Eff values are better now for new energy efficient motors, but if you don't know, 0.8 is a "safe" bet in that if you use it, you will end up estimating a bit high as opposed to a bit low. But large motors, i.e. above NEMA frames (roughly over 250HP), are typically 90-95% efficient, so I use 90% most of the time. The best thing is to get the efficiency values from the motor nameplates or from the suppliers if it isn't there.The first thing I noticed is that you are missing the term for the motor's efficiency, which is due to hysteresis losses in the core and windage of the spinning rotor. The estimate that I use for smaller motors is an efficiency of about 0.8, but this would likely increase with larger motors, and you might find it on the motor nameplate.
P = I * V * eff * pf * 1000 = KW
KWhr = KW * time
Cost = KWhr * rate($ / KWhr) = $
D'oh! I just got caught doing the same thing...Oops. I am so accustomed to using this when converting hp to current draw that I think I made a mistake by including the efficiency in this equation. I think that term comes in only when dealing with the shaft output power.
Another Edit: Also, if your amperage is per-phase, then don't forget the squareroot-3 term.
Eff values are better now for new energy efficient motors, but if you don't know, 0.8 is a "safe" bet in that if you use it, you will end up estimating a bit high as opposed to a bit low. But large motors, i.e. above NEMA frames (roughly over 250HP), are typically 90-95% efficient, so I use 90% most of the time. The best thing is to get the efficiency values from the motor nameplates or from the suppliers if it isn't there.
Energy efficiency has nothing to do with calculating the cost of electricity for an existing system.