Newton's third law of motion ( Action and Reaction ! )

When a bullet is fired from a gun equal and opposite forces are exerted on the bullet and the gun during the time the bullet is passing down the barrel

Newton's second law of motion ( momentum , force , weight , free falling )

A motortruck requires a larger force to set in motion when it is heavily loaded than when it is empty

The nature of friction: Static and dynamic friction

Friction is the name given to the force which opposes the relative sliding motion of two surfaces in contact with one another. It plays a notable part in our daily lives. For example

Newton's Laws of motion:1- Newton's first law of motion

In 1687 Newton published a book written , as was the routine in those days , in Latin and given the title , Philosophiae naturalis principia mathematica Translated , this means

Derivation of Kinetic energy formula and worked examples

Where there are no opposing forces, a moving body needs noforce to keep it moving with a steady velocity ( Newton's first law of motion ). If, however, a resultant force does act on a moving body

Monday, October 20, 2014

field effect transistor What It Does,How It Works,Variants,Values,How to Use it and What Can Go Wrong

The term field-effect transistor encompasses a family primarily consisting of the junc­tion field-effect transistor (or JFET, which is the simplest generic type) and the metal-oxide semiconductor field-effect transistor (or MOSFET, also sometimes known as an insulated- gate field-effect transistor, or IGFET). Because the principles of operation overlap consid­erably, the entire -FET family is grouped in this entry.

What It Does

A field-effect transistor creates an electric field to control current flowing through a channel in a semiconductor. MOSFETs of microscopic size form the basis of complementary metal oxide semiconductor (CMOS) integrated circuit chips, while large discrete MOSFETs are capable of switching substantial currents, in lamp dimmers, audio amplifiers, and motor controllers. FETs have become indispensable in computer elec­tronics.

A bipolar transistor is generally thought of as a current amplifier because the current passing through it is controlled by a smaller amount of current passing through the base. By contrast, all FETs are considered to be voltage amplifiers, as the control voltage establishes field intensity, which requires little or no current. The negligible leakage through the gate of an FET makes it ideal for use in low-power applications such as portable hand-held devices.

How It Works

This section is divided into two subsections, de­ scribing the most widely used FETs: JFETs and MOSFETs.


A junction field-effect transistor (or JFET) is the simplest form of FET. Just as a bipolar transistor can be of NPN or PNP type, a JFET can have an N- channel or P-channel, depending whether the channel that transmits current through the de­ vice is negatively or positively doped. A detailed explanation of semiconductor doping will be found in the bipolar transistor entry.

Because negative charges have greater mobility, the N-channel type allows faster switching and is more commonly used than the P-channel type. A schematic symbol for it is shown in Figure 29-1 alongside the schematic for an NPN transistor. These symbols suggest the similarity of the de­ vices as amplifiers or switches, but it is important to remember that the FET is a primarily a voltage amplifier while the bipolar transistor is a current amplifier.


Figure 29-1. A comparison between schematic symbols for N-channel JFET (left) and NPN bipolar transistor (right) suggests their functional similarity as switches or amplifiers, although their behavior is markedly different.

Three JFETs are shown in Figure 29-2. The N- channel J112 type is supplied by several manu­facturers, the figure showing two samples, one from Fairchild Semiconductor (left) and the other from On Semiconductor (right). Although the full part numbers are different, the specifications are almost identical, including a drain-gate voltage of 35V, a drain-source voltage of 35V, and a gate current of 50mA. The metal-clad 2N4392 in the center has similar values but is three times the price, with a much higher power dissipation of 1.8W, compared with 300mW and 350mW for the other two transistors respectively.


Figure 29-2. Junction Field Effect Transistors (JFETs). See text for details.

Schematic symbols for N-channel and P-channel JFETs are shown in Figure 29-3, N-channel being on the left while P-channel is on the right. The upper-left and lower-left symbol variants are both widely used and are functionally identical. The upper-right and lower-right variants likewise

mean the same thing. Because the upper variants are symmetrical, an S should be added to clarify which terminal is the source. In practice, the S is often omitted, allowing some ambiguity. While the source and drain of some JFETs are in fact interchangeable, this does not apply to all types.

The circle around each symbol is occasionally omitted when representing discrete compo­nents, and is almost always omitted when mul­tiple FETs are shown connected to form an inte­ grated circuit.


Figure 29-3. Schematic symbols for junction field-effect transistors (JFETs). Left: N-channel. Right: P-channel. The symbols at top and bottom on each side are functionally identical. Circles may be omitted. Letter S may be omitted from the symmetrical symbol variants, even though this creates some ambiguity.

The internal function of an N-channel JFET is shown diagrammatically in Figure 29-4. In this component, the source terminal is a source of electrons, flowing relatively freely through the N- doped channel and emerging through the drain. Thus, conventional current flows nonintuitively from the drain terminal, to the source terminal, which will be of lower potential.

The JFET is like a normally-closed switch. It has a low resistance so long as the gate is at the same potential as the source. However, if the potential of the gate is reduced below the potential of the source—that is, the gate acquires a more rela­

tively negative voltage than the source—the cur­ rent flow is pinched off as a result of the field cre­ated by the gate. This is suggested by the lower diagram in Figure 29-4.


Figure 29-4. At top, conventional current flows freely from drain to source through the channel of an N-doped JFET. At bottom, the lowered voltage of the gate relative to the source creates a field effect that pinches off the flow of current.

The situation for a P-channel JFET is reversed, as shown in Figure 29-5. The source is now positive (but is still referred to as the source), while the drain can be grounded. Conventional current now flows freely from source to drain, so long as the gate is at the same positive potential as the source. If the gate voltage rises above the source voltage, the flow is pinched off.

A bipolar transistor tends to block current flow by default, but becomes less resistive when its base is forward-biased. Therefore it can be re­


Figure 29-5. At top, conventional current flows freely from source to drain through the channel of a P-doped JFET. At bottom, the higher voltage of the gate relative to the source creates a field effect that pinches off the flow of current.

ferred to as an enhancement device. By contrast, an N-channel JFET allows current to flow by de­ fault, and becomes more resistive when its base is reverse-biased, which widens the depletion lay­ er at the base junction. Consequently it can be referred to as a depletion device.

The primary characteristicts of a junction field- effect transistor relative to an NPN bipolar tran­sistor are summarized in the table in Figure 29-6.

JFET Behavior

The voltage difference between gate and source of a JFET is usually referred to as Vgs while the voltage difference between drain and source is referred to as Vds.


Figure 29-6. This table contrasts the characteristics of an N-channel JFET with those of an NPN bipolar transistor.

Suppose the gate of an N-channel JFET is con­nected with the source, so that Vgs=0. Now if Vds increases, the current flowing through the channel of the JFET also increases, approximately linearly with Vds. In other words, initially the JFET behaves like a low-value resistor in which the voltage across it, divided by the amperage flow­ ing through it, is approximately constant. This phase of the JFET’s behavior is known as its ohmic region. While the unbiased resistance of the channel in a JFET depends on the component type, it is generally somewhere between 10Ω and 1K.

If Vds increases still further, eventually no addi­tional flow of current occurs. At this point the channel has become saturated, and this plateau zone is referred to as the saturation region, often abbreviated Idss, meaning “the saturated drain current at zero bias.” Although this is a nearly constant value for any particular JFET, it may vary somewhat from one sample of a component to another, as a result of manufacturing variations.

If Vds continues to increase, the component fi­nally enters a breakdown state, sometimes re­ferred to by its full formal terminology as drain- source breakdown. The current passing through the JFET will now be limited only by capabilities

of the external power source. This breakdown state can be destructive to the component, and is comparable to the breakdown state of a typical diode.

What if the voltage at the gate is reduced below the voltage at the source—such as Vgs becomes negative? In its ohmic region, the component now behaves as if it has a higher resistance, and it will reach its saturation region at a lower cur­ rent value (although around the same value for Vds). Therefore, by reducing the voltage on the gate relative to the voltage at the source, the ef­ fective resistance of the component increases, and in fact it can behave as a voltage-controlled resistor.

The upper diagram in Figure 29-7 shows this graphically. Below it, the corresponding graph for a P-channel JFET looks almost identical, ex­cept that the current flow is reversed and is pinched off as the gate voltage rises above the source voltage. Also, the breakdown region is reached more quickly with a P-channel JFET than with an N-channel JFET.


MOSFETs have become one of the most widely used components in electronics, everywhere from computer memory to high-amperage switching power supplies. The name is an acro­ nym for metal-oxide semiconductor field-effect transistor. A simplified cross-section of an N- channel MOSFET is shown in Figure 29-8.

Two MOSFETs are shown in Figure 29-9.

Like a JFET, a MOSFET has three terminals, iden­tified as drain, gate, and source, and it functions by creating a field effect that controls current flowing through a channel. (Some MOSFETS have a fourth terminal, described later). Howev­er, it has a metal source and drain making contact with each end of the channel (hence the term “metal” in its acronym) and also has a thin layer of silicon dioxide (hence the term “oxide” in its acronym) separating the gate from the channel, thus raising the impedance at the gate to at least


Figure 29-7. The top graph shows current passing through the channel of an N-channel JFET, depending on gate voltage and source voltage. The lower graph is for a P-channel JFET.

100,000 gigaohms and reducing gate current es­sentially to zero. The high gate impedance of a MOSFET allows it to be connected directly to the output of a digital integrated circuit. The layer of silicon dioxide is a dielectric, meaning that a field appled to one side creates an opposite field on the other side. The gate attached to the surface of the layer functions in the same way as one plate of a capacitor.


Figure 29-8. Simplified diagram of an N-channel MOS- FET. The thickness of the silicon dioxide layer has been greatly exaggerated for clarity. The black terminals are metallic.


Figure 29-9. Two MOSFETs. At left, the TO-220 package claims a drain current of up to 65A continuous, and a drain-to-source breakdown voltage 100V. At right, the smaller package offers a drain current of 175mA continuous, and a drain-to-source breakdown voltage of 300V.

The silicon dioxide also has the highly desirable property of insulating the gate from the channel, thus preventing unwanted reverse current. In a JFET, which lacks a dielectric layer, if source volt­ age is allowed to rise more than about 0.6V high­er than gate voltage, the direct internal connec­tion between gate and channel allows negative

charges to flow freely from source to gate, and as the internal resistance will be very low, the re­sulting current can be destructive. This is why the JFET must always be reverse-biased.

A MOSFET is freed from these restrictions, and the gate voltage can be higher or lower than the source voltage. This property enables an N- channel MOSFET to be designed not only as a depletion device, but alternatively as an en­ hancement device, which is “normally off” and can be switched on by being forward-biased. The primary difference is the extent to which the channel in the MOSFET is N-doped with charge carriers, and therefore will or will not conduct without some help from the gate bias.

In a depletion device, the channel conducts, but applying negative voltage to the gate can pinch off the current.

In an enhancement device, the channel does not conduct, but applying positive voltage to the gate can make it start to do so.

In either case, a shift of bias from negative to positive encourages channel conduction; the depletion and enhancement versions simply start from different points.

This is clarified in Figure 29-10. The vertical (log­ arithmic) scale suggests the current being con­ ducted through the channel of the MOSFET, while the green curve describes the behavior of a depletion version of the device. Where this curve crosses the center line representing 0 volts bias, the channel is naturally conductive, like a JFET. Moving left down the curve, as reverse bias is applied (shown on the horizontal axis), the component becomes less conductive until final­ ly its conductivity reaches zero.

Meanwhile on the same graph, the orange curve represents an enhancement MOSFET, which is nonconductive at 0 volts bias. As forward bias increases, the current also increases—similar to a bipolar transistor.

To make things more confusing, a MOSFET, like a JFET, can have a P-doped channel; and once


Figure 29-10. The current conduction of depletion and enhancement N-channel MOSFETs. See text for details. (Influenced by The Art of Electronics by Horowitz and Hill.)

again it can function in depletion or enhance­ ment mode. The behavior of this variant is shown in Figure 29-11. As before, the green curve shows the behavior of a depletion MOSFET, while the orange curve refers to the enhancement version. The horizontal axis now shows the voltage dif­ference between the gate and the drain terminal. The depletion component is naturally conduc­ tive at zero bias, until the gate voltage increases above the drain voltage, pinching off the current flow. The enhancement component is not con­ductive until reverse bias is applied.

Figure 29-12 shows schematic symbols that rep­ resent depletion MOSFETs. The two symbols on the left are functionally identical, representing N- channel versions, while the two symbols on the right represent P-channel versions. As in the case of JFETs, the letter “S” should be (but often is not) added to the symmetrical versions of the sym­bols, to clarify which is the source terminal. The left-pointing arrow identifies the components as N-channel, while in the symbols on the right, the right-pointing arrows indicate P-channel MOS­


Figure 29-11. The current conduction of depletion and enhancement P-channel MOSFETs. See text for details.

FETs. The gap between the two vertical lines in each symbol suggests the silicon dioxide dielec­tric. The right-hand vertical line represents the channel.


Figure 29-12. Schematic symbols for depletion MOS- FETs. These function similarly to JFETs. The two symbols on the left are functionally identical, and represent N- channel depletion MOSFETs. The two symbols on the right are both widely used to represent P-channel depletion MOSFETs.

For enhancement MOSFETs, a slightly different symbol uses a broken line between the source and drain (as shown in Figure 29-13) to remind

us that these components are “normally off” when zero-biased, instead of “normally on.” Here again a left-pointing arrow represents an N- channel MOSFET, while a right-pointing arrow represents a P-channel MOSFET.


Figure 29-13. Schematic symbols for enhancement MOSFETs. The two on the left are functionally identical, and represent N-channel enhancement MOSFETs. The two on the right represent P-channel enhancement MOS- FETs.

Because there is so much room for confusion re­ garding MOSFETs, a summary is presented in Figure 29-14 and Figure 29-15. In these figures, the relevant parts of each schematic symbol are shown disassembled alongside text explaining their meaning. Either of the symbols in Figure 29-14 can be superimposed on either of the symbols in Figure 29-15, to combine their functions. So, for instance, if the upper symbol in Figure 29-14 is superimposed on the lower sym­bol in Figure 29-15, we get an N-channel MOSFET of the enhancement type.


Figure 29-14. Either of the two symbols can be combined with either of the two symbols in the next figure, to create one of the four symbols for a MOSFET. See text for details.


Figure 29-15. Either of the two symbols can be combined with either of the two symbols from the previous figure, to create one of the four symbols for a MOSFET. See text for details.

In an additional attempt to clarify MOSFET be­ havior, four graphs are provided in Figure 29-16, Figure 29-17, Figure 29-18, and Figure 29-19. Like JFETs, MOSFETs have an initial ohmic region, fol­ lowed by a saturation region where current flows relatively freely through the device. The gate-to-

source voltage will determine how much flow is permitted. However, it is important to pay close attention to the graph scales, which differ for each of the four types of MOSFET.


Figure 29-16. Current flow through a depletion-type, N- channel MOSFET.

In all of these graphs, a bias voltage exists, which allows zero current to flow (represented by the graph line superimposed on the horizontal axis). In other words, the MOSFET can operate as a switch. The actual voltages where this occurs will vary with the particular component under con­ sideration.

The N-channel, enhancement-type MOSFET is especially useful as a switch because in its normally-off state (with zero bias) it presents a very high resistance to current flow. It requires a relatively low positive voltage at the gate, and effectively no gate current, to begin conducting conventional current from its drain terminal to its source terminal. Thus it can be driven directly by typical 5-volt logic chips.

Depletion-type MOSFETs are now less common­ ly used than the enhancement-type.


Figure 29-17. Current flow through a depletion-type, P- channel MOSFET.


Figure 29-18. Current flow through an enhancement- type, N-channel MOSFET.


Figure 29-19. Current flow through an enhancement- type, P-channel MOSFET.

The Substrate Connection

Up to this point, nothing has been said about a fourth connection available on many MOSFETs, known as the body terminal. This is connected to the substrate on which the rest of the compo­ nent is mounted, and acts as a diode junction with the channel. It is typically shorted to the source terminal, and in fact this is indicated by the schematic symbols that have been used so far. It is possible, however, to use the body ter­minal to shift the threshold gate voltage of the MOSFET, either by making the body terminal more negative than the source terminal (in an N- channel MOSFET) or more positive (in a P- channel MOSFET). Variants of the MOSFET sche­matic symbols showing the body terminal are shown in Figure 29-20 (for depletion MOSFETS) and Figure 29-21 (for enhancement MOSFETS).

A detailed discussion of the use of the body ter­minal to adjust characteristics of the gate is be­ yond the scope of this encyclopedia.


Figure 29-20. Schematic symbol variants for depletion MOSFETs, showing the body terminal separately accessible instead of being tied to the source terminal.


Figure 29-21. Schematic symbol variants for enhancement MOSFETs, showing the body terminal separately accessible instead of being tied to the source terminal.


A few FET variants exist in addition to the two previously discussed.


The acronym stands for MEtal Semiconductor Field Effect Transistor. This FET variant is fabrica­ ted from gallium arsenide and is used primarily in radio frequency amplification, which is outside the scope of this encyclopedia.

V-Channel MOSFET

Whereas most FET devices are capable of han­dling only small currents, the V-channel MOSFET (which is often abbreviated as a VMOS FET and has a V-shaped channel as its name implies) is capable of sustained currents of at least 50A and voltages as high as 1,000V. It is able to pass the high current because its channel resistance is well under 1Ω. These devices, commonly re­ ferred to as power MOSFETs, are available from all primary semiconductor manufacturers and are commonly used in switching power supplies.

Trench MOS

The TrenchMOS or Trenchgate MOS is a MOSFET variant that encourages current to flow vertically rather than horizontally, and includes other in­ novations that enable an even lower channel re­ sistance, allowing high currents with minimal heat generation. This device is finding applica­ tions in the automobile industry as a replace­ ment for electromechanical relays.


The maximum values for JFETs, commonly found listed in datasheets, will specify Vds (the drain- source voltage, meaning the potential difference between drain and source); Vdg (the drain-gate voltage, meaning the potential difference be­ tween drain and gate); Vgsr (the reverse gate- source voltage); gate current; and total device dissipation in mW. Note that the voltage differ­ ences are relative, not absolute. Thus a voltage of 50V on the drain and 25V on the source might be acceptable in a component with a Vds of 25V. Similarly, while a JFET’s “pinch-off” effect begins as the gate becomes “more negative” than the source, this can be achieved if, for example, the source has a potential of 6V and the gate has a potential of 3V.

JFETs and MOSFETs designed for low-current switching applications have a typical channel re­ sistance of just a few ohms, and a maximum switching speed around 10Mhz.

The datasheet for a MOSFET will typically include values such as gate threshold voltage, which may be abbreviated Vgs (or Vth) and establishes the relative voltage at which the gate starts to play an active role; and the maximum on-state drain current, which may be abbreviated Id(on) and es­tablishes the maximum limiting current (usually at 25 degree Centigrade) between source and gate.

How to Use it

The combination of a very high gate impedance, very low noise, very low quiescent power con­sumption in its off state, and very fast switching capability makes the MOSFET suitable for many applications.

P-Channel Disadvantage

P-channel MOSFETs are generally less popular than N-channel MOSFETS because of the higher resistivity of P-type silicon, resulting from its low­er carrier mobility, putting it at a relative disad­ vantage.

Bipolar Substitution

In many instances, an appropriate enhancement-type MOSFET can be substituted for a bipolar transistor with better results (lower noise, faster action, much higher impedance, and probably less power dissipation).

Amplifier Front Ends

While MOSFETs are well-suited for use in the front end of an audio amplifier, chips containing MOS­ FETs are now available for this specific purpose.

Voltage-Controlled Resistor

A simple voltage-controlled resistor can be built around a JFET or MOSFET, so long as its perfor­mance remains limited to the linear or ohmic re­ gion.

Compatibility with Digital Devices

 A JFET may commonly use power supplies in the range of 25VDC. However, it can accept the high/

low output from a 5V digital device to control its gate. A 4.7K pullup resistor is an appropriate val­ue to be used if the FET is to be used in conjunc­tion with a TTL digital chip that may have a volt­ age swing of only approximately 2.5V between its low and high thresholds.

What Can Go Wrong
Static Electricity

Because the gate of a MOSFET is insulated from the rest of the component, and functions much like a plate of a capacitor, it is especially likely to accumulate static electricity. This static charge may then discharge itself into the body of the component, destroying it. A MOSFET is particu­larly vulnerable to electrostatic discharge be­ cause its oxide layer is so thin. Special care should be taken either when handling the component, or when it is in use. Always touch a grounded object or wear a grounded wrist band when han­dling MOSFETs, and be sure that any circuit using MOSFETs includes appropriate protection from static and voltage spikes.

A MOSFET should not be inserted or removed while the circuit in which it performs is switched on or contains residual voltage from undis ­charged capacitors.


Failure because of overheating is of special con­cern when using power MOSFETs. A Vishay Ap­ plication Note (“Current Power Rating of Power Semiconductors”) suggests that this kind of com­ponent is unlikely to operate at less than 90 de­ grees Centigrade in real-world conditions, yet the power handling capability listed in a data­ sheet usually assumes an industry standard of 25 degrees Centigrade.

On the other hand, ratings for continuous power are of little relevance to switching devices that have duty cycles well below 100%. Other factors also play a part, such as the possibility of power surges, the switching frequency, and the integ­rity of the connection between the component

and its heat sink. The heat sink itself creates un­ certainty by tending to average the temperature of the component, and of course there is no sim­ple way to know the actual junction temperature, moment by moment, inside a MOSFET.

Bearing in mind the accumulation of unknown factors, power MOSFETs should be chosen on an extremely conservative basis. According to a tu­torial in the EE Times, actual current switched by a MOSFET should be less than half of its rated current at 25 degrees, while one-fourth to one- third are common. Figure 29-22 shows the real- world recommended maximum drain current at various temperatures. Exceeding this recommendation can create additional heat, which cannot be dissipated, leading to further accu­mulation of heat, and a thermal runaway condi­ tion, causing eventual failure of the component.


Figure 29-22. Maximum advised drain current through a power MOSFET, related to case temperature of the component. Derived from EE Times Power MOSFET Tutorial.

Wrong Bias

As previously noted, applying forward bias to a JFET can result in the junction between the gate and the source starting to behave like a forward- biased diode, when the voltage at the gate is greater than the voltage at the source by ap­proximately 0.6V or more (in an N-channel JFET). The junction will present relatively little resist­ance, encouraging excessive current and de­ structive consequences. It is important to design devices that allow user input in such a way that user error can never result in this eventuality.

bipolar transistor What It Does,How It Works,Variants,Values,How to Use it and What Can Go Wrong

The word transistor, on its own, is often used to mean bipolar transistor, as this was the type that became most widely used in the field of discrete semiconductors. However, bipolar transistor is the correct term. It is sometimes referred to as a bipolar junction transistor or BJT.

What It Does

A bipolar transistor amplifies fluctuations in cur­ rent or can be used to switch current on and off. In its amplifying mode, it replaced the vacuum tubes that were formerly used in the amplifica­tion of audio signals and many other applica­tions. In its switching mode it resembles a re­ lay, although in its “off” state the transistor still allows a very small amount of current flow, known as leakage.

A bipolar transistor is described as a discrete semiconductor device when it is individually packaged, with three leads or contacts. A pack­ age containing multiple transistors is an integra­ted circuit. A Darlington pair actually contains two transistors, but is included here as a discrete component because it is packaged similarly and functions like a single transistor. Most integrated circuits will be found in Volume 2 of this ency­clopedia.

How It Works

Although the earliest transistors were fabricated from germanium, silicon has become the most commonly used material. Silicon behaves like an insulator, in its pure state at room temperature, but can be “doped” (carefully contaminated) with impurities that introduce a surplus of elec­trons unbonded from individual atoms. The re­sult is an N-type semiconductor that can be in­ duced to allow the movement of electrons through it, if it is biased with an external voltage. Forward bias means the application of a positive voltage, while reverse bias means reversing that voltage.

Other dopants can create a deficit of electrons, which can be thought of as a surplus of “holes” that can be filled by electrons. The result is a P- type semiconductor.

A bipolar NPN transistor consists of a thin central P-type layer sandwiched between two thicker N- type layers. The three layers are referred to as collector, base, and emitter, with a wire or contact

attached to each of them. When a negative charge is applied to the emitter, electrons are forced by mutual repulsion toward the central base layer. If a forward bias (positive potential) is applied to the base, electrons will tend to be at­ tracted out through the base. However, because the base layer is so thin, the electrons are now close to the collector. If the base voltage increa­ ses, the additional energy encourages the elec­trons to jump into the collector, from which they will make their way to the positive current source, which can be thought of as having an even great­ er deficit of electrons.

Thus, the emitter of an NPN bipolar transistor emits electrons into the transistor, while the col­ lector collects them from the base and moves them out of the transistor. It is important to re­ member that since electrons carry a negative charge, the flow of electrons moves from nega­tive to positive. The concept of positive-to- negative current is a fiction that exists only for historical reasons. Nevertheless, the arrow in a transistor schematic symbol points in the direc­tion of conventional (positive-to-negative) cur­ rent.

In a PNP transistor, a thin N-type layer is sand­ wiched between two thicker P-type layers, the base is negatively biased relative to the emitter, and the function of an NPN transistor is reversed, as the terms “emitter” and “collector” now refer to the movement of electron-holes rather than electrons. The collector is negative relative to the base, and the resulting positive-to-negative cur­ rent flow moves from emitter to base to collector. The arrow in the schematic symbol for a PNP transistor still indicates the direction of positive current flow.

Symbols for NPN and PNP transistors are shown in Figure 28-1. The most common symbol for an NPN transistor is shown at top-left, with letters C, B, and E identifying collector, base, and emitter. In some schematics the circle in the symbols is omitted, as at top-right.

A PNP transistor is shown in the center. This is the most common orientation of the symbol, since its collector must be at a lower potential than its emitter, and ground (negative) is usually at the bottom of a schematic. At bottom, the PNP sym­bol is inverted, allowing the positions of emitter and collector to remain the same as in the symbol for the NPN transistor at the top. Other orienta­tions of transistor symbols are often found, mere­ ly to facilitate simpler schematics with fewer con­ductor crossovers. The direction of the arrow in the symbol (pointing out or pointing in) always differentiates NPN from PNP transistors, respec­tively, and indicates current flowing from posi­tive to negative.


Figure 28-1. Symbols for an NPN transistor (top) and a PNP transistor (center and bottom). Depending on the schematic in which the symbol appears, it may be rotated or inverted. The circle may be omitted, but the function of the component remains the same.

NPN transistors are much more commonly used than PNP transistors. The PNP type was more dif­ficult and expensive to manufacture initially, and

circuit design evolved around the NPN type. In addition, NPN transistors enable faster switch­ing, because electrons have greater mobility than electron-holes.

To remember the functions of the collector and the emitter in an NPN transistor, you may prefer to think in terms of the collector collecting pos­itive current into the transistor, and the emitter emitting positive current out of the transistor. To remember that the emitter is always the terminal with an arrow attached to it (both in NPN and PNP schematic symbols), consider that “emitter” and “arrow” both begin with a vowel, while “base” and “collector” begin with consonants. To remember that an NPN transistor symbol has its arrow pointing outward, you can use the mnemonic “N/ever P/ointing i/N.”

Current flow for NPN and PNP transistors is illus­trated in Figure 28-2. At top-left, an NPN transis­tor passes no current (other than a small amount of leakage) from its collector to its emitter so long as its base is held at, or near, the potential of its emitter, which in this case is tied to negative or ground. At bottom-left, the purple positive sym­bol indicates that the base is now being held at a relatively positive voltage, at least 0.6 volts higher than the emitter (for a silicon-based tran­sistor). This enables electrons to move from the emitter to the collector, in the direction of the blue arrows, while the red arrows indicate the conventional concept of current flowing from positive to negative. The smaller arrows indicate a smaller flow of current. A resistor is included to protect the transistor from excessive current, and can be thought of as the load in these circuits.

At top-right, a PNP transistor passes no current (other than a small amount of leakage) from its emitter to its collector so long as its base is held at, or near, the potential of the emitter, which in this case is tied to the positive power supply. At bottom-right, the purple negative symbol indi­cates that the base is now being held at a rela­ tively negative voltage, at least 0.6 volts lower than the emitter. This enables electrons and cur­ rent to flow as shown. Note that current flows

into the base in the NPN transistor, but out from the base in the PNP transistor, to enable conduc­tivity. In both diagrams, the resistor that would normally be included to protect the base has been omitted for the sake of simplicity.


Figure 28-2. Current flow through NPN and PNP transistors. See text for details.

An NPN transistor amplifies its base current only so long as the positive potential applied to the collector is greater than the potential applied to the base, and the potential at the base must be greater than the potential at the emitter by at least 0.6 volts. So long as the transistor is biased in this way, and so long as the current values re­ main within the manufacturer’s specified limits, a small fluctuation in current applied to the base will induce a much larger fluctuation in current between the collector and the emitter. This is why a transistor may be described as a current amplifier.

A voltage divider is often used to control the base potential and ensure that it remains less than the potential on the collector and greater than the potential at the emitter (in an NPN transistor). See Figure 28-3.

See Chapter 10 for additional information about the function of a voltage divider.


Figure 28-3. Resistors R1 and R2 establish a voltage divider to apply acceptable bias to the base of an NPN transistor.

Current Gain

The amplification of current by a transistor is known as its current gain or beta value, which can be expressed as the ratio of an increase in col­ lector current divided by the increase in base current that enables it. Greek letter β is customarily used to represent this ratio. The formula looks like this:

β = ΔIc / ΔIb

where Ic is the collector current and Ib is the base current, and the Δ symbol represents a small change in the value of the variable that follows it.

Current gain is also represented by the term hFE, where E is for the common Emitter, F is for For­ ward current, and lowercase letter h refers to the transistor as a “hybrid” device.

The beta value will always be greater than 1 and is often around 100, although it will vary from one type of transistor to another. It will also be affected by temperature, voltage applied to the transistor, collector current, and manufacturing inaccuracies. When the transistor is used outside of its design parameters, the formula to deter­ mine the beta value no longer directly applies.

There are only two connections at which current can enter an NPN transistor and one connection where it can leave. Therefore, if Ie is the current from the emitter, Ic is the current entering the collector, and Ib is the current entering the base:

Ie = Ic + Ib

If the potential applied to the base of an NPN transistor diminishes to the point where it is less than 0.6V above the potential at the emitter, the transistor will not conduct, and is in an “off” state, although a very small amount of leakage from collector to emitter will still occur.

When the current flowing into the base of the transistor rises to the point where the transistor cannot amplify the current any further, it be­ comes saturated, at which point its internal impedance has fallen to a minimal value. Theoreti­cally this will allow a large flow of current; in practice, the transistor should be protected by resistors from being damaged by high current resulting from saturation.

Any transistor has maximum values for the col­ lector current, base current, and the potential

difference between collector and emitter. These values should be provided in a datasheet. Ex­ceeding them is likely to damage the compo­ nent.


In its saturated mode, a transistor’s base is satu­ rated with electrons (with no room for more) and the internal impedance between collector and emitter drops as low as it can go.

The cutoff mode of an NPN transistor is the state where a low base voltage eliminates all current flow from collector to emitter other than a small amount of leakage.

The active mode, or linear mode, is the intermedi­ ate condition between cutoff and saturated, where the beta value or hFE (ratio of collector current to base current) remains approximately constant. That is, the collector current is almost linearly proportional to the base current. This lin­ ear relationship breaks down when the transistor nears its saturation point.


Small signal transistors are defined as having a maximum collector current of 500 mA and max­ imum collector power dissipation of 1 watt. They can be used for audio amplification of low-level inputs and for switching of small currents. When determining whether a small-signal transistor can control an inductive load such as a motor or relay coil, bear in mind that the initial current surge will be greater than the rated current draw during sustained operation.

Small switching transistors have some overlap in specification with small signal transistors, but generally have a faster response time, lower beta value, and may be more limited in their tolerance for collector current. Check the manufacturer’s datasheet for details.

High frequency transistors are primarily used in video amplifiers and oscillators, are physically small, and have a maximum frequency rating as high as 2,000 MHz.

Power transistors are defined as being capable of handling at least 1 watt, with upper limits that can be as high as 500 watts and 150 amps. They are physically larger than the other types, and may be used in the output stages of audio am­plifiers, and in switching power supplies (see Chapter 16). Typically they have a much lower current gain than smaller transistors (20 or 30 as opposed to 100 or more).

Sample transistors are shown in Figure 28-4. Top: A 2N3055 NPN power transistor. This type was originally introduced in the late 1960s, and ver­sions are still being manufactured. It is often found in power supplies and in push-pull power amplifiers, and has a total power dissipation rat­ ing of 115W. Second row, far left: general purpose switching-amplification PNP power transistor rated for up to 50W power dissipation. Second row, far right: A high-frequency switching tran­ sistor for use in lighting ballast, converters, in­verters, switching regulators, and motor control systems. It tolerates relatively high voltages (up to 700V collector-emitter peak) and is rated for up to 80W total power dissipation. Second row, center-left and center-right: Two variants of the 2N2222 NPN small signal switching transistor, first introduced in the 1960s, and still very widely used. The metal can is the TO-19 package, capa­ble of slightly higher power dissipation than the cheaper plastic TO-92 package (1.8W vs. 1.5W with a collector temperature no greater than 25 degrees Centigrade).


Traditionally, small-signal transistors were pack­ aged in small aluminum “cans” about 1/4” in di­ameter, and are still sometimes found in this form. More commonly they are embedded in buds of black plastic. Power transistors are pack­ aged either in a rectangular module of black


Figure 28-4. Samples of commonly used transistors. See text for details.

plastic with a metal back, or in a round metal “button.” Both of these forms are designed to dissipate heat by being screw-clamped to a heat sink.


Often a transistor package provides no clue as to which lead is the emitter, which lead is the base, and which lead is the collector. Old can-style packaging includes a protruding tab that usually points toward the emitter, but not always. Where power transistors are packaged in a metal enclo­sure, it is typically connected internally with the collector. In the case of surface-mount transis­ tors, look for a dot or marker that should identify the base of a bipolar transistor or the gate of a field-effect transistor.

A through-hole transistor usually has its part number printed or engraved on its package, al­

though a magnifying glass may be necessary to see this. The component’s datasheet may then be checked online. If a datasheet is unavailable, meter-testing will be necessary to confirm the functions of the three transistor leads. Some mul­timeters include a transistor-test function, which may validate the functionality of a transistor while also displaying its beta value. Otherwise, a meter can be put in diode-testing mode, and an unpowered NPN transistor should behave as if diodes are connected between its leads as shown in Figure 28-5. Where the identities of the transistor’s leads are unknown, this test will be sufficient to identify the base, after which the collector and emitter may be determined empir­ically by testing the transistor in a simple low- voltage circuit such as that shown in Figure 28-6.


Figure 28-5. An NPN transistor can behave as if it contains two diodes connected as shown. Where the functions of the leads of the transistor are unknown, the base can be identified by testing for conductivity.

How to Use it

The following abbreviations and acronyms are common in transistor datasheets. Some or all of the letters following the initial letter are usually, but not always, formatted as subscripts:


Forward current gain


Same as hFE


Voltage between collector and emitter (no

connection at base)

clip_image019ICM IBM PTOT


Maximum peak current at collector Maximum peak current at base

Total maximum power dissipation at room temperature

Maximum junction temperature to avoid damage

Figure 28-6. This simple schematic can be used to breadboard-test a transistor empirically, determining its functionality and the identities of its collector and emitter leads.


Voltage between collector and base (no con­

nection at emitter)


Voltage between emitter and base (no con­

nection at collector)

VCE sat

Saturation voltage between collector and


VB ,Esat

Saturation voltage between base and emit­



Current measured at collector

Often these terms are used to define “absolute maximum values” for a component. If these max­ imums are exceeded, damage may occur.

A manufacturer’s datasheet may include a graph showing the safe operating area (SOA) for a tran­sistor. This is more common where power tran­sistors are involved, as heat becomes more of an issue. The graph in Figure 28-7 has been adapted from a datasheet for a silicon diffused power transistor manufactured by Philips. The safe op­erating area is bounded at the top by a horizontal segment representing the maximum safe cur­ rent, and at the right by a vertical segment rep­ resenting the maximum safe voltage. However, the rectangular area enclosed by these lines is reduced by two diagonal segments representing the total power limit and the second breakdown limit. The latter refers to the tendency of a tran­sistor to develop internal localized “hot spots” that tend to conduct more current, which makes them hotter, and able to conduct better—ulti­mately melting the silicon and causing a short circuit. The total power limit and the second breakdown limit reduce the safe operating area, which would otherwise be defined purely by maximum safe current and maximum safe volt­ age.

Uses for discrete transistors began to diminish when integrated circuits became cheaper and started to subsume multi-transistor circuits. For instance, an entire 5-watt audio amplifier, which used to be constructed from multiple compo­nents can now be bought on a chip, requiring just


Figure 28-7. Adapted from a Philips datasheet for a power transistor, this graph defines a safe operating area (SOA) for the component. See text for details.

a few external capacitors. More powerful audio equipment typically uses integrated circuits to process inputs, but will use individual power transistors to handle high-wattage output.

Darlington Pairs

Discrete transistors are useful in situations where current amplification or switching is required at just one location in a circuit. An example would be where one output pin from a microcontrol­ler must switch a small motor on and off. The motor may run on the same voltage as the mi­crocontroller, but requires considerably more current than the typical 20mA maximum avail­ able from a microcontroller output. A Darlington pair of transistors may be used in this application. The overall gain of the pair can be 100,000 or more. See Figure 28-8. If a power source feeding

through a potentiometer is substituted for the microcontroller chip, the circuit can function as a motor speed control (assuming that a generic DC motor is being used).

In the application shown here, the microcontrol­ler chip must share a common ground (not shown) with the transistors. The optional resistor may be necessary to prevent leakage from the first transistor (when in its “off” state) from trig­ gering the second. The diode protects the tran­sistors from voltage transients that are likely when the motor stops and starts.


Figure 28-8. Where the emitter of one NPN transistor is coupled to the base of another, they form a Darlington pair (identified by the dashed rectangle in this schemat- ic). Multiplying the gain of the first transistor by the gain of the second gives the total gain of the pair.

A Darlington pair can be obtained in a single transistor-like package, and may be represented by the schematic symbol shown in Figure 28-9.

Various through-hole Darlington packages are shown in Figure 28-10.


Figure 28-9. When a Darlington pair is embedded in a single transistor-like package, it may be represented by this schematic symbol. The leads attached to the package can be used as if they are the emitter, base, and collector of a single NPN transistor.


Figure 28-10. Various packaging options for Darlington pairs. From left to right: The 2N6426 contains a Darling- ton pair rated to pass up to 500mA continuous collector current. The 2N6043 is rated for 8A continuous. The ULN2003 and ULN2083 chips contain seven and eight Darlington pairs, respectively.

Seven or eight Darlington pairs can be obtained in a single integrated chip. Each transistor pair in these chips is typically rated at 500mA, but they can be connected in parallel to allow higher cur­ rents. The chip usually includes protection di­ odes to allow it to drive inductive loads directly.

A typical schematic is shown in Figure 28-11. In this figure, the microcontroller connections are hypothetical and do not correspond with any ac­ tual chip. The Darlington chip is a ULN2003 or similar, containing seven transistor pairs, each with an “input” pin on the left and an “output”

pin opposite it on the right. Any of pins 1 through 7 down the left side of the chip can be used to control a device connected to a pin on the op­ posite side.

A high input can be thought of as creating a neg­ ative output, although in reality the transistors inside the chip are sinking current via an external device—a motor, in this example. The device can have its own positive supply voltage, shown here as 12VDC, but must share a common ground with the microcontroller, or with any other compo­ nent which is being used on the input side. The lower-right pin of the chip shares the 12VDC sup­ ply because this pin is attached internally to clamp diodes (one for each Darlington pair), which protect against surges caused by induc­ tive loads. For this reason, the motor does not have a clamp diode around it in the schematic.

The Darlington chip does not have a separate pin for connection with positive supply voltage, be­ cause the transistors inside it are sinking power from the devices attached to it.


Figure 28-11. A chip such as the ULN2003 contains sev- en Darlington pairs. It will sink current from the device it is driving. See text for details.

A surface-mount Darlington pair is shown in

Figure 28-12. This measures just slightly more

than 0.1” long but is still rated for up to 500mA collector current or 250mW total power dissipa­tion (at a component temperature no higher than 25 degrees Centigrade).


Figure 28-12. A surface-mount package for a Darlington pair. Each square in the background grid measures 0.1”. See text for additional details.


Two basic types of transistor amplifiers are shown in Figure 28-13 and Figure 28-14. The common-collector configuration has current gain but no voltage gain. The capacitor on the input side blocks DC current from entering the ampli­fier circuit, and the two resistors forming a volt­ age divider on the base of the transistor establish a voltage midpoint (known as the quiescent point or operating point) from which the signal to be amplified may deviate above and below.

The common-emitter amplifier provides voltage gain instead of current gain, but inverts the phase of the input signal. Additional discussion of amplifier design is outside the scope of this encyclopedia.

In switching applications, modern transistors have been developed to handle a lot of current compared with earlier versions, but still have some limitations. Few power transistors can han­ dle more than 50A flowing from collector to emitter, and 1,000V is typically a maximum value. Electromechanical relays continue to exist be­ cause they retain some advantages, as shown in the table in Figure 28-15, which compares switching capabilities of transistors, solid-state relays, and electromechanical relays.


Figure 28-13. The basic schematic for a common- collector amplifier. See text for details.


Figure 28-14. The basic schematic for a common-emitter amplifier. See text for details.


Figure 28-15. A comparison of characteristics of switching devices.

What Can Go Wrong
Wrong Connections on a Bipolar Transistor

Failing to identify a transistor’s leads or contacts correctly can obviously be a potential source of damage, but swapping the collector and emitter

accidentally will not necessarily destroy the tran­sistor. Because of the inherent symmetry of the device, it will in fact function with collector and emitter connections reversed. Rohm, a large semiconductor manufacturer, has included this scenario in its general information pages and concludes that the primary indicator of trans­ posed connections is that the β value, or hFE , drops to about 1/10th of specification. If you are using a transistor that works but provides much less amplification than you expect, check that the emitter and collector leads are not transposed.

Wrong Connections on a Darlington Pair Chip

While a single-component package for a Dar­lington pair functions almost indistinguishably from a single transistor, multiple Darlington pairs in a DIP package may create confusion because the component behaves differently from most other chips, such as logic chips.

A frequent error is to ground the output device instead of applying positive power to it. See Figure 28-11 and imagine an erroneous connec­tion of negative power instead of the 12VDC pos­ itive power.

Additional confusion may be caused by reading a manufacturer’s datasheet for a Darlington pair chip such as the ULN2003. The datasheet depicts the internal function of the chip as if it contains logic inverters. While the chip can be imagined as behaving this way, in fact it contains bipolar transistors that amplify the current applied to the base of each pair. The datasheet also typically will not show the positive connection that should be made to the common-diode pin (usually at bottom-right), to provide protection from surges caused by inductive loads. This pin must be dis­

tinguished carefully from the common-ground

pin (usually at bottom-left). The positive connec­tion to the common-diode pin is optional; the common-ground connection is mandatory.
Soldering Damage

Like any semiconductor, transistors are vulnera­ble to heat and can be damaged while soldering, although this seldom happens if a low-wattage iron is used. A copper alligator clip can be applied as a heat sink to each lead before it is soldered.

Excessive Current or Voltage

During use, a transistor will be damaged if it is subjected to current or voltage outside of its rat­ed range. Passing current through a transistor without any series resistance to protect it will al­ most certainly burn it out, and the same thing can happen if incorrect resistor values are used.

The maximum wattage that a transistor can dis­sipate will be shown in its datasheet. Suppose, for example, this figure is 200mW, and you are using a 12VDC supply. Ignoring the base current, the maximum collector current will be 200 / 12 = approximately 15mA. If the transistor’s emitter is connected to ground, and the load applied to the transistor output has a high impedance, and if we ignore the transresistance, Ohm’s Law sug­gests that a resistor that you place between the collector and the supply voltage should have a resistance of at least 12 / 0.015 = 800 ohms.

When transistors are used in switching applica­tions, it is customary for the base current to be 1/5th of the collector current. In the example dis­ cussed here, a 4.7K resistor might be appropriate. A meter should be used to verify actual current and voltage values.

Excessive Leakage

In a Darlington pair, or any other configuration where the output from one transistor is connec­ted with the base of another, leakage from the first transistor while in its “off” state can be am­ plified by the second transistor. If this is unac­ceptable, a bypass resistor can be used to divert some of the leakage from the base of the second transistor to ground. Of course the resistor will also steal some of the base current when the first transistor is active, but the resistor value is typi­cally chosen so that it takes no more than 10% of the active current. See Figure 28-8 for an example of a bypass resistor added to a Darlington pair.