You’re great, I so appreciate you digging in on this with me. Honestly I called e comfort today and they ran their own calculation and told me 15k BTU… which I was suspicious about.
The eComfort tool is downright TERRIBLE, guaranteed to oversize (often by 2x or more!). Note that they call it a "sizing tool" rather than a "load calculation tool" (which it absolutely isn't.)
I’ll try the tool again and see if I can get the room numbers. It has a weird requirement where the rooms you add in need to add up to some percentage (I think 80%) of the house square footage. I guess I could just trick it and do a fake room that is the house square footage and so the addition as the second room.
It's been a few months since I've used the tool, it was spot-on for load numbers on an ~800 square foot house I was consulting on in the Boston 'burbs, coming within 10% of the numbers produced by an engineer using Wrightsoft's professional tools.
No real load tool is as easy to use as the eComfort tool- they need real data on the R-values and construction types since they are shooting for better accuracy. If the BetterBuiltNW tool seems too klunky for you, the freebie online
CoolCalc or
LoadCalc tools can also get you pretty close. LoadCalc isn't quite as flexible as the BBNW or CoolCalc, in that it won't let you adjust U-factors, and the defaults on air leakage are pretty high. I've found that it often overshoots reality by 25% or more. CoolCalc is more adjustable, but (like LoadCalc) has lots of ways to accidentally have your thumb on the scale, so for newbies unaccustomed to using aggressive inputs (as instructed in Manual-J) it can also oversize by quite a large number. The BBNW tool is good because the defaults are all fairly aggressive, yet it allows you to adjust U-factors when if needed.
If you have to guess what the R-values are, use the highest R that is possible or likely given the age and construction quality. eg: If it's a 2x4 wall, unless you KNOW it's only R11 or R8 econobatts in the cavities assume it's R13. Similarly, set all of the air tightness assumptions for the house or ducts the tightest the tools allow.
Are the windows low-E double panes? If yes, if you don't have the actual manufacturer's specs, assume an U-factor of no greater than U0.35, with SHGC (solar heat gain coefficient) of no greater than 0.30. If you are not sure if they're low-E, look at the reflections of a candle, lighter, or LED flashlight. You'll see 4 reflections, inner & outer surface reflections from both panes. If one of the reflections is a different color than the others, it's a low-E window. Pretty much all double pane skylights built since the mid 1980s are low-E, some with ultra-low SHGC.
I’m all for getting better equipment if the value is there. I’m confused, are you saying an oversized Mitsubishi will still be efficient to minimum modulation? I’m guessing the comfort and other things will still not be as good?
The half-ton and 3/4 ton Mitsubishis don't modulate down as low as the LGs, but they modulate low enough for good comfort. With a minimum output of 1600-1700BTU/hr @ the ARHI test temps (47F and 17F) vs. 1023BTU/hr for the LGs, the difference is roughly the BTU equivalent of one active adult walking around the room, or three humans fast asleep. Both will be running nearly 100% duty cycles under almost any real load, but the Mitsubishi would be using only 1/3-1/2 the power under lighter load conditions, and measurably better during shoulder season average conditions. Without the load numbers and better weather data for your location it's hard to say what that would add up to in annual kwh (or $$$).
I see what you’re saying, that the btu rating is not the maximum heat transfer rate or there’s a performance curve and that rating is based on some heat transfer rate capability at some specific temperature? I’m with you, I don’t want to over size. My last house had an over sized system and it drove me crazy with the short cycling (fortunately the air was dry, we lived in LA). I knew it was over sized when it was getting installed and I just stayed out of the way as the “professional” insisted that a 1500sqft house equals 4 ton unit (that was his extensive load calculation).
The manufacturers bury the "extended temperature capacity tables" in the manuals. The AHRI submittal sheets spec some minimum & maximum capacities at the ARHI test conditions as well as the efficiency at the "rated" or "nominal" capacities, and sometimes the capacities at a couple of other temps. The ARHI test also requires that the inverter frequency be locked at 60 hertz for those tests at those temperatures ( rather than free modulating), so the true efficiency is often slightly better during normal operation than specified. The short-sheet specs on the
NEEP pages are from data submitted by the manufacturers under normal operation, but the heating capacity & efficiency hit from defrost cycles is
not included, so under some conditions (such as rime-icing fog) the capacity won't be nearly as high as specified, event though it will be pretty close under most conditions. But the NEEP pages show the COP efficiency at multiple modulation levels, rather than just at the AHRI rated modulation levels.
Silly somewhat unrelated question: would it be bad to the equipment if we run the AC with windows open? Wouldn’t it just be in a continuous cycle?
Bad for the
equipment, no, but it's bad for efficiency, and under some conditions bad for comfort. Running with windows open in MD would add a significant amount of
latent load (unlike in LA or Seattle) in summer, and a lot of the time you would end up cool but sticky/clammy rather than cool & dry. (That's easy to fix- close the windows and run the minisplit in "DRY" or "DEHUMIDIFY" mode for awhile.)
If you need or want ventilation, running a heat recovery ventilator (HRV) or energy recovery ventilator (ERV) would be preferred, since it's a controlled rate not much affected by wind, and most of the heat in the air transfer is retained rather than wasted. (There are ductless single-room HRVs out there that are comparatively easy to install compared to a ducted whole house ERV/HRV. I can point you to a few if you're interested.)
Mini-splits that modulate as low as the LGs or the cold-climate Mitsubishis would have pretty much continuous operation with or without the windows open.