|Home | Articles | Forum | Glossary | Books|
Before delving into the action within a semiconductor, and the operation of circuits employing transistors, it will be necessary to review some basic definitions, and the physical principles which support them. The well-known electron-drift process within electrical conductors is a logical starting point.
THE ELECTRON-DRIFT PROCESS
Fig. 1 shows a simplified germanium atom. It consists of a positively charged nucleus and four planetary electrons revolving in orbit around it. Each electron carries a negative charge of electricity, so that they together will just neutralize the positive charge of the nucleus, making the entire atom electrically neutral.
(This atomic picture is simplified, in that the germanium atom has a total of 32 planetary electrons in orbit, and the central nucleus has a resulting positive charge of 32 units. Twenty eight of these electrons are so tightly bound to the nucleus, however, that they are completely unavailable for purposes of being dislodged to form an electron current. These 28 electrons are not shown in Fig. 1.) The electrons shown in the illustration are in the valence band, or ring, of the atom. Germanium is normally a good insulator and a poor conductor. This means that only with great difficulty can one of the planetary electrons be dislodged from its orbit by the normal electron drift process. In a good conductor, on the other hand, electron drift occurs quite easily. When an electric field (electric potential, or a voltage) such as from a battery is applied across the terminals of a good conductor, electrons will be driven through the conductor, from the negative to the positive end. This flow of electron is called electric cur rent, and it consists of a long series of "domino" actions between the free electrons in the material (including planetary electrons which can easily be set free). An individual electron, moving into the conductor very quickly, finds itself approaching a head on collision with another electron, perhaps one in orbit around an atomic nucleus. Since electrons carry a negative charge, they mutually repel each other, and no electron collision occurs.
Instead, this orbiting electron will be "knocked out" of its orbit, not by collision, but by the electrical repulsion from the approaching electron. Once set free in this manner, the released electron will assume the velocity and direction of the approaching electron.
The approaching electron then finds itself slowed down and in the vicinity of a positively charged atomic nucleus; as a result, it "falls into" the recently vacated orbit. Thus, one complete sequence of electron drift action has taken place-an electron has been repelled from its orbit around a nucleus, and replaced by another electron.
ACCEPTOR ATOMS AND P-TYPE SEMICONDUCTORS
Germanium, the insulator, becomes germanium, the semi conductor, with the addition of selected impurities. Two very common impurities are boron and arsenic. Fig. 2 shows the atomic realignment that occurs when a small portion of boron is added to germanium. An atom of boron has a total of three planetary electrons in its valence band. One atom of boron will "lock" itself very firmly in place with four adjoining germanium atoms. When this happens, the resulting combination needs or wants one additional electron to complete itself, and will take ("accept") it from a nearby germanium atom. This action gives it the title of an acceptor atom. Before it takes this extra electron from its neighbor, the impurity atom is electrically neutral neither positive nor negative. After it takes this electron, the acceptor becomes permanently negatively charged.
The Concept of "Positive Holes"
The adjacent germanium atom becomes positively charged after relinquishing one electron. This positively charged atom becomes what is known as a "hole," or the carrier of a positive electrical charge; and the material is called a P-type semiconductor. The hole is indicated by the arrow in Fig. 2.
If we consider only this one positive atom and the negatively charged acceptor which "stole" its electron, we might well ask why the stolen electron does not return to its original orbit.
How can both positive and negative charges exist side by side without immediately neutralizing themselves? The answer can be found in the physics of atomic structure which is beyond the scope of this guide. We must be content with the picture of the acceptor atom, taking and holding tightly the extra electron which gives it an over-all negative charge.
For every atom of impurity, there is eventually created one negatively charged acceptor atom and its positive counterpart, the germanium hole, which is deficient by one electron. Even though the germanium atom by itself will not participate in the electron drift process (it is a poor conductor), the germanium "hole" becomes a fairly good conductor of electricity. It is said that holes move freely through a semiconductor, but this statement requires some elaboration. The positively charged atoms do not move through the semiconductor. By capturing planetary electrons from nearby atomic orbits, the "hole" can be given the appearance of moving fairly rapidly through a series of germanium atoms. In reality, each such atom in turn loses an electron and becomes a hole, then in turn captures an electron from the next atom, causing it to become a hole, and so on.
Thus, the positive charge seems to move in the direction of applied positive voltage-meaning away from a plus voltage and toward a minus voltage such as the negative terminal of a battery This process of hole-current movement resembles conventional electron drift, but with the important difference that the electron drift process presupposes extra electrons traveling through the atomic structure and, in effect, "pushing" planetary electrons out of their orbits. "Hole" current seems to travel in the opposite direction through a semiconductor, by "pulling" planetary electrons out of orbit.
"Mobility" of Charge Carriers
Electron charge is concentrated at a single point, whereas the positive charge of a hole seems to be distributed somewhat if not over the entire volume of the atom, then at least over a portion of it or a portion of the orbit. This would seem to indicate that the "pulling" of electrons from orbit might be more difficult than the normal process of pushing them from orbit. That this is the case may be verified from the "mobility" of electrons and holes in germanium.
The mobility of an electric charge is a measure of the relative ease or difficulty with which the charge can be moved by an applied electric field (voltage). It is usually expressed as the "drift velocity" of the charge in centimeters per second. The mobility of free electrons in a germanium semiconductor, when the applied electric field has a strength of one volt per centimeter, is about 3,600 centimeters per second.
The mobility of positive "holes" through a germanium semi conductor for the same applied electric field of one volt per centi meter, is about 1,700 centimeters per second. Thus, the drift velocity of free electrons through a germanium semiconductor is more than twice as great as the drift velocity of holes.
DONOR ATOMS AND N-TYPE SEMICONDUCTORS
Fig. 3 shows the atomic structure that results when an impurity atom such as arsenic is added to the germanium. Arsenic is chosen as an impurity because it has five planetary electrons orbiting about the nucleus. The arsenic atom will "lock" in place with four germanium atoms; and one of the five valence electrons will be released, or "donated," after this combination occurs.
This action gives the resulting atom the name "donor atom." The extra electron is indicated by the arrow in Fig. 3. Before giving up one electron, the atom has a neutral electric charge; and afterward, it becomes a positively charged "donor" atom.
Like the negatively charged acceptor atom discussed previously, the donor atom holds on very tightly to its new electrical condition, and does not recapture the electron it released. Consequently, both the positive atom and the negative electron can exist side by side in the same portion of the semiconductor without recombining. Although the total charge of this semiconductor will be zero (since the number of electrons released equals the number of positively charged donor atoms), the semiconductor becomes known as an N-type because of the availability of negative carriers (electrons). The positively charged donor atoms do not act as current carriers, and are not "holes." We have now created the two basic types of semiconductors.
The N-type has an excess of free electrons available for carrying current: whereas the P-type has a series of positively charged germanium atoms, called holes, also available for letting electron current flow through the semiconductor.
IMPURITY CONTENT OF SEMICONDUCTORS
Very small portions of an impurity are required to convert the insulator germanium to a semiconductor germanium--about one part impurity to a million parts germanium. This is due to the conductivity of the material, which varies directly with the density of current carriers (free electrons and holes) contained therein. The number of current carriers in a particular sample will equal the number of impurity atoms.
THE JUNCTION DIODE
When an N-type semiconductor is bonded to a P-type semi-conductor, a junction diode is formed. Fig. 4 shows the junction diode under three voltage conditions.
The natural tendency for the electrical charges in the junction diode, when no external voltage is applied (Fig. 4A), is for excess electrons in the N-material to cross the junction (moving to the left in Fig. 4A) and recombine with the positive holes in the P-region. A small amount of such recombining does occur, but it is prevented from happening on a wholesale basis. As soon as a few electrons have left the N-region, it will no longer be electrically neutral, but will have a slightly positive charge. Likewise as soon as the P-region has acquired a few excess electrons, it will no longer be electrically neutral, but will possess a slightly negative charge. This charge redistribution is shown in Fig. 4A. The current carriers in both semiconductors are repelled from the area of the junction. The area close to the junction is variously known as a depletion zone (because it is depleted of current carriers) or as a transition region. The particular charge distribution is referred to as a potential hill across the junction, a potential barrier, a dipole charge layer, a space charge layer, differing energy levels, among others.
In Fig. 4B a negative voltage is shown being applied to the P-type material and a positive voltage to the N-type. This condition, known as reverse bias, is not conducive to current flow through the junction. Electrons will try to flow from the P- to the N-material, but will encounter high opposition or resistance in their attempt to flow in the opposite direction. The reverse bias adds to the potential barrier set up by the junction when no bias is present.
Fig. 4C shows conditions across the junction when forward bias is applied. This term implies that the normal tendency of electron current to flow from N-type to P-type will be aided or encouraged by the applied voltage, and that a substantial electron current will cross the junction and flow through the external circuit.
The foregoing two paragraphs reveal the possibility of using semiconductors as a rectifying device for converting an applied alternating current to a unidirectional current. The rectifying principle finds wide application in both rectifier power supplies and detection or demodulation circuits.
THE JUNCTION TRANSISTOR
We are now ready to assemble three semiconductors in such a fashion that they become a transistor. Fig. 5 shows a P-type, an N-type, and another P-type semiconductor bonded together to constitute a PNP transistor. The charge distribution, when no external voltage is applied, is as indicated. The so-called charge dipole layer forms at each junction, so that both edges of the center semiconductor are slightly positive. This positive charge is composed of the fixed donor atoms, which do not move and hence do not participate in the electron drift process. The free electrons in the N- material are seen to be bunched near the center. They are held in position by the fixed negative acceptor atoms on the two adjoining faces of the P-type semiconductors, the repelling electric field of which extends through the positive concentrations of donor atoms on either edge of the N- material to the center.
This charge dipole layer at each junction could be shown as being produced by a small simulated battery across each junction, resulting in the existing polarities. It is important to under stand that, even under this static or equilibrium condition of no external voltage applied, there actually are small voltages (or electric fields) existing across each junction in a transistor.
A graphical representation .of this charge distribution is given immediately below the symbolic representation in Fig. 5. This charge distribution is also called an electric field-or more simply, a voltage. The fixed positive and negative atoms are omitted from this diagram. However, they would be distributed around each junction, just as they are around the single junction of Fig. 4A. The graphical picture (Fig. 5B) tells us the same story as the symbolic picture in Fig. 5A-namely, that within the base, but adjacent to each junction, there is a positive electric field due to the relative scarcity of free electrons. In the center of the base, however, there is a negative electric field due to the presence of excess free electrons. Within the emitter and collector, but immediately adjacent to each junction., there is a negative electric field due to the relative absence of positively charged "holes." As we move away from each junction, this electric field changes from negative to positive-due now to the presence of an excess of positive holes.
Because the collector is usually manufactured with a much lower conductivity factor than the emitter, considerably fewer free charge carrier (positive holes) will be milling around within it. The conductivity of a semiconductor is directly proportional to the density, or concentration, of these charge carriers. Frequently, the emitter material will have a hundred times greater conductivity than the collector material.
In Fig. 5A, the lower charge density indicated in the collector reduces the electric field across the base-collector junction, as shown graphically in Fig. 5B. And the smaller this electric field or voltage is, the easier will it be to breach the natural tendencies of the junction and get electron current to flow from the P-material of the collector into the N-material of the base.
(A) Normal charge distribution.
(B) Reverse bias condition.
(C) Forward bias condition.
( A) Symbolic representation of charge distribution. (B) Graphical representation of charge distribution.
As explained more fully in connection with Figs. 8, 9, and 10, this is accomplished by using a "signal" current ( or voltage) to vary the electron concentration within the base and thereby vary the strength of the small electric field across the base-collector junction. In this manner, it is possible to regulate the quantity of electrons flowing (as a result of the collector bias voltage) into the collector and through the transistor.
The transistor becomes a useful circuit device when external leads are attached to each semiconductor, and the leads brought to appropriate voltages and circuitry. The transistor can be compared, in some respects, to the triode vacuum tube. As explained in earlier books of this "Basic Electronics" series, the cathode of a triode emits electrons into the tube. The control grid, by virtue of the voltage applied to it, then regulates, or controls, the flow of these electrons through the tube; and the plate receives them after they have completed their journey.
The symbolic representations of the basic types of transistors, PNP and NPN, are shown in Fig. 6 and 7. In each illustration the left-hand semiconductor is labeled the base, the bottom right-hand one the emitter, and the upper right-hand semiconductor the collector. Since the function of the emitter is to emit current through the base to the collector, it is usually compared to the cathode of a tube. By virtue of the biasing conditions (current and voltage) applied between base and emitter, it is possible to control, or regulate, the flow of emitted current to the collector; consequently, the base of a transistor performs the same function as the grid of a tube. The collector receives, or collects, the current emitted by the emitter and, in that sense, corresponds to the plate of a tube.
ELECTRON-FLOW DIRECTIONS IN TRANSISTOR CIRCUITS
Fig. 6 indicates an arrow pointing into the base from the emitter, a symbol which immediately identifies this as a PNP type transistor. The arrow indicates the direction in which the positive units of current-in other words, holes-flow. As we shall see in a later example of circuitry, electron flow through a PNP transistor is into the collector from the external circuit, then into the base, from there into the emitter, and out the emitter to the external circuit again.
Fig. 7 indicates an arrow pointing into the emitter from the base, a symbol which immediately identifies any transistor as an NPN type. The arrow again corresponds to the more or less theoretical direction of flow of the positive units of current-holes again. In NPN circuitry, it is universally accepted that the electron flow is into the emitter from the external circuit, then into the base, from there into the collector, and out the collector to the external circuit. Thus, the direction of the electron flow is the same through the NPN transistor and the vacuum tube from emitter ( or cathode) to collector ( or plate). Hence, the analogy between a transistor and a vacuum tube is more precise for an NPN type.
The electron-flow directions through the two types of transistors are indicated in color in Figs. 6 and 7. The main electron stream (collector current) is shown in red; the base current, which usually carries the applied signal and a biasing current, is in green.
Forward Biasing of Emitters
Fig. 8 shows the PNP transistor of Fig. 5 when biasing voltages are applied to the emitter and collector. The voltage applied to the emitter (left-hand semiconductor) is positive.
Its polarity is such that electrons are drawn from the base into the emitter. Since this is the normal direction of electron flow through an NP junction, the emitter is said to be "forward biased" with respect to the base.
Reverse Biasing of Collectors
The voltage applied to the collector (right-hand semiconductor) is negative. Its polarity is such that electrons are driven from the collector into the base. Since this is the reverse of the normal direction of electron flow through a PN junction, the collector is said to be "reverse-biased" with respect to the base.
One fact should be especially noted. The two external voltage sources (batteries in this example) are connected so that they would drive electron current in a counterclockwise direction from the negative terminal of the "reverse bias" voltage and into the collector. From there the electron flow is through the base and emitter, to the positive terminal of the "forward bias" battery, then continues through the rest of the external circuit and re turns to the positive terminal of the reverse-bias voltage source.
In other words, these two voltage sources are not opposing each other. Rather, they are in series with each other and of like polarity-meaning both are trying to drive electrons in the same direction, or counterclockwise through collector, base, and emitter in that order.
Electrons in Base
In Figs. 5, 8, 9, and 10, we are concerned with free electrons from several sources. Since an understanding of each source is essential to an understanding of transistor action, each electron group has been shown in a different color.
Fig. 5 shows the free electrons of the base material in blue.
These are bunched along the center line of the base, whereas the positive holes in the adjoining P-type semiconductors, (in gray) are bunched along the outer edges. The fixed negative acceptor atoms in the P-material, and the fixed positive donor atoms in the N-material, have been omitted in this diagram, and also in Figs. 8 through 10.
Fig. 8 deals with the free electrons which exist within the N-type material; they are shown in blue, as in Fig. 5. It is the concentration, or density, of these free electrons that determines the normal conductivity of the base.
Also shown (in red) are the free electrons driven from right to left through the transistor by the two biasing voltages. When there is no signal voltage applied between base and emitter ( as in Fig. 8), this current assumes a normal, or "equilibrium," value. It is the main electron stream through the transistor, and corresponds to the plate current in vacuum-tube circuits.
Finally, we have an equilibrium value of electron current (green) flowing from base to emitter. When electrons from this current are within the base, they increase the electron density in the base, and consequently, the conductivity of the base.
The volume of the main electron stream from collector to emitter (red) is regulated by the conductivity of the base. Therefore, in Fig. 9, when the base is "biased" so that additional electrons are driven into it, the volume of current flowing across it from collector to emitter is drastically increased. The attempt has been made in this figure to show the concentration of blue electrons in the base as relatively unchanged from Fig. 8. There are slightly more green electrons in the base than were shown in Fig. 8.
However, because of this slight increase in electron concentration in the base, a considerably larger main collector-to-emitter current flows. These electrons. shown in red. now flow in some profusion from right to left through the three sections of the transistor.
When the polarity of the signal voltage is reversed (Fig. 10), the base-to-emitter current is restricted to a low value. This accounts for the low number of electrons (green) in the base. The resulting decrease in conductivity of the base substantially reduces the main electron stream (red) through the transistor.
It is desirable now to consider the transistor in operation in a sample circuit. This requires us to assign representative values to the applied voltages and to the two resistances in the emitter and collector circuits, and also to provide a means for changing the bias conditions between emitter and base. (In a vacuum-tube circuit, this latter function is known as "driving" the grid.)
TRANSISTORS AS CURRENT AMPLIFIERS
Transistors are referred to as "current-operated" devices, because the driving signal applied between base and emitter is usually referred to as a "current" rather than a voltage. Tubes, on the other hand, are considered to be "voltage-operated" de vices, because it is the instantaneous value of voltage at the tube grid that determines the amount of electron current which can pass through the tube.
Transistors, on the other hand, are referred to as "current amplifiers" because a small change in the bias current flowing between emitter and base will cause a much greater change in the amount of current flowing between emitter and collector. The necessary small change in bias current is usually provided by a very weak signal current (on the order of micro-amperes). Obviously, such a signal current will have a companion signal volt age, since one cannot reasonably exist without the other. Convenience and custom, however, have established the signal current as the prime agent in changing the instantaneous bias conditions of the transistor and thus regulating the flow of emitter-to-collector current.
Fig. 9 shows the same PNP transistor of Fig. 8, but with an additional current source in the line leading to the base.
This current source is shown as a small battery in order to establish the desired direction of electron flow from its negative terminal into the base, and then into and out of the emitter.
From here the electrons travel into the positive terminal of the "forward-bias" battery ( after passing through any resistor included in the emitter circuit), and finally out of the negative terminal and back to the positive terminal of the added battery.
A little reflection will confirm that the emitter-voltage source (battery) and the additional voltage source (shown as a battery) in the base circuit have polarities which drive the electron current in the same direction (counterclockwise) , whereas the polarities of the base and collector voltage sources are such that these two voltages oppose each other. The negative voltage applied at the base tends to drive the electron current clockwise through the collector circuit, but is prevented from doing so by the much larger negative voltage applied to the collector; the latter would like to drive the electron current counterclock wise into the base and the emitter.
Signal Current Action
We are now almost down to the water's edge in understanding how a transistor can "amplify" a current, and thus give both voltage and power gain. When this tiny additional electron cur rent is made to flow into the base and emitter, it manages some how to open the flood gates and let a much larger electron current flow from collector to emitter.
Should the polarity of this signal voltage be reversed, an opposite effect will occur and the electron current flowing from collector to emitter will be reduced considerably. This is depicted in Fig. 10, where the new biasing voltage added in the base circuit has a positive rather than negative polarity; consequently, it tends to draw electrons out of the base, rather than push new ones in. The consequent reduction in quantity of electrons flowing from collector to emitter is shown pictorially.
Current amplification is considered to have occurred in Figs. 8 and 9 because the resulting fluctuations in collector-to emitter current are much larger than the fluctuations in the signal current applied to the base. (The ratio between the two may easily be as high as 50 to 1.)
THE CHARACTERISTIC CURVE
A typical characteristic curve for a PNP transistor is given in Fig. 11. The horizontal scale is graduated in volts applied to the collector, and the vertical scale is graduated in milliamperes of collector current. The third variable, the amount of current flowing in the base, is shown by the lines running upward to the right. For any given values of collector voltage and base current, we can locate a single point on the base-current line and, projecting horizontally, find the amount of collector (to emitter) current flowing through the transistor.
The circuit of Fig. 12 is known as a "common-emitter" or "grounded-emitter" configuration. It differs somewhat from the examples shown in Figs. 8 through 10, which were common base configurations that required two separate biasing voltage sources (the forward-bias and reverse-bias batteries). A common-emitter configuration such as the one in Fig. 12 has the special advantage that a single voltage source serves very adequately for both biasing functions.
The common-emitter circuit corresponds most closely to the conventional vacuum-tube circuit, wherein the cathode of the tube is either grounded directly or is connected to ground through a low-value resistor which provides cathode biasing.
Circuit diagrams for the common-emitter configuration are normally drawn with the transistor in the "vertical" position that is, with the collector at the top, the base in the middle, and the emitter at the bottom. The signal is applied to the base from the left side of the diagram, and is "coupled" away to the next stage on the right side.
Note the convenience of this analogy to conventional vacuum tube circuitry, where the tube plate is shown at the top of a tube diagram, the grid in the middle, and the cathode at the bottom.
This convention of displaying a transistor vertically, with the collector at the top, should ease your mental transition from tube circuits to transistor circuits. That is to say, those who have learned to visualize the various electron currents moving around a tube circuit should have a minimum of difficulty in visualizing the comparable electron currents at work in a transistor circuit.
USE OF A "LOAD LINE"
Fig. 12 shows the PNP transistor with typical values of applied voltage at emitter and collector, and typical values of resistance in the emitter and collector circuit. Using these values of voltage and resistance, you can work a sample problem on the characteristic curve of Fig. 11. First, it is necessary to con struct what is known as a "load line" on the characteristic curve. This can be done by determining two important points corresponding to zero collector current and to zero collector voltage-and drawing a line between them.
The first point is evident from a consideration of Ohm's law, which states that the voltage developed across a resistor is proportional to the current flowing through that resistor.
If no current flows through the load resistor, R2, then there is no voltage drop across it. Thus, its two ends must be at the same voltage, which is the value of collector supply voltage Ee.
In this example, a value of 4 volts has been chosen, so this is also the voltage at the junction of base and collector when zero collector current is flowing. This value determines the location of the point corresponding to the zero collector current in Fig. 11.
Ohm's law can again help us locate the second point. If the collector voltage is truly zero, then the voltage "dropped," or used up by the collector current in flowing through R2, must equal the collector supply voltage. Since, by Ohm's law: Current (" ) _ Voltage (in volts)
m amperes - Resistance (in ohms)
... we can calculate:
Ee I (collector current) = R2 4
= 2 milliamperes
This determines the point corresponding to zero collector voltage. The line between the two points is the so-called "load line"; at any given instant, it relates the exact voltage at the collector to the exact amount of current flowing through the collector (toward the emitter). There will always be a small resistance to electron flow through the transistor; consequently, a small "voltage drop" must exist across the transistor. Since this resistance is no more than 10 or 20 ohms at the most, it is negligible in comparison with load resistance ( which in this example is assumed to be 2,000 ohms) and may be ignored in a qualitative example like that chosen here.
The next step in using the characteristic curve is to determine the operating point on the load line. This is the point corresponding to no applied signal in the base circuit. Common sense dictates that it should be somewhere near the middle of the load line, so that signal fluctuations in a positive direction will decrease the collector-to-emitter current by the same amount that a signal fluctuation in the negative direction will increase this current.
We can arbitrarily choose the point where the 20-microampere base-current line intersects the load line. Also, let us assume that the signal current to be amplified has a peak-to-peak swing of 20 microamperes, and that it is applied in the external circuit between the base and emitter. Any one of several standard coupling methods could be used. The method chosen is capacitive coupling, and the new signal will be driven in and out of the base circuit through capacitor C1.
The special virtue of the load line is that it at all moments will relate the three important variables of collector current, collector voltage, and base current. Thus, when the base current increases to -30 microamperes, the collector current increases to -1.6 milliamperes and the collector voltage decreases to -0.9 volt. (This is comparable to vacuum-tube circuitry, where an in crease in plate current through a resistive load reduces the plate voltage.)
Alternatively, when the signal current reduces the total base current to -10 microamperes, the collector voltage increases to -2.95 volts while the collector current is decreasing to -0.55 milliampere.
THE BETA FACTOR
Current amplification is said to have occurred in the previous example because we started with a signal amplitude of 20 micro amperes in the base-emitter circuit and came out with a variation of 1.05 milliamperes in the collector current. The ratio between these two current swings is:
1.05 X 10-a = 52 S 20 X 10-(1 •
Or, slightly over 50-to-1.
Such a high ratio is not uncommon among transistors. It is an important figure of merit known as the "transport factor" and symbolized by the Greek letter Beta, β. For this reason it is some times called the "beta factor" of a transistor.
To understand why a small signal current can subject the collector current to these wide variations, we must further study Fig. 5. Obviously, the explanation is not going to be as simple, nor as straightforward, as for vacuum tubes! In Fig. 5, note especially how the two junctions will assume certain charge distributions. Within the base section (which is N-type material), the negative free electrons will be repelled slightly from each of the interfaces. Thus, the outer edges of the base are slightly positive and its interior is slightly negative.
Likewise, the so-called free "holes" of positive charge in each of the P-sections (emitter and collector) will be slightly repelled from the interfaces. Both interface areas in the P material are therefore slightly negative. The free electrons in the base region would like to cross both junctions and combine with the positive holes. Referring to Figs. 8 and 9 you can see that, with the emitter "forward biased," some of these actions are now going on. But it is important to note that the unusual charge distribution at the interfaces has only been reduced in strength, not wiped out entirely. Free electrons are crossing the junction from base to emitter-in spite of the fact that fixed negative charges at the junction edge in the emitter are still exerting an electric repulsion field on them, trying to keep them out. According to the characteristic curve of Fig. 11, when the normal base current of -20 microamperes is flowing, the collector current has a value of -1.02 milliamperes.
THE ALPHA FACTOR
Electrons from both the base and collector currents will flow through the emitter. Thus, the total emitter current will be the sum of these two, or 1.04 milliamperes-only slightly larger than the collector current. The ratio between collector and emitter currents, known as the alpha gain, is another important figure of merit for transistors. In a junction transistor it cannot be greater than unity, but is normally very close to it.
In this example: Al ha= 1.02 p 1.04 or, about 98%. In Fig. 9, a small external voltage is applied to the base with such a polarity that it drives additional electrons into the base and through it to the emitter. This corresponds roughly to what happens when the base current is increased from -20 to -30 microamperes.
What happens within the base itself under these conditions? For one thing, the concentration, or density, of the negative current carriers (electrons) is increased, and the conductivity of any semiconductor varies directly with the density of current carriers in the material. Thus, the base becomes a much better conductor when more electrons are driven in. Conversely, when fewer base-current electrons flow (as in Fig. 10), the conductivity of the base is reduced.
Hence, the base acts as a variable resistor situated between emitter and collector, and having two power sources-the reverse bias and the forward bias.
How is it possible for such a small bias current to actuate this variable resistor? Fig. 9 shows the momentary increase in electron density in the base. Thus, the positive carriers (the so-called "holes") in the emitter are repelled less strongly by the "dipole charge· layer" at the junction, and will therefore move closer to it. We can assume that this shorter distance between the holes in the emitter and the free electrons in the base greatly expedites the electron drift process.
The "dipole charge layer" between base and collector will also have been diminished in strength, and consequently in width, so that the positive holes in the collector can draw closer to the excess of free electrons in the base. This will also tend to improve the conditions conducive to current flow-the inter change of free electrons from orbit to orbit of positively charged atoms which so closely resembles the electron drift process.
One additional fact of great importance is the width of the base itself. This is normally less than a thousandth of an inch, so that the two sections of P-type material are separated by only this minute distance. Thus, any increase in electron flow between base and emitter will increase the electron density in the base. In turn, more electron current will flow out of the collector, toward the emitter. It is almost as if the increased electron density in the base had created, between the collector and emitter, a series of "electron bridges" across which relatively large quantities of electrons can flow. Of course, this flow is in conformity with the special requirements of the electron drift process.
When the polarity of the applied signal voltage is such that it reduces the amount of electron current flowing from base to emitter, (as depicted in Fig. 10), the opposite effect can be created. There are fewer intrinsic electrons within the volume of the base; this in turn decreases the conductivity of the base, since conductivity varies directly with the density of current carriers in the material. Even if there were no other contributory effect to consider, the main electron stream through the transistor would still be substantially reduced, by virtue of having to flow through the higher base resistance.
SIZE OF AN ELECTRON
Even though the base is limited to a width of about a thousandth of an inch, it is worthwhile to consider how much "maneuvering room" is provided for the electrons within. The electron has a diameter of less than 10^-12 centimeter. This means that ff electrons could be arrayed end to end, a line one centimeter long would contain more than one trillion of them. Thus, a base width of a thousandth of an inch can still accommodate about two and a half billion electrons laid end to end.
This should give some idea of how drastically the number of electrons shown in Figs. 8, 9, 10 and others has been reduced and oversimplified in order to make a pictorial representation that can even be comprehended.
Figs. 8, 9, and 10 suggest a series of possible electron paths through the base of the transistor under three conditions of applied signal current. Fig. 8 assumes the condition of normal bias--i.e., zero signal applied to the base, and a forward bias such that an "equilibrium" electron current of -20 micro amperes is driven from base to emitter. This permits an "equilibrium" electron current of 1.05 milliamperes through the collector. Note the bunching of free electrons towards the vertical center line of the base, and also that an electron drift process is occurring throughout the three elements of the transistor. From the negative terminal of the reverse-bias battery, the electrons move from right to left into the P-type collector, jumping from hole to hole, then across the first junction and into the base.
Here they impart their energy to other free electrons. The latter in turn cross the second junction into the emitter, jumping from hole to hole until they emerge into the external circuit and enter the positive terminal of the forward-bias battery.
Fig. 9 shows the same transistor when the signal current from base to emitter is increased from -20 to -30 microamperes.
Being composed entirely of free electrons, this larger signal current increases the electron density in the base and also reduces the width of the "dipole charge layer" at the two junction interfaces. Simultaneously, the electron current from collector to emitter is vastly increased. This can be due to the greater ease of the electrons in crossing the existing "electron bridges" in the base, to the opening of new bridges, or a combination of both. If the beta factor of the transistor is 50, then for each additional signal-current electron that passes through the base, a total of about fifty additional electrons will be made to flow across the base from collector to emitter.
Fig. 10 shows the same transistor when the signal current from base to emitter is reduced from -20 to -10 microamperes.
Now the width of the charged layer at the two junctions will be increased. This restricts rather than encourages the electron drift process across the two junctions, although it still goes on at a reduced rate ( - .55 milliampere) . The smaller electron current can be due to fewer electron bridges through the base, to the increased difficulty in getting across the existing bridges, or a combination of both.
It is easy, from these analogies, to see how a greater width and consequently greater volume for the base would require more signal current in order to increase the base conductivity by increasing its electron density. It is obvious that a wider base would mean longer electron paths or "bridges" through the base, and that to open up new ones ( or improve the conductivity of existing ones) would require many more new electrons driven by the signal current. Consequently, the beta factor-or ratio between a change in collector current for a corresponding change in signal current-would be much lower, and the transistor would not be as good an amplifying device.
MEANING OF NEGATIVE CURRENT VALUES
Those of you who are new to transistors and their circuits should not become confused by the fact that the base and collector currents are both shown as being "negative" in the characteristic curve of Fig. 11. Whether a particular current is negative or positive depends on whether it is made up of negative electrons, or of positive charges such as "holes." It also depends on which direction the current is flowing in a circuit.
These matters have already been settled for us-the term "current" in transistor work is normally taken to mean "conventional" current, which is composed of positively charged particles such as "holes." Additionally, the "normal" direction of "conventional" current flow through any PNP transistor is from emitter to base to collector. However, in the NPN transistor, the polarity of the two biasing batteries would be reversed; the direction of electron flow would be from emitter to base to collector. Obviously, positively charged particles such as holes will flow in the opposite direction-from collector to base to emitter.
The flow direction of positive hole current through the NPN transistor has been chosen as conventional, or normal, perhaps because the flow directions of electron and hole currents through the NPN transistor correspond to the respective flow directions of electron and conventional or "positive" currents through vacuum tubes. Thus, a characteristic curve for an NPN transistor shows positive values for collector voltage ( corresponding to positive plate voltage in a vacuum tube) , and also positive values for base current and collector current. Therefore, in making any qualitative analysis of transistor-circuit operation, it is necessary to differentiate between the direction of electron current and that of positive, or hole, current.
THE COMMON-EMITTER CIRCUIT
Fig. 12 shows a sample circuit using the same PNP transistor whose characteristic curve appears in Fig. 11. This particular transistor, the 2N105, is designed for use at audio frequencies when the currents and powers involved are low. Additional circuit components required are:
MBias battery (-4 volts).
There are three electron currents which should be visualized in order to understand the operation of this circuit, which could serve as a low-power amplifier in the audio-frequency range.
These currents are:
1. Signal current (blue).
2. Base-emitter biasing current (green).
3. Collector-emitter current (red).
The signal current corresponds in many respects to the grid driving current in vacuum-tube operation. This signal current is driven by a signal voltage applied to input capacitor C1. In the so-called negative half-cycle depicted in Fig. 12, a negative signal applied at the input point drives a small electron current onto the left plate of capacitor C1. This in turn drives an electron current of equal size off the right plate of C1.
Some important facts of transistor operation can be relearned from Figs. 12 and 13. One is that a small increase in the number of electrons flowing from base to emitter encourages a large increase in the number of electrons which will flow from collector to emitter. This condition is depicted in Fig. 1~12.
During the half-cycle which we have termed "negative," the signal voltage is driving electron current onto the left plate of capacitor C1. Normal capacitor action requires that an equal number of electrons be driven away from the right plate of this capacitor, and these electrons are shown as being added to the base-emitter current.
The increased base-emitter current causes a substantial in crease in the collector-emitter current. The latter current, shown in red, is the main electron stream through the transistor.
This increased electron current flows through load resistor R1 and causes an increase in the voltage "drop" across it, in accordance with Ohm's law. Thus, the voltage at the top of the resistor (the output point of the circuit) will become less negative during this half-cycle. This has the effect of drawing electrons onto the right plate of capacitor C2, as shown in Fig. 12.
The flow directions of the original signal current flowing into C1, and the amplified signal current flowing beyond C2 tell us that a phase reversal occurs as the signal passes through this transistor circuit. This is comparable to our experience with vacuum tubes in a grounded-cathode configuration.
Fig. 13 is a "positive" half-cycle of the same circuit. The "positive" connotation is arbitrarily chosen to mean that half cycle during which less and less electron current is flowing into the base. The signal voltage applied to capacitor C1 acts as a sort of "pumping" action-alternately drawing off some of the electrons coming up through R1 ( on the positive half-cycles), and delivering them to the base (on negative half-cycles). In this way, the total current entering the base can easily be varied between -10 and -30 microamperes.
The main current stream from collector to emitter has been shown in red in these two diagrams. In Fig. 13, when the driving current decreases the total current entering the base, the col lector-to-emitter current stream is decreased in the same proportion. Using values previously calculated from the characteristic curve, the collector current during this second half-cycle drops to its minimum value of - .55 milliampere, and during the preceding half-cycle it reached its maximum value of -1.6 milliamperes.
The output point of this circuit is connected directly to the collector and, as we have seen, the voltage at this point fluctuates between the extremes of -.75 volt (at the peak of the first half cycle) to -2.90 volts (at the peak of the second half-cycle). Thus, we have a voltage swing exceeding 2 volts peak-to-peak at the collector.
Both the collector current and the base current are pulsating direct currents. This tells us they are unidirectional (flow in one direction only). Both currents are driven by the same bias battery of 4 volts. The path of the collector current begins at the negative terminal of this battery. Electrons are driven up ward through load resistance R2, then to the left and downward through the collector, base, and emitter to ground. From the ground point, the collector current has the necessary free access to the positive terminal of the battery. As is the case with the plate current in a vacuum tube, a closed path must exist through the power supply (battery) and the regulating device (transistor). The path of the base current also begins at the negative terminal of the battery. Electrons are driven upward through base resistor R1 and through the base and emitter to ground.
From here the base current can return freely to the positive (grounded) terminal of the battery. This closed path must also exist; and it is somewhat comparable to the grid return path provided in every vacuum-tube circuit.
One feature that makes the common-emitter configuration particularly attractive is that forward and reverse biasing can both be provided from a single voltage source. The important consideration in any transistor circuit is to assure that the voltage source or sources are wired so that their polarities will drive electrons in the proper direction throughout the circuit.
Regardless of whether the transistor is a PNP or an NPN, the collector current and base current must flow in the same direction through the emitter.
The second consideration is that a negative voltage must be provided in order to drive these currents in the PNP transistor, whereas those in the NPN require a positive voltage.
A word of caution: this circuit arrangement, including component values and the applied voltage value of 4 volts, has been chosen for one reason only-to clarify some of the important fundamentals of transistor operation in a sample circuit. They might not prove to be the best component values for low frequency audio amplification in a particular circuit. Also, even though the 2N105 transistor works equally well in a higher voltage range, that portion of its characteristic curve which applies in this range has been omitted from Fig. 11.