Radiant floor heating, Boiler vs. Air to Water Heat Pump
Hi all, first time poster, but have gotten some great advice from this site during my project. I finally signed up and decided to post regarding a question about how to heat my radiant floor as we move fwd on my home build. First a little background on my project.
Our house is a 2300 sq/ft 3 bed/ 2.5 bath, single level home located in Parkdale OR, with what looks to be zone 5 climate. We typically get a good solid several month run of temps around the low to mid 30’s with several weeks that will drop into the teens or single digits depending on temp.
Our housebuild has tried to balance cost efficency with whatever energy upgrades that we can afford to integrate and has somewhat followed a “pretty good house” model. Heating will be entirely in floor radiant floor concrete slab heating, and we have wrapped all walls with 2″ of polyiso. Windows are vinyl double pane low-e (can’t remember exact U values right now) with solar defense coating. The main part of the house is a 24×48′ two story space with a loft and cathedral ceilings (8/12 upper cord, 4/12 lower cord scissor trusses with a 12″ energy heel), and the bedroom wing is a 14×52′ single stor with a 4/12 catherdral ceilings throughout. Ceiling insulation will be R-49 in most of the main living space, and R-38 cathedral batts in the bedroom wing. Underslab we have R-15 of XPS foam with 2″ of foam at the perimeter of the slab. The whole house has been pretty carefully detailed to manage air infiltration, but is clearly not up to a passivhouse standard….just trying to go a little above and beyond basic construction code and be a little more efficient than a normal prescriptive home.
My question revolves around a quote I just recieved from the company that designed our radiant system and performed heat load calcs/ etc. They originally had spec’ed an Argo electric boiler for the project, but just informed me that for an extra 3,300 dollars we could convert to a (as yet unspecified brand) air to water heat pump which the company claims will dramatically lower cost to operate our radiant system. Right now the system is three zone (main living area, master suite, guest bedrooms) and I assume that the company is factoring in boiler/ heat pump sizing according to all of their prjected heat load calculations that they performed when designing the system.
At this point, I would like to hear pro’s/con’s of air to water heat pumps and electric boilers as I have had a devil of a time finding much info on air to water installs as they seem to be in their infancy in the US. We are building the house ourselves and at this stage 3,300 dollars extra added on to the budget has the potential to be significant, but do-able if we will see good long term cost savings with the heat pump. We will also b subbing out install of our final radiant controls and need to factor any changes in cost for install into the budget.
I appreciate any info or thoughts that you guys have on this matter, thanks for you time.
-Nick
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I'm guessing the mystery system is either the Daikin Alttherma system, or the Sanden Eco-Cute system. Either way, you'd get substantial savings vs. the electric heat: something like 1/3 the energy cost or better, and you'd get that savings on your hot water as well as your heat. So $3300 sounds like a pretty good deal. If you've got someone doing a heat load calculation, they should be able to project your annual heat cost in both cases pretty easily. Or if they can't, show us the calculation and we can convert to annual cost. On the other hand, if they can't, the chances that the did the load calc correctly are pretty slim.
The Daikin is a really nice system. The only thing I don't like about it is the R410-A refrigerant which has >1000X the global warming impact of the same amount of CO2. But with the "monoblock" configuration, the unit is factory-sealed and the chances of leaking are less than with a split system install.
The Sanden even better--higher efficiency and lower environmental impact because the refrigerant is just CO2, not the R-410A used in the Daikin. The only downside I know of is that it's a new system and you might have trouble finding technicians who know how to repair it when that's eventually needed. But given how much trouble I had finding a carpenter who could install my front door correctly, I can't imagine it's too much worse than trying to get anything else done right. Also unlike an R410A system, if the technician botches the job and lets the refrigerant leak, there's no significant climate impact.
If the cost adder of $3300 seems to much, you could consider a minisplit system , which would provide about the same 3X energy cost savings, and would probably cost the same or maybe even less than the electric boiler plus hydronic distribution system.
Is your construction underway, and if so, how far have you gotten?
The low cost adder and the "...as yet unspecified brand..." air to water heat pump makes me suspicious. The Daikin Altherma would be at least an $8K-12K cost adder. There are a few others out there, but they're not that cheap.
That said, at Parkdale's ~35F January mean temperature it doesn't take an extra-special heat pump to deliver a COP of 2.5 or better, which would be a reduction in heating cost by half.
What is your heat load, 99% outside design temp, and the anticipated water temp necessary to meet that 99th percentile heat load?
Thanks for the replies guys. To answer the questions on heat load calcs, the radiant company never sent us those specifics (probably because most owner-builders don't ask for them), but he was very meticulous in asking me initial design questions and he is used by a couple of very high en builders of green homes in the area. To date the rough in package that they sent was really nice, and the design schematics were good. I will email him to ask more about heat load calcs and projected annual costs.
I just recieved this reply back from the radiant company with some answers to some of my questions about brand of the unit and install. The newness of the system does worry me a bit in regards to finding good install/ service, but it does sound like the annual cost saving will be there with a heat pump vs. boiler package.
"All good questions. The heat pump unit is a “Unico” made by “Multi-Aqua” company. It has a nice looking slim line cabinet. All of the refrigeration work is done at the factory and is integrated in the cabinet along with a pump. Just set it on a pad next to the building a run a pair of insulated water lines into the mechanical room. You could use aluminum core pex tubing for those lines in lieu of copper. These lines will be insulated and contain some antifreeze.
Here is what you would need in addition to our package:
1. Equipment pad. You can pour a small slab, use patio blocks, or use a precast pad that we could provide.
2. Outdoor disconnect switch. We can supply this or your electrician can
3. Standard 50 gallon electric water heater. Pick one up anywhere to use as a buffer tank. The wiring and elements won’t be used.
4. Some antifreeze.
5. General installation supplies, tubing, fittings, clamps, pipe insulation, wiring, etc
There is a bit more installation labor and materials than just a boiler, due to use of a buffer tank.
Yes this unit reverses for chilled water cooling. You could do some floor cooling but it much nicer to hide away a small air handler and do it from above. (Something you could always add at any time)
These systems are more efficient that your standard air-to-air heat pump forced-air system. The design using a buffer tank smooth's out the operation of the compressor. It goes through a less frequent and shorter defrost cycle during freeze conditions, and there is no extra strip heat to keep the room air warm. A weather-responsive temperature control will vary the buffer tank temperature depending on how cold it is outside. These systems will drop your heating bill in 1/2 compared to resistance electric. About the only system that is more efficient is the geo ground-loop heat pump.
Let me know if you require further information."
Also, to answer Charlie's original question, the house is basically ready for rough in plumbing/electrical then drywall and insulation. Exterior siding/ insulation is totally done and the slab is obviously poured and finished with radiant loops embedded already...
Unico has been around awhile, so has Multiaqua.
I thought the smallest unit Multiaqua made was a single speed non-modulatino 3-ton , which would be overkill for most better-than-code 2300' houses in OR. Your design heat load is probably on the order of 20,000 BTU/hr, and with a radiant slab you'll probably only have 95F EWT on design day. Your outside design temp is probably between 15-20F, depending on altitude, and you'd probably have more than 35,000 BTU/hr of output with a 3 ton compressor under those conditions.
But without the real design numbers on the load & radiation, and the real specs for the heat pump, who knows?
The quick online search pulled up this:
http://www.multiaqua.com/pdf09/Multiaqua09_MAC_036,048,060.pdf
http://www.multiaqua.com/pdf09/Multiaqua09_Catalog_Hydronics.pdf
Not much info on heating capacity across outdoor temp & incoming water temp.
If the contractor thinks he can design around it and will stand behind the design, if it's only a $3300 cost adder it's probably going to be worth it. Be sure to install the outdoor unit where it won't be clobbered in a roof avalanche after a big winter storm.
It's worth finding out just how many of these the company has installed and getting some references. A hydronic heat pump is more than just a drop-in replacement for an electric boiler, and aren't nearly as tolerant of short-cycling as electric boilers are.
So I got a bit more info back from my radiant designer. He has not recalculated my heat loads/loss since we wound up adding the 2" of exterior foam insulation after he had originally calculated the job based on a standard R-21 job. I also forgot to mention in my original post that we purchased a used 4 year old Vermont Castings wood stove (for a "you'd be stupid to not buy it deal") for backup heat in case of power loss and ambiance, but do not plan on using it for regular heating.
Here is his reply on the heat pump model # and specs.
"This is a private labeled Mach-060 unit with optional low ambient controls. The technical description is a “Reverse Cycle Chiller”. It is very desirable to oversize this unit to minimize or eliminate any back-up heat. The buffer tank would be sized to the compressor in order to prevent short cycling. The heat pump would not directly see the house heating load, it’s target would only be the buffer tank. This is especially important in hydronic systems that are zoned."
At this point, it sounds like it will wind up being a cost savings to operate the system for sure, and may also set me up well in the future to install A/C using the heat pump (house currently is planned to passively cool with overhangs, solar defense glazing, and window layout designed to take advantage of stack effect...during our very hot summer it already did a good job staying fairly cool with no interior insulation).
The downside is obviously initial cost and whether I can fit it into the budget, as well as whether the increased complexity of install is worth it. After talking with my radiant guy again, it sounds like the reason he can pass on the low cost of the unit is because he bought out the inventory of a local HVAC store that was going under. He states that parts are available from the factory for the unit and that it would carry a standard factory warranty.
Here is a link I found to what looks like the exact unit.
http://www.multiaqua.com/pdf11/2011_MACH_CATALOG.pdf
On p.9 (p.10, pdf pagination) you'll see that with 115F EWT at 17F outdoor temp (near your design temp), the thing cranks out over 3 tons (1-ton=12,000 BTU/hr) even if glycol loaded.) That's 26-40,000 BTU/hr to cover a load that's probably less than 25,000 BTU/hr.
It's capacity at an even lower water temp (and higher outdoor temp) will be even higher.
Because the thing is going to be ridiculously oversized for even the whole-house load and radiation, sizing and control of the buffer tank are critical for both equipment longevity and efficiency.
Nick,
I've got one of the 25Kw Argo units and can tell you at best you get 1:1. I have easily spent $400 + in a month when trying to use my floor heat. My situation does differ in the fact that the builder didn't know what air sealing was and completely forgot to insulate the slab.
Looks like it's way oversized, but you are getting a good deal on it. And unlike a regular forced air system, there's no penalty for oversizing, as long as the buffer tank is big enough. It sounds like your contractor understands that. A big tank would also be a good investment for being able to allow the utility to time your electricity use according the needs of the grid once there are systems and incentives to do that.
If you run lower temperature water than the 115 it's spec'd for, you'd get higher COP and higher heating capacity. Ideally there would be a smart control system to automatically adjust that setpoint as needed but I haven't seen one I like for a system like this. But even if you leave it at, say 105 year round. you still save a lot vs. the electric boiler.
Do you have any nearby neighbors? You have enough capacity that you could run an insulated pipe to your neighbor and sell them some heat.
Thanks for all the replies guys, this has made my decision process a lot easier. I talked to a buddy of mine who builds high end green homes in the area and he said that he is actually considering starting to go to the air to water heat pumps too.
One thing that seems like a potential nice benefit to having the unit be oversized is that in the next 3-5 years we plan on adding a detached garage/ shop/ ADU about 20' from the side of the house where the compressor will be located. I would assume that given it's oversizing for the job I would probably be able to run radiant heat to the shop/ADU and not max out the units capacity, which is a nice way to future-proof my project.
I doubt building a buffer tank big enough to time-shift the power use makes financial sense, even with time of use rates, but undersizing the buffer and having it short-cycle would be a problem.
I'm still trying to figure out how the contractor can do this with only $3.3k more than an electric boiler. Buffer tanks of any significant size can be pretty expensive in themselves, and I doubt very much that the 5 ton chiller costs less than $2K f.o.b. the distributor's loading dock, then there's the design time, the extra control hardware, etc.
Putting crayon to napkin here ( lipstick to mirror? ) a 50 gallon buffer has ~400 lbs of water in it. If controlled to swing from 100F to 125F that 25F delta takes 25F x 400lbs= 10,000 BTU. At 50, 000 BTU/hr (about the capacity this unit has has at your ~35-40F mid-January average temp with 115F LWT) that's about a 0.2 hour cycle= 12 minutes. When it's 45F outside the thing puts out about 60K, which shortens the cycle to about 10 minutes, which is about as short as you'd want it to be. If controlled to a much narrower temperature swing or with much less thermal mass it incurs an efficiency penalty. When bumping along in sub-5 minute bursts it begins to look like an outright efficiency disaster.
It'll be hard to run it with swings more than 25F, and an LWT of 125F is already of dubious efficiency, so you really NEED the water temp requirements for the radiation to be lower than 100F or a much bigger buffer (to allow smaller temp swings in the buffer) to get decent efficiency out of it.
The question is , What is your lowest heat load expected to be ? Then and only then can you begin to properly size your buffer to eliminate short cycling of the HP .
Minimum firing rate - smallest zone = BTU surplus
BTU surplus x desired minimum run time ( usually 10 minutes) = cycle factor
Cycle factor / Delta T x 500 = tank capacity . Delta T for radiant floor is a recommended 10*