Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Level]

Greetings.

First of all, this is long and I repeat myself because some people simply have a hard time understanding me. Hopefully this time, it is spelled out.

I bring this up not only for my understanding but also others that may be enlightened by responses of those in the field. Without any further ado, lets discuss absolute polarity in the audio chain.

Not polarity per speaker but phase as it travels down the chain in absolute terms of staying coherent to the recording event.

This weekend I have studied absolute phase in the chain. To my surprise, I am consistent in my chain all the way through whereas I wonder who actually pays attention to it. This is basic stuff, granted but to verify my results, I would like a few basic questions answered.


Given the above scenarios, we would assume “compression” to the microphone diaphragm (diaphragm being pushed from initial sound event) should produce a positive wave (looking at single cycles) or rarefraction, with the loudspeakers. (outward cone motion)

JBL woofers (4310/4311/4312) have woofers with inverted phase meaning when a PNP transistor bank is pushing them, the cone is in compression mode (moving backwards). or pulling them (A JBL bright idea??)

1. In terms of absolute phase, what is correct here?

Should a compression event to a microphone cause a positive voltage swing to make a driver move outwards or should the JBL model be normal in this?

Most high quality monitors being used today work with compression events on the negative and rarefraction events on the positive. To clarify..... Microphone compressing means positive and loudspeaker moving forward. The signal on the positive.

2. When viewing the extreme zoom of a track, should the first movement of the wave file be in positive terms or negative terms? Can we hear the difference?  

Some in the field of audio says it does matter and a difference is apparent in sound quality.

How does absolute phase “travel” from microphone, to recorder, to mixer, to 2 track recorder, to mastering, to pressing plant to consumer CD player to speakers and then to the listener. How much of it is screwed up and that by simply changing the polarity at BOTH loudspeakers make that much of a difference?

In terms of absolute phase (input to output of your chain in mastering), here are some links to ponder. I would like some expounding on my questions if you will based on your experience and study of this.

Should we give a damn actually? It took all day to check my chain out down to the other systems as well. Only with the JBL’s did the events change at the loudspeakers. I have one CD player and one CD burner that inverts polarity. I am happy to have found this for sake of not wasting the day.



Here are my findings.

A.   Some equipment inverts phase internally (1 channel) and then re-associates, back at the outputs. From my understanding, this is to put “less strain” on the power supply and or rails
B.   Some equipment inverts phase on both channels.
C.   Some cables come phase inverted.
D.   A mix-down that sounds as though it is “fighting itself” may have one or more channels “out of phase” channels mixed in other channels that are of opposite phase.
E.   Some microphones may have phase inversion.

Here are some web links to ponder.

http://www.positive-feedback.com/Issue1/cjwoodeffect.htm

http://www.siber-sonic.com/broadcast/polarity.html


     http://www.econcrisis.homestead.com/MiscEssaysAbsolutePolari ty.html

     http://www.enjoythemusic.com/magazine/rickerinterview/ricker 9.htm

http://www.gcaudio.com/resources/howtos/absolutephase.html

[Wyn Davis]

Bill,

I have over the years found that this subject is almost like discussing religion or politics. Some believe, some don't.

Here is my take. The polarity at the output of any device should absolutely match the polarity of the signal being fed to the input. Further, I think that at any point where a signal can be intercepted during a process, it should match the signal fed at the input. I have no opinion about whether it matters if a signal is inverted during some part of a process prior to an output stage. I have never done any listening with that particular in mind. I am sure others around here have.

In the mixing environment I believe there are occasions when a signal inverted against the majority can be useful. In fact, (stepping out on a limb here), I think I recall eavesdropping on a lecture by our host all these many years ago at UCLA where he said he sometimes inverts the polarity of one overhead against the other. Point being, in the mixing environment one does whatever it takes to get the desired result.

In building a large audio system, like a studio, polarity should be positive, absolutely.

Finally, phase is in the time domain, polarity is not. Lets not confuse this issue by mixing those terms.

That is my position, I'm sticking to it.

Digging in ...

Wyn Davis
Total Access Recording
Redondo Beach, Ca

[Level]

So, compression of a microphone diaphram (the inital wavefront hitting it) should produce a positive start and push the loudspeaker cone outward..like a real sound event. Right?

I bet many systems get this wrong..in fact, the JBL's I spoke of certainly must have had a reason for being 180 out.

This is one factor I am trying to understand is "Why" and also, does anyone pay attention to it...in mastering.

[Nika Aldrich]

You want your recording chain to ideally be set up so that an event that causes air compression on the recording end causes air compression on the playback end.

If a microphone diaphragm moves inwards (away from the source) then the speaker cone should move outwards, simulating a move away from the source.

Issues are caused with instruments wherein the microphone does diaphragm does not move in this fashion - a mic on the underside of a snare, on the beater side of a kickdrum, on the backside of an amplifier, or on the reverse side of any other transducer.  In these situations you would typically flip the phase of the microphone to accommodate the air pressure movements.

As for whether or not absolute phase is audible, it varies with material.  Recent tests that I read about done at the University of Waterloo in Canada, if I recall correctly, showed that the ear is indeed sensitive to absolute phase upon critical testing.  An explanation is given in the relative movement of the basilar membrane.

Nika

[Bob Olhsson]

For some reason people in the hi-fi world keep making some kind of an issue out of polarity along the lines of exotic speaker wire arguments. It's been very well established for 70 years that polarity has frequently been observed to be audible and hence important to keep an eye on.

[JGreenslade]

On one occasion I was pitching a song to some media mooks (go easy on me); they heard the original arrangement and wanted some changes so I went back, not changing the desk, but wiring up the o/p from the HD recorder feeding the main vocal with a different cable for reasons that escape me that this moment.

When I returned the next day (having been up to 4am arranging) the mooks and label management were up in arms saying that the vocal sounded awful, even though I knew the singer was fine technically. When I returned to the studio I realised the cable carrying her vocal was wired inverting.

I wonder if the difference could've been to electrical consideration though. i.e: The desk + recorder had inverting op-amp i/p/o/p stages, and the one big advantage of transformers is CMR, whereas an op-amp will need careful evaluation to approach the CMR performance even a cheap xformer has as standard? Uneven DC-offset in electrically-coupled circuit perhaps?

I always check absolute phase religiously (a DK is a favoured means), although am not entirely sure why... Have done for years, maybe it was all those hi-fi magazines during childhood.

Cheers,
Justin

[Level]

We still don't have the JBL mystery sorted out yet..afterall, they were the dominant monitors from 1970-1980 and some of the finest productions were done during this time.

[JGreenslade]

Quote:

A. Some equipment inverts phase internally (1 channel) and then re-associates, back at the outputs. From my understanding, this is to put “less strain” on the power supply and or rails

Love to hear comments from the more academically inclined members as to why it could exert less strain on pwr rails.

The only scenario I could envisage would be if one rail had more current supply than the other, but that would be a weirdo supply right?

We should also reinforce for reader's benefit that polarity and relative phase are different creatures.

You can't hear relative phase, everyone knows that :-)

J

[JGreenslade]

While we're at it Bill (or anyone else), have you got any tips for using 4333Bs? (sorry for the thread creep)

Dispose of internal x-overs and go fully active, or just bi-amp and bypass the low inductor?

J

[Level]

I am sorry, I am not familiar with the 4333B's. I would have to have them here and test, to do a deterimination what x-over frequencies, slopes et al would be proper for them. I have never used that model.

Ok, I looked it up.

http://www.jblpro.com/pub/obsolete/4331b_4333b.pdf

Looks simalar to the L-300. I have used L-200's and I actually have a new pair of 077 tweets sitting around which would make an L200 (with proper x over), an L-300, with the sloping front.

Xover is at 800 (12dB/oct) and 8500 (12dB/Oct) from the specs.

If you multiamp them, you would either need a calibrated DB meter or balance the levels by ear.

Quote:

Love to hear comments from the more academically inclined members as to why it could exert less strain on pwr rails. The only scenario I could envisage would be if one rail had more current supply than the other, but that would be a weirdo supply right? We should also reinforce for reader's benefit that polarity and relative phase are different creatures. You can't hear relative phase, everyone knows that

That would be an interesting exchange indeed. I look forward to it myself. AFAIHH, I single supply should not have as much "swing" if two channels, out of phase, draws from it VS in phase. I am going by the observations of others on this..not my own. This was pertaining to Power amplifiers and line amplifiers. I will try to dig up the links. I simply remember some designers hammering this out and measuring the results on the famous "joe list" which most designers pay attention to.
YMMV.

[Giovanni Speranza]

How  can you check if your audio chain results in an absolute phase? A low freq would show the speaker movement but it should be so low that it would toast it.

[bobkatz]

Joe Speranza wrote on Wed, 17 November 2004 20:56

How  can you check if your audio chain results in an absolute phase? A low freq would show the speaker movement but it should be so low that it would toast it.

You can get a polarity tester. It's actually an excellent tool that will expose a wiring fault anywhere in the system. When working at a new studio I used to go round with a polarity tester in front of each microphone and with an assistant in front of the loudspeaker with the other half of the tester.

I just sold my old polarity tester as the Audio Toolbox already has one built in. Basically, the generator puts out a positive going pulse on pin 2 and the detector should detect a compression from the loudspeaker. Where it gets hairy is whether to detect the woofer, the tweeter or the aggregate.

[David Schober]

I agree about the polarity tester.   It's a must for every studio and essential when doing a live recording.

For those of you old enough to remember (like me), the old direct-to-disk records that Doug Sax's Sheffield Labs made had an interesting note somewhere in the liner notes.  It read something like, "Audiophiles note: for maximum results, please switch the polarity of your speakers when playing this album."  I remember reading that when I lived back in Texas wondering why they'd say something like that.

Years later when working with Doug, I asked him about it.  He explained to me that the cutting system he had in those days produced a reversed polarity LP.  I don't remember the exact reason, but fixing it would have required another amp in the circuit.  This was something he was unwilling to do as he felt the sonic detriment, if discernible on home systems, was a better compromise than the extra amp.  I don't know why he couldn't just rewire the output stage.  He didn't seem to be putting one over on me, but I was still wet behind the ears back then....

[Level]

David, I remember that too on the Sheffield Lab records. I had the JBL 4311WX and Klipschorns in 1976 when that came about, I had to invert the 4311's anyway, due to the woofer. So this was, invert the Klipschorns, and JBL's were back to zero phase!... and it was madness!

Sheer killer sound those albums are. It gets very little better.

I still have all my Sheffields on vinyl and a safety of Dave Grusin "discovered again" which I am remastering. Fine stuff!


Imagine Klipschorns with custom crossovers triamped and 4311WX hanging above them. I still have the picture somewhere. It was unreal then..still is!

[bblackwood]

"They made my guitar go straight ahead in a line for miles, and sound like it's coming through a telephone."

- Steve Miller (on the Klipsch speakers)