Yet Another Mini Split Sizing Question

[Elton Noway]

I requested bids from four different professional HVAC contractors in my area, to install a mini split system in our sunroom which is attached to the back of our house. The prices quoted from all four are within a hundred bucks of each other that isn't part of my decision. (I'm guessing they all know their competition). My confusion is they don't agree on the capacity of the unit I need.

The room is relatively small... only 12' x 15' 6" ( 186sq ft). The side of our house makes up one of the 15' walls. The ceiling is pitched following the roof line. 8 ft at the edge sloping up to 9' 6" at the peak. The ceiling is insulated with R30 fiberglass batts. The floor is insulated with R30 Roxul (rock wool batts). The walls are the weak link because they are pretty much glass. Although they aren't floor to ceiling, the space above them is solid wood, as are the window partitions and the R Value of lumber is negligible. All the glass is double pane Low-E argon insulated... but the sunroom faces west and in the summer from noon on it pretty much gets full sun. Location: Raleigh, North Carolina

Long story short... all the vendors claimed they did load a calculation. One vendor said since its only 186 square feet a 9K BTU would be more than sufficient. Two of the vendors said I needed a 12K unit and one said I needed 18K! I want to make sure the unit I choose will have no problem maintaining a comfortable temperature without struggling to keep up... but don't want to oversize the unit either. Although as I understand it thanks to variable capacity... over sizing an AC unit doesn't apply to the mini splits. I understand the bigger the unit the faster the pulldown but I was concerned a bigger unit might not run long enough to remove the humidity.

NET Based on my above specs... any suggestions to which of the suggested systems might be appropriate for our sunroom? NOTE: Attached picture (interior) was taken under construction (pre sheet rock and flooring.)

 

[Dana]

Are you sizing it for the heating load, the cooling load, or both?

If cooling, which direction does the long side of the room face? You clearly have some shading factors to work with too.

The square footage of the room doesn't much matter, but the square footage of the glass (and orientation, from a cooling perspective) DOES matter, as do the U-factors and SGHC of the windows, R-values of the walls & ceiling, etc.

While I agree a 3/4 tonner will probably do it for both heating and cooling, there is no substitute for doing some semi-real math on it, not a square feet x BTUs or some crude garbage-in = garbage-out with pro-forma numbers in a Manual-J tool using conservative rather than aggressive assumptions on shading factors, etc. You might try to squeeze some reasonable numbers out of this online tool if the options are close enough to what you have, or go at it a little deeper into the weeds with these cooling and heating calculation tools, but you'll need something better if you need to factor in shading factors.

Clearly if the cooling load comes in under 9000 BTU/hr at Raleigh's 1% outside design temp of 90F outside, 75F inside using the cruder tools even without factoring in the shade anybody's 3/4 tonner will have it covered for cooling. Better class 3/4 ton mini-splits can deliver 12,000 BTU/hr of cooling at 95F outside/80F inside, even though their "rated" or "nominal" cooling (the modulation level at which it's efficiency is tested) is only 9000 BTU/hr.

If you want to calculate the thermal bridging R-values, assume R1.2/inch for the framing lumber. Since you have the pre-finished pictures to work from you should be able to calculate the square footage of stud edge, and go from there. Add another R1 -R2 for the sheathing, siding & air films.

 

[Sponsor]

 

[Elton Noway]

Are you sizing it for the heating load, the cooling load, or both?

If cooling, which direction does the long side of the room face? You clearly have some shading factors to work with too.

The square footage of the room doesn't much matter, but the square footage of the glass (and orientation, from a cooling perspective) DOES matter, as do the U-factors and SGHC of the windows, R-values of the walls & ceiling, etc.

While I agree a 3/4 tonner will probably do it for both heating and cooling, there is no substitute for doing some semi-real math on it, not a square feet x BTUs or some crude garbage-in = garbage-out with pro-forma numbers in a Manual-J tool using conservative rather than aggressive assumptions on shading factors, etc. You might try to squeeze some reasonable numbers out of this online tool if the options are close enough to what you have, or go at it a little deeper into the weeds with these cooling and heating calculation tools, but you'll need something better if you need to factor in shading factors.

Clearly if the cooling load comes in under 9000 BTU/hr at Raleigh's 1% outside design temp of 90F outside, 75F inside using the cruder tools even without factoring in the shade anybody's 3/4 tonner will have it covered for cooling. Better class 3/4 ton mini-splits can deliver 12,000 BTU/hr of cooling at 95F outside/80F inside, even though their "rated" or "nominal" cooling (the modulation level at which it's efficiency is tested) is only 9000 BTU/hr.

If you want to calculate the thermal bridging R-values, assume R1.2/inch for the framing lumber. Since you have the pre-finished pictures to work from you should be able to calculate the square footage of stud edge, and go from there. Add another R1 -R2 for the sheathing, siding & air films.

Thanks for the quick reply and great response! Yes, the idea is to size for both heating and cooling. As to cooling, the long side faces west and catches the afternoon sun. You are correct about factoring in shading. I do get some shade but not until around 4pm in the afternoon when the sun drops below the tree line. From about 11am till 4PM it just bakes in the sun. After reading your response now my concern is "none" highly rated and recommended contractors who came out to size the job ever mentioned orientation or shading, asked about the SGHC value of the windows or type of insulation in the ceiling or floor. They may have made a mental note about the shading and being familiar with the neighborhood also made a mental note on orientation but I would have liked them (even one of them) to ask the questions you did. I almost get the impression they are guessing... i.e., being familiar with the efficiency of mini split systems and just looking at the size of the room and saying to themselves... "This will work". I'm beginning to wonder if I should contact some other firms for quotes.

Anyway... in the meantime I'll take a stab at the tool you provided the link for. Thanks! I did a quick glance but I'm afraid some are over my head and I'll be guessing at some of the answers.

 

[Dana]

In the middle of the day the sun is high in the sky. At high sun angles the overhangs shade part of the window, and the exterior surface of the window reflects most of the heat due to the oblique angle. The killer would be if there were no shade after 4PM, when all of the window is in full sun, with a more direct angle.

At mid-day the west facing windows don't get any sun at all. But since it's pretty shaded by the landscape from 4PM on you don't get the kind of heat load spike that you would get from low angle sun with no shade.

Do the math a couple of different ways, but you're almost certainly within the capacity of a decent 3/4 tonner such as the Mitsubishi -FH09NA or Fujitsu -9RLS3.

The heat load is going to be dominated by the windows, and it looks like you have on the order of 200 square feet of window (?) At 20F outdoors (your 99% outside design temp) and 70F indoors that's a 50F difference, and even with crummy code-minimum U0.35 windows the window losses would be U0.35 x 200' x 50F= 3500 BTU/hr. The wall, floor and ceiling losses won't add up to more than 1000 BTU/hr, and the infiltration/ventilation even less, but let's call it another 1000 BTU/hr as a worst-case. Even worst-casing you're at less than 6000 BTU/hr of heat load, and the average wintertime load is probably less than half that. Run anybody's load tool, see where it comes in. Keep in mind that any decent 3/4 ton mini-split puts out more than 10,000 BTU/hr @ @ +17F (a standard HSPF efficiency test temperature), some more than 10,000 BTU/hr.

Do the cooling numbers a bit more formally than that. If the cooling load is within the capacity of the Mitsubishi -FH09 it would be preferable to most others, since it can modulate down to 1600 BTU/hr output, whereas the Fujitsu 9RLS3 only drops back to 3100 BTU/hr before it starts cycling on/off. Both efficiency and comfort is better when it's modulating, running nearly continuously rather than cycling on/off. The 3/4 ton LG Art Cool Premier can throttle down even lower, to about 1000 BTU/hr, and may also be a candidate. It's usually more value-priced, but it's not as quiet (but not loud or cheap), and isn't as well supported as Mitsubishi or Fujitsu in many areas (including mine.) It's max-cooling at 95F outdoors/80F in, is nearly 13,000 BTU/hr.

HVAC contractors rarely if ever do real load calculations- it takes more time and expertise than they have, and that only adds cost. Even those that do run some numbers through a load tool have a lousy record on getting it right. It's (sadly) unrealistic to expect good load analysis out of them, especially on a project this small, and getting more quotes from other contractors isn't going to solve it. Figure out the numbers yourself, decide what models fill the bill, then solicit quotes for EXACTLY that equipment, and let them know you're not interested in upsizing or going with a different model. That makes the process quicker and easier for both the contractor and for you.

 

[Elton Noway]

In the middle of the day the sun is high in the sky. At high sun angles the overhangs shade part of the window, and the exterior surface of the window reflects most of the heat due to the oblique angle. The killer would be if there were no shade after 4PM, when all of the window is in full sun, with a more direct angle.

At mid-day the west facing windows don't get any sun at all. But since it's pretty shaded by the landscape from 4PM on you don't get the kind of heat load spike that you would get from low angle sun with no shade.

Do the math a couple of different ways, but you're almost certainly within the capacity of a decent 3/4 tonner such as the Mitsubishi -FH09NA or Fujitsu -9RLS3.

The heat load is going to be dominated by the windows, and it looks like you have on the order of 200 square feet of window (?) At 20F outdoors (your 99% outside design temp) and 70F indoors that's a 50F difference, and even with crummy code-minimum U0.35 windows the window losses would be U0.35 x 200' x 50F= 3500 BTU/hr. The wall, floor and ceiling losses won't add up to more than 1000 BTU/hr, and the infiltration/ventilation even less, but let's call it another 1000 BTU/hr as a worst-case. Even worst-casing you're at less than 6000 BTU/hr of heat load, and the average wintertime load is probably less than half that. Run anybody's load tool, see where it comes in. Keep in mind that any decent 3/4 ton mini-split puts out more than 10,000 BTU/hr @ @ +17F (a standard HSPF efficiency test temperature), some more than 10,000 BTU/hr.

Do the cooling numbers a bit more formally than that. If the cooling load is within the capacity of the Mitsubishi -FH09 it would be preferable to most others, since it can modulate down to 1600 BTU/hr output, whereas the Fujitsu 9RLS3 only drops back to 3100 BTU/hr before it starts cycling on/off. Both efficiency and comfort is better when it's modulating, running nearly continuously rather than cycling on/off. The 3/4 ton LG Art Cool Premier can throttle down even lower, to about 1000 BTU/hr, and may also be a candidate. It's usually more value-priced, but it's not as quiet (but not loud or cheap), and isn't as well supported as Mitsubishi or Fujitsu in many areas (including mine.) It's max-cooling at 95F outdoors/80F in, is nearly 13,000 BTU/hr.

HVAC contractors rarely if ever do real load calculations- it takes more time and expertise than they have, and that only adds cost. Even those that do run some numbers through a load tool have a lousy record on getting it right. It's (sadly) unrealistic to expect good load analysis out of them, especially on a project this small, and getting more quotes from other contractors isn't going to solve it. Figure out the numbers yourself, decide what models fill the bill, then solicit quotes for EXACTLY that equipment, and let them know you're not interested in upsizing or going with a different model. That makes the process quicker and easier for both the contractor and for you.

Hi Dana... THANKS for taking the time to reply and providing so much insight. I never stopped to consider the "angle" of the sun versus direct. Yep... pretty much from 4PM on I have the benefit of many large shade trees. Unfortunately I don't know the SHGC numbers for my windows. At the time I installed them they were called Pella Proline... but that was 14 years ago! My paperwork shows they are Low E glass, Argon Filled dual pane... but no SHGC or U Values.

Glass - 96sq ft facing North
Glass - 96sq ft facing South
Glass - 124sq ft facing West
16" overhang above all windows window/door glass
Top of all all windows /doors 12" below overhang

Wall surface Total area - sheathing and lumber (excluding glass & the side of house where the sunroom is attached to) is 105 sq ft with no insulation Ceiling Surface area: 210sq ft R30 insulated. Anyway... using the link you provided to loadcalc.net... I entered all the pertinent information to the best of my ability. Here's what it came up with.

So... if my total BTU's cooling is 8136 I believe I will be pushing the upper limit of a 9K unit. Correct?
Just FYI: At this minute the outdoor temp is 94F... the temp in my sun-room with ceiling fan running is 90F

Units being quoted as follows:
The contractor with the 9K unit suggested: Mitsubishi MUZ-HM09NA with MSU-HN09NA
One 12K contractor suggested: Mitsubishi MUZ-HM12NA with MSZ-HM12NA
One 12K contractor suggested: : Mitsubishi MUZ-GE12NA2 with MSZ-GE12NA-9
The 18K contractor recommended MUZ-HM18NA with MSZ-HM18NA

Out of curiosity since the install price for the 12K units was the same, I did a compare between the two units. The MUZ-HM12NA is 18 Seer... the MUZ-GE12NA is 20 Seer. The MUZ-HM12NA is being quoted by a MEUS Diamond contractor so it has a 12 year warranty on compressor and parts. Whereas the MUZ-GE12NA is not a Diamond contractor so it only has a 10 year warranty. Since the price is the same is one unit a better option over the other? i.e., model vs warranty vs seer.

 

[Dana]

Most 9K minisplits can deliver at least 10,000 BTU/hr of coolking at maximum speed, some can deliver more than 12,000 BTU/hr of cooling, so you are in no way going to be pushing it too hard. The 9000 BTU/hr number is the "rated" sizing, the modulation level at which they did the SEER efficiency testing.

The HM09 is also good for 10,000 BTU/hr cooling (@ 95F outdoors, 80F indoors), 7200 BTU/hr heating (@ +17F out, 70F in) per the submittal sheet.

That means it has about a 1.25x oversizing factor for you cooling, a 1.15x oversize factor for heating. It couldn't be more ideally sized for your loads from a maximum output perspective. The 1-ton would be overkill.

With online tools like that you can't tweak in most of the particulars for higher accuracy, and it's more likely than not to be overstating reality by 10-15%. It looks like you didn't use a shading factor for the windows other than overhangs(?), which means the cooling numbers could be quite a bit overstated.

When comparing units, look at the minimum modulation levels relative to your load, not just capacity and SEER/HSPF, since that affects both comfort and as-used efficiency. The minimum modulation levels on the HM09 are pretty high- 3800 BTU/hr minimum cooling is nearly half your design cooling load, and the 4500 BTU/hr min @ 47F is more than half your design heating load (and well over the actual heating load at 47F), which means it will be cycling on/off a lot, and probably won't quite hit it's efficiency numbers.

The FH09 is usually sold into cold climates, and it'll be more money up front, but it modulates down to 1700 BTU/hr cooling, 1600 BTU/hr heating @ 47F, which means it will run nearly continuously, and would likely hit it's (substantially higher) efficiency numbers. At internet store pricing it's about $450 more expensive than the HM09, but in most places the higher efficiency would pay that back several times over within the lifecycle of the unit. It has additional margin for both heating & cooling at max-modulation too, with 12,000 BTU/hr max cooling (almost as much as the HM12NA ), and 12,200 BTU/hr @ +17F.

 

[Elton Noway]

Most 9K minisplits can deliver at least 10,000 BTU/hr of coolking at maximum speed, some can deliver more than 12,000 BTU/hr of cooling, so you are in no way going to be pushing it too hard. The 9000 BTU/hr number is the "rated" sizing, the modulation level at which they did the SEER efficiency testing.

The HM09 is also good for 10,000 BTU/hr cooling (@ 95F outdoors, 80F indoors), 7200 BTU/hr heating (@ +17F out, 70F in) per the submittal sheet.

That means it has about a 1.25x oversizing factor for you cooling, a 1.15x oversize factor for heating. It couldn't be more ideally sized for your loads from a maximum output perspective. The 1-ton would be overkill.

With online tools like that you can't tweak in most of the particulars for higher accuracy, and it's more likely than not to be overstating reality by 10-15%. It looks like you didn't use a shading factor for the windows other than overhangs(?), which means the cooling numbers could be quite a bit overstated.

When comparing units, look at the minimum modulation levels relative to your load, not just capacity and SEER/HSPF, since that affects both comfort and as-used efficiency. The minimum modulation levels on the HM09 are pretty high- 3800 BTU/hr minimum cooling is nearly half your design cooling load, and the 4500 BTU/hr min @ 47F is more than half your design heating load (and well over the actual heating load at 47F), which means it will be cycling on/off a lot, and probably won't quite hit it's efficiency numbers.

The FH09 is usually sold into cold climates, and it'll be more money up front, but it modulates down to 1700 BTU/hr cooling, 1600 BTU/hr heating @ 47F, which means it will run nearly continuously, and would likely hit it's (substantially higher) efficiency numbers. At internet store pricing it's about $450 more expensive than the HM09, but in most places the higher efficiency would pay that back several times over within the lifecycle of the unit. It has additional margin for both heating & cooling at max-modulation too, with 12,000 BTU/hr max cooling (almost as much as the HM12NA ), and 12,200 BTU/hr @ +17F.

Whew... thats a lot of information to take in with my limited knowledge of HVAC. However, after reading through your response several times you've convinced me a 9K unit is all I need. Now all I have to do is convince the contractors to quote a 9K... (which won't be easy since they all seem eagar to sell the larger units). I may have to send them the link to this article on mini split systems entitled: Don’t oversize, choose the right equipment

The minimum modulation levels of the HM09 were an eye opener. The idea of it cycling off and on "a lot" and not quite hitting its efficiency numbers is a no-go for me. I've scratched that one off the list.

Based on the information we've exchanged... it sounds like "if you were me" you'd strongly consider the getting the FH09. My only confusion is when you stated "it will run nearly continuously". I'm aware when cooling that dry air feels more comfortable than humid air so running the unit longer will give the unit more time to dehumidify... but the idea of "running almost continuously" sounds like the unit would wear out faster. I know frequent cycling is a lot of wear and tear... but running continuously sounds like it would shorten its lifespan. Yes..no?

Last... since you said the FH09 is usually sold in cold climates, would there be any issue or anomalies I should be aware of with installing one here in what is considered to be a warm climate? The 9K FH09 is $400+ more than the 12K MUZ-HM12 but if I'll realize a more comfortable room year around... and with more efficiency... I can live with the price difference.

 

[Dana]

There are no problems with using a cold-climate mini-split in a warmer climate. Even cold climates get 100F days sometimes, even though warmer climates don't see -10F very often.

Running continuously doesn't wear out motors or compressors as much as frequent starting & stopping. The start up current on a compressor motor that is cycling has a bit of back-pressure to overcome. They are designed to be able to take continuous operation.

When starting and stopping in cooling mode moisture on the coils in the head re-evaporate into the room, so the latent cooling isn't very effective. These things all have a "Dry" or "Dehumidify" operating mode if it's not quite doing it in regular cooling mode (whether cycling or modulating), but the SEER efficiency isn't nearly as good in that mode, so using the normal cooling mode is recommended until/unless it's not keeping up with the humidity.

I AM a fan of the FH09 (and LG's Art Cool Premier 3/4 tonner) for the wide modulation range it provides.

You'll have to open a free trial subscription to read it, but this blog bit is worth looking at too.

 

[Elton Noway]

There are no problems with using a cold-climate mini-split in a warmer climate. Even cold climates get 100F days sometimes, even though warmer climates don't see -10F very often.

Running continuously doesn't wear out motors or compressors as much as frequent starting & stopping. The start up current on a compressor motor that is cycling has a bit of back-pressure to overcome. They are designed to be able to take continuous operation.

When starting and stopping in cooling mode moisture on the coils in the head re-evaporate into the room, so the latent cooling isn't very effective. These things all have a "Dry" or "Dehumidify" operating mode if it's not quite doing it in regular cooling mode (whether cycling or modulating), but the SEER efficiency isn't nearly as good in that mode, so using the normal cooling mode is recommended until/unless it's not keeping up with the humidity.

I AM a fan of the FH09 (and LG's Art Cool Premier 3/4 tonner) for the wide modulation range it provides.

You'll have to open a free trial subscription to read it, but this blog bit is worth looking at too.

Dana... THANK YOU... for all the insight and information.. as well as the link to greenbuildingadvisor. In addition to the article you referenced I found additional minisplit articles. Thanks to you the anxiety I was experiencing due to all the different "stories" I was hearing from the contractors on this subject has been alleviated. I now feel like a more informed shopper and have a better handle on the world of mini splits rather than just viewing them as another HVAC system!

As mentioned earlier I'm cooling/heating less than 200 sq ft and questioned the quotes for 12K and 18K units. Whenever I asked if a 12K or 18K was too big (i.e. oversized) I heard: "Not to worry... the good thing about mini splits is they are variable so a 12K/18K unit will modulate up or down as needed to meet the requirements of the load." I've now learned that although statement is true in a general sense... the contractor(s) fail to disclose, (i.e., either because they lack the knowledge, don't understand or don't care) the problems associated with oversizing such as short cycling, inadequate humidity control and the impact on the units efficiency.

I now have my contractor of choice quoting a FH09 ... he said the price difference dropping from a 12K to a 9K will offset some of the cost but as you hinted it looks like I'll have to come up with another $450. He said he had to agree the minimum modulation levels of the FH09 makes it a much better match for my sunroom. He said they don't usually quote the FH models because the higher cost of the high seer units will would make their bid less attractive when bidding against other contractors who will be quoting units with lower seer ratings as most customer will go for the lowest price with all things being equal. NET: Yes any of the units originally quoted would have cooled the room quickly, making average Joe homeowner happy... but without them ever realizing the unit was oversized, inefficient, short cycling thus resulting in added wear and tear and impacting the lifespan of the unit.

 

[Dana]

BTW: The capacity of their half-tonner would also cover your calculated loads. Its a hundred or so cheaper, with identical low-end modulation numbers as the FH09. It won't be significantly more efficient than the FH09, but a bit cheaper up front, but less maximum capacity, for a longer recovery period in instances when you turned it off when you went away for the weekend, to came back to a steam-bath of a sunroom. They're all more efficient if you "set and forget", rather than turning them off or up/down unless it's going to be a full day or more.

 

[Elton Noway]

Its been just a little over a year since the installation of my Mitsubishi FH09. Thanks to this forum I felt obligated to check back in with a follow up on this Mini split thread and the final results. A special thanks to Dana for pointing me in the direction of the FH09.

As eluded to in my reply of July 15th, several contractors I talked to downplayed the FH09 suggesting the Hyperheat model was overkill for our southern climate and mild winters, although they all agreed the wider modulation range it provides would be of greater benefit ... they usually don't quote the unit due to its higher cost.

WOW... I'm so glad I went with Danas' advice! We had a unseasonably cold winter this year setting two record lows in January. Granted not cold compared to Minnesota and other cold climates but very cold for Raleigh. To make matters worse our 23 year old whole house HVAC unit died early in the morning on Jan 1st.

Due to the record cold HVAC systems that were failing all over NC we were told it would be at least a week before anyone could respond to our service request. Fortunately.. our sunroom with its FH09 saved the day! It had absolutely no problem keeping the sunroom toasty warm. (amazing how much warm air it throws out!) Notice the snow in the top half of the pic below. Again not Minnesota but its cold for us folks down south.

As you might imagine we had no room to bargain or negotiate on the price of a new HVAC unit when the contractors know your back is against the wall (i.e., winter and no heat). Rather than buckle under the pressure... we set up temporary living quarters in the sunroom and decided to bide our time and enjoying the comfort of our sunroom. When spring arrived we had the luxury of time and were able to interview multiple contractors to find the most reputable and qualified as well as negotiate the price.

Anyway... fast forward to today with summer in full swing... temperatures in the 80's and upper 90's with humidity to match! (feels like a sauna outside) but... our sunroom is nice and cool with no humidity issues.

NET: No short cycling, controllable year around temperatures, awesome humidity control and great efficiency. Couldn't be happier.

 

[Dana]

Glad it all worked out for you, and that you stuck to your guns on that one!

The low-low minimum output numbers for the FH09 make it pretty flexible for sizing to lower-load rooms, but it still delivers the goods at the temperature extremes. Even the 1-tonners recommended have lower output than the FH09 even a +17F, and nowhere near the modulation range. The vapor-injection scroll compressors that give it the "hyper heating" capacity also gives it higher efficiency. While the FH09 is optimized for heating efficiency, it's cooling efficiency is better than the simpler compressor units too. It's hard to beat. (I hope you take the time to update the contractor who installed it too. Even in New England most installers don't really know what they're dealing with, and just how superior these units are to those with simpler compressors.)

For somewhat higher loads than yours Fujitsu's 3/4 ton -9RLS3 is another solid performer using a vapor injection scroll compressor yielding very-high efficiency and good low-outdoor temperature capacity. But since it can only modulate down to 3100 BTU/hr (nearly twice that of the FH09), the FH09 was the right choice here.