Electronics/Transistors
Template:Electronics Template:Mergefrom
What is a Transistor?
We've seen "resistors" that change their resistance in response to light (photoresistors) or to mechanically turning a knob (potentiometers).
A transistor can be initially thought of as an "electronically-controlled resistor", although the name is quite misleading. Two of the pins act like a normal resistor (more or less). The other "control" pin controls the resistance "seen" between the other 2 pins.
The "control" pin is called the gate in a FET transistor (the other 2 pins are the source and drain).
The "control" pin is called the base in a BJT transistor (the other 2 pins are the emitter and the collector).
Two electrical quantities can be used to control the resistance between the two terminals - current and voltage. In an FET, the voltage at the gate controls the resistance between source and drain, while in the BJT, the current flowing into the base controls the resistance between the emitter and collector.
General
While often referred to as an amplifier, a transistor does not create a higher voltage or current of its own accord. Like any other device it obeys the Kirchoff's laws. The resistance of a transistor dynamically changes, hence the term transistor. (Actually the word came from a contraction of "transconductance varistor" or "transfer varistor", but this is a useful mnemonic for remembering its function.)
One of its popular uses is in building a signal amplifier (note that it is not like a transformer), but it can also be used as a switch. Transistors were the second generation devices, and have revolutionalized the world of electronics. They have almost completely replaced vacuum tubes and are as ubiquitous as resistors.
The transistor revolution gave way to the Integrated circuit technology around the early 50s when Jack Kilby from Texas Instruments made the first Integrated Circuit (IC) of the world. Today's transistors are mostly found inside ICs. Stand-alone transistors are used mostly only in high power applications or for power-regulation.
Both the BJT and the FET are popular today, (among the FETs, the MOSFET being the most popular form of transistor) each one having certain advantages over the other. BJT's are much faster and high current devices, while FETs are small-sized low-power devices. Attempts are underway to integrate both features on a single chip (BiCMOS technology) to produce extremely fast, dense logic ICs, that can perform complex functions. Understanding the function of a transistor is a key to understanding electronics.
Field Effect Transistor
The most common transistors today are FETs.
Field Effect Transistor: These transistors are characterized as having a conductance between source and drain dependent on the voltage applied between the gate and the source terminals. The dependance is linear if the gate to drain voltage is also high along with the gate to source voltage. It turns into a square-law relationship if the gate to drain voltage is not enough.
One of the issues that comes up in circuit design is that as chips get smaller the insulator gets thinner and it starts to look like swiss cheese. As a result the insulator starts acting like a conductor. This is known as leakage current. One solution is to replace the insulator by a material with a higher dielectric coefficent.
Two types: enhancement and depletion. Enhancement is the standard MOSFET, in which a channel must be induced by applying voltage. Depletion MOSFETs have the channel implanted, and applying voltage causes the channel to cease being conductive.
FET transistors respond to the difference in voltage bias between the gate and the source.
MOSFET (Metal-Oxide-Semiconductor FET): standard FET
JFET (Junction Fet): When voltage is applied between the source and drain current flows. Current only stops flowing when a voltage is applied to the gate.
MESFET (MEtal-Semiconductor FET): p-n junction is replaced with Scottky junction. Not made with Silicon.
HEMT (High Electron Mobility Transfer): A MESFET
PHEMT (Pseudomorphic HEMT):
Complementary Metal Oxide Semiconductor
CMOS is not a type of transistor. It is a logic family, based on MOS transistors.
Complementary Metal Oxide Semiconductor CMOS is made of two FETs blocking the positive and negative voltages. Since only one FET can be on at a time, CMOS consumes negligible power during any of the logic states. But when a transition between states occurs, power is consumed by the device. This power consumed is of two types.
Short-circuit power: For a very short duration, both transistors are on and a very huge current flows through the device for that duration. This current accounts for about 10% of the total power consumed by the CMOS.
Dynamic power: This is due to charge stored on the parasitic capacitance of the output node of the device. This parasitic capacitance depends on the wire's area, and closeness to other layers of metal in the IC, besides the relative permittivity of the quartz layer separating consecutive metal layers. It also depends (to a much smaller extent) upon the input capacitance of the next logic gate. This capacitance delays the rise in the output voltage and hence the rise or fall in the output of a gate is more like a that in a resistor-capacitor (RC) network. Thus the dynamic power consumed due to switching action in one gate is given by:
Bipolar Junction Transistor
Bipolar Junction Transistor: The current through the collector and emitter terminals of a BJT is controlled by the current entering the base. If one applies the Kirchoff's current law on the device, the current entering the device through all the terminals must add up to zero. Hence is not the same as .
Construction
A lightly doped region called base is sandwiched between two regions called the emitter and collector respectively. The collector handles large quantities of current, hence its dopant concentration is the highest. The emitter's dopant concentration is slightly lesser, but its area is larger to provide for more current than the collector. The collector region should be heavily doped because electron-hole pair recombine in that region, while the emitter is not such a region. We can have two varieties in this kind of transistor.
NPN
Here a lightly doped p-type semiconductor (semiconductor with more holes than electrons) is sandwiched between two well-doped n-type regions. It is like two pn-junctions facing away. An IEEE symbol for the npn transistor is shown here. The arrow between the base and emitter is in the same direction as current flowing between the base-emitter junction. Power dissipated in the transistor is
- , where is the voltage between the collector and the emitter and is the collector current.
PNP
Here everything is opposite that of npn. This one is more like two pn-junctions facing each other. Its IEEE symbol is shown here. Again, note the direction of the arrow.
Operation
The BJT functions
Other materials
Nearly all transistors are built into integrated circuits on slabs of high-purity silicon.
Some high-speed transistors are built of GaAs.
Some space-rated integrated circuits are silicon-on-sapphire.
Because transistors are limited by the amount of power they can dissipate, it seems that diamond (which has an unusually high thermal conductivity for such a good insulator) would be a good choice.
Some people speculate that tiny transistors can be built out of individual carbon Wikipedia:nanotubes.