GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter 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

Heat pump decision time

Jon_Lawrence | Posted in Energy Efficiency and Durability on

I have been holding off making a decision as long as possible in the hopes that a perfect solution would magically show up, but it hasn’t and I can’t hold off any longer.  I hired an engineer to perform a Manual D, J and S and the mechanical subs have done the same.  The results were pretty consistent.  The “consensus” design includes a single-zone system each for 1) family room/kitchen, 2) living room/foyer, 3) second floor lounge; and 1 multi-zone system with a medium static slim duct unit for the bedrooms/conditioned and unvented attic and a ductless wall mount for the basement. The 2 manufacturers I am considering are Mitsubishi and Daikin.  Fujitsu and LG are not options as there are no qualified installers in my area.  The issue I am having is a common one on GBA, low design loads coupled with units that don’t modulate low enough to keep the system from cycling.  This only applies to the bedroom/attic unit in my case.  My heating and cooling design loads for the 3 bedrooms and attic combined are 5964 and 7765, respectively.  In a multi-zone setup, both the Mitsubishi and Daikin 9k units minimum operating levels are greater than my design loads which should result in cycling and less than adequate dehumidification.  In fact, the Mitsubishi Manual S showed that the 9k unit could not handle the latent load. Daikin has better dehumidification capabilities and latent is not an issue with their unit.  However, the Daikin installer recommended using their zoned ducting kit so each bedroom and the attic would be its own zone, which would just make the problem 4x worse.  The alternative I am looking at is a single-zone for the bedrooms and a single-zone for the basement – or none in the basement, we don’t really have a dire need for AC/dehumidification/heat in dry NJ basements.  The 9k single-zone medium static slim ducted unit from Mitsubishi would modulate down to about 4k btu’s and would handle my heating and cooling needs at design loads (I need to confirm this for the Daikin unit).  However, neither company offers hyper-heat when their slims ducts are configured for a single-zone.   Mitsubishi locks out at -7F, and Daikin’s engineering manual only shows btu output down to 5F, so I am not sure if it locks out or not.  Although my design temp is 14, we do get an occasional polar vortex.  This would require back-up heat, which I am fine with.  I can pull out portable heaters for those really cold days/nights of the year.  So with this info, my questions are:

1)   Would I be better off with a single-zone setup for the bedrooms/attic?

2)   Should I upgrade to a 12k vs 9k unit if the minimum capacities are virtually the same, the dehumidification is better, and the max BTU’s at low single digits would make the need for back-up heat very unlikely?  I assume there is a slight energy penalty for this with a bigger compressor, but when I look at btu’s/watt at 5F, it is virtually the same for both the 9k and 12k units.


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. Jon_R | | #1

    > minimum operating levels are greater than my design loads which should result in cycling and less than adequate dehumidification

    If the indoor fan stops during the off cycle, the cycling will have little effect on dehumidification. And go very far the other way (min operating level well below design load) and dehumidification is often worse.

    > Daikin has better dehumidification capabilities and latent is not an issue with their unit.

    Really? What do they say the latent removal (pints/hr) is when AC load is near zero? Count on needing some dehumidifier use with any of them. Probably only a dehumidifier in the basement.

    > would be its own zone, which would just make the problem 4x worse.
    What is your logic here? Zoning typically reduces CFM/ton which improves latent removal.

    Without specific data (temp, %RH, load, CFM, etc) showing otherwise, you should probably throw out all your assumptions about latent capability.

    Specific models numbers would help.

  2. Expert Member
    Akos | | #2


    In a multistory house with an open staircase, a fair bit of your heating will come from the units on the lower level, as long as those can handle the low temperatures, I would not worry about the low temp heating capability of the unit for the bedrooms.

    During the polar vortex and the bedroom unit is not putting out enough heat, you can always just open the doors. Almost all splits will continue to put out heat well bellow the lowest point in the engineering manual, the output will just drop.

    If you are using a medium static ducted unit, they can handle a fair bit of ducting, maybe you can save one extra wall mount by feeding the 2nd floor lounge with it just by upsizing it a bit.

    Basement loads on an insulated finished basement are pretty low, I would check the head you are putting in there is not way oversized especially on a multi split.

  3. Jon_Lawrence | | #3

    Guys - thanks for the responses.

    Jon - as to your first comment, I agree that if the fan is off, there is no dehumidification, but wouldn't having lower minimum operating levels result in longer run times and therefore better dehumidification?

    I can't find any information on latent removal at low loads, only max levels.

    So my point on the zoning is that the Daikin dealer is suggesting a 9k head that does not modulate below 9k to service a total load of 7765, and then breaking that down into 4 zones with roughly 2kbtu's load each. So now I have a 9k head controlled by 4 separate thermostats to service loads of only 2kbtu's. Wouldn't that result in worse cycling?

    The Mitsubishi unit is PEAD-A09AA7 ( and the Daikin unit is FDMQ09RVJU


    I have noticed the stratification effects during construction and with the open staircases (front and back) the 2nd floor is noticeably hotter and more humid that the 1st floor now. Based on the experience of others that I have read on GBA I am expecting the 1st floor to provide a good portion of the heat for the 2nd floor lounge as you noted. I also am expecting the 2nd floor lounge to provide much of the cooling for the 1st floor in the summer time. The single-zone wall mount units spec'd for those two locations are hyper-heat so they will keep pumping out through the polar vortex. I was planning on upsizing those units since they will probably be serving more load than indicated by the Manual J, just at opposite time of the year. Also, thanks to Dana, I am now aware of the Quaternity unit from Daikin which has a dehumidification only mode. That unit would go high on the wall of the cathedral ceiling in the second floor which is the most humid area of the house now and I would expect it to be so once completed. As for the basement, it probably does not need any heating or cooling. The garage entrance is through the basement so there will be some heat gain/loss through there, but minimal.

  4. Jon_R | | #4

    > wouldn't having lower minimum operating levels result in longer run times and therefore better dehumidification?

    Not significantly. What causes you to think this?

    > I can't find any information on latent removal at low loads, only max levels.

    A common problem. Unfortunately, it almost always operates at less than design load. Other than trying to discover the CFM/ton, there isn't a good solution to this lack of critical information.

    > So now I have a 9k head controlled by 4 separate thermostats to service loads of only 2kbtu's. Wouldn't that result in worse cycling?

    But the total design load on the head is about 8K (4 x 2K), which is close enough to 9K to not cycle much. At lower loads, who knows how all the controls interact? It could synchronize thermostats and/or modulate zone airflow to avoid short cycle times.

    > I am now aware of the Quaternity unit from Daikin which has a dehumidification only mode.

    It can't do dehumidification without some cooling. But it will do better latent removal than other designs.

    1. Jon_Lawrence | | #5


      I am assuming lower operating levels leads to longer run times and more dehumidification as the air is running over the coil for a longer period of time. Outside of those few weeks of the year when humidity is an issue, I am also assuming lower operating levels will allow the unit to better match the load and provide greater comfort.

      I agree that the 8k load for a 9k sized head would not cycle much if that was a constant load. However, my concern is a 9k constant output unit servicing 4 separate loads of 2k coming on and off independently of each other would have the compressor turning on and off constantly. And that is my design load, not my load at 47F.

      The Quaternity unit allows you to set a specific humidity level if desired. It will then maintain that level without cooling the air by "mixing the cool dry air with warm air via its intelligent indoor heat exchanger technology." Apparently they are able to turn the indoor unit into both an evaporator and condenser at the same time by dividing it into 2 parts. The evaporator part cools and dehumidifies, the condenser part warms as needed.

  5. Jon_R | | #6

    > more dehumidification as the air is running over the coil for a longer period of time.

    This assumption is often incorrect. One can easily find systems that provide more dehumidification when running for shorter times. Focus on CFM/ton (huge effect), not run time (very little effect).

    > It will then maintain that level without cooling the air

    Daikin says a vague "controls the indoor humidity level" (not your "without cooling") and has done a poor job of communicating about this. Elsewhere they say "without lowering the temperature", but then you realize they are talking about not lowering the thermostat (over-cooling down to 70F) to get more latent removal . You won't find data indicating that they can provide enough warming to fully offset the cooling (SHR = 0).

    1. Jon_Lawrence | | #7

      The max airflow rate in CFM for the FH09 is 381 dry/328 wet and the moisture removal per hour is .6 pints. The max for the same size Quaternity is 420 (they don't specify wet or dry) and 3.3 pints per hour.

      That is not my quote that is Daikin's. Also, they are not mixing it with warm interior air per se, it is heat that would have otherwise been evacuated to the condenser outside. Not sure if that make a difference or not. Even if it does cool the air slightly in dehumidification mode, won't the air still be warmer than if operated in dry mode? Isn't that better if you are not looking for cooling, just moisture removal?

      1. Jon_R | | #9

        It's definitely an improvement over other systems.

        > 3.3 pints per hour

        This works out to only ~3200 btu/hr of moisture removal in ?? conditions. Not conclusive, but I expect that too much heat is being dissipated outside to reach zero sensible cooling. Would be trivial to measure output air temp if anyone has one.

  6. joshdurston | | #8

    PEAD-A (medium static ducted)
    9000 btu / 12000btu
    min CFM 318 /318
    min cool btu 4300btu/5200btu
    Looks to me like the 1 ton unit might be the way to go since that extra 900btu is going to be likely mostly latent capacity. But 900 btu isn't much and may not be worth the cost delta. Dry mode will up the pints/hr dramatically, but the downside is you have to baby sit it in dry mode to avoid overcooling in mild weather. A portable dehum might be worth having regardless.

    IMHO, don't zone it, that's asking for trouble. Spend your money on higher end duct work instead. This should include some means of balancing the airflow.

    And get the MHK1 wall controller, or the Kumo Cloud interface with the remote space temp/humidity sensor. Both of these solutions are wireless and will let you position your space temp sensor to optimally sense and respond to the load.

    I don't have any experience with the Daikin unit, but I think some of the disappointment with Mini Split dehum is due to misguided expectations and some due to control algorithms being too efficiency focused.
    I suspect that during cooling operation the Daikin unit can modify the coil temperature to run colder if the humidity is high to enhance it's moisture removal rate, whereas Mitsubishi seems to vary the coil temp and fan speed based on dry bulb alone, unless you manually set it to dry mode. So if you have any sensible cooling load, the Daikin probably would outperform the Mitsubishi for dehumidifying. But if you take away the sensible load than it's going to be more satisfactory to use a dedicated space dehumidifier. It's unclear what Daikin is doing with their coil, I'm not sure if they are actually simultaneously heating and cooling, or if they are just shrinking the evaporator (by isolating a portion of the coil so room air can bypass and mix into the supply) to run it colder longer without overcooling as much (but still inevitably overcooling). To actually reheat as a condenser function I think you would need a minimum of a 3 or 4 pipe lineset, does the Daikin have more than 2 pipes in the lineset? Historically there has been some disappointment with the Daikin system not meeting expectations, but the expectation of not needing a dehumidifier at all may have been too high.

    In the quest for higher SEER ratings I think manufacturers have sacrificed latent capacity since it is much more energy intensive to keep a coil cold enough to dehumidify. SEER ratings above the low 20's in a humid climate are a farce in my opinion (as someone who has a SEER 30 rated FH09). I don't know the details but I suspect the testing to get a SEER rating doesn't have enough of a humidity load component, or is done at a unreasonably high space temperature of like 80F, where it's easier to have low relative humidity readings (because relative humidity is "relative" to temperature).

    1. Jon_Lawrence | | #10

      Thanks Josh. I agree the 12k is probably the way to go for the better latent control.

      I have attached a link to the Daikin Quaternity landing page with the explanation of how it is able to run in dehumidify mode without cooling the room. I am not sure where reality ends and marketing takes over though.

      1. joshdurston | | #11

        Hmm, their literature clearly says that they are using part of the coil as a condenser, but to do so they have to hold the expansion valve open. I suspect this is why it only works in a limited range of outdoor temps (apparently it doesn't work that well with below 70f ambient outdoor temperatures). If it gets too cold outside, even with the outside fan stopped the gas is probably fully condensed into a liquid before it gets inside leaving very little reheat capacity. They must have a secondary metering device like an orifice or capillary line in between the two indoor coil sections in dehum mode, maybe with a solenoid to switch it out when in regular mode. Fascinating setup. Perhaps if they had some OD coil bypass mechanism and maybe motorized louvers on the outdoor unit you could trap more heat at lower temps. (I've worked on some low ambient cooling LG Multi V systems that relied on dampers and hoods to keep the outside coils from getting overcooled in the winter).
        Even with it's shortcomings, I suspect it would in fact dehumidify much better than most units. Using only part of the coil is similar to having less CFM since only a portion of the air gets cooled. It's better to cool a little bit of air a lot (well below the space dewpoint), than a lot of air a little to dehum with a fixed BTU/hr capacity.

        Other units are definitely more efficient heaters with better HSPFs.

        1. Jon_Lawrence | | #13

          Josh - How were you able to determine that the unit does not operate well below 70F?

          1. joshdurston | | #14

            Anecdotal reports on other forums. Not direct experience. I still think its a worthwhile feature as long as you're aware of its limitations. I think they technically say it can work down to 50f, but it appears the reheat falls off as it gets colder.
            Once it's colder than 50 you can ventilate to dehumidify anyways.

            And some of my thoughts on how it might work are merely musings and informed guesses.

  7. Jon_Lawrence | | #12

    Josh - Thanks for the thorough analysis. It sounds like this unit really is a good fit for what I am asking it to do - provide dehumidification and/or cooling in a 2nd floor space with a cathedral ceiling. Since this space is open to below, it will probably be cooling the first floor as well, so I will size it as such. I will not be expecting it to do much heating, which as you noted is not its strong point, as the first floor unit will probably provide most of the heating. That first floor unit will either be a Mitsubishi FH12 or a Daikin Aurora FTX12.

  8. Jon_Lawrence | | #15

    Josh - I think I found one of the threads you may have seen - someone trying to use this unit to dehumidify 59F, 70% air in San Fran. Thankfully, my situation is different, trying to dehumidify 75F, 60%+RH.

    1. joshdurston | | #16

      Don't take it that I'm trying to dissuade you. The Quaternity is a interesting unit with some unique capabilities that I wish I had in my home sometimes.
      As long as your usuage aligns with the intended usuage case (AC with enhanced dehum), not low temp dehum (where a dedicated dehum will better suite), I think you will be satisfied.

      1. Jon_Lawrence | | #17

        Josh - No I did not take it that way at all. I am always looking to hear pros and cons before making a decision. You have highlighted a very good point about heat pumps which is that they have their limitations. Yes, they can be more energy efficient/comfortable/quieter, etc., but only if the right unit is spec'd. It appears that the Quaternity is the right unit for the location/use I plan for it. I would not try to use this, or probably any other heat pump for dehumidification of a cold basement.

  9. 1910duplex | | #18

    Jonathan, did you end up getting the Quaternity for upstairs? Did it provide the comfort you are seeking?


  10. frasca | | #19

    Also interested to hear how this performed!

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |