You need a mid / side matrix.

I can't tell you specifically how to do it in pro-tools (I'm a Tracktion user: its dead easy in Tracktion!) but the process is really quite simple.

1. Add your left and right signals together to create a mono sum. This is your "mid" or "sum" channel.

2. Invert the polarity of the right channel and mix this with the left again. The result is your "side" or "difference" channel.

3. Route the mid channel to the mix bus, panned centre, with 6dB of attenuation.

4. Split the side signal down two channels. Route both to the mix bus but pan one hard left, then invert the polarity of the other and pan it hard right.

You can now adjust the stereo width of the mix by adjusting the level of the "side" signal: any higher than -6dB will widen the mix, any lower will narrow it.

And when you get it to its widest state, it will be 180° out of phase and will completely disappear when your stereo signal is collapsed to mono. Also frequently referred to as a "vocal removal" technique for your own home made karaoke from original recordings.

Getting wider at 51 1/2.... I mean... WISER!
Ms. Remy Ann David

Especially dont do this if.....

Dont try that at home if u use mid-side on the mix tracks you'll lose the mid-side part of the mix quick....
However, my bro showed me a way to create a user defined stereo image effect.
the Idea is similar to IIRs
I He would create STEREO premaster mixdown Then he would mix(premaster) it down again using six copies of the stereo tracks....

II He used tracks 1/2 for the main unfiltered stereo mixdown and turned down the tracks master fader so it was peaking around -6 or something of that magnitude

III The idea for the four remaining tracks (3-6) is to pan 3/4 hard left and then pan 5/6 hard right. Then invert tracks 4/5 and this creates a midside with a balanced approach.

tracks 3/4 (left channel panned hard left) are one inverted mid-side
stereo and a balancing normal stereo track turned down to -9 to -12 or so

the same was done for tracks 5/6 (right side) turned down to -9 to -12

Basically, this is very much like using a plugin to create a stereo image, but I have to tell ya it sounds really good still. When I use the S1 waves plugin it sucks alot of low end out of the mix, leaving my mix sounding hollow.
My new concept towards enhancing stereo images is mostly being done in my mixing tho. I wonder how u can use this in mastering????.....
I also use a spectral analyzer on every instrument individually as to see the midside part of the graphic.
Well I hope any of this makes any sense at all, its all experimental for me right now, but its interesting as hell!!!
:-? its not enough that we have to play, record, mix, master we also have to create a fake room that sounds real :P

Using WAVES S1 for this.

Waves S1 plug in does this for you.

Put the MS- S1 splitter up on the top part of the insert tree of your DAW
Put a stereo plug(s) in next
Put the MS-S1 combiner up last to combine the signals again.

Now the left side of the stereo plug(s) are the mid or middle of the mix and the right is the sides. If you take a little of the mid frequencies out of the mid (left) side of the EQ and put them back in on the sides (right) the soundstage begins to widen.

peace
Phil