The word "Transistor" is a shortening of "Transconductance Resistor". This is a
solid state
device in which the
resistance property of its output is controlled by an input.
Principal uses of a transistor are in switching and amplification. The transistor is also the
basic building block of
integrated circuits.
A Transistor
is made from an element with semiconductor properties. Such elements are in group 4 of the
periodic table, and include germanium and silicon; such elements have a tetrahedral
crystal structure. Interestingly enough, carbon is in the same group, but although the
tetrahedral form of carbon may have semiconductor properties, this is actually diamond.
It is much cheaper to make transistors out of silicon.
History
The existing technologies before
semiconductors for signal
amplification were thermionic
valves -
vacuum tubes, and for switching were
relays. In the 1930s, Bell Lab's director
of research,
Mervin Kelly realised that the technology behind
telephone exchanges needed
improvement. The valves were troublesome because they generated huge amounts of heat, were
prone to
warm up problems, and required maintenance.
He was aware of certain chemical elements with strange electrical conductive properties,
which were later to be called
semiconductors.
After the end of the second world war, Kelly assembled a team of research scientists
with a view to replacing the thermionic valves with solid state semiconductors. Bill Shockley was selected as team leader, together with Walter Brattain and
John Bardeen.
Interestingly, to solve the problem of amplification, Shockley focused on using an electric
field to control flow of current through semiconductor material - later to be called the
field effect. He was working alone while developing this, mainly at home.
At the same time, Brattain and Bardeen were investigating the properties of impure
semiconductor material. They made an accidental discovery that when semiconductors were coated
with a different element, the result had some interesting electrical properties. They had
discovered that impurities changed the behaviour; they went on to develop this into the P-N
junction.
In 1947, they made the first working "point contact transistor", and demonstrated this to
Shockley. Shockley was furious, and shortly after this, the team split up. Both types of
transistor became very important. Brattain and Bardeen's transistor formed the basis of the
bipolar junction transistor (BJT), and Shockley's, the field effect transistor (FET: types of FET include the JFET and MOSFET).
BJT
Bipolar transistors are relatively low
impedance devices. The input
current is proportional to
the output current. The ratio of these two currents is called the
gain or H
FE.
BJTs do not have a linear characteristic, hence the incremental gain, written H
fe
(what change of output happens as a result of a change of input), is different depending on the
starting current. Fortunately, the middle of the curve is straight, which is good news if we want to
use the device for amplification. To achieve this, amplifier cicuits use
bias resistors to
provide the
quiescent current necessary to operate in the middle of the graph.
The output of a BJT is two terminals referred to as the emitter and collecter, and the third,
input terminal is called the base. The current flowing between emitter and base controls the
current flowing between emitter and collecter. The direction of the currents depends whether
the device is NPN or PNP. In a PNP device, the current flows from emitter to base, and emitter
to collecter. for NPN, the current flows the other way, i.e. towards the emitter.
FET
Field effect transistors are high
impedance devices. The input
voltage controls the output current.
The high impedance nature means that FET circuitry can be used to measure voltages without
draining current. This lends the technology to such applications as
solid state memory.
Although FETs can be used for amplification, their most common use is in
switching; FETs are the building blocks behind the
CMOS microtechnology of modern computers.
Early chips such as
TTL did use bipolar technology, but
MOSFET technology proved more suitable,
with lower
power dissipation and higher density of transistors on a chip.
The output terminals of a FET are the source and drain, and the input terminal is called the
gate.
Identifying the terminals
Transistors come in a variety of cases and
shapes. However, the case styles are numbered, such as TO92, TO18 etc. which are standard across the industry. The case is never completely symmetrical, and there will always be some
identifying mark such as a painted dot, or a nick in the plastic case.
To identify the pins, you need to look up the part number in a technical reference -
electronics vendors provide this information free of charge.
Sources
- http://www.pbs.org/transistor/album1/
- http://encyclobeamia.solarbotics.net/articles/transistor.html