Bi-Wiring From Amplifier To Loudspeaker

[admin]

"Bi-wiring" is a controversial topic. Some people are quite certain it makes an audible difference. Some others are convinced that it can't actually make any difference at all. The purpose of this analysis is to try and decide whether it is at least theoretically feasible that bi-wiring can make any difference. This is a tech article meant for propeller heads and not a general overview.

Discuss "Bi-Wiring From Amplifier To Loudspeaker" here. View the article at:
http://www.audioholics.com/education/cables/bi-wiring-from-amplifier-to-loudspeaker.

 

[davo]

What, is every one to chicken to post a comment?

At the risk of exposing myself as a fool... thanks for breaking it down for us lay-men. *cough why does my brain hurt cough* My high school physics were slowly re-awakening by the end of the article, but the graphs were fairly explainatry.
I have a simple idea where the concept of bi-wiring might have originated. As was stated..

" In addition, with any reasonable choice of cable, such thermal effects would be tiny, and orders of magnitude smaller than similar effects inside the loudspeaker itself!"

My point been , what if you had speaker wire of 40AWG for example? (0.0031inch/0.07987mm) Would that cause the sound to attenuate? (I'm not suggesting distortion) You wouldn't know until you had reason to double it up with more 40AWG. Suddenly it sounds more discernable, louder (because of less resistance)... you see my point. Hence the dogma of bi-wiring is born. Of course once you get to a wire size where the quote becomes applicable then the dogma goes out the window.
But some people don't know when to let go.


I could be a fool, just waiting for another fool to tell me ...

 

[gene]

My point been , what if you had speaker wire of 40AWG for example? (0.0031inch/0.07987mm) Would that cause the sound to attenuate? (I'm not suggesting distortion) You wouldn't know until you had reason to double it up with more 40AWG. Suddenly it sounds more discernable, louder (because of less resistance)... you see my point. Hence the dogma of bi-wiring is born. Of course once you get to a wire size where the quote becomes applicable then the dogma goes out the window.
But some people don't know when to let go.


I could be a fool, just waiting for another fool to tell me ...

I think 40AWG is an ultra extreme case. The wire would likely melt before you put any appreciable power through it.

You can however likely hear sonic differences of say 26AWG wire compared to 10AWG wire for long runs. Most notably would be in bass as the wire with higher resistance may actually sound like the system is producing more bass but with obvious less control.

To calculate insertion loss differences, simply use 20 * log of the voltage divider relationship between the cable resistance and the nominal speaker impedance.

For example: a 20ft speaker cable into an 8 ohm speaker
10AWG: 3mohm/ft x 20 = .06 ohms
26AWG: 81mohm/ft x 20 = 1.6 ohms

Rloss (10AWG) = 20 * Log [ 8 / (8 + .06) ] = - .065dB
Rloss (26AWG) = 20 * Log (8 / (8 + 1.6) ] = -1.58dB

Thus there will be an overall voltage difference of about 1.5dB using 26AWG wire at these lengths compared to using 10AWG cable.

 

[davo]

Yes, 40AWG is extreme and practically unusable but it was to make obvious the point that I was trying to demonstrate. Obviously I'm no tech head but it seemed logical the way I put it.
I'm just glad I didn't get banned for putting forth rediculous scenarios and propaganda

 

[jneutron]

The purpose of this analysis is to try and decide whether it is at least theoretically feasible that bi-wiring can make any difference.

No, it is not. "Trying to decide" imply's that a decision has not been made.

Leseuf misrepresents the clearcut, logical, accurate analysis I presented. As such, he creates a strawman argument, and debunks it. Nowhere has he discussed or rebutted what I have presented, but rather, a convienient construct of his own choosing. And, in point of fact, the construct he uses to bebunk is actually the construct I state does not show this dissipation modulation.

So, now AH presents the exact same strawman argument, and wished to discuss "analysis"

Interesting..which, btw, is what Jim called my analysis on the followup e-mail where he may actually understand what I've been saying all along.

This is a tech article meant for propeller heads and not a general overview.

Now that would be correct..

Again, what is the purpose of arguing against foregone conclusions???

Cheers, John

 

[tbewick]

No, it is not. "Trying to decide" imply's that a decision has not been made.

Leseuf misrepresents the clearcut, logical, accurate analysis I presented. As such, he creates a strawman argument, and debunks it. Nowhere has he discussed or rebutted what I have presented, but rather, a convienient construct of his own choosing. And, in point of fact, the construct he uses to bebunk is actually the construct I state does not show this dissipation modulation.

So, now AH presents the exact same strawman argument, and wished to discuss "analysis"

Interesting..which, btw, is what Jim called my analysis on the followup e-mail where he may actually understand what I've been saying all along.



Now that would be correct..

Again, what is the purpose of arguing against foregone conclusions???

Cheers, John

In comparison to bi-wired cables, how much more signal degradation occurs in ordinary non-bi-wired cable? Could you state this in terms conventionally used in audio - the amplitude of the cable-induced error in relation to original signal and the lifespan and characteristics of this error.

What types of signals would be worst affected by these degradations?

How often would such signals be encountered in ordinary listening?

What experiment would show that these signal degradations exist?

How audible do you think the signal degradations caused by bi-wired and non-bi-wired speaker cables are?

 

[jneutron]

In comparison to bi-wired cables, how much more signal degradation occurs in ordinary non-bi-wired cable? Could you state this in terms conventionally used in audio - the amplitude of the cable-induced error in relation to original signal and the lifespan and characteristics of this error.

The use of a monowire doubles the dissipation loss of the wire at times, and at times it zero's the loss. The net RMS losses are identical between the two.
If the wire losses are 5% of the signal from the amp, the dissipation will go from 10% to zero.

In terms conventional to audio, there are none. In fact, the biggest issue I've come across is the nature of the problem being outside current understanding.

What types of signals would be worst affected by these degradations?

It's not the type of signal, but rather, how the signal is divided up at the load. The reactive branches have to share one resistive channel...if you have a three way, the math gets even wonkier, nevermind music..how does one define music mathematically.

How often would such signals be encountered in ordinary listening?

Again, it's the nature of the branching. Whenever both loads are going, the issue exists.

What experiment would show that these signal degradations exist?

Take post #146 (if I recall correctly the number), throw a differential probe set across both tweeter loads.. run the highs only, balance the gains to zero differential, then apply the low frequency watching the highs differential.

How audible do you think the signal degradations caused by bi-wired and non-bi-wired speaker cables are?

Quite honestly, I do not know, nor do I care. I use #12 even for my 100 footers, and don't worry about it.

If it's so low that most humans cannot tell it exists, it's gotta be manifesting very little. I have my conjecture as to how it manifests audibly, but I have nothing in factual evidence to support such conjecture.


Honestly, I'd worry about room and eq, and leave the wires thick, and ....alone. At least, until such time as audibility proof comes along. That proof is not my concern. Just the analysis and it's ramifications.

Cheers, John

 

[Seth=L]

Honestly, I'd worry about room and eq, and leave the wires thick, and ....alone. At least, until such time as audibility proof comes along. That proof is not my concern. Just the analysis and it's ramifications.

Cheers, John

Heh, heh, this really makes me laugh a little. You are saying that untill someone can prove that it makes an audible difference you are just fine with fatty wire. But if someone does prove that it makes an audible difference, however unlikely or likely it may be, you would change your wiring based on that proof. But then you would have to see if you could tell the difference, and lets say for some reason you couldn't hear a difference, would you still have it this way.

 

[stratman]

Does anyone have a dead horse laying around somewhere? If you do, bring it here so I can club it and keep clubbing it!

 

[jneutron]

Heh, heh, this really makes me laugh a little. You are saying that untill someone can prove that it makes an audible difference you are just fine with fatty wire..

Give or take, yes. I am more than happy to cart around #12 wire to do the job. Hey, if I though I heard a difference, I'd worry about it. Since I don't hear it, either it's too low to worry about or I'm just deaf...a difference that makes no difference, is no difference. Why would I bother doubling the number of connections to make?

But if someone does prove that it makes an audible difference, however unlikely or likely it may be, you would change your wiring based on that proof.

Nope. I would not bother. It's bad enough carting one set of #12's, if somebody proves it is audible in some ridiculous conditions where one is sitting in the sweetspot, how does that affect a 450 seat venue application, or even at home where more than one person listens??

But then you would have to see if you could tell the difference, and lets say for some reason you couldn't hear a difference, would you still have it this way.

Nah, as I said, it's way too convienient to use one set of wires.

If I were intent on sitting in the sweetspot, and devoting my listening to image placement, maybe I'd bother, maybe not. My priorities in life currently do not include that level of nitpicking.

Cheers, John

 

[ajaynejr]

Without biwiring, when a heavy bass passage comes along, there will be a slight voltage drop in the speaker cable due to the larger current flowing. The difference in the mid range and higher frequency reproduction may or may not be sensed audibly.

If biwiring is used, the voltage drop will only occur in the cable feeding the woofer.

But if the dual cables are paralleled back together at the speaker (figure #3 in Audioholics' discussion) for whatever reason, then the voltage drop will again occur in the circuit to the midrange and tweeter speakers. This is because the current flow to the woofer will split between both cables. The total voltage drop will of course be smaller given that the cross section of both wires taken together is greater. The sonic effect is no different than using a single fatter cable to begin with (and may have been inaudible in the first place).

Video hints: http://members.aol.com/ajaynejr/video.htm

 

[mtrycrafts]

Without biwiring, when a heavy bass passage comes along, there will be a slight voltage drop in the speaker cable due to the larger current flowing.

Yes, but this is frequency dependent voltage drop. Why would the out of that frequency band be affected by that LF voltage drop?

 

[ajaynejr]

As far as I know, voltage drop across pure resistance (which a straight copper wire is) does not distort the waveform. This means that both the heavy bass content and any mid-range and treble frequencies present would be attenuated proportionately when the current in the wire increased due to commencing of heavy bass content.

 

[jneutron]

As far as I know, voltage drop across pure resistance (which a straight copper wire is) does not distort the waveform. This means that both the heavy bass content and any mid-range and treble frequencies present would be attenuated proportionately when the current in the wire increased due to commencing of heavy bass content.

Very close in your description. The problem is standard circuit analysis along those lines shows there to be zero effect..what I point out is standard circuit analysis does not consider the branches at the load which split the signal.

If you look at the way leseuf setup his argument, you note that he did not branch the loads, and used merely one. The problem stems from feeding a branch system composed of reactive elements.

Cheers, John

 

[mtrycrafts]

As far as I know, voltage drop across pure resistance (which a straight copper wire is) does not distort the waveform. This means that both the heavy bass content and any mid-range and treble frequencies present would be attenuated proportionately when the current in the wire increased due to commencing of heavy bass content.

You mean that the normal voltage dropped by the mid and high frequency due to their current draw, would be further affected by a large current through that wire at say 50Hz?

 

[ajaynejr]

Come to think of it, I agree that the amplifier-speaker circuit behaves differently for each frequency given the frequency dependent impedances of the crossover networks and speaker voice coils. I still contend that yes, the addition of a heavy current at 50 Hz will increase the speaker cable voltage drop incurred by higher frequencies sharing the same cable. Then, given the frequency dependent impedance of the crossover, etc. together with the different voltage drop across the cable, a completely new set of proportions of the higher frequencies will be present at the midrange and tweeter voice coils.

 

[PENG]

I thought this was a dead horse. Anyway, for those who have basic knowledge of electrical theory and circuit analysis will know that it is not hard to prove bi-wiring makes a difference electrically for the signal (as shown in the analysis) but it is difficult to prove if such difference is audible to most people.

 

[BMXTRIX]

I thought this was a dead horse. Anyway, for those who have basic knowledge of electrical theory and circuit analysis will know that it is not hard to prove bi-wiring makes a difference electrically for the signal (as shown in the analysis) but it is difficult to prove if such difference is audible to most people.

But what is lost on many is that this is the only thing that matters. Just like picture resolution vs. viewing distance is critical when considering screen size the question becomes most serious in far more practical terms than electrical theory.

An average person, with average hearing, with any average (or better) set of speakers: Are they going to hear a difference? And if they are not, then what else matters?

Why debate the number of atoms in your foot when you can still walk on it and still can't see them?

Bi-amping, of course, is a different consideration, but this is not bi-amping, it is bi-wiring and it seems that nobody has once offerred up any evidence that shows that real world audible differences will occur when using a good piece of 12 gauge wiring vs. a couple of thinner wires that add up to 12 gauge when put together.

 

[PENG]

But what is lost on many is that this is the only thing that matters. Just like picture resolution vs. viewing distance is critical when considering screen size the question becomes most serious in far more practical terms than electrical theory.

An average person, with average hearing, with any average (or better) set of speakers: Are they going to hear a difference? And if they are not, then what else matters?

Why debate the number of atoms in your foot when you can still walk on it and still can't see them?

Bi-amping, of course, is a different consideration, but this is not bi-amping, it is bi-wiring and it seems that nobody has once offerred up any evidence that shows that real world audible differences will occur when using a good piece of 12 gauge wiring vs. a couple of thinner wires that add up to 12 gauge when put together.

No disagreement at all! I also think it doesn't matter whether there is a physical/electrical difference, it is what you hear that counts. The only problem I have is that some people like to throw Ohm's Law and claim bi-wiring is the same as using thicker wire, and that it makes no difference whether you split the wires at the amp or at the speaker, or something to that effect. That's simply not true. Last time I commented on this topic was long time ago and in the end I failed to convince those people there is a difference (again, I am talking about physics/electrical difference, not audible difference) and they failed to convince me and may be a few others. That's why I thought it was a dead horse. It changed nothing. By the way, those who can't hear a difference bi-wiring may not hear a difference bi-amping neither, especially in DBT, but that's another topic for another day.

 

[jneutron]

An average person, with average hearing, with any average (or better) set of speakers: Are they going to hear a difference? And if they are not, then what else matters?

I believe the answer is...NO

I certainly have not heard a difference using music.

And, since I haven't, I choose not to biwire. If I were shown that it indeed is audible, I still will not biwire, as it does not matter to me.

Biamping, of course, is a different consideration, but this is not bi-amping, it is bi-wiring and it seems that nobody has once offerred up any evidence that shows that real world audible differences will occur when using a good piece of 12 gauge wiring vs. a couple of thinner wires that add up to 12 gauge when put together.

Concur..audibility has not been shown under rigorous test conditions. I am not confident dbt's are designed to find them accurately if they did exist, but there is a huge body of evidence so far that state no audibility.

Cheers, John