In practice, what the guys did back in the "Golden Age" was to get a test amp running with an approximate output transformer, then vary a load resistor on the secondary. This directly varied the plate to plate loading without having to rewind transformers or have to worry about zillions of taps.
They measured the output power (= voltage *current) in the variable load resistor, and dinked with the setting of the load resistor until they got a maximum. They also noted that they got one maximum and one minimum: a maximum power loading and a minimum distortion loading. These were different plate-to-plate load points. Once they got the best power and distortion points down, they could wind a fixed transformer that did the right thing for what they were designing.
They measured the output power (= voltage *current) in the variable load resistor, and dinked with the setting of the load resistor until they got a maximum. They also noted that they got one maximum and one minimum: a maximum power loading and a minimum distortion loading. These were different plate-to-plate load points. Once they got the best power and distortion points down, they could wind a fixed transformer that did the right thing for what they were designing.
Comment