Re: impedace matching differences at low freq & High fre
There are a couple reasons why impedance matching is important.
One reason why it is important to impedance match a load to its source (which will probably have a transmission line connecting the two) in high frequency circuits is because if they do not match (or actually the source does not match the transmission line or the load does not match the transmission line) the voltage signal will reflect back to the source as a ratio of the mismatch. This is usually not desired. If you are trying to drive a lot of voltage on an antenna for high transmitting power, these reflections can exceed the dielectric breakdown in the antenna cable or ruin the power amplifier.
The other reason you also mentioned is the maximum power transfer. High-frequency circuits are often required to amplify very small voltages, such as the voltage signal that comes from an antenna. And the antenna represents a voltage source with a source impedance (for example, 50 ohm) that is a non-adjustable characteristic. To transfer the maximum power from the source to the receiver requires the receiver impedance to match the voltage source impedance.
You also pointed out:
in a lot of audio amplifiers voltage matching is done rather than impedance matching.i.e..(it is desired to have low output impedance of stage and high input impedance of next stage for maximum voltage transfer)....
The maximum power transfer condition does not get the maximum voltage from the source to the load, in fact it causes the delivered voltage to be half of the source voltage. So why use a power match, especially if it attenuates the voltage signal that is already very small?
It would be possible to use a high impedance load (like in the audio amp) and obtain the full source voltage at the load, but the power in the signal received at the load would be very small, P=(V²)/R (we might double the voltage level but decrease the power level by 1000.) Which brings up the issue of noise sources. Any resistive element generates thermal noise: resistors, semiconductors, wires, etc. and these noise generators are power sources (not voltage sources,
per se) that compete with the received power signal at the input for amplification. By matching the source and load according to the maximum power transfer theory, the transferred signal has the best chance to overpower the competing noise "signals" at the load. The voltage source and the noise sources all deliver their power into exactly the same load impedance. If the signal power delivered to the load is greater than the noise power delivered to the load, so too will the signal voltage at the load be greater than the noise voltage at the load. Even though the absolute value of the load voltage is half of the source voltage for maximum power transfer condition, the impedance match helps overcome the effect of noise at the inputs.
(This explains why high-frequency circuit design is interested more with power measurement than voltage or current.)
Using the maximum power transfer theory makes less sense in other applications, like the audio amp, because maximum power transfer condition is not a power efficient condition (the zero source impedance, infinite load impedance is the efficient condition.) In low-freq circuits, strong signals are available and are far greater than the noise generators in the circuit. Voltage sources are often very low impedance by design, and we can treat the voltage sources as capable of supplying infinite power (by that I mean we often neglect that the voltage source has a source resistance.) So, using the maximum power transfer condition between source and amplifier would waste a lot of power, when it is only necessary to transfer the voltage signal level from source to load. Of course the maximum power transfer condition will apply at the final output stage where it is desired to transfer power from the amp to the loudspeaker and get real work done.