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

Request for Energy-Monitoring Data on Single-Zone Mitsubishi Minisplit

maheatpumpguy81 | Posted in General Questions on

Hi All,

I posted about a year go with this post: High December electric bill with Mitsubishi heat pumps. I wanted to follow up with all that I’ve learned, as well as put out a request for data others may have.

After many conversations with my installer and their conversations with Mitsubishi, it turns out that the multi-zone units change their compressor frequency as a result of outdoor temperature. I was clearly seeing a big jump in power consumption once outdoor temps hit about 36F. My two 3-zone outdoor units would more than double their electric consumption from about .80 kW/hr to ~ 2kW/hr. I’ve attached a picture of this happening as the temperature dropped below this threshold.

The installer is working with Mitsubishi to revamp my system to only run on single-zone heat pumps. It seems like this is going to happen as I have provided them a lot of my own data as well as building science literature in support of this change. The efficiency / modulating abilities of these multi-zone heat pumps is not at all impressive.

My question to this community is if anyone has similar data to mine, but for single-zone Mitsubishi heat pumps? I’d like to know whether this bump in compressor frequency (and in turn, electric consumption) is also typical for these single-zone set-ups. I have seen others’ monitoring data for non-Mitsubishi single zone heat pumps, but to make a good apples-to-apples comparison, I’d love to see this data specific to Mitsubishi units.

Thanks to all, especially the moderators, for helping with this journey.

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. vashonz | | #1

    I have 2 single zone Mitsubishi heat pumps (12k and 6k), 12k is just set to run all the time, and we basically leave the Master bedroom 6k off. I don't have any energy monitoring (yet) beyond the smart meter from the power company. I can share some (relatively useless) data.

    Large spikes are likely from cooking (9am, induction cooktop) and evening (plugging in the car to charge)

  2. walta100 | | #2

    I suspect you are being baffled with bull stuffings.

    The frequency of the power supplied to the compressor is directly corresponds with the speed the compressor turns and directly with the volume of refrigerant moved. I see no reason for the speed change to make a large change in the number of BTUs per watt. I would be very surprised if higher frequencies say 100 Hz did not have lower losses than 15 Hz does electrically.

    My wild guess is you have a hyper heat model and you are seeing when it changes from a one stage compressor to reinjection 2 stage compressor.


  3. jameshowison | | #3

    Thanks for this, very interesting. Could you explain your monitoring system? What needs to be in place to get this sort of data?

    When you say "changed to single-zone heat pumps" could you give more info? Do you mean changing the inside units so that they are a 1-to-1 match for the outside? Or do you mean replacing the outdoor compressors with more to achieve the 1-to-1 match?

    I'm currently trying to decide whether to use an existing compressor with a 1-to-1 match or with two indoor units (to avoid late afternoon sun making a bedroom on the shady side used as a home office cold)

  4. maheatpumpguy81 | | #4

    Thank you for the responses, guys.

    Chris, it is tricky to interpret your data. Looking at the second graph, I see a 50% drop in electric consumption at 4am, even as the outdoor temp drifts lower. I wonder if this is a consequence of your EV finishing up its nightly charge.

    Walta, your understanding and expertise definitely exceeds my own with this equipment and I feel unqualified to give you a useful response. That said, my two 3-zone outdoor units are indeed Hyper Heat units. I would hope that this notion of modulation by Mitsubishi isn't simply reduced to what you said of changing from a one-stage compressor to a reinjection 2-stage compressor.

    James, I am looking to match three of my six indoor wall units with same-sized single zone outdoor units (two 9k's on the first floor, one 6k in the basement). We are removing the multi-zone outdoor units completely. For the second floor, we'll remove all indoor wall units and instead use the existing attic duct work from the old AC system that was never removed and 2nd fl ceiling registers... we'll use a single zone outdoor unit connected to an air handler in the attic to heat and cool the 2nd floor rooms. I'll supplement with resistance heaters to get me through cold snaps. Does that effectively answer your question?

    1. jameshowison | | #7

      Thanks, very helpful!

    2. vashonz | | #8

      MA guy,
      Drop off at 4am, then spike at 8pm was because car was charging
      Right screenshot was using L1 120v charger (15a 120v max), left photo was with a spike at 3pm because I installed our L2 charger (30a, 240v) and plugged in the car to test.

      Iotawatt is on the list of things for the house, not sure if it's a priority for us yet.
      Separating house vs garage vs EVSE vs well usage is getting more difficult as we add components. We just ordered the steam shower/generator. I'd like to see how much each different component costs (HPWH, Steam generator, Split units, EV charger, Induction stove) and how our usage changes monthly cost (charging at work instead of at home (it's free at work!), maintaining the split units set at higher vs lower temp, Rheem Eco mode vs high demand, ect)

  5. maheatpumpguy81 | | #5

    James, I forgot to answer your monitoring question.

    I bought the Iotawatt kit. It involves putting ferrite CT rings around the hot lines of the circuits you want to monitor and connecting to the Iotawatt controller. That controller has WiFi and you monitor through a computer connected to your home network.

  6. bfw577 | | #6

    For reference here is what the modulation looks like on my 12k single zone system. Unit is a 12k Midea Premier floor console.

    I am using a $120 Efergy Engage hub to monitor my solar, whole house and both Mini splits.

  7. GBA Editor
    Kiley Jacques | | #9

    This podcast with Chris Briley, Phil Kaplan, and Dana Fischer (of Mitsubishi) might interest you. They answer the question: Why do multi-zone air-source heat pumps usually perform worse than single-zone ductless minisplits?

    1. PBP1 | | #16

      Great podcast, again, warning as to sizing - don't oversize, multi-zone min is about 35% of max. Thus, a "big" multi-zone ASHP rated at 40k will have a min of 14k for any heat demand, however, small that may be.

  8. Ben_McElroy | | #10

    I’m not alone! I know this is an older thread but I wanted to share my similar experience with a Mitsubishi 3 zone mini split (MXZ-3C24NAHZ2). At 36F and lower outside temperatures it starts short cycling like crazy (4-5 minutes on at 1500-1600W power draw then off for approx 3 minutes). All the indoor units seem to run in unison too during this condition. Above 36F outside it modulates ok at 600-1000W on 25 min on/5 min off cycles. Not ideal but I figured it was the nature of a multi split.

    My 18k 1:1 unit on the other end of the house (MSZ-FS18NAH) also had short cycle issues but I think it was unrelated to outside temps. More likely due to a single head in a large drafty space (new windows, air sealing, and insulation upgrades in progress). A wireless remote thermostat seemed to cure the short cycling; However, now it refuses to modulate. It runs more like a 2 stage conventional system either operating at a fairly steady 1050W or 1550W, a shutting down to switch between the two levels. It ramps up to those levels, “modulates” for a few minutes then runs steady until thermostat is satisfied then abruptly shuts off.

    I’m curious how/if you got Mitsubishi to replace your system? Mine was a combo self install with local HVAC company (with Mitsubishi experience) running the line sets and wiring.

  9. DinaR | | #11

    MAHeatPumpguy81 and others - I have this exact same issue. We installed 2 muti-zone Mitsubishi Hyperheat Heat Pumps 3 years ago. We live in Franklin MA, and have a 1940 sft colonial home built in 1984, with a basement of ~500sft semi-finished. Insulation is not very good, I did fill my attic with cellulose spray from MassSave, and part of the basement has Owens-Corning with some older bat-insulation. I have solar panels that generate about 600kWh during the Dec/Jan/Feb timeframe. The solar panels + heat-pump offsets our power usage completely in the summer months, and then I get screwed in Dec, Jan. In Jan 2021 the usage was 1300kWh , and this year Jan 2022 I see 1621kWh....after using about 500 to 600 kWh from the panels. I have an MXZ-3C30NAHZ2 that feeds 3 indoor heads (15k, 12k and 9k BTU) and a MXZ-4C36NAHZ that feeds 3 indoors heads (18k, 15k, 9k BTU). It looks like its the same exact problem where i constantly hear my outdoor units struggling to keep up when temps drop below 35F. Did you guys find a solution to this? I notice a lot of people are having issues with these units, is this something we can collectively represent to Mitsubishi? HyperHeat was advertised as the best there is, I cannot believe they are performing so badly. Thank you!

    1. bfw577 | | #12

      Does the 1621kWh of usage include the electricity for the rest of the house? That doesn't seem that bad if that's your total usage for your entire house and both Mitsubishi. That would work out to what a $350 or so bill in Massachusetts? Seems reasonable for heat and electricity for a home of that age. Its also been quite cold in New England recently.

    2. Ben_McElroy | | #13

      I just installed 3 remote wireless thermostats to the tune of $850 for my 9+9+9 wall mount heads running off a MXZ-3C24NAH. Short cycling issue was significantly reduced but the sub 36F temperature issue is still very present. When outside temps are sub 36F, the unit will draw up to 2000 W when only one 9k head is calling for heat. In comparison, the 18k single split never exceeds 1600 W in the same conditions at full load. I’m having issues getting through to Mitsubishi Tier 2 service to get some explanation since I self installed most of my system. I’ll be sure to post here if I get an answer. The big question is can I change the system output/operating parameters via dip switches or programming? I’m definitely oversized and my 9+9+9 system uses the same 24K compressor as the 6+6+6 set up. I only went with the 9+9+9 because the local Mitsubishi Diamond Elite installer said it could modulate as low as the 6+6+6 (Haha!). They wanted $23k for the 2 systems so I did it myself for $11k (plus another $1k for remote thermostats to stop the short cycling).

      I feel there is solid ground for some sort of case against Mitsubishi. Their literature for the M-series boasts about all the usual mini split benefits, improved comfort, steady temperature, inverter drives with a range of output to perfectly match the load, etc. My 9k heads run at full blast when the temp drops 1 degree below set point then shut off when it hits 1 degree above, no modulation (I know this is old news to some). Seems like false advertising to me.

    3. PBP1 | | #15

      I have MXZ-3C30NAHZ2 that feeds 3 indoor heads (15k, 12k and 9k BTU) for 2100 sq ft in Montana w/max monthly usage in past two years at about 1500 kWh (w/subzero temps, ave around 20F). The MXZ 36 is rated at around 40k, so seems like you have nearly 70k rated heating total. Per podcast (link from Kiley, thanks!), min for multi-zone is around 35% of max, so your two units min is: 9.8k + 14k = 23.8k. Hence, if both ASHPs call for heat, you're consuming substantial energy. If I understand, you're Jan 2022 is 1621 kWh + ~500 kWh (solar), for a total of 2121 kWh for two ASHPs w/6 zones. Ave December temperature in Franklin MA was ~ 39F (Warwick RI).

  10. PBP1 | | #14

    If my 3 zone Mitsubishi ASHP (MXZ 30 w/15k, 12k and 9k ducted) had this issue, wouldn't there be an expected shift in the heating slope, which is quite linear, around ave temp less than 36F? Like another poster, I have an Efergy TPM, but don't have it hooked up to my ASHP yet, only the non-ASHP panel. The plot is total home kWh/day versus ave monthly temperature, subtract 13 kWh/day to get ASHP. With no/low demand, I believe consumption for the ASHP and 3 heads is around 160 W.

    Regarding the multi-zone podcast (thanks for the link), my ASHP is rated at 28K and heat load was calculated to be 28K and indoor temp can be set as high as 74F (or more) in winter. The HVAC guys/HERS rater were in agreement with sizing of ASHP and heads (no oversizing). Maybe my benefit is that the 15k head serves a 20'x30' greatroom 9' to 13.5' ceilings that is completely shaded and north side such that the big head is first to call for heat and calls for a substantial amount?

    1. paul_wiedefeld | | #17

      The linear energy consumption vs. temperature of these heat pumps is always surprising to me. Between modulation and Carnot, I’d except the hit to be greater but my daily data has shown the same thing.

  11. bcc22 | | #18

    Having this same issue, 2 multi splits mxz-3c30nahz2 and 3c24 same series. Both with two indoor heads. Heating costs are astronomical, using over 3000kwh per month in the winter. Southern VT here. Help!!

    1. bcc22 | | #19

      Small house, 1500sqft. Insulation is pretty good and heat pumps are sole heat source. Just posted about this in a separate thread but found this discussion afterwards. Any ideas!? Such an expensive system to install and bleeding money on electric bills :(

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |