So I've paid for one HVAC manual J, done two online (coolcalc and loadcalc)

They basically all say the same. Numbers came in around 9000-9500k BTU Cooling, 15k heating.

Design temp is 15F winter, 87F summer. Mountains of North Carolina so yes, winter is colder than summer is hot.

Looking at Fujistu 9RLS3 (the 9,000 BTU one) or 12RLS3 (the 12,000 BTU)

Looking at Fujitsu Engineering tables for Cooling Capacity on the 9k it says on the line of 87F outside, 75F inside:

9300 BTU Total Capacity, 8000BTU Sensible Heat Capacity

For the 12k the line says:

12400 BTU Total Capacity, 10140 Sensible Heat Capacity

If anyone cares I'm looking at the RLS3H manual here, though I plan on getting the RLS3 (non-H) as below -5F is a rare occurrence (yes I have backup plan) - https://portal.fujitsugeneral.com/files/catalog/files/DTR_AS115E_01--ASU9-15RLS3_H1.pdf

So I don't know. Sure I could look at the line for 70F instead of 75F but look, it gets to 86F like from one week out of summer. I can deal with oh dang it's 77F in here and not 72F for a week four hours a day.

But I don't know if I'm looking at Total, or Sensible. I don't want to oversize so I can be in a 65F room when it's 90F out for one day every 12 years. But I don't want to be undersized.

If I look at TC the 9K seems about right. If I look as SHC then do I need the 12K?

And I have no idea what sensible HEATING capacity means in cooling table. Is that related to cooling, or does that mean if I need to run heat when it's 75F outside that's what I get for heat? That makes no sense as that value is lower than heat output when it's 5F outside. So that can't be it.

Question all boils down to so I need the 9k for summer, or the 12k for summer.

Either have plenty of the 15K BTU's I need for winter at 15F outside, and enough to heat on the oh-shoot-it's-2F outside put a spaceheater in the bedroom for 3 nights of the year.

And I certainly don't want to just go "gosh, the 12K is only $100 more, why not??"

I'd prefer right-sizing over just-in-case-fudge-it-sizing. If 12K is right, then right-on. But I don't want to decide on some "better safe than sorry" gut feeling then scratch my butt.

The 9K and 12K models appear to be the same in every way except something minor (perhaps gearing on the motor). Both modulate down to the same min output. So I'd get the 12K to leave 13% more cooling margin for sun, cooking, parties, recovery from setback, etc and 33% more for cold days (check the max amp draw). This is not a case where the heat exchanger is over-sized or the "larger" unit will cycle more frequently.

I agree with jonr. The usual reason not to oversize has more to do with the minimum than the maximum, and the minimum is the same.

I think that when they list "sensible heat" in the cooling section, they are talking about removing that heat from the conditioned space, in other words cooling.

That's "Sensible heat capacity", not "...sensible HEATING capacity...", which is the amount of air-temperature heat it's removing.  Comparing the total capacity to the sensible capacity, the difference is an indication of it's latent-cooling (dehumidification) it is delivering when running at max speed under those conditions.  From that you can derive it's worst-case latent / sensible ratio.   When operating below it's maximum blower speed the latent/sensible ratio increases- it dehumidifies better.

eg: At your 1% design temp of 87F and an indoor temp of 75F dry bulb/63F wet bulb , which is ~52% relative humidity (pretty comfortable conditions for most people) the 3/4 tonner delivers 9310 BTU/hr total cooling, 8010 in sensible cooling.  The difference is 1300 BTU/hr of latent cooling. The latent/sensible ratio is only 1300/8010, or about 16%. The 1-tonner can deliver 12,410 BTU/hr total, 10,180 BTU sensible, for a latent cooling of 2230 BTU/hr, a latent to sensible ratio of 2230/10,180 = 22%.

Under other conditions the 9RLS3 can deliver a total capacity of about 12,000 BTU/hr even when it's 95F outside, so it's really not likely to become a comfort problem.  The submittal efficiency test conditions for cooling are 80F dry bulb /67F wet bulb indoors , and 95F dry bulb / 75F wet bulb, and they tested its efficiency modulating at 9000 BTU/hr total, not sensible.  Most of the time it's running at lower speed for a higher latent/sensible ratio, keeping the place dry. Unless you have a high ventilation rate or lots of indoor moisture sources (don't be boiling pasta on the stove all day with the lid off when it's 87F outside) most of your load will be sensible cooling by the time temps actually reach 87F.

In the Manual-J they should have broken out the sensible and latent cooling numbers, which they added up to the 9000-9500 BTU/hr number.

But if you need 15,000 BTU/hr  of heating @ +15F outdoors you won't always get that out the 9RLS3 into a 68F room under all conditions.  The capacity tables show an output of 16,000 BTU/hr max at in to a +70F room at  +14F dry bulb/+12F wet-bulb (66% RH), but that's it's maximum, not including defrost cycles.  At higher outdoor wet bulb conditions (say, when it's snowing or rime-icing out) the max will be higher still, but it'll be spending even more time in defrost cycles.  The 12RLS2 delivers 17,400 BTU/hr max under those same conditions, which gives you a bit of margin, some room for defrost cycle time.  That alone is argument for using the 12RLS3 rather than the 9RLS3.

12RLS3 ordered

Was going to get the RLS3H but going to spend the money saved towards a better backup heat source. We're on a mountain and bad conditions means maybe not going anywhere, plus below -5F isn't super common here.

Quick update

12RLS3 installed and running for a few months.

Two nights ago it got down to about 3F outside.

Temperature inside apartment at wake up, 8am, was 65F

Thermostat was set to 72F

So as far as I'm concerned, that's freaking perfect.

Was that an actual wired thermostat, or was that the setpoint programmed by the remote? The room air stratification when it's actually cold out causes an offset between the average room temp measured by a thermometer 3-5 ' from the floor, and the incoming air to the mini-split head at the top, so the mini-split "thinks" it's warmer in the room than it actually is. A 7F offset when it's 3F outside is not unusual. If a 65F room temperature is ever considered a problem, unless the thing was running flat-out (probably not), you can probably just program the setpoint a few degrees higher when you know it's going to be single-digits or colder overnight.

I found that running a ceiling fan or two on 'Low' in reverse nets me about 3 more degrees without changing the remote. We have 9-1/2' ceilings.

Posted By jdebree on 13 Jan 2017 05:46 AM
I found that running a ceiling fan or two on 'Low' in reverse nets me about 3 more degrees without changing the remote. We have 9-1/2' ceilings.

At what outdoor temperature?

Just using head unit.

We kind of figured that out when it was about 15 outside so cold nights it's been set about 6f higher.

Maybe a wall thermostat is in its future.

We've been down in the teens a couple times this winter. Our house is about 1400 square feet, and we generally just run the bigger 12K Mits HyperHeat unit. We have a 9K in the master bedroom, but only use it when it's down in the low single digits; rare in our area.