EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Beamin on September 06, 2017, 04:28:24 pm
-
So I understand how you can have a load with a different impedance a resistor with 5k ohms vs say a speaker with a few ohms resistance and 8 ohms inductance.
But how can the source have high/low impedance? When I see a source I think it has two parameters voltage and amps and maybe frequency. So what is the difference in say an audio amp that works with 8 ohm speakers and a 4 ohm amp with 4 ohm speakers? Also I see when using meters there are high and low impedance load that work with different meters
-
Impedance matching is used for RF circuits, to avoid reflections. The cables used in audio amplifiers are far too short to worry about transmission line effects.
An audio amplifiers should not have its output impedance matched to the load impedance. Its output impedance should be as low as possible, to give a good damping factor. Otherwise the speaker will be undamped and very boomy, especially at its resonant frequency.
An amplifier designed to drive a 4 Ohm load, will have half the output voltage, but double current rating, of one designed for an 8 Ohm load, with the output power being equal.
-
Source impedance is just how much the source voltage changes with load current. If you have a 1 volt source with 100 ohm impedance, the short circuit current will be 10 milliamp, and if you drive a 100 ohm load, the load will see 0.5 volt / 5 milliamp.
Audio amplifiers are almost always very low impedance (< 1 ohm), but they have two issues related to load impedance. The first is that they have some current and/or power limit. An audio amp producing 20 Vrms will provide 2.5 amp and 50 watts into an 8 ohm speaker. If you connect a 4 ohm speaker, it will try to supply 5 amps and 100 watt. If this is beyond the amplifiers capacity the result will not be as desired. The second limitation is stability. Audio amplifiers, like many amplifiers, use negative feedback to reduce distortion, control gain, and provide a low output impedance. However, the load impedance affects the feedback. An audio amplifier connected to a too-low load impedance might become unstable and oscillate which might destroy itself or the speaker.
So audio power amplifiers will have a specification for the minimum allowed load impedance. Normally the banner (max) power rating will be at the minimum impedance. Many amplifiers will rate for power into multiple common speaker impedance values. None of this has to do with the amplifiers output impedance which as I said is usually very low.
-
But how can the source have high/low impedance? When I see a source I think it has two parameters voltage and amps and maybe frequency.
Let's leave frequency aside for the moment and concentrate on the other two.
The voltage will, indeed, come from within the source - but the amps that flow will be dependent on how easily it can pass through the source-load circuit. The correct term for this is impedance, but for the sake of understanding, we can just talk about resistance if we stick to a DC example.
In an oversimplification - think of a source as providing a voltage - say 10V. Now, since this is a real world supply, there is going to be a limit as to how much current it can deliver. Assuming this source can survive such a test, to find out what this limit is, we short it out and measure the current that flows. (Assume we are using an ideal current meter.)
If we do this and find that we get 2 amps, then by using Ohms Law, we come up with a resistance of 5 ohms. This is the internal resistance of the source.
If your source has a slightly more whimpy transformer and other components, you may do the shorting exercise and get 1 amp. This would make the source internal resistance 10 ohms.
If your source had a really beefed up transformer and other components, you might do the shorting exercise and get 10 amps. This would make the source internal resistance 1 ohm.
As you can see, the lower the source resistance, the more current it is able to deliver - and this means the voltage it puts out will be less affected by any given load.
If we stick to passive loads, the resistance of the load is fairly straightforward.
Using Ohms Law, the current that flows through the source/load circuit is the source voltage divided by the total resistance - which is the internal resistance of the source plus the resistance of the load.
When you get into AC signals, there is a reactive component that comes into the equations - but the basic principles are the same.
So what is the difference in say an audio amp that works with 8 ohm speakers and a 4 ohm amp with 4 ohm speakers?
This is taking things to a new level - and this has been mentioned in other posts above. It is somewhat more complicated to explain, so I will not try and give a full lesson on the subject - but give you a couple of basic points to help with understanding (hopefully).
First, there is the obvious - that the source must be able to provide enough current to drive the load properly. If the source can supply 5A and the load wants 10A to function properly, then this is a problem. It may be that the load just operates poorly - or the source is overloaded - or both. If, however, the source can supply 10A and the load wants 5A to function properly, then all is well ... for starters, that is.
But with some types of loads - and your traditional voice coil speakers are one - it is a little more involved. You see, the speaker is not truly passive. You run a current through it, which generates a magnetic field that reacts with another magnetic field causing motion. This motion, however, introduces some complications.
First, there is the matter of inertia. Speaker cones may be light, but they do weigh something. When the speaker cone is stopped, it takes a jolt of energy to get it going - and once it's going, it wants to keep going. Imagine what it would be like at the transition from the rising edge of a square wave to the top of the waveform ... the speaker cone is meant to go from maximum velocity to a dead stop. For anything that has any mass, that is a big ask! Next is the force that opposes movement which comes from the surround and the spider. Following on from these two is the "back emf".
Rolling all these factors together makes for a few extra considerations in amplifier design - and why load impedances should be "matched" to that design. If you don't, you get poor performance - one of which is the "boominess" mentioned above.
Also I see when using meters there are high and low impedance load that work with different meters
If you mean what I think you mean, this is something that is a little bit different in usage....
Normally, you want a meter to offer as small an effect on a circuit as possible. When making voltage measurements, you want its input impedance to be as high as possible, so that it doesn't load the circuit.
But, sometimes, this will cause you problems.
There are times when you might be checking the voltage on a wire that you expect to be unpowered, but it might show as having a voltage of, say 90V. You could then spend a lot of time trying to figure out where it came from. You've checked both ends, examined the cable and can't see where that voltage is coming from.
The truth could easily be very simple: The wire is acting as an antenna and is picking up spurious emf that surrounds us. When you have a meter with an exceptional input impedance - like 10M ohm or more - connecting the meter does not load the wire much at all, so it is able to detect the induced voltage.
The solution is to load the circuit.
You can do this by adding a resistor of your own - but that is neither convenient nor safe - so some meters have a low impedance (LoZ) range where there is a deliberate load between the probes, which is meant for exactly this situation. By putting a load on the circuit, these "ghost" voltages will be greatly reduced - or may even disappear. The BM235 has just such a range.
-
As hinted at by the other answers, the "impedance" of a voltage source is (loosely) its equivalent series resistance -- think Thevenin equivalent circuits, if you're familiar with that.
I'm very relieved to see that the responses here are correct; and do not reflect the widespread belief that matching source and load impedance is useful for audio.
I can see where the mistake comes from, though. If you have a voltage source with an ESR of x ohms; and you want to choose a load resistor in order to get the maximum amount of power dissipated in the load resistor; you reach that maximum when the load resistor is also x ohms. So if you had no choice in amplifier and were choosing speakers with the bizarre goal of getting the maximum power dissipated in the speakers, you would match the speakers to the amplifier.
However, the story is different if you reverse the scenario. If you have a load resistor of y ohms, and have the opportunity of choosing different amplifiers with the goal of delivering the most power (or efficiency) to the load, then you want your amp to have an ESR of as close to zero ohms as you can manage. So if you have some speakers, and want to choose an amp, matching impedances is a stupid idea, even if maximum power deliver was your goal.
-
If you have a voltage source with an ESR of x ohms; and you want to choose a load resistor in order to get the maximum amount of power dissipated in the load resistor; you reach that maximum when the load resistor is also x ohms.
This is the Maximum Power Transfer Theorem. It's a simple enough concept, but it is all too often taken far too literally - and applied incorrectly - based on the title.
The key issue with this theory is that the power dissipated in the source is the same as the power dissipated in the load. To understand why this is not a good idea - imagine a power station that could provide 100MW to the grid. If it was designed to operate at the maximum power transfer point, it would have to dissipate 100MW on site. That would not only be a huge waste of power - it would heat up the power generating equipment to destruction very quickly.
Maximum power availability for power generation does not lean on the maximum power transfer theorem - it is based on the impedance of the source (as well as the distribution system) being as low as possible.
This principle extends into far more real world examples than the MPT theorem.
-
Might add that the notion of load impedance matching of audio gear harks back to the valve era. Valves had relatively high internal resistance, hence the load impedance of the speaker (in that case via a transformer) had to be at least equal to it, or hardly any power would be developed in the speaker.
In fact, the 'matching' was to a suggested load impedance published by the valve manufacturer, not the actual internal resistance of the valve, which was somewhat lower. Again the consideration was that an exact match would have led to max output but poor efficiency, and hence very hot running.
-
To the Ineffable All,
It is correct that the maximum power transfer occurs when the load is equal to the source, provided that the source cannot be made smaller. For instance, a source of 4 ohms will not transfer as much power to a 8 ohm load as it would to a 4 ohm load. But, if the source can be changed to 2 ohms, then even more power can be transferred to the 4 ohm load even if the source and load resistances are not equal. The ultimate best situation is for the source resistance to be zero. Then all the power is transferred to the load.
Ratch
-
Might add that the notion of load impedance matching of audio gear harks back to the valve era. Valves had relatively high internal resistance, hence the load impedance of the speaker (in that case via a transformer) had to be at least equal to it, or hardly any power would be developed in the speaker.
In fact, the 'matching' was to a suggested load impedance published by the valve manufacturer, not the actual internal resistance of the valve, which was somewhat lower. Again the consideration was that an exact match would have led to max output but poor efficiency, and hence very hot running.
I was going to make this point that for audio work the impedance is pretty much ignored these days. It better to have the stiffest drive possible or lowest source impedance. Modern amplifier generally have only a fraction of a ohm output impedance. It becomes much more important as the frequency increase but even at RF people generally worry far more than necessary about matching.