It is my understanding that a sizing run for the baseline building must be done. The COP (for OS:Coil:Cooling:DX:SingleSpeed) is to be determined via COP = 7.84E-8 × EER × Q + 0.338 × EER, where EER is from an appropriate table in Appendix G, based on Q, and Q is (presumably) the result of the sizing run. Both EER and Q are to be "net" (includes fan power), while COP is "gross" (excludes fan power).
Said Q, obtained from the sizing run, might be the OS:Sizing:System Cooling Design Capacity (I think this might be net, let's call it Qn). Alternatively, it might be the OS:Coil:Cooling:DX:SingleSpeed Rated Total Cooling Capacity (though I think this might be gross, let's call it Qg).
Either way, prior to the sizing run, the COP for OS:Coil:Cooling:DX:SingleSpeed is just a best guess. It is then to be adjusted via the above equation, using a suitable Q, which may be Qn or Qg. If one were to do another sizing run with this updated COP, the new Qn or Qg would be different from the original Qn or Qg, giving a different COP, and so forth. This would have to be repeated a few times to "converge", I suppose, if it even did.
I believe even for the great and almighty Appendix G, such iteration would go too far. Instead, is it correct to just use a ballpark Q after the first (and only) sizing run to obtain and update the COP? if so, what is to be used - Qn or Qg?
