Chicago and jdd: Both of your methods look like they work. I personally use the method jdd is using because I find I make fewer errors that way (in chicago's method I always forget if it's Sold/Snew or Snew/Sold). But as always, you should use whatever makes the most sense to you.
Cabby: looks like you've hit the nail on the head. I've always calculated per-unit problems on a phase basis too (it takes more calcs, but I have to remember fewer equations that way).
Cabby: looks like you've hit the nail on the head. I've always calculated per-unit problems on a phase basis too (it takes more calcs, but I have to remember fewer equations that way).