in/out of phase

Posted on Jan 07, 2004 02:39 pm
TheTincanbug
sloppy dice, drinks twice
Member Since: Aug 05, 2003

What exactly do these terms mean? Thanks, gang.

Walt
Chief Cook and Bottle Washer
Member
Since: May 10, 2002

Jan 07, 2004 02:58 pm

Sound travels at given speeds. In recording we consider two of these speeds, through air, and through electrictronics. If the two sounds originate at different distances from the receiver e.g. microphone, ear; they will reach the receiver at different times. Also in electronics if signals can be delayed and reach the final destination at different times. This difference in time is refered to as phase. A good example is two mics 10 feet apart (front to back) from a speaker. The speakers voice will reach the second mic approx. 10 milliseconds later than the first. So you will hear the speakers voice in two phases. The term phase makes more sense when visulized by an ociloscope which displays waveforms in an electrical circuit. Then you can actually see a signal go through a full wave form which is considered to be a circle or 360 degrees. When two waveforms are monitored at once you can see how many degrees one signal lags behind another.

A vivid example of phase is when one signal is input into two speakers where the speaker leads are reversed. A positive pulse in the signal will cause one speaker cone to push while the other pulls. The air movement from the speakers will be 180 degrees out of phase and will cause considerable cancelation of sound if the speakers are close enough together.

Good question. Nasty to answer as the term phase is used in many different aspects of sound and electronics. As well as of the moon.

TheTincanbug
sloppy dice, drinks twice
Member
Since: Aug 05, 2003

Jan 07, 2004 03:19 pm

I see - thank you for your answer!

As an example, you described 2 mics recording the same source, 10 ft apart from each other... Is that not also a good way to describe delay?

Out of phase... I understand that the first mic would record a source several milliseconds ahead of the 2nd mic recording the same source. Would the out of phase term, rather than referring to the time delay between the two recorded signals, refer instead to the dip/rise of the waveforms from each signal, and describe a situation where the two signals are made audible while signal #1's oscillation was at its peak while signal #2's oscillation was at its lowest point, and synced so that signal #1's oscillation would descend (on an oscilloscope) while signal #2's oscillation rose to the peak of its wave?

Noize2u
Czar of Midi
Administrator
Since: Apr 04, 2002

Jan 07, 2004 07:43 pm

You have it mostly right, but out of phase refers to the entire wav cycle (the up and the down curve) being slightly ahead or behind the other in any measure of time. Wether the rise and fall match or not is not important other then the effect, just that the actual wave cycle is not at the exact same time as the other. If you expand one of your tracks and then copy it to another track, and then bump the copy ahead by several milliseconds then they are out of phase. But indeed, delay is also a term to define exactly what we are talking about.

So in the end, even if the mic is 10 feet from the other, theoretically we can have them both in phase with each other, it is just a matter of moving one mic so the signal reaches it with the cycle of the wave hitting it at exactly the same time.

But you can also cheat if you recorded them out of phase, bump your track ahead or back by milleseconds and match it up.

TheTincanbug
sloppy dice, drinks twice
Member
Since: Aug 05, 2003

Jan 07, 2004 09:59 pm

Ok, I think I get it now. Thanks, guys...

Walt
Chief Cook and Bottle Washer
Member
Since: May 10, 2002

Jan 08, 2004 11:49 pm

I think you do have it. Let us say that you are producing two tones both 1k HZ or Cycles per second. That means that one full cycle or 360deg of the wave form will take 1/1000 of a second to complete. If the two waves where 180 degrees out of phase one of the wave forms would be laging the other by 1/2000 of a second or half the cycle time as 180 degrees is half of a full cycle 360 degrees. Again it is expressed in degrees of phase instead of time but the two have an obvious interelationship. Leads or lags (perspective) of more than 360 degrees are usually expressed in time rather than phase. A lag of .5 msec would work out to 180deg for the 1Khz wave forms. The same lag time for a 500Hz signal would be 90 degrees as the frequency is half and the time for a full wave form would be double.

This is really good. I needed to go back over that. Been a long time. I'm supprised this old phart can still grind the numbers.