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

Community and Q&A

Hybrid heat pump water heater questions & recommendation

akg4y23 | Posted in Green Products and Materials on

I am building a home in the central VA area and have some questions regarding the optimal water heater set up for my situation. I have a wife and 2 kids. We all like relatively hot showers. We tend to shower/bathe in AM, do laundry a few times a week usually in mid afternoon on weekends and evenings on weekdays, and run the dishwasher late at night around 9-11pm. Essentially we dont have much high flow simultaneous usage except AM showers once the kids get older.

Floorplans for 1st and 2nd floor are attached. We will be using the 2nd floor MBR and the 1st floor MBR will be unused for now.

My question is what would the ideal setup be for water heating in this house? I was planning on doing a heat pump water heater with either a 50 or 80 gallon tank, possibly with point of service electric heaters at certain locations such as one to feed both jacuzzi tubs in the master baths.

My plan is:
1. 50 or 80 gallon hybrid heat pump water heater in the top left corner of the garage so short distance from the master baths, laundry, and one of the upstairs bathrooms for one kid. This will reduce the waste each time hot water is used.
A) Is that location ok for noise etc from the heat pump?
B) What are the recommended models for heat pump water heaters at the 50 and 80 gallon level?

2. 15-18KW Point of service electric water heater located in the first floor master bath with hot water input from the main water heater and feeding the first and second floor Master bath jacuzzi tubs. I am doing this primarily to make sure the jacuzzi tubs can be filled with hot water since the tank water heaters tend to have issues filling them before becoming lukewarm.
A) Should I also have this POS heater feed the showers since they will not be used at the same time and since the input will be from the hot water tank? I was considering doing this to be able to leave the tank water heater at a lower temperature and just have the POS make up the difference for showers since we rarely use water that hot elsewhere. ie set the tank water heater at 100 and the POS at 120 for example. The reduction in the tank water heater usage will have an added benefit of reducing the time the heat pump is on and noise from the heat pump.

3. Similar 15-18KW POS electric water heater with input from the hot water tank to feed the remaining 4 upstairs bedroom shower/tubs in a similar fashion.

4. 10KW POS electric water heater to feed the faucets of the two upstairs bathrooms on the right side of the house.

5. POS electric water heater with low flow rate for kitchen sink and island sink.

6. Direct flow from the hot water tank to the master bath faucets, dishwasher, laundry, one upstairs bath faucet, powder room, bar, and basement bathroom.

The purposes of this setup in my mind are:
1) Provide essentially unlimited hot water and more instantaneous warm/hot water to the showers and tubs.
2) Reduce draws from the hot water tank from the faucets.
3) Reduce the usage of the heat pump water heater for both noise and general efficiency since keeping the tank 20 degrees cooler will probably result in efficiency savings which offset the cost of using a small amount of electricity each day.

Any help & thoughts you all could offer would be much appreciated.

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. GBA Editor
    Martin Holladay | | #1

    A KG,
    Let's see. The house has two floors, and each floor measures about 56 feet x 56 feet. You have 6 bedrooms for a family with 2 kids. You have 5 bathrooms, two of which have a Jacuzzi. And you'd like to know how to optimize your hot water system.

    I know that this suggestion probably doesn't appear helpful. But one way to optimize your hot water system would be to eliminate 3 of your 5 bathrooms.

  2. akg4y23 | | #2

    Youre right that isnt helpful. This isnt a commentary on what I can or cannot afford to build, nor do you have any idea what the situation is at my house in terms of guests or parents or in-laws coming to visit.

    I could say more but Ill leave it at that and hope other members here have actual suggestions on the question itself.

  3. curtkinder | | #3

    First of all, I'm compelled to echo Martin's observations about sheer profligacy and add a couple of my own. I'm no advocate of "saving the planet" by shivering in the dark, hungry, naked and dirty, but there is something to be said for living more simply so that more others may simply live.

    1) It is hard to wrap my mind around offering free technical assistance to someone, who, bathroom count aside, "needs" a 400 SF master bedroom. The 200+ extra SF in that room has diddly-squat to do with forecast visitations by guests / parents / in-laws / asylum seekers.

    2) You seem bound and determined to acquire a heat pump water heater, which I applaud, and then set about finding schemes to minimize its operation. That seems counterproductive and pointless.

    All that said, here are my answers to a couple of your questions. Garage is an excellent location for a HPWH in Virgina. Noise varies among models, but none should present a problem in a garage location given that garage walls are required to be both insulated and air sealed from interior rooms.

    The GE HPWH is quite quiet, suitable for installation almost anywhere appropriate ventilation can be provided. So far, GE only offers a 50 gallon model; I quite wish they'd release the 80...promised in January 2013. AirGenerate models claim to be quiet, but I have no experience. Same with Stiebel Eltron.

    I have limited experience with the AO Smith, State, Reliance, American, Kenmore, Whirlpool, Craftmaster models (They are, to my knowledge, identical 60 or 80 gallon units with the same powerhead). My experience is that they are quite a bit noisier than the GE, featuring beefier compressors and much higher airflows for faster recovery. I'm reluctant to deploy them owing to the noise, but probably OK in a typical garage.

    I tentatively support POS electric heaters fed by somewhat distant HPWH, but I believe you have a sub-optimal setpoint scheme in mind. Better to set the HPWH setpoint somewhat higher (~125) than the POS heaters (~115) so that line losses are accommodated, allowing the POS to cease firing when HPWH-heated water arrives.

    Tankless electric heaters have other limitations:

    1) Vulnerability to hard water
    2) large electric service requirements
    3) limited flow rates at realistic temperature rises - Math: an 18 kW heater will only flow about 2 GPM while providing a 60*F temperature rise, which I'd expect to be needed in Virginia. 2 GPM is a real slow Jacuzzi fill rate.
    4) High first cost (not an issue, I guess)
    5) Heaven help your light bill if in future some form of demand charges appear, perhaps amounting to $5 per peak kW every month, and you have deployed several electric tankless heaters that come on near same time.

    Another option is demand based recirculation pumping - while preparing to bathe, touch a button that runs a small pump for a few minutes, sending cold water standing in long hot water lines into bottom of water heater and filling said lines with fresh hot water - faster supply, minimal waste.

    I hope the foregoing helps with your design.

  4. akg4y23 | | #4

    Curt, thank you very much for your response. Let me walk you through my thought process on why I was planning on minimizing the HPWH operation using the electrics, and why I was planning on setting the HPWH below the electric temp and maybe you can let me know if there is an error in my logic.

    I see two major cons for the electrics, first being that they take 1-2 min to warm up, and the second being that cost-efficiency wise they cost more to heat the water than gas. The pros being that they only heat water being used, versus the HPWH which has to keep a 50 gallon tank heated, provide a larger supply of hot water, and they take up a small space.

    For the HPWH I see the major con still being the fact it keeps a 50 gallon tank heated even though significant hot water usage in the house pretty much only occurs at 2 times of day (on weekdays at least), and it makes a lot of noise.

    So my thinking with the setup I describe is as such:
    1) The HPWH provides a continuous supply of lukewarm water (90 or so degrees), reducing the wait time for warm water since the water will reach the baths on the left side of the house, the laundry, and the bathroom near the laundry within seconds from the garage location of the HPWH.
    2) At 90 or so degrees the HPWH will not need to run as often and thus will create less noise and likely increase the efficiency as more of the heating will be via heat pump versus gas, and the heat pump should be able to maintain the baseline temperature most of the year.
    3) Having the input of the electrics connected to the HPWH output means the input water temp will be around 80 degrees after line losses and I can use a low powered electric (I think even the EcoSmart 11 might be enough) to heat the water before it reaches the faucet/shower/etc. This provides the benefit of both the HPWH and the on-demand efficiency of the tankless, essentially creating a system where you have reduced the energy required to keep a tank heated (120 degree water will lose heat much faster than 90 degree water) by providing a method to heat the water just a small amount on-demand.

    If I was to run the system as you said with the HPWH at 120 degrees with the delay I read about for the electrics to fire up I feel like there would be no benefit to having the electrics other than an unlimited supply of hot water since the majority of the water used will still be fully heated by the HPWH in that situation.

    Thanks again.

  5. curtkinder | | #5

    I think you are overly concerned with the standby loss from an ELECTRIC storage tank heater. It is about 5% in normal circumstances. The tankless heater marketing armada loves and fosters the belief that the loss is much higher. Their lies are based upon a grain of truth - a tankless heater IS markedly more efficient than an old fashioned center flue GAS fired storage heater - standby losses of those are 40-50%, but that option is not on your table. Tankless heaters have some losses; they are not 100% efficient.

    Your scheme is equivalent to running errands using a 10 MPG F250 so as to reduce gas used by a 30 MPG car parked right next to it. Whether an HPWH makes noise 3 hours per day or 4 hours per day makes little difference. They do NOT become noisier at higher temperature setpoints.

    If you size the tankless units for lukewarm water at the inlets rather than cold, you give up the one clear advantage offered by tankless - unlimited hot water during occasional high demand situations, such as hosting guests. Once the 90*F water from the HPWH is exhausted, output from underpowered tankless electric will become tepid, pleasing no bathers.

  6. akg4y23 | | #6

    Thanks again, very insightful. Im not sure why I thought the hybrid water heaters were gas + heat pump not electric + heat pump, maybe because the original AirTap/AirGenerate was an add on for traditional water heaters which were primarily gas. Given that these are electric hybrids I think youre right might not have any real purpose to doing the combination system. I wish I had information on the total amount of energy the GE would lose in standby losses over the course of a year if set at 120 degrees vs 90 degrees though... cant seem to find standby loss reporting or really any loss information on the GE at all.

    Regarding the analogy though I think I would say it would be more like if you had a hybrid car that gets 35mpg that you leave on all the time and just use it in short bursts and the occasional long trip versus using the F150 for short trips and the hybrid for longer trips.

    Regardless though I think you are probably right, even if there are any gains in efficiency with my scheme it probably isnt worth the complexity, increased likelihood for problems, and cost.

    Thanks for your help. If anyone has a link to where I can find information on standby losses for HPWH please let me know.

  7. akg4y23 | | #7

    Just an update, I found the info (on the prior model GE Geospring). The tank loses approximately 3.8 BTU/(hr*F). Total tank loss with the tank set at 140 degrees over the course of a year is 450-650 kW/h, about $50-60. Considering the 2.4 EF of the unit that means the cost of running a separate tankless electric heater in series would probably reduce energy efficiency if using 75-100 gallons a day of hot water.

  8. curtkinder | | #8

    I'm glad you found that #, I remember it from somewhere, but lacked motivation to dig it up. Assuming a 60*F average ambient air temp where heater is mounted, I get the heat equivalent of 780 kWh. The HPWH may operate at a COP of 2.0 during maintenance reheating from standby losses, so that would be around 400 kWh.

    IMO you shouldn't need a setpoint of 140. At a lower setpoint of 125-130, two good things happen - standby losses are linearly reduced, and COP of HPWH rises.

    From an energy and complexity point of view, an on-demand pump system seems more attractive than heaters in series. OTOH, the need to accommodate occasional high demand (houseguests) would most reliably be met with full-rated tankless units able to meet flow and heat rise independent of the HPWH - deals with the scenario of an exhausted HPWH passing cold water through to tankless heaters.

  9. MJDesigns | | #9

    If you have access to natural gas, why not just install a product like AO Smith's Vertex 100 ... 50 gallon tank that will provide you with more than enough hot water for the entire family. That's the fundamental purpose ... having the hot water when you need it. Natural gas is the way to go if it's available to you as opposed to electric.

  10. curtkinder | | #10

    I suspect (hope?) OP would have mentioned NG if it was available. Not many have it away from older core metro areas.

    Some folks object to flammable gas in any form in their home, even if NG is available.

  11. akg4y23 | | #11

    I believe NG is available however I am looking at the HPWH due to the energy efficiency. 50 gallons should be more than enough for us, as it has been enough in our current house. I am not looking at tankless NG because it takes 20 years for those to pay themselves off. The way I figure it the HPWH has a first hour rate of 63 gallons which is more than enough for 3 10 minute showers, and that would also presume the showerheads are spouting out 2 gpm of pure hot water which rarely happens since typically if the heater is set at 130 degrees there is some mixing of cold water at the showerhead. Hopefully GE will release the 80 gallon version before we build and it will be a moot point.

    Curt, I found the numbers for the tank losses on the DoE evaluation of the first GE Geospring model, I figure the new one is likely similar if not better regarding standby losses.

  12. Expert Member
    Dana Dorsett | | #12

    You can't really fill a Jacuzzi with a 50 gallon HP electric tank + 18kw tankless at a reasonable rate unless you really crank the setpoint of the tank to the max and are monitoring the temp closely to throttle back flow once the tank is nearing depletion.

    Methinks you're misreading the meaning of first-hour gallons, which has a prescribed temp rise in the fine print, and has nothing to do with the set point. With 50 gallons at a 130F setpoint with 50F mid-winter mix water you have a theoretical max of 68 gallons of ~105F showering (its actually less due to turbulent mixing in the tank), which is really squeaking it for three 10 minute showers at the low-low recovery rate of heat pump water heaters, especially if you ever have more than one shower running at time, or with ANY other draws during showers.

    If you care about showering performance & efficiency (and you probably should, if heating hot water with electricity with a heat pump water heater) a drainwater heat recovery heat exchanger downstream of the shower drain flows can cut the energy use literally in half. When looking at the carbon footprint and other air/water pollution externalities of Virginia's power grid, even with a heat pump water heater it's "worth it", even if your electricity costs are low. Your electricity costs could easily double if carbon ever gets taxed at the true cost of the externalities- think of it as a hedge. It also doubles the "apparent capacity" of the tank heater, which may be a real issue with heat pump water heaters when the kids hit the "endless shower" teenage years.


    No matter how you heat your hot water, it's cost-effective to go with at least the IRC 2012 code min of R3 of pipe insulation on all hot water distribution plumbing. The 3/8" wall closed cell pipe insulation sold at box sold at box stores is only R2, but it's not hard to find 5/8" wall R4 goods online or through plumbing supply houses.

  13. curtkinder | | #13

    Dana, an even more dedicated and knowledgeable blogger than I in many areas, makes several good points, to which I offer some tempering comments:

    1) Drainwater heat recovery. It's a simple concept that works well under certain circumstances, chief among them showering, not bathing, since the concept depends upon pulling heat from departing water in real time as new cold water replaces consumed hot water.

    Secondary considerations include effectiveness dependent upon cold incoming water - shower drain water likely runs 85-90*F, an excellent source of preheat energy for 40*F incoming water found in Canada, but well nigh useless for 70+*F incoming water found in the deep south. Virginia likely exhibits incoming water around 60*F, so your mileage will vary.

    Drainwater heat recovery systems require several feet of vertical drainwater travel, no problem for homes with basements but unattainable in cases of slab-on-grade or most raised floor crawl space homes.

    Despite my great enthusiasm for HPWH, it is worth noting that focusing on first hour rating conceals the Achilles heel of HPWH - extremely long full recovery times. Reheating a fully exhausted HPWH in heat pump mode can take 2-5 hours, depending on circumstances. Most HPWH heat the whole tank slowly, whereas conventional dual element storage electric resistance water heaters prioritize the upper element, making another shower's worth of hot water available 20-40 minutes following a complete exhaustion event.

    Finally, it is arguable that the longer shower teenage years should NOT be met with an endless hot water solution. That empowers ridiculously long showers and triggers moral hazard in the form of profligate consumption. It is perhaps better for teenagers to experience both an "end-of-shower-signal" in the form of tepid water from an exhausted tank, as well as the opprobrium from the next person needing a shower - both circumstances are opportunities to jerk teenagers from their self-centered reverie.

  14. Expert Member
    Dana Dorsett | | #14

    While I'd agree with Curt that drainwater heat recovery is nearly useless in FL, mid-winter in VA, it'll make a significant difference in showering performance in terms of what you can get out of a 50 gallon tank. The annual average incoming water temps will be in the mid-50s, using deep well water temps as the proxy:

    But mid winter incoming water temps will likely drop well into the 40s F. And since during mid-winter when the incoming water temps or low, the heat going into the heap pump water heater is a space heating load, and the net benefit of the drainwater heat recovery on total energy use is beyond what would be calculated by the simple testedEF of the water heater. In mid-summer when incoming temps are Curt's estimated 60F or higher the efficiency of the drainwater heat recovery is lower, but also needed less, and the heat pump water heater is reducing the cooling load. Drainwater heat recovery in VA would be a far greener approach than an in-line tankless, for comparable or lower money, and a much better financial ROI.

    FWIW: The measured entering drain water temp ahead of my DWHR is always over 95F, usually close to 100F, but it's pretty close to the shower drain. With a long drain run to the heat recovery unit it could drop to the low 90s I suppose. My incoming water temps at my MA location vary from the mid to high 30s in winter to the high 50s bumping on 60F in the summer. In VA your incoming water temps WILL be higher, but not 10s of degrees higher.

    The local grid in VA is currently pretty high-carb compared to some parts of the US, and even a standard ~0.55EF gas-fired tank is likely to have a lower annual carbon footprint than even a heat pump water heater on it's own, but if most of the hot water use is showering, heat pump water heaters + drainwater heat recovery would probably beat a condensing gas tankelss. But the VA grid is also lightening up it's carbon at a fairly quick pace with the displacement of 25-30% thermal efficiency coal by ~50-55% thermal efficiency combined cycle gas, along with the thin edge of the wedge of a rising renewables fraction. If you can make it work at your tub-fill volumes, heat pump water heaters are going to be a clear low-carb winner going forward. You'll never get the endless shower experience out of a heat pump water heater (even with drainwater heat recovery), but you pretty much could with a condensing gas-fired tank + DWHR. In my case the time-out on the bathroom light's occupancy sensor is the prompt for the teenager that it's time.

    The spa-filling capacity is still the biggest mechanical system design hurdle. Those loads are often served by big burner low mass finned water tube gas fired pool-heater type boilers when you have multiple spas to fill in a short period, but that's none too green and not a very cheap solution. If you can make it work with the larger heat pump water heaters it'll be a lot lower carbon and still cheaper to install than a mega-BTU boiler (that's only used intermittently.) An electric tankless would be a pathetic band-aid gaining only marginal capacity- it takes FOREVER to fill even half a spa with an 18kw tankless. Size the tank for the tub, and crank up the set point if you have to, increased standby losses be damned.

  15. akg4y23 | | #15

    I gotta say Im really happy I posted here, this is much more involved than I had thought initially.

    Two interesting perpectives. I guess on the one hand with 60 degree incoming water temp the HPWH will use even less energy and use the heat pump more than the electric elements, but if the goal is to increase the available hot water at peak usage then the DWHR system would be the perfect supplement. I definitely agree with Curt that providing unlimited hot water for the endless teenage showers will probably be counterproductive.

    Im still struggling with what would be the best method in my situation. In some respects a tankless NG heater still seems like a simpler solution although the intall cost is high and overall cost will not be recouped for 20 years and still less efficient than a HPWH let alone HPWH+DWHR.

    Thinking of the situation as three separate problems to approach:
    1. Increasing energy efficiency/reducing energy use:
    2. Providing a sustained stream of hot water for something like a Jacuzzi and the occasional extended shower.
    3. Instant hot water with reducing waste of water.

    So looking at these, Im pretty set on the HPWH because I feel like at least for the next 5-7 years with younger kids that will serve our needs and be efficient. The question is if it is worth it to install:
    a) A recirculating pump for $300-500 to provide instant hot water at the taps and save reportedly 5000-10000 gallons of wasted cold water per year (only costs about $10-20 in supply + sewage per year). Solves problem #3 but worsens #1. This seems worth it because the cost is the lowest and it provides two benefits of saving water and time. Cost of the slight increased demand on the HPWH will likely be fully offset by the water savings. The problem Im seeing is that these have valves that tend to fail often. Is there a good brand/model that is recommended?
    b) A drain water heat recovery system for $750-1000 to assist the HPWH in maintaining higher temperatures given the slow recovery rate. Improves problems #1 and #2. Although the benefit of this is probably less beneficial in VA, even if it raises the incoming temp 25-30 degrees that should be enough to significantly improve the recovery rate and overall peak demand supply from the HPWH.
    c) Supplemental tankless heater to the tubs and possibly master shower ($500). As long as the tankless heater is sufficiently powered it will help with #2 and #3 but be worse for #1. I still think this solution will work, even if it is a 'bandaid' because usage of the tubs is only occasional. If heating cold water the flow rate would not be enough from an 11 or 18kW heater but if the supply is the hot water from the HPWH even after peak demand temps will probably be on the order of 70-90 degrees significantly increasing the ability of the small electric to reach the desired temp at higher flow rates.

    Problem is all three of those solutions combined would cost $1000 for the HPWH, $1000 for the DWHR, $500 for the recirculating pump, and $500 for the small tankless. Total of $3000, some of which would be necessary even for just a standard water heater installation. Even if total energy and water savings were $400 yearly the payoff would be 6-7 years. Not too bad but at some point there are diminishing returns and increase complexity and future costs of repair etc.

    Im probably more confused now than before I started in terms of what would be the best thing to do. HPWH + DWHR seem like a good idea plus maybe the recirculating pump, but that still doesnt solve the problem we have now which is that the soaking tubs cant be used because the water heater cant fill them.

  16. akg4y23 | | #16

    In the process of discussing with my wife the possibility of removing the jacuzzi tubs from the two master bathrooms. Wish me luck.

  17. GBA Editor
    Martin Holladay | | #17

    Good luck. Load reduction is always the first step in any energy conservation plan.

  18. Expert Member
    Dana Dorsett | | #18

    Don't expect a 30F rise out of the drainwater heat recovery unless you get a truly tall one. Those rated in the ~50% range on the NRCAN apples-to-apples model comparison page will roughly split the temperature difference between what's coming out of the shower head (~105F), and the incoming water temp, so long as your flow rates are 2.5gpm or less. So when you have 55F incoming water (probably your annual average), at the output of the DWHR you'll get something like a 25F rise. With a higher-flow shower head you'll get less, with a lower-flow shower head you'll get a bit more, but it has to be in the 1gpm range to hit a 30F.

    With a 0.90EF electric tank and 30 minutes of showering time per day there's a reasonable net present value even with 7 cent electricity in cold-water country- see:

    With a heat pump water heater the financial model gets more complicated, but from an extended apparent-capacity point of view it can still pay off in family-harmony at any energy-price, eh? ;-) The recovery times are definitely shorter, and it takes a much longer shower to actually deplete the outflow at the top of the tank to below human body-temp.

    But it doesn't touch the tub-filling aspects even a tiny amount- only a tank big enough (and a setpoint high enough) can get you there, unless you have a HUGE amount of heater/burner behind it. At a reasonable tub fill rate of 8gpm (4000lbs/hr) it takes 200,000BTU/hr (59 kilowatts) of heat to deliver a 50F rise.

    A finned water-tube gas boiler with a 300-400K 80-85% efficiency burner like a Laars or RayPak and a ~50 gallon buffering heat exchanger like an Ergomax E44 can get you there without huge standby losses or short-cycling losses but it's not cheap to install. But that's really what's called for with multiple spa tubs. A standard gas-fired tankless won't cut it on spa-sized tubs unless you're very patient, and willing to put up with the quirks of gas-fired tankless water heaters under other types of use.

  19. curtkinder | | #19

    Wow - as I said, Dana routinely out-blogs me...few others do.

    Since you apparently DO have natural gas, its low cost per Btu (now and likely for the life of whatever water heater you choose) changes my thoughts a bit. Assuming you don't have a 500 gallon spa, there are higher efficiency gas fired storage tank heaters that might meet 99% of your needs with most of the efficiency but much less of the complexity of high Btuh tankless.

    AO Smith and Polaris spring to mind...Polaris claims up to 386 GPH at a 60*F rise at 95+% efficiency. I can't vouch for them.

  20. MJDesigns | | #20

    You're thinking this through way too much ... simplify your life and move on.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |