Waveguide has many practical uses in high-powered RF systems, as covered in the last tutorial. Specifically, it is used as a transmission line for higher frequencies ranging from the UHF band up through microwave, where it is used exclusively. But, the unique properties of waveguide also make it an ideal component for the required high-power filters on today’s digital transmission systems. Back at the inception of analog broadcasting, coax filters were used for the filter/diplexer that was required for notching out a place for the aural (4.5MHz ) carrier to be inserted, and then for the suppression of the 3.58MHz harmonics created from the color subcarrier. Today, many of the bandpass filters for masking the 8-VSB signal and channel filters used in combiners are made up of waveguide.
Why high-power filters?
Most filtering and shaping of the 8-VSB signal is performed at low levels within the exciter itself. But as with any high-powered amplifier, out-of-band signals are generated and amplified along with the desired signal. These out-of-band signals can interfere with adjacent TV channels or other services, so they must be suppressed. The FCC is very strict on the subject of out-of-band signals and great care is taken to ensure they are not transmitted. In the past, the UHF channels were separated by spacing them six channels apart. So, if there was a Channel 20, the next channel would be 26, then 32, then 38 and so on. At the time, the tuners inside the TV sets were the reason for this wide spacing. Today, DTV channels are packed side by side within the TV spectrum, so it’s even more important to suppress out-of-band signals to avoid interference.
Waveguide devices
In making waveguide filters or combiners/splitters, several different parts come into play that perform important functions.
The basic parts are an H-plane tee, E-plane tee, magic tee, tuned cavities and the hybrid. All of these parts have qualities that when put together create high-powered filters and combiners. (See Figure 1.)
In an H-plane tee, if power in applied to the center port, the power output from the two remaining ports are in phase with each other. With an E-plane tee, when power is applied to the top port, the power coming from the remaining two ports are out of phase (180 degrees) in relation to each other. These two properties are the basis for the magic tee.
The way to remember the two different types of waveguide bends (used in tees and elbows) is to look at them and think about how the “H” bend is hard and the “E” bend is easy (just look at their shapes and imagine trying to bend them). These two bends, or elbows, come either swept (long curve) or mitered (sharp straight bends); the ones used in today’s transmission systems are mitered because it takes up less space.
The hybrid is another splitter/combiner made from waveguide for high power levels, whereas coax hybrids are used for lower power levels. Hybrids are four-port devices: When power is applied to one port, it is split equally between two of the remaining ports, making it a 3dB divider; the fourth port receives no power. Hybrids can be constructed with a 90-degree phase shift so the two outputs are 90 degrees apart when the power is split. This comes in handy when building filters.
Magic tee
A magic tee is a combiner/splitter made of waveguide. The magic tee has four ports and is constructed by combining an H-plane tee and an E-plane tee. When power is applied to Port 1, it is split evenly and exits Ports 2 and 3, with none going to Port 4 (See Figure 2a.) If power is applied to Ports 2 and 3, and they are in phase, they combine and exit Port 1, with no power going to Port 4. (See Fig 2b.) If the phase of the input power is shifted by 90 degrees, the power is now split between Ports 1 and 4. (See Figure 2c.) And, if the phase is shifted even more to 180 degrees, then all power is now directed to Port 4. (See Figure 2d.)
Because of these properties, the magic tee makes it possible to combine it with other components to create a switchless combiner, or a magic tee, as it is called in the broadcast industry. These devices allow for two or more transmitters to combine their power in a way that allows for one to be shut down and all the power from remaining transmitter to be transferred to the antenna without having to go off the air.
With a standard combiner, if one transmitter is shut down, the output of the combiner drops to one-fourth of the total power (half the power of the remaining transmitter) with the other half of the power diverted to the reject load. To go to half power, the RF has to be shut off, and then either switches or patches must be moved to directly connect the remaining transmitter to the antenna by bypassing the combiner. With a switchless combiner, the output of the active transmitter can be directed to the antenna while remaining on the air. (See Figure 3.)
A switchless combiner uses a hybrid input connected to a folded magic tee. This arrangement allow the two inputs to combine and exit the output port, or to have either input directed to the reject port where it’s connected to a load. To do this, the phases of the two input signals are shifted via polypropylene paddles. Being denser than air, the polypropylene slows the propagation of the RF, thus shifting its phase. In addition, there is one paddle in the path of each input, and these paddles can be moved in and out of the path of the RF to change its phase via a motor. (See Figures 4 and 5.)
High-power filters
When stations share a common antenna, the output from their transmitters is combined and travels up a single transmission line to the antenna. To do this, channel filters are used before the combiner to keep one station’s signal from feeding back to the output of the other station’s transmitter. It’s basically a bandpass filter that only allows the station’s own signal to pass through. Because it does not block its own frequencies, any reflected signal or VSWR coming back from the transmission system will be seen by the transmitter and register on its VSWR metering. (See Figure 6.)
A so-called “waffle iron” filter is used at the output of many UHF transmitters to attenuate the upper harmonics that might interfere with GPS signals. These filters are made of waveguide with what looks like a waffle iron pattern inside, which is actually the filter. These harmonics do not carry much power, so their energy can be dissipated through the filter itself.
Today, mask filters are required for all 8-VSB transmitters to reduce the sidebands and allow adjacent channels. Mask filters are actually bandpass filters that reject any signals outside of the channel boundaries. Today, there are sharply tuned mask filters that cut off out-of-band signals more quickly; these filters introduce more group delay at the edges of the passband, and the 8-VSB exciters have to be able to correct for this distortion.
A mask filter consists of two hybrids, one at each end, connected by two sets of tuned waveguide filters and two loads. The first hybrid splits the input to feed the two filters. Any energy not within the bandpass is rejected back to the hybrid. When the signal passes through the hybrid, one side shifts it 90 degrees. And as it passes back through, another 90-degree shift occurs, equaling 180 degrees. Because of this, when the energy rejected by both filters combine, they exit the reject port of the hybrid where it is connected to a load. At the other end, the signals that pass through the filters enter the output hybrid, are combined and then exit the output port of the hybrid to feed the antenna.
The reject load, on each hybrid’s isolation port, provides the required termination of the fourth port of the hybrid. If the load or the connection to it is damaged and changes its impedance, the hybrid will become unbalanced and the filter will not perform as expected. This can also lead to the transmitter experiencing very high VSWR because most of the transmitter power can be reflected back to it. When trouble-shooting VSWR problems, be sure to check the reject load on the transmitter’s mask filter.
Next time
The next tutorial will cover the tuning of RF components.
