VOLTAGE REGULATORS AND INTEGRATED CIRCUITS
A voltage regulator is an electronic device connected in the output of a power supply to maintain the output voltage at its constant rated value. It reacts automatically within its rated limits to any variations in the output voltage. Should the output voltage rise or fall, the voltage regulator automatically compensates for the change and maintains the output voltage at the required value. Although there may be large changes in load current drawn from a power supply or changes in the applied voltage, the voltage regulator maintains a constant output voltage.
The regulating action of the voltage regulator is, in effect, that of a variable resistor that responds to any changes in the current flowing through it. This "variable resistance" may be in series or in shunt with the load.
Figure 1: Series Voltage RegulatorAs in any series circuit, the current through R and RL will depend upon the value of source voltage and the total amount of resistance (R + RL) in the circuit. It can be seen that the amount of current through RL, and therefore the voltage drop across RL, will be dependent on the setting of R. If the value of R is increased, the current will decrease and the voltage drop across RL will also decrease. If the value of R is decreased, the value of total resistance will also decrease and current will increase, causing a larger voltage drop to be felt across RL. If a fixed value for the voltage across RL is set, the regulator will ensure that this value remains constant by properly varying the resistor, R, to compensate for circuit changes.
If the input voltage increases without changing R, the output voltage will increase. However, if R is increased in value, a larger voltage drop across R results. The increases in the value of R and the voltage drop across it are proportional to the rise in input voltage (a 5 V increase in the input results in a 5 V increase in the drop across R) so that the output voltage (the drop across RL) remains exactly what it was before the increased input voltage. Conversely, a decrease in the input voltage requires a decrease in the value of R so that the voltage drop across R is proportional to the decreased input voltage (a 5 V decrease in the input results in a 5 V decrease in the voltage drop across R) and the output voltage remains constant.
Figure 2: Shunt Voltage Regulating CircuitAs in the simple series voltage regulator, a voltage dividing action is used to obtain regulation. The shunt-regulating device (RV) now determines the amount of current through RS. The larger the current through RS, the higher its voltage drop, and the lower the voltage across RL.
The shunt voltage regulator operates in the following manner. RS (the series resistance) is in series with RV (the shunt regulating device) and RL (the load impedance). RV and RL, being in parallel, will have the same voltage drop across them. RS, being in series, will have the total current flowing through RV and RL flowing through it. To keep the voltage constant, RV must be adjusted to compensate for changes in the input voltage as well as for changes caused by load impedance changes.
Analysis of an increasing source voltage will exemplify the regulating action. The voltage across RV and RL would tend to increase with this increase of source voltage. To compensate for this undesired change, the value of RV must be decreased. This results in more current flowing through the entire circuit (the total resistance has been decreased), and the increase in current through RS causes a larger voltage drop across RS. Consequently, the output voltage remains constant.
Conversely, when the source voltage decreases, the voltage across the parallel legs, RV and RL, would tend to decrease. To compensate for this change, the value of RV must be increased. This increases the total resistance and decreases total current. Less current through RS means less voltage drop across RS and more voltage available across RV and RL. Once again, the voltage is held constant. The amount of voltage increase or decrease from the source is the same amount of increase or decrease in the voltage drop across RS, resulting in the desired amount of voltage for the load.
If the value of RL were to decrease (the equivalent of adding another resistor in parallel with RV and RL), the total resistance would decrease. Current would increase, causing a greater voltage drop across RS. This would tend to cause the output voltage to decrease. To compensate for this change, the value of RV must be increased, reducing the amount of current through RS and bringing the output voltage to the desired amount. If the value of RL were to increase (the equivalent of removing a circuit from the system), the total resistance would increase. Total current would decrease, and the decrease in current through RS would result in a lower voltage drop across the element. This would tend to cause the output voltage to rise. To compensate for this change, the value of RV must be decreased, increasing the amount of current through RS, increasing the voltage drop across RS, and decreasing the voltage across the load to the desired amount.
Both the voltage regulators discussed rely upon a mechanical adjustment and would require continuous monitoring of the equipment to obtain, at best, barely satisfactory results. These voltage regulators were used for explanation purposes. Electronic devices (discussed next) accomplish the function of the variable resistors electronically and automatically to provide excellent results immediately.
Figure 3: Zener Diode Shunt Type Regulator CircuitThe voltage across the Zener is constant, thus holding the voltage across the parallel load RL constant. Although the circuit shown depicts a positive voltage output, it is a simple matter to have a negative output voltage and to reverse the Zener and the polarities shown in Figure 3.
The value of RS in Figure 3 has been fixed so that it can handle the combined currents of the diode and the load and still allow the diode to conduct well within the breakdown region. RS stabilizes the load voltage by dropping the difference between the diode operating voltage and the unregulated input voltage. The Zener diode is a PN junction that has been specially doped during its manufacture so that when reverse-biased it will operate at a specific breakdown voltage level. It operates well within its rated tolerance over a considerable range of reverse current.
If input voltage to the regulator circuit decreases, the voltage across the Zener diode must also decrease, causing Zener current to decrease. The total current in the circuit decreases, much the same as when the value of RV was increased in the simple shunt regulator earlier. The current through RS, having decreased, results in a lower voltage drop across RS. This results in a voltage drop across the Zener, bringing load back to the desired voltage.
When input voltage increases, the change is immediately felt across the Zener. This effectively biases the Zener so that there is an increase in Zener current. The increase in Zener current means an increase in total circuit current. RS, in series with the source, will have an increase in current through it, resulting in a larger voltage drop across it. With the larger drop across RS, the voltage across the Zener, and, therefore the load, is reduced to the desired output voltage.
The same basic materials and techniques that are used to construct a bipolar transistor are used to construct an integrated circuit. Therefore, the cost of manufacturing a complete integrated circuit, encompassing the equivalent of perhaps dozens of transistors and other semiconductors, is theoretically only slightly greater than that of a single transistor.
The small size of the integrated circuit is its most apparent advantage. A typical IC can be constructed on a piece of semiconductor material that is less than one tenth of an inch square. Even when the IC is suitably packaged, it still occupies only a small amount of space. In most applications where size and weight must be reduced to an absolute minimum, the IC can make a significant contribution.
The small size of the integrated circuit also produces other fringe benefits. The smaller circuits consume less power than conventional circuits, and they cost less to operate. They generate less heat and, therefore, generally do not require elaborate cooling or ventilation systems. The smaller circuits are also capable of operating at higher speeds because it takes less time for signals to travel through them.
The integrated circuit is also more reliable than a conventional circuit that is formed with discrete components. This greater reliability results because every component within the IC is a solid-state device, and these components are permanently connected together with thin layers of metal. They are not soldered together like the components in a conventional circuit; therefore, circuit failure due to faulty connections is less likely to occur. Also, all of the components within the IC are simultaneously formed, as opposed to conventional circuitry, which is assembled in a step-by-step manner. Integrated circuits are also thoroughly tested after they are assembled, and only those devices that meet the required specifications are considered suitable for commercial applications.
Integrated circuits also have certain limitations that make them unsuitable for some applications. Since the IC is an extremely small device, it cannot handle large currents or voltages. High currents generate heat within the device, and the tiny components can be easily damaged if the heat becomes excessive. High voltages can break down the insulation between the components in the IC because the components are very close together. This can result in shorts between adjacent components that would make the IC completely useless. Therefore, most ICs are low-power devices with low operating currents (in the milliampere range) and low operating voltages (5 to 20 volts). Also, most ICs have power dissipation ratings of less than 1 watt.
Many ICs are simultaneously formed on a single wafer. The number of ICs formed on a wafer will depend on the size of the wafer and the size of the individual ICs. All of the ICs formed on a wafer usually are of the same type and, therefore, have the same physical dimensions. Also, each IC contains the same number and type of components.
When all of the ICs have been simultaneously formed, the wafer is sliced into many sections. These sections are commonly referred to as chips. Each chip represents one complete integrated circuit and contains all of the components and wiring associated with that circuit. Once the ICs are separated into individual chips, each IC must be mounted in a suitable package and tested. Throughout this entire fabrication process, certain ICs are rejected because they have various physical or electrical deficiencies. Although an extremely large number of ICs are fabricated in any given process, only a relatively small number of these devices are found to be usable.
The components that are commonly used in ICs are diodes, transistors, resistors, and capacitors. These components can be formed by diffusing impurities into selected regions of a semiconductor wafer (substrate) to produce PN junctions at specific locations. The basic manner in which these four components are formed and the manner in which they are interconnected is shown in Figure 4.
Figure 4: IC ConstructionThe bottom of Figure 4 shows a simple electronic circuit consisting of a capacitor, a PN junction diode, an NPN transistor, and a resistor. Operating voltages and currents can be applied to the circuit through terminals 1, 2, and 3. This circuit could be easily constructed using four discrete components; however, it can also be produced as a monolithic integrated circuit.
Notice that all four of the components are formed within a P-type substrate or wafer. These components are simultaneously formed by diffusing N-type and P-type impurities into the P-type substrate to produce N-type or P-type regions. Several diffusion operations are necessary to form the entire IC, and it is necessary to accurately control the location, the size, and the depth of each N-type or P-type region that is formed. This is accomplished by depositing an insulating layer of silicon oxide on top of the substrate. Then, appropriate windows are cut in the oxide coating by means of an acid so that only the desired areas on the substrate are exposed. The N-type impurities are diffused through the windows and into the substrate to form the first and largest N-type regions. Next, the windows are covered with new oxide and new windows are formed. Then, a P-type impurity is diffused into the substrate at the appropriate locations to form P-type regions. This process is repeated again to form the final N-type regions.
Notice that the NPN transistor is constructed in basically the same manner as a conventional transistor. Its N-type collector was formed first, and then its P-type base and N-type emitter regions were formed. All three regions extend to the top of the substrate and lie in a flat plane.
The PN junction diode is also equivalent to a conventional diode. The larger N-type region that was first diffused into the substrate serves as the cathode, and the smaller P-type region that was diffused into the N-type region serves as the anode.
The resistor is formed by first producing a large N-type region and then forming a P-type region on top of it. This produces a long narrow strip of P-type material that is surrounded by N-type material. This long P-type region serves as a resistor; adjusting its length and width can control its resistance. Increasing the length will result in a higher resistance, while increasing the width will lower the resistance. The resistance value can also be adjusted by controlling the concentration of impurities within the P-type strip. A higher concentration of impurities will result in a lower resistance and vice-versa.
In general, the amount of space required to produce a resistor increases with its value, thus making it difficult to place large resistance values within an IC. Also, it is difficult to produce resistors with highly accurate values. However, the ratio between two resistors that are formed close together on a substrate can be regulated with a high degree of accuracy. Therefore, circuits that are produced in IC form generally are designed to take advantage of the accurate resistance ratios that are available, instead of specific resistance values.
The capacitor is constructed in a unique manner. The diffused N-type region serves as the bottom plate of the capacitor, and the layer of silicon oxide serves as the dielectric. The top plate of the capacitor is simply a layer of metal that has been deposited on top of the oxide coating. This type of capacitor is sometimes referred to as a metal-oxide capacitor.
The area of the plates, the thickness of the oxide layer, and the dielectric constant of the oxide layer determine the value of capacitance. In general, the amount of capacitance obtainable per unit area is very low, and it is necessary to make the capacitor quite large to obtain a significant value of capacitance. In most cases, it is not practical to construct capacitors with values that are much higher than a few hundred picofarads.
All of the components are simultaneously formed by coating the substrate with silicon oxide, cutting appropriate windows into the oxide coating, and diffusing impurities into the substrate. This process is repeated several times until all components are formed. In the final steps of the process, an oxide film is deposited over the substrate, and windows are cut into the oxide to expose the various regions within each component. Then, a layer of aluminum or gold is deposited over the oxide and allowed to make contact with the exposed regions. Next, acid is used to etch away certain portions of the metal film, leaving behind narrow strips of metal that serve as conductors to interconnect the various components and form a complete circuit. Notice in Figure 4 that the four components are connected together by thin metal strips. The three terminals that provide access to the integrated circuit are also shown. The metal strips that connect to terminals 1, 2, and 3 are brought out to the perimeter of the IC and enlarged to form rectangular pads. These pads provide relatively large metal surfaces to which fine metal wires can be bonded when the IC is permanently installed in a suitable package.
Like transistors and other types of solid-state components, ICs are mounted in packages that protect them from moisture, dust, and other types of contaminants. The packages also make it easier to install the ICs in various types of equipment, since each package contains leads, which can be either plugged into matching sockets or soldered to adjacent components or conductors.
The most popular IC package is the dual in-line package, which is commonly referred to as a DIP. Figure 5 shows a typical dual in-line package. Notice that the package has two rows of mounting pins, or leads, that can be either inserted in a matching socket or inserted in holes on a printed circuit board and soldered in place. Dual in-line packages may be constructed from either plastic or ceramic materials and commonly have a total of 14 to 16 leads, although larger and smaller units are available.
Figure 5: IC Dual In-Line Package (DIP)Another type of IC package that is widely used is the flat-pack, shown in Figure 6. The flat-pack is similar to the dual in-line package, but it is smaller and much thinner. Notice that the device is very thin and that its leads extend horizontally outward around its edges. The flat-pack can be mounted almost flush with the surface of a printed circuit board, and its leads usually are soldered directly to adjacent conductor pads on the board. The flat-pack is, therefore, used where space is limited. It is often made from metal or ceramic materials and can be used over a wide range of operating temperatures. The popular units often have a total of 10 or 14 leads, although larger and smaller units are also available.
Figure 6: Flat PackIntegrated circuits may also be mounted in metal cans that are similar to those used to house transistors. A typical metal can is shown in Figure 7. The metal can shown has only eight leads, although larger devices with more leads are available. Their long metal leads make it possible to install them in a variety of ways.
Figure 7: Metal Can ICAfter the integrated circuits are packaged in their respective containers, they are put through a series of tests to make sure that they meet certain electrical specifications. Since the performance of any circuit is affected by changes in temperature, the ICs must be tested over a wide range of operating temperatures to be sure that they are suitable for commercial applications.
Figure 8: TTL Logic CircuitThe TTL circuit performs an important logic function. It is capable of comparing two input voltage levels equal to either 0 volts or approximately 5 volts and providing an output voltage level dependent on the input combination. The circuit performs what is commonly referred to as the NAND function, and the circuit itself is referred to as a NAND gate. The NAND gate usually is represented by the symbol shown in Figure 9. Notice that only the two inputs and the output of the circuit are represented.
Figure 9: NAND Function SymbolIt is common practice to construct four of these NAND gate circuits on a single IC chip and mount the chip in a single package. Both dual in-line packages and flat-packs are widely used with ICs of this type.
A typical example of a digital IC formed using MOS construction techniques is shown in Figure 10. It contains both P-channel and N-channel enhancement-mode MOS field-effect transistors and is commonly referred to as a complementary metal-oxide semiconductor (CMOS) IC. CMOS circuits consume less power than other types of digital ICs, and they have good temperature stability. They can operate over a wide range of supply voltages, typically 3 to 15 volts. CMOS circuits also have high input resistance, which makes it possible to connect a large number of circuit inputs to a single output without loading down the output and disrupting circuit operation. This is an extremely important advantage in digital equipment, where thousands of circuits are used.
Figure 10: CMOS Digital ICThe circuit shown in Figure 10 contains four MOSFETs that are interconnected to perform a useful logic function. The resulting circuit is referred to as a NOR gate, and its symbol is shown in Figure 11. This symbol generally is used in place of the actual schematic. Four of these NOR gates usually are formed on one IC chip and mounted in a single IC package.
Figure 11: NOR Logic Function SymbolBoth TTL and CMOS circuits can be used to perform NAND or NOR functions, or a variety of other logic functions that must be performed in a highly complex digital system.
A typical operational amplifier circuit contains transistors, resistors, and capacitors that are interconnected to form a highly efficient amplifying circuit. The circuit has two inputs and one output, as shown. One input is commonly referred to as the non-inverting input (+), and the other is referred to as the inverting input (-). The circuit will amplify either DC or AC signals applied to either input. Signals applied to the (+) input are not inverted when they appear at the output. As the input voltage goes positive or negative, the output voltage correspondingly goes positive or negative. When a signal is applied to the (-) input, inversion takes place. In other words, the polarity of the output signal is always opposite to that of the input signal.
When the input voltage is equal to zero, the output voltage should also be equal to zero. However in practice, the output voltage may be offset by a slight amount since component tolerances make it impossible to construct a perfectly balanced circuit. Therefore, two offset null terminals are provided so that the circuit can be appropriately balanced. Connecting the opposite ends of a potentiometer to the offset null terminals while the arm of the potentiometer is connected to circuit ground does this. The potentiometer may then be adjusted to balance the circuit. The operational amplifier is commonly represented by the symbol shown in Figure 12. Notice that the (+) and (-) inputs are identified in the symbol.
The operational amplifier circuit generally is packaged in a variety of ways, such as a dual in-line packaging or in a metal can, to suit a broad range of applications.
Figure 12: Operational Amp Linear IC