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
Q&A Spotlight

Air-Source or Ground-Source Heat Pump?

A homeowner weighs the choices for a heating and cooling system in Climate Zone 6

Ductless minisplit

A ductless minisplit system includes an outdoor condenser and one or more indoor fan-coil units, but no dedicated duct system. The best of these units are more efficient than most conventional air-source heat pumps.
Image Credit: Scott Gibson

Dana is building a tight, well-insulated house in climate zone 6 and now faces a choice between a ground-source heat pump and an air-source heat pump for heating and cooling.

“After the 30% tax incentive, there is not much increase in cost for the geo system,” Dana writes in a Q&A post at GreenBuildingAdvisor. “I am being told different stories in regard to system performance and longevity of equipment (depending on what side of the fence you’re on).”

It’s no short-range decision, either. Dana hopes to stay in the house “until my kids have to make the choice of where to send me.”

Ground-source heat pumps extract heat from the ground, usually via tubing that’s either buried in the ground or encased in vertical wells. Air-source heat pumps, which rely on an air-to-air exchange, are less expensive to install. But conventional systems have had difficulty operating in cold weather, as someone in climate zone 6 can expect to see.

There are arguments for both system, but what’s the best choice for Dana? That’s the topic of this month’s Q&A Spotlight.

Other issues to consider

“There’s no easy answer to your question,” writes GBA senior editor Martin Holladay.

Manufacturers of ground-source heat pumps list the unit’s coefficient of performance (its COP), which is the ratio of energy produced to energy consumed, but as Holladay points out the numbers can be somewhat misleading.

“One caveat: manufacturers of ground-source heat pumps (GSHPs) don’t include the energy used by pumps when calculating efficiency or COP ratings,” Holladay says. “As a result, actual COPs of these systems vary widely; many experts have been called to troubleshoot GSHP systems with grossly…

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. Expert Member
    Dana Dorsett | | #1

    I have this disussion often.
    It can sometimes come down to competing subsidies, but sans-subsidy, when you have the building envelope performance up to where the design heat load is well under 3 tons, a ductless + rooftop photovoltaics solution can be a better financial investment than a ground-source heat pump solution at the same up-front cash investment. Summertime output on the PV more than makes up the difference in annual energy use from the (still substantial) difference in operating efficiency between a best-practices GSHP system and a better class ductless.

    This pencils out differently in every situation, since it depends on the real electricity costs, as well as the system costs, but with the falling installed costs of PV and ever increasing efficiency of ductless there's a real trend in favor of that solution. In places like MA where PV is heavily subsidized and granite ledge drilling makes geo 8-9, even 10 grand/ton, even with the 30% tax credit it's still hard to make the case for geo for heat loads under 3 tons. It takes a sharp pencil to figure exactly where the crossover is and the subsidies are moving targets, but taking the PV out of the equation, when in doubt, a higher performance building envelope + ductless air source heat pump will offer more comfort than a lower-performance house with higher-efficiency ground source heat pump system.

    Since the daily PV output leads the regional grid-load air condition peaks by only a couple of hours (and is still delivering a large fraction during the grid-load peak), the ductless + PV solution is nicer overall to the grid operator to boot.

    There will be hours in the winter where the outdoor dew point is too close to the outdoor temp that air source heat pumps spend effectively all their time in defrost mode with severely diminished output capacity (yes, even for the Mitsubishi Hyper Heating units), but those events are limited in duration (typically a very few hours at a time) and a high-R building would have a sufficiently long time constant to be ride out those shortfall periods without discomfort. The drier the winter climate, the rarer those low-capacity events will be.

  2. user-984364 | | #2

    Mini-splits & distribution
    I think with mini-splits you have to be sure to consider distribution of conditioned air as well (sure, you can do the little short-duct in-ceiling deals, but then you start losing some of the fantastic efficiency). We currently have a single mini-split head in our house, primarily for cooling, and while I knew it wasn't going to evenly cool the whole house, it requires a bit more effort than I thought to distribute the cool air very well at all. I probably should have known better. ;)

    WIth a GSHP presumably you have ducts and all that, and at whatever final efficiency, at least you're getting the heat to all the right places.

    I'm sure mini-splits can do a good job in some home layouts, esp. with multiple heads. But the point-source nature of the conditioning maybe something else to consider.

  3. GBA Editor
    Allison A. Bailes III, PhD | | #3

    Good overview
    Nice job summing up the discussion, Scott. Another thing to consider, which goes along with what Yost and Holladay were saying, is that GSHPs often don't get close to their rated efficiencies because of the pumps and loops. Often the pumps use much more energy than anticipated, and I've seen two reasons cited more often than others: (i) The pipes are often undersized for the flow, resulting in higher resistance, and (ii) the designers often neglect the issue of head, the weight of the column of water in vertical wells.

    Yost's point about the confusion over the energy source drives me crazy as well. I read a Thomas Friedman column once, and he was raving over his renewable energy 'geothermal' heat pump. I hate the term 'geothermal' applied to these systems because of that confusion. There's no lava or geysers involved here, no steaming pools of hot water at the surface or volcanoes blowing their tops. The ground is acting only as heat source or sink, but, as Peter points out, the 'fuel' that drives these systems is electricity, not geothermal heat.

    Finally, one little language quibble: In the fourth paragraph you wrote: "Ground-source heat pumps extract latent heat from the ground..." 'Latent' has a very specific meaning when attached to the word 'heat,' and it's not the meaning you were implying here. Actually, GSHPs extract sensible heat from the ground, even if it is just existing beneath the surface as a potential. Latent heat has to do with phase changes, and when GSHP loops deposit or remove heat from the ground (or even groundwater), nothing boils, condenses, melts, or freezes.

  4. Expert Member
    Dana Dorsett | | #4

    Re: Mini-splits & distribution (response to Eric)
    Houses of in climate zones 5 or higher with design-day heat loads within the output of a mini-split are by definition high-performance houses with way-better than code windows and high R values all around. This dramatically reduces the need for room-by-room heating/cooling distribution, but careful consideration needs to be taken on where to place the head(s). Most multi-story high-performance houses end up with 2-4 head multi-splits to handle the different loading of different floors, etc. but it's far from 1-head per room.

    Marc Rosenbaum (Peter Yost's reference) heats his 2-story house with a 1-head 1-ton mini-split that was done as a retrofit (yarding out the oversized oil boiler that had been heating the place.) With some management of convection between floors it's been working reasonably, but he has stated that most people would probably opt for more than one head or with localized supplemental heaters to better control temps in the remote rooms when occupied. See:

    1. DoggoneGreen | | #45

      Exactly this. If you put the heads in the right place, or use one or two ducted interior heads, you can keep the whole home comfortable. *IF* your house is well sealed and insulated.

  5. user-1151665 | | #5

    Condensor in the attic?
    Wondering if you could place a condenser of a mini-split in the attic instead of outside? This makes sense in your colder climates. Obviously you would need to have proper ventilation and a drain pan sitting under it. This would keep it operating in a much more controlled environment. The main obstacle to over come would be keeping the temperature in the attic low enough in the summer months. A metal roof with good ventilated should work well. Just an idea. What do the experts think?

    1. DoggoneGreen | | #46

      This would ruin your efficiency in the summer months. Outside works fine.

  6. Expert Member
    Dana Dorsett | | #6

    Mount them where they can breathe!
    Putting the outdoor unit in the attic above the insulation puts about R1 of roof deck & roofing (more when there's a snow-loads) and the mass of air that you're trying to extract the heat from. It would very quickly super-chill the attic below the outdoor temp when under any real heat load, and would melt the shingles off the attic under a cooling load, cutting dramatically into both efficiency and capacity in all seasons.

    It's good to mount them where they won't be buried in a roof avalanche or drifts. Bracket mounting on the exterior walls above the 50 year snow pack level, and under eave/rake overhang or small shed-roof to protect from roof avalanches/icicles/ice-storm encrustation is good. But installing them in attics or crawlspaces, not so much. Air source heat pumps really need access to their source, which is the great mass of outdoor air.

  7. wjrobinson | | #7

    The posts are very
    The posts are very informative. Much more useful than led light bulb blogs or electric lawn mowing. One point about picture saying outside condenser. Heat pump split airs have dual use fan coils inside only the compressor being outside.

    Dana, you are a welcome addition to GBA.

  8. johoro | | #8

    ASHP and HRV
    Will a home with an HRV (effectively) distribute treated air from a mini-split, thus minimizing the drawback of point-source heating and cooling? I realize open plan versus rabbit warren may impact this.

  9. Expert Member
    Dana Dorsett | | #9

    While it's possible to get
    While it's possible to get SOME redistribution of ductless heat with HRVs, that's best done if designed in from the get-go, where the low-loss rooms remote from the open area with the ductless head are only on the exhaust duct tree of the HRV, and getting their ventilation air from the directly-conditioned space via jump-duets, etc. This is not a huge heating effect, but can be enough if the direct heat loss/gain of that room is pretty low, which usually means limited amounts of window area and U-factors under 0.25, and a whole-wall R of at least 1.5x code-min.

    Bumping the temp in the directly-conditioned spaces a few degrees when it's cold out helps the indirectly heated rooms from chilling too low too, as does leaving the doors open- it's often a matter of empirical investigation, but most home heating systems aren't perfectly balanced in the first place, and people figure out by trial & error how to make it comfortable. Point-source heating with a ductless is usually more comfortable than doing it with wood-stoves, because when you need to overheat the directly conditioned space to keep the rest of the house cozy you don't have the direct radiant-roasting of people in the room with the point-source. But people put up with wood stoves, and find the direct-heating of heat radiating off the stove a plus when the room temp is low.

  10. user-1012653 | | #10

    I will chime in again...
    I will chime in again being as I was quoted infavor of the geo. While I still am, however I am actually back to the single head hyper heats now that they released a 21k head that will nicely condition my 1600 sqft main level heat load including temp drop off with the mini split. Costing was a bit one sided originally as I was factoring rebates into the geo numbers, but later found out (after the post) I can get similar rebates from the local electrical suppliers making a net investment in 2 mini splits installed (21k on main level 12k in basement) right around 2k. Head placement on the main level can cover the open living, dining, kitchen, and then direcly down a short hall to the 2nd bedroom, bathroom and laundry room (all are located in the "throw" range of the head). The master bedroom door sits on the same wall a couple feet away, so hopefully that picks up plenty. The master has a large solar gaining window that I also hope to help with the space. I will still throw a couple cheap 2' base board heaters in the bedrooms and back laundry room, hopefully never need to use them.

  11. user-1012653 | | #11

    as soon as my current home
    as soon as my current home sells, which should be any day now. the design stage will freeze once I break ground :)

  12. wjrobinson | | #12

    Jesse, when r u freezing the
    Jesse, when r u freezing the design and starting to build this endless design stage home?

    Are you going to blog your design and build process?

  13. user-1127636 | | #13

    Air-source via HRV vs Air-source vs Ground-source?
    How does the comparison change if an HRV is introduced? A product I'm looking at apparently is placed in-line with the outdoor part of a split-system heat-pump such that the exhaust HRV air is fed to the inlet instead of regular outdoor air. It does add a few $k to the price but would that bring the performance closer to the ground-source system?

    Could you still use an HRV as well as a ground-source system or are you not gaining much for your $?

  14. Brian Knight | | #14

    Design, Distribution and Comfort
    Eric brings up the concern that is often overlooked with mini-splits. If I were building a home for myself, I would probably go the mini-split route as I would be building a very small and open design that would lend itself well to a point source system. I think its risky however, to do it for clients even with high performance envelopes.

    Not everyone is going to be satisfied with uneven distribution or even worse, the need to keep bedroom doors open. This strategy is fine for energy conscious control freaks like me but not all families are going to accept the small sacrifices needed to make it work.

    My HVAC guy has several examples of homes he was brought in to consult on that have gone the mini-split path with poor results. In general, I dont think these homes meet the high performance envelope standard that me and this site champions but it still gives me a wait and see approach.

    When comparing costs of traditional ASHP to mini-splits, I have always considered 3 heads to be the point that it makes more sense to stick with a traditional system in our Zone 4 climate. The # of bedrooms and levels is what mainly drives the head count and even for high performance homes, 3 bedrooms usually means 3 mini-split heads.

    I happen to live in a location with very competitive Geo pricing, competent installers and great state tax credits which makes for a tough decision even for high performance envelopes with long paybacks. In general, I prefer traditional ASHP but am looking forward to serving a 3BR2BA home with mini-splits without the concern of a dreaded dis-comfort callback.

    I would love to hear more examples of 3BR2BA or larger homes that were built for "normal" american families that have 3 or less mini-splits and no problems of discomfort and distribution.

    As to Brendan's question, a little confused by the terms and layout but an HRV/ERV is independent of the heat pump and can be placed in any configuration. Usually the HRV/ERV supply air is fed into the air handler's return plenum but on homes with more than 2 levels with the HRV on the top or bottom level its probably best to keep the ducts separate to ensure adequate distribution.

  15. Expert Member
    Dana Dorsett | | #15

    They're out there (response to Brian Knight )
    In US climate zone 4 it's pretty easy for whole-house heating loads to end up within the output of a mini-split, but with higher U-values the balance issues become more problematic. That said, I know of a circa ~1990 code min 2-story 3br house on Whidbey Island WA currently heated with a single-head mini-split, with the head mounted mid-height in the "great room" open to both levels. The occupants manage the balance issues for the doored off sections by leaving the doors open during the coldest weather. Previously heated with fossil-fired propane hot air, they claim they'd never go back, and find heating with the mini-split far more comfortable than the ducted air solution. The openness of the floor plan and the full 2-story open area access means there's never more than door between any given room and the fully-conditioned space- it works (for them.)

    Nearby in Port Orchard WA I know of a '60s vintage 1br heated with a single-head mini-split, where they occasionally use supplemental resistance heat in the bedroom/bath during the wintertime temperature extremes to manage the comfort issues. It had previously been heated with a wood stove + ducted hot air electric furnace. Again, reported comfort is much improved with the mini-split over the ducted-air solution and wood-stove has been pulled out. But unlike the Whidbey Island house, some resistance heating is necessary during the coldest weather.

    Both are retrofits on existing houses, both well below current code min for tightness, R/U values, etc. With better windows and a modest uptick in R values the balance issues would take even less to manage. On new construction (or major rehabs) it's possible to design for ductless when that's considered a likely heating option. In any new construction the house and the mechanicals can be designed iteratively to work well as a system. It's a bit tougher when ductless is only an afterthought to a presumptively-ducted house.

    Using the heating/cooling ducting for the HRV isn't really best practice, but is often done for cost reasons. Ducting the ventilation air separately works better for the much lower air volumes moved.

    When/where subsidized and inexpensive there it's much to argue against going with geothermal, but in most places it's a huge up-charge even after subsidy, and doing the math on how much envelope improvement (or even PV) the cost delta might buy is strongly encouraged. Even at 5-6 grand a ton, if you can lower the load by a ton with envelope improvements, even at 8 grand/ton it's a better investment in comfort & energy use, since the "missing" tonnage of load has no operating costs, and provides more comfort via better balanced radiation-temperature from all surfaces.

  16. Brian Knight | | #16

    Discomfort callbacks seem to be cooling problem not heating
    Great stuff Dana thanks! The mini-split problems in our area were in the summer from homes built in the past three years. I think the complaints from at least two homes were on upper levels with more isolated bedrooms. I think at least one of them had separate ERV ducts which apparently were not adequately distributing the AC.

    I can see how lower heat distribution to bedrooms in Winter is less of a problem because many prefer cooler temps when sleeping. This preference becomes a problem come cooling season.

  17. user-1031655 | | #17

    Are there any ductless minisplits that don't look hideous? I know beauty is in the eye of the beholder, but these things are ugly. Growing up in a tiny house in Wisconsin, we had a wall mounted air conditioner in the living room, and to say the thing was an eyesore would be an understatement. As soon as I was out of the house, mom had it pulled and put in a small central air system. Even the one in the picture in this article is a hideous to have to look at every night, and I suspect, loud enough to interfere with good sleep, especially if one is not feeling so well to begin with when they lay down for the night.

    Also, I read that some comments about sleeping with bedroom doors open. As a child, I was always taught that sleeping with the bedroom door closed is done to slow smoke and give the occupants a few more seconds/minutes to get out the house in case of a fire. Is this not a common practice outside of WI? I'll accept that my aesthetic complaint may be just a personal pet peeve, but shouldn't design of hvac be done with basic life safety in mind?

    1. DoggoneGreen | | #47

      Honestly, most people don't even notice the heads in our house.

  18. celloman | | #18

    positioning of the outlet for heating
    In Australia the reverse cycle mini systems are the most common form of air conditioning. They are most commonly placed high on the internal wall as in the image in this article; this helps the system perform excellently in Summer to cool the interior. However, if the main demand is for heating, then outlets that are placed closer to floor level, will perform better and give a more instant warming. A friend had one installed in a dining room of a stone house which had little solar gain and it was extremely effective in heating that area.
    As for aethetics Paul W, I have to agree that they have a way to go before being described as attractive; the manufacturer LG did make some that had various 'modernist' fascias that attempted to add some style. Having the outlets stand on the floor does reduce their impact somewhat, I think. Some of the latest mini splits are exceptionally quiet now though, especially if it is a smaller sized unit for a bedroom.

  19. av8eer | | #19

    Coastal Ground Source Heat Pumps
    I live in the Tidewater region of Virginia and found a Geothermal installer who owned their own drill. They were able to perform the entire installation for $10K...that gave me a 2.5 Ton closed loop system with 3 ground loops at 300' depth each. While I did end up doing lots of cosmetic work on the back side, the proof was in the electric bill and I found that I save $60/month in peak season (I would have come close to this with a 16 SEER AC and 85% AFUE Furnace too). With the tax credit, they system was actually cheaper than a comparable ASHP. Our soil is about 8' of clay with water baring sand below, so it seems to me that the quotes I got for $15-20K for the same job were inflated because they knew that the tax credit existed. My contractors only worked 3 or 4 days a week and seemed to get by just fine economically with many observed inefficiencies in their work patterns. Drilling 3 holes was one day's work (not terribly long days), then one more day was needed to run the lines to the attic and install the sealed GSHP. Looking at it, it is much simpler that the high efficiency ASHPs. The evaporator is one large single pass coil and the heat exchanger is fairly small--about the size of a soccer ball. My end thought was that the sealed unit probably cost much less to construct than a 15 SEER style ASHP (substitute in appropriate COP since we are talking heat pumps). At any rate, in the end, I feel the market is paying a price for that tax credit just like "Cash for clunkers" forced people into paying over sticker price for the desirable models of new cars a few years back. Because they save you money over the long haul, there is a premium for GSHPs. In time, I believe the market will correct this, as it really looks like a much simpler unit for a company to make with tangible power savings for the consumer.

  20. GNiessen | | #20

    GSHP vs Air Sourced - sizing for extremes
    We just built and choose a GSHP. And while I do feel we still paid a premium for the system compared to a similar air source unit, I am happy with it. We did not go with as large a unit as was recommended by our contractor. Yet it seems to handle the extremes (record highs and lows) much better then our neighbors air units. We did go with a very well insulated house. But we have no problems cooling it in the mid afternoon when it is over a 100 with no wind. Due to the nature of the ground source, our unit does not work nearly as hard as an air source would have too. And it does matter that you use more electricity during the peak periods. Even if it does not show it on your electricity bill, the utility is paying more for it and do pass that on to the customers. I enjoy the peace of mind that I am going to be comfortable during the extreme conditions we have had lately.

  21. DBConstruction | | #21

    What about ASHP-to-water/radiant?
    I'm currently renovating a house in NH, zone 6, with high R, as well as tight with HRV. I want economy plus comfort, including convenience of being able to close doors, etc. At this point, ASHP to water/radiant looks like the best option for heat, with a summertime switch to fan/coil for occasional cooling. Only trouble? finding the equipment! Mitsubishi Ecodan/Zubadan is coming along as well as a couple others from Europe (Daikin, Robur-which is gas fired, Danfoss) Not a lot of experience yet in States, but I'm very interested. Does anyone have any experience, opinions?

  22. JCP_KES | | #22

    Any advice on Carrier Infinity 20 HP?
    Great conversation! Does anyone have experience or advice regarding the Carrier Infinity 20 Heat Pump w Greenspeed? It has a variable speed scroll compressor and 13 HSPF. How does it compare with mini-splits? A ducted system would solve those pesky distribution issues with bedrooms. Do you really loose that much with a ducted system if it's designed and installed properly?

  23. user-917153 | | #23

    Ground Source Heat Pump economics
    We tell customers to compare the fuel costs of alternatives, especially if natural gas is available, and if building a new home, to build one w/ low heat loss so that the choice of heating/cooling system is less important.

    We're still looking for cold-climate field studies that suggest GSHPs are cost-effective. The studies cited at the following link (studies by Manitoba Hydro and Steven Winter Associates) are underwhelming.


  24. kevin_in_denver | | #24

    PTACs and PTHPs Keep First Cost Way Down
    Ducts aren't required in a superinsulated house. Because the movement of heat through the exterior walls is so much slower than heat movement due to natural air convection within the walls, you don't get large differences in room temperatures, whether the interior doors are open or closed.

    Minisplits are pretty darn efficient.

    However, in a low energy house with some passive solar gain, you will find that the backup heat almost never comes on.

    Therefore, why spend a lot on a system that rarely gets used?

    That's why I'm always pushing PTHPs which are installed in minutes by the electrician, making the HVAC specialist another subcontractor you can eliminate.

    How does $509 for whole house Heating and Air Conditioning grab you? It includes a little bit of optional ventilation as well.

  25. David_Gregory_CZ3_CA | | #25

    A few more things on GSHP's...
    ...from a dilettante, in case it's useful:

    Air temperatures are pretty well known, making ASHP's less risky on this count (in addition to GSHP's component choice/install variability risks, already mentioned).

    Ground temperatures are 'stable', UNTIL YOU INSTALL A GSHP. Then it gets complicated...

    Short version (as I understand it) is that whether the soil is simply a heat source/sink or more like a 'battery' depends on the underlying geology and hydrology.

    For instance, @Jack Parker, in the Tidewater of Virginia, has GSHP pipes primarily in water-bearing sand. Depending on the groundwater velocity, any heat 'dumped' there in the summer is likely gone by winter, and any 'cool' 'dumped' there (sic) is gone by summer...relatively predictable.

    But, as @Dana Dorsett notes, MA and PA have lots of heat added to the stone during summer AC season is, in a way 'available' for winter heating. Sounds good, until you face the painful decision of whether to open windows in summer for 'natural ventilation', or run the AC more, to 'bank' heat for winter! >.<

    I'm only half joking. So called 'unbalanced' systems (poor match between heating and cooling loads) require more drilling, longer pipe runs, more materials / friction / pump energy, etc., or may simply not work. A study on GSHP's for bridge deck de-icing (RAMAMOORTHY, 2001) showed that 'air-conditioning' the road surface on the hottest days in summer could help 'balance' the system, saving more money on install costs than the extra energy (don't recall what discount rate was used to get NPV). Which suggests that in heating-dominated climates with soils with poor conductivity, you might get a net benefit from occasionally air-conditioning a hot attic (esp. on a sunny day with surplus PV?); or in cooling-dominated (and certain soils) from blowing off excess by heating the great outdoors on the coldest day of the year...or simply running the heat with the windows open!

    It would take a sharper mind (and pencil) that I have available to know for sure; I offer the conundrum here in hopes it finds such. But the essential point is simply reinforcing what has been stated already: that GSHP's are much harder to get right, which is likely why many 'underperform'. If you live in one of those places that's miserable in both summer _and_ winter, they'll probably do better that ASHP's. Otherwise...make sure there's a good engineer (who will adequately include local ground conditions) behind the evaluation.

  26. GBA Editor
    Martin Holladay | | #26

    Response to David Gregory
    I agree with your conclusion that "GSHPs are much harder to get right, which is likely why many 'underperform'." Balanced against all of the uncertainties that you list is one certainty: a GSHP system costs much more than several ductless minisplit units.

  27. user-1005777 | | #27

    Ground source augmented by solar
    I live in a trailer park, so the ASHP was my only choice. Having said that, I had played with the idea of a GSHP augmented by a small solar panel on the reurn to the sink. Since my heat load is 32 million BTU for the winter I would only need 80 square feet of solar panel to generate the energy needed. Since the temperature needed is very low, a swimming pool heater would serve very well. Any thoughts?

  28. Maribago | | #28

    by user-1031655:
    "Are there

    by user-1031655:
    "Are there any ductless minisplits that don't look hideous?"

    Thank you, user-1031655, for bring this up. key point: If my wife can't live with how it looks then, economics aside, it's a no-go. :-)

    Anyone worked with baffles/valences or other ways to improve the industrial chinese-restaurant look of these units?

    Gotta' be a aesthetic approach out there that still lets us benefit from the performance/economic of minisplits.

    1. DoggoneGreen | | #48

      If they really bother you, use an in-ceiling installation of a ducted head. Honestly though, they're just not that bad for most folks. Everyone's opinion varies, but in our home they tend to just blend into the background.

  29. [email protected] | | #29

    Air source heat pumps
    I recently went a talk on heating. Bruce Gray of Solaris Manufacturing in Qualicum Beach Vancouver Island British Columbia presented his system that is already in use in Spain. He claims a COP of 750 as opposed to a minisplit of 250. In bright sunlight the COP can reach1200. The system works on cloudy days and even by moonlight.Installation only takes one day but is suitable only for water radiant heat. This sounds too good to be true. Here a minisplit costs $4000 while this system is about $5000 for just the domestic hot water. Is this for real? Thanks

  30. GBA Editor
    Martin Holladay | | #30

    Response to Michael Legge
    Q. "He claims a COP of 750 ... In bright sunlight the COP can reach 1200. ... Is this for real?"

    A. No. Someone has misplaced their decimal point.

  31. [email protected] | | #31

    Air source heat pumps
    Martin-thank you for your prompt reply. Whether the decimal point is in the wrong place,or not, the comparison of COP is still significant. The "patented" source panel does not require the visible radiation spectrum and they are going to run a trial using the heat in attics to scavenge. The panel can be put on the North side a house and still functions. Am I seduced by a "snake oil" mirage? Have you visited the web page of Bruce Gray, who seemed ever so reasonable?Thank you Mike

  32. GBA Editor
    Martin Holladay | | #32

    Response to Michael Legge
    I'm not sure what you are talking about. Solaris Manufacturing has a website here:

    I don't see any information on air-source heat pumps or ground-source heat pumps on the Solaris website. All I see is information on solar thermal systems, like the one shown below.

  33. [email protected] | | #33

    Air source heat pumps
    Martin-my apologies for giving inadequate information for a web search.
    "Solaris uber efficient system"
    "Solar hotwater heating CSA certified solar products-Solaris Manufacturing"
    These may give enough information for you to assess the worth of this system as to the claims made. Thanks

  34. GBA Editor
    Martin Holladay | | #34

    Response to Michael Legge
    Both searches brought me to the same Solaris page:

    Again, this is a page showing solar thermal equipment. This has nothing to do with air-source heat pumps or ground-source heat pumps -- the topic of this article. You can't rate the performace of a solar thermal system with a COP number -- so your claims are mysterious and confusing.

    1. nilst | | #43

      Your link took me to a media library page. Seems to be some sort of kink in the space time continuum.

  35. user-1127834 | | #35

    hi Martin and Dana:

    Solar peaks to normal (perpindicular) to sunlight had its knowns and then Heat content retention calculated over time in no sun pointed to some "COP"'s for thought vs one collector to another...
    Bottom line was only newer low heat capacity (of storage ability in a collector) allowed annually the greatest useful input btuh to a load, year to year, as I am sure you know.

    I do not see any address to Dana's (and thank you always) articulation in of having reviewed any of the simultaneously HW Heating in AirConditioning GeoThermal Systems, still one forced-air or reversing chiller-to-like-boiler system. [not to confuse with that now gone parallel-refrigerated off in cooling to make HW others tried ] Series-flow refrigeratiion lke an oversized full condensing desuperheater but not a desuperheater as much as another full heat exchanger (hx).

    Installed with knowns of as the others indicated:
    Lower loop pressuredrops, units with oversized heat exchangers surpassing Energy Star (then 2012) having systems well exceeding 22 compressor-tons per pump hp... designing under 22ft tdh water (unit inclusive) has allowed for 1 up26-99 grundfoos pump to handle 4 loops and 3.1/2 ton compressors [like rated size 4's] easily at 50-deg Earth available. Manufacturer only had to self rate "COMPLIES WITH ENERGY STAR" as all needed, too.
    Using not more than 18% methanol (see its msds) and sizing the loop for guaranteed 34f or more return source in coldest average winter's peaks delivered to the GT inlet, kept a good install great.

    I have seen folks discuss higher loop inlets, and above 38 I have used no antifreeze, but around Northern Ohio, I was told how I did things differently by loopers of thousands,,, to find in 6/2010 standards for instance (to rem a number) of 186 vertical damp-borehole ft per compressor ton was complimented, since 1983 has kept me from adding to my performance contracting. Substantially allowing for 3 tons of compressors (some were dual, 3-staging, closest to variables since the 1970's) to fit retrofiting 3000-3400 sq ft 44-50,000 btuh designload homes with uninsullated basements. (Zone 6 pocket 6700-6800 deg days).

    We never counted on mass heat transfer season to season as some never turn on the a/c but a couple weeks a year.

    a smallest Dual compressor of 3.1/4 tons [like some size 4's for 98% of the other OEM's, would have load matched well enough for comparing to any 5 ton MINI equipment at 5f(-15C) and Air Source rateds size 4.1/2's... but really, about at 12f and below, with minimal ducting and a high central return air (regardless of cooling season) higher than Mini's, - --? (( I see the 'percentile' now...))
    Well, an old vertival air stabilization expert from GA, Lynn Perkins (now gone vertically up for some time) showed a better resulting dehumidification letting the older a/c systems run longer, stirring more total volumes of the air -- yes I know what I might hair about that...
    GT-Recovering 100% in A/C-
    And LOOP PUMPS OFF LINE in HW Heat reclaim of 100% to the 40-gal (all that is needed) tank with this GeoT method,HW Plumbing with the tank as ALL GEOTHERMAL in a credit (ask your advisor).
    Seems lowest first cost , in any 5-7 year preview. -life-cycle thirty-plus years to others replacements and multiple cleanings not so with GEOTHERMAL-Proper.

    I hear you Martin, and learn a lot fom you too!.
    But I hardly see anyone discussing the very successful 100% on-demand Priority in A/C and PROGRAMMABLE in heating mode to top off HW Tanks constantly in winter (90%) such that we have to have another 3/4 to 1 loop- also having OEM built in 4 zoning and 8 or 9 logging and program feature sensors.
    Another loop-'ton' in the ground for this very useful full HW feature.
    Now it is Variable IQ drive after 6 years field testing by/of Bristol's V-Star 25% to 100% compressor.

    GEOTHERMAL is any final use Earth-Heat below grade exchange, not just heat production only, I have reason(s) to believe.

  36. GBA Editor
    Martin Holladay | | #36

    Response to Jon Pierce
    Sorry, Jon -- you lost me after the first (incomprehensible) sentence.

    1. nilst | | #42

      I think I figured out the first sentence but the following sentences must be translated from Klingon.

  37. greenscape02 | | #37

    Response to article
    I really enjoyed this article. The point about heat pumps not being a "renewable energy source" was very much needed. Even though the source and sink are renewable energy sources, for those who don't know any better it is a bit misleading. We are still using electricity to power the compressor, fans, and pump/s.

    Also, the information regarding the mini split systems needs to be spread within the HVAC industry. Geothermal is viewed as the "economical and green" solution by many people. But the mini split discussion needs to be figured in to any new installation. For a continued discussion on geothermal heat pumps, see While there isn't a discussion about mini split systems, there are a many tips and techniques that compare the many geothermal system out there.

    Thank you for a fantastic article!

  38. ronaldsauve | | #38

    I vote for minisplits
    I have been impressed with minisplits for some time, but mainly as an "armchair quarterback" until recently. I believe they can be a very effective source for heating and cooling, if the space is very tight and well insulated. [email protected]

    But I had not installed one until this last year, (2013), when I installed one in a walkout basement living space in my basement. While until then, the space remained at close to 60 degrees in winter, in order to make the approximately 650 square feet of living space, I was closing it off from the heat source, the small boiler. So, in order to make the space consistently comfortable, I installed a Mitsubishi 9,000 btu, 21 SEER minisplit to provide heat, and (unneeded, as it is in the basement, cooling). However, it is now used to keep the humidity level at acceptable levels, even though it is not needed for cooling per se.

    This space, as well as the entire extensively remodeled house is very tight, and very well insulated. While I have not done anything with the basement slab to insulate it, as that would have involved more than I was willing to do, the rest of the house is very tight and well insulated. The basement wall effective R value is about R20, until it reaches ground level, when it is about R35. The main house wall effective R value continues at about R35, and roof R value is about R60-65. The windows are nothing special, just Andersen lowE casements. If I had it to do over again, it would be interesting to use some of the now available much higher performance windows. But more than that, the house is extremely air tight. I also use a 200 cfm (on high) HRV, but it runs on low all the time except when it is bumped up temporarily with bath use. The total square footage for both levels of the ranch home with full basement is 2600 square feet. I use one 250 gallon tank of oil to heat and provide hot water for 4 adults for the entire year for the entire space.

    The resident of the walkout living space likes to keep the space quite warm, so I needed to have a reliable means to keep it that way. I have been very pleased with the result. This past winter here in Maine, we had a number of below zero days, and the resident was very pleased with its performance, as they never had any time when they were anything but very comfortable. And after keeping a very close watch on the electric bill, I also am very pleased, as the total added cost of electricity never hit $20 a month.

    Needless to say, I am now more than an "armchair quarterback" when it comes to minisplits! I am now considering installing an additional Mitsubishi 9,000 btu minisplit upstairs to provide cooling and supplemental heating, especially during the swing portion of the year, for the main level of the house. At present, we use an 8,000 btu window AC unit, that cools the entire 1300 square feet of the main floor. And that has its impact on our electric bill, even with an Energy Star 10.8 EER unit. But this time, I will go for the new, 30.5 SEER minisplit unit. I had meant to have the then available 26 SEER unit for downstairs, but only discovered the wrongly provided 21 SEER unit after I had already started installing it. It will be interesting to see what difference the 30.5 unit will make on the cooling bill. But perhaps even more so, I am interested in seeing what effect it will have on the heating bill, both in terms of the oil use, and the electrical use.

  39. BrianRJH | | #39

    hydronic minisplit?
    I am wondering about the same idea, with cast iron radiators.

  40. Expert Member
    Dana Dorsett | | #40

    Richard's last response was in July 2012...
    Air-to-water heat pumps exist and are somewhat popular in Europe, less so in the US. The Daikin Altherma works in most US climate zones, and could be used with cast iron radiators of sufficient EDR area to cover the 99% design load with 110F water.

    I know of at least one Altherma installation in zone 6A Maine, but it's a very low temp radiant slab floor that only needs ~95F water at design condition. That would take a LOT of cast iron radiator to pull off in that climate.

    There are also a few pre-packaged air source reversible chillers that you could design a system around, but it's an engineering project. The Chilltrix modulating chillers have some nice specs, but I've yet to see a system in a cold US climate using them.

  41. GBA Editor
    Martin Holladay | | #41

    Response to Brian Higgins
    Like Dana, I'm guessing that you are unlikely to hear from Richard Berg, who may or may not be still checking in on GBA.

    A few months ago, I wrote a detailed article on air-to-water heat pumps. Here is the link: Air-to-Water Heat Pumps.

  42. DoggoneGreen | | #44

    Our 2500 sq ft. Unity home in NJ (zone 6a) relies on a Mitsubishi Ductless mini-split with 4 interior zones. It has worked flawlessly over the 5 years we've been in the home. If your home is will insulated and sealed, there is no need for a ground-source unit. They're more expensive, more difficult to repair and maintain, and won't perform any better in this climate zone.

  43. donjahnke | | #49

    I always see the full value of 30% taken off the cost of the GSHP systems yet most of the time they are attached to a home loan. If you have a 50K GSHP system and you get a 17K rebate in a perfect world the full cost is still 50K on the loan I see this all the time where they show a bill for electricity and at times is raising the mortgage by 150-200 dollars and no one accounts for it.
    We have started using the Dettson smart duct systems where I can tie a 23 SEER ASHP to a 94% fuel fired furnace down to 30k that modulates to 15k and zone a house to distribute cost effective system that takes the most efficient approach it can Air Source or Fuel fired.
    We finished a 4000 sq ft home with a 1.2 ton load on cooling and it is zoned into 3 zones with 52 ports running at 15cfm each great results
    Would never touch GSHP again

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |