Q: "Everything I record is out of phase. Is this a problem?"
A question from an Audio Masterclass reader...
"It seems that everything I record is out of phase, or in inverted polarity. Could it be the power coming into the house? Or not being grounded properly? Do I lose quality when this happens? And when I invert to the correct phase, do I lose anything there?"
Let me start by saying that this has nothing to do with mains power or grounding. Having got that out of the way I can begin to address this issue. Firstly, what exactly is the problem?
When people use the terminology 'out of phase', they generally mean that the signal is inverted. There is a lot else that could be said about phase, but the issue here is signal inversion, so that is the terminology I will stick to.
Imagine a microphone in front of a bass drum. The drummer whacks the beater and the head moves outwards towards the microphone. Conventionally, this will cause the microphone to generate an initial positive electrical pulse. Inward pressure on the diaphragm of a microphone creates a positive voltage.
The signal from the capsule of the microphone goes to pins 2 and 3 of its XLR connector. If this is wired correctly, pin 2 will carry a positive voltage compared to pin 3.
This pulse will then travel through preamplifier, audio interface, digital audio workstation, bounced .wav file to mp3, all the way around the world via the Internet, be downloaded onto someone's iPod and ultimately be reproduced by a loudspeaker in their iPod dock...
The acoustic pressure pulse created by the bass drum will cause the diaphragm of that loudspeaker to move outwards. And the listener's eardrum will move inwards just as if he or she had been standing right in front of the drum.
This is how things should be. But at many points in the chain, the signal can become inverted. The most likely cause is a cable where one of the connectors is wired with pins 2 and 3 the wrong way round.
It can also happen that a piece of equipment is incorrectly designed and inverts the signal. You can find some interesting comments on poorly designed equipment here...
Does signal inversion matter?
While it may seem like a horrendous distortion to invert an audio signal, the human ear doesn't seem to mind. Indeed, it is very difficult to tell whether or not a signal has been inverted.
Take the example of a drum kit with close mics. Whereas the bass drum pushes towards its microphone when struck, thus producing an initial positive pulse, every other drum initially moves away from the microphone diaphragm. Thus one drum in the drum kit creates positive-going initial pulses, every other drum creates negative-going pulses. Whoever worries about that? Nobody. It makes no practical difference.
If I were to write a list of one hundred things to worry about in audio, signal inversion, in itself, wouldn't even be on that list.
By the way, when considering whether a signal is correct or inverted, the phrase 'absolute polarity' (or 'absolute phase') is often used. If you look this up, you will find that a lot of people are fascinated by the topic. It's interesting, but not so relevant in day-to-day recording.
When signal inversion absolutely does matter
Although the absolute polarity of a single signal is not that much of an issue, when you start to combine signals it becomes very important.
For instance, if you have a stereo signal and, through some kind of error, the left channel becomes inverted, the stereo sound stage completely breaks down. If, for instance, an instrument is panned center, the left loudspeaker will be pushing out while the right is sucking in and vice versa. There is nothing in the natural world that creates this effect and the brain has no sensible way of interpreting it. It sounds bad indeed, and if you have not experienced it you should try some experiments in your DAW.
Alternatively, you might be processing a signal in some way that requires that you mix in the processed version with the original. If the processed signal is inverted, then it will partially cancel the original when it is mixed in.
Signal inversion has the potential to create havoc in any sound engineering scenario, which is why it is so important that absolute polarity is maintained at every point in the signal chain.
Back to the question...
If everything that the questioner records is inverted, then it would seem to be the case that there is a single point in the system where the inversion is happening. If so, this should be sought out and the problem resolved. Inverting back to the correct polarity will not cause any further problems, but the issue should not be allowed to arise in the first place.
In summary, although absolute polarity is of less relevance than many other factors in recording, any item of hardware or software, or any process, that inverts the signal has the potential to cause significant problems. There is a correct way to handle signals, and that is not to invert them. Not unless you have a very good reason to do so.