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

Estimating cost to run minisplit system per hour

MSmithGBA | Posted in General Questions on

Hi folks,

I just got a new Mitsubishi mini split system to provide heat and A/C to my basement. There is a 30K BTU outdoor unit with Hyper heat (MXZ-3C30NAHZ2U1) and 2 indoor wall units at 18k and 12k (MSZ-EF series).

I’m trying to understand how much it would cost to me to turn on one of the indoor units and run it for an hour so that I know how conservative I need to be with it. There’s different fan speeds too, not sure if that impacts it or if A/C vs heat impacts it. My current electric costs in MA are 0.217 per kWh.

Thanks for any help you can provide!

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. bfw577 | | #1

    Run the fans on high all the time unless comfort is a factor. The manufacturers seem to put a dent on efficiency over keeping the fan speeds low for comfort with the wind chill effect. The efficiency is significantly increased with the fan on high. I have monitoring equipment and temperature data on both my units that easily shows this. There is another post on here that I did to support this.

    I suggest installing an electricity monitor. I am using an Efergy Engage ($140) to monitor both my mini splits. The data has been invaluable on seeing how these units modulate and how to efficiently run them.

    I am in CT with a similar electricity cost but I have solar panels with net metering that covers almost all my load. I have done extensive comparisons and the mini splits are about half the cost of heating oil with a much warmer house.

    Here is a snapshot of yesterday's data that includes many defrost cycles in the morning to give you an idea on what you would see.

    1. MSmithGBA | | #3

      Interesting. I don't think I'm reading that chart right though. Mind helping me out quick as I'm new to energy monitoring. I read that chart as:

      1) On average you're using about .5 kW per minute (roughtly), so about 30 kW per hour
      2) At .217 kWh per kWh, means that it's costing on average $6.51 per hour to run

      But that seems too high, so I think I have a misunderstanding on how it works.

      Thank you again! And will checkout the energy monitoring. I originally looked into Sense, but it doesn't work well with variable loads so will need to look at a circuit breaker based one. Also thanks on the fan speed tip. Assumed the opposite, so good to know!

      1. bfw577 | | #6

        Its per hour. At full capacity it will draw around 1-1.2 kwh hour. You can see in the graph it briefly ramps up to max capacity when it first starts but then settles down to match the heat load. Total consumption that day was 8 kwh so around $1.60 for the day at New England electricity prices. Though my solar panels produced significantly more than that. Heating oil is around 2.69 a gallon here so the savings are quite clear.

        1. MSmithGBA | | #7

          Thanks! I was just researching and realized my misunderstanding. Think I get it now. Really helpful!

          I should have asked, what are your Btu's for that unit? Wondering if it's comparable to mine.

          Yeah...I'm gonna need an energy monitor... That's really valuable data.


    2. kenmarcou | | #24

      BFW577, Thanks for this post! I have been thinking of doing the same thing for over a year now. I’m so glad to see it’s working well for you. I’m in a similar place - Zone 5 Central MA hills. How many mini splits do you have and what’s yours solar capacity? I have to do pretty intensive building envelope improvements to the house which is 180 years old first but since the oil furnace quit a few years ago it’s been using baseboard electric which is wildly expensive. I don’t think I’ll be able to get the construction done before winter and think I’ll have to get a couple mini splits first to mitigate the winter damage.
      Also, I was thinking of getting the Emporia Smart Home Energy Monitor which seemed a little better than the Efergy ones but Erergy has been good for you? I want to find out how much electricity my 80 gallon 20 year old hot water tank is using for starters and then monitor the mini splits like you are when I get them.

  2. _jt | | #2
    1. MSmithGBA | | #4

      Wow...what a steal of a price! Thank you for sharing that! Will probably pick one up. Researching now.

      1. Expert Member
        BILL WICHERS | | #5

        I just bought one :-)

        There is a note that it uses the modbus protocol on its data port. If that’s really the case, then a lot of industrial controls and monitors should be able to work with it. B and B electronics makes a converter that will let you read modbus devices over Ethernet too.


    2. CollieGuy | | #8

      One of the comments tells us that it only registers up to five amps and another suggests four (960 to 1,200-watts at 240-volts).

      1. _jt | | #10

        I haven't had any capacity issues. I looped into one of the 120v legs so my current is off by two.

        I have measured 1800 watts when the outside unit was iced up. (900 on the display)

        You can use the other leg to be more accurate but I prefer to stay 120v. To the original question I am using about 600 watts per hour 7 am-10 pm, and a little less at night (reduced set point). Western NJ power is 0.1 per kwh so operating costs are comparable/ slightly less than my previous hydronic system but my measured temp is much higher. (Hydronic couldn't get past 68 in the cold)

    3. johns3km | | #21

      Something like this is easy and you won’t have to go down to your panel to view it.

      Efergy Elite 4.0 Wireless Electricity Monitor

  3. dnicekid | | #9

    I also have a mit system. Outside is a 42 or 48 hyper heat. Inside I have three 9kbtu units , 12kbtu and a 5kbtu.
    I am trying to figure how much electricity they use as well. The installer told me to just leave it on all day at a set temp and the system will take care of itself. With so many individual units will this be cheaper than my boiler (95%) using gas at $1.30 a therm?
    Btw I live in western Massachusetts and electricity is about .249

    1. DavidSilva | | #23

      My system will be very similar. Did you get a sense as to how many kwh per month it uses? Thinking of putting in solar, but not sure of how much to pad the existing electric bill based on the multi-split. I'm in central Mass, so electricity costs are very similar.

  4. _jt | | #11

    This is more of a math problem. 30 kwh per therm. COP of 3 means 10 kwh per therm. At your energy rate its twice as expensive unless factors like delivery losses (ducting, pipe loss, zones) get in the way.

    Correct me if I am wrong!

    1. CollieGuy | | #12

      Just to sharpen your numbers slightly.... one therm contains 29.3 kWh(e) of heat energy, and at an AFUE of 95 per cent you would net 27.8 kWh of usable heat; at $1.30 per therm, your cost per kWh(e) is effectively 4.7-cents. At 24.9-cents per kWh for electricity and at a COP of 3, your cost per kWh(e) of heat for the Mitsu is 8.3-cents, so pretty much twice that of gas as you suggest.

      1. dnicekid | | #13

        Thank you for the answer!
        Are BTus btus? Meaning total spent btus using baseboard heat vs mini split. Would a room heat faster recirculating air vs convection therefore could the splits possibly not use the “same amount” of btus?

        1. _jt | | #16

          More or less. Keep in mind real COP is much higher when it is in the 40s and 50s.

          But... delivery mechanism makes a big difference. Long runs of hydronic under the house or ducting can eat a lot of btus compared to a mini split.

          1. dnicekid | | #17

            That’s what I was thinking
            My boiler is a condensing unit, but only 2 zones(up and downstairs) often times takes about 45-hr to go up 4-6 degrees at current settings. When I have turned on the split you feel the heat in minutes. Also my split often starts in defrost mode when it’s over 32 degrees is that normal?

          2. _jt | | #18

            You shouldn't see defrost when it's over 32. Check for ice build up and make sure you have proper condensate drainage - you may have a sensor covered with built up ice.

            Make sure it's off the ground, and you have a decent tilt - (say 10-15 degrees) - it makes a big difference.

        2. CollieGuy | | #19

          Our situation is a little different. For us, our ductless heat pumps are our most economical (and environmentally benign) heat source by far. We pay 15.603-cents per kWh for electricity plus an additional 2-cent premium for 100 per cent renewable energy; at a seasonal COP of 3.0, each kWh(e) of heat clocks in at 5.9-cents. Fuel oil sold by the discount players currently retails for 93.9-cents per liter locally, and at an AFUE of 85 per cent our oil-fired boiler serves-up a kWh of heat at 10.3-cents.

          However, in terms of comfort, our boiler wins hands down. With hot water baseboard heat, temperatures are uniform throughout, and exterior walls and windows are gently warmed so you don't feel as if the heat were being literally sucked out of your body. There are no drafts and it goes about its job quietly.

  5. davidsmartin | | #14

    I installed the sense monitor. It does an excellent job of displaying usage for the whole house and identified several other appliances but it has not been able to identify the minisplit.

    If you are curious about 120v appliances it is easy to add a smart plug that reports back to Sense but that does not work for 240v hard wired appliances like the minisplit.

    Nevertheless, looking at usage meters for the whole house it is very easy to see how much the minisplit draws.

  6. bfw577 | | #15

    Here is a pretty good explanation of a ASHP.

    An Air Source Heat Pump (ASHP) will typically produce around 4kW thermal energy for every 1kW of electrical energy consumed, giving an effective “efficiency” of 400%. It is thermodynamically impossible to have an efficiency of more than 100%, as this implies that more energy is being produced than is being put in. For this reason the performance is expressed as a Coefficient of Performance (COP) rather than an efficiency. The above example would be expressed as having a COP of 4. The reason that it appears that more energy is being produced than is consumed, is because the only “valuable” energy input is electricity used to drive the compressor and circulating pumps. The remainder of the energy simply transferred from a heat source that would otherwise not be used (such as the ambient air, ground or a river) so is not considered as an energy input.

  7. CollieGuy | | #20

    dnicekid: "Also my split often starts in defrost mode when it’s over 32 degrees is that normal?"

    These are the defrost adjustment factors for the Mitsubishi M-series heat pumps. The biggest hit appears to fall between -2°C and +4°C.

  8. dnicekid | | #22

    Just realized something that may change things for those who have monitoring.

    My mit doesn’t really run independently as advertised. If I have my 12k unit heating and turn my 9k unit on fan mode it to pushes heat although not as hot as the 12k but enough to warm the room. Apparently the branch box “bleeds” heat to the other heads. This is part of the design from what I have read online. This brings me back to those who monitor electricity. If the outside unit is modulating for the one room yet still heating the other rooms wouldn’t it always be less of a load if you just ran the other units on fan mode rather than them asking for more load from the outside unit?

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |