Hero999 said:
I'm aware of that, but you can by an RF amplifier with an output impedance of 50ohms and it's 60% efficient. How is this possible when the when if for maximum power transfer the load must be 50ohms and it'd be only 50% efficeint?
I presume that the output really isn't 50ohms, it's just designed to drive a 50ohm load, is this correct?
Your presumption is quite correct. In fact, most amplifiers are designed to deliver a certain performance when driving a rated load, such as the popular 50 ohms used in RF work. They are not designed to have an output impedance of 50 ohms. They are designed to drive a load of 50 ohms, that's all. There are many other characteristics that must be traded off to achieve optimum performance and the best tradeoff does not occur when the amplifier's output Z is 50 ohms.
Maximum conversion efficiency is usually achieved by setting the amplifier's output impedance low relative to the load impedance. At times like that, you can get DC to RF conversion efficiencies of up to 90% even at RF frequencies (ref: "The Electronics of Radio", David Rutledge, pp 180-196). This is certainely not the point of maximum power transfer. It is common to find power amplifiers for GSM cellphones that do indeed deliver efficiencies of 60%, at frequencies like 900 or 1800 Mhz. These amplifiers do not have 50 ohm output impedance, and they are not linear but instead operate in Class C.
Another point to make is to reiterate the maximum power theorem, which says that with a given thevenin voltage source with fixed output impedance you will deliver the maximum power to a load if the load impedance is the complex conjugate of the source impedance. However, and this is the key point, if you can vary the source impedance as well as the load impedance you can have better efficiency. Consider taking the source impedance to the extreme of 0 ohms. In this case, all the power available from the source is delivered to the load since there are no internal source losses at all. So the most efficient setup is to set the output Z as close to zero as you can.
The reason this isn't done in some RF circuits is because is impossible to get the RF amp down to zero ohms of output Z while making all the other critical things as good as possible (things like gain, stability, distortion, efficiency etc.). But they do try.
Conversely, the efficiency also tends towards 100% when the load R tends towards infinity but this time the total power also tends towards zero.
A good explanation of this with further references is here:
https://en.wikipedia.org/wiki/Maximum_power_theorem