AVS Forum banner
1 - 20 of 20 Posts

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #1 ·
I did some searching and I didn't find the following addressed directly so I thought I would throw out some questions. It might also help others down the road.


I have a 7.2 setup and I have used many receivers in the past and have never received an Out of Phase error with any of them. Well I am just now installing another receiver and it reported both of my Fronts being Out of Phase. I double-checked my connections at the receiver, verified the wires were correct from the wall plate to the speakers and even took the wall plate apart to verify it was marked properly. Everything was as expected. The only other piece of info that might help is my Fronts have built-in powered subs and they do not have a Phase setting. I haven't yet tried running EQ with them unplugged but I think I will.


So I swapped wires on one of the Fronts and reran the EQ this time only the other Front was reported as being Out of Phase. I swapped its wires and according to EQ I was good to go. On purpose I'm trying to leave out the brand of receiver, etc. as I'm trying to understand being Out of Phase rather than if it's simply being reported incorrectly.


I can't reveal the speakers themselves to check how they react which I think is one test. So how would I actually know if they are in Phase or not? Could the built-in powered sub be throwing the EQ off? Although my sides also have powered subs and were reported fine.


Finally, I am using Audyssey MultiEQ... should I just take for granted it's incorrect (others have never complained) or make it happy by swapping the wires?


Thanks in advance.
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #3 ·
I can (sort of) understand the EQ software guessing incorrectly. However, I have a harder time understanding why switching the wires makes it believe the issue has been resolved.
 

·
Registered
Joined
·
1,807 Posts
Quote:
Originally Posted by Charles R
I can (sort of) understand the EQ software guessing incorrectly. However, I have a harder time understanding why switching the wires makes it believe the issue has been resolved.
Many loudspeakers that use higher order crossovers have some drivers intentionally wired in inverted polarity with respect to each other. For a recent example, see below:


http://www.stereophile.com/content/s...r-measurements


This Sony speaker has a tweeter wired in negative polarity, a midrange wired in negative polarity, and a woofer wired in positive polarity. Your FL/FR speakers may be wired similarly, and Audyssey MultEQ may detect that one of the drivers (probably the tweeter) is wired in inverted polarity compared to drivers in the other speakers in your system.


AJ
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #5 ·
Quote:
Originally Posted by WiWavelength
Your FL/FR speakers may be wired similarly, and Audyssey MultEQ may detect that one of the drivers (probably the tweeter) is wired in inverted polarity compared to drivers in the other speakers in your system.
Stupid question... then when I switched wires why didn't it find the other speakers (within the speaker) Out of Phase? Here's the spec on the speaker in question...

The VR970 features a powered, 100-watt, side-firing 10-inch (250mm) Deep Channel Design (DCD) bass unit and a passive 12-inch (305mm) radiator. The VR970 also includes two 4-1/2-inch (115mm) copolymer midranges and a 1-inch (25mm) anodized aluminum VR tweeter. Crossover 150, 2900Hz.


The sides (VR960) are very similar minus the 12-inch radiator. Crossover 150, 3000Hz. They all are so old I'm surprised they still sound rather sweet.
 

·
Registered
Joined
·
1,807 Posts
Quote:
Originally Posted by Charles R
Stupid question... then when I switched wires why didn't it find the other speakers (within the speaker) Out of Phase?
I am not entirely sure how Audyssey gauges polarity, but I presume it does so by observing compression (i.e. positive) or rarefaction (i.e. negative) during the first half wavelength of arrival, which, in a higher order crossover design, would be the output of the tweeter. So, when your VR970s are wired normally, the tweeter may be internally wired in negative polarity.


AJ
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #7 ·
That was the only condition I could see... it simply (only) tests the one speaker.
 

·
Registered
Joined
·
1,466 Posts
It's funny to see this question cause I've had this same issue. I can see how the internals of the speakers being wired could cause this. However, the weird thing for me is, I had two Denons report my speakers being out of phase, a 3310 and a 4810. When I hooked a Yamaha and a Pioneer up to the same speakers, they reported everything being fine. Odd to me.
 

·
Registered
Joined
·
22,557 Posts
None of the AVRS are checking electrical polarity, they are checking acoustic phasing. 2 different things.


Early room reflections are a common cause for "out of phase" errors during setup. So in that respect, getting the error might be a signal for you to look for causes. You may also see a small distance error on the same speaker if you actually measure it.


In any case, such phase errors can and do come and go depending on room conditions, and very small differences in mike placement will change it too.
 

·
Registered
Joined
·
1,807 Posts

Quote:
Originally Posted by Charles R /forum/post/20904252


That was the only condition I could see... it simply (only) tests the one speaker.

Not exactly. If my supposition is correct about how Audyssey determines polarity, then it does not explicitly test only "the one speaker" (i.e. tweeter). Audyssey has no control over the passive crossover network inside the loudspeaker; it can only measure the the output first arrival as it is emitted from the loudspeaker. And, with a higher order crossover, a characteristic (some would say flaw) of that output is that it is spread in time, such that the drivers react sequentially, not simultaneously.


For a recent counterexample of a loudspeaker with polarity and time aligned crossover and drivers, see below:


http://www.stereophile.com/content/t...r-measurements


AJ
 

·
Registered
Joined
·
1,807 Posts

Quote:
Originally Posted by MrEastSide /forum/post/20904281


It's funny to see this question cause I've had this same issue. I can see how the internals of the speakers being wired could cause this. However, the weird thing for me is, I had two Denons report my speakers being out of phase, a 3310 and a 4810. When I hooked a Yamaha and a Pioneer up to the same speakers, they reported everything being fine. Odd to me.

Why is that "funny" or "weird"? The two Denon AVRs utilize Audyssey MultEQ, while the Yamaha and Pioneer AVRs use YPAO and MCACC, respectively. In other words, they rely upon room correction processes with different methods and results.


AJ
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #12 ·

Quote:
Originally Posted by rdgrimes /forum/post/20904420


Early room reflections are a common cause for "out of phase" errors during setup. So in that respect, getting the error might be a signal for you to look for causes. You may also see a small distance error on the same speaker if you actually measure it.

My Fronts have been virtually located at the identical position for years and the room is somewhat treated (it's a dedicated screening room) with a few acoustical panels. However I do see your point that it can only measure what it hears. Distance wise I haven't verified this receiver but up to now every receiver has nailed it to within an inch or less. Mike wise the location has always been just as identical.
 

·
Registered
Joined
·
1,807 Posts

Quote:
Originally Posted by rdgrimes /forum/post/20904420


None of the AVRS are checking electrical polarity, they are checking acoustic phasing. 2 different things.

How so? The so called "out of phase" warning is there to tell the user, "Hey, check your wiring. One of your loudspeakers seems to output in inverted polarity compared to your other loudspeakers."


Now, I agree that "phase" and "polarity" are two different things. A change in phase indicates an advance or delay in the time domain, whereas a change in polarity indicates a mirror image inversion in the amplitude domain. But both changes can have similar effect, hence people commonly (erroneously) use "out of phase" to refer to either a 180° phase difference or an inverted polarity relationship.


AJ
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #14 ·

Quote:
Originally Posted by WiWavelength /forum/post/20904442


If my supposition is correct about how Audyssey determines polarity, then it does not explicitly test only "the one speaker" (i.e. tweeter). Audyssey has no control over the passive crossover network inside the loudspeaker; it can only measure the the output first arrival as it is emitted from the loudspeaker. And, with a higher order crossover, a characteristic (some would say flaw) of that output is that it is spread in time, such that the drivers react sequentially, not simultaneously.

So you are saying there is an (increased) delay because of the crossover? That makes sense but how does swapping wires correct it according to Audyssey? I can't see how it would change or fix the delay... wouldn't swapping wires change the timing of all of the speakers (within the speaker itself)? Perhaps the new timing falls into its acceptable level and if so should you leave them swapped or are there other considerations?


Again to a large degree I'm trying to keep Audyssey out of the discussion and rather focus on Out of Phase itself... although I know it's hard not to drag it into the arena.
 

·
Registered
Joined
·
12,368 Posts
The weakness of the typical Room EQ systems is that they attempt to correct the system's frequency response @ the arrival to the listener's (far-field) position..


So if there is a problem with a loudspeaker's polarity it may not detect this and/or indicate an inaccurate warning about phase. The Harman/JBL solution to this is to do (2) measurements, the 1st is near-field and the 2nd is far-field. By doing the near-field 1st the software can better detect if a loudspeaker is connected properly. This capability is found in the higher-end Harman/JBL components that utilize a TI DSP processor.


Just my $0.02...
 

·
Registered
Joined
·
1,807 Posts

Quote:
Originally Posted by Charles R /forum/post/20904618


So you are saying there is an (increased) delay because of the crossover? That makes sense but how does swapping wires correct it according to Audyssey? I can't see how it would change or fix the delay... wouldn't swapping wires change the timing of all of the speakers (within the speaker itself)?

No, reversing the polarity of the wiring does not change the group delay imposed by the higher order crossover. Rather, it just inverts the waveform.


Now, if the tweeters in your FL/FR are internally wired in negative acoustic polarity, while the tweeters in your other loudspeakers are internally wired in positive acoustic polarity (or vice versa), then Audyssey may raise the "out of phase" flag. Reversing the polarity of the wiring to your FL/FR may unite their tweeters in polarity with the other tweeters in your system. However, the mids or woofers may then be in inverted polarity with respect to the rest of the system. So, pick your poison.


All in all, this is another good reason to use five identical loudspeakers.


AJ
 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #17 ·

Quote:
Originally Posted by WiWavelength /forum/post/20904688


No, reversing the polarity of the wiring does not change the group delay imposed by the higher order crossover. Rather, it just inverts the waveform.

I wonder why swapping the wires corrects the issue. Presumedly the new waveform comes closer to what it expects...

Quote:
Now, if the tweeters in your FL/FR are internally wired in negative acoustic polarity, while the tweeters in your other loudspeakers are internally wired in positive acoustic polarity (or vice versa), then Audyssey may raise the "out of phase" flag. Reversing the polarity of the wiring to your FL/FR may unite their tweeters in polarity with the other tweeters in your system. However, the mids or woofers may then be in inverted polarity with respect to the rest of the system. So, pick your poison.

I'm going to wing it here... the delay of the Fronts is different enough from the other speakers to suggest that the fronts are Out Of Phase. When in actuality it might simply be they perform differently (by design). Is that possible? And perhaps the other EQ software was more lenient in allowing larger delay differences.
 

·
Registered
Joined
·
1,807 Posts

Quote:
Originally Posted by Charles R /forum/post/20904790


I wonder why swapping the wires corrects the issue. Presumedly the new waveform comes closer to what it expects...


I'm going to wing it here... the delay of the Fronts is different enough from the other speakers to suggest that the fronts are Out Of Phase. When in actuality it might simply be they perform differently (by design). Is that possible? And perhaps the other EQ software was more lenient in allowing larger delay differences.

Nope, you are missing the point. Group delay is an inherent characteristic of a higher order passive crossover, but delay is not the issue. Acoustic polarity seems to be the issue.


To illustrate, imagine the step response that I posted previously is that of your FL/FR VR970s. The tweeter and mid output in negative acoustic polarity, the woofer in positive acoustic polarity:


http://www.stereophile.com/content/s...r-measurements


Now, suppose that this is the step response of your SL/SR VR960s. The tweeter, mid, and woofer all output in positive acoustic polarity:


http://www.stereophile.com/content/b...r-measurements


To get the tweeter and mid in the VR970s to match the polarity of the tweeter and the mid in the VR960s, you would have to reverse the wiring on the VR970s. With inverted polarity, the step response of the VR970s would look like this:




Does that make more sense?


AJ


 

·
Registered
Joined
·
13,306 Posts
Discussion Starter · #19 ·
Quote:
Originally Posted by WiWavelength
Does that make more sense?
Yes it does. I was getting stuck on delay based on this...

And, with a higher order crossover, a characteristic (some would say flaw) of that output is that it is spread in time, such that the drivers react sequentially, not simultaneously.


Sequentially meaning the time would be increased. My (basic) misunderstanding is how (or what) frequencies are produced based on how the wires are connected. To summarize the polarity determines the direction the path takes but not the course itself?.. if that makes sense.
 

·
Registered
Joined
·
1,466 Posts

Quote:
Originally Posted by WiWavelength /forum/post/20904462


Why is that "funny" or "weird"? The two Denon AVRs utilize Audyssey MultEQ, while the Yamaha and Pioneer AVRs use YPAO and MCACC, respectively. In other words, they rely upon room correction processes with different methods and results.


AJ

Even though the programs are different, I just assumed the other two brands would report a problem as well. That said, I also had an Onkyo 809 hooked up to the same speakers and that didn't report a problem either.
 
1 - 20 of 20 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top