Write on the board: semiconductor; p, n, u, m, ---k, M, G,
Draw on board: box with "Input", "Output", "Control". Lines.
Plug-in the LED demo box, with the whole variety of lights (requires a 12-vdc power supply).
2.0 Review
a. p pico 10-12
b. n nano 10-9
c. u micro 10-6
d. m milli 10-3
e. 1 (one) 100
f. k kilo 103
g. M mega 106
h. G giga 109
i. T tera 1012
3.0 Theory and demonstration
3.1 Today, we're going to learn about semiconductor devices, such as diodes and transistors, and learn how they work. We're also going to see how the electrostatics of the 18th century are related to the vacuum tubes and transistors.
3.2 Semiconductor material: We know that diodes and transistors are made from semiconductor material, which allows only a very small current flow. We call pure semiconductor material "intrinsic", because there's nothing added to it However -- and here's the first secret of transistors -- we add a pinch of an impurity, about one sprinkle of a big salt shaker to a railroad car full. This is called "doping" of the intrinsic material. The impurity is either "N" type material, which has a few free electrons, or "P" type material, which is lacking a few free electrons. These free electrons can wander through the atomic stew pot because they're not bound tightly as the intrinsic electrons are.
3.3 Movement of charges. There are two forces that cause the electrons to move. The first force is the random motion that all electrons go through from the presence of heat. This is called "diffusion" [non-directed wandering]. The other motion is caused by one that you know well: it's the repulsion force from two like bodies getting near to each other. That is called "drift". [compare electric drift to that of a river. [Draw a circuit with P and N impurities, and show free charges. Connect a battery to show energy source]
3.4 Semiconductor device construction. [draw schematic of a PN junction with a battery bias]. Note that the P section has an excess of "holes". What are those? [a hole is a place in the electronic structure of a solid where an electron ought to be but isn't! It acts like a positive charge, attracting an electron. Think of it as a bubble that floats through the semiconductor structure]. Now note the N section; it has a surplus of electrons.
3.5 Semiconductor diode. This is simply a PN junction (show the schematic symbol]. It passes current in the forward direction. Which way is that? [In the direction of the arrow]. Without a battery no current flows. Why? [because we need a little voltage to get the charges moving. The current flow in the diode is a function of junction properties of which the ohmic resistance is only one of many]..
3.5.1 With forward voltage, the current through the NP junction increases. Why? [The battery + terminal repels the holes in the P-type material, driving them toward th junction. Meanwhile, the negative battery terminal repels the electrons, driving them toward the junction also. A small voltage, typically about 0.2 to 0.5 volts depending upon the semiconductor material, produces a rapid increase in forward current.
3.6 Transistor operation. The transistor consists of two junctions within one solid mass. For the NPN transistor [draw the block NPN, and label the three terminals], there is an NP junction from the emitter to the base. The other side of the base has a PN junction for the collector. There is electron flow from emitter into base, and then through the base into the collector. With a little more electron flow from emitter to base, there is a lot more current flow to the collector. Here are the details on this flow: [discuss electrons moving by drift, the result of a + voltage from emitter to base. Show charges in P, N. P.
3.6.1 Current Travel. Current travel through the transistor [draw a simple transistor circuit, with a bias pot connected to the base for adjustment]. There are three different zones: E, B, C. Electrons are sucked into the emitter -- if there is no voltage on the base [turn pot in schematic to zero volts], the charges won't go in. Why not? So, with bias on the base, the charges flow in the emitter.. They move through the emitter primarily by drift [like a boat going down-river because of current flow], but also by diffusion [helter-skelter]. When they get to the NP emitter-base junction, they are pulled across by the + field and by attraction to the holes [+ charges]. They move through the base region by drift and diffusion, until they come under the influence of the collector N region. They streak through the collector by strong drift, heading for the output terminal where the + battery is calling them. Now, what do we have for the result? Current into the emitter, some flows down and out of the base lead and the rest goes to the collector.
3.6.2 Minority Current flow. The junction transistor operates because of minority current flow through the base region. Before we talk about this, let's consider the meaning of "majority" and "minority." When the Mayflower dropped anchor in the bay just off the coast of North America, there were about 120 pilgrims who came ashore. They were met by 5000 (well, maybe less) Indians. Who were in the minority? [the pilgrims]. What if we have 27 Indians and 27 Pilgrims -- or maybe 27 electrons and 27 holes? This ratio doesn't have a majority/minority. When talking about semiconductors, we call this "intrinsic."; the + and the - cancel each other out. However, the junction transistor works because of minority flow…..So, where is there a minority current flow in an NPN transistor? [show charges in each section; note that minority flow occurs when the emitter charges get into the base region]
3.6.3 Control power -- Here's how a small current controls a larger one. A large quantity of electrons moving from the emitter to the collector has to go through the base; however, only a small amount of the current goes down the base lead, and a lot goes to the collector. This is a big improvement in controls. You can see that a small amount of current controls a very large amount. This is called the
b ("beta") of the transistor. A poor transistor might have a beta of 10; a good transistor will have a beta of 100 to 300 or more. This depends on the geometry -- narrow base, more electrons moving to the collector by diffusion.3.6.4 Control similar to water flow. When you're dealing with electrical current or water, remember: If you keep putting in a current -- either water or electric -- it has to come out someplace. [draw a picture of an electric junction. How much control does a transistor have? Let's look at the water analogy: you have a hose with just a dribble of water coming out. Half the water is leaking out of the back of the faucet, and the other half comes out of the faucet output. Is this good control? [No -- you use half the water in maintaining control. Now try it again with a good controller. The nozzle on your garden hose is very good; just a little water leaks out of the trigger on the nozzle. Where does the rest of the water go? [out the output terminal -- no place else to go!] Same as with the transistor -- a good unit will send almost all of the input current to the output.
3.6.5 An efficient transistor control. Once the charges go into the minority carrier region [the base region], they can do three things: go straight to the collector junction; go to the base lead; recombine in the base with holes and just cancel out. To make the transistor efficient, we need to make the base just as thin as possible, so that the charges spend a minimum time in the base.
3.7 FET (Field Effect Transistor) Introduction
3.7.1 Difference between Junction and Field Effect Transistor. There's one basic difference between the junction transistor and the FET: The junction transistor works on a minority carrier; the FET operates with majority carriers. Here's the thing to remember: the FET is like a big resistor that can be varied with input voltage. You may remember in Electrostatics, that "like charges repel, unlike charges attract." The vacuum tube control is built on this principle. The control grid repels the electrons more or less, depending on the control signal. It can turn the tube on full, or cut off all electron flow.
3.7.2 FET History. Back in 1935, when I was about your age, a German by the name of Oskar Heil, patented an FET based on this principle. It didn't work! At Bell Laboratories, there was a team trying to invent a solid-state replacement for the vacuum tube. The leader - boss - of the inventor team, whose name was Bill Shockley, felt that the FET had possibilities. He reasoned that a strong electrical field could regulate the flow of electricity in a semiconductor. He tried to build one, but it didn't work. Three years later, Brattain and Bardeen built the first working transistor, a germanium point-contact unit. Two years after that, Shockley then designed the junction (sandwich) transistor, which was manufactured for several years. But in 1960 Bell scientist John Atalla developed a new design based on Shockley's original field-effect theories. By the late 1960s, manufacturers converted from junction type integrated circuits to field effect devices. Today, most transistors are field-effect transistors. We are now using trillions of them. Each computer has a microprocessor using at least 5 million (!) FETs, and there are millions of computers. If that's not enough, most of the logic chips used for all digital work are FET-based.
3.7.3 Enhancement Mode FETs: [draw figure 4-3.2.3] The main power path -- the input and the output circuit -- can be either a P-type, or an N-type semiconductor bar. The control element is formed alongside of the main path. With no voltage applied to the control element, it is inactive. Nothing happens. The main path is through the two diodes. However, one of those diodes is backwards for any current flow. Result? No current flows through from input to output. Now, let's put some voltage on the gate. Electrons come streaming through the P layer from the negative terminal of the battery. The gate keeps pulling electrons through the P layer until finally, it has filled up all the holes in the P layer, and then added more electrons. The P layer now has more electrons than holes -- and it changes from P to N. This makes a continuous bridge from N1 to N2, all N-layer. The FET is turned on and the current now flows very easily
3.7.4 Types of FETs: Earlier transistors controlled the current flow with a P-N junction -- a diode . Later on, the P-N junction was replaced with metal-oxide, which turned out to be easier to mass-produce in microchips. Today's transistors are "MOS-FETs", or Metal Oxide Semiconductor Field Effect Transistors. They were developed by Bell Labs, Fairchild Semiconductor, and hundreds of Silicon Valley, Japanese and other electronics companies.
3.7.5 CMOS (Complementary MOSFET) transistors: CMOS units are widely used in all manner of application. A CMOS transistor consists of two MOSFETs connected in series, to make a very fast, positive switch..[draw simplified schematic of this, with an N-MOS on top, and a P-MOS on the bottom.]
3.7.6 CMOS versus Vacuum Tubes: The CMOS control circuit does not draw any power. In that sense, it is much like the grid of a vacuum tube. Old-timers who had been designing with vacuum tubes really liked CMOS. It was so similar in usage.
There are some other classes of semiconductor products that you should know about. They're used for many products in your home and they are not transistors, but they are similar to transistors.
3.8 SCRs (Silicon-Controlled Rectifiers). The SCR is one of the few basic semiconductor devices used in extremely high-power work. An example of that is the "North South Intertie", a dc (direct current) connection, at about 300-kVdc between the Los Angeles main power lines and the Bay region main power lines. SCRs also are used as control elements for ac motors, high-power ac heaters and lamp controls.
3.9 Triac This is a bi-directional triode varistor used for ac control of small motors or appliances. Probably every home dimmer switch in the world uses a triac as its control element. Small kitchen blenders, egg-beaters, choppers, Cuisinarts etc. each use a triac for control. They are preferred to SCRs because a triac can control both the positive and the negative phase of an ac wave; this same function requires the use of two SCRs, one for each phase.
3.10 Light-sensitive diodes The LSD (light-sensitive diode) serves as an electronic "eye", watching a process. It detects a bottle on the Coca-Cola bottling machine; it watches for a light bulb going through the sealing machine; it counts each customer walking through the store doorway; it detects when a security door has been opened, and then initiates an alarm bell.
3.18 Opto-Isolator The opto-isolator can sense an action in one piece of electronic apparatus, and transmit a light signal to another apparatus. With this action, it can communicate between a radar transmitter floating deck operating at 50 or 100 kVdc, down to ground level at the operator's desk. For these extremely high voltage, an opto-isolator is used in conjunction with a fiber-optic transmission cable. The cable is made of quartz and plastic, and does not conduct electricity.
4.0 Workshop
Page last updated on January 28, 2002