Microwave Concepts | Communication Systems | Size and Frequency

Microwave Concepts

Microwave are the ultrahigh, super high, and extremely high frequencies directly above the lower frequency ranges where most radio communication now takes place and below the optical frequencies that cover infrared, visible, and ultraviolet light. The outstanding benefits for radio communication of these extremely high frequencies and accompanying short wavelengths more than offset any problems connected with their use. Today, most new communication services and equipment use microwaves or millimeter-wave bands.

Microwave Frequencies and Bands

The practical microwave region is generally considered to extend from 1 to 30 GHz, although some definitions include frequencies up to 300 GHz. Microwave signals of 1 to 30 GHz have wavelengths of 30 cm (about 1 ft) to 1 cm (or about 0.4 in).

The microwave frequency spectrum is divided up into groups of frequencies, or bands, as shown in Fig. 16-1. Frequencies above 30 GHz are referred to as millimeter waves because their wavelength is only millimeters (mm). Note that parts of the L and S bands overlap part of the UHF band, which is 300 to 3000 MHz. Recent developments in semiconductor technology, such as smaller-geometry silicon and GaN, have made millimeter waves practical and useful. Frequencies above 300 GHz are in the submillimeter band. Currently, the only communication in the submillimeter ranges is for research and experimental activities.

Benefits of Microwaves

Every electronic signal used in communication has a finite bandwidth. When a carrier is modulated by an information signal, sidebands are produced. The resulting signal occupies a certain amount of bandwidth, called a channel, in the radio-frequency (RF) spectrum. Channel center frequencies are assigned in such a way that the signals using each channel do not overlap and interfere with signals in adjacent channels. As the number of communication signals and channels increases, more and more of the spectrum space is used up. Over the years as the need for electronic communication has increased, the number of radio communication stations has increased dramatically. As a result, the radio spectrum has become extremely crowded.

The use of the radio-frequency spectrum is regulated by the federal government. In the United States, this job is assigned to the Federal Communications Commission (FCC). The FCC establishes various classes of radio communication and regulates the assignment of spectrum space. For example, in radio and TV broadcasting, certain areas of the spectrum are set aside, and frequency assignments are given to stations. For two-way radio communication, other portions of the spectrum are used. The various classes of radio communication are assigned specific areas in the spectrum within which they can operate. Over the years, the available spectrum space, especially below 300 MHz, has essentially been used up. In many cases, communication services must share frequency assignments.

In some areas, new licenses are no longer being granted because the spectrum space for that service is completely full. In spite of this, the demand for new electronic communication channels continues. The FCC must, on an ongoing basis, evaluate users’ needs and demands and reassign frequencies as necessary. Many compromises have been necessary. Technological advances have helped solve some problems connected with overcrowding. For example, the selectivity of receivers has been improved so that adjacent channel interference is not as great. This permits stations to operate on more closely spaced frequencies.

On the transmitting side, new techniques have helped squeeze more signals into the same frequency spectrum. A classic example is the use of SSB, where only one sideband is used rather than two, thereby cutting the spectrum usage in half. Limiting the deviation of FM signals also helps to reduce bandwidth. In data communication, new modulation techniques such as PSK and QAM have been used to narrow the required bandwidth of transmitted information or to transmit at higher speeds in narrower bandwidths. Digital compression methods also transmit more information through a narrow channel. Multiplexing techniques help put more signals or information into a given bandwidth. Broadband schemes such as spread spectrum and orthogonal frequency-division multiplexing (OFDM) allow many radios to share a single bandwidth.

The other major approach to solving the problem of spectrum crowding has been to move into the higher frequency ranges. Initially, the VHF and UHF bands were tapped. Today, most new communication services are assigned to the microwave and millimeter-wave regions.

To give you some idea why more bandwidth is available at the higher frequencies, let’s take an example. Consider a standard AM broadcast station operating on 1000 kHz. The station is permitted to use modulating frequencies up to 5 kHz, thus producing upper and lower sidebands 5 kHz above and below the carrier frequency, or 995 and 1005 kHz. This gives a maximum channel bandwidth of 1005 – 995 =10 kHz. This bandwidth represents 10/1000 = 0.01 or 1 percent of the spectrum space at that frequency.

Now consider a microwave carrier frequency of 4 GHz. One percent of 4 GHz is 0.01×4,000,000,000= 40,000,000 or 40 MHz. A bandwidth of 40 MHz is incredibly wide. In fact, it represents all the low-frequency, medium-frequency, and high-frequency portions of the spectrum plus 10 MHz. This is the space that might be occupied by a 4-GHz carrier modulated by a 20-MHz information signal. Obviously, most information signals do not require that kind of bandwidth. A voice signal, e.g., would take up only a tiny fraction of that. A 10-kHz AM signal represents only 10,000/4,000,000,000 = 0.00025 percent of 4 GHz. Up to 4000 AM broadcast stations with 10-kHz bandwidths could be accommodated within the 40-MHz (1 percent) bandwidth.

Obviously, then, the higher the frequency, the greater the bandwidth available for the transmission of information. This not only gives more space for individual stations but also allows wide-bandwidth information signals such as video and high-speed digital data to be accommodated. The average TV signal has a bandwidth of approximately 6 MHz. It is impractical to transmit video signals on low frequencies because they use up entirely too much spectrum space. That is why most TV transmission is in the VHF and UHF ranges. There is even more space for video in the microwave region.

Wide bandwidth also makes it possible to use various multiplexing techniques to transmit more information. Multiplexed signals generally have wide bandwidths, but these can be easily handled in the microwave region. Finally, the transmission of high-speed binary information usually requires wide bandwidths, and these are also easily transmitted on microwave frequencies.

Disadvantages of Microwaves and Millimeter Waves

The higher the frequency, the more difficult it becomes to analyze electronic circuits. The analysis of electronic circuits at lower frequencies, say, those below 30 MHz, is based upon current-voltage relationships (circuit analysis). Such relationships are simply not usable at microwave frequencies. Instead, most components and circuits are analyzed in terms of electric and magnetic fields (wave analysis). Thus techniques commonly used for analyzing antennas and transmission lines can also be used in designing microwave circuits. Measuring techniques are, of course, also different. In low-frequency electronics, currents, and voltages are calculated. In microwave circuits, measurements are of electric and magnetic fields. Power measurements are more common than voltage and current measurements.

Another problem is that at microwave frequencies, conventional components become difficult to implement. For example, a common resistor that looks like pure resistance at low frequencies does not exhibit the same characteristics at microwave frequencies. The short leads of a resistor, although they may be less than an inch, represent a significant amount of inductive reactance at very high frequencies. A small capacitance also exists between the leads. These small stray and distributed reactances are sometimes called residuals. Because of these effects, at microwave frequencies, a simple resistor looks like a complex RLC circuit. This is also true of inductors and capacitors. Fig. 16-2 shows equivalent circuits of components at microwave frequencies.

To physically realize resonant circuits at microwave frequencies, the values of inductance and capacitance must be smaller and smaller. Physical limits become a problem. Even a 0.5-in piece of wire represents a significant amount of inductance at microwave frequencies. Tiny surface-mounted chip resistors, capacitors, and inductors have partially solved this problem. Furthermore, as integrated-circuit dimensions have continued to decrease, smaller and smaller on-chip inductors and capacitors have been made successfully.

Another solution is to use distributed circuit elements, such as transmission lines, rather than lumped components, at microwave frequencies. When transmission lines are cut to the appropriate length, they act as inductors, capacitors, and resonant circuits. Special versions of transmission lines known as strip lines, microstrips, waveguides, and cavity resonators are widely used to implement tuned circuits and reactances.

In addition, because of inherent capacitances and inductances, conventional semiconductor devices such as diodes and transistors simply will not function as amplifiers, oscillators, or switches at microwave frequencies.

Microwave Concepts | Communication Systems | Size and Frequency

Another serious problem is transistor transit time—the amount of time it takes for the current carriers (holes or electrons) to move through a device. At low frequencies, transit times can be neglected; but at microwave frequencies, they are a high percentage of the actual signal period.

This problem has been solved by designing smaller and smaller microwave diodes, transistors, and ICs and using special materials such as gallium arsenide (GaAs), indium phosphide (InP), and silicon-germanium (SiGe) in which transit time is significantly less than in silicon. In addition, specialized components have been designed for microwave applications. This is particularly true for power amplification, where special vacuum tubes known as klystrons, magnetrons, and traveling-wave tubes are the primary components used for power amplification. However, newer GaN semiconductor transistors are now useful well into the millimeter-wave range.

Another problem is that microwave signals, as do light waves, travel in perfectly straight lines. This means that the communication distance is usually limited to line-of-sight range. Antennas must be very high for long-distance transmission. Microwave signals penetrate the ionosphere, so multiple-hop communication is not possible. The physics of electromagnetic waves indicates that the shorter the wavelength and the higher the frequency, the shorter the transmission range for a given power or antenna gain.

Microwave Communication Systems

Like any other communication system, a microwave communication system uses transmitters, receivers, and antennas. The same modulation and multiplexing techniques used at lower frequencies are also used in the microwave range. But the RF part of the equipment is physically different because of the special circuits and components that are used to implement the components.


Like any other transmitter, a microwave transmitter starts with a carrier generator and a series of amplifiers. It also includes a modulator followed by more stages of power amplification. The final power amplifier applies the signal to the transmission line and antenna. The carrier generation and modulation stages of a microwave application are similar to those of lower-frequency transmitters. Only in the later power amplification stages are special components used.

Fig. 16-3 shows several ways that microwave transmitters are implemented. The special microwave stages and components are shaded. In the transmitter circuit shown in Fig. 16-3(a) a microwave frequency is first generated in the last multiplier stage. The operating frequency is 1680 MHz, where special microwave components and techniques must be used. Instead of tuned circuits made of loops of wire for inductors and discrete capacitors, microstrip transmission lines are used as tuned circuits and as impedance-matching circuits. SAW filters are the most commonly used filters in low-power circuits. One or more additional power amplifi ers are then used to boost the signal to the desired power level. Both bipolar and MOSFET microwave power transistors are available that give power levels up to several hundred watts.

When FM is used, the remaining power amplifiers can also be class C, which provides maximum efficiency. For phase modulation and QAM, linear amplifiers are needed. If more power is desired, several transistor power amplifiers can be paralleled, as in Fig. 16-3(a). If AM is used in a circuit like that in Fig. 16-3(a), an amplitude modulator can be used to modulate one of the lower-power amplifier stages after the multiplier chain. When this is done, the remaining power amplifier stages must be linear amplifiers to preserve the signal modulation.

For very high-power output levels—beyond several hundred watts—a special amplifier must be used, e.g., the klystron.

Fig. 16-3(b) shows another possible transmitter arrangement, in which a mixer is used to up-convert an initial carrier signal with or without modulation to the final microwave frequency.

The synthesizer output and a microwave local oscillator signal are applied to the mixer. The mixer then translates the signal up to the desired final microwave frequency.

Microwave Concepts | Communication Systems | Size and Frequency

A conventional crystal oscillator using fifth-overtone VHF crystals followed by a chain of frequency multipliers can be used to develop the local oscillator frequency. Alternatively, one of several special microwave oscillators could be used, e.g., a Gunn diode, a microwave semiconductor in a cavity resonator, or a dielectric resonator oscillator.

The output of the mixer is the desired final frequency at a relatively low power level, usually tens or hundreds of milliwatts at most. Linear power amplifiers are used to boost the signal to its final power level. At frequencies less than about 10 GHz, a microwave transistor can be used. At the higher frequencies, special microwave power tubes are used. The tuned bandpass circuits are shown in Fig. 16-3(b) can be microstrip transmission lines when transistor circuits are used, or cavity resonators when the special microwave tubes are used.

Modulation could occur at several places in the circuit in Fig. 16-3(b). An indirect FM phase modulator might be used at the output of the frequency synthesizer; for some applications, a PSK modulator would be appropriate.


Microwave receivers, like low-frequency receivers, are the superheterodyne type. Their front ends are made up of microwave components. Most receivers use double conversion. A first down-conversion gets the signal into the UHF or VHF range, where it can be more easily processed by standard methods. A second conversion reduces the frequency to an IF appropriate for the desired selectivity.

Fig. 16-4 is a general block diagram of a double-conversion microwave receiver. The antenna is connected to a tuned circuit, which could be a cavity resonator or a microstrip, or a stripline tuned circuit. The signal is then applied to a special RF amplifier known as a low-noise amplifier (LNA). Special low-noise transistors, usually gallium arsenide FET amplifiers, must be used to provide some initial amplification. Another tuned circuit connects the amplified input signal to the mixer. Most mixers are of the doubly balanced diode type, although some simple single-diode mixers are also used.

The local oscillator signal is applied to the mixer. The mixer output is usually within the UHF or VHF range. The 700- 800-MHz range is typical. A SAW filter selects out the difference signal, which in Fig. 16-4 is 12 GHz -11.2 GHz = 0.8 GHz, or 800 MHz.

The remainder of the receiver is typical of other superheterodyne. Note that the desired selectivity is obtained with a SAW filter, which is sometimes used to provide a specially shaped IF response.

Many of the newer microwave cell phone and LAN receivers are of the direct conversion type, and selectivity is obtained with RC and DSP filters.

In more recent microwave equipment such as cell phones and wireless networking interfaces, the microwave frequencies are generated by a phase-locked loop (PLL) operating as a multiplier.

See Fig. 16-5. The VCO produces the desired local oscillator (LO) or final transmitting frequency directly. The VCO frequency is controlled by the phase detector and its low-pass loop filter. The frequency divider and input crystal determine the output frequency. Recent advances in quartz crystal design permit input oscillator frequencies up to 200 MHz. In Fig. 16-5, the 155-MHz crystal combined with a frequency divider of 20 produces a VCO output at 155 x 20 = 3100 MHz, or 3.1 GHz. The ÷20 divider reduces the 3.1-GHz output to 155 MHz to match the input crystal signal at the phase detector, as required for closed-loop control. Of course, the divider can also be part of a microcontroller-based frequency synthesizer that is designed to permit setting the output to multiple channel frequencies as required by the application.

Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency

A more common architecture today is direct-conversion transmitters and receivers using I/Q modulators and demodulators, as shown in Fig. 16-6. Most modulation, such as QPSK or QAM, is performed by DSP in a baseband processor. The baseband processor produces the digital in-phase (I) and quadrature (Q) signals defining the modulation from the data input. In the transmitter, the I/Q signals are sent to DACs where they are converted to analog signals and filtered in a low-pass filter (LPF), and then sent to mixers. The mixers receive the local oscillator (LO) signals from a 90° phase shifter driven by the frequency synthesizer that sets the transmit frequency. The mixer outputs are added to create the final signal, which is amplified in a power amplifier (PA) before being sent to the antenna.

The I/Q modulator is in IC form and can usually handle signals in the 200-MHz to the 6-GHz range. If a higher final frequency is needed, the modulator output is sent to an up-converting mixer with its own local oscillator. The resulting higher frequency is then sent to the PA.

Microwave Concepts | Communication Systems | Size and Frequency

At the receiver, the signal from the antenna is fed to a low-noise amplifier (LNA) and a band-pass filter (BPF) to define the bandwidth. The signal is then sent to the mixers in the I/Q demodulator. If higher frequencies are involved, the LNA signal may go to a down-converting mixer first and then to the I/Q demodulator mixers. The signal is then mixed with the LO signal from the synthesizer at the signal frequency. The mixer outputs are then filtered and sent to ADCs, where the digital I/Q signals are generated. These I/Q signals are then processed in the baseband circuits to recover the original data.

Transmission Lines

The transmission line most commonly used in lower-frequency radio communication is a coaxial cable. However, coaxial cable has very high attenuation at microwave frequencies, and conventional cable is unsuitable for carrying microwave signals except for very short runs, usually several feet or less. Newer types of coaxial cables permit lengths of up to 100 ft at frequencies to 10 GHz.

Special microwave coaxial cable that can be used on the lower microwave bands—L, S, and C—is made of hard tubing rather than wire with an insulating cover and a flexible braid shield. The stiff inner conductor is separated from the outer tubing with spacers or washers, forming a low-loss coaxial cable known as hardline cable. The insulation between the inner conductor and the outer tubing can be air; in some cases, a gas such as nitrogen is pumped into the cable to minimize moisture buildup, which causes excessive power loss. This type of cable is used for long runs of transmission lines to an antenna on a tower. At higher microwave frequencies, C band and upward, a special hollow rectangular or circular pipe called waveguide is used for the transmission line (see Sec. 16-3).


At low microwave frequencies, standard antenna types, including the simple dipole and the one-quarter wavelength vertical antenna, are still used. At these frequencies, antenna sizes are very small; e.g., the length of a half-wave dipole at 2 GHz is only about 3 in. A one-quarter wavelength vertical antenna for the center of the C band is only about 0.6 in long. At the higher frequencies, special antennas are generally used (see Sec. 16-6).

Microwave Lines and Devices

Today, although vacuum tubes and microwave tubes such as the klystron and magnetron are still used, especially for higher-power applications, most microwave systems use transistor amplifiers. Over the years, semiconductor manufacturers have learned to make transistors work at these higher frequencies. Special geometries are used to make bipolar transistors that provide both voltage and power gain at frequencies up to 100 GHz. Microwave FET transistors have also been created such as the MESFET described. The use of gallium arsenide (GaAs) and silicon-germanium (SiGe) rather than pure silicon has further increased the frequency capabilities of MOSFETs and
bipolar. Both small-signal and power MOSFETs are available to operate at frequencies up to about 50 GHz. GaN power FETs are now commonly used at frequencies of 50 GHz. Because most microwave communication activity takes place in the lower-frequency ranges (L, S, and C bands), transistors are the primary active components used.

In the following sections, we discuss both discrete transistor types with microstrip tuned circuits and monolithic microwave integrated circuits (MMICs).

Microstrip Tuned Circuits

Before specific amplifier types are introduced, it is important to examine the method by which tuned circuits are implemented in microwave amplifiers. Lumped components such as coils and capacitors are still used in some cases at the high UHF and low microwave frequencies (below about 2 GHz) to create resonant circuits, filters, or impedance-matching circuits. However, at higher frequencies, standard techniques for realizing such components become increasingly harder to implement. Instead, transmission lines, specifical microstrip, are used.

Microwave Concepts | Communication Systems | Size and Frequency

These are readily implemented at the microwave frequencies because one-half or one-quarter wavelength transmission lines are only inches or some fraction thereof at those frequencies. Microstrip is preferred for reactive circuits at the higher frequencies because it is simpler and less expensive than stripline, but stripline is used where shielding is necessary to minimize noise and cross talk. The tuned circuits are created by using a copper pattern printed-circuit board (PCB)
upon which are mounted the transistors, ICs, and other components of the circuit.

Fig. 16-7 shows several views of the microstrip transmission line used for a reactive circuit. The PCB is usually made of G-10 or FR-4 fiberglass or a combination of fiberglass and Teflon. The bottom of the PCB is a thin solid copper sheet that serves as a ground plane and one side of the transmission line. The copper strip is the other conductor of the transmission line.

Fig. 16-7(a) is a perspective view of a microstrip line, Fig. 16-7(b) is an end view, and Fig. 16-7(c) and (d) sides views. Both open and shorted segments of the line can be used, although shorted segments are preferred because they do not radiate as much as open segments. One-quarter wavelength sections are preferred because they are shorter and take up less space on the PCB. Fig. 16-8 summarizes the open- and shorted-line possibilities for microstrip.

An important characteristic of a microstrip is its impedance. As discussed, the characteristic impedance of a transmission line depends on its physical characteristics, in this case, e.g., on the width of the strip and the spacing between the strip and the copper ground plane, which is the thickness of the PCB material. The dielectric constant of the insulating material is also a factor. Most characteristic impedances are less than 100 Ω; 50 Ω is the most common, followed closely by 75 Ω. Values higher than 100 Ω are used for cases when impedance- matching requirements demand it.

The one-quarter wavelength transmission line can be used to make one type of component look like another. For example, in Fig. 16-9(a), the λ/4 microstrip line can make a resistor at one end look like the resistance of another value; specifically,

R2 = Z02/R1

Microwave Concepts | Communication Systems | Size and Frequency

Here R1 is the resistance value of the resistor connected to one end of the line, and Z0 is the characteristic impedance of the microstrip, such as 50 Ω. If R1 is the characteristic impedance of the microstrip, such as 50 Ω, the line is matched and the generator sees 50 Ω. If R1 is 150 Ω, then the other end of the line will have a value of R2, or

502/150 = 2500/150 = 16.67 Ω

This application is the same as the one-quarter wavelength matching transformer or Q section discussed. Recall that two impedances can be matched by using a line of length λ/4 according to the relationship Z0 = R1R2, where Z0 is the characteristic impedance of the one-quarter wavelength line and R1 and R2 are the input and output impedances to be matched.

A quarter-wavelength line can also make a capacitor look like an inductor or an inductance look like a capacitance [see Fig. 16-9(b) and (c)]. For example, a 75-Ω microstrip λ/4 long will make an inductive reactance of 30 V look like a capacitive reactance of

Xc = Z02/Xl= 752/30= 5635/30 = 187.5 Ω

Fig. 16-10 shows the physical configurations for equivalent coils and capacitors in microstrip form. The thin segment of microstrip shown in Fig. 16-10(a) acts like a series inductor. Fig. 16-10(b) shows a short right-angle segment whose end is grounded; this microstrip acts as a parallel inductor. When series capacitance or capacitive coupling is needed, a small capacitor can be created by using the ends of microstrip lines as tiny capacitor plates separated by an air dielectric [Fig. 16-10(c)]. A shunt capacitor can be created by using a short, fat segment of microstrip as in Fig. 16-10(d). As these general forms demonstrate, it is often possible to visualize or even draw the equivalent circuit of a microstrip amplifier by observing the pattern on the PCB.

Microstrip can also be used to realize coupling from one circuit, as illustrated in Fig. 16-11. One microstrip line is simply placed parallel to another segment of the microstrip. The degree of coupling between the two depends on the distance of separation and the length of the parallel segment. The closer the spacing and the longer the parallel run, the greater the coupling. There is always signal loss by such a coupling method, but it can be accurately controlled.

Although microstrip performs best when it is a straight line, 90° turns are often necessary on a PCB. When turns must be used, a straight right-angle turn, like that shown in Fig. 16-12(a), is forbidden because it acts as a low-pass filter across the line. A gradually curved line, like that shown in Fig. 16-12(b) (or Fig. 16-11), is preferred when the turn radius is much greater than the width of the line. An acceptable alternative method is shown in Fig. 16-12(c). The cut on the corner is critical.

Note that the dimensions must be held to λ/4, which is one-quarter the width of the microstrip.

Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency

A special form of microstrip is the hybrid ring shown in Fig. 16-13. The total length of the microstrip ring is 1.5λ. There are four taps or ports on the line, spaced at one-quarter wavelength (λ/4) intervals, which can be used as inputs or outputs.

Now, a signal is applied to port 1, and some interesting things happen. Output signals appear at ports 2 and 4, but their levels are at one-half the power of the input. Thus the circuit acts as a power divider to supply two signals of equal level to other circuits. There is no output at port 3. The effect of applying a signal at port 4 is similar. Equal half-power outputs appear at ports 1 and 3, but no signal appears at port 2.

If individual signals are applied simultaneously to ports 1 and 3, the output at port 2 will be their sum and the output at port 4 will be their difference. The unique operation of the hybrid ring makes it very useful for splitting signals or combining them.

Microstrip can be used to create almost any tuned circuit necessary in an amplifier, including resonant circuits, filters, and impedance-matching networks. Fig. 16-14(a) shows how a low-pass filter is implemented with microstrip sections. The component shown is a highly selective low-pass filter for use in the 1- to 3-GHz range depending on the exact dimensions. The lumped constant equivalent circuit is shown in Fig. 16-14(b). The transmission line segments are formed on the PCB itself and connected end to end as required. The transistors or ICs are then soldered to the PCB along with any resistors or larger discrete components that may be needed.

Microwave Concepts | Communication Systems | Size and Frequency

Example 16.1
A one-quarter wave Q-matching section made of microstrip is designed to match a source of 50 ohms to a load of 136 ohms at 5.8 GHz. The PCB dielectric constant εr is 2.4. Calculate (a) the required impedance of the microstrip and (b) its length.

Microwave Transistors

Microwave transistors, whether they are bipolar or FET types, operate just as other transistors do. The primary differences between standard lower-frequency transistors and microwave types are internal geometry and packaging.

To reduce internal inductances and capacitances of transistor elements, special chip configurations known as geometries are used that permit the transistors to operate at higher power levels and at the same time minimize distributed and stray inductances and capacitances.

Fig. 16-15 shows several types of microwave transistors. Fig. 16-15(a) and (b) is low-power small-signal microwave transistors; these types are either NPN silicon or gallium arsenide FET. Note the very short leads. Both packages are designed for surface mounting directly to microstrip on the PCB. The transistor in Fig. 16-15(b) has four leads, usually, two emitters (or source) leads plus the base (or gate) and the collector (or drain). The two emitter leads in parallel ensure low inductance. Some transistors of this type have two base or two collector leads instead of two emitter leads.

Microwave Concepts | Communication Systems | Size and Frequency

Fig. 16-15(c) shows an enhancement-mode power MOSFET. The short, fat leads are thick strips of copper that are soldered directly to the microstrip circuitry on the PCB. These wide loads also help transfer heat away from the transistor. Fig. 16-15(d) is an NPN power transistor with two emitter leads. The short, fat leads ensure low inductance and also permit high currents to be accommodated. The fat copper strips help to dissipate heat. The devices in Fig. 16-15(c) and (d) can handle power levels up to several hundred watts. Transistors for small-signal amplification and oscillators are available for frequency up to about 100 GHz. For power amplification, transistors are available for frequencies up to 50 GHz.

Most microwave transistors continue to be made of silicon. As transistor geometry sizes have decreased below 0.04 μm (40 nm), switching speeds and amplification frequencies have increased well into the microwave region. CMOS digital integrated circuits, which are made with MOSFETs, can operate up to 10 GHz. RF and linear/analog circuits made with CMOS and used in low-power microwave radios as well as in optical fiber transmission circuits can achieve operation up to 10 GHz. But beyond 10 GHz, special devices are necessary.

You have already seen how GaAs MESFETs, a type of JFET using a Schottky barrier junction, can operate at frequencies in excess of 5 GHz. A variant of the MESFET called a high electron mobility transistor (HEMT) extends the frequency range beyond 20 GHz by adding an extra layer of a semiconductor material such as AlGaAs.

A highly popular device known as a heterojunction bipolar transistor (HBT) is making even higher-frequency amplification possible in both discrete form and integrated circuits. A heterojunction is formed with two different types of semiconductor materials. Some popular combinations are indium-phosphide (InP) and silicon-germanium (SiGe). Other combinations include AlGaAs/GaAs and InGaAsP. The InP HBTs operate at frequencies up to 50 GHz, and SiGe HBTs have been developed to operate up to 200 GHz. Both small-signal and power amplification versions are available.

For power amplifiers, the LDMOS enhancement-mode FETs are popular at frequencies below 6 GHz. Such transistors can handle power levels to several hundred watts. At frequencies to 50 GHz, the new gallium nitride (GaN) pHEMET transistors can supply power levels up to about 100 watts.

Small-Signal Amplifiers

A small-signal microwave amplifier can be made up of a single transistor or multiple transistors combined with a biasing circuit and any microstrip circuits or components as required. Most microwave amplifiers are of the tuned variety. That is, their bandwidth is set by the application and implemented by microstrip series or parallel tuned circuits, and then microstrip lines are used to perform the various impedance-matching duties required to get the amplifier to work.

Another type of small-signal microwave amplifier is a multistage integrated circuit, a variety of MMIC. Besides amplifi ers, other MMICs available include mixers, switches, and phase shifters.

Transistor Amplifiers

Fig. 16-16 shows a microwave amplifier for small signals. This type of amplifier is commonly used in the front end of a microwave receiver to provide initial amplification for the mixer. A low-noise transistor is used. The typical gain range is 10 to 25 dB.

Most microwave amplifiers are designed to have input and output impedances of 50 ohms. In the circuit shown in Fig. 16-16, the input can come from an antenna or another microwave circuit. The blocks labeled TL are microstrip sections that act as tuned circuits, inductors, or capacitors. An input microstrip TL1 acts as an impedance-matching section, and TL2 and TL3 form a tuned circuit. TL4 is another impedance-matching section that matches the tuned circuit to the complex impedance of the base input. The tuned circuits set the bandwidth of the amplifier input.

Microwave Concepts | Communication Systems | Size and Frequency

A similar sequence of impedance-matching sections and tuned circuits are used in the collector to set the bandwidth and to match the transistor collector impedance to the output. And C2 and C3 are variable capacitors that allow some tuning of the bandwidth. All the other components are of the surface-mounted chip type to keep lead inductances short. The microstrip segments TL9 and TL10 act as inductors, forming part of the substantial decoupling networks used on the base bias supply VBB and the collector supply VCC. The base supply voltage and the value of R1 set the base bias, thus biasing the transistor into the linear region for class A amplification. RFCs are used in the supply leads to keep the RF out of the supply and to prevent feedback paths through the supply that can cause oscillation and instability in multistage circuits. Ferrite beads (FB) are used in the collector supply lead for further decoupling.

MMIC Amplifiers

A common monolithic microwave integrated-circuit (MMIC) amplifier is one that incorporates two or more stages of FET or bipolar transistors made on a common chip to form a multistage amplifier. The chip also incorporates resistors for biasing and small bypass capacitors. Physically, these devices look like the transistors in Fig. 16-15(a) and (b). They are soldered to a PCB containing microstrip circuits for impedance matching and tuning.

Another popular form of MMIC is the hybrid circuit, which combines an amplifier IC connected to microstrip circuits and discrete components of various types. All the components are formed on a tiny alumina substrate that serves as both a base and a place to form microstrip lines. Surface-mounted chip resistors, capacitors, transistors, and MMIC amplifiers are connected. The entire unit is packaged into a housing, usually metal for shielding, and connected to additional circuits on a PCB.

Power Amplifiers

A typical class A microwave power amplifier is shown in Fig. 16-17. The microstrip lines are used for impedance matching and tuning. Most microstrip circuits simulate L-type matching networks with a low-pass configuration. Small wire loop inductors and capacitors are used to form the decoupling networks to prevent feedback through the power supply, which would cause oscillation. The input
and output impedances are 50 Ω. This particular stage operates at 1.2 GHz and provides an output power of 1.5 W. Typical power supply voltages are 12, 24, and 28 V, but can go to 36 or 48 V for high-power applications.

Microwave Concepts | Communication Systems | Size and Frequency

Note that bias is not supplied by a resistive voltage divider. Instead, it usually comes from a separate bias-current circuit like that shown in Fig. 16-18, which is an ordinary constant-current source. Resistors R1, R2, and R3 form a voltage divider to set the base voltage on Q1. A voltage is developed across emitter resistor R4. This voltage divided by R4 gives the value of the current supplied by Q1 to the transistor in the microwave amplifier. A diode in series with the voltage divider provides some temperature compensation for variations in the emitter-base voltage that occur in Q1. Resistor R1 is adjustable to set the bias current to the precise value for optimum power and minimum distortion.

Most power amplifiers obtain their bias from constant-current sources; this provides a bias current that is relatively temperature-independent and thus provides superior protection against damage.

The bias is applied to the base of the amplifier in Fig. 16-17 through a small inductor. This and the bypass capacitors keep the microwave energy out of the bias circuit.

Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency

The single-stage FET power amplifier shown in Fig. 16-19 can achieve a power output of 100 W in the high UHF and low microwave region. The transistor is an enhancement-mode FET; that means that the FET does not conduct with drain voltage applied. A positive gate signal of 3 V or more must be applied to achieve conduction. The gate signal is supplied by a lower-power driver stage, and the gate bias is supplied by a Zener diode. The gate voltage is made adjustable with potentiometer R2. This allows the bias to be set for linear class B or class C operations. For linear class A operation, a quiescent value of drain current is set with the bias. This bias circuit can also be used to adjust the power output over a small range.

In the circuit shown in Fig. 16-19, a combination of LC-tuned circuits and microstrip is used for tuning and impedance matching. Here L1 is a tiny hairpin loop of heavy wire, and L2 is a larger inductor made of multiple turns of wire to form a high impedance for decoupling.

Most modern GaN power amplifiers are complete circuits in a single package. Each targets a specific frequency range and application, such as cellular or radar.

Waveguides and Cavity Resonators

Long parallel transmission lines, such as 300-V twin-lead, radiate electromagnetic energy while transporting it from one place to another. As the frequency of operation gets higher, the amount of radiation from the line increases. At microwave frequencies, virtually all the energy is radiated; almost no energy ever reaches the end of the transmission line. Despite their high attenuation, coax cables are widely used to carry a microwave and even some millimeter-wave signals over short distances. For the lower microwave frequencies, special coaxial cables have been developed that can be used up to approximately 6 GHz if the length is kept short (less than 100 ft). Above this frequency, coaxial cable loss is too great except for lengths of several feet or less. Short lengths of coaxial cable, several feet or less, can be used to interconnect pieces of equipment that are close together, but for longer runs, other methods of transmission must be used. To minimize loss, special types of coaxial cables with large inner conductors and shields have been developed; however, in most cases, these cables are rigid rather than flexible, which makes them expensive and difficult to use.

Another problem is the power-handling limit of coaxial cables. The larger cables can carry power levels up to about 1 kW. Higher power levels needed in radar and satellite applications require another solution.


Most high-power microwave energy transmission above about 6 GHz is handled by waveguides, which are hollow metal conducting pipes designed to carry and constrain the electromagnetic waves of a microwave signal. Most waveguides are rectangular. Waveguides can be used to carry energy between pieces of equipment or over longer distances to carry transmitter power to an antenna or microwave signals from an antenna to a receiver.

Waveguides are made of copper, aluminum, or brass. These metals are extruded into long rectangular or circular pipes. Often the insides of waveguides are plated with silver to reduce resistance, keeping transmission losses to a very low level.

Signal Injection and Extraction

A microwave signal to be carried by a waveguide is introduced into one end of the waveguide with an antennalike probe that creates an electromagnetic wave that propagates through the waveguide. The electric and magnetic fields associated with the signal bounce off the inside walls back and forth as the signal progresses down the waveguide.

The waveguide totally contains the signal so that none escapes by radiation. The probe is shown in Fig. 16-20(a) is a one-quarter wavelength vertical antenna at the signal frequency that is inserted in the waveguide one-quarter wavelength from the end, which is closed. The signal is usually coupled to the probe through a short coaxial cable and a connector. The probe sets up a vertically polarized electromagnetic wave in the waveguide, which is then propagated down the line. Because the probe is located one-quarter wavelength from the closed end of the waveguide, the signal from the probe is reflected from the closed end of the line back toward the open end. Over a one-quarter wavelength distance, the reflected signal appears back at the probe in phase to aid the signal going in the opposite direction. Remember that a radio signal consists of both electric and magnetic fields at right angles to each other. The electric and magnetic fields established by the probe propagate down the waveguide at a right angle to the two fields of the radio signal. The position of the probe determines whether the signal is horizontally or vertically polarized. In Fig. 16-20(a), the electric field is vertical, so the polarization is vertical. The electric field begins propagating down the line and sets up charges on the line that cause the current to flow. This in turn generates a companion magnetic field at right angles to the electric field and the direction of propagation.

A loop can also be used to introduce a magnetic field into a waveguide. The loop in Fig. 16-20(b) is mounted on the closed end of the waveguide. Microwave energy applied through a short piece of the coaxial cable causes a magnetic field to be set up in the loop. The magnetic field also establishes an electric field, which is then propagated down the waveguide. Probes and loops can also be used to extract a signal from a waveguide. When the signal strikes a probe or a loop, a signal is induced that can then be fed to other circuitry through a short coaxial cable.

Microwave Concepts | Communication Systems | Size and Frequency
Microwave Concepts | Communication Systems | Size and Frequency

Waveguide Size and Frequency

Fig. 16-21 shows the most important dimensions of a rectangular waveguide: the width a and the height b. Note that these are the inside dimensions of the waveguide. The frequency of operation of a waveguide is determined by the size of a. This dimension is usually made equal to one-half wavelength, a bit below the lowest frequency of operation. This frequency is known as the waveguide cutoff frequency. At its cutoff frequency and below, a waveguide will not transmit energy. At frequencies above the cutoff frequency, a waveguide will propagate electromagnetic energy. A waveguide is essentially a high-pass fi lter with a cutoff frequency equal to

fco = 300/2a

where fco is in megahertz and a is in meters.

For example, suppose it is desired to determine the cutoff frequency of a waveguide with a dimension of 0.7 in. To convert 0.7 into meters, multiply by 2.54 to get centimeters and divide by 100 to get meters: 0.7 in= 0.01778 m. Thus,

fco = 300/2(0.01778) = 8436 MHz = 8.436 GHz

Normally, the height of a waveguide is made equal to approximately one-half the dimension, or 0.35 in. The actual size might be 0.4 in.

Example 16.2
Does a rectangular waveguide have a width of 0.65 in and a height of 0.38 in. (a) What is the cutoff frequency? (b) What is a typical operating frequency for this waveguide?

Example 16.3
Would the rectangular waveguide in Example 16-2 operate in the C band?

The C band is approximately 4 to 6 GHz. Because a waveguide acts as a high-pass filter with a cutoff of 9.085 GHz, it will not pass a C band signal.

Signal Propagation

When a probe or loop launches energy into a waveguide, the signal enters the waveguide at an angle so that the electromagnetic fields bounce off the sidewalls of the waveguides as the signal propagates along the line. In Fig. 16-22(a), a vertical probe is generating a vertically polarized wave with a vertical electric field and a magnetic field at a right angle to the electric field. The electric field is at a right angle to the direction of wave propagation, so it is called a transverse electric (TE) field. Fig. 16-22(b) shows how a loop would set up the signal. In this case, the magnetic field is transverse to the direction of propagation, so it is called a transverse magnetic (TM) field.

The angles of incidence and reflection depending on the operating frequency (see Fig. 16-23). At high frequencies, the angle is large and therefore the path between the opposite walls is relatively long, as shown in Fig. 16-23(a). As the operating frequency decreases, the angle also decreases and the path between the sides shortens. When the operating frequency reaches the cutoff frequency of the waveguide, the signal simply bounces back and forth between the sidewalls of the waveguide. No energy is propagated.

Whenever a microwave signal is launched into a waveguide by a probe or loop, electric and magnetic fields are created in various patterns depending upon the method of energy coupling, frequency of operation, and size of the waveguide. Fig. 16-24 shows typical fields in a waveguide. In the end view, the lines represent the electric field E lines.

The dots represent the magnetic field H lines. In the top view, the dashed lines represent the H field, and the 3s and dots represent the E field. The 3 means that the line is going into the page; the means that the line is coming out of the page.

The pattern of the electromagnetic fields within a waveguide takes many forms. Each form is called an operating mode. As indicated earlier, the magnetic or the electric field must be perpendicular to the direction of propagation of the wave. In the TE mode, the electric field exists across the guide and no E lines extend lengthwise along with the guide.

In the TM mode, the H lines form loops in planes perpendicular to the walls of the guide, and no part of an H line is lengthwise along with the guide.

Subscript numbers are used along with the TE and TM designations to further describe the E and H field patterns. A typical designation is TE0,1. The first number indicates the number of half-wavelength patterns of transverse lines that exist along the short dimension of the guide through the center of the cross-section. Transverse lines are those perpendicular to the walls of the guide. The second number indicates the number of transverse half-wavelength patterns that exist along the long dimension of the guide through the center of the cross-section. If there is no change in the field intensity of one dimension, a zero is used.

The waveguide in Fig. 16-24 is TE because the E lines are perpendicular (transverse) to the sides of the guide. Looking at the end view in Fig. 16-24 along the short length, we see no field intensity change, so the first subscript is 0. Along the long dimension of the end view, the E lines spread out at the top and bottom but are close together in the center. The actual field intensity is sinusoidal—zero at the ends, maximum in the center. This is one-half of a sine wave variation, and so the second subscript is 1. The mode of the line in Fig. 16-24 is therefore TE0,1. This, by the way, is the main or dominant mode of most rectangular waveguides. Many other patterns are possible, such as the two shown in Fig. 16-25.

Waveguide Hardware and Accessories

In some ways, waveguides have more in common with plumbing equipment than they do with the standard transmission lines used in radio communication. And, like plumbing, waveguides have a variety of special parts, such as couplers, turns, joints, rotary connections, and terminations. Most waveguides and their fittings are precision-made so that the dimensions match perfectly. Any mismatch in dimensions or misalignment of pieces that fit together will introduce significant losses and reflections. Waveguides are available in a variety of standard lengths that are interconnected to form a path between a microwave generator and its ultimate destination.

Connection Joints

Fig. 16-26 shows a choke joint, which is used to interconnect two sections of the waveguide. It consists of two flanges connected to the waveguide at the center. The right-hand flange is flat, and the one at the left is slotted one-quarter wavelength deep at a distance of one-quarter wavelength from the point at which the walls of the guide are joined. The one-quarter waves together become a half-wave and reflect a short circuit at the place where the walls are joined. Electrically, this creates a short circuit at the junction of the two waveguides. The two sections can actually be separated as much as one-tenth wavelength without excessive loss of energy at the joint.

This separation allows room to seal the interior of the waveguide with a rubber gasket for pressurization. Some long waveguides are pressurized with nitrogen gas to reduce moisture buildups. The choke joint effectively keeps the RF inside the waveguide. And it introduces minimum loss, 0.03 dB or less.

Curved Sections

Special curved waveguide sections are available for making 90° bends. Curved sections introduce reflections and power loss, but these are kept small by proper design. When the radius of the curved section is greater than 2λ at the signal frequency, losses are minimized. Fig. 16-27 shows several different configurations. Flanges are used to connect curve sections with straight runs.

T Sections

It is occasionally necessary to split or combine two or more sources of microwave power. This is done with T sections or T junctions (Fig. 16-28). The T can be formed on the short or long side of the waveguide. If the junction is formed on the short side, it is called a shunt T. If the junction is formed on the long side, it is called a series T. Each T section has three ports, which can be used as inputs or outputs.

If a signal is propagated along a waveguide connected to the C port of a shunt T like that in Fig. 16-28(a), equal amounts of the signal will appear in phase at output ports A and B. The power level of the signals at A and B is one-half the input power. Such a device is called a power divider.

The shunt T can also combine signals. If the signal input to A is of the same phase as the signal input to B, they will be combined at the output port C. The output power is the sum of the individual powers. This is called a power combiner. In the series T shown in Fig. 16-28(b), a signal entering port D will be split into two half-power signals that appear at ports A and B but that are 180° out of phase with each other.

Hybrid T’s

A special device called a hybrid T can be formed by combining the series and shunt T sections (see Fig. 16-29). Sometimes referred to as a magic T, this device is used as a duplexer to permit simultaneous use of a single antenna by both a transmitter and a receiver. The antenna is connected to port B. Any received signal is passed to port D, which is connected to the receiver’s front end. Port A is terminated and not used. The received signal will not enter port C, which is connected to the transmitter. A transmitted signal will pass through to the antenna at port B, but will not enter port D to the receiver. If a transmitter and receiver operate on the same frequency or
near the same frequencies, and the transmitter output signal is at a high power level, some means must be used to prevent the power from entering the receiver and causing damage. The hybrid T can be used for this purpose.


In many cases, it is necessary to terminate the end of an unused port of a waveguide. For example, port A in Fig. 16-29 would be connected to a load of the correct impedance to prevent high reflections, unfavorable SWR, and loss. If the length of the line is some multiple of one-quarter or one-half wavelength, it may be possible to open or short the waveguide. In most cases, this is not possible, and other means of termination are used.

One approach to termination is to insert a pyramid-shaped metallic section at the end of the line, as shown in Fig. 16-30(a). The taper provides a correct match. Usually, the tapered section is movable so that the termination can be adjusted for minimum SWR.

It is also possible simply to fill the end of the line with a powdered graphite resistive material, as shown in Fig. 16-30(b). This absorbs the signal and dissipates it as heat so that no reflections occur.

Termination can be accomplished by using a resistive material shaped like a triangle or wedge at the end of a closed line [see Fig. 16-30(c)]. The tapered resistive element is oriented to match the orientation of the electric field in the guide. The magnetic component of the wave induces a voltage into the tapered resistive material, and current flows. Thus the signal is absorbed and dissipated as heat.

Directional Couplers

One of the most commonly used waveguide components is the directional coupler. Directional couplers are used to facilitate the measurement of microwave power in a waveguide and the SWR. They can also be used to tap off a small portion of a high-power microwave signal to be sent to another circuit or piece of equipment. The directional coupler in Fig. 16-31 is simply a short segment of the waveguide with coupling joints that are designed to be inserted into a long run of waveguide between a transmitter and an antenna or between some source and a load. A similar section of waveguide is physically attached to this short segment of the line. It is terminated at one
end, and the other end is bent away at a 90° angle. The bent section is designed to attach to a microwave power meter or an SWR meter. The bent section is coupled to the straight section by two holes (X and Y) that are one-quarter wavelength apart at the frequency of operation. Some of the microwave energy passes through the holes in the straight section into the bent section. The amount of coupling between the two sections depends on the size of the holes.

If energy from the source at the left moves through the waveguide from left to right, microwave energy passes through hole X into the bent section. The energy splits in half. Part of the energy goes to the left, where it is absorbed by the terminator; the other half moves to the right toward hole Y. This is indicated by the dashed lines.

Additional energy from the straight section enters hole Y. It, too, splits into two equal components, one going to the left and the other going to the right. These components are indicated by the solid lines. The energy moving from hole Y to the left toward hole X in the curved section cancels the energy moving from hole X to Y. The signal moving from hole Y to hole X travels a total distance of one-half wavelength or 180°, so it is out of phase with the signal at hole X. But between Y and X, the signals are exactly out of phase and equal in amplitude, so they cancel. Any small residual signal moving to the left is absorbed by the terminator. The remaining signal moves on to the power or SWR meter.

The term directional coupler derives from the operation of the device. A portion of the energy of signals moving from left to right will be sampled and measured. Any signal entering from the right and moving to the left will simply be absorbed by the terminator; none will pass on to the meter.

The amount of signal energy coupled into the bend section depends on the size of the holes. Usually only a very small portion of the signal, less than 1 percent, is extracted or sampled. Thus the primary signal is not materially attenuated. However, a sufficient amount of signal is present to be measured. The exact amount of signal extracted is determined by the coupling factor C, which is determined by the familiar formula for power ratio:

C=10 logPin/Pout= coupling factor, dB

where Pin = amount of power applied to straight section.
Pout = signal power going to power meter.

Most directional couplers are available with a fixed coupling factor, usually 10, 20, or 30 dB.

The amount of power that is actually measured can be calculated by rearranging the power ratio formula:

For example, if the input power is 2000 W to a directional coupler with a coupling factor of 30, the actual power to the meter is 2000/103= 2000/1000 = 2 W.

A variation of the directional coupler is the bidirectional coupler. Using two sections of line to sample the energy in both directions allows the determination of the SWR. One segment of the line samples the incident or forward power, and the other is set up to sample the reflected energy in the opposite direction.

One of the most widely used forms of directional coupler today is that made with a microstrip transmission line, such as shown in Fig. 16-11. Directional couplers can be made right on the printed-circuit board holding the other circuitry. Directional couplers are also available as separate units with coaxial connectors for inputs and outputs.

Cavity Resonators

A cavity resonator is a waveguide-like device that acts as a high-Q parallel resonant circuit. A simple cavity resonator can be formed with a short piece of waveguide one-half wavelength long, as illustrated in Fig. 16-32(a). The ends are closed. Energy is coupled into the cavity with a coaxial probe at the center, as shown in the side view in Fig. 16-32(b). When microwave energy is injected into the cavity, the signal bounces off the shorted ends of the waveguide and reflects back toward the probe. Because the probe is located one-quarter wavelength from each shorted end, the reflected signal reinforces the signal at the probe. The result is that the signal bounces back and forth off the short ends. If the signal is removed, the wave continues to bounce back and forth until losses cause it to die out.

This effect is pronounced at a frequency where the length of the waveguide is exactly one-half wavelength. At that frequency, the cavity is said to resonate and acts as a parallel resonant circuit. A brief burst of energy applied to the probe will make the cavity oscillate; the oscillation continues until losses cause it to die out. Cavities such as this have extremely high Q, as high as 30,000. For this reason, they are commonly used to create resonant circuits and filters at microwave frequencies.

A cavity can also be formed by using a short section of circular waveguide such as that shown in Fig. 16-33. With this shape, the diameter should be one-half wavelength at the operating frequency. Other cavity shapes are also possible.

Typically, cavities are specially designed components. Often, they have hollowed out sections in a block of metal that have been machined to very precise dimensions for specific frequencies. The internal walls of the cavity are often plated with silver or some other low-loss material to ensure minimum loss and maximum Q.

Some cavities are also tunable. One wall of the cavity is made movable, as shown in Fig. 16-34(a). An adjustment screw moves the end wall in and out to adjust the resonant frequency. The smaller the cavity, the higher the operating frequency. Cavities can also be tuned with adjustable plugs in the side of the cavity, as in Fig. 16-34(b). As the plug is screwed in, more of it intrudes into the cavity and the operating frequency goes up.


A circulator is a three-port microwave device used for coupling energy in only one direction around a closed loop. A schematic diagram of a circulator is shown in Fig. 16-35(a). Microwave energy applied to port 1 is passed to port 2 with only minor attenuation; however, the signal will be greatly attenuated on its way to port 3. The loss from port 1 to port 3 is usually 20 dB or more. A signal applied to port 2 will be passed

with little attenuation to port 3, but little or none will reach port 1. The primary application of a circulator is as a diplexer, which allows a single antenna to be shared by a transmitter and receiver. As shown in Fig. 16-35(a), the signal from the transmitter is applied to port 1 and passed to the antenna, which is connected to port 2. The transmitted signal does not reach the receiver connected to port 3. The received signal from the antenna comes into port 2 and is sent to port 3, but is not fed into the transmitter on port 1.

A common way of making a circulator is to create a microstrip pattern that looks like a Y, as shown in Fig. 16-35(b). The three ports are 120° apart. These ports, which are formed on PCBs, can be connected to coaxial connectors.

On top of the Y junction, and sometimes on the bottom, is a ferrite disk. Ferrites are ceramics made of compounds such as BaFe2O3. They have magnetic properties similar to those of iron or steel, so they support magnetic fields, but they do not cause the induction of eddy currents. Iron is often mixed with zinc or manganese. A widely used type of ferrite is yttrium-iron-garnet (YIG). A permanent magnet is then placed such that it forces magnetic lines of force perpendicular to the device through the ferrite. This strong magnetic field interacts with the magnetic fields produced by any input signals, creating the characteristic action of the circulator.

The type of circulator described here is for low and medium power levels. For high-power applications, a circulator made with a waveguide-like structure containing ferrite material is used.


Isolators are variations of circulators, but they have one input and one output. That is, they are configured like the circulator diagrammed in Fig. 16-35, but only ports 1 and 2 are used. An input to port 1 is passed to port 2 with little attenuation. Any refl ected energy, as would occur with an SWR higher than 1, is not coupled back into port 1. Isolators are often used in situations where a mismatch, or the lack of a proper load, could cause reflection so large as to damage the source.

Watch Video to understand more

Storage Area Networks | SCSI | Internet Security | Security Measures ( Microwave Concepts | Communication Systems | Size and Frequency )

Internet Working | SONET | TCP/IP | Internet Transmission Systems ( Microwave Concepts | Communication Systems | Size and Frequency )

Radio Wave | Reflection | Refraction | Diffraction | Ground, Sky Waves ( Microwave Concepts | Communication Systems | Size and Frequency )

Mcqs of Electronics | online Test of Electronics | Thevenin’s | Nortons ( Microwave Concepts | Communication Systems | Size and Frequency )

Antenna | Antenna Operation | Antenna Types | Radio Waves | Dipoles ( Microwave Concepts | Communication Systems | Size and Frequency )

Click Here To Learn More ( Microwave Concepts | Communication Systems | Size and Frequency )

Click Here To Learn ( Microwave Concepts | Communication Systems | Size and Frequency )

Leave a Reply

%d bloggers like this: