phase issues

Massili

Member
Jun 3, 2005
431
0
16
well, i know it's an old question but i'm really working hard to comprehend phase issues! I know what its effect in the music and know (basically) what is it...
i suppose it can be a good thread to those, like me, dont know how to determine/find out/fix it in the host program (i use cubase sx3)

.... A friend of mine asked me to record his band and we tracked the drums and i'm trying to check all the possibilities to correct phase problems in the recorded material but i dont know how to check/find out how to fix it!

I found that Voxengo PHA-979 can correct phase problems.... i'm very confuse with this thing and i'd like help

I read oz article (drum guide) and there is a part there explaining about phase , i read that but didnt get to apply that....

thx in advance
 
"I found that Voxengo PHA-979 can correct phase problems...."

Nothing can fix phase except you....
 
The concept is not really that difficult. When zoom in really close on 2 waves recorded from two different mics (lets say two adjacent rack toms) the peak and valleys of the two waves should be more or less reinforcing each other. I mean, hit the left tom only and check the waves it records on its main mic and adjacent mics (spill) and compare.
If you got peaks on one wave (Mic A) and valleys on the other (Mic B) at the same time instants then the energy will cancel each other out and your sound will sound thin. Adjust the adjacent mics that give you trouble then rinse and repeat.
 
There are two components of phase in my mind.

Time and amplitude.

If two or more pick up points are recieving the same source signal chances are they will be at different times. If these times are within about 50 ms or so they will cause a summing of the two signals to occur, the result may be a adverse or benificial, it is entirely subjective. A doubling of amplitude may cause an unwanted colour to the sound, or highlight an element of the room that is undesirable.

The thing about phase is that is everything in audio. Whenever you have the same frequency playing at the same time at the a similar amplitude there will be phase interraction. So, if you have 1k ringing on the snare and then 1k on the guitar there will be phasing if there timing iss close enough together.

Accept that there is no such thing in reality, at least in my experience as perfect phase. Attending to phase is a damage limitation excercise. So when you hear the life being sucked out of a track and no bass, by remebering that phase is caused by relative amplitude and timing, you can adjust your gain structure/mic positioning/polar pattern choice to minimise this.

I have never had any luck correcting phase after the fact, just minimising its effect, but getting it right on the way in is sadly the best way to do it.
 
thank you guys for the explanations. About the peaks and valleys i know that's a phase problem. Im trying to line up the waveforms of a drum track in cubase by adjusting the delay time (ie. top and botton snare mic) but in fact i dont know if what i'm doing is really fixing the phase issue also i dont know what can be my start point (which waveform of the drum kit) to fix the others from that point, could i be clear? ... i can post pictures on the waveforms if anyone would like to help me with this!

thx
 
thank you guys for the explanations. About the peaks and valleys i know that's a phase problem. Im trying to line up the waveforms of a drum track in cubase by adjusting the delay time (ie. top and botton snare mic) but in fact i dont know if what i'm doing is really fixing the phase issue also i dont know what can be my start point (which waveform of the drum kit) to fix the others from that point, could i be clear? ... i can post pictures on the waveforms if anyone would like to help me with this!

thx

Are you commited to these tracks. If not re-track and make sure you are completely out of phase between the two snare mics. Then flip it on your pre if you can. If not do like others said and flip it in the host software.
 
You shouldn't have too much phase cancellation in those two snare tracks (if the snare's heads went completely bonkers out of phase then we wouldn't hear the snare at all) so quit tinkering with numbers and zoom in as much as you can.

The thing you're looking for as far as alignment goes will be the big snappy attack that we're hoping is visible in those tracks. You'll have to be zoomed out just far enough to see the difference in amplitude between attack and decay, but be close enough to know when the two are lined up more or less as best as they're going to get with a reasonable amount of work. For right now all you have to do is line the two up as best as you can. Once you've gotten that part in play through the track, alternating the phase switch in the track settings just to see if before you had the things out of phase, and figure out which has the most constructive interference on the low part of the snare's sound.

----------

A quick side note, in case you were wondering if the heads should be delayed on disk because there's a slight natural delay between the two in the room. Assuming the temperature in your room is about 21* C (or about 70* F if you don't like Celsius) and you don't have a completely whacked air composition (if you're not in a sealed oxygen room, or the Hindenburg, or on the surface of Venus, this is a no) sound will be traveling at about 771.2 miles an hour. We can put this in slightly more convenient terms - 1129.5 feet per second - and figure out how much delay you'd have in the room if you could separate completely the sound of the two heads. If you're using a six-inch snare (which you are, because six inches is a half of a foot and it makes the arithmetic easier) and you're standing directly over (or under) the snare, that distance of six inches is traveled by the sound in .443 milliseconds:

1129.5 ft/s ^ -1 = .886 milliseconds (sound travels 1 foot in .884 milliseconds)
.886 * 1/2 = .443 milliseconds (six inches is half of a foot, so it takes half of the time for sound to go that far)

If you must use a five-inch snare, the delay is .369 milliseconds.

What's more, this delay gets smaller as you move farther away from directly over or directly under the snare - you can best approximate this by taking the sine of the angle from directly overhead to the center of the drum to the observer, so assuming you're sitting a fair distance away from nutsonthesnare the delay goes down to .384 milliseconds for the six-inch and .320 milliseconds for the five-inch.

So yes, we're talking very small amounts of time, but if you must throw in a delay to complicate things past 'the big lines mean hit and we line them up YAY!' then you have that. Merry Consumermas.

----------

If you've lined up the attack and tinkered with phase reversal in the host software and you're still not getting things sounding right, post an image of a small part of the wave forms - preferable two to five hits, nothing more. Hopefully we can scribble things back in a paint program and we'll make sense of that.

Jeff
 
Jeff, thx alot for your helping

well, i have to read it a few more times to understand better (i already 2 times - lol). Your explanation was beyond than my brain can accept today! heh

I didnt line up anything yet. I mute all the others pieces and hearing to the snare mics i phase reverse in the host and got a better sound than before, it sounded thin and with a more treble sound and now fuller and heavier, unmute all the pieces and played with the phase reverse button and didnt hear any difference. Did the same think with the over mics and hear different responses in the stereo image, cymbals sounded a bit treble with no phase reversed channel. Of course we always should "hear" what sound best and im following my ears and making sure that i can deal with phase problems as much as i can to get better results in my recordings and also learning, thx for helping me Jeff... i will do more tweaks tonite, i have got no much free time and hopefully i can today! sry for my bad english!

:)
 
This might sound stupid but I really haven't had a lot to deal with one source that's been recorded through multiple mics and the occuring phasing problems so...

...why can't we just align the the phase in our DAW by moving the waveform a little left/right?
 
I think one important thing that everyone is overlooking is the fact that if you have to adjust for phase that much in the daw then you messed up recording your source tracks. You have to remember when you start moving things you might be aligning them with one other track but you are now un-aligning them everywere else. There are times were you have no option but to fix something (i.e. snare top and bottom, kick in and out, bass di and mic) but even with that said you should be doing it during tracking.
 
so, basically you'd have to put all mics up, record every sound, watch how the it all alignes and then adjust until it's good, over and over and over and over again?


Yes...How else are you going to learn....The point being is that set your mics up so that you have the least amount of phase as possible. Heck some times the phase you do get can work to your advanatge. Messing with things on a track level like you describe is a fix for something you should have done correct while tracking.

Also, just so everyone is on the same page. It is 100 percent impossible to get everything to be in phase. Even if you start adjusting it in your host.
 
That is a true conclusion - there could have just been a botched placement thing in there (Cubase tends to play with my timing on occasion) and that's why I'm trying to see what could have gone on.

I do, however, believe that phase anomalies can be very fun to play with (see the 'That Clayman Sound' thread) and haven't yet decided that phase oopsie means death. Being of a very curious and easily amused nature when it comes to things I can boil down to simple math or science I prefer to see what this can yield.

Jeff
 
By accepting that phase is omnipresent and infact the foundation upon which you can build a great recording/mix, you will never be stuck going, why does this sound boxy or why does this sound weak.

By utilising the physical nature of it, relative TIME and relative AMPLITUDE, you will always be able to be in charge of the situation. Once there is more than one signal in any system phase relationship between the two will be an issue, in reality even bad phase can be used to your advantage, but I haven't yet managed to make that one truly work.

I'll admit it is very trying to make a crap room sound good, but by addressing the problems you have one by one, and working around them with the tools you have, preamp gain, mic postion, mic pattern, brain, you can easily overcome them.

Time and Amplitude,