Speaker Connections

[der]

It begs the question - if we have no control of the source then how can it matter? Just something else to keep you up at night if you choose to be concerned?

derr

[iso]

It begs the question - if we have no control of the source then how can it matter? Just something else to keep you up at night if you choose to be concerned?

derr

As I said AES has made recommendation that recordings are made correctly phased. Also I would say that more than 50% of all previous recordings have been phased correctly... so even this is not perfect world, correctly phased signal chain makes sense to me.

JBL used to mark positive speaker terminal with black and negative terminal with red... this was perfectly OK when you knew this... but did not look too smart to me. Maybe I am a bit pedant... but I like to make things right way.

Best Regards

Kimmo

[der]

As I said AES has made recommendation that recordings are made correctly phased. Also I would say that more than 50% of all previous recordings have been phased correctly... so even this is not perfect world, correctly phased signal chain makes sense to me.

JBL used to mark positive speaker terminal with black and negative terminal with red... this was perfectly OK when you knew this... but did not look too smart to me. Maybe I am a bit pedant... but I like to make things right way.

Best Regards

Kimmo

It makes sense to me as well but I'm unsure I'd know if I was listening to a recording that inverted phase. Apparently, at least one of the speakers in my AR3as is wired out of phase. Why, I don't know. I would guess it was a conscious choice made by a designer or an engineer at the time.

Good postings Kimmo!

der

[iso]

Thank you for your kind words. Most peoples probably think that person tinkering with audio gear like me should be hospitalized immediately.

But... "Just something else to keep you up at night if you choose to be concerned?" is actually salt of the earth.

Best Regards

Kimmo

[RoyC]

It makes sense to me as well but I'm unsure I'd know if I was listening to a recording that inverted phase. Apparently, at least one of the speakers in my AR3as is wired out of phase. Why, I don't know. I would guess it was a conscious choice made by a designer or an engineer at the time.

der

Reversing polarity of a driver(s) is only one of a number of factors that affect phasing of a speaker system. In fact, depending on the crossover and driver properties, reversing the polarity of a driver compared to other drivers in the cabinet can actually bring all of the drivers more "in phase" with each other. There are few speaker systems, if any, where all drivers are absolutely in phase with each other, regardless of the polarity connections of each driver. (This is an argument used by advocates of single driver speaker systems.) I find this "absolute" phase discussion rather meaningless with respect to the real world...too many variables.

der,

There were no AR-3a drivers wired with reverse polarity from the factory. What are you referring to?

Roy

[Carlspeak]

Keep in mind all that a loudspeaker is an AC device and thus which binding post the pos. speaker wire is connected to isn't critical. What is very important (as others here have already noted) is that both speakers have the same connections.

With recordings available on the market today with no known polarity information, you are shooting in theh dark when wiring the pos. speaker wire to the #2 post vs the #1 post. Absolute polarity is elusive with commercial recordings. Perhaps we can only hope with connecting the pos. speaker wire to the Ar3a #2 post and the neg. wire to the #1 post that MOST recordings polarity will correspond to those connections.

[der]

Reversing polarity of a driver(s) is only one of a number of factors that affect phasing of a speaker system. In fact, depending on the crossover and driver properties, reversing the polarity of a driver compared to other drivers in the cabinet can actually bring all of the drivers more "in phase" with each other. There are few speaker systems, if any, where all drivers are absolutely in phase with each other, regardless of the polarity connections of each driver. (This is an argument used by advocates of single driver speaker systems.) I find this "absolute" phase discussion rather meaningless with respect to the real world...too many variables.

der,

There were no AR-3a drivers wired with reverse polarity from the factory. What are you referring to?

Roy

I have no actual knowledge of AR speakers being wired out of phase. I have heard it stated that they were. I should have been a little clearer with that statement. I'll be the first to admit that I'm no where near an engineer.

der

[JKent]

There were no AR-3a drivers wired with reverse polarity from the factory. What are you referring to?

Roy

Roy,

I have encountered at least one AR woofer (maybe a 2ax) that was incorrectly marked with a red plus on the negative terminal. I thought you said you had seen that as well.

It may have been mentioned somewhere in these pages and I'm guessing that's what der was referring to.

Or... maybe the practice of wiring the tweeter out of phase.

-Kent

[DavidR]

......................... maybe the practice of wiring the tweeter out of phase.

-Kent

I recently upgraded some Yamaha 3-way mini speakers. The tweeters were dead and the mids might have well have been. Both the mid and tweets were wired out of phase. I bought mylar dome tweeters and a full range driver for the mids. I made a better xover than the single NPE cap from the original design. I left the tweeter out of phase (originally) but wired the mid in phase with the woofer. I got curious one day and switched the phase of the tweeters. I could hear no difference.

I'm not an electrical engineer so maybe I'm over simplifying it but a voice coil is just a winding of wire. Current in one end out the other. Does it make a difference which way the current goes around if both speakers are the same?

(@JKent: I used some of the 10uF PP and 2uF mylar surplus caps from Madisound. Good deal!)

[RoyC]

Roy,

I have encountered at least one AR woofer (maybe a 2ax) that was incorrectly marked with a red plus on the negative terminal. I thought you said you had seen that as well.

It may have been mentioned somewhere in these pages and I'm guessing that's what der was referring to.

Or... maybe the practice of wiring the tweeter out of phase.

-Kent

The primary issue is AR installed voice coils of woofers (and some early mids) inconsistently, so the + and - may be on either side. It is always prudent to use a battery to check the polarity of an AR woofer before re-installing, especially if the + label is lost.

[owlsplace]

The primary issue is AR installed voice coils of woofers (and some early mids) inconsistently, so the + and - may be on either side. It is always prudent to use a battery to check the polarity of an AR woofer before re-installing, especially if the + label is lost.

Hmm, how to check the mids? I suppose I could hook them up to my Denon and let it tell me if it is out of phase.

Roger

[der]

I'll be pulling the woofers in my 3a's soon for crossover work. I guess I can check the woofers polarity then.

der

[RoyC]

AR knew which side was which, (usually) marked them accordingly, and had the drivers connected correctly. It is mostly when pulling drivers or using replacement parts that the issue can become a problem.

Some early front-wired AR-3a and AR-5 midrange leads were crossed to correct for the voice coil reversal. The + side usually has a red mark on those. These are not common, nor is there an easy way to check it.

Btw, reversing polarity connections of a driver in a speaker does not necessarily "reverse phase" or put it "out of phase" with other drivers any more than it was before reversing the connections! It changes or "shifts" phase within a speaker system. "Polarity" and "phase" are two different things.

Polarity, mechanical/electrical properties of the drivers, the crossover, and cabinet design (and technically the listening space) must all be taken into account to discuss phasing. Reversing polarity of a tweeter and or midrange will alter frequency response, and, subjectively, can be a matter of sonic preference for audio enthusiasts within a given listening space. It is also not likely to have any correlation with achieving "absolute phase".

Roy

[der]

I replaced my original 3a midranges with new factory sourced units many years ago and very carefully wired the replacements (front wired) to match the markings on the drivers and the orientation of the wiring. My woofers have been out once before and were factory marked and were installed and wired as they were when removed. From what you have said RoyC I need not be concerned about phase. Forgive me for interchanging polarity and phase. As I stated previously I'm not an electrical engineer. I'm just a hobbyist that desires the best his vintage loudspeakers can offer.

der

[RoyC]

I replaced my original 3a midranges with new factory sourced units many years ago and very carefully wired the replacements (front wired) to match the markings on the drivers and the orientation of the wiring. My woofers have been out once before and were factory marked and were installed and wired as they were when removed. From what you have said RoyC I need not be concerned about phase. Forgive me for interchanging polarity and phase. As I stated previously I'm not an electrical engineer. I'm just a hobbyist that desires the best his vintage loudspeakers can offer.

der

Hi der,

I'm sure you have nothing to worry about in relation to "phasing" of your 3a's or its drivers.

My comment was not pointed at you. It was meant as a general comment, and point of clarification in a thread which has become more complicated than it needed to be.

Roy

[tysontom]

I replaced my original 3a midranges with new factory sourced units many years ago and very carefully wired the replacements (front wired) to match the markings on the drivers and the orientation of the wiring. My woofers have been out once before and were factory marked and were installed and wired as they were when removed. From what you have said RoyC I need not be concerned about phase. Forgive me for interchanging polarity and phase. As I stated previously I'm not an electrical engineer. I'm just a hobbyist that desires the best his vintage loudspeakers can offer.

der

der,

With regard to the polarity of wiring on AR's 3-way speakers, there was indeed a change made to reverse the polarity of the midrange with respect to the woofer, and early AR speakers had this wire crossed at a point between the midrange driver and tweeter-terminal board (3-wire terminal) on the front of the speaker. This put the midrange 180˚ out of phase with the woofer (this would occur only at overlapping frequencies around the crossover point). When Roy Allison began work at AR in 1959, he began to realize that there was a small—but measurable and audible—peak in the woofer-to-midrange crossover region (this was in the AR-3 and AR-2a). This was noted in listening rooms at AR and in the anechoic chamber. RA devised a DPDT toggle switch with long wires to allow him to switch the midrange "in" and "out" of phase with the woofer to do A-B comparisons back in the reverberant listening field and measurements. It was found that the response was slightly smoother with the speaker wired out of phase between the woofer and midrange (the tweeter remained wired with the same polarity as the midrange), and the reversal took place. At first Edgar Villchur was skeptical, but after verification, he agreed that the change was an improvement, but he did not agree to a change in the tweeter's polarity with respect to the midrange, so this remained unchanged.

AR's 3-way speakers thus became wired out of phase. Subsequently, instead of reversing the wiring on top of the baffle board, AR decided to reverse the polarity of either the magnet (N-S orientation) assembly or the voice coil leads, Roy Allison wasn't sure which, when he told me this years ago. I believe this reversal went on well into the AR-3a years as well, but beyond that I don't know. This reversal is very insignificant with 12 dB LCR crossovers, but the 6 dB crossover were somewhat affected by it due to the fairly large overlap of the woofer and midrange.

Regarding the "absolute phase" of speaker-amplifier-source inputs, the only thing that is truly important is to maintain speaker "polarity" at the speaker input terminals, as we've mentioned earlier. As for absolute phase, it's impossible to do throughout the entire sound-reproduction chain, and doesn't matter. If you look at the instruments in an orchestra, from violins to horns to bass drums, etc., each reproduces musical waveforms with wildly varying wavelengths and phases, and these reach the listener (or the recording microphone) in a completely jumbled fashion of phase—which is completely normal. There is no correct phase relationship in the original instruments producing music, and there is no rational purpose in attempting to recreate a phase relationship in the playback chain.

—Tom Tyson

[der]

Two fascinating posts from Roy and Tom. I'm not sure how exactly we got here but what great information. And, from one that knew Roy Allison! The leads on my midranges are not crossed and the positive terminal is clearly marked. (I just looked again) I bought my 3a's in the Spring of 1969. The midranges were replaced some 10 years of so later although I can't recall the exact year. I'm getting ready to recap my crossovers and am hoping I get an audible improvement. My tweeters still have output but my ears are unlikely to detect much of it!

Thanks to both of you gentlemen for your very enlightening responses.

der

[ra.ra]

der, I totally agree with you that Tom's story of how this sort of engineering revision actually occurred is very interesting.

I understand that this topic is primarily directed toward your 1969 AR-3a's, but it was also interesting to see way back in post 2 of this thread where David attached the schematic for the AR-3a 'Improved', which shows the tweeter polarity reversed from the other two drivers. Also, in the smaller and simpler two-way AR-7 and some versions of the AR-6, the polarity of the drivers is inconsistent with regards to the speaker terminals 1 and 2. While I've never been able to understand these obvious wiring differences between speaker models, with Tom's report of this history it is good to know that these decisions were all very deliberate outcomes resulting from rigorous laboratory measurements and extensive test listening.

[owlsplace]

Yeah, thanks for that explanation, Tom. Isn't phase differences how the human ear judges spatial relationships?

Roger

[RoyC]

Polarity considerations have always been part of crossover design. The AR-11/10pi (near the same era as the AR-3a Improved) had reversed tweeter leads as well.

Just to clarify....Tom's post is in relation to the design of the AR-3. If the mid's voice coil reversal went into the 3a years, it was compensated for in the 3a...though I have not seen evidence that it went "well into" the 3a years. I believe only some very early AR-3a mids were reversed, and marked as such. I have verified polarity on many AR-3a mids going back to the very early serial numbers. It is not common to find the polarity of the voice coil leads to be reversed, and when they are, the front-wiring had been adjusted for like-polarity.

This does not explain the reversal found in some AR-5 mids. I've actually worked on pairs of 3a's and 5's with very close serial numbers where one mid was cross-wired and the other was not. Once again, removing the metal grills of the cross-wired specimens and verifying polarity showed that the voice coil leads were indeed reversed compared to its mate, and corrected.

Unfortunately AR-3 mids are so inconsistent these days, it probably doesn't matter how they are connected.

Roy