GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Mitsubishi minisplit 8-head system is not keeping up

simpsoap993 | Posted in GBA Pro Help on

Back in mid-November 2016, we had a Mitsubishi Hyper Heat MINI Split installed in our house after opening up the first floor of our home and installing some beams to make it more open. We lost duct work feeding the second floor, and had to come up with a solution. We decided on a Mitsubishi ductless system, since I had a smaller version installed in our detached garage and was very happy with it. My uncle did the garage, while I went with one of the Mitsubishi Diamond Dealers to handle our house. After 3 quotes, we decided on a company and proceeded.

We went with a 4 ton unit outside, and a total of 8 heads inside the house. They are connected via two branch boxes in our attic, one branch box supports 3 heads, the other supports 5. Our house is 2320 sq ft, with 3 heads on the first floor, 12k head in the living room/family room, 9k head in the Kitchen and a 6k head in the breakfast nook. Upstairs there are 5 heads installed, two 6k heads in the master (one in the bedroom and one in the bathroom) and then one 6k head in each of the other 3 bedrooms. The total cost of this system was roughly $22k, and I must say, I haven’t been very pleased with this in our house.

My concern is that I feel that the system doesn’t keep up with the temp set on the remotes. The remotes are set at either 68-68F, and I feel that the heads run most of the time. It has just started getting pretty cold here in Ohio, but these Hyper Heat units are suppose to handle -13F temps. I’m concerned that there is an issue with the system somplace, and my installer has been slightly challenging to deal with. During most days, the 3 heads downstairs are turned on, and the upstairs heads are turned off, due to heat rising. When we go to sleep, we tend to turn the master bedroom on, as well as our son’s room. The temp in the bedroom’s seem fine, as the heads are likely oversized for the rooms, but the Mitsubishi doesn’t have anything smaller than a 6k head.

I’d like to know what the community here thinks, or possibly what I could check. The detached garage runs like a champ, and I have had zero issues with the two 18k heads installed out there. My other concern is our electric bill/usage. Our past two months have been skyhigh, and according to our meter, we have used twice the amount of electricity than we normally use. November bill stated a use of approx 2500kwH and December was approx 4400kwH. My wife and I are both confused as the system we removed from the house was a 10 year old 13 SEER heat pump and the new units are supposedly 18.9 SEER.

Any help would be appreciated.


GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    Lots of issues here. It seems clear that your equipment is grossly oversized. Did anyone perform a Manual J calculation?

    If I understand correctly, you have two complaints.

    1. "The system doesn't keep up with the temp set on the remotes." Do you mean that the minisplits can't make your rooms warm enough, no matter how high you set the thermostat? Or do you merely mean that the room temperature differs from the thermostat setting (a minor problem -- fixable by increasing the thermostat setpoint)?

    2. "Our electric bill and usage have been sky-high." This is a real problem -- one that is difficult to diagnose over the internet. You may need to contact a Mitsubishi rep to get to the bottom of this.

  2. Expert Member
    Dana Dorsett | | #2

    As the weather gets colder the temperature stratification in the rooms becomes greater, with a bigger difference between the air near the ceiling and near the floor. Mini-splits sense the room temperature by the temperature of the incoming air, so wall coils mounted high on a wall sense a temperature that is warmer than the average room temperature. To deal with that you can either bump up the setpoint with the remote, or (for a bunch of money) install wall-mounted thermostats.

    The GE06 heads don't have as much turn down ratio as the FH06 heads- which versions do you have?

    The SEER rating of the heat pumps have no direct relation to the HSPF (heating season performance factor). What was the HSPF of the older system vs. the Mitsubishi?

    Is your outdoor unit the (now discontinued) MXZ-8B48NA (HSPF= 8.7) or is it the MXZ-8C48NA (HSPF= 11.4) ?

    You installed TWO 18K heads in a garage, and you're wondering why the power use is high?

    Is there some reason why you haven't contacted the installer with these concerns?

    If you are activly turning the heads off & on, you are ruining the potential system efficiency. They work far better with a "set & forget" approach, only bumping the temp up or down a few degrees at a time. When it's say 30F outside a head running at it's minimum speed is more than 2x as efficient as when running at max speed. Any energy savings you get from the lower heat loss when they are off gets rapidly eaten up during the temperature recovery ramp by the drastically lower efficiency. By turning heads off you are also forcing the remaining heads to run a higher speed (=lower efficiency) much of the time to handle the compressor's minimum output without cycling the compressor off/on, which would be even worse for efficiency.

    Exact meter reading dates & kwh are far more informative than the mere month, given the large differences in weather from year to year for roughly the same month, especially in the fall & spring shoulder seasons. If you have the old electric bills and the new ones with exact meter reading (not billing) dates and a ZIP code you can look up the heating degree days for those periods.

  3. simpsoap993 | | #3

    Martin - a Manual J calculation was performed on the house, and I was told that I had a heat loss of approx 45,000BTU, which I thought seemed a little high. They addressed their concerns to me that the 4 ton unit might not provide enough heat on the really cold days, but I was skeptical, perhaps I should have been.

    As for responses, I haven't set the remotes beyond 68-69 degrees, and it just seems that they are kicking on often to keep the temp. I've purchased a couple of thermostats that I have around the house, and most of the time they read about 66 degrees. However, this morning it was roughly between 2-5F outside, and the thermostats were reading 59F. Should I adjust the setpoint for something like this?

    As for our electric bill, the only thing in the house that has changed was removing our old heat pump and blower from the basement, and the installation of the new Mitsubishi.

    I've attempted to call Mitsubishi twice now, and both times they have told me I need to contact my contractor who performed the install.

  4. simpsoap993 | | #4

    Dana -

    The 6k heads that are installed in the house are the FH model.

    The outdoor unit is the MXZ-8C48NAHZ. As for the old equipment, I'm not sure what the HSPF rating was, the blower in the basement was orginal with the house, which was built in 1993, so I'm sure it had lost its efficiency, or at least some of it. It had a 30AMP breaker for the old heat pump, and a 100AMP breaker for auxillary backup.

    In regards to turning the heads on/off, how many do you suggest we turn on? Should all 8 be on? We have our master bathroom, office and spare room off, as we aren't in there often. I tend to turn our bedroom and our son's bedroom off in the morning, when we leave for the day. the downstairs heads are always on.

  5. Expert Member
    Dana Dorsett | | #5

    If the heads are sized correctly for the loads they should not be "...kicking on often to keep the temp...", they should in fact be running nearly continuously any time the temperatures are below 40F. Only when they have enough load to run in a modulating fashion at a very high duty cycle will the system hit anywhere near it's HSPF numbers.

    The exact model numbers of the heads & compressor would be useful for this discussion. FE and GE and FH heads have different turn down ratios & efficiencies, and all are compatible with MXZ-8C48NA compressors.

    Of course the installer would be the first place to call for support. Since they specified & installed it, it's THEIR responsibility to make it right. Mitsubishi would be the right place to call if the contractor went out of business or was unable/unwilling to make corrections/adjustments/ explanations.

    The offset between the programmed set point and room temperature is normal (as explained in my prior post.) Just bump up the setpoint to where it's comfortable and don't sweat the fact that it's a higher number. If you care to know the true room temperature, a cheap digital thermometer mounted on the wall or on a table top etc works just fine. Depending on where the heads are mounted and the room's heat load it's not uncommon to see an offset of 10F between the setpoint and the average room temp when it's in single digits F outside. If it's mounted near the ceiling on an interior wall it'll be a much bigger offset than when mounted below a window on an exterior wall.

  6. Expert Member
    Dana Dorsett | | #6

    "I tend to turn our bedroom and our son's bedroom off in the morning, when we leave for the day."

    Do NOT turn ANY the heads off- that's exactly the scenario that ruins it's modulating efficiency! It may seem counterintuitive for those who only had non-modulating HVAC, but really, "set & forget" works best, only bumping the setpoints up/down a few degrees at a time.

    If the Master Bath stays at a reasonable temperature most of the time even with the ductless head off it may be better simply left off, with the door(s) open most of the time to provide load for the head(s) in the adjacent room(s).

    If the downstairs heads are cycling off during winter weather with a low on-time duty-cycle, try turning one off to see if that improves the duty cycle of the others. Seriously, if sized correctly they should be running nearly constantly this time of year, only modulating speed up/down with the weather.

  7. GBA Editor
    Martin Holladay | | #7

    Rule number one: If you aren't warm enough at the current thermostat setting, try bumping up the thermostat.

  8. simpsoap993 | | #8

    Dana -

    So you suggest leaving the heads on in the rooms we aren't even using, but just setting them as low as the remote goes, which is 61F?

    Below are the exact models of the heads/units I have in my house:

    Outdoor unit: MXZ-8C48NAHZ
    Family Room: MSZ-FH15NA
    Kitchen: MSZ-FH09NA
    Breakfast Nook: MSZ-FH06NA
    Master Bedroom: MSZ-FH06NA
    Master Bath:MSZ-FH06NA
    Office: MSZ-FH06NA
    Spare Room: MSZ-FH06NA
    Son's Room: MSZ-FH06NA

    As I mentioned, the temp upstairs has been fine, it's the temp downstairs which I feel is lower than I would like, but I can adjust the setpoint on the remotes. Still doesn't explain the wattage usage on our bill though.

  9. Expert Member
    Dana Dorsett | | #9

    If the room is a spare room that isn't used for at least part of the day most days, there's no point to running the head, and powering it off will use less power than letting it cycle maintaining a low temp. If the office is only used on weekends or something, you might leave it off during the week, but leave it on all weekend, don't turn it off on Saturday if there's a chance you'll be using it on Sunday.

    The FH06 and FH09 units each can modulate down to about 1700 BTU/hr even during shoulder-season weather and should be running nearly continuously during cold weather for the master bedroom (but not the bath), and possibly the other bedroom. But if you have turned off most of the other heads it would have to run nearly full-blast to keep up with the minimum output of the MXZ-8C48NAHZ, which is about 7000 BTU/hr. For a room with a design heat load of say 3500 BTU/hr @ +5F it would then only cycle on/off, along with the compressor, which is bad for efficiency on all fronts. You would need to have at least four FH06/FH09 heads active for them to run at minimum modulation and super-efficiency while the compressor is running at it's very efficient minimum speed without the compressor cycling on/off.

    The minimum output of the FH15 is about 5000 BTU/hr, so as long as that room has a LEAST that much load at the average winter outdoor temp it (and the others) should be able to modulate well. But the FH15 can deliver 18,000 BTU/hr @ +5F on it's own, which is about half the heat load of my whole (antique 2x4 framed) house @ +5F, and would be oversized for most zones. If there is good air communication between the family room and kicthen or family room and breakfast nook, try turning off the FH15, but leaving the bedrooms and kitchen (or breakfast nook) head on. If the kitchen & breakfast nook are adjacent and not doored-off from one another, turn one of those off, since they are probably forcing each other into cycling rather than modulating.

    Knowing the Manual-J room loads associated with those head (and the indoor & outdoor design temperatures) would be useful.

    Systems with oversized heads & compressors that are cycling rather than modulating most of the time will only deliver a seasonal average coefficient of performance (COP) of 2- 2.5, which is an HSPF of 6.8 - 8.5. If sized such that most of the heads (and the compressor) are all modulating well into the shoulder seasons it should deliver a seasonal COP of at least 3 (HSPF= ~10) if not the meeting or exceeding the tested & specifed HSPF 11.4 (COP=3.34) in any Ohio location. Almost all of Ohio is in the heating-hours Zone IV region by which heat pump heating efficiencies are characterized (see: ), so if it was all sized & installed correctly it should be able to at least come close.

    October/November 2015 was pretty mild compared to October/November 2016 for some parts the northeastern US, which is why you need to be using a kwh per heating degree-day comparison rather than just the usage for the month.

    BTW: A typical 2320' house with 2x4 framing and clear-glass double panes (or clear storm windows over single pnes) over an UN-insulated basement might have a load of 45,000 BTU/hr @ 0F, but if it's pretty tight slab-on-grade or the basement is insulated it'll be closer to 30, 000 BTU/hr @ 0F. If it's pretty tight with 2x6 framing and low-E windows (or low-E storms it could even be under 25,000 BTU/hr. There are many exceptions to prove the rule which is why you use a Manual-J rather than WAGs like this, but most HVAC contractor performed Manual-J's are conservative rather than aggressive, and Manual-J only works well when the assumptions are aggressive.

  10. simpsoap993 | | #10

    My apologies, the head in the living room is FH12, not FH15. The upstairs is perfectly fine, the temp is good. The downstairs is the issue. Take today for an example, it is about 4F outside right now, and the heat is barely coming out of the heads. The setpoint is set to 72F, but my thermostat sitting on our table is reading 63F. I'm frustrated beyond belief, especially after spending $20k.

  11. ohioandy | | #11

    Andy, I''m also in Ohio. It's 10 degrees here now. My two 9K Fujitsus are keeping my 1,000 sq.ft. house nice and warm. Dana's right about stratification in frigid weather: these last couple days it's noticeably colder along the downstairs floor, and I've had to bump the thermostat to 70 to achieve 67. Upstairs, the unit is set at 66 and achieving 66.

    Just to echo previous advice, get your installer out to look. Minisplits can do far more than you're asking of them--other than the design issues, another possibility is you've got some failing controllers or maybe leaky linesets. One of my units seemed to fade in its first year, and my installer came out, checked things over, adjusted the coolant charge, and it's been fine ever since. FWIW, he didn't even charge me for that. Sounds like you've got an unenthusiastic installer. In my corner of Ohio, it's tough to find minisplit contractors who have a clue.

  12. simpsoap993 | | #12

    I feel your pain, I feel like my contractor was more than happy to take my money, but unresponsive to come out. I just don't understand why the lack of air flow when it gets this cold. Could it be low on coolant?

  13. simpsoap993 | | #13

    For a point of reference, it's currently 0F here in Ohio, my downstairs heads are set on 72F and 75F and the thermostats are reading 54/55F. That's just unacceptable to me. Even with the hyper heat, I feel the outdoor unit should be doing a better job.

    I'd love some opinions.

  14. simpsoap993 | | #14

    Looks a phone call into the contractor is in order, last night around 230am, I woke up and noticed that none of the heads in the house were getting power. Breaker isn't tripped, but every head is off. House is currently at 50F. These contractors where a Mitsubishi Diamond Dealer, but that doesn't seem like it means much.

  15. DIYJester | | #15

    You could also be having issues with the branch box. From my understanding the branch box is what is sending the call for heat from the head to the compressor.

    Did they use the proper wire for the M-Net signal cabling, this should be 16-2 shielded wire according to Mitsubishi.

  16. simpsoap993 | | #16

    I'm going to ask all these questions when they come today at 1230. Looking at the communication wire, it appears to be 16-2. Both branch boxes are mounted in the attic, not exactly sure what happened last night, but the system is dead in the water at the moment.

  17. DIYJester | | #17

    I would definitely raise hell with the installer. I just bought the 42NA, 18k, and 3 6k heads at just under $8k with line sets and branch box. I would guess they have close to $10k in labor on the install as the 6k heads are only about $475 each plus line sets.

    It also seems your system is pretty oversized. I have close to 3300 sq. ft. and will only have 4 heads. The largest in the living/kitchen and one 6k in each bedroom. The extra 6k for the master bath and breakfast nook seem excessive.

  18. simpsoap993 | | #18

    The contractor came out around 1230 today. It suggests the quick disconnect was loose which resulted in a blown fuse. I can only assume this happen due to poor installation. We have heat again, and I've taken some advice from this forum and adjusted which heads are on.

    Supposedly, the contractor is going to come back out this week to address my other concerns. They stated that the weekend is only for emergency service. They are suppose to call me on Monday.


  19. DIYJester | | #19

    Do you know how much refrigerant they added to they system? With that many heads and line sets etc. I would guess they have to charge a good 10-15 lbs.

  20. simpsoap993 | | #20

    11lbs 3oz were added based upon 340ft of line set.

  21. simpsoap993 | | #21

    I wanted to provide an update to anyone following this thread.....

    - After the tech came to the house on Saturday, the system was back online, but not doing well at keeping up. Adjusting the setpoints on the remotes to 68-75 yielded pretty much no difference at all. The heads on the first floor of our house ran continously, while the ones on the second floor were cycling on/off. Even with heat rising, I don't feel like our upstairs should be 10 degrees warmer than the downstairs.

    - I called the HVAC company this morning, because the Saturday visit was just for emergency and to get the system going. They plan to come back out to check for leaks, and possibly pull vacuum on the system and see if they can find any issues. This perhaps would tell them how much refridgerant is in the system as well?

    - I've asked for a copy of my Manual J calculation, but who knows if they will provide me with a copy. I think for the amount of money that I spent, i'm not asking for too much here.

    - From what I've read on this site, this system should be able to keep our house at a comfortable 70F, which it has yet to do, since being installed.

    thanks for all the help with questions/comments. Of the 8 total heads in the house, we have 6 running, the master bathroom and office head is turned off.


  22. DIYJester | | #22

    I would be very interested to find out what happens. I just got a Mitsubishi 4 zone system and will be doing most of the install myself. I have the EPA license to get the refrigerant from my Navy time and have worked on larger units (100+ tons, even though 410a doesn't need it), but I don't have the tools to pull the micron vacuum etc. or I'd probably try to do it myself. Depending on his price I may do just that. When I was looking at possibly a ductless system for 3 heads (with ducts) and commissioning, they wanted almost $5000 and were not a Diamond dealer, wanted extra for a Manual D, and didn't really want to look at my Manual J.

    I'm working with a Diamond Installer now to see how to get the 12 year warranty still. Unfortunately as you may be finding out, most of these will not cover labor.

    My last central air unit was installed in 2012, professionally. By 2015 cooling season it had leaks 11 lbs or R134. Just to charge the system some guys wanted $30 a lb., that didn't include finding the leak or replacing anything that was covered "under warranty". I ended up having them charge it for one more season and had it evacuated the next. Every year since the install I've also had water leaking from the poorly routed and insulated flexible ducting in my attic. The installer says it's not his problem. I should mention he was the same company that installed a hydronic floor heat system in this home when it was built (not by me) without any sub-slab or perimeter insulation.

  23. srenia | | #23

    Having loss 7k on a couple low temp mini splits in a rental I understand a little bit. 22k for a system that is basically junk. 22k you'll never see a return on the heat bill. In my case I replaced the brand new mini splits with mini PTAC unit that are used in hotels. Just under 1k for the heat pump versions, but reliable, easy to fix and work! Sometimes the easier solution is the better. Air sealing and insulation gave more energy bill savings. Some people have good results with mini split type systems, but I would never recommend it for the average person. Mini split manufactures laugh at you, installers are unreliable and the energy savings don't offset initial cost compared to other strategies. Hopefully your story is a caution for others as mine is.

  24. Jon_Lawrence | | #24


    Are you able to measure the temp of the air coming out of the heads?

  25. simpsoap993 | | #25

    Jonathan -

    I purchased a probe to check the heat coming out of the heads, over the weekend, the temp coming out of the heads downstairs was only about 85-87F, which seems very low.

  26. iLikeDirt | | #26

    This is an excellent cautionary tale on the pitfalls of complexity. The system profiled here was horrendously expensive and complicated, and has proven unreliable. When you need help with your expensive system, it can turn out that the pros are anything but and the cost keeps spiraling out of control.

    As much as I want to love mini-split heat pumps, I keep hearing these stories over and over again. In theory they should be reliable, efficient, simple to install, and reasonably inexpensive. In practice, the state of the American HVAC industry often precludes it.

    PTAC/PTHP units look like a great alternative to me, especially the ducted versions. Get the load down with insulation and air sealing, then feed the house with one central ducted PTHP. When it breaks, you just switch it out with a new one for less than $1000--a task that mechanically-inclined homeowners may be able to do themselves. And they're easy to replace as the technology improves and efficiencies climb higher. Sure, the running cost will be higher. But the installation cost is likely to be so much lower that you can install more rooftop PV to make up the difference.

  27. bencarsan | | #27

    Andrew, if you aren't happy you have an install problem. Properly installed, this stuff works great.

    Are you using the hand held remote or the optional wireless wall mount MHK1? We always recommend clients use the MHK1 because the supplied hand held units don't have their own thermostat and instead rely on the temperature reading at the head itself. This can lead to comfort issues when the temp at the head is different than other parts of the room. The MHK1 allows the user to locate the thermostat wherever it makes the most sense to take the temperature of the room.

    As Dana points out you should let your comfort be the guide--don't rely on numbers.

    Also, keep in mind that temperature set backs generally don't make efficiency sense with air source heat pumps because they like to run long and low. I set mine back at night, but that's for better sleep and comfort not energy savings, and I program them to come up to temp in stages rather than all at once.

  28. simpsoap993 | | #28

    Bennett -

    I'm using the remotes, as I wanted the ability to control each of the heads individually.

    I understand this is an install issue, and not equipment. My uncle did my detached garage, which is why I wanted to go with Mitsubishi's for the house. However, my uncle was too busy at the time to tackle the whole house install, and winter was coming and we needed heat upstairs. It's sad to think that installers don't know what they are doing when it comes to equipment like this. Yes it was an expensive install, but all three quotes I received for the work were north of $20k.

    As for comfort, we don't rely on the numbers, but if i turn the head to 70-72F, it never actually achieves that temperature. I've listened to my wife complain about not being warm in the house, and I agree with her, and then I get to hear my 3 year old son tell me he is cold as well. We use our propane fireplace reguarly to help take the edge off.

    I do appreciate everyone's inputs on here, I have actually learned a lot -- but without a good installer, it isn't like I could have installed a system like this myself.

  29. STEPHEN SHEEHY | | #29

    I think the real benefit to ductless minisplits is their simplicity. My house is very well insulated and very airtight. For a simple open plan like ours, it was easy to decide on two minisplits, each with separate indoor and outdoor units. For the almost two years we've had them, they have worked flawlessly.
    But for a multiroom house that needs 8 heads, you lose the simplicity. Andrew would probably have been betterr off with a more typical furnace or boiler.

  30. Expert Member
    Dana Dorsett | | #30

    At more than $5K/ton for a 4-tonner it's on the high side compared to what I've seen on competitively bid multi-split installs in my area. Most come in between $3500-$4000 /ton. Often separate mini-splits can be even cheaper.

    This system absolutely SHOULD be able to keep up with the load, and it's worth pushing the contractor to make good on it. The fact that it's running at piss-poor efficiency (we think, based on the power-use ), even when it's not delivering the heat is an indication that something is fundamentally wrong, but anything that far off should be diagnosable by a competent tech.

    You're still too focused on the absolute numbers if you're concerned that it never achieves 70-72F when the remote is set to those numbers. My relatives in the comparatively mild Pacific Northwest heating with mini-splits regularly bump the setpoint up to 78F or so to achieve an average room temp of 70-72F when it's "cold" outside (= under 30F, by local standards.) Setting it to 85F, or whatever the max is would be the only way to tell if it really can't keep up. The temperature stratification issue affecting the sensed temperature is real, and only solved by installing the heads much lower on the wall, with the top of the wall unit no more than ~ 4' off the floor. Mounting them high on the wall makes far more sense in cooling dominated climates than in heating dominated climates like yours, yet most installers in my area (MA) seem to put them at the minimum clearance from the ceiling, as if it were just an air conditioner.

  31. simpsoap993 | | #31

    Dana -

    I understand that the thermostats in the house are not going to read what the same as what the setpoint is at. I was only mentioning that, because if the heads are set somewhere around 68-70F when I go to bed, I wouldn't think the thermostats in the house would be at 59F. I mean, the air coming out of the heads isn't what I would expect either.

    I think its apparent that I spent too much on the system, but honestly I didn't have much of a choice in my location. Their price was in the middle of the three quotes I received, but they seemed the most knowledgeable. I think the issue is the complexity of the system, and I'm actually not sure if they have ever done a whole house with the Mitsubishi, but instead just a single head or two here or there.

    I'm hoping to hear from them tomorrow and know what day they are coming out this week. I'm pretty stressed and frustrated about the whole ordeal.

    Thanks again!

  32. bencarsan | | #32

    The MHK1 controls heads individually. $250 ea, you can order online and install yourself. Interesting previous GBA thread on this here:
    These things don't work like other kinds of heating equipment and there are a couple of reasons you might not get "hot" air when expect to: if room air is stratified and you are using a hand held remote, then the unit doesn't see demand; and if you hit the button to raise the temp you might have to wait 20 minutes or more to get full output in response depending on where it is in its cycle. I had one unit that normally took over an hour to come up to full temp after being turned off.
    BTW, $22K for 8 heads is significantly less than I'd charge and less than most people I'd know would charge--and my margins are low.
    If this is not a thermostat problem, then your installer needs to go through diagnostic procedures. Mitsubishi has great technical support, so they should be able to walk him through.

  33. JC72 | | #33

    "...after opening up the first floor of our home and installing some beams to make it more open.."

    Do you have can lights in the ceiling of this now two-story open space? I'm thinking that you are experiencing some serious stack effect because your problem sounds just like my home (leaks like a sieve).

    What sort of air sealing have you done within the house?

  34. simpsoap993 | | #34

    Bennett - I understand what the MHK1 does, but at @ $250ea, and assuming I would need 8 of them, that's an additional $2k. What temp do you guys usually read from the heads if you use a temp probe? I'd like to know this for comparison sake.

    John - There are about 6 can lights in our kitchen, however only an additional two were added after the remodel, so I can't image we are losing that much heat from the can lights.

    As for air sealing, i've begun the process of replacing windows, but just ran out of time last year. I have a company coming to the house tomorrow to quote me on getting closed-cell spray foam place in the rim joists within our basement.


  35. simpsoap993 | | #35

    I forgot to ask, are you guys leaving the i-See sensor on with these heads, are turning them off and controlling the vanes/fan speed manually? I've read that the i-See sensor disables adjusting the vanes and such, and directs the air where it thinks it needs to be.

  36. Expert Member
    Dana Dorsett | | #36

    The output air temperature at the heads varies (a lot), but is generally north of 100F even at temps below zero, often north of 115F. It's the temperature of the intake air (at the top of the wall units) where it's sensing the room temperature.

    Infiltration losses from can-lights are big enough to be worth fixing, but it's not enough to cause system to not keep up with the load.

  37. simpsoap993 | | #37

    Well I'm lucky if i see 102F, which has happened on occasion - the other when i was about 10F outside, the air coming out was about 86F.

    What's the best solution for fixing infiltration losses from can lights?

  38. JC72 | | #38

    @Andrew Simpson

    Leaking can-lights will contribute to heat loss, but I'm think that opening up that space has in effect created a super-highway for heat to congregate in the second story ceiling. I would also think that the velocity of incoming exterior air as also increased, but someone with more experience would know.

    It's a problem with homes that have these multi-story open family rooms. Your ducted system along with a first floor ceiling would at least keep some warm circulating on the first floor.

    My unprofessional opinion of course.

  39. simpsoap993 | | #40

    @ John Clark - we opened our first floor up, but it is still a traditional style home. We don't have a multi-story family room. Essentially what was done, was my wife wanted the load bearing wall between the kitchen and dining room removed. The wall was removed, a beam put in place, and then drywalled. Hopefully this clears up the confusion a bit!

    An update for the others -- my contractor called me this morning, and it sounds like he is awaiting a call back from the Mitsubishi rep, who would like to come to the house with them to check things out. He is suppose to get a call today at 1pm, and then call me with a time they are going to come out. I'm looking forward to this. The funny thing, while on the phone with him, he still decided to state the fact that when it was cold this past weekend (10F) that the system might not keep up. I sort of just laughed at him, since these hyper heat units work down to -13F, and from what I've read on these forums, should be more than adequate at keeping up.

  40. Expert Member
    Dana Dorsett | | #41

    There's a difference between something that will " down to -13F..." and sizing things so that it will always "...keep up...". Yes, they still have a specified BTU/hr output at -13F, but it's only something like 80% of the specified max output at +5F (which by marketing-design happens to be the modulation level at which it is rated for efficiency at +47F.)

    But none of that has any bearing on whether it will keep up with your actual heat load at +10F or -13F or any other outdoor temperature. Your heat load is much higher at -13F than it is at +10F, and much higher at +10F than it is at +25F. But getting the head capacities right for the zone loads is the responsibility of the contractor, and is why they (hopefully) did a room-by-room Manual-J.

    If anything the system as a whole is more likely to have excess capacity- it's oversized for most homes that size. The fact that the contractor is waffling is an indication that they aren't very comfortable with calculating loads, which doesn't bode well for right-sizing the zone heads.

  41. simpsoap993 | | #42

    @DANA - I plan to ask for a copy of the Manual-J tomorrow, if they provide it to me is another question.

    Based on this link here:

    It appears that @ 5F that my outdoor unit has a maximum capacity of 54,000BTU/hr, am I understanding that correctly?

  42. simpsoap993 | | #43

    The contractor called and they are coming Friday morning to pump down the system and see how much refrigerant is in the system, from there they will check for leaks. I also asked about getting a copy of the Manual J, and they won't provide me a copy, stating it is proprietary information.

  43. GBA Editor
    Martin Holladay | | #44

    When you have paid an HVAC contractor $22,000 to design and install a heating system, and the contractor refuses to show you the Manual J calculation, the only possible response is, "That's outrageous."

    We need to address this problem in the industry head on. I'm all in favor of public shaming. If an HVAC contractor knows that they are at risk of public shaming on GBA, they might think twice before taking such a stand.

  44. DIYJester | | #45

    You should also note that the equipment in your setup can be purchased for just over $11,000, not including any dealer incentives or mark downs. That would leave more than half of the install cost as labor, knowledge, and expertise, which it seems like are all missing. You paid for a Manual J. I would guess they never did one, licked their finger, stuck it in the air and said "Yeah, 48,000 btu/hr will do!"

    I would also make sure the Mitsubishi is informed their Diamond contractor is refusing to give you property you paid for. It was your home, your money, it's proprietary to you, not them.

    The installer of the AC system in my house did the same thing, they threw a 5 ton unit on a 2100 sq. ft house. This meant the unit ran for about 10 minutes an hour. We had a cool, damp house all summer because it would not stay running long enough to dehumidify. When I called to ask if they ever did a load calculation he sounded as if he'd never heard of one.

  45. GBA Editor
    Martin Holladay | | #46

    It should also be noted that a Manual J is required for code compliance in many jurisdictions. (This fact is often unenforced, but it is certainly true for new construction almost everywhere in the U.S., and in some jurisdictions is also required for heating appliance replacement jobs.)

    Code compliance documents can be requested by building inspectors, and can therefore be considered (in some sense) public documents. It is entirely reasonable for a homeowner to tell a contractor, "I want to make sure that my new heating system is code-compliant. Will you please give me a copy of the code-required Manual J?"

  46. simpsoap993 | | #47

    @Mike - I understand what these systems cost, as my uncle did my detached garage for me, and it was much much less. However, he was unable to do our house, so I had to find a local contractor to perform the install. I tend to agree with you, when he told me yesterday that he wouldn't give me a copy, I immediately thought, he didn't perform one!

    @Martin - At this point, I'm discouraged and upset with the contractor and how this has played out. I'm beyond stressed and have lost many sleepless nights laying there thinking about this system. It bothers me as well that the contractor keeps saying my issue is the fact that I have an unconditioned basement. I feel like that is just an excuse, as it seems he is good at coming up with them. I guess we'll see what tomorrow brings.

    I'm also picking up 3 MHK1 thermostats tomorrow as well, and plan to install them Friday/Saturday in the downstairs and see how much of a difference that makes.

    If you want the name of the contractor, please feel free to email me at simpsoap at gmail dot com.

  47. DIYJester | | #48

    I had to file a Freedom of Information Act FOIA request with my county's lawyers to get drawings for my original structure. The people at the desk said it wasn't mine since I didn't build the house. About thirty minutes after sending my email I got a call from the county's lawyer saying all the drawings were ready for me to pick them up an 10 cents a page. We also have the Sunshine Law in Missouri which reenforces FOIA.

  48. Expert Member
    Dana Dorsett | | #49

    @ANDREW SIMPSON : The 54,000 BTU/hr max capacity @ +5F in the submittal sheet doesn't include defrost cycles, so your real average performance will be somewhat less,varying a bit with the outdoor humidity, but yeah, that's about it. If an aggressive Manual-J comes in under 50,000 BTU/hr at +5F, this system should be able to cover it. (Some vendors specify the output at a given wet-bulb temperature rather than dry bulb, which should account for defrost issues.)

  49. simpsoap993 | | #50

    @Dana - I was told I have a 48,000 head load loss, but yes, without that Manual J, I guess that was only from the contractors mouth, nothing that I have on a piece of paper or anything.

    How hard are the MHK1's to install? Pretty easy?

  50. Expert Member
    Dana Dorsett | | #51

    I'm not sure what they even mean by "head load loss". Or is that a typo, and you meant to write "heat load loss"?

    A heat load of 48,000 BTU/hr for a 2300 square foot house is on the high side of (energy use measured ) reality for most houses that size, unless your outside design temperature is well into negative double-digits. The 99% outside design temps for most of Ohio is in positive single-digits, and some locations are in double-digits:

    The heat load of my sub-code 2400' antique 2x4 framed house is a bit under 40,000 BTU/hr @ 0F, and it has a very inefficient shape (more than a dozen corners to the foundation the footprint, and three dormers). Most of it is on one floor, with a head-banger ceiling kneewalled "crazy aunt suite" bedroom & full bath in the built out dormered attic space. Your house sounds like it has full second floor, and likely far fewer corners, so at similar insulation levels & air tightness your load @0F should be quite a bit below mine (unless you have a air leaky uninsulated basement or something.)

    According to your list you have:

    1 x FH15 , for 18,000 BTU/hr of head capacity @ +5F

    1 x FH09, for 10,900 BTU/hr of head capacity @ +5F

    6 x FH06 for (6 x 8700=) 52,200 BTU/hr of head capacity @ +5F

    Add it all up and it's 81,000 BTU/hr of heating capacity for the heads, but the compressor can only deliver 54,000 BTU/hr, but you have quite a bit more head capacity than the estimated load, and more than enough head to deliver the full 54,000 BTU/hr.

    I'm not sure what limitations the branch boxed might impart to the numbers, but the heads are probably ALL oversized for their room/zone loads, and would never be even trying to deliver 80,000 BTU/hr except in an initial power on cold start situation. How their system manages partitioning the refrigerant & heat flows to the branch boxes & heads is proprietary, though some details might be buried in the engineering manuals.

  51. simpsoap993 | | #52

    @Dana - Sorry, I meant "heat load loss". Based on that document you just posted, my 99% outside design temps is 3, I'm located in Springfield, OH.

    As for my list, I have the following:

    1 x FH12
    1 x FH09
    6 x FH06

    As for my basement, it is currently unfinished and uninsulated, I have been contemplating gettign the rim joists sprayed with closed cell foam, and using batts between the joist.

  52. Expert Member
    Dana Dorsett | | #53

    Insulating the foundation walls is usually more effective than insulating between the joists. A poured concrete foundation is good for about R1, so the U-factor for the above-grade section is about 1 BTU/hr per square foot per degree difference. You probably have about 150 linear feet of perimeter, and if it has an average of 18" of exposure (1.5') that's 225 square feet of U-1 wall. When it's 3F outside and a cool 53F in the basement (barely above your deep subsoil temperatures in Springfield) that's a 53F temperature difference for a loss of 53F x U1 x 225' = 11,250 BTU/hr just for the above-grade section, and probably another 5-6K or more for the below grade walls.

    Insulating the foundation walls to IRC 2012 code-min typically takes more than 10,000 BTU/hr off the heat load of the basement, (despite the now higher basement temperatures, which will probably be north of 60F, and raises the average floor temperature of the first floor, which is a subtle comfort change, and reduces the heat load of the first floor rooms.

  53. bencarsan | | #54

    Andrew, re your contractor, what Martin said. And if you don't get satisfaction I'd definitely send a letter to Mitsubishi documenting your experience. As a diamond contractor, your installer is trading on Mitsubishi's reputation and recommendation, so Mitsubishi should be concerned if they are not competent.

    Yes, MHK1s are easy--open up the electrical panel on the head, connect the cable to the receiver, push the connect button, tell the thermostat to connect, press the button again and done. One thing they don't tell you--if you want to run the receiver unit up in a ceiling or something you can do that--they don't have to be mounted line-of-sight with the wall unit. The MHk1s don't handle all the features on the hand held, so it's useful to keep them around.

  54. simpsoap993 | | #55

    If I want to get this basement insulated, as is sounds like it should be, can someone tell me if the Dow Thermax is any better than they stuff Lowes or Home depot carries? I know the Thermax comes in white and can be left alone, can the stuff from Lowe's or home depot be painted or does it have to be covered with drywall?

  55. DIYJester | | #56

    Foamboards will need to be covered with an approved fire barrier, usually 1/2 gypsum drywall. Depending on how the wall is built this is normally required where the top of the foam meets the joist to limit the chances of a flue forming. The Dow does not appear to need this.

  56. simpsoap993 | | #57

    I wanted to give everyone an update from the contractor coming out. They ended up finding a leak at one of the connections. They first did a pump down of the system and it only had 12lbs 6oz total in the system. From the factory it comes with 10lbs 9oz, so we essentially lost everything that they had added for the additional line set. Once the leak was fixed, they filled the system back up and it's been running as I expected.

    This morning when we woke up it was actually 72F downstairs which is the first time that's happened with getting the new system installed. I have 3 MHK1's to install downstairs but havent decided if I should install them or not. Guess time will tell how it works or if we have e further issues.

    Thank you all for the help!

  57. charlie_sullivan | | #58

    I'm glad it's finally working right! I hope it stays that way!

    The party line on GBA is that one of the big advantages of mini-splits is that the installation is easy and it's rare to have a botched installation. This story at least shows that we shouldn't be complacent about that. Anyone who has spent much time on building projects knows that nothing is foolproof and there are plenty of fools around.

    The potential for leaks is a bigger concern than just the hassle and cost involved in getting it sorted out. The refrigerant is a potent greenhouse gas--releasing 12 lbs of R410A is the equivalent of about 10 tons of CO2 emissions, or the equivalent of burning 1000 gallons of fuel. That's not a catastrophe--many houses in my neighborhood burn that much fuel every winter--but it's a significant impact.

    So I come back to liking the idea of a heat pump with hydronic distribution, like the Chiltrix system. Refrigerant leaks are less likely since the refrigerant loop is only within a factory sealed unit, and the quantify of refrigerant is less. Getting the installation right is still a challenge, and it's actually more of a challenge because the skills needed don't land squarely in the domain of a trade that's already well established. But in the long term I think we need to get ramped up on doing that type of system. Leaks in the distribution system are still a serious problem, and can actually produce more damage to the building in the worst case, but at least the water that leaks out is cheap to replace and doesn't pose environmental hazards.

    We also have the question of how to deal with the upcoming phase-out of HFC refrigerants. HFOs will probably be used in mini-splits, but with a factory sealed unit that sits outdoors, CO2 and propane are viable choices for the refrigerant. I think that will make hydronic distribution start to look more attractive and become more popular.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |