[fmw]

The problem I have with polarity is that some amps have inverted polarity and some don't.  For a HT system, that's problematic, as I have to remember that my Jeff Rowland amp that runs my R/L speakers is reversed in polarity but my Bryston 9B or Tube Amp or Nuforce amp, which runs my center and/or rear speakers, is not.  Thankfully, Avia has test tones to discern if speakers are phased properly.

Probably not a big deal.  Center channels normally carry different content from the mains.  When they do carry the same content, they normally carry it at different times such as tracks that follow a moving subject.  You aren't likely to hear any bass cancellation or other phase related anomalies in such a situation.  It isn't the same thing as having stereo speakers out of phase with each other.  I wouldn't spend a second worrying about it.

[Kevin Haskins]

Do you have any research you can site?

[Steve Eddy]

Do you have any research you can site?

Who?

se

[ted_b]

I assume they do it to pander to the audiophile community.  I even have some CD sets with the same content on two CD's with opposite polarity, so others besides amp manufacturers do it also.  I'm pretty sure no one has ever heard a difference in polarity in a properly conducted blind test.  So it would be one of those "belief" things.  Is there a technical reason for it?  I don't believe so.

Steve Eddy has just explained the technical reason for it.  And in fact it's very easy to hear polarity differences blind - I've done it myself using a PC-based ABX program and cheap headphones, 10/10 correct - if you construct a test signal designed to accentuate the differences (which I did).  It's extremely difficult on music, but I do know of two successful ABX trials using carefully chosen short musical selections.

You are on a different subject.  Let's keep things on track.  The OP asked about polarity switches on amplifiers.   He's wondering what benefit it provides to his listening.  If you think you can tell which way the polarity switch is set on an amplifier by listening to any music in a properly conducted blind test, I'd be happy to put up some money as a bet that you can't.  How do I know?  I've done the tests many times with many audiophiles.  Polarity isn't audible in music.  You can believe it is as many do, but it isn't.  The polarity switch is just an audiophile "feature" with no real world purpose except, perhaps, to play with "specially designed test signals" designed to point out the polarity.

Once again...???  The OP did not ask about polarity switches on amps.  Are you reading the same OP as me?
Here is his exact question:
"I was curious if anyone here could tell me why some manufacturers allow their components to invert polarity.  Is there actually a technical or design reason why this would be an optimum way to build a circuit?"

He is, in fact, asking if inverting polarity is a design feature or is it a byproduct of a design, and why.  Many here have tried to explain that many analog stage designs take a simple route, and if they go through certain cathode paths, will invert absolute polarity (it's an electrical thing).  It is then much simpler to inform the user than add more sand and glass to the path, rather than do the re-inverting at the ouptuts per se.   

You are arguing that polarity is inaudible in music, which is a fine and fair argument, but more off topic than the OP.  He asked why Rolex's have certain design charateristics and you respond to say "it's just a watch, nevermind the design, time is irrelevant."

[fmw]

I appreciate you correcting me.

[opaqueice]

Acoustical waveforms are made up of compressions and rarefactions. That doesn't change simply because you have multiple microphones at different locations. You can only change that electronically somewhere between the microphone and the loudspeaker.

Well, kind of.  Some sound sources (most, actually) have a polarity which depends on angle.  For example if you pluck a string towards you you'll hear one polarity, and if you pluck it the other way you'll hear the opposite.  Same thing if you strike a single membrane or gong.  Depending on where the mic/ear is, the leading edge of the signal can be either a rarefaction or a compression.

Secondly, when you mix channels from multiple mics you often have to delay or invert the polarity of some of them to avoid phase cancellations.  Actually mics routinely have a polarity inversion switch for that kind of thing.  So by the time the recording gets mixed down to two tracks, any notion of a correct polarity is gone.

[Kevin Haskins]

Do you have any research you can site?

Who?

se

You.... I'm not trying to be a smartass.  I'm just curious because I've not read anything about the ear's ability to discern between a rareification, and a positive pressure wavefont.    Its plausible at least.... but plausible is different than correct.   

Do you have a source?   AES article or otherwise would be great.   I'd be interested in reading more. 

[Steve Eddy]

Well, kind of.  Some sound sources (most, actually) have a polarity which depends on angle.  For example if you pluck a string towards you you'll hear one polarity, and if you pluck it the other way you'll hear the opposite.  Same thing if you strike a single membrane or gong.  Depending on where the mic/ear is, the leading edge of the signal can be either a rarefaction or a compression.

Point taken.

I was simply pointing out that with asymmetrical waveforms, moving the microphone some short distance closer to or farther away from the source as was stated in the post I was replying to isn't going to change the polarity of that asymmetry.

Secondly, when you mix channels from multiple mics you often have to delay or invert the polarity of some of them to avoid phase cancellations.  Actually mics routinely have a polarity inversion switch for that kind of thing.  So by the time the recording gets mixed down to two tracks, any notion of a correct polarity is gone.

Sure. In which case the advocates of absolute polarity suggest you simply set polarity to whatever sounds best to you.

se

[Kevin Haskins]

Acoustical waveforms are made up of compressions and rarefactions. That doesn't change simply because you have multiple microphones at different locations. You can only change that electronically somewhere between the microphone and the loudspeaker.

Well, kind of.  Some sound sources (most, actually) have a polarity which depends on angle.  For example if you pluck a string towards you you'll hear one polarity, and if you pluck it the other way you'll hear the opposite.  Same thing if you strike a single membrane or gong.  Depending on where the mic/ear is, the leading edge of the signal can be either a rarefaction or a compression.

Secondly, when you mix channels from multiple mics you often have to delay or invert the polarity of some of them to avoid phase cancellations.  Actually mics routinely have a polarity inversion switch for that kind of thing.  So by the time the recording gets mixed down to two tracks, any notion of a correct polarity is gone.

Yes... and on the opposite end, the loudspeaker, some ugly things are done with phase.   Not only do  you have a phase wrap with respect to frequency due to crossover, you have 180 degree phase issue with respect to ported bass systems.   The crossover wouldn't be an issue with absolute polarity but the ported output, 180 degrees out of phase with the transducer output would pose a problem.    Also, keep in mind that when you record something, the electrical phase is 180 degrees out of phase with the original.   When you play it back through the loudspeaker it would revert back because once again, you see 180 degrees of phase rotation with respect to the electrical signal when its converted back to acoustical output.

[Kevin Haskins]

Sure. In which case the advocates of absolute polarity suggest you simply set polarity to whatever sounds best to you.

se

And that gets us nowhere now does it.   

[Steve Eddy]

You....

Ah...

I'm not trying to be a smartass.

Oh no, I never thought you were. Just wanted to be clear who you were asking it of as there were a couple of posts between mine and your asking.

I'm just curious because I've not read anything about the ear's ability to discern between a rareification, and a positive pressure wavefont.    Its plausible at least.... but plausible is different than correct.   

Do you have a source?   AES article or otherwise would be great.   I'd be interested in reading more.

Actually nothing I said had anything particularly to do with whether or not polarity is actually audible. I was just pointing out that asymmetrical acoustic waveforms do have an "absolute" polarity which isn't changed by simply moving the microphone a couple of inches.

Having said that, you might find these of interest:

Observations on the Audibility of Acoustic Polarity
Greiner, R. A.; Melton, Douglas E.
Volume 42 Number 4 pp. 245-253; April 1994

Comments on -Observations on the Audibility of Acoustic Polarity- and Author's Reply
Monforte, John; Karley, Brent; Greiner, R. A.
Volume 43 Number 3 pp. 147-149; March 1995

Polarity and Phase Standards for Analog Tape Recorders
Vanderkooy, John; Lipshitz, Stanley P.
Preprint Number: 1795

Proofs of an Absolute Polarity
Johnsen, Clark
Preprint Number: 3169

Another Look at the Importance of Transducer Polarity in the Recording Studio
Kaiser, James A.; Hedden, Gary H.
Preprint Number: 3172

se

[MaxCast]

Hey, as long as Ralf sounds like Sam says he should sound you got nothing to worry about. 

[opaqueice]

Yes... and on the opposite end, the loudspeaker, some ugly things are done with phase.   Not only do  you have a phase wrap with respect to frequency due to crossover, you have 180 degree phase issue with respect to ported bass systems.   The crossover wouldn't be an issue with absolute polarity but the ported output, 180 degrees out of phase with the transducer output would pose a problem.    Also, keep in mind that when you record something, the electrical phase is 180 degrees out of phase with the original.   When you play it back through the loudspeaker it would revert back because once again, you see 180 degrees of phase rotation with respect to the electrical signal when its converted back to acoustical output.

Don't most multi-way speakers literally invert polarity on some drivers so as to avoid cancellations in the crossover region?  I suppose that depends on the order of the crossover, mod 2?

Here's an interesting link explaining the mechanism for the compression/rarefaction asymmetry of human hearing:

http://epl.meei.harvard.edu/~keh/cd846/

4th lecture, page 7.

[Ethan Winer]

I was able to prove that the effect I was hearing could not be due to speaker effects (I'll explain how if you care).

Yes, I care.

Very roughly speaking the ear acts as a half-wave rectifier, so asymmetrical waveforms sound different when inverted.

That makes no sense. If the ear were a half-wave rectifier at normal volume levels it would create THD and IM distortion. Of course, ears do distort and create IM products, but not as a matter of course with music played at normal levels.

I downloaded Lecture 4 from the link in your later post, but the file seems corrupt and page 7 in particular does not display. But if I'm reading page 6 right, it shows what agrees with my understanding - nonlinearity is dependent on level. So let's say it another way:

At normal volume levels where the ear is not driven largely into distortion, absolute polarity is not readily audible. Now, one might be able to contrive a special test signal that reveals absolute polarity better than normal music. But the usual "purist" position that absolute polarity is very important and must be maintained defies my own experience. Just because it might be slightly discernible in certain situations doesn't mean it's critically important for music reproduction.

--Ethan

[Daygloworange]

At normal volume levels where the ear is not driven largely into distortion, absolute polarity is not readily audible. Now, one might be able to contrive a special test signal that reveals absolute polarity better than normal music. But the usual "purist" position that absolute polarity is very important and must be maintained defies my own experience. Just because it might be slightly discernible in certain situations doesn't mean it's critically important for music reproduction.

--Ethan

Although I haven't done a specific test for absolute polarity and it's effects on what we hear, I suspect much along the lines that Ethan and Steve Eddy suggest, that retaining absolute polarity of recorded music is not that big of a deal.

I read an article somewhere where they suggest that the leading edge of any musical performance is where preserving proper polarity is absolutely critical in playback. In other words to make sure that a compression is not inverted to a rarefraction.

The problem I have with that is that the leading edge "attack" portion of a musical note is mostly "noise" (as in asymmetrical) that's followed by the symmetrical waveform.

Cheers

[Kevin Haskins]

I'll do some more reading on the subject.   For the moment though... I think that the following would be an accurate summary.   Do you agree Steve/Ethan?

1.   Absolute polarity can be audible under very specialized test situations.    Audible says nothing about better/worse, it just means its detectable.   

2.  Under the vast majority of situations its not audible. 

[Steve Eddy]

I'll do some more reading on the subject.   For the moment though... I think that the following would be an accurate summary.   Do you agree Steve/Ethan?

1.   Absolute polarity can be audible under very specialized test situations.    Audible says nothing about better/worse, it just means its detectable.   

2.  Under the vast majority of situations its not audible.

That seems to sum it up as far as the published research goes. However many of the proponents of absolute polarity would disagree with it and believe that it's rather trivially easy to hear.

se

[Kevin Haskins]

I'll do some more reading on the subject.   For the moment though... I think that the following would be an accurate summary.   Do you agree Steve/Ethan?

1.   Absolute polarity can be audible under very specialized test situations.    Audible says nothing about better/worse, it just means its detectable.   

2.  Under the vast majority of situations its not audible.

That seems to sum it up as far as the published research goes. However many of the proponents of absolute polarity would disagree with it and believe that it's rather trivially easy to hear.

se

Great... it should be trivial for them to publish a peer reviewed paper on it then and set us all straight.

[opaqueice]

That makes no sense. If the ear were a half-wave rectifier at normal volume levels it would create THD and IM distortion. Of course, ears do distort and create IM products, but not as a matter of course with music played at normal levels.

I think you're forgetting something - ears are connected to brains, and brains do a lot of processing.

I downloaded Lecture 4 from the link in your later post, but the file seems corrupt and page 7 in particular does not display. But if I'm reading page 6 right, it shows what agrees with my understanding - nonlinearity is dependent on level. So let's say it another way:

The file works fine for me.  But in any case look at the bottom plot on page 6.  That's the electric potential (i.e. the signal generated in the auditory nerves) versus pressure.  Notice how it's asymmetrical?  The magnitude of the voltage is much smaller for a rarefaction than for a compression.  If you manage to view the next page, you'll see a plot of the response to an oscillating pressure, and the negative parts are (mostly) chopped off.  (Although only for the inner hair cells - not for the outer ones - but in any case it's clear the neural response is NOT symmetric between rarefaction and compression.)

Now, here's how I did my test:  I generated an asymmetrical tone (on my laptop) by adding a pure tone to some harmonics with rotated phases.  DC offset was zero (not that it would matter).  Then I polarity reversed it, so I had two files - one with more power in the positive half, one with more power in the negative half. 

Using a PC ABX program, I listened through Grado headphones to the two tones.  The effect wasn't obvious, but listening carefully you could hear a slight difference in pitch.  Let's say A sounded higher than B, and I could pretty easily score perfectly on the ABX by listening for it.  So the polarity reversal was audible, but was it my ears or was it the headphones or something else in the playback chain?

Grados, as you may know, are open-backed.  Furthermore you can rotate the drivers around horizontally, so you can wear them with the speakers pointing out, away from your ears.  Obviously not something you'd want to do normally, but in this case when I flipped them around, B sounded higher than A!  Why?  Because the drivers are dipoles, and turning them around turned rarefactions into compressions. 

I thought that was pretty clever, if I do say so myself...    aa  but in any case it proves beyond reasonable doubt that the asymmetry is in my head (literally!) and not in the audio chain.

[Steve Eddy]

Great... it should be trivial for them to publish a peer reviewed paper on it then and set us all straight.

One would think.

se