CIRCUIT TECHNOLOGIES
Lumped and Distributed Circuits
RF components are frequently made up of several different electrical components called discrete components. For instance, while an RF amplifier is an RF component, it is made up of several discrete electrical components like diodes, transistors, and the big three: resistors, capacitors, and inductors. Resistors, capacitors, and inductors are small, inexpensive passive electrical components used to shape electrical signals and are utilized in some combination in every electrical circuit.
When electrical components are combined together in a defined area to perform some prescribed function, the components are said to form a circuit. In the world of RF, there are two philosophies behind circuit design: lumped element circuits and distributed circuits.
Lumped element and distributed circuits both use the same semiconductor devices (transistors, diodes, and MMICs) in their designs. Where they differ is in the nature of the passive components they use, specifically the big three, as well as others like transformers and couplers.
Lumped Circuits
In lumped element circuit designs, the capacitors and inductors are real things which can be seen and touched. A sampling of "real" capacitors and inductors is shown in Figure 5-2. In lumped element designs, couplers are really just transformers. (I told you they were used for other things.)
Figure 5-2. Lumped element ("real") capacitors and inductors. Courtesy of Sprague-Goodmn Electronics, Inc.
Distributed Circuits
Distributed circuits are where things get interesting. As you recall from the introduction, signals in the RF world get around in a circuit by cruising around on a conductor which is often just a small, thin piece of metal (called a trace) on a printed circuit board (PCB) or other substrate. An interesting thing happens to these metal traces at RF frequencies: they begin to act as discrete components. In distributed circuits, RF engineers can shape the metal traces in very specific ways to make them behave like capacitors, inductors, transformers, and even couplers, which is why in pure distributed circuits the only "real" devices are semiconductors, the rest are just a bunch of odd-shaped metal traces. Figure 5-3 shows an example of a distributed circuit. The circuit traces in the small square area on the left side of the figure actually form a coupler, which looks quite a bit different from a lumped element (transformer) coupler (refer to Figure 4-21 in the previous chapter).
Figure 5-3. A distributed circuit. Courtesy of JCA Technology.
How is the choice made between circuit philosophies when designing a circuit? While both have their advantages and disadvantages, the decision is usually very simple. Remember in the RF world, the higher the frequency, the smaller things get, and vice versa. If a circuit is to operate at a "low" RF frequency, then the components will be relatively bigger. And if the components need to be bigger, then the size and shape of the circuit traces needed to realize the components in a distributed design become prohibitively large. All this is a long-winded way of saying, below a certain frequency, the only circuit choice is a lumped element design. Conversely, above a certain frequency, the only reasonable choice is a distributed circuit design, and in between, you flip a coin.
Discrete, Hybrid, and MMIC Circuit Choices
Once a circuit philosophy is selected (or dictated), the next choice is what circuit technology to use. Once again there is a choice to be made, this time from three different circuit technologies: discrete, hybrid (also called MIC, for microwave integrated circuit), or MMIC (microwave monolithic integrated circuit). Table 5.2 details the three choices and their respective advantages and disadvantages. An example of each type is shown in Figures 5-4a, 5-4b, and 5-4c. Figure 5-4b (the MIC) contains both packaged and bare "chip" semiconductors.
| Technology | Description | Advantages | Disadvantages |
|---|---|---|---|
| Discrete | Combines semiconductor devices (diodes, transistors and MMICs) and lumped passive devices as individually packaged discrete components onto a printed circuit board (PCB). | Utilizes existing discrete components, fast design time, superior performance at high power. | Takes up a lot of space, reduced performance at high frequency, expensive in large quantity. |
| Hybrid (MIC) | Combines both packaged and "chip" semiconductor devices (diodes, transistors, and MMICs), and passive devices (both lumped and distributed), along with metal traces onto a ceramic substrate. | Smaller and better high frequency performance than discrete, cheaper than discrete in large quantity, superior high frequency performance. | Expensive in small quantity, longer design time than discrete, more delicate handling and troubleshooting than discrete. |
| MMIC | Combines semiconductor devices (diodes and transistors) and distributed passive devices onto a single piece of semiconductor. | Smaller than any other approach, less expensive than any other approach in high volume. | Very expensive in small quantity, very long design time, some degradation in performance compared to hybrid approach. |
Figure 5-4a. A discrete circuit. Courtesy of Alpha-Industries.
Figure 5-4b. A hybrid circuit. Courtesy of Micro Networks.
Figure 5-4c. A MMIC. Courtesy of TriQuint Semiconductor, Inc.
Did You Know?
The predominant way that discrete components get mounted onto RF printed circuit boards (PCB) today is by what is known as surface mount technology (SMT). In SMT, the individual components get soldered right to the surface of the PCB. Now this may seem obvious, but in the early days of electronic manufacturing, another technique called feed-through technology, was used in which the individual components were actually fed through little holes drilled in the PCB. So if nothing else, SMT saves on drill bits.
Did You Know?
The newest MIC technology used today is called Low Temperature Co-fired Ceramic, or LTCC for short. LTCC is just like a PCB (with its multilayer traces), only it is made from ceramic instead of a plastic composite. LTCC allows the MIC to be made very small, which is ideal for use in very high frequency components.
Subassemblies
There are certain RF manufacturers who make a living by combining more than one RF component, to perform more than one RF function, into a single package. When a box of RF "stuff" performs more than one basic function, like a single mixer or a single amplifier, the box of stuff is referred to as a subassembly.
Did You Know?
Sometimes a subassembly is referred to as a subsystem. Now there is no technical difference between the two descriptions, but companies selling "subsystems" have been known to fetch a higher price than those selling "subassemblies." That's marketing for you.
As an example of a subassembly, refer back to the block diagram of a receiver in Figure 3-1. A receiver has a mixer which is fed by an oscillator and followed by a filter. If some enterprising RF manufacturer thought it made good business sense, they might combine all three components (mixer, oscillator, and filter) into a single box and call it a mixer-oscillator-filter subassembly (or a mixoster?). The good news regarding subassemblies, at least as far as the manufacturer is concerned, is that because they are complicated to make, they fetch a high price. On the flip side, however, subassemblies take a long time to develop, require a lot of engineering and, because they are so specific, tend to have only a single customer. Because of this, manufacturing subassemblies is considered risky business. Of course, that doesn't keep a lot of manufacturers from trying.
Cavities
There is one final way an RF component can be manufactured. Unlike the other technologies (discrete, hybrid, and MMIC), cavity type components do not use conductors to carry the RF signal. Instead, RF signals move as waves inside cavity components.
A cavity circuit is some sort of hollow container made out of metal with the RF signal bouncing around on the inside. Cavity technology is a fairly old RF technology, and many different RF components can be made as cavity components, like couplers, oscillators, and even amplifiers. When an amplifier is of the cavity type, it is called a traveling wave tube amplifier or TWTA (or TWT for short). (It makes sense: an amplifier with a wave traveling around inside a tube-shaped cavity should be called a traveling wave tube amplifier!) A TWT is shown in Figure 5-5.
Cavity components are used for one reason and one reason only: high power. When RF engineers need to amplify a signal really big—bigger than any transistor can amplify it—they use a cavity amplifier (TWT). When they need to couple a high power signal or filter a high power signal, they use cavity components. The output filters of cellular basestations are (guess what?) cavity filters, because the output power of basestations is relatively high. (Cavity filters also happen to have the superior filtering characteristics required by cellular transmitters, but that's another story.)
Figure 5-5. A traveling wave tube. Courtesy of Litton Electron Devices.
