My experiment: 3 hybrid heat pump water heaters in series to heat superinsulated house.
I’m headed toward using 3 5o gal Rheem hpwh to heat my house. Large house, but R50-R70 shell. I’ve built a number of superinsulated homes in Fairbanks, Alaska and 50k btu was more than enough, so I’m hoping 10-20k BTU will work here in Michigan (Traverse City).
The HP BTU/hr is 3×4200 w/ backup elec 3×5000 = 27,600 BTU but I think 15k or less would handle heating most days.
Location: Large underground utility room. Will duct conditioned air into HRV ducting for free summer AC.
I’ll probably run the heated water through a dual coil tank and use both that tank and my slab as the heat sink/storage. Domestic hot will come from one coil and I’ll hook up solar thermal later.
I have not seen this set up before, but I want the following:
* Easy to install
* Cheap, commoditized product that’s easily replaced.
* Does not require dealer/licensed plumber for install/replacement
* Defacto back up: If one unit fails I’m not 100% down.
* Expandable for little cost.
Thought I’d post this goofy idea here and see what the experts think.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
A heat pump just moves heat around. In your case, where will the heat come from? From the earth surrounding an uninsulated underground room?
Put the three water heaters in parallel (vs series). Consider the duty cycle a HPWH is designed for and an air to water heat pump (eg, Chiltrix or Arctic) instead.
The underground (earth sheltered) utility room. Eventually, there will be a sauna in there. I'll also pump solar thermal into that room. I'm not concerned about available btus from the surroundings/location.
The floor of a superinsulated basement is a really lousy heat exchanger for pulling heat out of the soil. It gets even worse if you insulate the slab.
If you're pumping solar thermal into the room only to store it in a heat pump water heater you're wasting steps, and energy. Solar thermal does just fine storing the heat in low-pressure insulated tanks.
I strongly recommend doing at least the napkin-math on this. It looks like you're hacking away at a Rube Goldberg contraption in an attempt to violate some pretty basic laws of thermodynamics. The laws of physics are self-enforcing.
The setup you are proposing is similar to a horizontal geothermal loop, just built under the house. Since it is a house, your limits are slightly more severe as you can't freeze the ground around the foundation. Depending on your soil type, and 4' depth, you might be looking at 3 to 5 BTU/sqft. You'll need a very large utility room to get 15000k btu out of it. No amount of heat pump magic will get you around this fact.
There are some low temp ducted HPWH which would take air from outside, but since they don't have a vapour injection compressor, the wintertime heat output will quickly fall off once the temperature drops much bellow 30F.
The best you can hope is to heat the house during the shoulder seasons with the heat pumps and heat during the real winter with a wood, gas or electric(resistance or low temp mini split).
Akos, Dana: Thanks for the thoughtful replies. Akos I don't understand the first part of your comment. I am pulling heat from air. The details of that have not been posted. There is an exterior passively heated air source entry into the room as well. It's not superinsulated. Etc.
The heat I get is going through the slab to heat the house. As an add on, I have another tank that will store solar thermal (basically pumping btus into the room) the heat pump loops that heat the house will route through the heat exchanger in that extra tank, so in general, the theory would be: 1) get heat from solar thermal, if that's not adequate, heat pumps kick in, if that's not adequate, electric resistance elements kick in. Is there a reason that Rube Goldberg contraption wouldn't work if the air temp in the 2000 c.f. utility room stays between 40F and 80F?
But you are hitting on an important point and one of the reasons I posted here. I am having difficulty finding details on the heat pump compressors themselves. AO Smith claims theirs work at lower temps, but there's no detail or even btu/hr numbers.
Since the cans of worms will likely be pried open one by one, I should also mention I have a 10kw solar array and net metering, so using electric resistance 2-3 months out of the year does not seem like an issue I'd care about.
The heat in the utility room has to come from somewhere, which is through the ground.
The setup you are proposing is a ground source geotherm but instead of getting nice heat transfer through direct contact of the geo loop, you are coupling through air. This makes the heat transfer even less efficient. Also since you won't be down to 8' or 10' depth, the soil temp will be colder, so you get even less heat out.
What you'll find is that if you start running those 15kbtu of heat pumps for any extended period of time, unless the space is large (think underground garage size) the air temperature will drop very quickly.
Heat pump water heaters in an unconditioned basement of a house work well for DHW because they don't need to extract that much energy. A water heater needs to typically supply an average of 1500btu. Roughly 65% of that will come from the space, the rest from electricity to run the heat pump, so you are taking just under 1000btu out of the space. That is significantly less than what you are proposing.
If you have a 10kw array, take the cost of all that complexity above and buy a nice cold climate mini split. Even better, take the cost of the solar thermal and buy a couple of extra PV panels. This extra PV energy feeding a min split will make much more heat than the thermal panels ever will.
I know it sounds pretty boring compared to your idea. When it comes to houses, boring is good.
There was another recent (last day or two) thread with a similar idea. It won’t work. Heat pumps don’t “make” heat — they just scavenge heat from somewhere else. In the case of a regular heat pump water heater, heat is scavenged from the air in the basement. As you pull heat out of the basement, the air gets colder (less heat), and you eventually reach a point where the air is too cold for the heat pump to effectively operate. With normal air source heat pumps with outdoor coils, this “too cold” point is the point at which you start having to use auxiliary heat sources instead of the heat pump itself.
As Dana suggested, I also think you’ll find that you can’t get enough heat into the basement through the slab to make your idea work. I think you’d be much better off insulating the basement to limit heat loss and installing a regular air source heat pump with an outdoor coil. If you want to use geothermal heat, you could go with a ground source heat pump.
Dana also recently mentioned, in that other thread, a heat pump water heater that has an outdoor coil. That could potentially work for what you’re trying to do, but I still think your idea is overly complex for what you’re trying to accomplish.
Bill
Zephyr7, I know how heat pumps work. I made no mention of trying to heat through a slab into the basement. Not possible or planned. It's only a utility room separated from the house, and the heat comes from the ground and outside air (preconditioned).
I've look at the Sanden unit. I like it. It'll work. But what I'm trying to accomplish is listed in my original post, and such units (all of them) do not meet any of those criteria.
However, while I can easily calculate how many btus I need, I'm not clear how many btus I can in reality get given the compressor type (no data), temperature variation (no data), and available btu transfer via air/ground in my setup (don't know how).
I don't think there's any question I can supply my domestic hot water (but I'm still investigating the potential of space heating). See this excellent report of Sanden vs Rheem: https://energy350.com/wp-content/uploads/2018/11/FortisBC-Heat-Pump-Water-Heater-Presentation-Slides-7.18.2018.pdf
You may be able to get a close estimate of how many BTUs you can move through the compressor if the hot water heater gives a refresh rate. If you know how many gallons you can get per unit time, and you know how much you’re raising the temperature of that volume of water, you can calculate how many BTUs went into heating that volume of water that much that quickly. Might be something to help you work out the info you need.
BTU transfer through the ground is tricky. You can work it out based on thermal differential and the surface area, but the ground isn’t an infinite source of heat, so you’re thermal transfer will slow down over time and how much and how quickly I do not know. That’s a bit of an interating problem, actually. The ground source heat pump guys have deal with this, so you might be able to get some info from those people.
It’s always tricky when trying to use things outside of their original design intent. I’ve done a little of this trying to use waste heat from data center air conditioners to heat adjacent office space in the winter.
Bill