Impedance Matching - What, How, Why ???

[parisphoto]

I have heard that it is important to match the impedance of your separates.
Why ? How is it done / verified ? Speakers too ? What is the result of a mismatch ? Thanks to all.

 

[whoaru99]

Nothing personal, but it's mostly much ado about nothing. Not that it's not important, but in nearly all cases it's taken care of itself by the design of the gear. Meaning, it's not something you really need to worry about deciding unless, frankly, you're getting into (pardon the "French") somewhat oddball gear.

Most components have relatively low output impedance and most preamp and power amps have relatively high input impedance. Basically, this means the source doesn't really know the preamp or power amp is even connected to it. This is good because it means the output characteristics of the sources are not changed (or, very little so, anyway).

Speakers are a bit different, inasmuch as for most amps and receivers the preferred speaker impedance is 8 ohms, some handle down to 4 ohms or even 2 ohms. Here you have some choices and need to ensure the speakers you choose fit with the impedance recommendations for the amplifier/receiver. As I mentioned earlier, there are some exceptions, but relatively few and far between for the typical stereo buff.

 

[Jon_Logan]

It's not important to impedance match your analog audio.

You do want to impedance match your coaxial digital out, your video connections, and your antenna connections, however.

 

[parisphoto]

Wadia dock to Cambridge Dac Magic to pre amp to power amp.
Do I have to pay attention to impedance ?

 

[whoaru99]

It's not important to impedance match your analog audio.

We probably need to clarify what we are talking about for proper impedance matching.

Some would consider it proper impedance "matching" to combine low output impedance feeding high input impedance with minimum ratio of 1:10. This is what you want, generally, for audio, and nearly all audio components will have inherent properties to make this reality.

However, sometimes that is confused with the meaning of "impedance matching" as you mentioned for digital connections and video connections where the system impedance should be maintained as close to 75 ohms as possible, including using cable of the proper characteristic impedance.

 

[whoaru99]

Wadia dock to Cambridge Dac Magic to pre amp to power amp.
Do I have to pay attention to impedance ?

Yes and no.

If using a coaxial connection from dock to DAC, then yes, you should use a cable with 75-ohm characteristic impedance. A common "video" cable has this characteristic.

As to between the audio components, no, there is nothing to match, per se. See my comments above.

 

[Jon_Logan]

Wadia dock to Cambridge Dac Magic to pre amp to power amp.
Do I have to pay attention to impedance ?

For the Wadia dock to Cambridge DAC Magic, that sounds like a 75 ohm impedance matching situation on that link. "normal" patch cords from the Cambridge's audio out to the preamp, though.

 

[parisphoto]

Thanks. It's always nice to have another thing NOT to worry about.

 

[ferninando]

matched impedance insures maximum transfer of power. Mismatch either way reduces it, power.

 

[parisphoto]

How about sound quality ?

 

[whoaru99]

matched impedance insures maximum transfer of power. Mismatch either way reduces it, power.

I'm not sure this is true in all cases regarding the OPs concern, but again it depends on what you mean by "matched impedance".

For example, the connections from source component to preamp and preamp to power amp are intended to be optimized for voltage/signal transfer, not power transfer, thus low impedance feeding high impedance is best. If this is what you mean by "matched impedance" then we agree.

Also, power amplifier to speaker connections work best, afaik, when the amp output impedance is low compared to the speaker, creating as close to possible as a true voltage source from the amplifier.

These are my understandings, anyway, with regard to audio circuits. I understand that certain other circuits such as video and or RF/antenna systems do work best with matched impedances, but again, not the same thing as with audio circuits.

 

[ferninando]

I'm not sure this is true in all cases regarding the OPs concern, but again it depends on what you mean by "matched impedance".

For example, the connection between source component and preamp is intended to be optimized for voltage/signal transfer, not power transfer, thus low impedance feeding high impedance is best. If this is what you mean by "matched impedance" then we agree.

Also, power amplifier to speaker connections work best, afaik, when the amp output impedance is low compared to the speaker, creating as close to possible as a true voltage source.

These are my understandings, anyway, with regard to audio circuits.

just read the text book. If impedance is matched max power will be transfered. thats all im saying.

 

[whoaru99]

just read the text book. If impedance is matched max power will be transfered. thats all im saying.

No, I'm not going to read the text book....if you have a reason to say why "matched impedances" are proper for audio interconnection of components and speakers, please post them here so that all may learn.

 

[belgianbrain]

Improper impedance matches can sometimes have weird effects, like reflected or standing waves or the introduction of harmonics and other forms of distortion - but as far as I remember from my classes 10 years ago these tend to happen more often at higher frequencies than we're dealing with in the audible spectrum.

As noted above, I'm not sure the term is being used correctly in this context. Impedance matching is usually done to maximize power transfer in electrical power systems. Again, as noted above, that's not what we're doing here.

 

[Audio_Works]

For example, the connections from source component to preamp and preamp to power amp are intended to be optimized for voltage/signal transfer, not power transfer, thus low impedance feeding high impedance is best.

I agree with this, today most outputs are designed as voltage sources not power sources.

Also, power amplifier to speaker connections work best, afaik, when the amp output impedance is low compared to the speaker, creating as close to possible as a true voltage source from the amplifier.

This needs more discussion. Altho what you say is true that the lower the amps impedance compared to speaker the more it will behave as voltage source its not necesarily desired. With a big motor speaker with low Q one might want higher output impedance from the amp to achive desired damping and good low frequency response.

Improper impedance matches can sometimes have weird effects, like reflected or standing waves or the introduction of harmonics and other forms of distortion - but as far as I remember from my classes 10 years ago these tend to happen more often at higher frequencies than we're dealing with in the audible spectrum.

Problems would occur if the cable lenght is very high. There used to be a time when the standard impedance in audio was 600ohms.


All in all impedance matching has benefits when you deal with wave transmission. For example acoustic waves in Speakers! The reason why Horns are so efficient is that Impedances are matched between speaker and air load.

 

[33&athird]

matched impedance insures maximum transfer of power. Mismatch either way reduces it, power.

No, I'm not going to read the text book....if you have a reason to say why "matched impedances" are proper for audio interconnection of components and speakers, please post them here so that all may learn.

I think he's just pointing out that mis-matched impedances reduce power, in any circuit- not necessarily audio connections. I believe this is true, a matched impedance will certainly transfer the most power, potenitally.

For interconnects, etc. I think matching is a bit overdone for components, though way-mismatched things can cause problems, as stated, most gear kind just handles it. If it sounds good, it's probably alright:yes:.

 

[EchoWars]

oldhifiguy is right, but the only time it is important is when you're dealing with weak signals where a lot of amplification is expected (phono carts, radio signals, microphones).

 

[whoaru99]

Agreed, and I don't believe I disputed the significance in an abstract sense, only with regard to the Op's specific question(s). Generally I try not to obfuscate, but sometimes fail.

 

[Jon_Logan]

oldhifiguy is right, but the only time it is important is when you're dealing with weak signals where a lot of amplification is expected (phono carts, radio signals, microphones).

If it's HF, even large amplitude signals require impedance matching. You don't want reflections to sum with your freshly launched signals.

 

[Audio_Works]

Considering the fact that most preamplified outputs have a very low output impedance and that the cable is pretty short and that the audio bandwith is just 20kHz you dont really care about reflections because the source impedance will damp them. However when you need to transmit signals over long distances(rf, telephone systems, etc) thats when it becomes a problem. In this situations your generator can not have very low impedance because of stability issues caused by reactive elements in the transmission line (i.e. cable). Most line outputs and power outputs have zobel networks (impedance eq) to insure this stability.
Look at power amps, with SS you dont care because the "generator" has low impedance but with tubes you need a transformer otherwise you lose power. In many tube amps even between stages you will find transformers.
With weak signals you need adaptation to get the most signal out and to ensure a good bandwith.