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

Help with Fujitsu ducted minisplit selection

alan72 | Posted in General Questions on

Fujitsu 9RLFCD versus 12RGLXD

CZ 5 (northern part of it)

I have included Manual J calculations for 2 zones in our home build that we are specifying ducted minisplts.  The HVAC contractor wants to use the 12RGLXD unit due to concerns about static pressure issues.  The engineer who did the Manual J initially recommended the 9RLFCD unit, but thinks the 12RGLXD will be ok because it’s variable capacity.

I don’t want to oversize the unit esp RE latent load while cooling.  Any advice?



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. MattJF | | #1

    So either would work. The 9RLFCD will produce more than 15,400btu at your design temp, so no capacity issues.

    The static pressure advantage is not what it appears on the specs at first. The 9RLFCD runs 383 cfm, while the 12RGLXD runs 500 cfm. That increase in cfm that needs to be moved is a big deal. Connected to the same ducts, the 12RGLXD will run at 2x the static pressure of the 9RLFCD (500^2/353^2=2.0). There is still a static advantage to the mid static unit, just smaller than expected. It’s about a 11% advantage vs what looks like a 120%

    The 9RLFCD has a .5 better HSPF. It should also be able to do a bit better on the days with high latent, low sensible loads due to lower fan speed. Central returns without transfer ducts are also easier with the lower cfm. You can get about 80cfm though most closed doors according to John Semmelhack.

    What’s the layout like? Do you really need two zones? One 12RGLXD could do both zones.

    *Edited, had the CFM for the 9RLFCD as 383, when is actually. In this scenario, there is very little static pressure advantage on the larger mid static unit

  2. alan72 | | #2

    It's a large ranch - we have separated it into 4 zones - these 2 zones are on opposite sides of the house... I attached a layout.

    The master zone (light blue) has south and west facing windows and no transfer ducts. - so maybe the 9RLFCD?

    The 'back right' zone (green) has transfer ducts and some south facing windows - maybe the 12RGLXD?

    If we use the 12RGLXD, should we re-design the ducts?

    I think we have the other zones worked out...


  3. MattJF | | #3

    See the edit to my post above, where I corrected an error in the 9RLFCD cfm and conclude there is barely any static pressure advantage to the 12RGLXD. The ductwork needs are pretty much the same for both.

    I would stick with the 9RLFCD for the better efficiency.

    That being said, I missed an important aspect in my original post. According to the actual heating capacity table, the 9RLFCD produces 15,400 BTU at 5F.

    That is more than 2.1 oversized, which is sub optimal. Did you specify the zones or did Energy Vanguard? I think I would try to get down to 2 or 3 zones. In relatively low load houses with properly sized hvac, zoning is less critical.

    I don't know why your manual J lists the equipment capacity as 12,000. Several of the specs in the Equipment Summary are not correct (300cfm vs 353cfm). The load calcs also don't appear to include ventilation.

    Your ventilation rate of 215 cfm seems high. I don't know how many SF the place is though. 40 CFM going into the master is on the high side though. Read this:

    1. Jon_R | | #7

      > That is more than 2.1 oversized, which is sub optimal.

      In what way? Do you have data supporting this?

  4. alan72 | | #4

    The zoning was designed by the engineer. I think static pressure issues were the main reason why there are so many zones...

    Our HVAC contractor is still concerned about the static pressures and suggested the mid static ducted unit. The 9RLFCD is a little oversized and the 12RGLXD more oversized... I'm not an engineer and I'm not a contractor so I posted the question... I'm not particularly interested in resdesigning the system - it has been a lengthy process to get to this point... I think the number of zones was a series of compromises - like most things in building a house...

    RE ventilation - I've read a lot about ventilation and standards for ventilation didn't make any sense to me w/o some sort of control... It's like a standard telling me that the heater needs to run at a certain rate without something telling the heater when to stop heating... my personal comfort should correlate with the temp setting. I know that doesn't exist for air freshness, but at least a co2 monitor can let me know that we are moving air, so we are using a CERV2 -- maybe 40 is excessive. I'll take it back to the HVAC contractor and the engineer...

    1. MattJF | | #9

      Make sure the contractor understands the impact of increasing or decreasing CFM on the static pressures. There is very little static advantage to the mid static system. The pressure drop in a duct is related the Velocity^2, so small changes to the flow rate/velocity have a big impact.

      For the same ductwork:
      CFM1^2/CFM2^2 = P1/P2

      Did Energy Vanguard do the duct design and sizing? As long as the duct velocities are sub 450 fpm and the filter is bigger than 2.4 SF/400 CFM, you should be good. I would stick with the 9RLFCD, but either will work.

      As far as sizing, from the manual J, you will hit the min capacity of the 3100 BTU at 42 Degrees and probably mid 30's in actual use due to the fact that manual J overestimates loads in the first place. That does leave the potential for the unit to be cycling during a significant amount of time. Fujitsu is the only ones that might be able to calculate the impact of 2.1x over sizing vs. 1.25x over sizing. I haven't looked at the daily temperature bins to get an idea of how much time might be spent cycling and what percent of the annual load that would be. Page 24 of the PDF for this study gives some idea of the COP impact of cycling:

      I didn't actually know what a CERV2 was. If the system is installed as spec'd, you can experiment with throttling it back. It seems like it is smart and should throttle back on its own, although I didn't details on if it modulates or turns on/off. It is useful to have some monitoring feedback to guide ventilation, but it is a bit different than temperature. With temperature you feel hot or cold while many aspects of ventilation involve how you will feel over a much longer time period. Using a monitor and manual adjustments is practical, unlike temp control. What you have should work.

      1. Jon_R | | #11

        > Page 24 of the PDF for this study gives some idea of the COP impact of cycling:

        Because the power was cycled (not what happens in real situations), their results are completely irrelevant. Actual degradation is far lower than the .42 they measured for the Fujitsu and usually higher than the negative value (cycling increased low load efficiency!) they got for the Mitsubishi. You need to look elsewhere for accurate cycling degradation coefficients. Probably about .12.

        1. MattJF | | #13

          What page are you getting the numbers you cite?
          Edit - I see you are referring to the cooling cycling data on page #23/pdf page 31.

          Heating cycling data is on page #16/pdf page 24. The Fujitsu has a .27 degradation factor, while the Mitsubishi had a .45 degradation factor.

          I don't know exactly how the cycling degradation plays into the overall efficiency. It would be informative to look at how much heating occurs during conditions that would cause cycling. From daily HDD data for the past 3 years, 29% of HDD occur in conditions that would cause cycling. It would be worse somewhere with a longer more moderately cool winter, like Seattle where the same home would see 84% of days in cycling conditions.

          Honestly, I am going working with the general guidance from people who know more than I do.

          1. Jon_R | | #15

            In summary, there doesn't seem to be good data on inverter heat pump sizing for optimal annual efficiency. 10% undersized (using supplemental resistance heat), sized to design load (also needs some supplemental heat), 25% over (about right for no resistance heat), 40% over, 100% over?

            And it all changes depending on if you are using design load or peak load and rated capacity or maximum capacity. The same unit can be called under-sized and over-sized.

  5. Expert Member
    Akos | | #5

    It looks like the ducting was sized for the max flow capacity of the unit. Since your load is quite a bit less than that, the unit should not need to run at max. If the the ducting with the low static head is "marginal" at max, it should be fine at a realistic load. I see little point for the larger unit (plus mid pressure units usually cost more).

    One thing, I see both supply and return in the master bed. I don't know how the ducting is in the basement, these units are pretty quite but that could still be a fair bit of noise. Couple of bends and some straight runs of flex duct might be good idea if you have the pressure budget.

  6. Jon_R | | #6

    Alan, the latent performance figures are largely nonsense. Take the minimum CFM of the indoor unit (eg, 300) and a typical (not design) cooling load. Calculate a CFM/ton. If this is high (like 650+ CFM/ton), then there will be NO latent removal. Using your numbers - you will need a dehumidifier.

    Do not rely on the outdated and often just plain wrong: "a closer match between rated capacity and design load will produce better latent removal". Ditto for efficiency.

    30-40 CFM of fresh air into a bedroom with two people is a good value for health.

  7. Jon_Lawrence | | #8


    Unless you see a real benefit to having the master on a separate unit than the other bedrooms, what about making that 1 zone using the 12 RGLXD? Assuming those other bedrooms combined have about the same load as your master, that should put your design load much closer to, albeit still with a buffer to the capacity of the 12RGLXD. As for latent removal, the low CFM output of the 12RGLXD is 300 vs 265 for the 9RLFCD, not a huge difference and I would think the bigger evaporator should help offset the slightly higher minimum CFM as it relates to dehumidification.

    I have a 12RGLXD servicing the master and 2 bedrooms in a space that is 63' x 17', so just about the same size as your bedrooms combined. We have not had real cold temps yet, and we are still under construction (just finished drywall) but I have had it running with temps in the mid-50's and I the only way to tell it was running was to stand under the supply register.

    Before the mid-static unit was released, I thought about using a slim-duct, but no one would install it because of the static pressure requirements (.7). I have the 9RLFCD installed in the open plan family/kitchen/living room with a single supply register. It works well in that space and I can actually feel the warm air coming out 30' away, but there is very little static pressure in that setup.

    1. Jon_R | | #10

      > the 12RGLXD is 300 vs 265 for the 9RLFCD ... as it relates to dehumidification

      It depends on output. For example, at a continuous 4800 Btu/hr, you have 300/(4800/12000) = 750 CFM/ton or 265/(4800/12000) = 662 CFM/ton. So no dehumidification either way.

      On the other hand, if you could get them to run at 9000 Btu/hr and cycle to match the load, dehumidification should be fine with either one.

      IMO, "dry mode" often has issues that make it a non-viable solution.

      1. Expert Member
        Akos | | #12

        With a ducted unit, you can always partly close the dampers to reduce the flow if there is not enough de-humidification. You can also program in lower delta P into the unit so that it will run the fans at lower speed which would have the same effect. Dry mode is not always the best as it tends to overcool the place.

        The bigger issue is over-sizing the unit to the point that it effects comfort (ie 9k head in a small bedroom) which doesn't seem to be the case here.

        1. Jon_R | | #14

          Agreed, reducing airflow is an option. How practical/convenient it is is unclear. For example, what happens when you have SP-Mode set for reduced airflow and then return from vacation and it runs at full output? Does it ice up? Shut itself off? Automatically increase the SP-Mode?

          Not clear to me is why they don't design these things with a wider fan range and controls such that % latent removal is increased (vs eliminated) at lower loads.

          1. MattJF | | #16

            I know Semmelhack has mentioned he adjusts the SP mode setting seasonally. I don't know how far he is pushing it.

            On the heating side you should be able to push CFM up until the supply air is too cool for comfort or the fan makes too much noise. This might decrease fan life.

            In cooling, you risk freeze up and maybe there is some aspect of the system not operating properly with a very cold indoor coil.

            The lack of fan turn down is a bit puzzling. I guess one issue is providing that turndown/up for all static pressures. If you design in the middle of the static range, it does seem like you have some options though.

          2. joshdurston | | #18

            There are enough temperature and pressure sensors on most minisplits to avoid freeze up. I agree about dehum being problematic.
            Dry mode would be more useful if it was simply cooling but with a lower cfm/ton while still running off of room temperature. I don't understand why most manufacturers don't program a minimum space temp into dry mode.

  8. alan72 | | #17

    This discussion is awesome - thanks!

    What are your recommendations? Stay with the 9RLFCD units or switch to the 12RGLXD units?


    btw - I have asked our engineer what he thinks about combining the bedrooms in one zone serviced by one 12RGLXD - I think we discussed this and went in the 2 zone design for some reason - I have failed to keep accurate notes that reflect why some decisions weren't pursued...

  9. MattJF | | #19

    I believe Fujitsu will control to the room temp set point running -1 on -3f off. See page 6.

    The 12RGLXD produces a bit more than 16,700 btu at you design temp. What is the combined load of both bedroom zones?

    1. alan72 | | #20

      Total heating load is 15895 Btuh (7171+8724) and total cooling is 7986 (3630+4346)

      I added the Manual J for the Bedroom zone in my original question above.


      1. Jon_R | | #23

        > Total heating load is 15895...

        More accurately "Total design day heating load is estimated to be 15895...". Note that it gets significantly colder than design day and you need either extra HP capacity or supplemental heat to cover this.

  10. Jon_Lawrence | | #21


    One of the conclusions I came to when I was trying to decide which unit/brand to go with was that there is nothing out there that is going to provide me with a perfect solution so I decided that I was going to focus on the heating requirement since I live in a heating dominated climate. I currently have forced air in a drafty house and it is always uncomfortable because it is either too warm and noisy when the forced air is on, or too cold when it is off. I figured that issue would go away with a heat pump system that was flexible enough to meet my design temp requirements but also modulate low enough to minimize cycling when loads are low. However, with such low loads, it was tough to find a unit that was not oversized and adding more rooms to any unit resulted in excessive static pressure. Thankfully the 12RGLXD came along it seems to be working well with the low loads as each of the bedrooms feels like a constant warm temperature. I am going to see how it performs in near design temperatures later this week as we are going to see low 20's on this weekend.

    While this system should work well for my main goal of heating, the concern I still have is that it may be less than optimal in the cooling season because of humidity. The reality is that we have about 3 weeks of uncomfortable dog days of summer and we get a few days in the spring when it is rainy but 72 degrees so humid but no real need for heating or cooling. I don't mind pulling out a portable dehumidifier if needed during those times in order to have a system that meets my heating needs.

    I just don't know that this will be a big issue simply because the house is so tight and well insulated that outside humidity does not have an immediate affect on the inside humidity. I had continuous insulation (not including dense pack yet) and air barrier finished before the start of the summer. Those dog days we had this year had little affect on the indoor humidity. We worked inside with the windows shut and temps stayed in the mid 70's and it was comfortably dry. More comfortable than my current house which was either freezing when the AC was running or warm and clammy when it cycled off. I did not have an ERV running yet which should introduce more humidity on days like that and we were not living there cooking, taking shower, etc., which should also add humidity. However, if it is an issue, I can always turn to the portable dehumidifier.

    1. MattJF | | #22

      A 12RGLXD should cover all the bedrooms. Jonathan's point that perfection is not really achievable is well taken. I think we have poked and prodded the original plan a fair bit and it holds up fairly well with reasonable compromises taken to make it all work. A very large rambling house like this is more difficult to optimize than a two story box.

      The other bedroom manual J lists a 12RLFC. If not combining zones, I would go with another 9RLFC, which more than cover the load.

      If you have a normal door on the master bedroom, I would put a transfer duct. If it is a double door or no door, you should be good.

      1. alan72 | | #24

        Yes, Jonathon's point is well taken - and all decisions seem to be a compromise - which is basically true for most things in life. My thinking is that I would rather manage the latent cooling load better and use auxiliary heat (sweater, space heating) as needed - I'd rather clear the humidity effectively.

        My impression is that the 12RGLX is too large for the Master zone. If we keep the Master and bedroom zones separate - I think we'll stay with the 9RLFCD for the master.

        The other bedroom zone has the 12RGLX specified - again for the static pressures... sounds like you would rec the 9RLFCD for the bedroom zone?

        Thanks again.

        1. MattJF | | #25

          Yes, with the zones as is, it seems like a 9RLFCD should work fine for the bedroom zone.

  11. alan72 | | #26

    Thanks again for all the input.

  12. Jon_Lawrence | | #27


    I think you are making the correct decision to design based on your primary objective and then plan for auxiliary/backup equipment to address any shortcomings. Good luck!

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |