• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Test CD to hear inverted polarity

onplane

onplane

What! No Wake???
Somewhere I read that there exists a CD designed to allow folks to test their amps for polarity inversion. Like "perfect pitch", I've read that there are people who can actually hear the difference between correct and inverted polarity audio signals. Now the reason for this that audio signals are asymmetrical. No one could hear any difference between a correct or inverted sine wave, because they are identical.

Now, I'm NOT talking about the situation where speakers are inverted to each other. I'm talking about the situation where all drivers are in phase, but somewhere in the amp, CD, tuner, pre-amp, etc. the entire signal gets inverted.

Is there a test CD that allows folks to hear the differences?

Of course, how can one be certain that their CD player doesn't do an inversion? Nevertheless, if one had such a CD, we theoretically could decide which way sounds best and adjust our systems accordingly.

Now, I happen to have an amp in my basement (collecting dust) that has a tone defeat switch. Purpose is to by-pass tone controls and provide a shorter, more direct signal path. In the process of doing that, however, signal inversion occurs (I know this because I saw it on my scope and in further checking I could see where it happens on the schematic). What I don't know, of course, is which way I'm going. That is, I never tested polarity on the entire system.

Finally, further confusing things, I don't know whether I could actually hear any difference between a correct and an inverted audio signal.

Regards,
Jerry
 
Register to hide this ad
I believe my memorex cleaning CD has a track for checking phase. If you were to run an audio sweep, you'd hear a dropout in sound at frequency if there's an issue.
 
ConradH said:
I'm sure you could make one using Audacity.
Click to expand...
This is the correct way to do it. Take a familiar song, make a copy of it. Make another copy with it inverted. Burn both to a CD. Jump back and forth between songs. Here a difference? If yes, which way is right for you? Little point in finding out what the whole system polarity is from player to speakers, because there is no standard on how songs are to be recorded. It's literally a case by case situation. That's why the polarity button on many preamps is there, so you can switch it on each song to find out which way sounds best to you.

If you really want to do this, for academic purposes, the best test signal to make is a raised cosine. Single cycle of anywhere from 400Hz to 1kHz. I usually make these test signals with a single cycle, 4 seconds of zeros, then repeat this duo over and over for up to two minutes.
 
In theory, a compression from the original sound source to an observer/listener of the live sound should result in a compression at your ears upon playback. The rest of the system should be configured in support of that. This starts with the original mic, recording system, mastering system, distribution system (none of which you have control over). Then when you get the recording, your system would need to support that, too.

Personally, I believe it does matter for critical listening. But that's an article of faith, because I've never done that testing in a home system. But many people do not believe that, and that's not really the 'topic' here.

Back in the days of disco, correct polarity meant the bass drum would punch you in the chest. You can definitely tell that, and it's not a subtle nuance.

Will intermediate stages of incorrect absolute polarity 'poison the well', even if corrected elsewhere? I don't think so, but I've not tested that, either...

House de Kris' single-cycle test would be an interesting one, for sure.

Chip
 
I suspect there's a very simple solution to this. A few years back, I was hunting for system test files and ran across a set of files put up by a PA pro named Bink. It has (had) some very useful files including some that check channel assignments and polarity. It's no longer available on the net but PM me your email address and I'll send those few files.
 
The absolute polarity (correct) of certain signals (sounds of certain instruments) is easily audible.
For those who are curious. I suggest checking out The Wood Effect by Clark Johnsen.
 
I have a CD entitled "Chesky Records Jazz: Sampler and audiophile test compact disc. Vol. 1" Chesky JD37. On tracks 13 to 20, there are two polarity tests. One with solo trumpet and the other, an excerpt from Marina by the David Chesky Band. Makes for an interesting comparison; I don't really hear any real differences between normal and reversed polarity with my system. This CD can be had used via Amazon for BIG bucks! It appears to be available on YouTube but I wonder if the test tracks are really meaningful at the level of a YouTube video?

Bill
 
Well, I did mention that I have an amp that will reverse polarity. I guess that I can drag it out of the basement and listen to see whether I can hear any difference.

Possibly using headphones? I'm thinking a single driver might show the difference better?

None of these proposed solutions, however, really addresses the problem. I think what I need is an asymmetrical signal recorded on a CD that clearly indicates polarity. Something like a 400Hz signal that is recorded as a pure sine in the minus half and possibly a square wave in the positive (or maybe a severely clipped wave in the positive half). With this, any scope could tell whether polarity is maintained by the system, and by the CD player.

Does this make sense?

Regards,
Jerry
 
Stanford Research offers this waveform in their SR1 analyzer for checking absolute phase. Its shape will survive passage through almost any audio processing device:

PhaseWave.png

Are there any Audacity experts here, capable of making and publishing something similar?
 
JoeESP9 said:
The absolute polarity (correct) of certain signals (sounds of certain instruments) is easily audible.
For those who are curious. I suggest checking out The Wood Effect by Clark Johnsen.
Click to expand...
I read the book, which is very interesting. This was a very hot topic for a few months about 30 years ago. I built a switcher that allowed me to switch absolute polarity. I took it to one of our audio club meetings and we all had a listen. Nobody could tell the difference between inverted and non-inverted polarity. The particular recording one uses to test absolute polarity may make a significant difference. Maybe, with just the right recording, one might be able to hear a difference. In any case, it was a fun project.
 
Last edited:
onplane said:
None of these proposed solutions, however, really addresses the problem. I think what I need is an asymmetrical signal recorded on a CD that clearly indicates polarity. Something like a 400Hz signal that is recorded as a pure sine in the minus half and possibly a square wave in the positive (or maybe a severely clipped wave in the positive half). With this, any scope could tell whether polarity is maintained by the system, and by the CD player.
Click to expand...
A raised cosine is absolutely asymmetrical. Much easier to see the effect than a wonky half sine / half square wave. This is the standard test signal to do the very measurement you seek. The bonus is that it will not excite a bunch of high frequencies like the half square wave portion would do. I have just one test CD with this signal. It is either the JAS Test CD or the Denon Test CD. I'm not sure which, hence why I didn't mention it. It is pretty easy to put this signal together yourself with free software.
 
It's easy to digitally synthesize an asymmetric waveform and then listen to it normally and then inverted- that proves absolutely nothing. Preserving any natural asymmetric waveform you can name, through analogue recording chains would be problematic.

Digital systems will of course, allow the preservation of some level of asymmetry, but let's be realistic, the capture of 99% of 'natural' events will include some types of analogue pistonic transducers and they are considerably symmetrical in their dynamic characteristics.

The Denon audio technical test disc has standard single sample impulse tests to test CD player absolute polarity, tracks 76/77 IIRC. The tracks have little other practical use other than showing the effectiveness (or lack thereof) of some methods of digital filtering. Compare early multibit machines with later Bitsream/PWM DACs with the impulse track/s. Often, the single bit implementations allow the 1 sample impulses to 'disappear'...
 
TPettenati said:
You could record these and rip to CD. Maybe not as extensive as you are looking for, more of the quick and dirty test.

Stereo Polarity (Phase) Sound Test
Click to expand...

TPettanati, thanks for that link! The files on that page are of no use, but the link in the page to Absolute Blind Polarity are of tremendous importance to this thread.

These people believe that our ears are simply NOT sensitive to absolute polarity. They further have a blind test so you get a chance to prove them wrong.

I took that test and failed miserably. This is exactly the result tcdrive mentioned in his post.

The blind test is here:

http://www.audiocheck.net/blindtests_abspolarity.php

I'd encourage everyone to take this test and tell us how you did.

Regards,
Jerry

PS: So if we really can't hear the difference, the obvious question is why do the guys who make expensive commercial gear care? I have a guess and it's only my guess, but I suspect the huge benefit is the ability to cascade dozens of amps and having all drivers moving in the same direction at the same time. Further, as amps die and need to be replaced, it's no problem as any generation will work the same.
 
Last edited:
How does the scoring system on that site know if your computer speakers, internal or external, system computer is feeding, or headphones are correctly configured? Or is it just if you hear a difference between the two?

The condition could exist that 100% wrong is just as accurate as 100% right. :)

I suppose the deal could be that, if you're interested enough to head to that site, your gear is already well sorted...

Chip
 
Chip Chester said:
How does the scoring system on that site know if your computer speakers, internal or external, system computer is feeding, or headphones are correctly configured? Or is it just if you hear a difference between the two?

The condition could exist that 100% wrong is just as accurate as 100% right. :)

I suppose the deal could be that, if you're interested enough to head to that site, your gear is already well sorted...

Chip
Click to expand...

Chip, the system doesn't know if your system inverts polarity and since the test is blind, it doesn't care. It provides two signals one of which is the inversion of the first and only asks you to identify which is which in the blind test.

I've tried now with Koss headphones and with my computer 2.1 system and for the life of me I can't tell which blind signal is which. They sound the same to me.

Jerry

PS: How did you do, Chip? Did you really get 100% wrong?
 
You must log in or register to reply here.
Back
Top Bottom