tony666
Well-known member
How come this is so mixed up in the audio industry? I find it very confusing always reading about 180° phase shifts etc. when they actually mean polarity reversed?
polarity reversal instead of out of phase
- Thread starter tony666
- Start date
Help Support GroupDIY Audio Forum:
..It's all the same. It's the old game of coming up with new and interesting technical terms..
nacho459
Well-known member
I have heard several people sternly say, "It's a polarity switch, not a phase switch!" I guess because phase is more or less a delay issue, and polarity is a mechanical issue.
MikoKensington
Well-known member
Yeah but the end result of a mechanical polarity reversal is a 180 degree shift in phase. :?
MikoKensington
Well-known member
That makes sense. I guess. To use the term 180 degrees out of phase you'd have to be hearing nothing at all. But no one here said that.
The terminology we use isn't always used properly. Phase shift is one waveform's lead or lag time in relation to another. When we say 180 phase shift we are thinking the waveform's relation with itself. Not in relation to the other waveform.
I think a lot of it comes from our use of 180's in day-to-day conversation. "I was ready to take this little hussie home form the bar, and she pulled a complete 180 on me."
I think the key is the term inversion. Polarity and Phase. At that point the two are the same. Or are they?
Here:
http://www.atis.org/tg2k/_phase_difference.html
http://www.atis.org/tg2k/_phase_inversion.html
[quote author="MikoKensington"]I think the key is the term inversion. Polarity and Phase. At that point the two are the same. Or are they?[/quote]
No, not for our purposes.
For the strict mathematical analysis of infinite waveforms, phase inversion (as described on the link you posted, Miko - I've never seen the term used in any formal text before) and polarity inversion of a periodical symmetrical waveform will give the same result.
In audio, however, even periodic, symmetrical signals (sine waves, for example) are discrete in time, i.e., they have a beginning and an end; so if you phase shift an audio signal 180 degrees, it will be time delayed by 1/f seconds. Try to use it in a balanced system and the very first and the very last cycles will still come through loud and clear.
Also, remember that a phase shift of 180 degrees at one frequency, can be 3 degrees, or 3000 degrees at other frequencies. So the very concept of a 180 degree phase shift of a harmonically complex waveform doesn't make sense at all!
No, not for our purposes.
For the strict mathematical analysis of infinite waveforms, phase inversion (as described on the link you posted, Miko - I've never seen the term used in any formal text before) and polarity inversion of a periodical symmetrical waveform will give the same result.
In audio, however, even periodic, symmetrical signals (sine waves, for example) are discrete in time, i.e., they have a beginning and an end; so if you phase shift an audio signal 180 degrees, it will be time delayed by 1/f seconds. Try to use it in a balanced system and the very first and the very last cycles will still come through loud and clear.
Also, remember that a phase shift of 180 degrees at one frequency, can be 3 degrees, or 3000 degrees at other frequencies. So the very concept of a 180 degree phase shift of a harmonically complex waveform doesn't make sense at all!
clintrubber
Well-known member
Sorry for nitpicking, you mean 1/(2*f), right ?
It'd be nice if people were talking about polarity-switches, not phase-switches. I'd vote for 'polarity-inversion'.
But to people who just want to get the job done this will all be a lot of :shock: :roll: :?: :!: :sad: :shock: I'm afraid.
electronaut
Well-known member
I think with the new "phase correction" tools that are starting to appear on the market (like the Little Labs "phase alignment" tool) and all this talk of adjusting phase by a few degrees here or there, it probably makes more sense for the regular old phase-reversal switch on a mic amp to be called "polarity". It's a matter of semantics, yes, but I think it lessens the likelihood of confusion.
[quote author="electronaut"]but I think it lessens the likelihood of confusion.[/quote]
On second thought... that's probably wishful thinking.
[quote author="electronaut"]but I think it lessens the likelihood of confusion.[/quote]
On second thought... that's probably wishful thinking.
MikoKensington
Well-known member
Why hasn't the AES done something about this? Polarity it is.
[quote author="clintrubber"]Sorry for nitpicking, you mean 1/(2*f), right ?[/quote]
Absolutely.
[quote author="electronaut"]It's a matter of semantics[/quote]
No, it's not. They are completely different things and we are right to bitch about it :evil:.
Absolutely.
[quote author="electronaut"]It's a matter of semantics[/quote]
No, it's not. They are completely different things and we are right to bitch about it :evil:.
