Another Mini Split (Over)sizing Question!
Hi everyone–first time caller, long time listener here.
I had a 4-zone ductless mini split system installed a couple weeks back and am hoping that you all might provide some clarity. I’m somewhere between needing to understand what’s going on with the system and having buyers remorse. I’ll start out with the basics and then get down to my question.
The house (floorplan attached):
1,680sqft conditioned according to plans, 3 bedroom, 2 bathroom, in northern New Mexico, built 1986.
Previously heated by a 100k BTU boiler (original to house) with radiant hydronic baseboards. Cooled with a swamp cooler. Ductwork exists for the swamp cooler, but according to all potential installers was likely undersized and uninsulated for any modern ducted system.
The system:
-MXZ-3C24NAHZ tied to MSZ-FS09NA (Master Bedroom~210sqft finished) and 2x MSZ-FS06NA (Daughter’s Bedroom [~117sqft finished] and Office [~104sqft finished])
-MUZ-FS18NAH-U1 tied to MSZ-FS18NA (Open concept living room/kitchen/dining room [~835sqft finished])
-No additional controllers/thermostats, just included remotes.
The story:
Boiler (Slant/fin cast iron) is original to the house and basically every valve, connection, etc. on it is leaking or has leaked. Having had some appliances die on us in the past we wanted to get out ahead of this expensive purchase and make a decently well-informed decision. We had a Mitsubishi single zone (18k) ductless in a previous house (~1,000 sqft) that we liked, though it didn’t really push air down a short hallway into our one bedroom and bathroom.
I ran a block calculation on loadcalc.net and got around a 33k BTU demand for heating. A room by room calculation (attached, hopefully) shows around 43k-47k depending on how generous I am with my estimates of our insulation. Having dealt with the cold/hot bedroom (ha!) before, and based on recommendations from the 8 HVAC companies we interviewed, we went with a head in each bedroom despite the wisdom on here that we were playing with fire with the small bedrooms in particular.
Since being installed, the living/kitchen/dining room has been quite comfortable, as has the master bedroom. The two small bedrooms, unsurprisingly, are the potential issue. The system was installed during a cold week (highs in the 20s), and the first several days I was having to set those two small rooms 4ish degrees above the actual desired temperature. Now that things have warmed up those two rooms are often 72-73 degrees, despite their setpoint being as low as 63. Doors are open as often as possible–closed only when sleeping. Again, I knew that the 6k heads were likely too big for the bedrooms but, perhaps foolishly, wanted to cover adjacent bathrooms and hallways if possible so was hoping to buy a little bit of leeway there. Also again, I was led to believe that between the flat roof (no attic space) and existing ductwork it would be impossible or financially impractical to install a ducted unit, though in theory I understand why that would have been ideal.
So my questions to you fine people are:
1. What’s the deal with the two small bedroom running cold then hot when the master has been comfortable and within a degree or two of its setpoint the whole time? Just a function of short cycling all around? Would have expected them all to just be hot.
2. Is there any way to control/automate/circulate myself into a slightly more efficient/more comfortable position? i-See is currently on, but no empty room detection for whatever that’s worth. It does seem to ramp/wake up when we enter a room though.
3. For those kind souls who are willing to take a peek at the attached load calculations (and I do appreciate that that site isn’t the same as having a professional run a real calculation), how would you respond to the “fact” that it shows the 6k heads as perhaps not grossly oversized? Any obvious flubs in my entries?
Much appreciated!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I'm no expert, but it seems the biggest factor between master and other bedrooms is that small bedrooms are getting way way more solar energy facing the southwest. I'm interested to see what anyone else comes up with regarding the system. Good luck!
Definitely a good observation. It’s been a mix of weather (clouds, snow, but also sunshine) since we got them installed so I’m also interested to see what people think. I’m unsure what to expect when we switch over to cooling. Those rooms bake in the afternoon so I’m hoping oversizing will be less of an issue when we’re cooling.
“3. For those kind souls who are willing to take a peek at the attached load calculations (and I do appreciate that that site isn’t the same as having a professional run a real calculation), how would you respond to the “fact” that it shows the 6k heads as perhaps not grossly oversized? Any obvious flubs in my entries?“
Looks like 60% of your heat loss is through the slab. Let’s hope the calculator used realistic u values and temperature values for that!
Also a good observation. For being a mid-1980s house I think it holds temperature fairly well (boiler went out once in 2020 when we were in single digit temperatures and the house stayed reasonably…not cold…just running the gas heater in the garage with the door open to the house. But it is just engineered wood on concrete with a pretty skimpy underlayment in between. Given the age of the place and my observations I’m guessing no additional insulation on the slab.
What happens if you shut off the units in the small bedrooms?
So far I’ve only had one off at a time (turned off the office for the weekend) and no obvious change in the master. That was when both small bedrooms were running a little colder than the master and when the weather was colder but I didn’t notice a big change in my daughter’s room either. I’ll try shutting both of them off for awhile tomorrow to see if it affects the master.
So I turned off the heads in the small bedrooms yesterday, which was a fairly cold day. The two small rooms stayed pretty even until the sun went down, dropping just a degree or two over the course of the day. The master was farther below its set point than normal. 4ish degrees below rather than the 1-2 degrees it normally is.
A Mitsubishi thermostat may solve your issue in the master. My guess is the other bedrooms will likely be fine with their units off during heating season. May not be the case in cooling season.
Adding on to my post again, I used the process outlined here: https://www.greenbuildingadvisor.com/article/replacing-a-furnace-or-boiler
When I do the math I get numbers that aren’t far off from the rated output at 17F, which is the closest temperature on the spec sheet to the 99% outside design temperature for my area (though the link in the article is dead so I’m not 100% sure what value to use there).
Questionable math aside, between my loadcalc output and calculations based on my energy bill which fall in the same ballpark, does it seem reasonable to assume that my system is not horrendously oversized? Or am I overlooking something obvious?
But oversized on the whole house level or not I guess that doesn’t actually mean much for the reality of heating and cooling small bedrooms? Would it be worth looking into replacing the MXZ-24 outdoor unit with individual compressors for each bedroom head? If anyone has a link to an article quantifying energy use when short cycling vs efficient modulation I’d be curious to see!
Just in case this additional information is helpful to someone going down the Google rabbit hole in the future here’s one more update from me:
After an extended streak of mild weather where the units were off for long periods we just switched to some occasional cooling. My daughter’s room would go up to 10 degrees below the set point, while the other two rooms would hardly cool, if at all. Also, the coils on the air handler in another room would freeze over when I turned on a zone. Called my installer up this morning and they got someone out just now. Turns out the units were cross wired. Mitsubishi actually has a good video on YouTube about diagnosing this problem. With only 10-15 minutes of work everything seems to be functioning as it should (fingers crossed) and I’m hoping this will have a positive effect on the over/under heating/cooling now that the proper zone(s) are actually calling for refrigerant. Hopefully calling for heat/cold where intended also helps the system’s efficiency.