Voltage Follower

for a voltage follower, the input impedance must be high and the output low..

pardon me.. but wont the circuit get loaded if the output impedance is low?

No, by "output impedance" is meant the impedance for charge "going out of" the voltage follower.

If you connect the output of the voltage follower to some circuit X, then if X has a low input impedance, then the output of the voltage follower will be loaded.

// Mankku

low output impedance means that a given load doesn't load down the output voltage of the follower as much.

Yes, low output impedance renders a circuit less susceptible to loading, because a low output impedance means the circuit can supply more current before voltage drops too far.
Ohm's law.

google : "cathode follower" and "emitter follower" .
these are low impedance circuits .
these are normally used where the voltage swing is fairly small and to source current over voltage .

the output impedance of "voltage follower" (common collector or common cathode) circuit is usually mid to high impedance .
this is normally the case for amplifier circuits that produce a fairly wide voltage swing over current .

Hi CWB,
"this is normally the case for amplifier circuits that produce a fairly wide voltage swing over current". I think its the other way. Low output impedance mean very low voltage swing for a huge variation in current. Correct if I'm wrong.

google : "cathode follower" and "emitter follower" .

a cathode or emitter follower circuit is used when it is desirable to drive a (usually) low impedance load (think 75 ohm video) ... the phase follows the input signal .
the voltage swing/drop across the emitter (or cathode) resistor is much less than that of a common emitter (or cathode) amplifier where the output is taken from the collector (or plate) across a fairly high resistance .
the advantage is that appreciable amounts of current are available to drive a low impedance load .

the circuit originated in the tube era ... and most tubes are high impedance , voltage driven devices .
a way to obtain a low impedance (and wide band) output that did not require the use of a transformer (plate driven) was to insert a small resistance between the cathode and ground and take the output from across that resistance (with appropriate dc blocking) .

BJTs are low impedance , current driven devices (when compared to tubes) .
however , things being relative , it is still possible to achieve the same results with an emitter follower as those of the cathode follower .

FETS behave in a manner similar to tubes .
as an example , i read many years ago that one could liken them to a triode ; they are more of a voltage driven device .
MOSFETS are fairly close to tubes in some manners ... exceedingly low current requirements for the gate that can be likened to small signal triodes (etc) .


there is another similar circuit that is used (not so much these days) ...
it is commonly called a "phase splitter" .
it is used where two signals of opposite phase are needed from a single signal input and a minimum of parts required .
these types of circuits were used quite a bit in tube type audio amplifiers .
a google search of the term will provide a lot of information .

that's kind of true: low output impedance means small CHANGES in voltage swings when the CHANGES in current swings are big.

In other words, if your low-output impedance amp is outputing 5Vpp into a 10k load, it will output close to 5Vpp into a 1k load.

the same voltage swing (5Vpp) is maintained over a wide range of load.

Be it any circuit, typically all circuit will try to load the next stage. Loading effect will always be there, unless it is going to be a pure theoretical mode. Generally, when the 1st stage is very susceptible to load conditions and if the load of the next stage causes huge variation on the 1st stage, an intermediate stage such as voltage follower are used, which in turn takes in all the loading of the next stage & produces less load on the 1st stage. So, an voltage follower will always try to load the next stage. Amount of loading depends on the 2ns stage characteristics.

But it does come up with lot's of problem. It cannot be used where larger BW is required. In typical MOS, at higher frequency, when Cgs becomes effective, it looses the low impedance characteristics of the output, which requires additional Cap to solve. But again the value of cap depends on capacity of the 1st stage.