|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
98.203.125.33
In the current issue of enjoythemusic.com an article about polarity dating to 1989 is republished. I think you will all get something out of it.More About Absolute Polarity
Article By Dan Shanefield
From BAS Volume 17 Number 3, November 1989Below see my response.
Where to begin?
Scientists in every field often have contrary opinions, yet Dan Shanifield offers only one source, JAES! A shill perhaps?
I offer a more even-handed view. John Roberts in his 1983 article Exposing Audio Mythology says "Just because the effect of Absolute Polarity is not particularly audible today, doesn't mean that it won't be in ten to twenty years, as speakers and other equipment improve."
Results are all I care about, and I can tell you for a fact that polarity is not just audible, is it critical to hearing the recording properly, which leads to a greater ability to suspend dibelief.
Apparently, there are ways to employ a polarity circuit that validates my experience, and some that would not.
The reason why many people fail to hear polarity is the preponderance of phase incoherent loudspeakers, or what Dr Richard Greiner has called the crossover catastrophe.
FYI: Going back some 15 years, I have easily heard the effects of polarity on Accoustat 2+2s and custom made 4+4s. Genesis Vs, Reference 3A De Capos (one resistor between tweeter and woofer!), Nearfied Acoustics Pipedreams (point source, almost zero distortion), and on my current speakers- Magenpan 3.5Rs; of these all are phase coherent designs. I replaced the OEM passive XOs with a Behringer CX2310 active XO.
Not only does the Behringer (and I would venture any decent active XO)allow the speakers to breathe, it features independent polarity of the L and R inputs, AND individual polarity control for bass, and hi-frequencies, and don't forget the subwoofer, it has polarity control of it, too!
Point 1/3 in Dan Ss article "Not strongly audible to most people". And what planet did he pull that scientific data from- Uranus?
Hmmph! Clearly polarity is important to the Pro industry. Although, like the author, most of them can't hear it, and as a result polarity on recordings is all over the place, especially multimiked recordings, and recordings made from different studios, and times IE "Best of" compilations.
Just because polarity is not strongly audible to Dan S, the editor, or to "most people", doesn't mean it is not a critical component to hearing music in the proper polarity. I can readily hear it on TV broadcasts
I have had many audiophiles hear my systems over the years. They came in polarity virgins, however, it didn't take long to train their ears to appreciate the benefits of polarity, especially when the entire recording polarity is either 0 or 180.
Edits: 12/18/08Follow Ups:
I do believe many people can hear polarity issues, but many do not recognize the sound of polarity issues. They often attribute what they hear to other factors: speakers, amps, whatever. I had been attending CES since 1982 and only stopped a couple of years ago. In the last few years I would ask, usually when the rooms were empty, if I could switch polarity on the demo speakers in many rooms. Most exhibitors were rather stunned to hear the difference.
I remember one manufacturer who looked at me politely but acquiesced to my request since he had a Wadia player with the polarity inversion switch. He first requested that he stand behind my chair in order to hear what I was hearing. In flipping the switch he was stunned. He told me ( I was already a dealer for his other products) that while he understood the theoretical ramifications of polarity inversion, he has never really bothered to attempt to identify the hearing with the theory.
I did that to a Chinese made speaker exhibition too. It was called the Morkai, IIRC, and as I entered the room I could hear the exhibitors anxiously whispering that something was wrong and the sound was not what it should be; A complex design it had a tri wired crossover, IIRC. I told the exhibitor I believed his woofer was out of phase and we swapped wires on the woofer module and he was instantly gratified. I was rather surprised he mentioned me in his CES report.
The fact is many people can recognize sound when something is wrong. They may not be able to finger the culprit immediately and most have learned to ignore the effects of polarity. When so many designs consist of drivers in mixed polarity the issue becomes even more confused. I believe every audiophile should live with a speaker like the old Spicas or a full range electrostatic or a pair of Vandersteens for at least a year so their ear can recognize polarity issues.
Stu
No wonder I have heard people who listen to music 180 out of phase
are known to use the term "that sucks" when they dont care for the music
content
:)
The only time 'polarity' comes into it during recording is when more then one mic is used to capture an instrument. Thats usually the drum kit where each drum is close mic'ed and an overhead pair is used to capture the whole kit, the cymbals and some of the room sound.
In this case one or more mics need to be inverted to avoid large cancellations.
Often the snare drum is mic'ed top and bottom in which case one mic has got to be inverted.
It has nothing to do with overall 'acoustic polarity' which once something has been recorded is impossible to reconstitute. Whos to say if a drum was mic'ed from above, in which case the soundwave starts with a rarefaction, or below in which case the wave begins with a compaction? If you stand a few yards away from a drum kit there is equally no way of telling if you hear the compaction or rarefaction first.
The engineer knows in the case of close mic'ing a drum kit but if you mic a clarinet nobody knows and nobody cares, the mic just goes where it sounds best.
The reason xovers have polarity switches is because pa speakers are hardly ever time aligned and the switch is again used to avoid the worst cancellations.
Some xovers feature 360deg adjustement of phase for the same reason, mine do anyway.
Drivers should be run the way they were designed. In case of my Tannoys they were designed to move backwards when faced with a positive going waveform, just like JBLpros. They sound better that way. The Volt drivers in my woofers were designed for either and do not sound in any way different if I flip the polarity.
Drivers should be run the way they were designed. In case of my Tannoys they were designed to move backwards when faced with a positive going waveform, just like JBLpros. They sound better that way. The Volt drivers in my woofers were designed for either and do not sound in any way different if I flip the polarity.
First off they are not designed to move backwards
The terminals are simply labled red on positive and blk on negative
The OLD JBL line speakers were labled out of phase
The new JBl stuff has been labled properly for positive and negative
for sometime now
Many people had no clue of the reverse phase on the old JBL stuff
and ran systems with out of phase JBL speakers mixed with other
pro loudspeaker gear
Which basically sucked
At least JBL has it right on the phase issue now
And that's all it is! Nothing but a means to describe which way a driver will move when a + battery post is touched to the red terminal.
Good thing though that a general agreement has been reached that "+" means an outwards-going excursion (nvver-minding the usual out-of-phase crossover wiring).
Poles Apart
Regarding JBL woofers I could almost agree, Tannoys (of a certain vintage) are somewhat different as woofer and tweeter coil share the same magnet.
Again if you have issues with Tannoys 'being designed' to move inwards take that up with their engineers. Could prove difficult though as I expect most of them to have retired a long time ago.
Until I have their word on it I stick with what they wrote in the technical manual (the one for service engineers)… and my ears.
I see you have issues with the Tannoy technical manual. You might want to take that up with the Tannoy design engineers.
If you care to research the AES recommendations for preservation of polarity, they recommend a forward motion of the driver cones as being the preferred way to mark the positive terminals. Bearing in mind that these standards were promulgated fairly late in audio history, many traditional older companies have elected to retain their older standards. I suspect Tannoy would be one of them. The fact that JBL has altered their designs reflects an acceptance of that AES recommendation (also part of the IEC recommendations, IIRC).
People like AKG, last I checked their website, openly admit that their headphones are reversed in polarity but say that is part and parcel of the AKG sound. Thus some companies retain their non standard procedures in order to satisfy their particular niche market.
The fact is, much like the dual standard for XLR pin outs, again covered by AES and IEC standards, there is absolutely no law forcing manufacturers to adhere to the standards recommended. Hence there is much confusion in the world of audio. Some can hear the differences, others can not. But with the various "standards" comes tremendous confusion, and many listeners are simply overwhelmed by trying to understand what problem is causing what anomaly in sound.
In addition, polarity inversions are used to "doctor" the sound. I have long stated that Phil Spector's "wall of Sound" was his use of inverting the polarity of the background instrumentals while keeping the vocals in correct polarity. The inverted polarity of the background instruments gives a large soundstage because it smears the sound of those players and seemingly enlarges the soundstage.
You can also hear this in the movie "The Commitments". In the newly released directors cut, the director explains how they recorded every one in the band playing their own instruments but reversed the polarity in respect to the lead singer in order that his voice project far more and sound significantly better in order to emphasize the pint in the story line.
In the Michael Crawford album, I notice that Crawford's voice is reversed in polarity to the other people in the duets he sings. I find that his voice is not the greatest, at least not in comparison to Barbara Bonney, a noted operatic singer, so to underscore that this is his album, there are polarity inversions.
In addition, the use of vocal synthesizers creates issues with polarities. The Aphex Aural Exciter, IIRC, inverts polarity and adds some EQ. Many singers use it inn order to get that husky voice which seems to be popular these days. Streisand, Ronstadt, Neil Diamond all use it. Compare their very early recordings to their latest ones and the quality of their voice has significantly changed. Then try inverting the polarity of their recordings and you will immediately hear the sound of their early recordings.
Polarity has become a tool to further emphasize certain aspects of music, when sheer volume differences are not enough. In a world where electronics are pretty much mandatory, that may not be so much of an issue. It is definitely an issue when you come to acoustical instruments and voices performing in real time and space without electronics, however.
Stu
Hi,
and on my current speakers- Magenpan 3.5Rs; of these all are phase coherent designs
Dude you sure about that? Maybe I am messing up phase with polarity like Hare mentions, but I though anywhere you have a 12db slope you typically change the polarity of one driver.
ANyhow, I certainly can't hear polarity on my maggies and I don't think that it I am alone on that :)
Scroll down and it shows the 3.5s crossovers
I don't think that anyone will be able to hear polarity issues. Take two 1k sine waves, invert polarity on both channels, they will be the same I betcha. Phasing is different. Revers polarity on one channel. If the wave cancels, you have a properly set up system. If not, you have improper speaker placement and/or system issues.
If you have sine waves they will sound the same in either polarity, as the waveform of an undistorted sine wave is completely symmetrical. If you want to try detecting polarity with constant tones, you can mix in 2nd harmonic distortion at various phases. This can be done easily using audio editing software which has waveform generating features. You can create a mixture of fundamental and second overtone at any amplitude and phase you wish. Detection results will be best if tones are well below 1 kHz as that is pushing the range at which the auditory nerve fires synchronously with the acoustic waveform. Above a cutoff only amplitude information makes it to the brain, according to the research that I have read. I have had better results detecting absolute polarity when listening to music rather than test tones.
It is equally convenient for me to adjust playback polarity as volume. I use both controls about the same amount when adjusting playback so that a given recording sounds at its best. However, most of the recordings that I listen to are minimally miked Classical and Jazz recordings, which tend to show up polarity much more than multiple miked complex mixes.
After trying both polarities with many different Classical and Jazz recordings, if you still don't hear differences it may be that your speakers aren't phase coherent enough. Try playing back through a single driver speaker setup. It need not be fancy. A cheap table radio should do just fine.
I have found it best not to listen to the tone of the instruments, just the general ambiance. If you can achieve a Zen mind it will be easier to detect polarity differences. Over time you will gain experience and be better able to hear polarity right away and predict whether the sound is likely to improve if you reverse the polarity. However, at least in my case, this ability varies from day to day.
Enjoy. It is a fun game to play and a cheap tweak.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Right.
What I was getting at is that if the speakers by design have the tweeters connected with reverse polarity (phase???) relative to the mids or the bass, then that sinewave will be at different polarity at the different drivers making it impossible to hear.
Now lets say I have a single full range driver, are you saying that I cant hear reverse polarity?
You may be hearing an effect of the filter. But if you were to do a double blind test, there would be no beneficial obvious difference.
but not true with actual music which contains many overtones. The relationship of the fundamental tone to the overtones will be changed, and one obvious aspect easily heard is excessive sibilance.
Stu
Uhhhh. Sine waves are harder for gear to reproduce than music. Harder for digital and analog. Do you know what a sine wave consists of?
Music with its complex overtones and harmonics are much harder to reproduce accurately. IM starts to rear its ugly head along with TIM distortion in the case of amps with slow transient response. Typically you do not get those issues with a pure sine wave, particularly in the 1KHz range.
But perhaps I am not understanding the scope of your question. Perhaps you ought to elaborate a bit more.
Stu
There is an issue with this that you audiophiles do not address: Positive polarity. You guys are missing the point with absolute polarity. 99.9 percent of the speakers made today move out first, using negative polarity. That is opposite of how sound hits the microphone diaghragm. So, by flipping all channels to acheive "absolute polarity", you are getting closer to positive polarity, which would make for absolute polarity IMO.
JBL was the big proponent of positive polarity, and Altec Lansing was into it. If you go to JBL's website, you can find the white papers on this topic. JBLs drivers had their leads reveresed, compared to the rest of the world, to facilitate positive polarity. They did this until the late 80s. Then they gave in and switched to be like everyone else.There is also a misconception that phase and polarity are the same. They are not. The two terms cannot be interchanged. You can have two signals with the same polarity, out of phase. You can have to signals, one with polarity inverted, in phase.
The fact that a Behringer or any other processor allows for polarity flipping does not mean that it facilitates or allows for absolute phase. In fact, the Behringer is utter crap, because it's processes cause latency. This latency is not compensated for. all you high end baffoons that think this box is stellar have no ears and/or poor acoustics, because any time you use these processes, the latency is adding delay, which is smearing your signal, because this box does not allow for perfect time alignment. It cannot. It is a dumb unit. There is no delay compensation.
The dbx driverack is the same way. Only the dsp on the INPUTS has delay compensation. And it is not perfect, It is course. There is no delay compensation for dsp on the outputs. So with these units, you really need SIA SMAART, TEF or some other time measurement software/device that allows you to set up your system, and then measure time arrival for each band, and then add delay to the ones with lesser processing so that everything is in phase from there to amp to speaker.
Of course, the room and physical placement will have some effect on phasing. So that is where an all pass filter will come in handy. You might need 225Hz rotated 65 degrees. This is what that is for.
Edits: 12/16/08
If the microphone diaphragm moves backwards in the recording session that means your ear drum will also move backwards. If a speaker pulses outward given a signal, that re creates the the situation heard in real life and mimics the microphone diaphragm motion.
In examining JBL speakers, the pro series (L series) all have their drivers moving inboard when using a battery test. AKG headphones have a similar motion. In most other drivers as you point out the movement is primarily outwards when positive current is applied to the positive terminal.
I believe the original poster is very much correct in stating that a great majority of speaker systems today employ drivers wired in mixed polarity. In addition many are not time aligned which further clouds the issue and the perception of it.
Many prominent designs employ inverted drivers. I haven't kept up with all models in all designs but allow me state a few. The Proacs all have the tweeters inverted in relation to the woofer, even up to the Future models where at least they place the midrange and tweeter in correct polarity relative to each other. The early Alons and Hales Transcendent series all have their midrange drivers inverted to their woofers and tweeters.
I believe a lot of this is deliberately done because of the initial review in The Absolute Sound of the Proac Mini Tablettes. The Proac has the woofer inverted in polarity to the tweeter, but the reviewing editor (HP) proclaimed the speaker a wonder in that it had image specificity but a huge soundstage which belied its diminutive size. As you know, inverted polarity smears the image presenting a seemingly huge soundstage. Having the tweeter in correct polarity then gives a lot of image specificity although there is a distinct discontinuity evident that HP did not comment on nor seemed to have be sensitive to. This created a whole following on using inverted driver designs which now seems to more the norm than the exception.
However, when speaker designs incorporate both time and correct polarity, differences of polarity in sound are very audible. With practice, it is easily heard even with problematic designs
Stu
I think you will find that almost all speakers which use a passive crossover with 12dB slope will invert the tweeter. This is because a passive 12dB xover introduces a 180deg phase shift between woofer and tweeter. Inverting the tweeter reduces the resulting problems but does not eliminate them. This can all be easily avoided by using an active xover.
One of the many reasons active xover are inherently superior to passive ones.
PS: if one hears a marked improvement using Behringer xovers it only shows that even the worst active xover is better then a well-designed and executed passive one.
with even ordered slopes invert polarity, You can correct this by simply inverting the leads to the drivers which simply follow the signal and is passive in that regards.
The issue is that many designers invert polarity in order that the adjacent driver(s) produce a steeper crossover slope due to the cancellation effects. This can give a more even amplitude response but totally screws up the timing information, which ultimately gives a better cues for soundstaging, individual tonality and such. Properly selecting crossover points to minimize the overlap can work equally well, in my expereince.
IMHO there is no real advantage for an active crossover except in the case where you really need more power. After all, the crossover circuit in the active unit still follows the designs of a passive one, except in a digital crossover design. Being an analog guy, I find that adding such a design to my system destroys certain aspects of the sound I am achieving and seeking.
Of course, YMMV.
Stu
Stu
IMO the real advantage of an active xover is that they do not smear transients like passives do.
And the lack of phase issues and distortion. They do add a bit of noise though.
Analogue active xovers (like mine) do follow the design of passives with one major difference: Inductors are replaced by opamps. An infinitely superior solution, your transients will thank you and the bass response will be much more defined (due to a) the lack of an inductor in the signal path and b) the full benefit of the associated amps damping).
I have had the chance to directly compare a number of active speakers with their otherwise identical passive brethren and the result has always been the same: Actives are more accurate and passives 'mushier'. If 'mushiness' is what you seek then actives are not for you; in all other cases actives are clearly superior.
Again, I do admit my experience is limited with active rossovers. One with which I have extensive experience with has been the Marchand which is remote controlled and adjustable which makes for inteesting comparisons of speaker topologies. It also has a choice of different order crossover networks, IIRC.
Still, if a speaker is properly time aligned and phase aligned from inception I find that the added complexity of the system takes away from the overall presentation. Curiously I get tighter and more coherent sound in my experience from top through bottom in the avoidance of an active system. I'm not saying that it can't be done, however, just that it makes things more difficult.
As I said YMMV.
I have also listened to several active loud speakers and have found that the vast majority of those I have listened to have serious issues with time and phase. To be quite honest I do not recall any that were as well aligned as some passive speakers I have heard. Of course part of the issue is the amplification modules which are used with many speakers. Many are not up to high end, high resolution standards, a problem I find with many subwoofers these days. Again, YMMV.
Stu
I on the other hand am not aware of any higher quality main monitors which are not phase compensated. This may not be immediately visible but its easy to do electronically.
The problem is that they are all pug-ugly as they are designed to be soffit mounted but can also easily be used free-standing if one can live with the looks.
My own 4way actives are also completely phase compensated and pretty much retain the point-source character of the Tannoy DC they are based on. There is a slight timing error of <3microsec though. Not actually audible.
For high quality active monitors you might want to check out ATC 150, 200 and 300 from their pro series or Quested HM412 and HM415. They might seem expensive but I doubt you could buy the components for less.
Sadly there are no more high-quality analogue active xovers being made today (I have no experience with Marchand products) I am aware of so I will keep modifying my old BSS 360s.
It is actually a lot easier to design active speakers from scratch then passives.
It just costs a lot more to implement…
I am far from an expert, but this looks like an analogue active xover.
"99.9 percent of the speakers made today move out first, using negative polarity." Huh? HUH???
The writer means "drivers", not "speakers", but let it pass. Or, not -- the huge majority of designs today incorporate crossovers that require the polarity of the bass driver, say, to be opposite that of the mid. So which one may be discovered to "move out first"?
Besides which, the writer displays a manifest confusion here between electrical polarity and acoustic polarity. Most designers will tell you they have wired their drivers so that when the + on a battery is connected to the red terminal, the driver will move forwards. This is an unwritten convention of the trade, but one that JBL alone (I believe)ignored, originally, springing as they did for the opposite. But that's electrical polarity.
Acoustic polarity, the subject of both Dan Shanefield's letter and HiOnFi's post, is altogether different and separate. Acoustic polarity recognizes the undisputable fact that most acoustic instruments have a polarity signature. They produce, most commonly, compression transients -- although a few do produce rarefaction. When these signatures are reversed by an audio system, the instruments sound wrong. Specifically, the compression transients sound muffled . That is why acoustic polarity really matters and electrical polarity remains tertiary at best.
Throughout recorded history, owing to a lack of specific directions or standards, our delivery media have incorporated a mixture of the two polarities. Sides can be different from each other, cuts can even sometimes alternate, and worst of all, a track mix can involve both recorded polarities, making a mish-mash of the music. This is just another reason why simply-miked and unprocessed (un-toyed-up) recordings can sound so great, so natural.
But, a further caveat: Whatever the polarity aspect of the recording or the reproducing system may be, the requirement is for a compression-type instrument to produce a compression transient from the loudspeaker. And compression transients are produced only by out-going driver motions. There can be no compromise.
When the poster says, "By flipping all channels to acheive 'absolute polarity', you are getting closer to positive polarity, which would make for absolute polarity IMO" one can only be perplexed by the sloppy thinking and/or writing.
The man does however get a couple of things right:
There is also a misconception that phase and polarity are the same. They are not.
Indeed not. Although, polarity is one specific and singular condition of phase. Also it is binary, either 0 or 180.
The fact that a Behringer or any other processor allows for polarity flipping does not mean that it facilitates or allows for absolute phase.
One notes some additional confusion there (besides that poignant substitution of "phase" for "polarity"). Nothing, but nothing "facilitates or allows" for correct polarity (i.e. absolute polarity) except for the human ear. The ear here is the ultimate, and just about the only arbiter.
We remain...
Poles Apart
I have read many things about what audiophiles say about absolute polarity. Some say that the polarity should be reversed on all channels, which is said to improve something. I think that is a bunch of crap. That is why I said what I did about inverting the channels on the Behringer. If inverting both channels is your version of absolute polarity, then go for it. But that is not the answer.From the Rane Pro Audio Reference:
Polarity:
A signal's electromechanical potential with respect to a reference potential. For example, if a loudspeaker cone moves forward when a positive voltage is applied between its red and black terminals, then it is said to have a positive polarity. A microphone has positive polarity if a positive pressure on its diaphragm results in a positive output voltage. [Usage Note: polarity vs. phase shift: polarity refers to a signal's reference NOT to its phase shift. Being 180° out-of-phase and having inverse polarity are DIFFERENT things. We wrongly say something is out-of-phase when we mean it is inverted. One takes time; the other does not.]
Edits: 12/17/08
If you are dumb enough to put DC on a driver....go for it. I would never put a battery on a speaker. That's stupid.
Touching battery leads to a driver is a common industry practice to verify the pole markings. It would be stupid to accept those markings without verification.
And no, we're not talking car batteries here. Give me a break.
Poles Apart
Actually, using a 9v battery is a very common way to quickly verify + and - terminals on a speaker. Ive been in the pro audio industry for 20 years and people have been doing this (for good reason) commonly all along.
Why do you need to verify pole markings? Your high end manufacturers can't do that right? What hard stuff are they screwing up.
It is not an industry accepted practice. I know too many people in recone shops to know the truth.
You can put a 9volt battery on a decent loudspeaker and it wont
hurt it whatsoever
To think putting a small DC voltage on a loudspeaker will damage
it is dumb
I wouldnt recomend a DC voltage like that on a tweeter or a compression
driver
Your talking a fraction of xmax from a small DC battery on a woofer
or mid driver
ff
Do it on a new speaker from any of the major manufacturers and see if your warranty is still valid.
Stupid because.....?If 1.5V DC damages a woofer, the woofer must be a real P-O-S.
Guy's, I implore you...do not blow your woofers using
a whopping 1.5V D cell battery :)
Edits: 12/17/08 12/17/08
9V has been proven to burn some coils. A little DC may not do the thing in right away. It may contribute to a future failure. This applies to low W drivers mainly. DC is a speakers enemy.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: