Ok, I've understood phase issues and cancellation for a while...
Some of the things I try to do include keeping the microphones in a 1-4-1 relationship to their instruments... so if the mic is 1 foot from the guitar, then there must be 4 feet to the next mic that is 1 foot from another source like a vocal... more likely to be 1/2 that but the point remains...
So, now I have a vectrascope and I was trying to 'see' phase issues. I have a track that has a banjo player and another where she sings...
The banjo is pretty loud so it definitely goes in both microphones. I pan the tracks 100% Left and right to accent the phase differences in the vectrascope. The correlation meter shows better phase relationship when the phase is flipped but it makes her voice sound nasal.
Obviously, I'll go with what sounds good, but I wanted to ask if I'm generally going about it right. Her voice isn't going into both microphones at all equally. Perhaps that pushes the correlation down? Flipping the phase actually serves to cancel some of her voice allowing the correlation to look a little better since it is only correllating the banjo?
I tried to make a screen capture but got frustrated.