How big a mini-split does this room need?

The spine of this tool is the room air conditioner sizing table ENERGY STAR publishes, reproduced below (the first two columns as printed; the third column is our arithmetic)^[1]. It is the only government-backed area-to-capacity table we could find that also publishes its own adjustment rules, which is what makes it usable rather than merely quotable. It starts at 100 square feet and stops at 2,500. Ask for a room outside that and this site returns no number at all — extrapolating past the end of a source’s own table is how invented figures get born.

Read the third column before anyone sells you a rule of thumb. At the bottom of the table a room needs 50.0 BTU per square foot; at the top it needs 17.0. The published table asks for more BTU per square foot in small rooms than in large ones, so a single multiplier applied across the whole range oversizes the big rooms and undersizes the small ones.

Three of the worksheet’s controls — sun, people and kitchen — map onto ENERGY STAR’s four published adjustment rules (sunny, shaded, occupants, kitchen). Very sunny adds 10 percent^[2]; heavy shade takes 10 percent off ^[3]; each person past two adds 600 BTU per hour ^[4]; and a kitchen adds a flat 4,000^[5] for the cooking. Those are the source’s numbers, not ours.

The envelope setting is ours and we say so on the control itself. There is no insulation term anywhere in the ENERGY STAR table, and we could not obtain a published residential envelope multiplier we were willing to cite. So the three envelope settings — 0.85 for tight, 1.00 for average and 1.2 for leaky ^[7] — carry an assumed label everywhere they appear. Our own labeled adjustment. The ENERGY STAR table has no insulation term; a room whose envelope is much better or worse than typical will sit outside these multipliers. If a contractor has run a Manual J load calculation on your house, that number is a better guide than this one.

Ceiling height sits in between. The source states the eight-foot baseline and, for higher ceilings, suggests a unit with a higher CADR, but gives no capacity formula, so scaling in proportion to room volume is our arithmetic on a sourced baseline. That makes it derived: the input is published, the operation is ours, and the operation is printed on the methodology page where you can disagree with it.