It doesn't get any better either. With modulation the dissipation does not decrease even though the efficiency increases at the peak power levels. The carrier power is still there 100% of the time at 30% efficiency. The audio is in the form of separate side bands that is additional power that the tube has to handle.
Actually, it should get better. A properly operating amplifier with low level modulation, linear or grid modualted, should run a steady carrier output level and a constant DC input level, regardless of modulation. For the sake of discussion, let us assume 100 watts carrier output. The DC input will be about 300 watts @ roughly 30@ efficiency. With no modulation present, we have 300 watts input with 100 watts output. That leaves 200 watts dissipated in the tube plates. Now let's modulate 100% with a sinewave tone. The average rf power output will now be 150 watts, 100 watts carrier power plus 50 watts average sideband power. The DC input is still 300 watts, so the tubes are now dissipating only 150 watts as opposed to 200 watts with no modulation. The overall efficiency of the amplifier rises from 30% to 50%. If the tube plates are showing a glow, they should actually dim a little with modulation.
Of course, the instantaneous efficiency is constantly varying over each cycle of the envelope waveform, from 0% at no output, to approximately 60% at maximum peak output capability of the amplifier. But it is AVERAGE power that determines how red the tube plates get, how loud the signal sounds over the air, and how much interference the signal produces. That is why the FCC's method of determining power output by p.e.p. is bogus.
Don K4KYV
