Maddocks,
In about 1999 I did some overall efficency calculations on a ethylene plant demethanizer using the method described in faq124-1340, with VLE per a high pressure modified RKS method that I had used in other parts of the plant. This was a trayed tower, and a good match was achieved at relatively low overall efficiency- meaning I needed few theoretical stages (relative to actual trays) between the top of the tower and upper feed point. I think 40% overall efficiency seems in the ballpark, although I can check my files if this is of interest.
While this result was fine for the energy study we were doing at the time, it was pointed out to me during review that there were some more "accurate" VLE correlations available for methane, hydrogen, and C2 systems. These correlations predicted more stages required, and therefore a higher overall efficiency. As a result of this realization, I caution against getting too serious about column efficiency numbers for which the full context is not known.
best wishes,
sshep