What if you reverse the connections on the microphone?
What if there is no microphone to generate the voltage signal, but a guitar pickup, for example?

Provide concrete evidence that the phase of the received acoustic wave is discernible to the human ear.

 

for something like the 1812 overture with an actual cannon, psycho-acoustic research shows that there is a difference in the perceived sound between that initial *crack* of the cannon being the compression part or the rarefaction part of a cycle, because positive pressure can be directed, but negative pressure cannot.

Even if that were to be true, you would have to have a controlled polarity all the way from the microphone to the speaker output to maintain the sound polarity.
I highly doubt that is done for any music recording.

 

What do you consider to be a reliable answer?
The physics of acoustic systems is well understood.

For one thing, it's insanely hard to model real world acoustics. There are so many variables including speaker position relative to each other, geometry of an enclosed space and geometry of every object in the room. On top of that, hearing capability ranges significantly from person to person. To make it worse, much of what is well known in physics is actually an extrapolation of related experiments with many small sample sizes to boot.

As great as computational models are, they are still in their infancy for a model with this many elements. It's similar to thinking we can or have modeled the environment.. it's not obvious what the "environment" really is from a metaphysical point of view because we don't know what we don't know as the expression goes. In these cases I default to the position of a skeptic simply because I haven't been exposed to sufficient evidence and a methodology I can understand.

Personally I find answers like "it's well known" to be one step above an assumption in reliability. Well known by who? A group of technocrats? There is a lot of bad science out there and I think it gets perpetuated in large by folks being told something is true rather than confirming it to be true themselves.

 

That's simple. Remove or restrict as many of the variables as possible. We do this in science and engineering all the time.
You want to study the acoustic effects of phase inversion at a loud speaker? You do this in an acoustic anechoic chamber.

You think we don't know enough about the theory of acoustics?




What you are looking at is the image of a cranial ultrasound.
Sound waves are generated by an array of transducers into the human scalp and we can analyzer the reflections returned to the transducers.

Or the ulrasound image of a baby in its mother's womb:



Does phase matter? Absolutely.

Does phase matter with a single loudspeaker while listening to music? Absolutely not!

 

Even if that were to be true, you would have to have a controlled polarity all the way from the microphone to the speaker output to maintain the sound polarity.

That is exactly correct, but it is not nearly as difficult as you might think. For the mic and speaker people, increasing pressure produces increasing voltage, and increasing voltage produces increasing pressure. For everyone else in the chain, and there can be dozens of pieces of equipment at play, it is very simple: no matter what goes on internally, the signal phase is non-inverting from input to output.

I highly doubt that is done for any music recording.

Actually, yes, it is, for almost all music recording. That's like doubting that Sony checks the speaker phasing on every amplifier they produce.

From the 1970's to the 2000's I've climbed around in maybe two dozen audio consoles, large and small. In that world, it is as natural as correct speaker phasing for stereo.

I was instructed in this while installing the cables in the station's new videotape room (*16* videotape machines plus 3 film chains). During a break I wandered down to the FM / radio master control room to see what the other half was up to. They had just received a shiny new SpectraSonics audio console, and were ringing out the connections. With a 2-channel scope and a test oscillator, they checked the signal phase from every input to every mix bus to every output. Years later I did the same thing in two different remote trucks.

From the diaphragm to the cone, it really is a thing.

ak

 

You think we don't know enough about the theory of acoustics

Yes, that's a good way to put it. I'm not being flippant but your example of ultrasound imaging is exactly what I'm talking about. That's a technique for the visual system not auditory. It's conflating different phenomenon. There are certainly properties that are applicable to both systems but the heart of the issue is where is the evidence for the thread starters question?

And for the record, I don't always personally hold the opinion that I'm arguing in favor for. Sometimes I just want to see how people push back on an idea with a solid argument of their own. I digress and bid you good day sir.

 

I've wondered this myself and never got a reliable answer.

I am responding to this comment you made.

What was the question you asked?
Whom did you ask?
What was the answer you received?
Why was the answer not reliable?
What makes for a reliable answer?

If the question was, "Does polarity on a single loudspeaker system matter?",
my answer is categorically "No".

 

I am responding to this comment you made.

What was the question you asked?
Whom did you ask?
What was the answer you received?
Why was the answer not reliable?
What makes for a reliable answer?

If the question was, "Does polarity on a single loudspeaker system matter?",
my answer is categorically "No".

My interest in the topic is very close to thread starters question. But I don't care for terms like matter because that includes a value structure. I prefer to ask something like: Are there any situations in which a polarity reversal will cause a loudspeaker to behave differently?

Framed this way, I sincerely think the evidence provided in this topic is insufficient and biased. We didn't really define speaker nor polarity and other common words. This is fine and often necessary for concise communication but the validity of any argument is hard to verify when the premises don't include a clearly defined casual relationship.

For example, take what you said here:

If the question was, "Does polarity on a single loudspeaker system matter?",
my answer is categorically "No"

This asserts an argument in the form of If x then y. The trouble is we haven't defined x locally in this conversation nor are we sure that x can be reduced to a single coefficient.

Stated another way, why don't you simply get in your echo chamber and perform an experiment so thread starter and I can see the test results for ourselves? Obviously you don't have access to one and the test setup requires very specialized and expensive equipment.

Since we will almost certainly not get the evidence I'm demanding in the sample sizes I'm demanding, we have to go on faith that what you said is true.

And to be frank, there are groups of intellectuals who seem to think because they have a degree or years of experience in one field, that expertise caries over to another and others should bow to their wisdom. Whenever I see that happen, I demand evidence and they often reply empty handed. Anyway, this has become a philosophical debate so I'll leave it there.

 

it is as natural as correct speaker phasing for stereo.

Yes, that is not in dispute. For stereo, both channels must be in phase. But this thread is about the phase of a single speaker.

 

Stereo is two distinct channels why is "phase" mentioned? Two microphones record the performance, their positions would determine the phase difference if recording a pure single frequency tone, the wavelength of a 1 KHz sound signal is about 1ft.

If there was a pure 1KHs sine sound to your left, your ears would receive the signal 180 degrees out of phase.

No human could tell the difference between the various possible pairings of the four wires in a test with the piece of music.

 

Are there any situations in which a polarity reversal will cause a loudspeaker to behave differently?

Ok. That is your question.

How many situations can you identify?
In the context of single loudspeaker usage, 99.9% of the situations would be a human listening to music content at any volume level.

As I pointed out, this situation can be tested. Put a human in an anechoic chamber and ask them to differentiate between sounds reproduced by loudspeakers at different phase shifts.

Have I tested this? No.
Do I have the ability to test this, yes. We have anechoic chambers in our audiology department.
I don't have to test this because my knowledge of acoustics and properties of wave motion tells me that the human ear and brain is insensitive to acoustic phase angle.

Specifically, the part of your question "polarity reversal will cause a loudspeaker to behave differently?"
The answer is NO.
Loudspeakers are driven with AC electrical signal. Unless the loudspeaker is driven beyond its maximum design capabilities, the pressure wave produced will follow that of the input signal.

Understand that loudspeakers are not "perfect" acoustic transducers. The voice cone has mass, inertia, and tendency to deform at certain frequencies and power levels. We are working on the assumption that the loudspeaker is not driven to these extremes to cause it to not perform as an ideal acoustic transducer.

 

No human could tell the difference between the various possible pairings of the four wires in a test with the piece of music.

I think I could tell if the two grounds went to one speaker and the two positives went to the other.

 

Stereo is two distinct channels why is "phase" mentioned? Two microphones record the performance, their positions would determine the phase difference if recording a pure single frequency tone, the wavelength of a 1 KHz sound signal is about 1ft.

If there was a pure 1KHs sine sound to your left, your ears would receive the signal 180 degrees out of phase.

No human could tell the difference between the various possible pairings of the four wires in a test with the piece of music.

As I have already stated, this is a simple test to determine if mono or stereo speakers are in phase or out of phase.
The difference is very noticeable.

 

I think I could tell if the two grounds went to one speaker and the two positives went to the other.

Touché !

 

Stereo is two distinct channels why is "phase" mentioned? Two microphones record the performance, their positions would determine the phase difference if recording a pure single frequency tone, the wavelength of a 1 KHz sound signal is about 1ft.

No human could tell the difference between the various possible pairings of the four wires in a test with the piece of music.

Not true.
Have you never had a stereo music system?

Stereo is two distinct channels but all stereo recording's bass frequencies are generally common to both channels.
The wavelength of a 100Hz tone is about 10 ft.
Thus there is a significant loss in apparent bass response if the two speakers are out-of-phase, that is readily heard by any human (and I would presume some animals).
Also there is a more noticeable separation that there are two separate speakers with apparent loss is center sound.

At one time in an early stereo system I had, I added a cross-coupled DPDT switch in one of the speaker leads to easily reverse the leads while listening to music.
I could easily tell which connection was in-phase and which was out.

 

To further expand on my position.

The human ear is sensitive to acoustic frequency and amplitude, not phase.
In order to discern differences from an acoustic transducer, you would need constructive and destructive interference which would affect acoustic pressure. This can happen when signals intersect having traveled via different paths such as encountered in a room, concert hall, auditorium, etc.

Maximum and minimum points can be detected with a microphone when a single sinewave frequency is emitted.
With music content, no maximum or minimum points occur.

 

Ok. That is your question.

How many situations can you identify?
In the context of single loudspeaker usage, 99.9% of the situations would be a human listening to music content at any volume level.

As I pointed out, this situation can be tested. Put a human in an anechoic chamber and ask them to differentiate between sounds reproduced by loudspeakers at different phase shifts.

Have I tested this? No.
Do I have the ability to test this, yes. We have anechoic chambers in our audiology department.
I don't have to test this because my knowledge of acoustics and properties of wave motion tells me that the human ear and brain is insensitive to acoustic phase angle.

Specifically, the part of your question "polarity reversal will cause a loudspeaker to behave differently?"
The answer is NO.
Loudspeakers are driven with AC electrical signal. Unless the loudspeaker is driven beyond its maximum design capabilities, the pressure wave produced will follow that of the input signal.

Understand that loudspeakers are not "perfect" acoustic transducers. The voice cone has mass, inertia, and tendency to deform at certain frequencies and power levels. We are working on the assumption that the loudspeaker is not driven to these extremes to cause it to not perform as an ideal acoustic transducer.

Aha! So you could test it but you haven't. I'm not saying you should either, I'm simply pointing out that no one has. However, I'm open to compromises, I'll settle for a study or statistics on the exact subject. I'm willing to bet these are few and far between for reasons I've laid out.

I suppose my argument can be summarized in two parts. First no one has provided real world experimental results and the experiments themselves are difficult to formulate, carry out, interpret and distribute. These are some very difficult and limiting problems that tend to get ignored in online discussion.

 

I'm simply pointing out that no one has...

...First no one has provided real world experimental results and the experiments themselves are difficult to formulate, carry out, interpret and distribute. These are some very difficult and limiting problems that tend to get ignored in online discussion.

This is categorically false.
A quick search on the internet will provide research on phase audibility.

 

The human ear is sensitive to acoustic frequency and amplitude, not phase.

Nope. Phase discrimination contributes to directional awareness.

ak

 

In response to the question about "Would a small DC offset affect the performance of a loudspeaker?"

The answer is, yes.
Any DC current would cause a displacement from the zero-current position and would add to the power dissipation of the voice coil.

How would it affect the sound quality?

The result would be similar to the performance of a Class-A audio amplifier that is biased away from its optimum quiescent operating point. At higher signal levels, the loudspeaker would begin to distort sooner on one side of the excursion. If the loudspeaker is not over driven, then you will not notice the difference.