GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

HPWH for radiant heat

brixton | Posted in General Questions on

Would love some advice.
Here’s the situation. We have a 65g HPWH for our potable HW that probably needs replacing. It’s having issues.  It is located in a 550 sq.ft. garage that never seems to go below 55 degrees F. The lower level of the house (1000 sq.ft), not counting the unheated garage,  has radiant heating through a concrete slab. The house is well insulated (for the Pacific Northwest area) and the heating load is around 6600 Btu. for the afore mentioned lower level. This heat is currently provided by a 5kw electric boiler which ends up using almost 50% of our annual (all electric) energy. I would like to use a HPWH for both the potable and hydronic. 
I’m aware of the Sanden equipment and have been told by the company rep. that we would be a perfect fit for their system based on the limited radiant system, our potable needs, and our temperate location. The downside when compared to the other option/s  I’m considering is the cost of the equipment and installation, plus there is no back-up if the unit stops working unless of course I add something.
So here’s the big question. What if I use a new HPWH in the garage to do what I’m sure the Sanden could do? Would it be asking too much of a new Rheem Performance Platinum 80 gallon unit? The compressor on that unit is rated at 4200 Btu, but there is also a 4500w element for high demand. When the element comes on obviously our efficiency is back to where the boiler is, but we do not have too many very cold days here so can we heat the slab on most days without the element coming on, assuming the potable draw is not too high? The radiant is a closed system so I know I will need to use a heat exchanger, I’m thinking the Taco X block could work. The other option is possibly combining 2 40 or 50 gallon versions of the Rheem. I think my garage air volume is big enough for 2 units, and this way I would have 8400 total Btu of heat pump capacity in the hope that the elements would rarely if ever come on. The upsides (over the Sanden), in my mind at least, are savings on equipment and installation, and some redundancy as there are 2 HPHW heaters that also have elements should there be an issue with either heat pump. I know the Rheems are not made for this kind of use so I’m sure any warranty would be voided. As for efficiency, I’m not smart enough to figure that out. I believe the Sanden operates at somewhere between COP 1.5 to 6 or 7 at peak, although because the unit sits outside and we will be using the radiant and more hot potable when it’s cold, my guess is that a realistic COP over the year is probably around 3. The Rheems are rated at 3.5 to 4 COP depending on tank size. As they are not outside, I’m thinking they should still be quite efficient when the heat loads are high.
Sorry for the long description, but this is not a common situation so the solutions are probably limited, and I wanted to provide enough information in the hope that there is a specialist out there that can steer me in the right direction.

Thanks for any advice.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. onslow | | #1


    Simple answer - it is not very sensible method using available HPWH. The Sanden would be much more efficient over the temp range you are going to experience in the PNW.

    Be very wary of COP claims. The COP is merely an expression of how much energy you expend to move BTUs. ALL COPs fall as you attempt to move BTUs into a higher temp reservoir. The typical HPWH needs to have a resistance element to make the stored water hot enough fast enough to be useful for normal household usage. Your system demand will likely place the load profile in a range that would demand the resistance heat be active. Cold weather, high demand, low temp in garage would mean low available heat to capture, so resistance kicks in.

    The idea of two in a garage only doubles the temperature drop you will experience as a result of "extracting" the heat value of the air in the garage. If the garage doesn't freely exchange air with the outside world, then the available heat to extract will fall as you cool the garage air volume and the efficiency will drop.

    The Sanden sits outside so the volume of air is limitless. The COP of 6 or higher is for fairly warm air temps, but the COP is still very high down to your low season temps. And they have a presence in your region, so they will know real performance numbers. Still, if an installed system is several thousand dollars, you need to look at the net cost of electric using your current heater versus the new system cost. Are the electric costs low in you area? How long is the payback? How much attention does each system require for maintenance and repairs?

    Hope this helps.

    1. brixton | | #2

      Thanks Roger.

      To answer your questions, I think our electricity costs around $0.11 per kwh.
      As for payback, if I am close in assuming the Sanden will heat the radiant system at 1/3 the cost of the 5kw boiler, and perhaps the potable 30% more efficiently than the current HPWH, then I should save around $500 a year, so the payback is perhaps 10 to 12 years depending on the labour costs for the Sanden. There's a pretty good chance we'll be in the house that long. I was trying to be a little conservative in the efficiency comparisons to the current equipment, but if anyone thinks I'm off base with that, please advise.
      The Sanden would be at least $1600 more to buy than two 50 gal. HPWH, and then there is the additional labour to do the exterior install with electrical connection.
      As for the temperature drop in the garage, I don't know what the drop caused by one HPWH is, as it has been there since day one. Hard to know how much below the 55 degree low point it would go with a second HPWH. It's an unheated garage so I really don't mind, and it would still be significantly warmer than the outside low temps where the Sanden would be.
      I know the Sanden is probably the ideal equipment for my situation, but I'm just trying to work out if there is another way to get the job done. One that costs less to install (based on this particular situation), has a little bit of built in back-up thanks to the 2 units that also have resistance elements when needed, and will cost quite a bit less to run than what I have now. When it comes to efficiency vs the Sanden, the 2x HPWH system would have at least a $2000 head start.

      1. charlie_sullivan | | #7

        You can't just assume the two water heaters would chill the garage twice as much: water heating is intermittent, whereas the heating could be continuous for a week or more during cold spells. If you don't mind your garage getting cold, and you are seeing it as a bonus if it does better than straight electric heat, you could try it, but it's not clear it would really give you much benefit.

        What about a Chiltrix air-to-water heat pump serving both needs? I think it's oversized for your needs and I'm not sure it's any cheaper than the Sanden, but maybe you can use it for heating other parts of the house too.

  2. Expert Member
    Akos | | #3

    Standard heat pump water heaters remove on average 2000btu out of the space they occupy. Adding in a 6600 BTU load on top of that will quickly over cool your garage to the point that the unit will be switching to resistance backup.

    The bigger issue is that these units are not designed to run 24/7, which is what you would be asking from the units. Even adding an uninsulated recirc loop to one can cause it to run continuously. Might work for a while, but it won't last.

    As for efficiency of the Senden setup, some studies out there have shown an average COP of around 2 in combi application. Much better efficiency is possible, but it requires very careful design and keeping the return water temperature to the unit very cold. So unless you are ready to dedicate some serious design/tweak time, you'll never get a high enough COP for the unit to pay for itself.

    Unfortunately in this case, with cheap electricity, the existing resistance boiler is your best bet.

    It would be great if there were more heat pump based options out there in the 15k to 20k BTU range.

  3. this_page_left_blank | | #4

    Just an FYI, the COP numbers you are quoting for Rheem are not COP numbers, they're UEF (Unified Energy Factor). I don't know how those two compare, but they're not the same thing.

  4. tommay | | #5

    Are you sure you would like to mix potable water with heat loop water?

  5. brixton | | #6

    The radiant loop is, and would remain closed, and would take the heat energy from the potable system via a heat exchanger. I understand that the Taco X Block does a great job with this and has a variable speed pump. It does cost at least $1600 though.

  6. Expert Member
    Dana Dorsett | | #8

    What is the water temp requirement of your radiation? (It really matters with heat pumps, not so much with electric boilers!)

    What others said about the folly of using heat pump water heaters (including Sanden's) in this application.

    A small R410A refrigerant reversible chiller (like a Chiltrix) would probably handle it at high efficiency assuming reasonably low water temp requirements.

    There are cheaper ways to isolate the potable from the heating water than the X-block- you're WAY ahead of yourself on the system design.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |