[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2007, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
69.200.83.8
In Reply to: RE: some thoughts on polarity posted by rick_m on June 04, 2007 at 19:50:34
It's true that a dense surface like a wall has higher acoustic impedance than air, but that means a wave in air that reflects off that surface will have inverted polarity.
Look here for example:
http://www.physicsclassroom.com/Class/sound/u11l3c.html
http://www.kettering.edu/~drussell/Demos/reflect/reflect.html
or here for some great animations:
http://www.gmi.edu/~drussell/Demos/waves/wavemotion.html
Thanks for the interesting links. Finally figured out how to turn on animation in the browser and that perked them up. I'm thinking that this one you referred to: http://www.physicsclassroom.com/mmedia/waves/free.html kinda applies here. Although it's showing a transverse impulse hitting an open end, I think the same condition applies to a longitudinal wave hitting a closed end. If you take the vertical displacement in the animation as the pressure of the longitudinal wave it looks good.
I believe confusion ensues when trying to think of longitudinal waves using a transverse mental model. I pawed around on the web and found a site that seems to apply, check out:
http://hyperphysics.phy-astr.gsu.edu/hbase/sound/reflec.html
They even have a scheme to visualize the phase of longitudinal waves with a string analogy.
Here's one that also applies although it's sonar: http://interactivesonicspaces.interfaculty.nl/ferenc_teglas/reflection.htm
I'm hanging on to my belief that sound waves do NOT invert when bouncing off of solid objects.
So, can we infer that this means that absolute polarity is twice as important when listening in a reflective environment?
Regards, Rick
Your belief is correct - what I said before was wrong. I apologize for that - one should never believe anything one reads on the internet, including this sentence... in fact I'm going to email the administrator of that site. It's pretty inexcusable to get that wrong on a website devoted to helping high-school kids learn physics!
Now, how can we see what happens to sound waves bouncing off a barrier? The easiest way is to simply visualize a compression pulse traveling towards a wall. If no energy is absorbed into the wall, the pulse must bounce off with the same amplitude. And because the number of air molecules is preserved during the bounce, the reflected pulse must also be a compression pulse.
More formally you can just solve the equations of motion, subject to the boundary condition that the air displacement at the wall is zero. That boundary condition is like a fixed end string, for which the reflected pulse is inverted. In other words there's a node in the displacement at the wall. But the pressure is the derivative of the displacement with respect to position, and so the reflected pressure wave has the same polarity as the incident wave, and there is an anti-node of pressure at the wall.
What does this mean for the issue at hand? Well, as I mentioned before, the moment there is more than one source of sound and more than one mic per channel you can't preserve polarity when you mix the mic channels. Furthermore the polarity of sound waves generally changes as a function of position. For example the polarity of a dipole radiator reverses as you move in a circle around it - so reflected waves from one side are of opposite polarity as reflected waves from the other. So only if we record one monopole radiator with one mic is there a correct phase which it makes sense to preserve, at least as far as I can see.
I read back on what I had originally written on HA about room reflections, and in hindsight, I completely misunderstood what I was responding to, as well as getting my references wrong. Glad to know that the situation is more nuanced than I had thought.
Thanks Rick and truthseekerprime!
I haven't read your postings on the subject, but am pleased that you found our exchange helpful. But as my psych prof used to say, "there's no stars in being right and no stigma in being wrong". I only happened to notice that it seemed backward because I'm used to working with RF transmission lines.
My thanks to you both for the gracious responses, it's great fun being able to kick things around and learn from one another.
Long Live Home Audio!
Rick
I noticed our postings crossed in the mail, as it were. But our stories are the same.
Multi-miking is the bane of recording. I think suffering all the problems that ensue with a single microphone are worth it. Frequency response is way overrated in my book as I believe that impulse response is far more important to ultimate satisfaction. As you've stated, it usually gets lost in mix-downs.
As far as reflections go, I don't think that they nullify the importance of correct phase. As you mentioned earlier, consider binaural recording, you end up recording the reflections, but you would also hear them if you were there. Mixing together Mics. from various places is unnatural and wipes out essential temporal information if their levels are anywhere close.
Regards, Rick
It's certainly true that binaural recordings played through headphones have by FAR the most realistic soundstage I've ever heard. It's really eerie how realistic they sound, even on medium to low quality phones.I think the future is probably sophisticated digital processors which compensate for the listening environment and speaker system and adjust the phase etc. appropriately. Suppose we replace CDs with a new digital format that contains information about the recording venue, including multiple tracks recorded by mics at different positions (that information is usually kept by record companies so they can make new mixes and re-release them). There would be a digital pre-amp that calibrates itself once to the speakers and the listening room with a few measurements, and then processes the recording based on that information and the information in the digital file so as to produce the most realistic result at the listening position.
Not simple, but well within reach of current technology, I would say.
I believe you correct both in the potential direction and that it's within reach of current technology. I suppose we would record everything as dry as possible and convolve it with the forum coefficients upon playback. But how many people care enough to bother? Surely not the music business. Plus the additional processing, if not done to "audiophile" standards could degrade the final result.
Perhaps a simpler approach would be to just standardize on binaural (with defined polarity of course). There are orders of magnitude more people listening with headphones now than there were 20 years ago and they could get great sound with no processing. Our home stereos could concentrate on delivering that sound to our ears sans headphones.
Actually for me, just simple stereo or even mono, can be very satisfying if it's recorded and mastered well. Wish more were...
Regards, Rick
It is interesting that you two are questioning the microphone techniques. I do believe it further advances an understanding of the issue of polarity.
In the case of many 'pop' recordings, I believe you'll find your first statement to be true. Mikes are set up very close to, and actually within the instrument at times. Mikes are within a drum kit, and are often clipped to the bells of the various instruments, and thus they are often actually employing what you are advocating (in saying that the recording should be 'dry' as possible.
As far as orchestral recordings or large scale ensembles, the distance does become critical. Binaural, or single stereo mikes are very good and avoid most issues. Decca, as you are probably aware of, used a 'tree', a 'T' shaped jig upon which the various mikes were mounted with different patterns depending on what was required. Here, the mikes employed were roughly about 3 to 4 feet apart at the maximum.
TSP (truthseekerprime) is very correct in stating that the distance between the mikes present a potential problem in recreating a phase correct experience. Audio wavelengths can be very long, over 30 feet in length and also very short. Here a cue can be determined by the average distance between a set of human ears: say about 6 inches or so.
However, the situation of recording live is somewhat ameliorated by the triangulation of the placement. In the case of the Decca 'tree' the mikes are placed, say, 40 feet away from the ensemble. While the mikes may be a slight distance apart from each other, they are roughly equidistant from the ensemble itself. The result is manifested in the sound obtained: things are more coherent and more consonant with what you hear in real life. Use of a cardioid pattern will eliminate a certain amount of crosstalk between the mikes, which can help.
In contrast, listen to the Decca recorded Phase Four albums. They are heavily multimiked and the perception is that you are sitting within the orchestra. Many of the classic RCA's have similar sound, though maybe not as apparent. The highly lauded RCA Reiner/Scheherazade (TAS super disc list) falls in that category. I believe only the first three stereo recordings were minimally miked: The Reiner/Zarathustra [LSC 1806], the Fiedler/Gaite Parisienne [LSC 1817]and the Munch/Daphnis and Chloe [LSC 1893]. These are minimally miked recordings and sound very different from other RCA's which typically use 6 to 7 mikes.
In the case of the Scheherazade, the listener can hear anomalies in the three dimensional soundstage presented. The various solo instruments zoom to the foreground and when their solo is over, retreat back into the orchestral mix.
The most interesting microphone technique currently employed is the one designed and utilized by Ray Kimber. His operating principle was to try and recreate the typical loudspeaker set up but to avoid the phase issues by having mikes set up say 10 to 12 feet apart, the typical distances most speakers are placed apart.
It is not an inexpensive set up, but it has fascinating results and is very realistic. The set up is only available at Weber State auditorium, IIRC, but the recordings there are superb, although not all the University's music students may be quite deserving of the quality of the recordings. In addition, Ray employs several other 'tricks' in order to get very realistic recordings. I've seen him advertise the sample discs he has on AA with all proceeds going to the Weber State Music program. If you have not heard his recordings I would highly encourage you to try them. The technique obviously will not be universally adopted, but they remain as a fascinating experiment in the recreation of a live event, and a very realistic one at that.
Stu
I'll sit down later today and calculate this. For the moment, read this from the link in my previous post, which very emphatically states that sound waves reflected off of hard surfaces are polarity inverted:
http://www.physicsclassroom.com/Class/sound/u11l3c.html
These principles of free and fixed end reflection can be applied to sound waves. Though a sound wave does not consist of crests and troughs, they do consist of compressions and rarefactions. If a sound wave is traveling through a cylindrical tube, it will eventually come to the end of the tube. The end of the tube represents a boundary between the enclosed air in the tube and the expanse of air outside of the tube. Upon reaching the end of the tube, the sound wave will undergo partial reflection and partial transmission. That is, a portion of the energy carried by the sound wave will pass across the boundary and out of the tube (transmission) and a portion of the energy carried by the sound wave will reflect off the boundary, remain in the tube and travel in the opposite direction (reflection). If the end of the tube is "open" or uncovered such that the air at the end of the tube can freely vibrate when the sound wave reaches it, then the behavior at the boundary resembles free end reflection. There is no inversion of the disturbance when reflecting off the open end (uncovered end) of a cylindrical tube. That is, if a compression is incident towards an open-end, it will reflect and return as a compression. Similarly, if a rarefaction is incident towards an open end, it will reflect and return as a rarefaction. The opposite occurs if the end of the tube is "closed" or covered up. If the end of the tube is "closed" or covered, then the air at the end of the tube is fixed and cannot freely vibrate when the sound wave reaches it. In this case, the behavior at the boundary resembles fixed end reflection. There is inversion of the disturbance when reflecting off the closed end (covered end) of a cylindrical tube. That is, if a compression is incident towards an closed end, it will reflect and return as a rarefaction. Similarly, if a rarefaction is incident towards an closed end, it will reflect and return as a compression.
The behavior of sound waves at open ends and closed ends will become important Lesson 5 during the discussion of musical instruments. Many musical instruments operate as the result of sound waves traveling back and forth inside of "tubes" or air columns. These waves reflect at either a closed end or an open end of the air column; and the fact that inversion occurs at a closed end will have a huge impact on the numerical pattern of frequencies produced by such instruments.
Well, he may be emphatic, but then so are politicians... I still think he's got it backward. Imagine an audiophile's wife getting to the mailbox first and opening the Visa bill. Seeing what his new preamp cost, she trots into the listening room and shoots him.
The pistol causes a rapid increase in the room's air pressure which propagates as a positive wavefront. When it strikes a wall (a high Z), it can't go any further and the pressure builds up at that point which in turn produces a positive wavefront which propagates outward back into the room. It reflects in phase.
In her haste she left the door open so when the positive wavefront going in that direction reaches it, it can escape and most of it does rapidly decreasing the net air pressure and causing a negative wavefront to propagate back into the room. It's a low impedance and reflects out of phase.
The moral is that high impedances reflect in phase, low impedances out of phase and that a new preamp can have hidden costs.
Regards, Rick
x
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2007, Audio Asylum, all rights reserved ]