Scrutinize the Specs to Identify the Right Switch

Scrutinize the Specs to Identify the Right Switch

Selecting an RF/microwave switch requires understanding the needs of an application and sorting through different switch types with considerably different performance levels.

Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicable.

Switches for RF/microwave applications have evolved into many forms, from slower, higher-power electromagnetic switches to the fastest diode switches for smaller signals. Microelectromechanical-systems (MEMS) technology, which has made its own distinct splash in this crowded field, further enriches the assortment of high-frequency switches and switch technologies for specifiers. Selecting the right switch for the job becomes a matter of understanding what these different switch technologies can do and how well each switch type suits different applications.

Fig. 1
1. The single-pole, four-throw (SP4T) switch shows the size limits incurred by packaging with multiple coaxial connectors. (Photo courtesy of API)

Perhaps the best way to compare and ultimately pinpoint the right microwave/RF switch is to study the different operating parameters, including frequency range, bandwidth, power-handling capability, switching speed, insertion and return loss, isolation, VSWR, power consumption, linearity, and reliability. High-frequency switches are based on a number of different technologies, such as as electromechanical, solid-state, and MEMS switches, and each offers advantages and drawbacks. Most applications can help specifiers define the performance limits required for a switch, and knowing the general capabilities of each switch type can simplify the selection process.

For example, insertion loss stems from a switch circuit’s parasitic circuit elements—resistance, capacitance, and inductance—and how they essentially form a lowpass filter with attenuation as a function of frequency. Switch insertion loss is unavoidable and some amount of attenuation is acceptable. Ideally, though, insertion loss should be as low as possible to minimize signal attenuation through a switch.

Voltage standing wave ratio (VSWR) is the ratio of reflected to transmitted electromagnetic (EM) waves. Reflections can occur when EM signals change from one media to another, e.g., from a coaxial connector to a switch’s printed-circuit board (PCB). Reflections will occur at any change in impedance.

When comparing switches and their parameters, it’s important that parameters are normalized, such as comparing loss at the same frequencies. Some parameters, such as switching speed, can be defined in various ways. For example, switching speed is typically interpreted as the duration from a certain level of a digital control signal to a certain level of an RF/microwave signal, such as the time from a 50% transistor-transistor-logic (TTL) signal to 90% RF signal

Switch Types

Electromechanical switches, commonly used in test systems, typically deliver outstanding electrical performance over wide frequency bandwidths, with high power-handling capabilities, low loss, and high isolation. They are physically larger and more expensive than other switch types. Moreover, they rely on the physical movement of a switch element to achieve conducting (“on”) or open-circuit (“off”) switch positions, which limits their operating lifetimes compared to solid-state and MEMS types of switches. Due to the physical movement within the switch, switching speed is also limited versus other switch types.

Solid-state switches, commonly based on PIN diodes or GaAs MESFET devices, employ electronic switching. These high-speed components often switch with the aid of logic control, such as TLL driver circuitry. Usually they’re limited to lower signal power levels than electromechanical switches, with narrower bandwidths. Switching speeds, though, are typically in the microsecond range, versus the typical millisecond speeds for electromechanical switches.

Fig. 2
2. This GaAs FET IC switch operates from 20 MHz to 6 GHz in a micro-lead-frame-dual (MLFD) surface-mount package measuring 1.0 ×1.0 × 0.5 mm. (Photo courtesy of Skyworks Solutions)

MEMS switches, relative newcomers to RF/microwave applications, exhibit some of the traits of the other two switch types. MEMS switches are essentially electromechanical switches formed with semiconductor-sized structures. They feature low insertion and return loss and high isolation, but are limited in speed compared to solid-state switches. Power-handling capability also is limited when compared to electromechanical switches.

All three switch types can be compared by means of a standard set of operating parameters, as noted earlier. However, one or two parameters often stand out as the most important for a particular application, such as insertion loss and isolation across a frequency range. In some cases, switching speed may be the key, and achieving extremely fast switching speed, such as with a PIN diode solid-state switch, will mean sacrificing considerable power-handling capability. A PIN diode switch can achieve microsecond switching speed (unlike MEMS and electromechanical switches that range in the milliseconds). Power-wise, though, it can only handle one or two watts (similar to a MEMS switch), compared to 100 W or more for some electromechanical switches.

Some tradeoffs are fairly clear for RF/microwave switches (e.g., RF power for switching speed), while others are more subtle (e.g., loss and isolation with bandwidth). Also, while electromechanical switches generally offer the best combination of loss and isolation with wide bandwidth, developers of solid-state monolithic-microwave-integrated-circuit (MMIC) diode and field-effect-transistor (FET) switches constantly strive to improve their products. Now solid-state switches feature bandwidths of 40 GHz and more, albeit with limited power-handling capabilities compared to electromechanical switches.

RF/microwave switches are further differentiated by several operating modes and traits. Switches can operate as failsafe or latching components, and be absorptive or reflective in nature. A failsafe switch is able to function without power applied; in this scenario, its common port will be connected to one of its output ports. On the other hand, a latching switch will not operate without power applied—its common port will not be connected to any of its output ports.

In addition, a switch may be reflective or absorptive. When turned off, a reflective switch offers open ports (in a high impedance state), while an absorptive switch has ports at the characteristic impedance of the system (typically 50 Ω). A make-before-break switch makes contact with a new switch channel before cutting power to the original switch path. A break-before-make switch breaks contact with one signal path before it forms another switch path.

Fig. 3
3. Gold-plated leads minimize loss in this compact pin package for a RF/microwave switch. (Photo courtesy of API Technologies )

Switches can also be sorted out in terms of whether they can perform cold or hot switching functions. Cold switching refers to no signal being applied to a switch when changing states, while hot switching is the opposite—a signal can be applied to a switch when changing states.

Switch physical configurations vary according to the number of poles and throws, as well as the package style. Many switch suppliers offer products ranging from simple single-pole, single-throw (SPST) switches to complex single-pole, 16-throw (SP16T) varieties. Such components can be used to control one or more receiver channels, depending on the number of throws, but will be limited in size according to the type of package.

Waveguide switches, for example, require housings large enough to mount the input and output waveguide flanges. Similarly, coaxial switches require package size sufficient to mount the required number of coaxial connectors (Fig. 1). Coaxial packaging will become smaller at higher frequencies due to the shrinking size of the connectors needed to handle higher frequencies. At lower frequencies, for example, larger Type N connectors are used for coaxial switch packaging, which limits the package’s ability to scale down to smaller sizes. At higher frequencies, connector types such as SMA, 2.9-mm, and 2.4-mm connectors can be mounted on more compact housings.

Of course, when miniaturization is required, solid-state and MEMS switches hold tremendous advantages over electromechanical switches, since they can do away with coaxial connectors and fit into surface-mount packages (Fig. 2). In this case, the limitation of one or two watts of power-handling capability for solid-state and MEMS switches is offset by the availability of surface-mount and compact pin packages (Fig. 3) that allow switches to be mounted on PCBs for compact circuits.

A number of suppliers offer good switch products across many different frequency ranges, performance levels, and sizes. Often times, an application’s requirements will define the type of switch for that application, such as power-handling capability or switching speed. Breaking down a switch search, by starting with the type of switch, can at least help to accelerate the specifying process and eliminate time spent considering inappropriate switch types for a particular application.

Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicable.
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.