[Solved] But why?

I"m mixing a song my son and I put together. My son did the vocal. I did the usual trick of duplicating his vocal track, shifting one track ahead 20ms and the other back 20ms. One channel was panned full left, the other full right. Nothing new.
 
I noticed though that the vocal was predomintally on the left speaker. I checked everything and all looked like it should. I then shifted the two signals so they were both lined up. Now the vocal was centered. I shifted the orignal back 20ms and the copy ahead 20ms. The vocal was now mostly on the right speaker.
 
Anybody have an explanation for this phenomena? Does the brain hear the first track and ignore essentially what is an echo? Sorry if this has been discussed before and I did Google it and could not find anything.
 
Walt

As it goes, I think the thing you're calling "the usual trick" is somewhat incomplete. There's a technique - which I personally don't care for - for thickening a vocal where you what you describe, but you also pitch shift the vocals panned to the side by a few cents, one up, the other down.
 
This creates (supposedly) enough difference between the two sides to create a sense of width. When the two sides aren't different enough, there's no sense of width. You have mono, effectively. And the precedence effect brought about by the delay means it will skew to one side.
 

Well.. everything in its place. It can be cool sometimes. But I find it to be a bit unsubtle. Better as a deliberate special effect, than a general thing, I think.

I'm with Craig on this. Get your son to record a second track and then pan the 2.
 
I've always underwhelmed by any sort of fake doubling techniques. But make sure he sings it as close to the original as possible and watch the timing!! (More important than pitch in my estimation)

40 ms difference between the copies sounds  much to me. Anyway, I never hard pan the vocals, I prefer the Lennon-method. The result sounds more natural to me, which suits the style of songs I make.

isn't this essentially throwing the vocal out of phase ?

The Haas effect is the brain figuring out the left hear heard something earlier than the right, so the sound came from the left. It works over delays less than 1ms, which is a foot or less.
 
20ms at the speed of sound is the length of a hall, so this is not the cause.

rbecker
 
Also, the Haas effect has a delay max of up to 35 ms, so that is close to where the OP is at.
A simple, definitive test the OP could do is switch sides and just see if the effect reverses.

I had tried that when I was trying to figure out what was happening and yes the effect reverses.
 
Walt

Bristol_Jonesey
I'm with Craig on this. Get your son to record a second track and then pan the 2.
 
I've always underwhelmed by any sort of fake doubling techniques. But make sure he sings it as close to the original as possible and watch the timing!! (More important than pitch in my estimation)

I appreciate all the comments and have Googled and studied the "Haas effect". Now I know. Thanks.
 
Now for the rest of the story. Normally I do the vocals when my son and I put together some material. Some times I do doubling of the vocals, but actually sing the second part. This was the first time my son did a vocal (plus harmony) on one of our recordings. It's a cover for "More today, than yesteday" by Spiral Staircase. He also did two Trumpets, two Tenor Saxes, and two Alto Saxes. When he left, I added Guitar, Bass, Organ, and Drums.
 
After doing a rough mix, I realized that his vocal needed to be fattened up a tad. I couldn't really ask him to come over and record a second part since he lives in Nagoya Japan and was visiting us last December. He will be renting some recording space in his home town so we can collaborate in the future, but for the time being, what I heard is what I had.

 
Now that I understand what was happening, I will try some of the suggestions in this tread. Any more ideas would be appreciated.
 
Thanks to all,
 
Walt

I don't think doubling is cheesy.  I like it.  I normally do this:
 
1.  Sing the vocal twice.  Use Melodyne to get them close but not exact in pitch and timing.  Pan them 40% to 60% left and right.  I think it is better if one isn't always ahead and one always behind in time, because that sounds less natural.
2.  You can use Craig's trick to simulate singing the track twice.
3.  I also have the Abbey Road Reel ADT plug in and it is really good.
 
Psycho-acoustics say that the track you hear first will always seem louder.
 
PS Visit my website to hear what the doubling sounds like.  I don't think it sounds as heavy as what some people do, but it is all relative.

konradh
I don't think doubling is cheesy.  I like it.  I normally do this:
 
1.  Sing the vocal twice.  Use Melodyne to get them close but not exact in pitch and timing.  Pan them 40% to 60% left and right.  I think it is better if one isn't always ahead and one always behind in time, because that sounds less natural.
2.  You can use Craig's trick to simulate singing the track twice.
3.  I also have the Abbey Road Reel ADT plug in and it is really good.
 
Psycho-acoustics say that the track you hear first will always seem louder.
 
PS Visit my website to hear what the doubling sounds like.  I don't think it sounds as heavy as what some people do, but it is all relative.

The cheesy statement was a play on Cactus Music's post. I sometimes double my vocals, but I sing them twice as in your number 1 option. I just didn't like the way it was turning out with the way I was doing it. I'll look in the the ADT plug-in you mentioned for possible future applications.
 
BTW I visited your WEB site a couple months ago and I'll never go back...You're too good and I get depressed! I am getting better though. I listened to stuff I did 3 years ago and said to myself, "What was I thinking!"
 
Walt

jpetersen

konradh
1.  Sing the vocal twice. 

That we have been told is apparently not an option.

 
True, for this song.
 

jpetersen
 
But if there is a chorus sung identically elsewhere, you could copy them under each other and try Vocal Sync.

 
There are two choruses. We did one and copied it for the 2nd. My son's range went into compliance trying to get to the high notes, so after several takes on the 1st chorus, we caved and copied it.

jpetersen
Leave verses undoubled for variation.

The verses needed the most attention, since it had a fairly soft background. He sang harmony in the chorus so that helped out along with 6 brass instruments.

jpetersen
But what I would really like to know is what was causing the effect?

As somebody pointed out, it's the Haas effect.

jpetersen 
If you go close to the speaker that is apparently "dead", do you then hear something?...

The speaker was not dead, just reduced in perceived volume such that the vocal seemed located between Center and Hard Left.
 
Walt
 

John T
One of the things I dislike, personally, about a lot of these techniques, is the way they make the positioning of the vocal kind of ambiguous. By having vocal dupes panned around, you often end up with a vocal that doesn't seem to sit anywhere in particular in the stereo field.

this is very true and a great point.
this is why I never, ever pan double vocal tracks.
for the exact reason you mentioned, it doesn't sit in the stereo field correctly and it tends to either bury other instruments / or stand out way too much.
 

yorolpal
If you use Melodyne you need to randomize pitch, formant and timing to achieve a great double track sound.

Thanks I'm going to try that. I'm traveling right now, but will check it out when I get back. I was going to purchase Abbey Road Reel ADT Plug-In. Do you know if this method gives as good results?

Walt