22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Products

Solutions

Store

Distributors

Libstock

C ontact Us

Ente r k e ywords

Login | C art (0)

Book: PIC Microcontrollers

TOC Introduction Ch. 1 Ch. 2 Ch. 3 Ch 4. Ch. 5 Ch. 6 Ch. 7 Ch. 8 Ch. 9 App. A App. B App. C

Introduction: World of microcontrollers

Featured Development Tools

EasyPIC v7 Development
System

The situation we find ourselves today in the field of microcontrollers had its beginnings in the development of technology of integrated
circuits. This development has enabled us to store hundreds of thousands of transistors into one chip. That was a precondition for the
manufacture of microprocessors. The first computers were made by adding external peripherals such as memory, input/output lines,
timers and others to it. Further increasing of package density resulted in creating an integrated circuit which contained both processor and
peripherals. That is how the first chip containing a microcomputer later known as a microcontroller has developed.

This is how it all got started...

In the year 1969, a team of Japanese engineers from BUSICOM came to the USA with a request that a few integrated circuits for
calculators were to be designed according to their projects. The request was sent to INTEL and Marcian Hoff was in charge of the project
there. Having experience working with a computer, the PDP8, he came up with an idea to suggest fundamentally different solutions instead
of the suggested design. This solution presumed that the operation of integrated circuit was to be determined by the program stored in the
circuit itself. It meant that configuration would be simpler, but it would require far more memory than the project proposed by Japanese
engineers. After a while, even though the Japanese engineers were trying to find an easier solution, Marcians idea won and the first
microprocessor was born. A major help with turning an idea into a ready-to-use product was Federico Faggin. Nine months after hiring
him, Intel succeeded in developing such a product from its original concept. In 1971 Intel obtained the right to sell this integrated circuit.
Before that Intel bought the license from BUSICOM which had no idea what a treasure it had. During that year, a microprocessor called the
4004 appeared on the market. That was the first 4-bit microprocessor with the speed of 6000 operations per second. Not long after that, an
American company CTC requested from Intel and Texas Instruments to manufacture an 8-bit microprocessor to be applied in terminals.

EasyPIC v7 is top selling PIC Development

Board for 250 Microchip PIC MCUs in DIP
packaging. It features USB 2.0
programmer/debugger and over 17
essential modules necessary in
development. Board comes w ith
PIC18F45K22. [more info]

Free Online Books

Book: PIC Microcontrollers Programming in C

Even though CTC gave up this project, Intel and Texas Instruments kept working on the microprocessor and in April 1972 the first 8-bit
microprocessor called the 8008 appeared on the market. It was able to address 16Kb of memory, had 45 instructions and the speed of
300 000 operations per second. That microprocessor was the predecessor of all todays microprocessors. Intel kept on developing it and
in April 1974 it launched an 8-bit processor called the 8080. It was able to address 64Kb of memory, had 75 instructions and initial price
was $360.
Another American company called Motorola, quickly realized what was going on, so they launched 8-bit microprocessor 6800. Their chief
constructor was Chuck Peddle. Apart from the processor itself, Motorola was the first company that also manufactured other peripherals
such as the 6820 and 6850. At that time many companies recognized the greater importance of microprocessors and began their own
development. Chuck Peddle left Motorola to join MOS Technology and kept working intensively on developing microprocessors.
At the WESCON exhibition in the USA in 1975, a crucial event in the history of the microprocessors took place. MOS Technology announced
that it was selling processors 6501 and 6502 at $25 each, that interested customers could purchase immediately. It was such a sensation
that many thought it was a kind of fraud, considering that competing companies were selling the 8080 and 6800 at $179 each. On the first
day of the exhibit, in response to the competitor, both Motorola and Intel cut the prices of their microprocessors to $69.95. Motorola
accused MOS Technology and Chuck Peddle of plagiarizing the protected 6800. Because of that, MOS Technology gave up further
manufacture of the 6501, but kept manufacturing the 6502. It was the 8-bit microprocessor with 56 instructions and ability to directly

What are microcontrollers, anyw ay?

Electronic components built into one
single chip capable of controlling a small
submarine, a crane or an elevator
anything. Its up to you to decide w hat
you w ant them to do. You just have to
w rite a program and dump it into the
microcontroller. [more info]

address 64Kb of memory. Due to low price, 6502 became very popular so it was installed into computers such as KIM-1, Apple I, Apple II,
Atari, Commodore, Acorn, Oric, Galeb, Orao, Ultra and many others. Soon several companies began manufacturing the 6502 (Rockwell,
Sznertek, GTE, NCR, Ricoh, Commodore took over MOS Technology). In the year of its prosperity 1982, this processor was being sold at a
rate of 15 million processors per year!
Other companies did not want to give up either. Frederico Faggin left Intel and started his own company Zilog Inc. In 1976 Zilog announced
the Z80. When designing this microprocessor Faggin made a crucial decision. The 8080 had already been developed and he realized that
many would remain loyal to that processor because of the great expenditures which rewriting of all the programs would result in.
Accordingly he decided that a new processor had to be compatible with the 8080, i.e. it had to be able to perform all the programs written
for the 8080. Apart from that, many other features have been added so that the Z80 was the most powerful microprocessor at that time. It
was able to directly address 64Kb of memory, had 176 instructions, a large number of registers, a built-in option for refreshing dynamic
RAM memory, a single power supply, greater operating speed etc. The Z80 was a great success and everybody replaced the 8080 by the
Z80. Certainly the Z80 was commercially the most successful 8-bit microprocessor at that time. Besides Zilog, other new manufacturers
such as Mostek, NEC, SHARP and SGS appeared soon. The Z80 was the heart of many computers such as: Spectrum, Partner, TRS703,
Z-3 and Galaxy.
In 1976 Intel came up with an upgraded version of the 8-bit microprocessor called the 8085. However, the Z80 was so much better that
Intel lost the battle. Even though a few more microprocessors appeared later on the market (6809, 2650, SC/MP etc.), the die had already
been cast. There were no such great improvements which could make manufacturers to change their mind, so the 6502 and Z80 along
with the 6800 remained chief representatives of the 8-bit microprocessors of that time.

Microcontroller versus Microprocessor

A microcontroller differs from a microprocessor in many ways. The first and most important difference is its functionality. In order that the
microprocessor may be used, other components such as memory must be added to it. Even though the microprocessors are considered
to be powerful computing machines, their weak point is that they are not adjusted to communicating to peripheral equipment.
Simply, In order to communicate with peripheral environment, the microprocessor must use specialized circuits added as external chips.
In short microprocessors are the pure heart of the computers. This is how it was in the beginning and remains the same today.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

1/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Fig. 0-1 Microcontroller versus Microprocessor

On the other hand, the microcontroller is designed to be all of that in one. No other specialized external components are needed for its
application because all necessary circuits which otherwise belong to peripherals are already built into it. It saves the time and space
needed to design a device.

BASIC CONCEPT
Did you know that all people can be classified into one of 10 groups- those who are familiar with binary number system and those who are
not familiar with it. You dont understand? That means that you still belong to the later group. If you want to change your status read the
following text describing briefly some of the basic concepts used further in this book (just to be sure we are on the same page).

World of Numbers
Mathematics is such a good science! Everything is so logical and simple as that. The whole universe can be described with ten digits only.
But, does it really have to be like that? Do we need exactly ten digits? Of course not, it is only a matter of habit. Remember the lessons from
the school. For example, what does the number 764 mean: four units, six tens and seven hundreds. Simple! Could it be described in a bit
more complicated way? Of course it could: 4 + 60 + 700. Even more complicated? Naturally: 4*1 + 6*10 + 7*100. Could this number look a
bit more scientific? The answer is yes: 4*10^0 + 6*10^1 + 7*10^2. What does it actually mean? Why do we use exactly these numbers: 100,
101 and 102 ? Why is it always about the number 10? That is because we use ten different digits (0, 1, 2, ... 8, 9). In other words, because
we use base-10 number system, i.e. decimal number system.

Fig. 0-2 The num ber 764 represented in three different w ays

Binary Number System

What would happen if only two digits would be used- 0 and 1? Or if we would not know to determine whether something is 3 or 5 times
greater than something else? Or if we would be restricted when comparing two sizes, i.e. if we could only state that something exists (1) or
does not exist (0)? Nothing special would happen, we would keep on using numbers in the same way, but they would look a bit different.
For example: 11011010. How many pages of a book does the number 11011010 include? In order to learn that, follow the same logic like
in the previous example, but in reverse order. Bear in mind that all this is about mathematics with only two digits- 0 and 1, i.e. base-2

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

2/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

number system (binary number system).

Fig. 0-3 The num ber 218 represented in binary and decim al system

Clearly, it is the same number represented in two different ways. The only difference is in the number of digits necessary for writing some
number. One digit (2) is used to write the number 2 in decimal system, whereas two digits (1 and 0) are used to write that number in binary
system. Do you now agree that there are 10 groups of people? Welcome to the world of binary arithmetic! Do you have any idea where it is
used?
Excepting strictly controlled laboratory conditions, the most complicated electronic circuits cannot accurately determine the difference
between two sizes (two voltage values, for example) if they are too small (lower than several volts). The reasons are electrical noises and
something called the real working environment (unpredictable changes of power supply voltage, temperature changes, tolerance to
values of built in components etc.). Imagine a computer which would operate upon decimal numbers by recognizing 10 digits in the
following way: 0=0V, 1=5V, 2=10V, 3=15V, 4=20V... 9=45V !? Did anybody say batteries? A far simpler solution is the use of binary logic
where 0 indicates that there is no voltage and 1 indicates that there is voltage. It is easier to write 0 or 1 instead of there is no voltage or
there is voltage. It is called logic zero (0) and logic one (1) which electronics perfectly conforms with and easily performs all those
endlessly complex mathematical operations. It is electronics which in reality applies mathematics in which all numbers are represented by
two digits only and in which it is only important to know whether there is voltage or not. Of course, we are talking about digital electronics.

Hexadecimal Number System

At the very beginning of computer development it was realized that people had many difficulties in handling binary numbers. Because of
this, a new numbering system had to be established. This time, a number system using 16 different digits. The first ten digits are the
same as digits we are used to (0, 1, 2, 3,... 9) but there are six digits more. In order to keep from making up new symbols, the six letters of
alphabet A, B, C, D, E and F are used. A hexadecimal number system consisting of digits: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F has been
established. What is the purpose of this seemingly bizarre combination? Just look how perfectly everything fits the story about binary
numbers.

Fig. 0-4 Binary and Hexadecim al num ber

The largest number that can be represented by 4 binary digits is the number 1111. It corresponds to the number 15 in decimal system.
That number is in hexadecimal system represented by only one digit F. It is the largest onedigit number in hexadecimal system. Do you
see how skillfully it is used? The largest number written with eightdigits is at the same time the largest twodigit hexadecimal number. Bear
in mind that the computer uses 8-digit binary numbers.

BCD Code
BCD code is actually a binary code for decimal numbers only. It is used to enable electronic circuits to communicate in a decimal number
system with peripherals and in a binary system within their own world. It consists of fourdigit binary numbers which represent the first ten
digits (0, 1, 2, 3 ... 8, 9). Even though four digits can give a total of 16 possible combinations, only the first ten are used.

Number System Conversion

The binary numbering system is the most commonly used, the decimal system is the most understandable while the hexadecimal system
is somewhere between them. Therefore, it is very important to learn how to convert numbers from one numbering system to another, i.e.
how to turn a series of zeros and units into values understandable to us.

Binary to Decimal Number Conversion

Digits in a binary number have different values depending on their position in that number. Additionally, each position can contain either 1
or 0 and its value may be easily determined by its position from the right. To make the conversion of a binary number to decimal it is
necessary to multiply values with the corresponding digits (0 or 1) and add all the results. The magic of binary to decimal number
conversion works...You doubt? Look at the example:
110 = 1*2^2 + 1*2^1 + 0*2^0 = 6
It should be noted that for decimal numbers from 0 to 3 you only need two binary digits. For greater values, extra binary digits must be
added. Thus, for numbers from 0 to 7 you need three digits, for numbers from 0 to 15- four digits etc. Simply speaking, the largest binary
number consisting of n digits is obtained when the base 2 is raised by n. The result should be then subtracted by 1. For example, if n=4:
2^4 - 1 = 16 - 1 = 15
Accordingly, using 4 binary digits it is possible to represent decimal numbers from 0 to 15, including these two digits, which amounts to 16
different values in total.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

3/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Hexadecimal to Decimal Number Conversion

In order to make conversion of a hexadecimal number to decimal, each hexadecimal digit should be multiplied with the number 16 raised
by its position value. For example:

Fig. 0-5 Hexadecim al to decim al num ber conversion

Hexadecimal to Binary Number Conversion

It is not necessary to perform any calculation in order to convert hexadecimal numbers to binary numbers. Hexadecimal digits are simply
replaced by the appropriate four binary digits. Since the maximum hexadecimal digit is equivalent to decimal number 15, we need to use
four binary digits to represent one hexadecimal digit. For example:

Fig. 0-6 Hexadecim al to binary num ber conversion

Marking Numbers
The hexadecimal numbering system is along with binary and decimal
number systems considered to be the most important for us. It is easy
to make conversion of any hexadecimal number to binary and it is also
easy to remember it. However, these conversions may cause
confusion. For example, what does the statement It is necessary to
count up 110 products on assembly line actually mean? Depending on
whether it is about binary, decimal or hexadecimal, the result could be
6, 110 or 272 products, respectively! Accordingly, in order to avoid
misunderstanding, different prefixes and suffixes are directly added to
the numbers. The prefix $ or 0x as well as the suffix h marks the
numbers in hexadecimal system. For example, hexadecimal number
10AF may look as follows $10AF, 0x10AF or 10AFh. Similarly, binary
numbers usually get the suffix % or 0b, whereas decimal numbers get
the suffix D.

Comparative table below contains the values

of numbers 0-255 in three different numbering
systems.

Bit
Theory says a bit is the basic unit of information... Lets forget this dry explanation for a moment and take a look at what it is in practice. The
answer is nothing special a bit is a binary digit. Similar to decimal number system in which digits in a number do not have the same value
( for example digits in the number 444 are the same, but have different values), the significance of the bit depends on the position it has
in the binary number. Therefore, there is no point talking about units, tens etc. Instead, here it is about the zero bit (rightmost bit), first bit
(second from the right) etc. In addition, since the binary system uses two digits only (0 and 1), the value of one bit can be 0 or 1.
Dont be confused if you find some bit has value 4, 16 or 64. It means that bits values are represented in decimal system. Simply, we have
got so much accustomed to the usage of decimal numbers that these expressions became common. It would be correct to say for
example, the value of sixth bit in binary number is equivalent to decimal number 64. But we are human and habits die hard... Besides,
how would it sound number: one-onezero- one-zero...

Byte
A byte or a program word consists of eight bits grouped together. If a bit is a digit, it is logical that bytes represent numbers. All
mathematical operations can be performed upon them, like upon common decimal numbers. As is the case with digits of any other
number, byte digits do not have the same significance. The largest value has the leftmost bit called the most significant bit (MSB). The
rightmost bit has the least value and is therefore called the least significant bit (LSB). Since eight zeros and units of one byte can be
combined in 256 different ways, the largest decimal number which can be represented by one byte is 255 (one combination represents
zero).
A nibble is referred to as half a byte. Depending on which half of the byte we are talking about (left or right), there are high and low
nibbles.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

4/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Fig. 0-8 High and Low nibbles

Logic Circuits
Have you ever wondered what electronics within some digital integrated circuits, microcontrollers or processors look like? What do the
circuits performing complicated mathematical operations and making decisions look like? Do you know that their seemingly complicated
schematics comprise only a few different elements called logic circuits or logic gates?
The operation of these elements is based on the principles established by British mathematician George Boole in the middle of the 19th
century- even before the first bulb was invented! In brief, the main idea was to express logical forms through algebraic functions. Such
thinking was soon transformed into a practical product which far later evaluated in what today is known as AND, OR and NOT logic circuits.
The principle of their operation is known as Boolean algebra. As some program instructions used by the microcontroller perform the same
way as logic gates except in the form of commands, the principle of their operation will be discussed here.

AND Gate
A logic gate AND has two or more inputs and one output. Let us presume that the gate used in this case has only two inputs. A logic one
(1) will appear on its output only in case both inputs (A AND B) are driven to logic one (1).
The table shows mutual dependence between inputs and output.
When the gate has more than two inputs, the principle of operation is the
same: a logic one (1) will appear on its output only if case all inputs are
driven to logic one (1). Any other combination of input voltages will result in
a logic zero (0) at its output.

When used in a program, a logic AND operation is performed by the program

instruction, which will be discussed later. For the time being, it is enough to
remember that logic AND in a program refers to the corresponding bits of two
registers.

OR Gate
Similarly to the previous case, OR gates also have two or more inputs and
one output. A logic one (1) will appear on its output if either input (A OR B)
is driven to logic one (1). If all inputs are at logic zero (0), the output will be
driven to logic zero (0).

In a program, logic OR operation is performed between the corresponding registers

bits- the same as in logic AND operation.

NOT Gate
This logic gate has only one input and only one output. It operates in an
extremely simple way. When logic zero (0) appears on its input, a logic one
(1) appears on its output and vice versa. This means that this gate inverts
the signal by itself. It is sometimes called inverter.

If a program, logic NOT operation is performed on one byte. The result is a byte with
inverted bits. If byte is considered to be a number, the inverted value is actually a
complement of that number, i.e. the complement of a number is what is needed to
add to it to make it reach the maximal 8 bit value (255).

EXCLUSIVE OR Gate
The EXCLUSIVE OR (XOR) gate is a bit complicated comparing to other gates. It represents a combination of all the previously described
gates. A logic one (1) appears on its output only when the inputs have different logic states.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

5/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

In a program, this operation is commonly used to compare two bytes. Subtraction

may be used for the same purpose (if the result is 0, bytes are equal). The
advantage of this logic operation is that there is no danger to subtract larger number
from smaller one.

Register
A register or a memory cell is an electronic circuit which can memorize the state of one byte.

Fig. 0-17 Register

Special Function Register

In addition to the registers which do not have any special and predetermined function, every microcontroller has a number of registers
whose function is predetermined by the manufacturer. Their bits are connected (literally) to internal circuits such as timers, A/D converter,
oscillators and others, which means that they are directly in command of the operation of the microcontroller. Imagine eight switches which
are in command of some smaller circuits within the microcontroller- you are right! Special Function Registers (SFRs) do exactly that!

Fig. 0-18 Special Function Register

Input/Output Ports
In order to make the microcontroller useful, it has to be connected to additional electronics, i.e. peripherals. Each microcontroller has one
or more registers (called a port) connected to the microcontroller pins. Why input/output? Because you can change the pins function as
you wish. For example, suppose you want your device to turn three signal LEDs and simultaneously monitor the logic state of five sensors
or push buttons. Some of ports need to be configured so that there are three outputs (connected to the LEDs) and five inputs (connected to
sensors). It is simply performed by software, which means that the pins function can be changed during operation.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

6/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Fig. 0-19 Input / Output ports

One of the more important specifications of input/output (I/O) pins is the maximum current they can handle. For most microcontrollers,
current obtained from one pin is sufficient to activate an LED or other similar low-current device (10-20 mA). If the microcontroller has many
I/O pins, then the maximum current of one pin is lower. Simply put, you cannot expect all pins to give maximum current if there are more
than 80 of them on one microcontroller. Another way of putting it is that the maximum current stated in the data specifications sheet for the
microprocessor is shared across all I/O ports.
Another important pin function is that it can have pull-up resistors. These resistors connect pins to the positive power supply voltage and
their effect is visible when the pin is configured as an input connected to mechanical switch or push button. Newer versions of
microcontrollers have pull-up resistors configurable by software.
Usually, each I/O port is under control of another SFR, which means that each bit of that register determines the state of the corresponding
microcontroller pin. For example, by writing logic one (1) to one bit of that control register SFR, the appropriate port pin is automatically
configured as input. It means that voltage brought to that pin can be read as logic 0 or 1. Otherwise, by writing zero to the SFR, the
appropriate port pin is configured as an output. Its voltage (0V or 5V) corresponds to the state of the appropriate bit of the port register.

Memory Unit
Memory is part of the microcontroller used for data storage. The easiest way to explain it is to compare it with a filing cabinet with many
drawers. Suppose, the drawers are clearly marked so that it is easy to access any of them. It is easy enough to find out the contents of the
drawer by reading the label on the front of the drawer.
Each memory address corresponds to one memory location. The content
of any location becomes known by its addressing. Memory can either be
written to or read from. There are several types of memory within the
microcontroller.

Read Only Memory (ROM)

ROM (Read Only Memory) is used to permanently save the program being executed. The size of a program that can be written depends on
the size of this memory. Todays microcontrollers commonly use 16-bit addressing, which means that they are able to address up to 64 Kb
of memory, i.e. 65535 locations. As a novice, your program will rarely exceed the limit of several hundred instructions. There are several
types of ROM.
Masked ROM. Microcontrollers containing this ROM are reserved for the great manufacturers. Program is loaded into the chip by the
manufacturer. In case of large scale manufacture, the price is very low. Forget it...
One Time Programmable ROM (OTP ROM). If the microcontroller contains this memory, you can download a program into this memory,
but the process of program downloading is a one-way ticket, meaning that it can be done only once. If an error is detected after
downloading, the only thing you can do is to download the corrected program to another chip.
UV Erasable Programmable ROM (UV EPROM). Both the manufacturing process and characteristics of this memory are completely
identical to OTP ROM. However, the package of this microcontroller has a recognizable window on the upper side. It enables the surface
of the silicon chip inside to be lit by an UV lamp, which effectively erases and program from the ROM.
Installation of this window is very complicated, which normally affects the price. From our point of view, unfortunately- negative...

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

7/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Flash memory. This type of memory was invented in the 80s in the laboratories of INTEL and were represented as the successor to the UV
EPROM. Since the contents of this memory can be written and cleared practically an unlimited number of times, the microcontrollers with
Flash ROM are ideal for learning, experimentation and small-scale manufacture. Because of its popularity, the most microcontrollers are
manufactured in flash versions today. So, if you are going to buy a microcontroller, the type to look for is definitely Flash!
Random Access Memory (RAM)
Once the power supply is off the contents of RAM (Random Access Memory) is cleared. It is used for temporary storing data and
intermediate results created and used during the operation of the microcontroller. For example, if the program performs an addition (of
whatever), it is necessary to have a register representing what in everyday life is called the sum. For that purpose, one of the registers in
RAM is called the sum and used for storing results of addition.
Electrically Erasable Programmable ROM (EEPROM)
The contents of the EEPROM may be changed during operation (similar to RAM), but remains permanently saved even upon the power
supply goes off (similar to ROM). Accordingly, an EEPROM is often used to store values, created during operation, which must be
permanently saved. For example, if you design an electronic lock or an alarm, it would be great to enable the user to create and enter a
password, but useless if it is lost every time the power supply goes off. The ideal solution is the microcontroller with an embedded
EEPROM.

Interrupt
The most programs use interrupts in regular program execution. The purpose of the microcontroller is mainly to react on changes in its
surrounding. In other words, when some event takes place, the microcontroller does something... For example, when you push a button on
a remote controller, the microcontroller will register it and respond to the order by changing a channel, turn the volume up or down etc. If the
microcontroller spent most of its time endlessly a few buttons for hours or days... It would not be practical.
The microcontroller has learnt during its evolution a trick. Instead of checking each pin or bit constantly, the microcontroller delegates the
wait issue to the specialist which will react only when something attention worthy happens.
The signal which informs the central processor about such an event is called an INTERRUPT.

Central Processor Unit (CPU)

As its name suggests, this is a unit which monitors and controls all processes inside the microcontroller. It consists of several smaller
subunits, of which the most important are:

Instruction Decoder is a part of the electronics which recognizes program instructions and runs other circuits on the
basis of that. The instruction set which is different for each microcontroller family expresses the abilities of this circuit.
Arithmetical Logical Unit (ALU) performs all mathematical and logical operations upon data.
Accumulator is a SFR closely related to the operation of the ALU. It is a kind of working desk used for storing all data
upon which some operation should be performed (addition, shift/move etc.). It also stores the results ready for use in
further processing. One of the SFRs, called a Status Register (PSW), is closely related to the accumulator. It shows at
any given moment the status of a number stored in the accumulator (number is greater or less than zero etc.).

Fig. 0-22 Central Processor Unit - CPU

Bus
Physically, the bus consists of 8, 16 or more wires. There are two types of buses: the address bus and the data bus. The address bus
consists of as many lines as necessary for memory addressing. It is used to transmit the address from the CPU to the memory. The data
bus is as wide as the data, in our case it is 8 bits or wires wide. It is used to connect all circuits inside the microcontroller.

Serial Communication
Parallel connections between the microcontroller and peripherals via input/output ports is the ideal solution for shorter distances- up to
several meters. However, in other cases - when it is necessary to establish communication between two devices on longer distances it is
not possible to use a parallel connection - such a simple solution is out of question. In these situations, serial communication is the best
solution.
Today, most microcontrollers have built in several different systems for serial communication as a standard equipment. Which of these
systems will be used depends on many factors of which the most important are:

How many devices the microcontroller has to exchange data with?

How fast the data exchange has to be?

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

8/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

What is the distance between devices?
Is it necessary to send and receive data simultaneously?

One of the most important things concerning serial communication is the Protocol which

Fig. 0-23 Serial com m unication

should be strictly observed. It is a set of rules which must be applied in order that the devices can correctly interpret data they mutually
exchange. Fortunately, the microcontrollers automatically take care of this, so the work of the programmer/user is reduced to simple write
(data to be sent) and read (received data).

Baud Rate
The term Baud rate is commonly used to denote the number of bits transferred per second [bps].
It should be noted that it refers to bits, not bytes! It is usually required by the protocol that each byte is transferred along with several control
bits. It means that one byte in serial data stream may consist of 11 bits. For example, if the baud rate is 300 bps then maximum 37 and
minimum 27 bytes may be transferred per second, which depends on type of connection and protocol in use.
The most commonly used serial communication systems are:
I2C (Inter Integrated Circuit) is a system used when the distance
between the microcontrollers is short and specialized integrated
circuits of of a new generation (receiver and transmitter are
usually on the same printed circuit board). Connection is
established via two conductors- one is used for data transfer
whereas another is used for synchronization (clock signal). As
seen in figure, one device is always the master. It performs
addressing of one slave chip (subordinated) before
communication starts. In this way one microcontroller can
communicate with 112 different devices. Baud rate is usually 100 Kb/sec (standard mode) or 10 Kb/sec (slow baud rate mode). Systems
with the baud rate of 3.4 Mb/sec have recently appeared. The distance between devices which communicate via an inter-integrated circuit
bus is limited to several meters.
SPI (Serial Peripheral Interface Bus) is a system for serial
communication which uses up to four conductors (usually three)one for data receiving, one for data sending, one for
synchronization and one (alternatively) for selecting the device to
communicate with. It is full duplex connection, which means that
data is sent and received simultaneously. The maximum baud
rate is higher than in I2C connection.

UART (Universal Asynchronous Receiver/Transmitter)

This connection is asynchronous, which means that a special line for clock signal transmission is not used. In some situations this
feature is crucial (for example, radio connection or infrared waves remote control). Since only one communication line is used, both
receiver and transmitter operate at the same predefined rate in order to maintain necessary synchronization. This is a very simple way of
transferring data since it basically represents conversion of 8-bit data from parallel to serial format. Baud rate is not high up to 1 Mbit/sec.

Oscillator
Even pulses coming from the oscillator enable harmonic and synchronous operation of all circuits of the microcontroller. The oscillator
module is usually configured to use quartz crystal or ceramic resonator for frequency stabilization. Furthermore, it can also operate without
elements for frequency stabilization (like RC oscillator). It is important to say that instructions are not executed at the rate imposed by the
oscillator itself, but several times slower. It happens because each instruction is executed in several steps. In some microcontrollers, the
same number of cycles is needed to execute any instruction, while in others, the execution time is not the same for all instructions.
Accordingly, if the system uses quartz crystal with a frequency of 20 Mhz, execution time of an instruction is not 50nS, but 200, 400 or 800
nS, depending on the type of Microcontroller Unit (MCU)!

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

9/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Power supply circuit

There are two things worth attention concerning the microcontroller power supply circuit:
Brown-out is a potentially dangerous state which occurs at the moment the microcontroller is being turned off or in situations when power
supply voltage drops to the limit due to electric noise. As the microcontroller consists of several circuits which have different operating
voltage levels, this state can cause its out-of-control performance. In order to prevent it, the microcontroller usually has built-in circuit for
brown out reset. This circuit immediately resets the whole electronics when the voltage level drops below the limit.
Reset pin is usually marked as MCLR (Master Clear Reset) and serves for external reset of the microcontroller by applying logic zero (0) or
one (1), depending on type of the microcontroller. In case the brown out circuit is not built in, a simple external circuit for brown out reset
can be connected to this pin.

Timers/Counters
The microcontroller oscillator uses quartz crystal for its operation. Even though it is not the simplest solution, there are many reasons to
use it. Namely, the frequency of such oscillator is precisely defined and very stable, the pulses it generates are always of the

Fig. 0-27 Tim ers/Counters

same width, which makes them ideal for time measurement. Such oscillators are used in quartz watches. If it is necessary to measure
time between two events, it is sufficient to count pulses coming from this oscillator. That is exactly what the timer does.
Most programs use these miniature electronic stopwatches. These are commonly 8- or 16-bit SFRs and their content is automatically
incremented by each coming pulse. Once a register is completely loaded - an interrupt is generated!
If the timer registers use an internal quartz oscillator for their operation then it is possible to measure time between two events (if the
register value is T1 at the moment measurement has started, and T2 at the moment it has finished, then the elapsed time is equal to the
result of subtraction T2-T1). If the registers use pulses coming from external source then such a timer is turned into a counter.
This is only a simple explanation of the operation itself.

How does a timer operate?

In practice pulses coming from the quartz oscillator are once per each machine cycle directly or via a prescaler brought to the circuit which
increments the number in the timer register. If one instruction (one machine cycle) lasts for four quartz oscillator periods then, by
embedding quartz with the frequency of 4MHz, this number will be changed a million times per second (each microsecond).

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

10/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

Fig. 0-28 Tim er Operation

It is easy to measure short time intervals (up to 256 microseconds) in the way described above because it is the largest number that one
register can contain. This obvious disadvantage may be easily overcome in several ways by using a slower oscillator, registers with more
bits, a prescaler or interrupts. The first two solutions have some weaknesses so it is preferable to use prescalers or interupts.

Using prescaler in timer operating

A prescaler is an electronic device used to reduce a frequency by a pre-determined factor. Meaning that in order to generate one pulse on
its output, it is necessary to bring 1, 2 , 4 or more pulses to its input. One such circuit is built in the microcontroller and its division rate can
be changed from within the program. It is used when it is necessary to measure longer periods of time.
One prescaler is usually shared by timer and watch-dog timer, which means that it cannot be used by both of them simultaneously.

Fig. 0-29 Using prescaler in tim er operating

Using the interrupt in timer operation

If the timer register consists of 8 bits, the largest number that can be written to it is 255 (for 16-bit registers it is the number 65.535). If this
number is exceeded, the timer will be automatically reset and counting will start from zero again. This condition is called overflow. If
enabled from within the program, such overflow can cause an interrupt, which gives completely new possibilities. For example, the state of
registers used for counting seconds, minutes or days can be changed in an interrupt routine. The whole process (except interrupt routine)
is automatically performed in the background, which enables the main circuits of the microcontroller to perform other operations.

Fig. 0-30 Using the interrupt in tim er operation

This figure illustrates the use of the interrupt in timer operation. Delays of arbitrary duration with minimal interference by the main program
execution can be easily obtained by assigning a prescaler to the timer.

Counters
If a timer is supplying pulses into the microcontroller input pin then it turns into a counter. Clearly, It is the same electronic circuit. The only
difference is that in this case pulses to be counted come through the ports and their duration (width) is mostly not defined. This is why they
cannot be used for time measurement, but can be used to measure anything else: products on an assembly line, number of axis rotation,
passengers etc. (depending on sensor in use).

Watchdog Timer

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

11/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

The Watchdog Timer is a timer connected to a completely separate RC oscillator within the microcontroller.
If the watchdog timer is enabled, every time it counts up to the program end, the microcontroller reset occurs and program execution starts
from the first instruction. The point is to prevent this from happening by using a specific command. The whole idea is based on the fact that
every program is executed in several longer or shorter loops.
If instructions which reset the watchdog timer are set at the appropriate program locations, besides commands being regularly executed,
then the operation of the watchdog timer will not affect program execution. If for any reason (usually electrical noises in industry), the
program counter gets stuck on some memory location from which there is no return, the watchdog will not be cleared and the registers
value being constantly incremented will reach the maximum et voila! Reset occurs!

Fig. 0-31 Watchdog Tim er

A/D Converter
External signals are usually fundamentally different from those the microcontroller understands (Ones and Zeros), so that they have to be
converted in order for the microcontroller to understand them. An analogue to digital converter is an electronic circuit which converts
continuous signals to discrete digital numbers. This module is therefore used to convert some analogue value into binary number and
forwards it to the CPU for further processing. In other words, this module is used for input pin voltage measurement (analogue value). The
result of measurement is a number (digital value) used and processed later in the program.

Fig. 0-32 A/D Converter

Internal Architecture
All upgraded microcontrollers use one of two basic design models called Harvard and von-Neumann architecture.
Briefly, they are two different ways of data exchange between CPU and memory.
von-Neumann Architecture
Microcontrollers using this architecture have only one memory block and one 8-bit data bus. As all data are exchanged by using these 8
lines, this bus is overloaded and communication itself is very slow and inefficient. The CPU can either read an instruction or read/write
data from/to the memory. Both cannot occur at the same time since the instructions and data use the same bus system. For example, if
some program line says that RAM memory register called SUM should be incremented by one (instruction: incf SUM), the microcontroller

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

12/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

will do the following:

1. Read the part of the program instruction specifying WHAT should be done (in this very case it is the incf
instruction for increment).
2. Read further the same instruction specifying upon WHICH data it should be performed (in this very case it is the
SUM register).
3. After being incremented, the contents of this register should be written to the register from which it was read
(SUM register address).
The same data bus is used for all these intermediate operations.
Harvard Architecture
Microcontrollers using this architecture have two different data
buses. One is 8 bits wide and connects CPU to RAM. Another
consists of several lines (12, 14 or 16) and connects CPU to
ROM. Accordingly, the CPU can read an instruction and perform a
data memory access at the same time. Since all RAM memory
registers are 8 bits wide, all data within the microcontroller are
exchanged in the same such format. Additionally, during program
writing, only 8 bits data are considered. In other words, all you can
ever change from within the program and all you can affect will be
8 bits wide. A program written for some of these microcontrollers
will be stored in the microcontroller internal ROM upon having
being compiled into machine language. However, these memory
locations do not have 8, but 12, 14 or 16 bits. The rest of bits- 4, 6
or 8- represents the instruction itself specifying to the CPU what to
do with the 8-bit data.
The advantages of such design are the following:

All data in a program is one byte (8 bit) wide. As the data bus used for program reading has several lines (12, 14 or 16),
both instructions and data can be read simultaneously by using these spare bits. Therefore, all instructions are
executed in only one instruction cycle. The only exception is jump instruction which is executed in two cycles.
Owing to the fact that a program (ROM) and temporary data (RAM) are separate, the CPU can execute two
instructions simultaneously. Simply, while RAM read or write is in progress (the end of one instruction), the next
program instruction is being read via another bus.
When using microcontrollers with von-Neumann architecture one never knows how much memory is to be occupied by
some program. Basically, each program instruction occupies two memory locations (one contains information on
WHAT should be done, whereas another contains information upon WHICH data it should be done). However, it is not a
hard and fast rule, but the most common case. In microcontrollers with Harvard architecture, the program bus is wider
than one byte, which allows each program word to consist of instruction and data. In other words: one program wordone instruction.
INSTRUCTION SET
Instructions that can be understood by the microcontroller are known as an instruction set. When you write a
program in assembly language, you actually tell a story by specifying instructions in the order they should be
executed. The main restriction in this process is the number of available instructions. The manufacturers stick to
one of the two following strategies:

RISC (Reduced Instruction Set Computer)

In this case, the microcontroller recognizes and executes only basic operations (addition, subtraction, copying etc.). All other more
complicated operations are performed by combining these (for example, multiplication is performed by performing successive addition).
The constrains are obvious (try by using only a few words, to explain to someone how to reach the airport in some other city). However,
there are also some great advantages. First of all, this language is easy to learn. Besides, the microcontroller is very fast so that it is not
possible to see all the arithmetic acrobatics it performs. The user can only see the final result of all those operations. At last, it is not so
difficult to explain where the airport is if you use the right words. For example: left, right, kilometers etc.
CISC (Complex Instruction Set Computer)
CISC is the opposite of RISC! Microcontrollers designed to recognize more than 200 different instructions can do much and are very fast.
However, one needs to understand how to take all that such a rich language offers, which is not at all easy...

How to make the right choice

Ok, you are the beginner and you have made a decision to go on an adventure of working with the microcontrollers. Congratulations on
your choice! However, it is not as easy to choose the right microcontroller as it may seem. The problem is not a limited range of devices,

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

13/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

but the opposite!

Before you start designing some device based on the microcontroller, think of the following: how many input/output lines will I need for
operation? Should it perform some other operations than to simply turn relays on/off? Does it need some specialized module such as
serial communication, A/D converter etc. When you create a clear picture of what you need, the selection range is considerably reduced,
then it is time to think of price. Is your plan to have several same devices? Several hundred? A million? Anyway, you get the point...
If you think of all these things for the very first time then everything seems a bit confusing. For that reason, go step by step. First of all, select
the manufacturer, i.e. the family of the microcontrollers you can easily obtain. After that, study one particular model. Learn as much as you
need, do not go into details. Solve a specific problem and something incredible will happen- you will be able to handle any model
belonging to that family.
Remember learning to ride a bicycle: after several unavoidable bruises at the beginning, you will manage to keep balance and will be able
to easily ride any other bicycle. And of course, you will never forget the skill in programming just as you will never forget riding bicycles!

PIC microcontrollers
PIC microcontrollers designed by Microchip Technology are likely the right choice for you if you are the beginner. Here is why...
The real name of this microcontroller is PICmicro (Peripheral Interface Controller), but it is better known as PIC. Its first ancestor was
designed in 1975 by General Instruments. This chip called PIC1650 was meant for totally different purposes. About ten years later, by
adding EEPROM memory, this circuit was transformed into a real PIC microcontroller. Nowadays, Microchip Technology announces a
manufacturing of the 5 billionth sample...
In order that you can better understand the reasons for its popularity, we will briefly describe several important things.
RAM
[bytes]

Clock
Freq.
[MHz]

ROM
[Kbytes]

PIC 10FXXX

0.375 0.75

16 - 24

6-8

4-8

0-2

0-1

PIC 12FXXX

0.75 - 1.5

25 - 38

4-8

0-3

PIC 16FXXX

0.75 - 3

25 - 134

14 44

20

0-3

PIC 16HVXXX

1.5

25

18 20

20

Pins

A/D
Inputs

Resolution
of A/D
Converter

Family

Comparators

8/16
bit
Timers

Serial
Comm.

PWM
Outputs

Others

1x 8

0-1

1x 8

EEPROM

0-2

1x 8

EEPROM

1x 8

Vdd =
15V

Base-Line 8 - bit architecture, 12-bit Instruction Word Length

Mid-Range 8 - bit architecture, 14-bit Instruction World Length

PIC 12FXXX

1.75 - 3.5

64 - 128

20

0-4

10

1-2x
81x
16

0-1

EEPROM

PIC 12HVXXX

1.75

64

20

0-4

10

1-2x
81x
16

0-1

PIC 16FXXX

1.75 - 14

64 - 368

14 64

20

0 - 13

8 or 10

0-2

1-2x
81x
16

USART
I2C SPI

0-3

PIC 16HVXXX

1.75 - 3.5

64 - 128

14 20

20

0 - 12

10

2x 81
x 16

USART
I2C SPI

High-End 8 - bit architecture, 16-bit Instruction Word Length

PIC 18FXXX

4 - 128

256 3936

18 80

32 48

4 - 16

10 or 12

0-3

0-2x
82-3
x 16

USB2.0
C AN2.0
USART
I2C SPI

0-5

PIC 18FXXJXX

8 - 128

1024 3936

28 100

40 48

10 - 16

10

0-2x
82-3
x 16

USB2.0
USART
Ethernet
I2C SPI

2-5

PIC 18FXXKXX

8 - 64

768 3936

28 44

64

10 - 13

10

1x 83
x 16

USART
I2C SPI

All PIC microcontrollers use harvard architecture, which means that their program memory is connected to CPU via more than 8 lines.
Depending on the bus width, there are 12-, 14- and 16-bit microcontrollers. The table above shows the main features of these three
categories.
As seen in the table on the previous page, excepting 16-bit monsters- PIC 24FXXX and PIC 24HXXX- all PIC microcontrollers have 8-bit
harvard architecture and belong to one out of three large groups. Therefore, depending on the size of a program word there are first,
second and third category, i.e. 12-, 14- or 16-bit microcontrollers. Having similar 8- bit core, all of them use the same instruction set and
the basic hardware skeleton connected to more or less peripheral units.
In order to avoid tedious explanations and endless story about the useful features of different microcontrollers, this book describes the
operation of one particular model belonging to high middle class. It is about PIC16F887- powerful enough to be worth attention and
simple enough to be easily presented to everybody.
table of contents | next chapter

MikroElektronika Embedded Solutions

PIC Solution

dsPIC Solution

Follow us on
PIC32 Solution

Add-on boards

Additional Software

PIC De ve lopm e nt Boards

dsPIC De ve lopm e nt Boards

PIC 32 De ve lopm e nt Boards

C lick Boards

Visual TFT

PIC C om pile rs

dsPIC C om pile rs

PIC 32 C om pile rs

m ik rom e dia shie lds

Visual GLC D

PIC Program m e rs/De bugge rs

dsPIC Program m e rs/De bugge rs

PIC 32 Program m e rs/De bugge rs

C om m unication boards

Pack age Manage r

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

14/15

22/01/13

Introduction: World of microcontrollers - Book: PIC Microcontrollers

PIC Kits

dsPIC Kits

PIC Book s

dsPIC Book s

AVR Solution

PIC 32 Kits

Storage boards

8051 Solution

ARM Solution

8051 De ve lopm e nt Boards

AVR De ve lopm e nt Boards

AR M De ve lopm e nt Boards

8051 C om pile rs

AVR C om pile rs

AR M C om pile rs

8051 Program m e rs/De bugge rs

AVR Program m e rs/De bugge rs

AR M Program m e rs/De bugge rs

8051 Book s

AVR Kits

AR M Kits

8051 Kits

Support

Forum

m ik roBUS

Le ts m ak e

Pre ss

GLC D Font C re ator

R e al Tim e C lock boards

Display boards
Me asure m e nt boards
Audio & Voice boards
Powe r Supply boards
GPS boards
GSM/GPR S boards

Le gal Inform ation

Product Archive

About Us

C opyright 1998 - 2013. MikroElektronika. All rights reserved. All trade and/or services marks mentioned are the property of their respective owners.

www.mikroe.com/chapters/view/1/introduction-world-of-microcontrollers/

15/15