Phase vs polarity

[Vaughan Odendaa]

Hi there,

I see that in some audio forums, phase and polarity are used interchangeably. But this is not correct IMO.

Because in my mind, phase would imply a time component. A delay in time is a phase relationship between two sound waves. But if one simply reverses the speaker leads from positive to negative, then one would be reversing polarity but the time relationship between both speakers haven't changed.

By reversing the speaker leads, the waveforum will be traveling in opposite directions. So the positive cycle will cancel with the negative cycle and there will be far less bass at the listening position.

Correct ? To me, the definition of phase means the point or angle that a sound wave will travel in the air in a point in time with other sound waves, measured in degrees.

Perhaps someone with better knowledge can explain the relationship (or lack thereof) and explain how reversing speaker leads does not equate to a change in phase but to that of polarity.

Thanks.

--Regards,

 

[jonnythan]

They're essentially the same thing, but used in different contexts.

When drivers are in phase, they are moving in and out at the exact same times as each other.

When they are 100% out of phase, one's movement is the exact opposite of the other.

Being out of phase, the signals have different polarization.

If you switch the leads of the speaker wire, you will have the speakers totally out of phase. One will be moving in while the other is moving out. They will create exactly opposite sound waves that will cancel each other out and generally sound quite awful.

 

[jonnythan]

Oh wait, I see what you're saying.

You're saying that an exact negative of the signal is not the same thing as a 180 degree phase shift. This is true, assuming the signal changes with time.

Reversing the polarity of the speaker wires does *not* create a true phase shift. It creates a polarity reversal, which is *very near* a phase shift.

 

[Vaughan Odendaa]

Thanks for the reply ! Yes, this is what I was suggesting. I am talking about AC signals. A musical waveform is AC. The speaker outputs AC which is dynamic and changes with time.

The other day someone told me that one speaker was connected out of phase when what really happened was the speaker's polarity was changed from positive to negative.

It isn't accurate to posit that the speaker is out of phase because that would imply that there would be a phase shift of the signal which, again, implies a time component.

--Regards,

 

[jonnythan]

I think I agree with you

 

[Vaughan Odendaa]

Cool.

--Regards,

 

[j_garcia]

While they are similar, polarity is only 2 conditions. Phase can be continuously varied and has to do with signal timing as you mentioned Vaughan. Phase is often used as a generic term for polarity, as in your above example.

 

[jonnythan]

Yes, but when you *reverse the speaker wires* the speakers are still in phase - that is, their timing is precisely the same - but the amplitude of the signal is exactly reversed. That is, their polarity is reversed but they are still in phase because there is no time differential between them.

I think I have Vaughan's point there.

 

[markw]

Yes, but when you *reverse the speaker wires* the speakers are still in phase - that is, their timing is precisely the same - but the amplitude of the signal is exactly reversed. That is, their polarity is reversed but they are still in phase because there is no time differential between them.

For all intents and purposes, what's been bolded is what's commonly referred to as "out of phase" because they are 180% out of phase with each other. The polarity of the wires connecting the amp to the speakers controls this.

 

[AbyssalLoris]

Sorry people, but from a mathematical/electronics perspective, most of these things are incorrect. At least using the standard conventions. I don't know anything about how drivers are responding to signals or the whole physics of that, but this much is true:

1. If you are talking about only a sinusoidal AC signal (as appears to be the case), reversing polarity is the same thing as being out of phase (that is 180 degree phase shift). While the observation that the signal with opposite polarity is just the same in-phase signal with -amplitude seems ok, it is really the same thing as saying that it is the the same +amplitude signal with 180 degrees phase shift. -A*sin(wt+d)=A*sin(wt+d-180degrees). So yeah, it is out of phase. What is causing this confusion is the convention that when you are doing a polar representation of a signal (amplitude, phase), the amplitude is defined as being non-negative. As in, you can represent any sine-wave with a non-negative amplitude and a phase.

2. Looking at time delay with periodic waveforms can be misleading. The two waveforms are in phase with each other at every multiple of the time period of the wave, but that does not mean there is no time delay between them.

3. Music is not just a sine wave. That would just be a single tone. It would be represented as a combination of such waves however, and the same principles hold true.

 

[jonnythan]

For all intents and purposes, what's been bolded is what's commonly referred to as "out of phase" because they are 180% out of phase with each other. The polarity of the wires connecting the amp to the speakers controls this.

I think that this is different than phase.

If one signal is out of phase with the other, it is literally following it in time. For a sine wave, this means that they have opposite amplitudes. For waves that are not perfect sine waves, they won't be exact opposites. They'll be close, but not quite.

Therefore, a reversal of the polarity of the speaker wires does *not* lead to an actual phase shift.

 

[markw]

whatever...

Therefore, a reversal of the polarity of the speaker wires does *not* lead to an actual phase shift.

We can both agree to disagree here. But I think we can both agree on the fact that it's audiable and degrades the system performance, right? Particularly if the same information is coming out of both speakers, right? Like, say, bass?

...but you're bordering on the "absolute phase" issue which is another matter entirely.

 

[jonnythan]

But phase, by definition, has a time component. If the signals are out of phase, there is a distance measured in time between peaks of the signals.

These signals are slightly out of phase:

So are these signals, but this signal isn't a sine wave:

http://64.233.169.104/search?q=cache:_k0PnY6Nf80J:www.musiciansfriend.com/document?doc_id=91427&src=3SOSWXXA+phase+vs+polarity&hl=en&ct=clnk&cd=1&gl=us&client=firefox-a

 

[no. 5]

For what it's worth, I like to use Polarity to describe the state of an electrical connection (i.e. + to +, or + to -), and Phase to describe the relationship of one signal to another.

It may not be correct, but I like it.

 

[markw]

You still didn't answer my questions.

I'll repeat them again.

But I think we can both agree on the fact that it's audiable and degrades the system performance, right? Particularly if the same information is coming out of both speakers, right? Like, say, bass?

So, what do you call this phenomena? I believe the rest of the world calls it "out of phase".

hint... remember, phase is relative.

 

[Vaughan Odendaa]

Excellent points everyone ! Just waiting for some EE's to pop in and set things straight.

--Regards,

 

[Vaughan Odendaa]

Just a few things. When someone "phase shifts a square wave by 180°", the outcome is equivalent no matter how you think of it, but if they "phase shift a square wave by 35°", you could think of it as shifting the entire square wave by a fraction of its period, or of shifting each individual frequency component by 35°, which would make it not a square anymore.

Does it make sense to talk about phase shifting arbitrary signals, or only sinusoids? As I'm sure everyone knows already, a phase shift in real life implies a time delay. So phase shifting an entire signal by 180° would require a delay of at least half a period of the lowest frequency present ?

Thoughts and suggestions ? Where are the EE's when you need them ?

--Regards,

 

[gmichael]

Are there any hairs left to split?

 

[OttoMatic]

Just a few things. When someone "phase shifts a square wave by 180°", the outcome is equivalent no matter how you think of it, but if they "phase shift a square wave by 35°", you could think of it as shifting the entire square wave by a fraction of its period, or of shifting each individual frequency component by 35°, which would make it not a square anymore.

Does it make sense to talk about phase shifting arbitrary signals, or only sinusoids? As I'm sure everyone knows already, a phase shift in real life implies a time delay. So phase shifting an entire signal by 180° would require a delay of at least half a period of the lowest frequency present ?

Thoughts and suggestions ? Where are the EE's when you need them ?

--Regards,

I have an MSEE degree, but I don't do a whole lot of actual EE work -- I'm a firmware guy for the most part anymore.

I guess I'm not clear on some of the questions.

Reversing the leads on your speakers will incur a 180 degree phase shift. I just don't see any way around it. Look at it on a scope, and they'll be 180 degrees out of phase.

Usually, we practice the theoretical side of it using sinusoids because it keeps the math simpler. However, two signals (i.e., your two speakers, because the leads are reversed) are out of phase, then that system will remain out of phase regardless of the signals run through them. And using Fourier's work, we can break down any signal, including music, into an infinite series of sinusoids. That means we can continue to do that "easy" math, by breaking down the signals into sine waves.

Anyway, that's the way I see it, just off the top of my head.

 

[Vaughan Odendaa]

Reversing the leads on your speakers will incur a 180 degree phase shift. I just don't see any way around it. Look at it on a scope, and they'll be 180 degrees out of phase.

But then what is the difference between polarity and phase ?

--Regards,