About Phase. Forward, March.

If you read the previous post and were confused about the term phase, be prepared to get really confused (I’m kidding, I hope.) Actually, an understanding of phase will be important for later posts, when we start to discuss balanced vs. unbalanced audio, microphone placement, standing waves and other stuff. OK. Now, about phase.

Phase

For anyone who has ever worked in broadcast, recording studios, duplication houses, or even anyone who has fallen asleep on the couch and woken up to see color bars and hear an annoying, steady-state high-pitched noise, chances are you’ve heard a one kilohertz sine wave. That’s 1000 Hertz, or 1000 cycles or periods per second. For illustration purposes, we are going to use a sine wave, because it is the simplest of all waveforms and hence the simplest way to demonstrate the concepts. Of course, in the real world, sound waves are much more complex, comprised of many different types of waves, including sine waves and others. There are many images, examples, tutorials and even motion diagrams of sine waves and sound waves accessible on the internet. And since you’re already on the internet, you can save me the time and I’ll save you the torture of subjecting you to any of my hand-drawn diagrams. Do a Google image search for “sine wave.” Then search for “audio waveform” and “complex waveform.” And check some wikis and other sites if you need to learn a little more about the above terms, or any other terms you may not be familiar with. It can be helpful to have a better understanding of sine waves and their motion before reading on.

OK, so now that you’ve done a bit more research, you should have somewhat of a better idea of where we’re going. A sine wave is the simplest waveform possible. Most sounds, or waveforms, are complex waveforms, and are actually made up of many different sine waves of varying frequencies (also called pitch) measured in Hertz or Kilohertz, and amplitudes (or volume), and measured in voltage or decibels, depending on what is being measured and where. Confused yet? Here’s an example: A wave, whether it is a sine wave or a complex waveform, which is in the range we can hear, called the audible frequency spectrum, goes through a series of different stages while on its way to your ears. Let’s use the example of an acoustic guitar which is being recorded. The sound begins as mechanical energy generated in the instrument. It then becomes acoustic energy as it is transmitted through the air as sound. Once it reaches the microphone, it is converted to electrical energy, and it begins a long journey through various stages as electrical energy – the mic pre-amp, various outboard devices such as reverbs, compressors, limiters, effects devices, then onto some storage medium such as magnetic tape or optical disc. Then, it is played back, edited, mixed, and perhaps put through another series of electronic devices until it is converted back to acoustic energy at the speakers and travels through the air to your ears where it is ultimately converted back to electrical impulses in the brain and perceived as sound.

But whether we are talking about acoustic, mechanical, or electrical signals, the theory of phase is the same.So, picture a sine wave, which, after its first complete 360 degree cycle, repeats over and over again to infinity. Now picture a second sine wave which is an exact duplicate of the first sine wave. As long as the first cycle for each wave starts at the exact same point in time, the two waves are said to be in phase. If you now take the first wave and send it to the left speaker of a properly set up stereo monitoring system, and take the second (identical) wave and send it to your right speaker, the waves will still be in phase. What happens with the speakers is that they both push out and pull in at the exact same times and by the exact same amounts, pushing out during the positive-going sections of the wave and pulling in at the negative-going sections of the wave. This is good, if you want them to be in phase, which most people do. Now, let’s say you want to put them out of phase by half the period, or 180 degrees, which, for sine waves, is as out of phase as you can get since with sine waves once you move past 180 degrees you are actually moving closer to 360 degrees at which point the two waves will be perfectly lined up again, except that one of them will be exactly one period behind the other. Still with me? OK, now, you could delay this second wave by 180 degrees electronically, or what you could do is take the terminals of one of the speakers (let’s say the right speaker) and switch them so plus goes to minus and minus goes to plus (black to red and red to black). Make sure the amp is off because if you accidentally short the terminals together while the amp is on you could blow the amp, the speaker, and your ears!

Now you have two speakers which are 180 degrees out of phase! That means that the one that is flipped is actually pushing when it should be pulling and pulling when it should be pushing. What does this sound like? Well, with a sine wave, the difference will probably be subtle, but try it with some music through the speakers (it won’t hurt them.) Depending on how “good” (or “trained”) your ears are, you will notice an effect ranging somewhere between a very subtle difference and having the feeling that some unseen force is attempting to twist your head off! For people with “good ears,” there are few sounds that are quite as objectionable as out of phase speakers. What is actually happening is that the sounds coming from each speaker are identical, but opposite, so depending upon where you and your ears are in the sound field created by the two speakers, there will be varying amounts of what is called cancellation of the waveforms.

Now here’s the cool part. If this phase reversal is done at one of the electrical stages, what will happen is complete cancellation, which will result in - you guessed it – no sound at all. How? If you have each channel on a separate input or fader of a mixer, for example, and they are in phase, what happens when you add them together? They get twice as big (in theory, anyway, but more on that later.) Let’s say you call the peak level of one channel a unit of one. Add an identical unit of one, and you get a total of two units – one plus one equals two. The same goes for when the minimum peak level is minus one. Minus one plus minus one equals minus two. And the same goes for all points in between. Except zero, of course, where zero plus zero is still zero (again, in theory, anyway, but much later on we’ll see how you can never actually have zero – with analog, it doesn’t exist, and with digital, you can’t have zero without running into big trouble. (We’ll discuss dither in a later chapter.)


Polarity

OK, ready to get more confused? Out of phase is not really out of phase. Usually. When we say out of phase, we usually mean reversed polarity. Sometimes, phase and polarity are interchangeable terms. Usually they’re not, they’re just incorrectly used as though they are.

Remember when we flipped the terminals on our right speaker? What we actually did was not flip the phase, but flipped the polarity. Since, in our current theoretical set-up, we are listening to music (complex waveforms, not sine waves) that means that all waveforms are opposite in direction, resulting in the speaker pushing when it should be pulling and pulling when it should be pushing. If we were playing just a sine wave through both speakers, with the terminals flipped on the right speaker, the end result would be the same. That is, whether you delay (or advance) the sine wave in the right speaker by 180 degrees, OR whether you reverse the polarity, the resulting sine waves are identical. So, in this case, with a sine wave, you could say the phase is reversed or the polarity is reversed, and both would be correct. But in the case of music, really what’s happening is the polarity is reversed but the phase isn’t.

Here it is in a nutshell: phase corresponds to time (x-axis) measurement, whereas polarity corresponds to voltage or amplitude (y-axis) measurement.


Absolute polarity

What is “absolute polarity?” No, it’s not icy-cold vodka. When we flipped the polarity of the right speaker but kept the left speaker correct, there was a significant difference in the sound as the left speaker would move out when it was supposed to, but the right speaker would move in when it was supposed to move out, and vice versa. What if we now take the terminals on the LEFT speaker and flip them also. So now BOTH speakers are reverse polarity. But now both are moving in the same direction at the same time, that is, both are moving out when they should be moving in, and both are moving in when they should be moving out. So what will THAT sound like? Well, again, that depends on how “good” your ears are. The difference is not nearly as dramatic or noticeable as when the speakers are reversed polarity, but to many people it is noticeable.

If you’ve ever done any waveform editing on a digital audio workstation (ProTools, Logic, Fairlight, etc.) chances are you’ve seen wave forms up close, or “zoomed in.” If you look really close at, say, a kick drum hit, you’ll see that the initial transient is a large positive-going wave. That means that, if your speakers are BOTH opposite polarity, the initial transient of your kick drum hit is going to be a large NEGATIVE-going wave. Both the left and right speakers will be operating in unison, in that they are both going in the SAME direction, it’s just that they will both be going in the WRONG direction. Some people can hear this, some people can’t. Try it yourself to find out if you can hear a difference. But whether you can hear it or not, make sure you get it right when setting up an audio system, whether it’s the speakers or anywhere further upstream in the audio path. Because there are few things more annoying to people with good ears than improperly set up monitors.

Let’s talk about length

OK, this is the first post, so there may be some terms that some readers are not familiar with. There will be more postings which will cover each term or concept in more detail. For the uninitiated, just wait until the subsequent posts. For everyone else, please don’t hesitate to ask a specific question, ask about a specific topic, or discuss or add your own comments.

Let’s talk about length

Most people who know even a little about speakers know that the wires should be as close to the same length as possible, right? Right, and it’s because if the lengths are different, you will get slight variations in the phase, right? Wrong! Here’s the correct reason:

To make the explanation simpler, let’s assume that you’re not setting up for 5.1 or more, and we’ll use the example of a simple stereo pair of speakers, left and right. And so there’s no confusion, what we’re talking about is non-powered speakers which are going to be fed a speaker-level signal, because when talking about powered speakers with a line-level input, none of the following applies. (more on all that stuff later.)

Speakers generally present very little electrical resistance, measured in units called ohms, and are more commonly specified using a similar unit of measurement called impedance, also measured in ohms, and typically range from 2 to 16 ohms. Because of many other factors, the amount of resistance measured may not be anything close to the actual amount of impedance, but in the case of non-powered speakers, these measurements are nearly identical. A speaker with 8 ohms impedance typically measures about 6.5 ohms in resistance. Are you still with me? OK, let’s use the common 8-ohm speaker in our example. Let’s also assume that you have already matched the speakers with an amplifier that is rated for 8-ohm speakers. This means that each amplifier output wants to “see” about 8 ohms across each pair of terminals, that is, 8 ohms across the left channel and 8 ohms across the right channel.

There are actually quite a few different variables involved, but for simplification, we can say this: As long as both left and right speakers are identical, and both left and right channels of the amp are identical, we can assume that each speaker will behave the same, and this will translate to them sounding the same (well, ok, provided you’re sending them each an identical mono signal instead of a two-channel stereo signal whose left and right channels will actually differ. With a true stereo signal, the left and right channels will have different program information and will sound different, but, nevertheless, you want to make sure that your monitor paths are identical, even if the program isn’t.)

Now, what would you guess would happen if you took one of those 8-ohm speakers away and replaced it with, say, a 10-ohm speaker? You don’t have to be a rocket scientist or even mediocre at math to be able to guess that they will sound different. And you probably wouldn’t recommend that anyone who wanted an accurate stereo monitoring environment do this. But there’s one important variable that we forgot to account for, which, when not taken into account, can have the same effect as replacing one of those correct speakers with a wrong one – the CABLE connecting the amp to the speakers.

See, everything has some resistance, including wire. In fact, resistance of speaker wire can be as high as 0.1 ohms per foot. So, let’s say you have 15 feet of speaker cable on the left side and 15 feet on the right. That totals 1.5 ohms of resistance just for the cable. And, since many 8-ohm speakers actually measure about 6.5 ohms, let’s add 6.5 (instead of 8) to 1.5 to get a total of 8 ohms resistance on each side of our speaker/cable path. That’s good – both sides of your monitor path have identical amp outputs, speaker cable length, and speakers. But what happens if you use only 5 feet of cable for one speaker and 25 feet for the other. You now have 7 ohms on one side and 9 ohms on the other, respectively. This difference may not seem like much, but because the numbers are so small to begin with, you’re talking about somewhere between 20 - 25% of difference in resistance between the two channels! That is a significant difference, and certainly enough to cause a drastic difference in sound between the left and right sides.

So now you’ve seen why even only 20 feet of difference in length can cause a significant difference in sound, or, to use the correct term, frequency response. Now here’s why even a difference of 200 feet or more will have absolutely NO NOTICEABLE EFFECT on phase:

Sound traveling through the air is very slow – only about 1000 feet per second. Sound traveling through wires is very fast. In fact, sound traveling through wires is actually electricity. Electricity travels very near the speed of light, or 186,000 miles per second, which translates to nearly a billion feet per second! Or, about a billionth of a second per foot of cable. So 20 feet of cable on one side and 220 feet of cable on the other side will result in sound arriving at only 200 billionths of a second later on the side with 220 feet of cable! For the mathematically-challenged, that’s less than a microsecond, which will not be audible as far as phase is concerned. But the resistance of the side with 200 feet of cable is now going to increase the overall resistance of that side to about 26.5 ohms (6.5 for the speaker and 20 for the cable.) That WILL result in a significant difference in the sound of the two speakers, but it has absolutely nothing to do with phase. So, in theory, there is a difference in phase, but you show me the “golden ears” that can actually hear that slight difference in phase, and I want some of whatever they are on!

Stay tuned for the next post, where we will discuss phase, or maybe resistance, or maybe Ohm’s Law. I don’t know yet…

Welcome to Audio Auditor

Welcome to Audio Auditor. This blog is the culmination of my years of experience in and dedication to the world of audio and sound - often confusing, sometimes complicated and ever evolving. I have toiled in some of New York City's top recording studios, installed audio and video systems in some of the highest priced real estate in the world, and worked around some of the biggest names in the music, entertainment, and media industries. I have also been "in the trenches" of some of the lowest-budget recording studios and operations that one can imagine.

Throughout the years, I have come to find out that all of us have one thing in common - We love sounds. Maybe we don't all love the same sounds, but we all love at least some sounds. And another thing I've discovered is that the way these sounds are created, manipulated, transmitted, stored, reproduced and duplicated is, in many cases, poorly understood.

There is a lack of information in this area, a lot of mis-information, and, in many cases, information out there that is just plain wrong. Often, information starts out being presented at a very basic, almost elementary level, and then suddenly seems to jump several steps ahead. The poor reader has not been given enough information to bridge the gap to understanding the next level and is left wondering what he missed or why he doesn't understand. In many cases, I think the information is just poorly explained or presented. And all too often, people who don't understand a concept are attempting to learn it from someone who is only slightly more informed than they are.

I hope this blog can help separate fact from falsehood. My goal is to explain and discuss audio theory and applications in a way that is understandable by everyone, and to help bridge the gaps of knowledge that lead to confusion and misunderstanding. I believe that the sharing of knowledge and discussion of ideas makes us all smarter.

For me, audio, sound, and music have been a lifelong obsession. There is no such thing as "perfect" sound, but that doesn't mean we have to stop striving to obtain it. After all, this is an art, and all art is subjective. I feel a motto coming on here:

"Let's strive for the best, but be perfectly happy to settle for better."