+

+-

-

-U1

UgRL

+

-

+

-

t

Ug

U

2 .UFU1M

UgM

Rectifier circuits using diodes2

U

AE

0897

Y

Robert Bosch GmbH

Rectifier diodeA rectifier diode allows current to pass inone direction (positive terminal connectedto p-type region) but not in the other (posi-tive terminal connected to n-type region). Itacts like a flow control valve and is thereforethe ideal component for rectifying alter-nating current (Figs. 1a and 2).

The current flowing in reverse direction(reverse current) is much smaller than theforward current (approx. 1/100 thestrength). It increases rapidly with risingtemperature.

Amongst other applications, rectifier diodesare used in automotive alternators to con-vert the alternating current into direct cur-rent. Because of the high ambient tempera-tures to which the alternator is subjected,the reverse current is a critical factor and hasto be taken into account in the design of thediodes used.

Rectifiers for high reverse voltagesIn order to obtain a high reverse voltage inrectifiers, at least one region of the rectifiermust have low conductivity. However, thatmeans a high resistance in forward directionand consequently high power loss and exces-sive heating.

Creating a very lightly doped i-typeregion between the heavily doped p and n-type regions produces a p-i-n rectifierwhich, despite having a high reverse voltage,has a low forward resistance (conductivitymodulation). The i-type region acts likeintrinsically conductive silicon.

Uses: all rectification applications involv-ing high voltages.

Zener diode The Zener diode is a semiconductor diodein which the reverse current rises abruptlyupwards of a certain voltage as a result ofZener breakdown and/or avalanche effects.Even though large numbers of these diodesdepend for their function on avalanche ef-fects rather than Zener breakdown, they arestill referred to as Zener diodes.

Zener diodes are rated for different break-down voltages depending on their particularapplication. They are designed for continu-ous operation at the breakdown voltage.They are used chiefly for voltage limitationand for generating a voltage reference, e.g. indirect-current power supply units.

Variable-capacitance diodeThe space-charge region at the p-n junctionacts like a capacitor; the semiconductor ma-terial depleted of charge carriers acts as thedielectric. An increase in the applied voltageincreases the width of the depletion layerand reduces the capacitance; reducing thevoltage increases the capacitance.Capacitance diodes are used mainly for reso-nant circuit tuning and frequency multipli-cation (e.g. in tuners).

Schottky diodesThe Schottky diode is a semiconductordiode with a metal-semiconductor junction.Because electrons transfer more easily fromthe n-type silicon to the metal film than inthe opposite direction, an electron-depletedboundary layer known as the Schottky bar-rier is created in the semiconductor. Trans-mission of charge is performed exclusivelyby electrons; as a result, extremely highswitching speeds are achieved because nominority storage effects occur.

Schottky diodes are suitable for use as fastswitches and microwave rectifiers.

Tunnel diodeThe tunnel diode (or Esaki diode) is a semi-conductor diode with a very heavily dopedp-n junction in which the tunnel effect,which is explainable only by quantum me-chanics, occurs to such a degree that whenoperated in forward direction a negative dif-ferential conductivity occurs over a certainrange of the current/voltage characteristic(Fig. 1b).

Tunnel diodes are used as oscillators inthe gigahertz (GHz) range and for low-noiseamplifiers.

Electronic components Semiconductor components 15

Robert Bosch GmbH

Semiconductor resistorsIn contrast with ohmic resistors, these com-ponents are voltage, current and tempera-ture-dependent. They generally consist ofpolycrystalline semiconductor materials.The effects that occur in semiconductor re-sistors are based in part on depletion-layerproperties that appear at the crystalliteboundaries.

VaristorsThe resistance of a varistor decreases as volt-age increases. The polarity of the voltagemakes no difference. A varistor consists ofpolycrystalline ZnO or SiC powder mixedwith a binder, compressed and sintered.

Uses: e. g. voltage stabilisers, surge protec-tors for semiconductor circuits.

NTC resistors (thermistors)NTC (Negative Temperature Coefficient)resistors (thermistors or thermal resistors)are, as the name suggests, resistors that havea marked negative temperature coefficient.Their electrical resistance decreases as thetemperature increases and consequently,NTC resistors conduct electricity better athigh temperatures than at low temperatures.

NTC resistors are made from polycrys-talline metal oxides such as Fe2O3, ZnTiO4or MgCr2O4 by a process of compressionand sintering. Their temperature coefficientscan be as much as 6 %/K.

Uses: e. g. temperature sensors.

PTC resistors (thermistors)PTC (Positive Temperature Coefficient) re-sistors are resistors that have a positive tem-perature coefficient (of the order of a few/K). Their electrical resistance increases asthe temperature increases and consequently,PTC resistors conduct electricity better atlow temperatures than at high temperatures.Most metals are thermal resistors with verylow temperature coefficients.

The term PTC resistors refers to resistorsmade of semiconductor materials. PTC re-sistors made of ferro-electrical ceramic ma-terial (e. g. polycrystalline barium titanate)

have a relatively narrow temperature rangeand a very high positive temperature coeffi-cient (+6 to +60 /K). PTC resistors madeof silicon have wide temperature range and a virtually constant positive temperaturecoefficient (approx. +0.8 /K).

Uses: e. g. liquid level sensors, heatingregulators.

MagnetoresistorsA magnetoresistor is a magnetically control-lable semiconductor resistor. Its resistanceincreases as the magnetic flux density B in-creases (Fig. 3).

The indium antimonide film (approx.25 m thick) contains minute needles ofnickel antimonide with very high electricalconductivity. The flux density prevents thecharge carriers from following a direct path.They move at an angle from one needle tothe next. Within the metallic needles, differ-ing charge-carrier densities are immediatelyequalised. As the flux density B increases, thecurrent paths become more and more an-gled and the distance travelled by the chargecarriers therefore greater. As a result, the re-sistance of the magnetoresistor increases.

Uses: e. g. magnetic-field sensors, control-lable resistors.

16 Electronic components Semiconductor components

Fig. 3Current path under theinfluence of a low (left)and a high (right)magnetic flux density B

B B

Magnetoresistor3

U

AE

0003

Y

Robert Bosch GmbH

TransistorsTransistors can be used control a large cur-rent with a small control current. Conse-quently, these semiconductor componentscan be used as power amplifiers or switches.

Transistors have three or more connec-tions. Figure 4 illustrates the classification ofthe transistor family. Transistor is a con-traction of the term transfer resistor.

Bipolar transistorsBipolar transistors consist of three semicon-ductor regions of differing conductivityarranged to form a p-n-p or an n-p-n con-figuration. These regions (and their connec-tions) are called the emitter (E), the base (B)and the collector (C).

Depending on their applications, transis-tors are categorised as low-signal transistors(up to 1 watt power loss), power transistors,switching transistors, low-frequency transis-tors, high-frequency transistors, microwavetransistors, phototransistors, etc.

Such transistors are called bipolar becausethey make use of charge carriers of both

polarities (holes and electrons). In an n-p-n transistor, positive charge carriers(holes) in the base current control roughly100 times as many negative charge carriers(electrons) flowing between the emitter andthe collector. This corresponds to a currentamplification factor of around 100. Bipolartransistors are thus controlled by means ofthe base current.

Method of operation of a bipolar transistor(with reference to an n-p-n transistor):The emitter-base junction (EB) is polarisedfor forward bias (Fig. 5 overleaf). Thismeans that electrons migrate into the baseregion. The base-collector junction (BC) ispolarised for reverse bias. This creates aspace-charge region with a powerfulelectrical field.

Discernible coupling (transistor effect)occurs if the two p-n junctions are very closeto one another (less than 10 m apart in sili-con). In that case the electrons crossing theEB junction pass through the base to thecollector. As soon as they come within the

Electronic components Semiconductor components 17

Fig. 4B BaseE EmitterC CollectorS SourceD DrainG Gate* Bulk connection

(connection desig-nation not usual)

E C

B

E C

B

G

DS

G

DS

G

DS

G

DS

G

DS****

G

DS

Bipolar transistorsInjection transistors

Unipolar transistorsField-effect transistors (FETs)

Junction-gate FETsJFETs, MESFETs

Insulated-gate FETs, IGFETs, MISFETsSpecial case: MOSFETs, or MOS transistors

Depletion-typeintrinsically conductive

Depletion-typeintrinsically conductive

Enrichment-type intrin-sically non-conductive

p-typechannel

n-typechannel

p-typechannel

n-typechannel

p-typechannel

n-typechannel

PNP NPN

Transistors

Transistor family with symbols4

U

AS

0924

E

Robert Bosch GmbH

range of the electrical field at the BC junc-tion, they are accelerated into the collectorregion and travel onwards as the collectorcurrent. The concentration differentialwithin the base thus remains and, therefore,so does the impetus for continued electronmigration from the emitter to the collector(Fig. 5).

In conventional transistors, 99 % or moreof electrons originating from the emitterpass into the BC space-charge region andform the collector current. The few that arelost have found their way into the electrongaps in the p-type base while passingthrough it. If this effect were left unchecked,the base would become negatively chargedand the resulting repulsion forces wouldcompletely prevent the continued flow ofelectrons within an extremely short space of time (50 ns). Such a build-up of negativecharge can be partially or entirely counter-acted in a transistor by a low base current of positive charge carriers (holes). Smallchanges in the base current bring about largevariations in the emitter-collector current.

With moderate base currents, the n-p-n transistor acts as a bipolar, current-controlled, amplifying semiconductorcomponent. If the base current alternatesabruptly between very low and very highlevels, the transistor acts as a switch.

Field-effect transistors (FETs)In a field-effect transistor (Fig. 6a), the cur-rent flowing from the source (S) to the drain(D) through a conductive channel is con-trolled by an electric field. That field is gen-erated by a voltage applied via a controlelectrode known as the gate (G). The use ofan electric field to control the current flowexplains the origin of the term field-effecttransistor (FET). In contrast with bipolartransistors, field-effect transistors use onlyone type of charge carrier (either electronsor holes) and are therefore also referred to asunipolar transistors. Field-effect transistorsare subdivided into junction-gate field-effect transistors

(junction FETs, JFETs) insulated-gate field-effect transistors, in

particular metal-oxide semiconductor(MOS) field-effect transistors (MOSFETs).

MOSFETs are particularly suited to use inhighly integrated circuits. Power FETs havesuperseded bipolar transistors for mostapplications.

Method of operation of a junction-gate FET(with reference to a JFET with an n-typechannel, Fig. 6a):DC voltage is applied to the ends of an n-type crystal. Electrons flow from thesource, S, to the drain, D. Two p-type regionsat the sides forming the gate (G), and thenegative voltage applied to them, determinethe width of the channel. If the negative gatevoltage is increased, the space-charge regionsextend further into the channel and constrictthe current path. The voltage at the controlelectrode, G, thus controls the current be-tween the source, S, and the drain, D. TheFET only requires charge carriers of a singlepolarity in order to function. The current iscontrolled virtually without power con-sumption. The junction FET is thus a unipo-lar, voltage-controlled component.

Method of operation of a MOS FET(with reference to a MOS FET with p-typeenrichment): while no voltage is present at

18 Electronic components Semiconductor components

Fig. 5N n-type siliconP p-type siliconE EmitterB BaseC Collector

C

_ +UEC

_ +UEB

E1%

100 %

++++++++

99%

+++

+++

+++

+++

---

---

---

---

---

---

N P N

Emitter Base CollectorEB BC

B

Bipolar n-p-n transistor5

U

AE

0458

-1E

Robert Bosch GmbH

the gate electrode, no current flows betweenthe source, S, and the drain, D; the p-n junc-tions are reverse biased (Fig. 6 b). A negativevoltage applied to the gate, G, forces theelectrons in the n-type region below theelectrode into the centre of the crystal anddraws holes which are always present asminority charge carriers even in n-type sili-con to the surface. A narrow p-type layer iscreated below the surface, in effect a p-typechannel. Current, which is comprised onlyof holes, can now flow between the two p-type regions (source and drain).

As the gate voltage acts through an insu-lating oxide layer, no current flows in thecontrol circuit and control therefore requiresno power. The MOS FET is thus a unipolar,voltage-controlled component.

PMOS, NMOS and CMOS transistorsIn addition to the p-type channel MOStransistor, or PMOS transistor, there is theNMOS transistor in which the p and n-typeregions are reversed in comparison with thePMOS transistor. Due to the higher mobilityof electrons, NMOS transistors are fasterthan PMOS transistors which, for physicalreasons, are easier to manufacture and weretherefore available first.

If PMOS and NMOS transistors are cre-ated in pairs on the same silicon chip, theyare referred to as CMOS transistors (comple-mentary MOS transistors, Fig. 6c). Theirparticular advantages are very low powerloss, high interference immunity, TTL-com-patibility (transistor-transistor logic in elec-trical circuits), low power-supply voltage andsuitability for analog signal processing.

The low power consumption, particularlyfor digital circuits, is due to the fact thatcurrent only has to flow when a digital cellchanges ( e.g. switches from 0 to 1). Re-tention of the information requires only thata voltage is applied. Since, with this configur-ation, the PMOS and NMOS transistors areconnected to one another, the leakage cur-rent is minimal. More than 80% of ICs arenow manufactured using CMOS technology.

ThyristorsThyristors are semiconductor componentswith at least three p-n junctions (one ofwhich may also be replaced by a suitablemetal-semiconductor contact) that can beswitched from a reverse-bias condition to aforward-bias condition (or vice versa). Theterm thyristor is used as the generic term forall types of component which conform tothat definition. It is a contraction of the twowords thyratron (gas-filled tube triode) andresistor.

Uses in power electronics: speed and fre-quency control, rectification and conversion,switching.

Electronic components Semiconductor components 19

Fig. 6a Depletion layer,

field-effecttransistor

b PMOS transistorc CMOS transistor

pairN n-type siliconP p-type siliconS SourceG GateD DrainUGS Voltage between

G and SUDS Voltage between

D and S

D

G

_ +UDS

UDS

UGS

UGS

SN

P

P

Space-charge zone (depletion layer)

Source Gate Drain

a

b

c

N

PMOS transistor NMOS transistor

Source

Output Input

GateDrain

Source

P P

silicon

NNp-type well

_+

_+

N

p-type channel

Source Gate

Gate oxide

Drain

PP+

+

++

+

silicon

Transistor designs (Schematic)6

U

AE

0768

E

Robert Bosch GmbH

Monolithic integrated circuits (ICs)

Monolithic integrationThe planar process is a method of manufac-turing semiconductors which makes it pos-sible to produce all components of a circuit(resistors, capacitors, diodes, transistors)and the conductive connections betweenthem on a single silicon chip in a single pro-duction process. It involves the creation ofmultiple p and n-type layers in a multi-stagedoping process using a pre-doped single-crystal silicon disc or wafer.An integrated circuit IC (Integrated Circuit,see section entitled Manufacture of Semi-conductor Components and Circuits) doesnot contain any separate (discrete) com-ponents but rather switching elements orfunctional elements.

Degree of integrationThe degree of integration is defined as thenumber of functional elements, transistorsor gates on a single chip. The following cate-gories are defined based on the degree ofintegration (and chip surface area): SSI (small-scale integration): up to

around 1000 elements per chip, averagechip area 3 mm2 (varies considerablydepending on level of power loss)

MSI (medium-scale integration): up toaround 10,000 elements per chip, averagechip area 8 mm2

LSI (large-scale integration): up to100,000 elements per chip, average chiparea 20 mm2

VLSI (very-large-scale integration): morethan 100,000 elements per chip, averagechip area 30 mm2

The number of functional elements onVLSI chips is constantly increasing. Micro-processors can now have up to 10 milliontransistor functions per chip. Memory mod-ules (DRAM) can even contain several hun-dred million transistors. The number ofsuch functional elements is subject to anexponential increase over time. This fact was first recognised by Gordon Moore

20 Electronic components Semiconductor components

Fig. 7a Increase in number of

transistors per chip inmemories andprocessors (GordonMoores Law)

b Fall in cost of transis-tors

c Decreasing size ofstructural units of new processors

Year

Stru

ctur

e si

ze

2

4

6

8

10m

01980 1990 20001970

8008: 10 m

8080: 6 m

III III

mmm

= 0.8 = 0.35= 0.18

Pentium I

II

III

8088: 3 m

386DX: 1,5 m

486DX: 1 m

US

cen

ts p

er tr

ansi

stor

0.01

0.1

1

10

100C

0.001

8080

8008

8088

386DX

486DX

PentiumI II

III

Tran

sist

ors

per

chip

10 K

100 K

1 M

10 M

100 M

1 Ga

b

c

MemoriesProcessors

1 K

8080

8008

8088

256

256

64

64

16

16

4

1 K

4

1 M

1 G

386DX

486DX

Pentium IIIIII

Development of monolithic integration on a singlechip in the period from 1970 to 2000(logarithmic representation in a and b).

7

U

AE

0759

E

Robert Bosch GmbH

(co-founder of the company Intel) and is ex-pressed by Gordon Moores Law (Fig. 7a):Elements per chip = 2(year 1956) 2/3

This law states that the number of func-tions per chip will double every 18 months.Since the structural units in which the chipsare manufactured are becoming smaller atthe same time as the number of functionalelements is growing, the size of the ships al-ters very little and the costs per transistor aresimilarly dropping exponentially (Fig. 7b).Thus the structural units of new processorshave shrunk over a period of 28 years from10 m in 1972 to 0.18 m in the year 2000(Fig. 7c).

The degrees of integration LSI and VLSIdemand methods such as CAD (computer-aided design). VLSI circuits in particular canonly be created with the help of highly so-phisticated programs which can convert en-tire function blocks into corresponding cir-cuit subdivisions. Such program systems arecalled HDL (Hardware Description Lan-guage) or VHDL (Visual HDL).

IC classificationThere are various systems for classifying ICs,although there are invariably mixed cate-gories within each system. Classification by method of production:

Bipolar/Unipolar (MOS);mixed category: e. g. BiCMOS, BCD

Classification by function:Analog/Digital;mixed category: e. g. mixed-signal IC

Classification by application:Standard IC/ASICmixed category: e. g. ASSP

ASIC (application-specific IC): developedand produced specifically for a particularapplication and exclusively for a particularcustomerASSP (application-specific standard prod-uct): developed and produced specificallyfor a particular application for a particularcustomer but is also sold to others for thesame type of application.

Apart from a few exceptions, ICs are nowalways based on MOS or combined technolo-gies, and for that reason bipolar ICs are notdescribed in any further detail at this point.

Analog circuitsAnalog circuits are required wherever elec-tronic equipment has to communicate withthe outside world which is always analog.This applies in particular to the areas ofpreparation and processing of input andoutput signals. Examples of analog-inputsignal generators are microphones (tele-phone) and sensors. Examples of analog-output signal receivers are speakers or actua-tors.Figure 8 illustrates the stages of develop-ment of an analog IC from conception toinstallation in an ECU.

Electronic components Semiconductor components 21

Fig. 81 Development and

simulation 2 Layout3 Production4 Testing, analysis and

release 5 Packing, final testing6 Installation in an ECU

MVcc

Vccs

On/off

2

1

3

4

6

5

Development of an analog IC from conception to installation

8

U

AE

0762

E

Robert Bosch GmbH

Basic structures: stabilised-voltage sources,stabilised-current sources, differential am-plifier stages, coupling circuits, potentialshifting circuits, output stagesApplication-based ICs: operational ampli-fiers, voltage regulators, comparators,timers, transducers, interface circuitsSpecialised ICs: voltage references, broad-band amplifiers, analog multipliers, functiongenerators, phase-locking circuits, analogfilters and switches

Digital circuitsDigital circuits (Fig. 9) are used whereverlarge volumes of data have to be processedwithin a short space of time. High-perfor-mance microcontrollers in particular canperform several hundred million computa-tional operations per second. They are thusable to utilise the input data to provide therequired, highly accurate response at theoutput of the particular device. Similarly,large volumes of data can be transmitted viaa cable link in coded form, e.g. several thou-sand telephone calls can be transmitted si-

multaneously via a fibre-optic cable withoutcross-talk.

Since inputs and outputs for communi-cation with the real world always have to be analog, analog-digital converters (ADCs)are used at the inputs of a digital circuit, anddigital-analog converters (DACs) at the out-puts.

Digital systems make use of a whole series of recurring basic circuits and variations ofthem. The range extends from simple gatesto memories, microprocessors and micro-controllers.

Digital modules can only be connected upto form a complete system if power supplyvoltages, logic signal levels, switching speedsand signal transmission times are compat-ible. This requirement is met within a familyof circuits. Since, apart from a few spe-cialised applications mainly in military sys-tems, CMOS logic circuits are the only typenow used, such compatibility is no longer aproblem.

22 Electronic components Semiconductor components

Digital circuit on a chip9

U

AE

0013

Y

Robert Bosch GmbH

Mixed-signal ICsICs in which both analog and digital circuitelements are combined are called mixed-sig-nal ICs. They are an essential requirementfor the combination of entire electronicsystems on a single piece of silicon (SoC:System-on-a-Chip).

With ever-increasing degrees of integra-tion, this type of circuit is becoming moreand more attractive and the number of ap-plications is continually increasing. It is nowconceivable, for example, to integrate theentire electronic circuitry for a mobilephone or a complete Internet-access systemon a single chip. Where only low levels ofelectrical power are required, such as in theexamples quoted, both the analog and thedigital sections are manufactured usingCMOS technology.

Mixed-signal ICs for applications involv-ing higher electrical power levels, such as arefrequently encountered in automotive sys-tems, are typically hybrids of various tech-nologies, such as BCD systems (where Bstands for bipolar, C for CMOS and D forDMOS).

Figure 10 shows how the functions, sens-ing, analysing and acting are integratedin a mixed-signal IC. In such circuits, thebipolar section is used for converting theanalog inputs. The CMOS section of the ICperforms the logical processing operationsusing digital technology. The DMOS sectionenables high analog output performance.

Elements of digital circuitsDigital circuits assign voltage signals one ofthe two signal levels 0 (low voltage level)or 1 (high voltage level) and process thosesignals digitally. As digital circuits make upthe greater part of all ICs and our day-to-day lives are no longer conceivable withoutthem, a few typical circuit elements aredescribed below.

Gates Gates are logical-operation circuits with twoor more inputs. The input signals are logicallylinked to one another so that their variouscombinations determine the output signalsaccording to a defined logic (e.g. AND gate).

Electronic components Semiconductor components 23

Power (DMOS)Digital logic (CMOS)Analog (bipolar)

Acting

Analysing

Sensing

Mixed-signal IC10

U

AE

0763

E

Robert Bosch GmbH

InvertersInverters are digital circuits that convert thedigital input signal 0 into the output signal1 and vice versa.

BusThe bus links up the individual elements ofa digital circuit; this relates both to connec-tions within an integrated circuit and theinterconnection of separate digital ICs.

A bus can connect up a large number ofseparate modules with different functionsbut with electrically identical interfaces. Aserial bus (e. g. for connecting up serialEEPROMs) transmits the data along a singlelead. A parallel bus (e. g. address bus, databus, control bus) is a bundle of parallel lines.The number of separate items of informa-tion that can be simultaneously transmitted(equal to the number of bus lines) is, to-gether with the data transmission speed, ameasure of the performance capacity of thedata bus. The most common types of busare 8-bit, 16-bit and 32-bit.

The dimensioning of the data bus is deter-mined by the capacity of the CPU (centralprocessing unit). In order for a system toachieve its maximum performance capa-bility (i.e. maximum processing speed) theCPU and data bus should have the samecapacity, i.e. an 8-bit CPU should use an 8-bit bus, a 16-bit CPU a 16-bit bus, etc.

In addition, the bus should be able totransmit the data as quickly as the CPU canprocess it. This is often not the case withexternal busses (e. g. CPU clock speedfCC = 400 MHz and bus clock speedfCB = 133 MHz).

Only one bus node at a time can write datato the bus. According to the bus type, thisaddresses the data in such a way that it canonly be read by the addressee, or else itswitches all other bus nodes off before send-ing the information and then on again after-wards.

Clock-pulse generator (CPG)A clock-pulse generator ensures that all op-erations on the microcomputer are synchro-nised with a defined timing pattern. Theclock-pulse generator must be matched tothe required speed of computing operations.

Input/output (I/O) unitThe I/O unit handles data exchange with theoutside world. The input signals are scannedat the required frequency. The output signalsare transmitted at the speed required by, andin the optimum order for, the applicationconcerned or are held in temporary storageuntil called for.

MicrocomputersThe microcomputer consists of the follow-ing interacting components (Fig. 11): A microprocessor as the CPU (central pro-

cessing unit). For its part, the micropro-cessor consists of the control unit and thearithmetic and logic unit. The arithmeticand logic unit performs arithmetical andlogical operations, as the name suggests,while the control unit carries out the in-structions from the program memory.

Input and output units (I/O units) whichhandle the exchange of data with periph-eral devices. Peripheral devices include in-put and output devices and external datastorage media.

A program memory in which the operatingprogram (user program) is permanentlystored (ROM, PROM or EPROM).

Data memories for holding the data beingprocessed at any particular time. This datachanges continually (RAM).

The bus system which links up the individ-ual components of the microcomputer.

A clock-pulse generator (oscillator) whichensures that all operations on the micro-computer are synchronised with a definedtiming pattern.

Logic circuits which are modules with spe-cialised tasks such as program interrupts.They are integrated in individual I/O units.

24 Electronic components Semiconductor components

Robert Bosch GmbH

Electronic components Semiconductor components 25

I/O Bus

cont

rolle

r

Com

mun

icat

ion

with

ext

erna

lch

ips

via

addr

ess/

data

bus

Mic

rop

roce

sso

rC

entra

l Pro

cess

ing

Uni

t (C

PU

)

Arit

hmet

ic a

nd L

ogic

Uni

t (

ALU

)4-

, 8-,

16-,

32-b

it

Vo

latil

e re

ad/w

rite

m

emo

ry (

RA

M)

for

varia

ble

data

Mem

ory

capa

city

64 b

ytes

3

2 kb

ytes

Rea

d-o

nly

mem

ory

(RO

M, E

PR

OM

,fla

sh E

PR

OM

)F

or p

rogr

ams

and

perm

anen

t dat

a re

cord

s

Mem

ory

capa

city

2 kb

ytes

5

12 k

byte

s

No

n-v

ola

tile

read

/w

rite

mem

ory

(EE

PR

OM

)

Mem

ory

capa

city

32 b

ytes

1

kbyt

e

Bu

s

4-,

8-, 1

6-, 3

2-bi

t dat

a ci

rcui

t C

lock

-pu

lse

gen

erat

or

(osc

illat

or)

I/OIn

terr

upt

cont

rolle

r

Mon

itorin

gci

rcui

t(w

atch

dog

)

I/OE

vent

coun

ter

Cou

nter

8

64

bit

I/OS

igna

l acq

uisi

tion

and

outp

ut w

ithtim

e re

fere

nce

(Tim

er, t

ime

proc

essi

ng u

nit,

inpu

t cap

ture

, ou

tput

-com

pare

re

gist

er)

Res

olut

ion

50

ns

Tim

e ra

nge

50 n

s

1s

I/OA

nalo

g/D

igita

l(A

/D)

conv

erte

r

Res

olut

ion

8

10 b

it4

3

2 ch

anne

ls

I/OD

igita

l in

puts

/ou

tput

s

8

32

chan

nels

I/O Ser

ial

inte

rface

(UA

RT

,S

PI,

CA

N)

Dat

a ra

te20

0 bi

t/s

1

Mbi

t/s

Mic

roco

mpu

ter

Microcomputers11

U

AE

0454

-1E

Robert Bosch GmbH

The chief components of a microcomputerare generally separate modules connected toone another on a printed-circuit board. Forsimpler tasks such as are more and more fre-quently required for Internet access on wire-less communication devices, single-chipcomputers are becoming increasingly com-mon. They integrate the functions detailedabove on a single silicon chip (System-on-a-Chip). The relatively small amount of RAMthat can be accommodated on a chip at rea-sonable cost limits the performance capabil-ities of such highly integrated systems.

MicroprocessorA microprocessor is a central processingunit (CPU) in the form of an integratedcircuit on a chip. The concept of the micro-processor avoids individualisation despitethe high degree of integration, and enablesadaptation to the multiplicity of practicaldemands by reliance on programming. Amicroprocessor is not capable of function-ing on its own; it is always part of a micro-computer. At present, there are 16-bit, 32-bitand 64-bit microprocessors. Processors aresubdivided into two main categories: PCs use CISC (Complex Instruction-Set

Computer) processors. These are ex-tremely versatile and permit unrestrictedprogramming.

Workstations normally use RISC(Reduced Instruction-Set Computer)processors. These perform specific taskssuch as are frequently required on work-stations much faster, but are distinctlyslower for all other tasks.

MicrocontrollersThe microcontroller is a component that in-corporates the following elements on a chip: CPU, random-access memory (RAM), peripheral modules (input/output, inter-

rupt, timer, serial interfaces) and an optional read-only memory (ROM).

With these integrated components, the mi-crocontroller is capable of stand-alone oper-

ation. For that reason, it is also referred to asa single-chip microcomputer.

The microcontroller is used to controlself-regulating systems such as an engine-management system.

At present, there is a choice of 4-bit, 8-bit,16-bit or 32-bit microcontrollers for thevarious types of application. Depending onapplication, they may also have expansionmodules connected to them (e.g. additionalmemory for data and program code).

The program that is run by the CPU ispermanently fixed in the read-only memoryand is not changed for different appli-cations. This distinguishes the single-chipmicrocontroller from the PC.

TransputerA transputer is a special type of micro-processor that is particularly suited to theconstruction of parallel computer networks.In addition to the usual components of amicroprocessor, it has communication andprocess-handling hardware on a chip.

ProgrammingThe only command form capable of directinterpretation by a microprocessor is a bitpattern, i.e. the binary representation of anumber. Since, however, this form of in-struction is not easy to work with for a pro-grammer, and is therefore susceptible toerrors, easily memorable abbreviations(mnemonics) are used. These are automati-cally translated by an assembler programinto bit patterns (machine code) that can beunderstood by the microprocessor. Micro-computers for single-purpose applicationsare programmed in the assembler languagespecific to the processor.

For more complex systems and programs,high-level programming languages such asC are needed, as otherwise it would be im-possible to keep extensive programs man-ageable and free of errors. Such languagesrequire sophisticated translation programs(compilers) which convert the text of thehigh-level language into a form that can beprocessed by the processor.

26 Electronic components Semiconductor components

Robert Bosch GmbH

Semiconductor memoriesApplicationsMemories are used to store large volumes of Digital signals representing data

(I/O data, statuses, intermediate resultsinvolving frequent and rapid reading andwriting),

Program code (usually permanentlystored), and

Constants (permanently stored)

Storage involves recording (writing), permanent retention (actual storage), and location and retrieval (reading) of infor-

mation.

Memories make use of physical effects thatmake two different statuses clearly distin-guishable and easy to generate and identify.The advantage of semiconductor memorieslies in their technological compatibility withthe modules used in other parts of a com-puter, and thereby in new opportunities forfunctional integration.

Classification of semiconductor memoriesDigital signals are stored by utilising thestatus alternatives conducting/not conduct-ing or charged/uncharged. The most impor-tant terms are explained below according totheir standardised definitions, where applic-able, or their most common usage (seeFig. 12 for overview).

Random-access memory (RAM)Random-access memory or RAM is a short-term memory that allows direct access toany storage location. Information can beread/written from/to the memory any num-ber of times. However, it must exist in bi-nary form, i.e. encoded as a series of yes orno statuses (logical 1 or logical 0). Sucha yes or no unit of information is called abit (binary digit). Random-access memories,like read-only memories, are organised on abit or word basis depending on the particu-lar application. A word is a group of bitsthat can be processed as a single unit. Theword length is equal to the number of bitsprocessed as a single unit. On microcompu-ters, word lengths of 4, 8, 16, 32 and 64 bitsare common. 8 bits make up one byte, i.e.1 byte = 8 bits.

Electronic components Semiconductor components 27

Programmable on a program-ming device

Programmablein the circuit

Semiconductor memories

Non-volatile memories Volatile memories

Factory programmed User programmed

Staticmemories

Dynamicmemories

Read-only UVerasable

ROMRead-only

memory

PROMProgram-

mable read-only memory

EPROMErasablePROM

FlashEEPROM

EEPROMElectricallyerasablePROM

SRAMStaticRAM

DRAMDynamic

RAM

Electrically erasable

Classification of semiconductor memories12

U

AE

0465

-1E

Robert Bosch GmbH

Memories can be organised on the basis of avariety of word lengths. The way the mem-ory is organised is generally indicated by amultiplication sign, e.g. an 8 M 8 RAM or8-megabyte RAM 8 million bytes = 64 million bits = 64 Mbit

That means that the memory has eight datainputs and eight data outputs at which theeight bits of one of the eight million storeddata words are simultaneously present. Itsword length is thus 8 bits. Since all memoryspecifications are based on the binary sys-tem, the exact number of bits is as follows:64 Mbit = 8 Mbyte 8 bit = 223 8 bit = 226 bit = 67,108,864 bits

Static RAM (SRAM)Static RAMs use bistable switching elementsas the data storage cells. Their function issimilar to that of a flip-flop, a simple circuit

with two transistors, of which either the one(logical 1) or the other (logical 0) con-ducts at any one time. In SRAM, the infor-mation remains stored until the storage cellconcerned is addressed and overwritten, orthe power supply is switched off.

Static RAMs are currently available insizes up to 16 Mbit. Although considerablymore expensive than dynamic RAMs, be-cause they can be written into and read sig-nificantly faster they are used as high-speedaccess computing memories (cache mem-ory) for the CPUs in computers. Typicalread-write times are currently less than30 ns.

Dynamic RAM (DRAM)The information in a DRAM is stored as anelectrical charge in the gate capacitor of aCMOS transistor which either passes (logi-cal 1) or does not pass (logical 0) a cur-rent as a result. As such capacitors are sus-ceptible to leakage, the charge is graduallylost. In order that the information (charge)is retained, the memory has to be refreshedat regular intervals (every few ms).

28 Electronic components Semiconductor components

5,00075,000

400

120

305 0.5 0.13 0.06 0.03

1973 1977 1981 1984 1987 1990 1995 1998 2001 2005

Car TVHouse Shirt PizzaBicycle1 Local phone

call

1 Stick ofchewing

gum

1 Paper clipStamp

CHEW

ING G

UM

Growth in productivity of dynamic RAM (DRAM) in hindsight. Cost of 1 Mbit of memory capacity in Euro ()

13

U

AE

0773

E

Robert Bosch GmbH

DRAMs are currently available commer-cially in sizes up to 256 Mbit (on one chip),and in laboratory conditions up to 1 Gbit(gigabit = 109 bit). Since every storage cellconsists of a transistor and a capacitor, thisrequires more than a billion transistors on asingle chip.

DRAMs are volatile memories. Thismeans that they lose their information whenpower is switched off.

DRAMs are very widely used nowadays assystem memories for all types of computer.The fall in price for 1 Mbit of memory overthe last few decades has been so dramaticthat ample memory can now be fitted in anyPC without substantially affecting its price(Fig. 13).

Read-only memory Read-only memory (ROM) is permanent-storage memory that allows any memorylocation to be accessed directly but as thename indicates allows the informationonly to be read and not altered.

A ROM is a non-volatile memory, i.e. theinformation it contains is retained evenwhen the power supply is switched off. It isusually used to store program codes (controlprograms) and fixed data (function tables,encoding rules, character generators, enginecharacteristic-data maps) that need to be re-trievable at any time. The information maybe indelibly entered in the memory either bythe manufacturer (in one of the last stages of production) or by the user by appropriateprogramming of specially prepared memo-ries (PROMs or programmable ROMs).

Erasable ROM There are also ROMs whose contents can beerased and reprogrammed as outlinedbelow.

EPROM (Erasable PROM)This type of erasable read-only memory canhave its contents completely wiped by irra-diation with UV light and can then be re-programmed. Such reprogramming is only

possible with special equipment at relativelyhigh cost.

EEPROM (Electrical EPROM) EEPROM is also referred to as E2PROM (Esquared PROM). This type of erasable read-only memory can be electrically wiped andreprogrammed. The wiping and reprogram-ming operation can be performed either in aseparate unit or in situ. Every storage cell ofan EEPROM be individually overwritten.For that reason, this type of memory mod-ule can also be used as non-volatile datamemory (e.g. for status information inengine-management systems).

Flash EEPROMYet another variation of EPROM andEEPROM is the flash EEPROM. In this case,electrical flash pulses are used to erase spe-cific storage areas or the entire contents ofthe memory. The erased areas can subse-quently be reprogrammed.

The flash memory can be reprogrammedon a programming station. However, the ad-vantage of flash EEPROM is that is can alsobe reprogrammed while still inside thesealed control unit. When this is done, thememory area which contains the program-ming routines must not be erased, or alter-natively, the programming routines must betransferred to the RAM before the memoryis wiped. The microcontroller then workswith the RAM as the program memory.Flash EEPROM is used wherever relativelylarge volumes of data need to be stored butalso have to be changed from time to time(e.g. in mobile phones, digital cameras andas program memories in electronic controlunits in motor vehicles).

Electronic components Semiconductor components 29

Robert Bosch GmbH

Opto-electronic componentsPhotoresistorA photoresistor is a resistor whose resistancedecreases when exposed to light (Fig. 14).Light (photons) generates free charge carri-ers in the semiconductor as soon as the en-ergy of a photon is equal to the energy hold-ing a charge carrier within the lattice (inter-nal photoelectric effect).

Photoresistors generally consist of polycrys-talline semiconductor materials in which theeffect is particularly marked such as CdS,CdSe, PbS, PbSe, CdTe, ZnO, Se, InSb, InAs,Ge or Si.

A photoresistor is not equally sensitive toall wavelengths. Its maximum sensitivity lieswithin a narrow wavelength band that isspecific to the material used.

Photoresistors are mainly suitable for usein cameras as light meters.

PhotodiodeA photodiode is a semiconductor diodewhich utilises the depletion-layer photoelec-tric effect. Reverse voltage is applied to thep-n junction. Incident light releases elec-trons from the crystal lattice. As a result,additional free electrons and holes are pro-duced. They increase the reverse current(photoelectric current) in proportion to theintensity of the light (Fig. 15).

Photodiodes can be used in light meters,photoelectric beams and in positioning andremote control applications that use infraredlight.

Photovoltaic cellLike the photodiode, the photovoltaic cellreleases charge carriers when exposed tolight; no external voltage is applied to the p-n junction, howe