22

I've learned that it's desirable in an ideal amplifier to have high input impedance and low output impedance. Why exactly? What are the implications an amplifier has the opposite- low input impedance and high output impedance.

I don't exactly understand how impedance input and output.

Prabhpreet
  • 553
  • 2
  • 6
  • 15
  • 3
    Just wanted to note this is true of any "driver". Even a buffer, for example, wants to have really high input impedance and really low output impedance. There may be exceptions to this but it's a good rule of thumb. – NickHalden May 08 '13 at 17:17
  • Basically it is ideal to have high input impedance and low output impedance to be able to cascade voltage amplifier blocks. But it's not necessarily always true as it is ideal for current amplifier to have low input impedance and high output impedance. – lucas92 Feb 19 '18 at 14:53

5 Answers5

47

Actually, the premise of your question is only true if the signals you are interested in are voltages. In that case, if the amplifier draws no current through its input (has infinite, or at least very high input impedance), then connecting it to a source won't affect the signal voltage, regardless of what the source impedance is.

Similarly, when you connect a load to the output of your amplifier, if the amplifier has zero output impedance, the signal voltage won't change, regardless of the current drawn by the load.

These properties make it much easier to analyze the behavior of the system overall.

However, if the signals you're interested in are currents rather than voltages, you want your amplifier to have zero input impedance and infinite output impedance for the same reasons.

Dave Tweed
  • 168,369
  • 17
  • 228
  • 393
  • 21
    and if you're interested in *power*, you want to match impedances between source and load. – markrages May 08 '13 at 17:27
  • +1 That's a good call - transimpedance amplifier for photodiodes springs to mind for zero impedance input and, maybe a current source for feeding power down a POTS line – Andy aka May 08 '13 at 18:13
  • 2
    A key insight by Edison was that for _power_ you did **not** want to match impedances. Matching impedances means 50% power loss. You sometimes need to match impedances to reduce reflection: seldom to maximise power transfer. – david May 09 '13 at 07:23
  • @david: You need to explain what you mean, or provide some citations. Minimizing reflected power is the same thing as maximizing the forward power transfer. – Dave Tweed May 09 '13 at 10:54
  • 2
    A key insight by Edison was that maiximizing the forward power transfer is seldom what you want to do. In his case, because loosing 50% of the power made electricity supply more expensive, less popular, and less profitable. (http://en.wikipedia.org/wiki/Maximum_power_transfer_theorem) – david May 10 '13 at 11:35
  • @david: You're falling into the trap described in your citation: Maximizing efficiency is not the same thing as maximizing power transfer. In the context of this particular question, we are indeed talking about power, not efficiency. – Dave Tweed May 10 '13 at 11:51
  • 2
    It's not a trap I fell into: it is the specific point: If you are interested in _power_, you are seldom interested in maximising the power transfer ratio. – david May 13 '13 at 01:09
5

The ideal amplifier should not draw any current at all from its input. Assuming a two input amplifier the signal current in both input probes is zero. In other words the input impedance must be infinite!

The output, shoul operate as the ouptut of an ideal voltage source. This means that the pottential between the output and the ground must be \$A(v_2-v_1)\$, no matter how much current would a load conected to the output would draw. In other words the output impedance must be zero!

For a real amplifier, the input impedance must be as large as possible while the output impedance must be as low as possible!

Thanos
  • 257
  • 2
  • 9
  • 1
    An audio amplifier should not draw any *power* from its input. That implies that the impedance mismatch between the source should and the amplifier input should be maximized. An amplifier driven by a low-impedance source should thus have the maximum possible input impedance, while one driven by a high-impedance source should have the lowest possible impedance. The threshold where one should regard a source as low- or high-impedance is primarily a function of how well the impedance is known. If it's "at most" some value, consider it low; if it's "at least some value", consider it high. – supercat Dec 10 '13 at 19:50
4

Simplified answer relating mainly to audio amplifiers: -

An audio amplifier with a low output impedance can deliver larger powers to its loudspeaker more efficiently than an amplifier with higher output impedance. Thus, you will find that audio amps have output impedances measured in less than 1 ohm and in a lot of cases in milli-ohms.

On the other hand, a weak and feeble signal from (say) a microphone doesn't want to struggle with feeding its signal into an amplifer with a too low an input impedance - this may potentially (and significantly) attenuate the signal and require higher levels of amplification to compensate thus increasing noise pick-up etc..

If, an audio power amplifier's impedances were reversed, as you suggest, it would potentially produce a noiser signal on the loudspeaker and, be power-inefficient to the point that it would get significantly warmer in generating the equivalent sound level from the speaker.

There are other problems with low input impedances in that reshaping of the frequency response of some microphones may occur. This is also true of the high output impedance - the electro-mechanical nuances of the speaker may cause some signals to appear louder than what they should be.

As an aside, there are many amplifiers that do have a fairly low input impedance and these are typically in the field of RF where you need to match impedances to prevent signal reflections.

No-doubt there are other examples I've missed.

Andy aka
  • 434,556
  • 28
  • 351
  • 777
2

Negative Feedback is used to 'linearize' the output of an Amplifier. An amplifier is linear if it's output is an exact amplified copy of the input. If an amplifier is not linear, the output is distorted. Linear Distortion is one of the specifications that is given for any good audio amplifier.

A side effect of Negative Feedback is that, in common configurations, it creates a low-impedance output. No matter what load you put on the amplifier, the output Voltage is the same: the output voltage is feed back into the input to linearize the amplifier, and that makes the output voltage insensitive to load conditions.

So: Good amplifiers use negative feedback: voltage feedback is common: good amplifiers commonly have low output impedance.

The voltage response of some loads (particularly speakers) is quite non-linear, and quite frequency sensitive. For this reason, high-impedence-output amplifiers, that is, current-drive amplifiers, are considered by some audio engineers to be superior to low-impedence-output audio amplifiers.

High-output-impedence amplifiers are not necessarily any less efficient at transferring power into speakers (loud speakers inherently sound better than soft speakers), and are not inherently any less efficient at transfering power from the power supply into the speakers (electrical power in that quantity is cheap anyway), but current-sense circuits have always been slightly more difficult, slightly more expensive, and slightly less linear than voltage-sense circuits.

david
  • 4,364
  • 15
  • 30
0

It is desirable to minimize the loading effect either when the amplifier(voltage) is used to drive a circuit or when it is being driven by some other circuit.

Arunish
  • 1
  • 1