With the recent surge in availability of RF switching products for test system development, choosing the right product for your application has become increasingly difficult. Most RF vendors describe their RF switching products using two main specifications – topology and bandwidth (such as the NI PXI-2594 2.5 GHz 4x1 multiplexer). While these specifications are indeed important and vital during the evaluation phase, they do not provide the buyer with enough information to make an informed purchasing decision. The goal of this tutorial is to introduce to you the following seven important specifications that must be considered when designing your RF switch network:
- Characteristic impedance
- Insertion Loss
- Return loss and voltage standing-wave ratio (VSWR)
- Isolation and crosstalk
- Rise time
Before discussing characteristic impedance and other RF switch specifications, it is first important to understand the difference between how signals propagate in DC circuits versus RF systems. In DC circuits or circuits where the propagating signal has low frequencies, the voltage of the signal at different points on a cable in the signal path varies minimally. This is not so in the case of RF or high-frequency signals where wavelength of the signal is considerably small in comparison to the length of the cable allowing multiple cycles of the signal to propagate through the cable at the same time.
Consider an example where two waves (signals) of different frequencies are made to propagate through a 1 m coaxial cable. The frequency of the first signal is 1 MHz while that of the second signal is 1 GHz. To calculate their wavelengths, we will use the following formula:
In the above formula, l is the wavelength of the signal, f is its frequency, and VF is the velocity factor of the cable. Let us assume that the coaxial cable being used to route both signals is of type RG8, which is known to have a velocity factor of 0.66.
Signal 1 where f = 1 MHz:
Figure 1. A 1 MHz Sine Wave Propagating through a 1 m Coaxial Cable
In the case of Signal 1, the length of the coaxial cable is considerably small in comparison to the wavelength of the signal propagating through it. Therefore, as can be seen in Figure 1, the variation in potential of the signal at different points in the cable is negligible.
For Signal 2 where f = 1 GHz:
Figure 2. A 1 GHz Sine Wave Propagating through a 1 m Coaxial Cable
In the case of Signal 2, the length of the coaxial cable is much larger (almost 5X) than the wavelength of the signal propagating through it. Therefore, at any given time, multiple cycles of the signal will travel through the cable simultaneously. Because of their small wavelengths, high-frequency signals travel through cables in waves. Such signals therefore suffer reflections and power loss when traveling between varying media (wave theory). In the case of electrical circuits, this variation in medium takes place when the signal (wave) is made to pass through system components that have varying characteristic impedances. Therefore, to minimize reflections and power loss, RF systems must be constructed using suitable components with matched impedances. As a rule of thumb, signal degradation due to power loss and reflections that occur in the transmission line become significant once the length of the cable exceeds 0.01 of the wavelength of the signal it is being used to route.