More RF Power To the People

In August, this column took up some aspects of the measurement of DTV transmitter power output. You might wonder why I used the phrase "DTV transmitter power output" instead of "effective radiated power," since ERP appears on your license and that's what the FCC regulates.

ERP depends on the radio frequency power delivered to the transmitting antenna and the antenna gain. Antenna gain is always specified in azimuth and elevation, so ERP varies with the path over which a signal is transmitted.

The antenna manufacturer specifies gain of a product, so you are dependant on those numbers; likewise, the efficiency of your transmission line. The power output of your transmitter is under your control, so it must be measured by calculating your ERP.

DIGITAL VS. ANALOG ERP

The ERP specified by the FCC is the average power of the DTV signal* quite the opposite of how ERP for an analog signal is specified. With analog signals, the transmitter power output is maximum during synchronizing pulses, while its average power varies with the average scene brightness. It is then obvious why we measure peak power for NTSC.

The transient peak power of a DTV signal varies with time, nanosecond by nanosecond. If you have a LeCroy digital oscilloscope, you can capture your DTV signal and then find its transient peak power, but that is not what the FCC needs and incidentally may be about 6 dB higher than average power.

As I indicated in August, average power is the squared rms voltage across a dummy load resistor. Why rms volts? Because average power represents its heating power, and the fundamental technique to measure RF power is to measure the increase in the temperature of a known quantity of a known liquid by the heating effect of this power. This is known as the "calorimetric" method of measuring RF power. The dummy load dissipates the transmitter power output into a known quantity of a liquid whose specific heat is also known.

Being an out-of-service measuring procedure, this is usually employed by transmitter manufacturers. But there is a small catch to this. Your DTV transmitter must comply with the FCC DTV RF mask, which may limit your transmitter power output by the nonlinear distortions it produces. In other words, your transmitter cannot be operated to provide more average power output than its distortion products allow, nor can it be operated at a power output above which your calculated ERP would be exceeded.

The calorimeter method sounds tricky and perhaps messy at times. It is tricky because you are dealing with a large quantity of a moving liquid. It is messy if it leaks. It is also an out-of-service measurement technique. There are more convenient ways to monitor your DTV transmitter output power.

But first let's calculate the rms voltage corresponding to your average power output. Let us assume that your transmitter is rated to provide an average DTV power output of 10 kW. Its rms output voltage is 707 volts. That is an awful lot of voltage, so we sample a small known fraction of its power output. Let's assume we can obtain a sample of your RF power that is 1/1000th of the total RF power. Now the RF voltage is 22.36 volts.

Note that you must know this fraction of the RF sample, and you may have to depend on the measurement provided by the manufacturer of the RF sampling device.

There is also another problem. The transmitter power output goes up the transmission line towards your antenna, which will not have a voltage standing wave ratio equaling 1.00; it will therefore reflect some of that incident power, which has no place to go but back down the line towards the transmitter.

What you want to measure is the power from the transmitter into the line, so you need an RF sampling device known as a "directional coupler" that will sample only the incident wave from the transmitter to the transmission line. This may be your most important RF test equipment because its accuracy determines the accuracy to which you can calculate your ERP for the FCC.


(click thumbnail)Fig. 1: Directional coupler to sample the power output of a transmitter as shown or by reversing it, measure VSWR of the antenna and its transmission line.A directional coupler may be considered black magic because it is able to sort out the incident power coming from your transmitter from the reflected power being returned from the transmission line. It is actually a transformer with both inductive and capacitive coupling to the sample port as shown in Fig. 1. These two couplings are carefully arranged so that they are equal. The current flowing to the load induces a current in the sampling loop. This voltage and the voltage capacitively coupled to the sampling port are in phase.

Reflected power flows in the opposite direction so the inductive component induced in the sampling loop and the current capacitively coupled to the sampling port are out of phase. This is how the directive coupler sorts out the incident and reflected power.

If you want to measure the power being fed to the load, connect the directional coupler as shown in Fig. 1. If you want to measure the power being reflected back to the transmitter, turn the directional coupler around. This inconvenience can be avoided by using a dual directional coupler as shown in Fig. 2. It has two transformers and two coupled ports: One provides a sample of the incident power from the transmitter to the load, the other provides a sample of reflected power coming back to the transmitter.


(click thumbnail)Fig. 2: Dual directional coupler which permits simultaneous measurement of the transmitter power output and VSWR of the antenna and its transmission line.The coupling to each sample port is the same, and the manufacturer provides this data for your channel. In any real directional coupler, some of the reflected power sneaks into the incident power sampling port. The ratio of incident-to-reflected power at this port is known as the "isolation factor," while the difference between the coupling factor and the isolation factor is known as the "directivity" of the directional coupler.

These three parameters determine how accurate your measurements will be. As your antenna will normally provide a low VSWR (good impedance match to the transmission line), very little power will be reflected. For a transmission line efficiency of, for example, 80 percent, 8,000 watts of power will arrive at your antenna. If the antenna reflects 160 watts, then the reflected power arriving at the directional coupler will be 128 watts. For a coupling factor of 30 dB, the reflected power at the reflected sample port would be 128 milliwatts.

Now consider the incident power, which should not be coupled to this port. The incident power is 10,000 watts, and the coupling factor is 30 dB, so there is 10 watts at the incident sampling port where it should be.

If the directivity of the directional coupler is 40 dB, then 1/10,000th of the incident power will appear at the reflected sample port. This is 1 milliwatt of incident power where it should not be. However, this is quite small relative to the reflected sample, so the error is negligible. But if the directivity is only 30 dB, then there is 10 milliwatts of incident power at the reflected sample port.

If the incident power (10 milliwatts) and the reflected power are in phase at the reflected sample port, a power meter connected there would read 138 milliwatts, but if these samples are out of phase, the reading would be 118 milliwatts. This sample is usually indicated in terms of VSWR, so you see there is an uncertainty in measuring VSWR due to the phase of the signals arriving at the directional coupler. Given a directivity of 30 dB, this uncertainty is small, but if the directivity of your directional coupler is less than 30 dB, VSWR is less certain.

The effect of directivity upon the measured incident power is negligible unless there is a lot of reflected power. At the incident sample port, you have (in our example) 10 watts. With zero directivity, the reflected power of 128 watts will be attenuated by 30 dB (the coupling factor) to 128 milliwatts, which introduces a negligible error in measuring 10 watts of incident power.

Some broadcasters intend to use their NTSC transmitter on their NTSC channel for DTV in 2009, or whenever the sunset provision goes into effect. Will the directional coupler and antenna now being used for NTSC serve for DTV? Those directional couplers and antennas were designed for the NTSC signal, so the coupling factor and directivity and the antenna voltage standing wave ratio were specified at the visual carrier frequency and the color subcarrier frequency and at the aural carrier frequency.

As most of the power in NTSC signals is near the visual carrier frequency, the performance of a directional coupler or antenna intended for NTSC was optimized at the visual carrier frequency. For DTV, these particular specifications may not be sufficient. The DTV signal has its power distributed evenly across the channel, and the directional coupler and antenna for DTV signals should be specified differently.

I hope this gives you a better understanding of the operation of directional couplers and the vital role they play in the proper operation of your DTV transmitter. Stay tuned.