Andy's a hard act to follow.
The one detail he left out that helps explain why a "one size fits all" solution won't work.
Using a fixed resistor would be easier, but the variation from one MOSFET to the next is what makes it necessary to adjust the radio's bias voltage to the final.
The behaviour of a MOSFET when you put enough voltage on the gate to "turn on" the thing is very abrupt.
A tenth of a Volt up or down can throw the MOSFET from a safe level of idle current to either "off", where it won't draw idle current at all.
-OR-
It could cause the transistor to pull a dangerous level of current with no drive power, and overheat it.
A tenth of a Volt doesn't sound like much. The turn-on threshold voltage will vary from one brand of MOSFET to the next, by as much as two or three tenths of a Volt. And individual parts from a batch of 100 can vary by a tenth of a Volt or more. Every batch we buy gets hooked to the tester, and the last digits of the Vth reading marked on the rear of the part with a Sharpie marker. A matched pair or set of four will then be two (or four) parts with the same or nearly-same test marking.
A fixed resistor will never be right for every transistor. The older SSB radios built before the MOSFETs nearly all had a trimmer pot for setting idle-current bias.
AM-only radios depend on DC voltage rectified from the driver transistor's RF input to the final. When there's no drive, the MOSFET is turned off.
SSB is different. No carrier. The driver and final have to be switched on when the mike is keyed, and that means using the radio's switched transmit voltage to provide that DC gate-bias voltage and turn them on before you say "hello".
73
