Hey all,
 
I'm working on the chorus of a song that's got melody, then harmony above and below that.  The chord changes are a bit tricky and my ears have finally adjusted so that I'm happy with the parts I've put down, then I doubled them to make them a little more solid.  Any recommendations on general best practices for mixing six vocal parts?  It sounds a little "flangey" to me at the moment.
 
Update and edit:  I adjusted the levels and that's helped.  The main melody is a little to the right with the double panned opposite and a good bit lower.  Same thing for the other two parts.  Not so flangey now but you can still hear everything.
 
Thanks!
 
Donny

Well the main vox should stay centred, the harmony vox's should be panned hard right and left.  You should EQ the main vox as you would normally, so for a main vox I would give a boost in the higher end cut the low end and then find the formant of the vocal and boost slightly.  You would do the opposite to the harmonies.  You could also try different reverbs on the harmonies and you could even chuck a little bit of chorusing on the harmonies too.  Just to thicken them and differentiate them from the main vox. 

Quadruple Post!! 

Double Post

Triple Post 

I use doubling and harmony a lot. 
 
My song Missing Person (2012) http://www.soundclick.com/bands/page_songInfo.cfm?bandID=443134&songID=11962059  has 10 vocal tracks.  2 singers with 1 main and 2 doubles and 2 harmonies each.
 
Footsteps has 5 vocal tracks http://soundclick.com/share.cfm?id=12048139
 
On any given project my average solo project has 5 vocal tracks in the end mix.  One is lead. the second and third tracks are also lead vocals that have been recorded live, never cloned. I keep the main vocal centered and about -6db but that level really depends on the way it sets in the mix. No FX in the track. Set up a vocal bus for reverb and EQ with all 3 leads going there. Pitch correct the lead vocal track and perhaps the doubles. The key to avoiding the flangy, phasey sound is to pan the 2 doubles wide..... 80% or more and keep the levels on them somewhere around -24db. This level will be set by trial and error but -24 is the starting point.  My aim is to let these tracks add a slightly fuller sound without being obvious. In other words, I can't really hear them in the main mix, but if I solo the vocal bus I can hear the doubles.
 
On the fourth and fifth tracks,,,,,, harmony tracks, these too are recorded independently. I pan them 50% to 80% in most projects and use -16db as the starting point.  I want the harmonies to "fill in" the vocal in the chorus and other places where I have them. Generally, I like a subtle harmony presence. You do have the choice to bring them up and create a "Statler Brothers" kind of harmony where all the harmony vocals have equal billing in the sound stage. I do not generally like to use that method of harmony but will do it as the song demands. I also will very often put the harmony tracks in their own bus and add a slightly heavier reverb to them. Nothing crazy, just something that moves them "back" a bit.
 
This is how I record vocals and harmonies. If you clone the lead, you will have exact duplicates of the tracks and more issues with comb filtering and for that reason I always record independent tracks. Even so, I have to be careful to avoid the comb filtering that is possible with three very similar tracks. Keeping the levels low on the doubles will help immensely in that area. It still gives you the fatter doubling but at more of a subconscious level and it leaves the main vocal sounding cleaner.  To see if you can hear the difference, simply use the GROUP function to group both of the doubled/panned vocal tracks and then as the song is playing in the parts where there is not harmony vocals, use the mute to mute and unmute the 2 doubles during playback.
 
Anyway.... that's how I do it. hope this helped you.

Thanks very much for the post.  Yes, in playing around with it, that's very close to what I ended up doing.  I believe I've ran into my first real encounter with comb filtering as well with four tracks of hand claps on the chorus.  I ended up using three instead of four and wide panning and level adjusting provides the fullness while avoiding cancellations I seemed to be getting?
 
Not sure if this is related at all, but is there a phase switch on the Sonar mixer?

Don,
 
I record and handle layers of vocals and harmony's almost identical to the way Herb does.  The only difference is in my panning of the main vocals.  Lead vocal is always dead center.  I usually do 2 other layers and pan them anywhere from 35% left and right up to 100%; it all depends on the song and how the other instruments are panned in the song.  I listen while panning and have them find their own space within the mix to sit.  It's all about competing space for me.  
 
I always record separate takes of vocals and sing them as closely to the lead vocal as I possibly can, for layers.  Most of the time I will tweak pitch and timing on those layers.  The key is to get them as close to the lead as possible, but still be different enough to sound good.  Again, it is an ear thing.  Levels are pretty similar to Herbs.  Depends on how much you want the layer or harmony to be noticed.
 
Another trick that I have used on a layered vocal is to automate volume so that the vocal layer is not always present, but comes in on certain phrases, groups of words or a particular word that you want to emphasize; that can be really effective.

Yes, Sonar has a 180 degree phase switch.
 
TO add to Mike's post and mine..... ENVELOPES..... simply put, I use them. Every track has a volume envelope and some have panning envelopes. It's just so easy to use them to get exactly what you want.

These observations are troubling:
 
    It sounds a little "flangey" to me at the moment.
and
    I believe I've ran into my first real encounter with comb filtering...
 
I believe your diagnosis is correct: you are hearing phase issues.
 
Normally, you'll only have this problem if you're mixing cloned tracks, because they're identical to the original except for delays or other effects that affect their phase relative to the original. However, I have also run into it when the tracks were not clones but were overly processed with Melodyne, V-Vocal or AudioSnap. 
 
One kludge that'll help disguise comb filter artifacts is to apply a chorus to the cloned tracks. Another trick is to roll off the highs on the clones. Only the primary track needs to convey consonants, so you can tolerate some loss of intelligibility on the reinforcing clone tracks. You might also be able to alleviate the phasey effect by nudging the clones a couple of milliseconds.