in band‐pass filters, and the inductance must be evaluated for each use to find the effective value considering the parasitic capacitance.
A common figure of merit for inductors is the self‐resonant frequency (SRF), above which they act more like a capacitor (impedance goes lower with increasing frequency) than an inductor. The value of the SRF can be estimated in one way by looking at the length of the wire used in making the inductor. The SRF will be less than the frequency for which the wire is one‐quarter wavelength.
1.13.4 PC Board Vias
The PC board via is perhaps the most common PC board component, and often the most overlooked. The effect of a via depends greatly upon how it is structured in the circuit. A single via to ground in the center of a transmission line appears as almost a pure inductance. However, a via between RF traces can have aspects of inductance and some parasitic capacitance (due to pads around the via) that can cancel, in part or all, the inductive effect. When a via is used in a mounting pad for a shunt element, such as a resistor used as a load, or a bypass capacitor, the mounting pad and via form a resonant structure such that the size of the mounting pad can increase the effective impedance of the via. Further, several vias are often used in parallel to ground devices, sometimes to lower their effective inductance and sometimes to provide greater heat sinking of an active device. Putting vias in parallel does lower their effective inductance, but not in a simple way. Rather than halving the inductance, mutual inductance between vias means that the value of effective inductance doesn't reduce as expected. For example, putting two 100 Ω resistors at the end of a line to ground, placed in parallel, may show much larger inductive effect the same two 100 Ω resistors place in a T pattern, where the ground vias are separated and the mutual inductance is less.
1.14 Active Microwave Components
With a few exceptions, passive components follow some fundamental rules that greatly simplify their characterization; principally, they are linear, so their characterization doesn't depend on the power of the signal used to characterize them, but only on the frequency. Active components, on the other hand, are sensitive to power, and their responses to both frequency stimulus and power stimulus are important. Often, passive components are operated well below any power level that causes a change in their response, but more and more active components are being driven into higher‐power operation to optimize their efficiency.
1.14.1 Linear and Non‐linear
In the measurement sense, one definition of linear devices is that they are devices in which the output power is a linear function of the input power. If the input power is doubled, the output power is doubled. Almost all passive devices follow this rule, and many active devices as well. An alternative definition of linear is one for which only frequencies that are available at the input appear at the output. In practice, the first definition is more useful for system response. Some important characteristics of active components are discussed next.
1.14.2 Amplifiers: System, Low‐Noise, High Power
1.14.2.1 System Amplifiers
System amplifiers are simply gain blocks used to boost signal levels in a system, while providing reverse isolation. They can have higher noise figures than LNAs as they are used in signal paths where the signal is well above the noise floor. They often follow an LNA stage, frequently after some pre‐filtering. They are also often used in the frequency converters as a local oscillator (LO) amplifier to isolate the RF signal from leaking out the LO port, or as an isolation amplifier to prevent LO leakage out the RF port. These tend to be broad band amplifiers, with good input and output match, emulating an idealized gain block. The important figures of merit for such amplifiers are gain (S21), input and output match (S11, S22), and isolation (S12). Occasionally, directivity of an amplifier is defined as isolation (a positive number in dB) minus the gain (in dB), or S12/S21. It is a measure of the effects of a load apparent at the input of the amplifier, or how the output impedance is affected by the source impedance (Mini‐circuits n.d.) and is important in cases where other system components have a poor or unstable match. Since these amplifiers have wide bandwidths, it is important that they have good stability as they can have a variety of load impedances applied. Other figures of merit for system amplifiers can include gain flatness (deviation of the gain from nominal value), 1 dB compression point (the power at which the gain drops by 1 dB), harmonic distortion, and two‐tone third‐order IM, sometimes expressed as third‐order intercept point (see Section 1.3).
1.14.2.2 Low‐Noise Amplifiers
Low‐noise amplifiers are found in the front end of communications systems and are particularly designed to provide signal gain without a lot of added noise power. The key figure of merit is noise figure, along with gain. But, in a system sense, the noise parameters (see Section 1.3) are quite important as they describe the way in which the noise figure changes with source impedance. LNAs are used in low‐power applications, and their 1 dB compression point is not a key spec; however, distortion can still be a limiting factor in their use, so a common specification is the input referred intercept point. A key trade‐off made in LNAs is between lowest noise figure and good input match. The source impedance where the LNA provides the lowest noise may not be the same as the system impedance, so a key design task for LNAs is to optimize this trade‐off.
1.14.2.3 Power Amplifiers
Many of the figures of merit for power amplifiers are the same as system and LNAs, but with an emphasis on power handling. In addition, the efficiency of the amplifier is one of the key specifications that one finds primarily with power amplifiers, implying that the DC drive voltage and current must also be characterized. Because power amplifiers are often used with pulsed RF stimulus, the pulse characteristics, such as pulse profile including pulse amplitude and phase droop, are key parameters.
Power amplifiers are often driven into a non‐linear region, so the common linear S‐parameters may not apply well to predict matching. Therefore, load‐pull characterization is often performed on power amplifiers. Gain compression and output‐referred intercept point are common for power amplifiers. Some amplifier designs such as traveling‐wave‐tube amplifiers (TWT) have a characteristic that causes the output power to reach a maximum and then decrease with increasing drive power, and the point of maximum power is called saturation. Gain at rated output power is another form of a compression measurement where rather than specifying a power for which the gain is reduced by 1 dB, it specifies a fixed output power at which the gain is measured.
Power amplifiers are often specified for their distortion characteristics including IMD and harmonic content. In the case of modulated drive signals, other related figures of merit are ACPR and adjacent channel power level (ACPL). A figure of merit that combines many others is EVM, which is influenced by a combination of compression, flatness, and inter‐modulation distortion among other effects.
1.14.3 Mixers and Frequency Converters
Another major class of components is mixers and frequency converters. Mixers convert an RF signal to an IF frequency (aka down‐converters) or IF frequencies to RF (up‐converters) through the use of a third signal, known as the local oscillator (LO). Typically, the output or input that is lower in frequency is called the IF (for intermediate frequency). The LO provided to the mixer drives some non‐linear aspect of the circuit, typically diodes or transistors that are switched on and off at the LO rate. Using the second definition, frequency converters and mixers would not be considered linear. In fact, in their normal operation, they are linear (under the first definition) with respect the desired information signal, and ideally the frequency conversion does not change the linearity of the input/output transfer function.
The