cold climate air to water heat pump auxiliary heat & AC
We are in the detail design phase of a 3 bedroom lake cottage outside of Rangeley, ME climate zone 6. Plans attached. It will be a vacation home for us and also rented when not in use. Any HVAC design needs to be “guest proof” i.e. relatively seamless to operate. Several of the articles I’ve read on GBA indicate using a wood stove, or even baking cookies to “help” the heat pump (usually an air-air) on the coldest days. Cottage will have propane available for backup power generator and gas fireplace in great room on main floor and also one on the lower level family room area. Expect that the gas fireplace used to take the chill off in shoulder seasons – not necessarily integral to maintaining comfort in the coldest of nights.
I would like to do in floor hydronic radiant heat for both levels. I’ve been researching cold climate air to water heat pumps such as those from space pak or arctic (these seem to be the top search results, open to other suggestions). They advertise performance down to -22F or so which in theory meets the design temperature. Architect and HVAC installer aren’t super familiar with the cold climate heat pumps and are suggesting that propane auxiliary heat will be necessary. From what I can tell, with the air-water system plus a good size water tank (thermal battery) a resistance heater backup should be able to cover it.
Would also like to have air conditioning. The cottage will have finished walk out basement so would like to limit intrusion for ducting so the smaller duct systems from space pak or unico are very appealing. I’m not sure how often the AC will even be needed, but given climate change trends, I believe best to install now. I’m a bit concerned about size balance of cooling needs being low compared to heating needs.
I assume and ERV or HRV is appropriate for a code built home (tight but not “super tight”)? Advice given was that maybe not necessary and prior experience that installed units don’t run much (I’m thinking the building wasn’t well air sealed).
Architect was thinking radiant in the basement slab and a forced air system for the first floor (using the small ducting) as most economical. We’ve experienced hydronic radiant floor (gas fired) in the past and loved it.
I know some will suggest the mini split option for A/C but we simply do not care for the aesthetics of them and I figure we should use something that can leverage the air-water heat pump in both seasons.
Planning to install a solar system (there will be a large storage barn on property with loads of roof space)
Primary consideration is comfort secondary are cost (installation and use) and minimizing the use of fossil fuels is.
TLDR – will electric aux heat be sufficient for air-water hydronic floor in zone 6? Can anyone recommend a firm in New England that can do some energy modeling to ensure a properly sized system and help optimize insulation/building efficiency and HVAC system sizing?
Design considerations:
9767 degree days, -13F winter design temp 97.5% (lowest ever recorded -34F)
81F dry bulb AC design temp record high 94F
Assume built somewhere between code (see attached) 3 ACH50 and R49 roof, R20+5 walls and Pretty good house standards 1 ACH50, R60 roof, R40 walls.
Thanks! Steve
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Why not use an A2WHP until the COP reaches 2.9 then shift to NG or Propane for the Heat? You may want to look at Veissman/Carrier for an air to water heat pump. The Vitocal 100 can use either electric or a boiler for backup heat.
Depending on the size of the house, a CERV2 / GeoBoost can provide some heat and cooling while also managing the IEQ.
Let's look at cold weather performance, just pulling some numbers off of the websites:
Arctic: At 7C(45F) the Arctic is rated 8.3 KW output, at -20C(-4F) it's 5.5 kW, 66% of it's 7C output.
Space-Pak: At 45F it's rated 37K BTU/hr, at 5F it's rated 19K, 51% of its 5C output.
Chiltrix: At 47F it's rated 33.8 BTU/hr, at 5F it's rated 17.0K and at -4F 14.4K, 50% and 43% respecitvely.
Arctic does the best job of keeping capacity but 66% isn't that much and -4F isn't that cold. The Chiltrix and SpacePak seem roughly comparable given that the comparison numbers don't exactly align.
By comparison, let's look at a vapor injection air-to-air heat pump, I picked the Mitsubishi SUZ-KA36NAHZ because it's about the same nominal tonnage. It maintains about 75% of its 47F capacity at -13F. (See https://ashp.neep.org/#!/product/34584/7/25000/95/7500/0///0 ; I wish NEEP did similar rating of air-to-water). At 5F it has 92.5% of its 47F capacity. Note that at -13F it's COP is 1.86, still above 1.0 so still cheaper than resistance electricity to run. This unit has much better cold weather performance than any of the air-to-water heat pumps.
Something like this Mitsubishi with a ducted air handler is going to be a much cheaper way of providing air conditioning than anything hydronic.
My recommendation would be to get a Manual J, room-by-room heating and cooling loads. Size a ducted air-to-air for at least the cooling loads. You're heating loads are going to be higher, but see where that puts you. If you still have your heart set on heated floors, price out an air-to-water heat pump with just enough capacity to cover whatever the ducted system doesn't provide.
Then there's the question of what to do when it's -34F again. Your choices are going to be propane or resistance heat, both have the advantage of operating independent of temperature. Resistance is a lot cheaper to install, and will be easier to integrate with the water system, I know Chiltrix has an inline heater that warms the water when the heat pump can't keep up. A big part of the decision will hinge on your local utility prices and what you expect them to do in the future.
Not sure why you're quoting 97.5 percentile heating temperature, 99th percentile is standard. If you're getting that number from a vendor I'd be suspicious that he's trying to hide something from you.
It appears the Chiltrix has stopped updating their website. The CX34 has been replaced now with the CX35, sold through Unico. It is closer to $7000. Arctic Heat is quickly becoming the cheapest option out there. I found that Arctic Heat is made by Macon. Some reddit folks have imported directly from them. Apparently Arctic doesn't have a great variable speed controller, they run it more On/Off.
https://www.macon.com.cn/product/evi-full-dc-inverter-heat-pumpshm32-strong-heating-stars.html
Also -- you want to plan your heating and cooling, ventilation, and hot water all at the same time.
I'm a fan of heat pump water heaters, but I'm not a fan of the integrated solutions that the air-to-water guys sell, they don't save any energy as far as I can tell and they make the system a lot more complicated and expensive.
You're going to want ERV in your climate, possibly integrated with the AC ductwork.
>From what I can tell, with the air-water system plus a good size water tank (thermal battery) a resistance heater backup should be able to cover it.
The water tank isn't going to pencil out. I can explain in detail but trust me. Be very skeptical of any vendor who seems excited about it. Be very skeptical of any vendor who throws around scientific-seeming but unscientific terms like "thermal battery." Really anything with the word "thermal" in it should raise your suspicion.
This thread covers a lot of what you're asking about as well:
https://www.greenbuildingadvisor.com/question/help-needed-from-hydronic-experts-xpress-trak-and-other-foam-underfloor-heating-panels-solid-performance-or-low-performing-shortcut
Electric resistance would be sufficient. However, you might find that it requires too much amperage. Propane is a fine backup choice too. Your propane backup could be a boiler or furnace, this is easy for an installer.
I think the most consequential decision would be 100% ducted heating or in-floor radiant + ductwork for cooling and potential heating. Both have pros and cons.
Absolutely no technical knowledge like the fine gents above. But here's a comment as a user that went deep towards Air to Water, and ended up Air-Air with dual fuel (Nat Gas), and just using some of your own words: Guest proof, seamless, rented out. Hydronic in my region is really expensive to install. Also experienced installers with Air-Water were impossible to find locally, we had one guy profess his experience, and he gave me a price over the phone---hahah not joking. Hydronic is also slower to respond from the the end-user thermostat changes - IE A renter cranking the heat after a day outside. Add complexity, Air-Water heat pump with tankage, with AC. We went with an Air-Air with a dual fuel ducted system. You have propane, consider the electric backup or propane at a set cutover temperature when the Air-Air is losing capacity at the lower temps - Someone mentioned check your local utility costs - run the numbers, there are lots of great resource on GBA and beyond to help you. Some level of redundancy as the guys above noted, propane when needed, if the heat pump fails. We can adjust temp on our heat pump from my phone, and boom, if needed the nat gas kicks on (or heat pump), and the house is toasty warm before we get into the neighborhood. Also, as a landlord of sort, consider what a call out looks like? Can you go troubleshoot on a Saturday night when the temp inside for a renter is declining? Can a regular plumber or only the original installer? As for AC, if that's a must, sounds like you're doing ducts....Again, not an expert in any way shape or form, but consider what happens on bad days when a more complex system starts to fail. Maybe someone will mention, even multiple split-heads installed inside may be more efficient and cost effective to support the heat loads? My sentiment is hydronic is fading as these heat pumps become more efficient and cost effective, and more installers have experience. Best of luck. I will keep reading this, as I'm interested in your outcome decisions.
Also, the hydronic and Air-Water install for us, was more than x2 the Air-Air...
>My sentiment is hydronic is fading as these heat pumps become more efficient and cost effective, and more installers have experience.
I would say the two factors working against hydronics are that air conditioning is becoming universal in just about every climate, and with houses that are better insulated and air sealed the interior temperature varies a lot less and it becomes less important to deliver the heat exactly where it's needed.
To paraphrase DC, even with hydronics for heating (or cooling) you still need to run ductwork for the ventilation system, I.e. ERV and AC. With R35 walls, R60 ceilings, and CFM50 of 300–you can install 4 inch or 6 inch duct and be comfortable. Just be sure to design floor/ceiling truss that handle the ductwork.
PEX in concrete makes sense for basements and garages but not for upper level living spaces. Do a room by room manual j and look at the floor temperatures required to heat the spaces. Hydronics makes a lot of sense if you have lots of glass or have a bathroom tile floor where you keep the floor heated. Just make sure you have realistic expectation before you write the check.
>Do a room by room manual j and look at the floor temperatures required to heat the spaces.
This is really the most important step. And realize that your floor won't be at the temperature generated by the Manual J all winter, it will be at that temperature 1% of the time, or about 88 hours per winter.
A big problem people run into is they install too much floor heat, for most of the winter the floor temperature is low and they don't get the toasty toes feeling they were looking for.
What works better in most situations is figure out what floor temperature you want, figure out how much floor you can heat to that temperature and meet your heating load on an average day in winter. Then supplement that with air handlers to handle colder weather.
DC & others - thanks for all the detailed feedback - I really appreciate it! I'll look at the air to air with ductwork. I wonder if one unit would cover heat on the 1st floor and AC for both floors and a second system for the radiant floor in the basement and garage slabs.
I'm not getting pushed in one direction or the other. I had a preference for the radiant floor given a prior experience - I figured it would be simpler to do air to water if the end result was hydronic floor heating. Given examples above, the air-air system maintains better performance at low temps. The prior experience was on a work assignment in Warsaw, Poland (2018-2021) and the structure was concrete and from what I could tell very well insulated and sealed (triple pane Euro windows etc.) it was powered by natural gas.
I've got a degree in Mechanical Engineering so I understand the theory, but its' been 25+ years since taking thermodynamics and there are many inputs to any heat loss calculations for sizing a system that I'm not qualified to make.
My architect and HVAC contractor are clearly more comfortable with conventional systems - I'm going to reach out to an energy engineering firm.
the 97.5% temp was from Rangeley building department. Reference weather station is a bit further north so perhaps 99% at my location.
>I wonder if one unit would cover heat on the 1st floor and AC for both floors and a second system for the radiant floor in the basement and garage slabs.
That would work. Also, if you run ductwork to both floors you don't have to rely completely on the floor heat, you can use the air to supplement the heat on the coldest days.
The reason that's important is that often if you size the floors for the maximum heating load, on just average days they're not very warm.
Looking online I see the average January temperature for Rangeley is 14F. I would size your floors so that they're running 100% when it's 14F outdoors. That way they're basically on all the time in the winter. The next thing to figure out is a target floor temperature. This is personal preference, but for most people anything over 90F is too hot, and anything below 80F isn't noticeable.
The rule of thumb is the floor produces 2 BTU/hour per square foot for every degree in difference between the floor surface and the air. So let's say you settle on 85F as a good floor temperature and 72F as your interior temperature, that's 13F of difference or 26 BTU/hr per square foot.
It looks to me like you're in Franklin County, ME, according to this: https://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/County%20Level%20Design%20Temperature%20Reference%20Guide%20-%202015-06-24.pdf your heating design temperature is -1F, cooling is 85F.
So at 14F you have 58F of heating, and at -1F you have 73F, which means at 14F you have 58/73=79% of the heating load at -1F. Let's say your heating load for the basement comes in at 15,000 BTU/hr. You want to size your radiant floor for 79% of that, 12,000 BTU/hr and have the ductwork supply the other 3,000. At 26 BTU/sf you need 460 square feet of heated floor. Next step would be to look at your floor plan and figure out which 460 square feet you would most like heated.
You should also then look at the performance curve of the air-to-water heat pump. You'll find it falls off as the temperature drops. Let's say it's only capable of 9000 BTU/hr at -1F, it would be perfectly legit to size the ducted system for 6K instead of 3K to make up the difference. Even with 9K of output your floor would still be warm -- 82F-- just not quite as warm. It would also be legit to equip the water with a resistance heat element capable of 3K to make up for the shortfall.
In terms of control electronics, the simplest thing is to put separate thermostats on the floor heat and the ducted heat, and set the floor for 72F and the ducted for 71F. The ducted will only run when the floor heat can't keep up. Easy-peasy. The air-to-water heat pumps usually have built-in outdoor reset, where the water temperature varies depending on the outdoor temperature, so your floor temperature will be matched to the load.
The size of your ducted system is going to be whatever the upper floor requires, plus the (hypothetical) 3,000 BTU/hr from the basement.
You have about ten degrees of cooling and 70 degrees of heating, or about a 7:1 ratio. Even if you have a huge amount of solar gain it's probably about 4:1. So whatever heat pump you get to meet the heating load is almost certainly going to be more than adequate for cooling as well. The only other thing to look out for is to make sure that the ductwork to the basement is sized for cooling as well as heating.
"The reason that's important is that often if you size the floors for the maximum heating load, on just average days they're not very warm."
Is having a warm floor the only reason folks put in radiant floor systems? On mild days, the air temperature output of a forced air system doesn't feel warm either. Does this mean it should be supplemented by a radiant floors on those days? :-)
Reducing radiant floor coverage just means you'll need more air backup to handle sensible cooling loads in summer (cooling is harder than heating from the floor). If you want some floors warmer than others, you probably want the rooms with these floors warmer too. Our experience is that we just need to set our bathroom t-stats a few degrees higher than the adjoining rooms to have the tile floors be pleasantly warm in winter. The other floors in the house are distinctly *not cold*, meaning we can walk around without socks or slippers all year round without, for example, feeling uncomfortable walking across the kitchen slate floor. There are other comfort benefits that I've mentioned in other comment threads that go far beyond "toasty floors" that I won't repeat here.
"I would say the two factors working against hydronics are that air conditioning is becoming universal in just about every climate, ..."
Hydronic systems support ducted or fan coil based air handlers that can be used for heating and cooling too. If you're going to the bother/expense to have any hydronic system at all, why further complicate things by adding in some mini-splits or a ducted air-to-air system? A proposed reason seems to be that the air-to-air systems are more efficient at the extremes near the design temps. These temperatures don't happen very often, spec'd COP vs COP for your installation can vary wildly, and the COP difference is rapidly closing as more air-to-water units are incorporating the same inverter based compressors as used in the air-to-air systems. So I think the design temp efficiency difference should not be the main driver for choosing one vs. the other because it is probably much smaller than stated. (e.g. we let our house coast over the coldest/hottest hours of the day and so almost never run our Nordic ATW-75 unit when it's COP is compromised by outdoor air temps).
"In terms of control electronics, the simplest thing is to put separate thermostats on the floor heat and the ducted heat, and set the floor for 72F and the ducted for 71F. "
The simplest thing (assuming you make the more complex choice of installing hydronics at all), is to install a control system that will do this for you. For example, the Tekmar 406 heat pump controller + 553 or 557 t-stats will already do multi-staging (e.g. if one heat pump isn't enough for your application), backup heat (electric/gas/oil, single/double stage or modulating boiler), blended floor and air heating/cooling, tempering of floor temperatures to avoid condensation, and humidity control. This solution has been around for almost a decade so there are probably other, potentially better, control options available today. The point is though, that you don't need to cobble together a control strategy for these applications. Competent controls manufacturers have already figured this out.
For those who haven't seen my other posts here, my experience is based on Zehnder ERV + WarmBoard + Nordic ATW-75 + Electro Industries 15k Boiler (runs a few hours a year and is probably over sized) + Tekmar controls to manage all heating and cooling of a two-story, 3400sqft, Bensonwood home in Boulder, CO. We pre-provisioned for one fan coil to handle dehumidification but haven't found a need to install it since we have almost zero latent load here. I'm happy to provide additional information about our system and what to expect comfort wise from this type of configuration.
Engineers like math do some. What will the temp of the floor be when it is 25°F outdoors and 68°F indoors Assuming the attic has R60, the walls are R15 without windows or doors and R14 under the slab floor. My wild guess is under 78° and your toes will tell you that tile floor is cold.
Walta
This is where I was trying to go to in post #14.
DC I will admit it my attention span is getting awfully short.
Walta
[responding to #17]
"If you're going to the bother/expense to have any hydronic system at all, why further complicate things by adding in some mini-splits or a ducted air-to-air system?"
Fair question. If you go back to post #2, I layout stats for three air-to-water heat pumps. There may be others, these three seem pretty representative. Our design temperature is -1F and all of three are going to be around 15K BTU/hr at that temperature. We don't know the detail of the OP's house, but I'm going on the assumption that's not going to be enough, he's going to need two units. If only one unit is needed, then you won't need both hydronic and air. But if two units are needed it makes more sense to go with one water and one air.
>Reducing radiant floor coverage just means you'll need more air backup to handle sensible cooling loads in summer.
This is Maine, where summers aren't that hot but are humid. Design temperature is 85F. Any cooling that meets the latent load is going to meet the sensible load, there's not going to be any excess sensible load that the floor can handle. There may even be a problem with an air handler meeting the latent load, running a dehumidifier all summer is very common in New England.
>On mild days, the air temperature output of a forced air system doesn't feel warm either. Does this mean it should be supplemented by a radiant floors on those days? :-)
Yes, that's exactly what I'm suggesting. Size and control the system so that the air only comes on when the floor is at 100% capacity and more heat is needed.
>The simplest thing (assuming you make the more complex choice of installing hydronics at all), is to install a control system that will do this for you. For example, the Tekmar 406 heat pump controller + 553 or 557 t-stats will already do multi-staging
What I suggested with two thermostats is the functional equivalent of a multistage thermostat. The Tekmar 406 is discontinued, but its replacement is here:
https://www.supplyhouse.com/Tekmar-tN2-406-tN2-House-Control-Heat-Pump-and-Backup-Four-Zone-Valves
Twelve hundred bucks just for the controller, the thermostats are extra. Functionally equivalent to what can be done with two wall thermostats.
" If only one unit is needed, then you won't need both hydronic and air. But if two units are needed it makes more sense to go with one water and one air."
The two feel orthogonal to me. If I need more capacity, I can add capacity many ways. In a heating dominated location, perhaps that's just adding an electric heating element in an already specified buffer tank, installing a standalone boiler, using staging to combine multiple A2W heatpumps, adding electric heat mat or baseboards in some locations, or installing some A2A capacity. The choice of how you add capacity may narrow the choices in how that capacity can be delivered, but the fact that you need capacity doesn't automatically force your hand to switch between technologies. There are so many options here that, given the limited information we have on the OPs goals, requirements, and constraints, saying any option makes more sense is much further than I'm willing to go.
"This is Maine, where summers aren't that hot but are humid. Design temperature is 85F. Any cooling that meets the latent load is going to meet the sensible load, there's not going to be any excess sensible load that the floor can handle. "
It depends on how tight the building is and how well indoor humidity is controlled. Without a carefully calculated manual J, who knows. But, having lived in a home with floor cooling, I can say that I am more comfortable, over a much wider indoor air temperature and humidity range, than I am in environments that only deliver cooling via air handlers. You aren't shocked coming indoors by a wash of frigid air. When your body heat radiates to a cool surface the effect is more subtle but still very effective.
My comment about maximizing floor area for cooling is the effectiveness of floor cooling is easier to compromise with furnishings (e.g. adding a living room area rug) and generally can reject far fewer BTUh/sqft than can be added by the same floor area in heating mode. So, if the plan is to incorporate floor cooling at all, I wouldn't artificially restrict its area.
"Yes, that's exactly what I'm suggesting. Size and control the system so that the air only comes on when the floor is at 100% capacity and more heat is needed."
There is nothing controversial in the above statement. What I disagree with is trying to slice and dice the covered floor area in some attempt to get a perfect ratio. The right questions to ask are around what is the primary method for space conditioning in any particular part of the building, are there any conditions where it must be supplemented, how often will that occur, and do the occupants have any preference on the nature of the additional conditioning. Having backup conditioning unobtrusively deploy across a range of conditions is a well solved controls problem assuming things aren't grossly mis-sized.
"The Tekmar 406 is discontinued"
Can you point to where you found information on the 406 being discontinued? The SupplyHouse.com link you provided is for the 406. It has been around a long time so it wouldn't surprise me to see it replaced, but I didn't find anything on the Tekmar site about this.
"Twelve hundred bucks just for the controller, the thermostats are extra. Functionally equivalent to what can be done with two wall thermostats."
Hydronic systems and their controls are expensive. I never claimed otherwise. But the capabilities of these controls go quite a bit beyond what can be achieved with just two t-stats. Anyone interested can make up their own mind by reading the very detailed manual (especially the section that covers sample applications): https://www.watts.com/dfsmedia/0533dbba17714b1ab581ab07a4cbb521/42881-source/406-d-03
"Hydronic systems and their controls are expensive."
Exactly. Look, I'm a hydronics fanboy, but I realize that the cost difference is real and it should only be used for things that it's better at. Heating floors is something that hydronics is good at, the only real competing technology is resistance heating and the difference in operating cost is significant enough to make up for the installation cost penalty for hydronics.
But when you're talking about provide heat or cooling to a coil with a fan blowing air over it, the cost advantage of a minisplit or a ducted minisplit is so great that it's compelling. Chiltrix has pricing on their website, the unit I discussed earlier, the CX34, two tons nominal output, is $4889. The Mitsubishi SUZ-KA36NAHZ which I brought in for comparison is three tons nominal and available online for slightly over half that. Hydronics are going to be twice as expensive up and down the line. There's not any fundamental technological reason for that, it's just that hydronics are niche products and conventional systems are broad market consumer goods.
With a conventional unit you're never going to have trouble finding someone to install it or service it or finding parts. Again, I think its fundamentally simpler to be sending water around the house than refrigerant, it's just a question of infrastructure.
I actually believe that hydronic fan coil units work better than minisplit heads, because you don't run into sizing issues nor issues with multiple heads on the same circuit. But the cost element is so compelling that it overwhelms the slightly better performance. And since minisplits are so ubiquitous you have a lot more choices when it comes to form factor. To my knowledge you can't get a hydronic fan coil unit that fits between 16" joists in a ceiling, for example, whereas I know Mitsubishi makes a minisplit head that does.
"But the capabilities of these controls go quite a bit beyond what can be achieved with just two t-stats. Anyone interested can make up their own mind by reading the very detailed manual."
Why on earth would you want to make the system more complicated than it needs to be if it does the job?
I fear this thread is now delivering diminishing returns for the OP, so I hesitate to respond at all. In fact, given the changing goal posts (from viability and complexity, to cost, and now some speculative concerns about sourcing equipment and serviceability) I’m somewhat regretting that I engaged here to share my lived experience operating a hydronic system of this type. Why post if DCcontrarian is just going to remind me that his search has yet to find a hydronic fan coil cassette that fits between ceiling joists set 16” apart? I came here for honest information exchange, not to entertain someone looking to win an argument.
“Why on earth would you want to make the system more complicated than it needs to be if it does the job?”
If you go back to the original post, the scenario is essentially “I want hydronic radiant floor heating + cooling with controls that are guest proof, in a two story cottage. “ I don’t think that two t-stats per-zone, manually configured to different temps to achieve balance between the floor and air handlers will be “guest proof.” I shared my experience because I know that all of these goals can be achieved, leveraging the hydronic system the OP is already planning to install, with a single user friendly t-stat per-zone, but it requires more advanced controls. The OP already suggested using Unico air handlers, which support hydronic fan coils. So one viable answer is Tekmar controls + backup boiler for capacity only required near or below the design heating outdoor temp + single A2W heatpump + Unico or other fan coils. The Tekmar 406 will seamlessly deliver the two different water temps required to simultaneously drive the floor and air handlers in both heating and cooling mode, deploy backup heat only when necessary, and ensure condensation can’t occur on the floor while in cooling mode. The controls can also support as many zones as you like (e.g. bathrooms warmer than bedrooms). If the A2W is sized for a heating dominant location like Maine in this case, it will have more than enough capacity to handle all cooling loads. My preference for backup heat would be an electric boiler because they are relatively cheap and ~zero maintenance, but given the onsite propane, a propane boiler could be used as well.
That’s all that I wanted to share here. Apologies for not communicating this clearly before, and the noise that caused on this thread.
Unico air handler capable of heating and cooling 2-2-1/2 tons
(https://shop.unicosystem.com/vertical-fan-coils/vertical-ahu-2430-b-coil-e-coated ):
$5,478.30
Chiltrix 2-ton nominal air to water heat pump: $4,889
Total $10,367.
Mitsubishi 2-ton heat pump, with air handler: (https://hvacdirect.com/mitsubishi-svz-ka24na2-24-000-btu-18-seer-ductless-mini-split-heat-pump.html)
$3,527.46. Total.
I'm sorry, I just can't ignore that. The Unico/Chiltrix combo does nothing the Mitsubishi doesn't do. And the Mitsubishi has far superior cold-weather performance.
The only reason to consider the air-to-water is for the heated floors. The only reason to consider not having an air-to-air heat pump is if the air-to-water can meet the entire needs of the house by itself and you only need one heat pump. The design temp is -1F. The Chiltrix can provide about 15K BTU/hr at that temperature, it's going to be a very tight house if the heating load is only 15K BTU/hr at -1F. If you need more than one heat pump the second one should be a cold weather ducted minisplit.
It appears LG is entering the North American market on A2W heat pumps?
https://www.rapidhvacparts.com/products/kphtc411m-lg-r32-air-to-water-heat-pump-monobloc-41-000-btu-h