Today, you’ll still hear about diodes, anodes, and cathodes. But rather than large, heavy, hot, high-voltage tubes, diodes are tiny things made from silicon or other semiconducting materials. Some diodes can handle voltages nearly as great as their tube counterparts. Semiconductor diodes can do just about everything that tube diodes could, plus a few things that people in the tube era probably never imagined.
The most common and important applications of a diode are the following:
The hallmark of a rectifier diode is that it passes current in only one direction. This makes it useful for changing ac to dc. Generally speaking, when the cathode is negative with respect to the anode, current flows; when the cathode is positive relative to the anode, there is no current. The constraints on this behavior are the forward break over and avalanche voltages.
Suppose a 60-Hz ac sine wave is applied to the input of the circuit in the below figure. During half the cycle, the diode conducts, and during the other half, it doesn’t. This cuts off half of every cycle. Depending on which way the diode is hooked up, either the positive half or the negative half of the ac cycle will be removed. Remember that electrons flow from negative to positive, against the arrow in the diode symbol.
The circuit and wave diagram shows a half-wave rectifier circuit. This is the simplest possible rectifier. That’s its chief advantage over other, more complicated rectifier circuits.
One of the earliest diodes, existing even before vacuum tubes, was a semiconductor. Known as a cat whisker, this semiconductor consisted of a fine piece of wire in contact with a small piece of the mineral galena. This bizarre-looking thing had the ability to act as a rectifier for small radio-frequency (RF) currents. When the cat whisker was connected in a circuit like that of the figure, the result was a receiver capable of picking up amplitude-modulated (AM) radio signals.
A cat whisker was a finicky thing. Engineers had to adjust the position of the fine wire to find the best point of contact with the galena. Tweezers and magnifying glasses were invaluable in this process. A steady hand was essential.
The galena, sometimes called a “crystal,” gave rise to the nickname crystal set for this low-sensitivity radio. You can still build a crystal set today, using a simple RF diode, coil, a tuning capacitor, a headset, and a long-wire antenna. Notice that there’s no battery! The audio is provided by the received signal alone.
The diode in the figure acts to recover the audio from the radio signal. This is called detection; the circuit is a detector. If the detector is to be effective, the diode must be of the right type. It should have low capacitance, so that it works as a rectifier at radio frequencies, passing current in one direction but not in the other. Some modern RF diodes are actually microscopic versions of the old cat whisker, enclosed in a glass case with axial leads. You have probably seen these in electronics hobby stores.
When current passes through a diode, half of the cycle is cut off. This occurs no matter what the frequency, from 60-Hz utility current through RF, as long as the diode capacitance is not too great. The output wave from the diode looks much different than the input wave. This condition is known as nonlinearity. Whenever there is a nonlinearity of any kind in a circuit—that is, whenever the output waveform is shaped differently from the input wave-form—there will be harmonic frequencies in the output. These are waves at integer multiples of the input frequency. Often, nonlinearity is undesirable. Then engineers strive to make the circuit linear so that the output waveform has exactly the same shape as the input waveform. But sometimes a circuit is needed that will produce harmonics. Then nonlinearity is introduced deliberately. Diodes are ideal for this. A simple frequency-multiplier circuit is shown in the figure. The output LC circuit is tuned to the desired nth harmonic frequency, nfo, rather than to the input or fundamental frequency, fo.
For a diode to work as a frequency multiplier, it must be of a type that would also work well as a detector at the same frequencies. This means that the component should act as a rectifier, but not like a capacitor.
When two waves having different frequencies are combined in a nonlinear circuit, new frequencies are produced. These new waves are at the sum and difference frequencies of the original waves. You’ve probably noticed this mixing, also called heterodyning, if you’ve ever heard two loud, sine wave tones at the same time.
Suppose there are two signals with frequencies fl and f2. For mathematical convenience, assign f2 to the wave with the higher frequency. If these signals are combined in a nonlinear circuit, new waves will result. One of them will have a frequency f2 − fl, and the other will be at f2 + f1. These are known as beat frequencies. The signals are called mixing products.
The above figure is a frequency domain graph. Amplitude (on the vertical scale) is shown as a function of frequency (on the horizontal scale). This kind of display is what engineers see when they look at the screen of a spectrum analyzer. Most of the graphs you’ve seen so far have been time-domain graphs, in which things are shown as a function of time. The screen of an oscilloscope normally shows things in the time domain.
The ability of diodes to conduct with forward bias, and to insulate with reverse bias, makes them useful for switching in some electronic applications. Diodes can switch at extremely high rates, much faster than any mechanical device.
One type of diode, made for use as an RF switch, has a special semiconductor layer sandwiched in between the P-type and N-type material. This layer, called an intrinsic semiconductor, reduces the capacitance of the diode so that it can work at higher frequencies than an ordinary diode. The intrinsic material is sometimes called I type. A diode with I-type semiconductor is called a PIN diode.
Direct-current bias, applied to one or more PIN diodes, allows RF currents to be effectively channeled without using complicated relays and cables. A PIN diode also makes a good RF detector, especially at frequencies above 30 MHz.
Most diodes have avalanche breakdown voltages much higher than the reverse bias ever gets. The value of the avalanche voltage depends on how a diode is manufactured. Zener diodes are made to have well-defined, constant avalanche voltages. Suppose a certain Zener diode has an avalanche voltage, also called the Zener voltage, of 50 V. If a reverse bias is applied to the P-N junction, the diode acts as an open circuit below 50 V. When the voltage reaches 50 V, the diode starts to conduct. The more the reverse bias tries to increase, the more current flows through the P-N junction. This effectively prevents the reverse voltage from exceeding 50 V.
The current through a Zener diode, as a function of the voltage, is shown in the below figure.
The Zener voltage is indicated by the abrupt rise in reverse current as of the reverse bias increases. A typical Zener-diode voltage-limiting circuit is shown in the below figure.
There are other ways to get voltage regulation besides the use of Zener diodes, but Zener diodes often provide the simplest and least expensive alternative. Zener diodes are available with a wide variety of voltage and power-handling ratings. Power supplies for solid-state equipment commonly employ Zener diode regulators.
The forward breakover voltage of a germanium diode is about 0.3 V; for a silicon diode, it is about 0.6 V. A diode will not conduct until the forward bias voltage is at least as great as the forward breakover voltage. The “flip side” is that the diode will always conduct when the forward bias exceeds the breakover value. In this case, the voltage across the diode will be constant: 0.3 V for germanium and 0.6 V for silicon.
This property can be used to advantage when it is necessary to limit the amplitude of a signal, as shown in the below figure.
By connecting two identical diodes back-to-back in parallel with the signal path (A), the maximum peak amplitude is limited, or clipped, to the forward breakover voltage of the diodes. The input and output waveforms of a clipped signal are illustrated at B. This scheme is sometimes used in radio receivers to prevent “blasting” when a strong signal comes in.
The downside of the diode limiter circuit is that it introduces distortion when limiting is taking place. This might not be a problem for the reception of Morse code, or for signals that rarely reach the limiting voltage. But for voice signals with amplitude peaks that rise well past the limiting voltage, it can seriously degrade the audio quality, perhaps even rendering the words indecipherable.
When a diode is reverse-biased, there is a region at the P-N junction with dielectric properties. As you know from the last chapter, this is called the depletion region, because it has a shortage of majority charge carriers. The width of this zone depends on several things, including the reverse voltage.
As long as the reverse bias is less than the avalanche voltage, varying the bias can change the width of the depletion region. This results in a change in the capacitance of the junction. The capacitance, which is always quite small (on the order of picofarads), varies inversely with the square root of the reverse bias.
Some diodes are manufactured especially for use as variable capacitors. These are varactor diodes. Sometimes you’ll hear them called varicaps. They are made from silicon or gallium arsenide.
A common use for a varactor diode is in a circuit called a voltage-controlled oscillator (VCO). A voltage-tuned circuit, using a coil and a varactor, is shown in the below figure.
This is a parallel-tuned circuit. The fixed capacitor, whose value is large compared with that of the varactor, serves to keep the coil from short-circuiting the control voltage across the varactor. Notice that the symbol for the varactor has two lines on the cathode side. This is its “signature,” so that you know that it’s a varactor and not just an ordinary diode.
Oscillation and amplification
Under certain conditions, diodes can be made to produce microwave radio signals. There are three types of diodes that do this: Gunn diodes, IMPATT diodes, and tunnel diodes.
A Gunn diode can produce up to 1 W of RF power output, but more commonly it works at levels of about 0.1 W. Gunn diodes are usually made from gallium arsenide. A Gunn diode oscillates because of the Gunn effect, named after J. Gunn of International Business Machines (IBM) who observed it in the sixties. A Gunn diode doesn’t work anything like a rectifier, detector, or mixer; instead, the oscillation takes place as a result of a quirk called negative resistance.
Gunn-diode oscillators are often tuned using varactor diodes. A Gunn-diode oscillator, connected directly to a microwave horn antenna, is known as a Gunnplexer. These devices are popular with amateur-radio experimenters at frequencies of 10 GHz and above.
The acronym IMPATT comes from the words impact avalanche transit time. This, like negative resistance, is a phenomenon the details of which are rather esoteric. An IMPATT diode is a microwave oscillating device like a Gunn diode, except that it uses silicon rather than gallium arsenide.
An IMPATT diode can be used as an amplifier for a microwave transmitter that employs a Gunn-diode oscillator. As an oscillator, an IMPATT diode produces about the same amount of output power, at comparable frequencies, as the Gunn diode.
Another type of diode that will oscillate at microwave frequencies is the tunnel diode, also known as the Esaki diode. It produces only a very small amount of power, but it can be used as a local oscillator in a microwave radio receiver.
Tunnel diodes work well as amplifiers in microwave receivers because they generate very little unwanted noise. This is especially true of gallium arsenide devices.
The behavior of Gunn, IMPATT, and tunnel diodes is a sophisticated topic and is beyond the scope of this book. College-level electrical-engineering texts are good sources of information on this subject. You will want to know about how these devices work if you plan to become a microwave engineer.
Some semiconductor diodes emit radiant energy when a current passes through the P-N junction in a forward direction. This phenomenon occurs as electrons fall from higher to lower energy states within atoms.
LEDs and IREDs
Depending on the exact mixture of semiconductors used in manufacture, visible light of almost any color can be produced. Infrared-emitting devices also exist. The most common color for a light-emitting diode (LED) is bright red. An infrared-emitting diode (IRED) produces wavelengths too long to see.
The intensity of the light or infrared from an LED or IRED depends to some extent on the forward current. As the current rises, the brightness increases up to a certain point. If the current continues to rise, no further increase in brilliance takes place. The LED or IRED is then said to be in a state of saturation.
Because LEDs can be made in various different shapes and sizes, they are ideal for use in digital displays. You’ve probably seen digital clock radios that use them. They are common in car radios. They make good indicators for “on/off,” “a. m. /p. m.,” “battery low,” and other conditions.
In recent years, LED displays have been largely replaced by liquid-crystal displays (LCDs). This technology has advantages over LEDs, including much lower power consumption and better visibility in direct sunlight.
Both LEDs and IREDs are useful in communications because their intensity can be modulated to carry information. When the current through the device is sufficient to produce output, but not enough to cause saturation, the LED or IRED output will follow along with rapid current changes. Voices, music, and digital signals can be conveyed over light beams in this way. Some modern telephone systems make use of modulated light, transmitted through clear fibers. This is known as fiberoptic technology. Special LEDs and IREDs produce coherent radiation; these are called laser diodes. The rays from these diodes aren’t the intense, parallel beams that you probably imagine when you think about lasers. A laser LED or IRED generates a cone-shaped beam of low intensity. But it can be focused, and the resulting rays have some of the same advantages found in larger lasers.