Replies

1. Expert Member
   Dana Dorsett | Oct 17, 2016 03:10pm | #1

   Not all 9K & 6K Mitsubishi heads are alike. The FH09 and the FH06 have identical output at the low end of the modulation range, (1700 BTU/hr in cooling mode, 1600 BTU/hr in heating mode) and swapping out an FH09 that's cycling on/off at low speed with an FH06 will have no benefit whatsoever- it will cycle just as often.

   You are definitely being constrained by the high minimum speed of the compressor in the MXZ-5C42NAHZ. The minimum operating output in cooling mode of the newest MXZ-5C42NAHZ is 6000 BTU/hr (see: https://www.goductless.com/hvac/pdf/mitsubishi/mitsubishi-multizone/MXZ-5C42NAHZ-sb.pdf ),which is 2/3 full output of an FH06. So when only one head is calling for refrigerant it MUST run at at least 6000 BTU/hr. That's only half-speed for an FH09, and 2/3 speed for an FH06, but any time the load is lower than a half-ton (which is nearly all the time) it's going to cycle rather than modulate, no matter which head you have, since the compressor won't modulate lower than a half ton, even though the individual heads could, if operated with a lower-modulating compressor.

   And you say the techs are telling you yours only modulates down to 12,000 BTU/hr? All of the newest versions modulate down to 6K in cooling mode, even the MXZ-8C48NAHZ:

   http://www.mitsubishipro.com/media/989222/mxz_h2i_productlaunchbrochure1stprint.pdf

   Read the nameplate model number carefully, see if it isn't a different model. Even the older non "-HZ" model of the MXZ-5CNA modulates down to 6000 BTU/hr:

   https://www.acwholesalers.com/hvac/pdf/mitsubishi/mitsubishi-multizone/MXZ5B42NA-sb.pdf

   It's true that the 3 ton air handler on the main level can't go that low, but the MXZ-5C42NA (and NAHZ) compressor can.

   It's isn't likely that even the peak cooling load of a 165 square foot or 150 square foot bedroom is anywhere near 6000 BTU/hr unless it's on the west side with a big west facing window, and similarly, peak heating loads are far below the output of an FH06, let alone an FH09. But since the low end of the modulation range is the same with either, you won't gain anything by swapping it out. A mini-duct cassette serving both rooms may have done a bit better, but not a lot better- you simply don't have enough average load to keep them spinning. Even the peak heating & cooling loads of those two rooms are probably only slightly above the minimum-modulated output of the individual heads.

   The 3 ton air handler for the 1223 square feet of main level is also likely to be extreme overkill. That's 407 square feet per ton of load, whereas a typical house that size would be on the order of 1000 square feet per ton, and a low-load home could easily be 2000' per ton, per Allison Bailes' compiled Manual-Js on real houses:

   https://www.greenbuildingadvisor.com/sites/default/files/images/Bailes%20graph%20for%20Manual%20J%20blog.preview.png

   The humidity issue is very likely to be aggravated by the fresh air duct on the HRV system, which should be closed any time the outdoor dew points are north of 55F if you're hoping to hold the line at 50% relative humidity.

   If you haven't already, be sure to run the FH09s in "dry " mode, which maximizes the latent cooling during the on-cycles even when it's cycling. (Unlike some cheaper mini-splits, I believe the FH series will still control to a temperature set point, and not just keep on cooling until the mode is changed.)

2. 
   benthrush | Oct 17, 2016 04:55pm | #2

   Thanks so much Dana for the information Dana, your input is most valuable and I appreciate the time. The 3 ton air handler is serving 1223 square feet on main level of house and 1223 square feet in full basement. Do you think I'm still way oversized? How low can the air handler modulate down? I was also concerned about the fresh air duct adding to the humidity problem and even tried the system for a few weeks with it completely shut which did not seem to help. Do you think the idea of taking one of the 9K mini splits and placing in hall and eliminating the other one all together is crazy? Should I consider adding a remote thermostat for mini..? I've also attache pics of exact outdoor unit and mini models.

   Thanks!!
   Ben

   
3. Expert Member
   Dana Dorsett | Oct 17, 2016 06:35pm | #3

   It's definitely the NAHZ version of the 5C42, and it should be able to modulate down to 6000 BTU/hr in cooling mode (not that it saves you from the cycling). If the Mitsubishi techs are telling you 12,000 BTU/hr min, make them tell you why it's twice what the submittal sheet specifies: http://meus1.mylinkdrive.com/files/MXZ-5C42NAHZ_Submittal.pdf

   Yes, I think the 3 ton air handler is way oversized. Even if the basement were a walk-out (is it?) , there's almost no way a tight dense packed 2x6 house with an insulated basement would have loads that high on the first floor unless you are over-ventilating the house at some insane level.

   The following is well short of a Manual-J, more of a WAG based on where real heating & cooling load numbers tend to fall.

   At modest ventilation rates the cooling load of an insulated basement is almost nil, dominated by the latent cooling load, probably no more than 1/4 ton unless it's a west facing walk out with a big slider. The first floor's cooling load is probably less than 1.5 tons unless it has a lot of unshaded west facing window area .

   The heating load of an insulated basement is also very low compared to fully above grade floors. A dense packed 2x6 wall house with code min windows and a reasonable ventilation rate at your area's ~+10F outside design temperatures will typically fall 10-12 BTU per square foot of fully above grade conditioned spae. With the basement included in that zone I'd expect your heat load for basement + first floor combined to come in between 15,000 -18,000 BTU/hr. You can probably heat & cool that zone with the 1.5 ton MVZ A18AA4 with margin:

   http://meus1.mylinkdrive.com/files/MVZ-A18AA4_For_MXZ_MULTI-ZONE_SYSTEMS_Product_Data_Sheet.pdf

   To specify "rated capacity" of 20,000 BTU/hr as a nominal heating capacity it also has to be tested at that level with a 70F room and a +17F outside design temperature. The"rated capacity" is the modulation level at which efficiency is tested at +47F, not it's maximum capacity at +47F. While capacity may fall a bit with temperature, it'll still have over 18K @ +10F (it might even have the full 20K, but I'd have to dig up other documentation to figure out it's +10F capacity when married to the 5C42.)

   As an absolute worst case the MVZ-A24AA4 would do it, but is likely to be overkill if it isn't a walk-out basement, with half of the wall area above grade.

   https://meus.mylinkdrive.com/files/MVZ-A24AA4_For_MXZ_MULTI-ZONE_SYSTEMS_Product_Data_Sheet.pdf

   Before making any changes (if any) to the configuration, run a room-by-room load calculation at the 1% and 99% outside design temperatures. Hire a competent engineer or RESNET rater (but NOT an HVAC sales-droid) to run those numbers, and tell them to use aggressive rather than conservative assumptions on infiltration rates. When you have the numbers we'd be able to guess whether an FH09 in the hall makes more sense than one per room. I suspect in cooling mode you could cool the whole house most days with a single FH09 in the upstairs hall, turning off the 3 ton air handler (or setting it's thermostat a few degrees higher) and letting the FH09NA run, but it might become a 5AM comfort issue during the depths of the heating season.

   That said, one of the guys in the office here lives in a 2x4 framed antique bungalow, and heats his 3-room upstairs with a single FH12NA upstairs, and another in the more open floor plan area downstairs, with an FH09 managing the doored off kitchen zone. His loads upstairs are within range of the FH09NA, but the higher cfm of the 1-ton head improves the open-door distribution a bit better when it's cold enough to matter. His house is in near suburbs of Boston, with a 99% outside design temp in high single-digits, maybe as high as +10F, comparable to or a few degrees cooler than Garrett County, MD. His total square footage above grade is comparable to yours, but his basement has no foundation insulation and is not heated (though he uses it daily.) In his case he was for-armed with the load calculations, and pushed back on (or rejected completely) most of the contractor proposals, most of which involved nonsensical amounts of inappropriate equipment. Out of a good half-dozen or more proposals only two made any sense at all, and we sculpted it a bit from there. I was still leaning toward keeping the upstairs unit a FH09, but he didn't want to risk a shivering unhappy spouse in January, and felt he could tolerate a hit in efficiency more readily than years of making amends to her. :-)

4. 
   benthrush | Oct 17, 2016 10:18pm | #4

   Ok wow, lots to digest. If i end up moving 9k unit to hall and switching out air handler to smaller MVZ-A24AA4 or MVZ A18AA4 would I still keep the same outdoor unit or switch it out as well? Another thing I noticed about the 3 ton air handler is that it produced very little condensate even during hot/humid days this summer. Another sign of oversizing? I do have the building analysis that engineering company provided which I'll scan and attach tomorrow if you don't mind reviewing. I do have several concerns about calculations as I feel they do not reflect the actual project given the construction quality of the home along with wrong location (used closest location in software program) etc...

5. 
   Dana1 | Oct 18, 2016 08:20am | #5

   The minimum modulation of the outdoor units is the same across the _____NAHZ multi-splits, so even if you down-size the air handler it won't need to be swapped out. I don't have the engineering manuals for the MVZ air hanlders, but IIRC they have roughly a 3:1 turn-down ratio, so the A18 should be able to cruise along a the ~6000 BTU/hr minimum speed of the 5C42, whereas the A36 can only modulate down to ~12,000 BTU/hr. The ducts may or may not have to change if the air handler is down-sized.

   If you had it set up for the A36 to run continuously whether it was actively cooling or not there would not be much condensation., but otherwise there should have been a decent amount. I don't know if there is a "dry" mode for those units the way there is with the FH09 heads, but if there is, that's the way you should be using it if the humidity levels are too high.

   It's possible to put an upper bound on the whole-house heating load by analyzing the electricity used during the shoulder seasons(to establish baseline for other power use) and the winter season bills, using exact meter reading dates and weather data from a nearby weather station, if you're up for that. The basic methodology live here, but it's a bit more complex when using variable speed (and variable efficiency) heat pumps:

   https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/out-old-new

6. 
   benthrush | Oct 18, 2016 11:03am | #6

   Thanks Dana - I thought you might find the attached calculations interesting. The A36 does indeed run continuously, is this good, bad, able to be changed? In an effort to get the system up and running sooner rather than later I'm thinking of going to single 9k mini split in upstairs hallway as change #1. The big question then is to keep the A36 or go down to A18 or A24. All of my ductwork is in place and concealed so changing ducts would be a huge project. I would say about 25% percent of my basement is day-lighted, slab and walls are all insulated with 2" of eps foam and walls have additional fiber glass in 2x4 walls. Home is on heavily shaded wooded lot. What is your gut instinct? All the best.
   Ben

   
7. Expert Member
   Dana Dorsett | Oct 18, 2016 02:55pm | #7

   First off using an outside design temp of -5F for Morgantown WV is just nuts! The gold standard is to calculate the load at the 99th percentile outdoor temperature, and up-size slightly from there. The 99th percentile temperature bin for Morgantown is +11F (that's a 16F difference!) using ACCA 's data set:

   http://articles.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf

   According to ASHRAE's datasets even the 99.6th percentile temperature bin for Parkersburg Wood County airport (colder than Morgantown) is +5.4F, and the 99th percentile bin is +11.8F :

   https://www.captiveaire.com/catalogcontent/fans/sup_mpu/doc/winter_summer_design_temps_us.pdf

   The load calculation details aren't included, and with errors that large it make me wonder what other errors of omission & commission were committed. (his has the hallmarks of an HVAC contractor's Manual-J, not an aggressive & precise engineering or RESNET rater's calculation. ) They described the construction as "average", but dense packed 2x6 walls and better-than-code insulated basement are not "average". Your house was described as:

   Main Level - 1,223 sq.ft. - ( Dense packed cellulose)
   Upper Level - 586 sq.ft - (Dense packed cellulose)
   Basement - 1,179 sq.ft - (2 inch EPS Foam followed by 2x4 wall with unfaced fiberglass)
   Attic - blown cellulose - R44

   They calculated the load at a 75F temperature difference (70F indoors, -5F outdoors), whereas the 99% design load is closer to 60F (70F indoors, +10F outdoors). Scaling their calculated 40,301 BTU/hr to a 60F delta-T gives us (40,301 x 60/75=) 32,241 BTU/hr @ +10F which is on the very high side for a tight house that size, (about 18 BTU/hr per square foot of above-grade conditioned space) which makes me suspect other errors.

   As reference points, I've seen 2x4 framed houses about that size with completely uninsulated basements come in at about 30-35K range @ 0F, and the house you described is better than that. I live in a ~2400' 2x4 framed house with sub-code attic insulation dense packed walls, clear glass antique windows with clear-glass storms, and 1500' of ~R15-ish insulated basement, with a whole-house load in the ~35K range @ +5F outdoors.

   My gut tells me your whole house load is probably not more than 27,000 BTU /hr for the whole thing, and probably closer to 25K. The basement + first floor is proably about 18K (but could be as high as 24K depending on total window area and type), and the second floor maybe 6000-7000 BTU/hr, 9K absolute max (depending on windows & configuration.) But this is nothing but a WAG, not a load calculation.

   Set up a spreadsheet tool to run a room-by room I=B=R type load calculation. For dense-packed 2x6 walls 16" o.c. framing will have a U-factor of about U0.065 with most siding types. For the R44 attic use U0.028. You'll have to describe the windows better to come up with a reasonable U-factor for the windows. We've determined that a 60F temperature difference is the approximate 99% design condition, so for each exterior surface type (window, wall, upper floor ceiling, door, etc) calculate:

   Area x U-factor x 60F = BTU/hr.

   For the insulated basement, only include the above grade wall area, but the U-factor of the a 2" XPS + 2x4/R13 + concrete or CMU wall is about U0.045.

   With those numbers we can play "what if" games on the likely infiltration rates, etc and hit pretty close to a full-on Manual-J.

   See: https://www.greenbuildingadvisor.com/blogs/dept/musings/how-perform-heat-loss-calculation-part-1

8. 
   qofmiwok | May 16, 2021 12:45pm | #8

   I'm going through the same thing. I have 3 rooms upstairs with max heat loads ranging from 3700-4800 BTU/hr. Max conditions don't happen very often, and based on average heat loads I'm imagining they'll usually be running below 2000. The smallest outdoor unit that can handle 3 heads is 24000, which modulates to 9000, or maybe I've seen a bit lower than that occasionally. So it seems like I'll be cycling.

   So we're currently looking at a ducted 18k unit, but it still only modulates down to 9k, and I'd rather not have ducts. I'm wondering if 3 individual 6k units with 3 separate condensers would be a dumb idea. https://hvacdirect.com/hvac/pdf/M_SUBMITTAL_MSZ-FH06NA_MUZ-FH06NAH-en.pdf This Mitsu is supposed to modulate down to 1600 (1700 for cooling.) And if I needed to go lower I could turn off one of the units and keep the doors open.

   I'm CZ 6B, downstairs max load 24k BTU.

1. 
   bongo30 | Apr 03, 2022 08:37pm | #9

   I’m in the same situation as you. Have you decided to install individual units or a multi-split. I’m leaning toward individual units (even though it will put us at 4-5 individual mini-splits).

1. 
   qofmiwok | Apr 03, 2022 09:13pm | #10

   After much research there were no individual splits that I wasn't worried were still going to be too powerful. But I also didn't want a ducted system, and multi-splits (1 compressor and multiple heads) don't seem to work that great according to my research. Lots of reports that they don't modulate they way they should. What I decided on was a VRV split system called Daikin VRV Life. We originally were looking for just 3 rooms upstairs, and were going to do radiant downstairs. But the radiant wouldn't work great either with super low loads, and we had a hot summer and for the first time thought AC might be a nice idea occasionally. So we ended up doing 2 systems with 3 heads each (5 wall mount, and one ceiling mount in the great room.) We had 5 heads on a single compressor, but we had long line lengths to the compressors and they thought 2 x 24's with 3 heads each would work better than 1 x 36's.
   It's not a cheap system, but I actually was quite happy with the $40k price compared to the original design which was over $100k.