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
Musings of an Energy Nerd

All About Water Heaters

Buying a new water heater isn’t necessarily the best way to lower your energy expenses for domestic hot water

An example of a common type of point-of-use electric water heater used throughout the Caribbean and Latin America. This showerhead heater is manufactured by Dur-O-Matic. [Photo credit: Cassie and Britton Kauffman,]
Image Credit:

If you want to save energy, there are lots of exciting appliances and building materials that you might want to specify for your home: triple-glazed windows, an efficient refrigerator, and compact fluorescent or LED lighting, for example.

When it comes to choosing a water heater, though, clarity evaporates. Simple, affordable water heaters aren’t very efficient, and efficient equipment is complicated and costly. So how do you go about choosing a water heater?

Emergency decisions

Most homeowners ignore their water heaters. About once every ten years, the average homeowner goes down to the basement or out to the garage and notices that the water heater is sitting in a pool of water.

Since the family needs to replace the leaking water heater right now, the $350 special at Home Depot (or whatever model the nearest plumber is willing to install) looks good. Although a rush replacement job is understandable under the circumstances, it’s not the wisest way to choose a water heater.

Choosing a water heater

The overwhelming majority of water heaters sold in the U.S. are tank-type water heaters heated by natural gas, propane, or electric resistance elements. Tank-type water heaters are widely available and inexpensive. Of the three most common fuels, natural gas is by far the cheapest, except in a few areas of the country with very low electric rates.

If natural gas is unavailable, an electric-resistance water heater makes more sense than a propane water heater, since electric water heaters avoid potential problems with backdrafting and flue-gas spillage.

What if you want to heat your water more efficiently — or in a more environmentally friendly way? Well, there are many options, all of which require a much bigger investment in equipment.

Options include:

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. stuccofirst | | #1

    Which is better?
    Am I better off with my current indirect system with a modulating boiler and natural gas, or an el cheapo atmospherically vented one. During the summer, I don't use much hot water at all, so firing up the boiler to maintain the hot water seems like a waste. zone 5a

  2. GBA Editor
    Martin Holladay | | #2

    Response to Shane Claflin
    If you have functioning equipment in good condition, I don't recommend that you run out and buy another appliance.

    At least your fuel is inexpensive. Natural gas is very cheap, and the price is still dropping.

    Consider implementing some or all of my suggestions for reducing hot water use:
    make sure your pipe layout is as efficient as possible,
    reduce the diameter of your tubing if necessary,
    insulate your hot water pipes,
    consider installing a drainwater heat recovery device,
    make sure you have low-flow showerheads, faucets, and appliances,
    wash your clothes with cold water, and
    be conscious of conservation every time you turn on the tap.

  3. gusfhb | | #3

    I think the horrible efficiency numbers for indirects are possible, but not likely. The system in my house may have been one of them. A 1970 oil boiler, kept hot year round, fired 3 times an hour for half a minute or something. With your condensing boiler you are no doubt more efficient since it is cold start[only runs when called] and over 90 percent efficient when running. I don't think I have ever seen my new[condensing] boiler run without hot water being drawn.
    A regular atmospheric hot water heater is a pig. Not only does it struggle to be 70 percent efficient, but slowly dumps air past the water tank and out the chimney, cooling the tank, Conditioned air heated by your furnace. Sealed combustion may be no more efficient than atmospheric but at least it stops most of that.
    A condensing gas boiler with an indirect is about as good as it gets.

  4. GBA Editor
    Martin Holladay | | #4

    Response to Keith Gustafson
    The summer efficiency of an indirect water heater varies depending on your level of hot water usage. Buildings with high levels of hot water usage (multifamily buildings, hotels, laundromats) will see much higher efficiencies than homes with average or low levels of hot water usage.

    Marc Rosenbaum is an engineer. He measured the efficiency of a ten-year-old Buderus oil-fired boiler hooked up to an indirect tank in his Massachusetts home and concluded that the summer efficiency of the system (used only for domestic hot water, not space heating) was 10%.

  5. gusfhb | | #5

    Of course it is.

    Because as you later state, he uses 1100kw to provide hot water for his house a year. About 31 gallons of heating oil in that buderus. Less than a tenth of a gallon a day. While I would not expect a Buderus to be wired anything but cold start, if it was setup to keep the boiler hot 24/7 it would be even worse.. If he ran a gas tank water heater it would probably have been 5 percent efficient.

    All storage hot water heaters are terrible if you do not use hot water, which is why those point of use ones are a great idea.

  6. Expert Member
    ARMANDO COBO | | #6

    Thank you
    Great article and lots of good information. Thank you.

  7. GBA Editor
    Martin Holladay | | #7

    No, it was worse than that
    If Marc Rosenbaum was only using 31 gallons of heating oil a year to make domestic hot water, he might have been able to live with the situation. But the situation was much worse -- he was burning 0.6 gallon of oil per day for domestic hot water only. Annualized, that's 219 gallons of oil a year.

    Of course, if the Buderus is used for space heating during the winter, the efficiency situation improves. But over the summer -- let's call that 6 months -- the Buderus was burning 0.6 gallons a day, or 110 gallons for half the year.

  8. gusfhb | | #8

    my numbers
    Just as a frame of reference. I have kept track of my boiler run time over the past year. The control on the Buderus [GB125BE] records run hours and it tracks very accurately to gallons. Family of 4 with two small children, lots of laundry, dishwasher, etc, all new fixtures. We averaged .45 gallons of oil per day in the summer. Now there were sporadic heat calls till mid June[yeah I know if it was up to me there wouldn't be] and starting again in mid September. this means that at least 8 months of the year the jacket losses are in the plus column. As a fully condensing power vented units, stack losses are very low.
    So we are probably at this stage of our family in the average/heavy hot water users. I have no way of tracking water usage so I cannot truly get an efficiency number.

  9. GBA Editor
    Martin Holladay | | #9

    Response to Keith Gustafson
    Thanks for sharing your numbers. It sounds like your family is using more hot water and burning less oil than Marc Rosenbaum and Jill. That's good. There are many possible reasons for the apparent difference in efficiency -- something could have been wrong with Marc's Buderus, or his basement could have been colder than yours, or the accuracy of the monitoring method at one of these houses may have been somewhat imprecise.

    In any case, it's always good to have the numbers, so thanks for sharing yours.

    (By the way, Marc installed a water meter on his hot water line, so he knows how many gallons of hot water he's using.)

  10. gusfhb | | #10

    Boiler cannot be set up correctly. It is a bit apples to oranges, but I am willing to submit that I use 5 times more hot water than he does. My boiler is only 6 percent more efficient than his. For space reasons my tank is ~20 feet of pipe from my boiler. And I use 25 percent less oil to heat my hot water?

    We would have to check with him, but his boiler must be setup to run all summer. There is no reason a Buderus needs to stay hot. He probably shuts it off in the summer now, but he needs to get a new boiler guy.....

  11. GBA Editor
    Martin Holladay | | #11

    Response to Keith Gustafson
    Marc got rid of his boiler -- he says he was happy to see it go up his basement stairs.

    He installed a Marathon electric water heater and a rooftop PV array.

    He recently installed a Geyser heat-pump water heater, and hooked it up to his Marathon tank. He's still gathering data on the Geyser. He reports that the efficiency of the Geyser is quite low during the winter.

  12. gusfhb | | #12

    He sure does love buying stuff..............

    We all know that outside of the NE boilers are as common as hen's teeth, but the reason I had to comment is that I would hate to see someone install a cheap gas water heater right next to their condensing gas boiler. That would be silly. A badly setup indirect[and their are certainly plenty of those] is certainly bad, Marc's was historically bad, no offense to Marc.

    If you use very little hot water any tank is inefficient, but a gas atmospheric must be the worst.

  13. wjrobinson | | #13

    Two people , but active
    Two people , but active people so we use water like three or four. 150 gallons of oil for summer 6 months. 80+ efficient standard boiler with in boiler water coil kept at temperature of course. 150/180=.83 gallons/day. 45 degree well water.

  14. gusfhb | | #14

    Ahh, do miss the old tankless coil, nothing like moving the waterbed and filling it with hot water.....of course tankless coils should go the way of waterbeds..............

    sounds like you could payoff an electric tank for summertime use in about a season

    do miss the waterbed tho.......

  15. user-965582 | | #15

    Gravity-fed recirculation

    Is there a reason why you didn't consider gravity-fed recirculation as a cheaper option than a hot water circulation pump? I have a long run to my kitchen and am not sure whether a small Stiebel unit or gravity-fed recirculation would better. A Stiebel unit would be easier to install, but I am attracted to the passive design of gravity-fed recirculation, although it would involve a lot of copper pipe.

  16. GBA Editor
    Martin Holladay | | #16

    Response to William Rau
    William Rau,
    Although it's possible to install sloped piping that creates a thermosyphon loop between your water heater and a remote faucet, most energy experts recommend against that approach. The major problem with a 24-hour-a-day thermosyphon loop is that the water in the pipe loses heat (even if the loop is insulated), so your energy bill will rise.

  17. user-1030561 | | #17

    installation savings?
    Martin, you say that a mini-tankless would cost too much for using at each faucet, but wouldn't there be some slight plumbing savings by eliminating the hot water piping? Of course you'd still need to pipe a full bathroom for hot water to supply the tub/shower. But I've always been intrigued by the idea of using these units at powder rooms, utility sinks, etc.

    Those shower mounted units are at once the ugliest and most beautiful things I've seen in quite some time!

  18. GBA Editor
    Martin Holladay | | #18

    Response to Jack Barnes
    You're right: if you can install an instantaneous electric water heater in your bathroom that is sized to handle the bathroom's load, you would save the cost of running hot-water piping to that room.

  19. user-729621 | | #19

    This is such an important
    This is such an important point, such a good article, about how simple systems are so often the best. I changed my water heater and furnace to a gas-fired instantaneous (on-demand) heater, connected to my hydronic heating system through a heat exchanger. (not legal everywhere but it is here in Ontario). The result: when it is cold out and the heater is firing to keep the house warm, I can't get a decent shower. Then there is the issue of maintenance; you have to properly flush the thing out every year. Nobody told me and last summer I had to rebuild the thing for $ 700, far more than the price of a conventional water heater. I am an architect and I know my way around a plumbing system, but this all so complex and high tech that I am in another world.

    Also, any gas savings are lost by the fact that I cannot get my teenage daughter out of the shower.

  20. GBA Editor
    Martin Holladay | | #20

    Response to Lloyd Alter
    The more complicated and expensive the equipment, the bigger the maintenance headaches. I don't think there is any exception to that rule.

  21. Expert Member
    KOHTA UENO | | #21

    Nice work!
    Martin--very nice column. A few points to add.

    I don't know if you're aware of Tom Butcher's work at Brookhaven, but he did a great study looking at efficiencies of various systems out there. He wrote a recent ASHRAE Journal article on his work, but it is summarized pretty well in a 2007 report he did:

    Performance of Integrated Hydronic Heating Systems (BNL-79814-2008-IR)

    The list of tested systems is on page 5, and the results summary on page 7. The standard off the shelf atmospheric tank wasn't that bad for summertime efficiency (58%), while the atmospheric boilers with either tankless coils or sidearm tanks were pretty miserable (~40%).

    This lines up with work I've been doing in my "copious free time"--logging the atmospheric boiler and sidearm tank (HT Products Superstor) in my apartment basement. 2 adults, normal working schedule, and we're burning ~20 therms/month in the summertime. The problem is that the boiler only runs maybe once or twice a day for hot water calls--and there's no purge control, so there's 12 gallons of water and a few hundred pounds of cast iron that cools from 180 F to room temperature every cycle. A quick calc suggests that this is ~8 therms/month of consumption. Why yes, I'm planning on building a roll-your-own purge control one of these days.

    Another great resource is ACEEE's white paper on hot water technologies--they point out that site energy-to-taps efficiency in multifamily buildings is so lousy (1/3 of site energy to taps in a bad system, and 1/2 in a good one) that point of use electric isn't a half bad idea. It's a big 94 page report, but the relevant section is pages 85 to 90.

    Emerging Hot Water Technologies and Practices for Energy Efficiency as of 2011

    A few projects that I have been involved with (new construction) used top of the line condensing tankless water heaters, and I'm pretty impressed at the end-of-the-day consumption--families of 2 or 4 burning ~3-5 therms/month. Yes, these are often houses owned by energy weirdos, um, I mean consumption-conscious occupants. But still--that's$4-9/month on natural gas.

  22. GBA Editor
    Martin Holladay | | #22

    Response to Kohta Ueno
    Thanks a lot for the links to the research papers.

    I'm glad to hear more voices agreeing that point-of-use electric heaters, in many situations, require less source energy (fossil fuel at the electric generating plant) than efficient oil-burning or gas-burning equipment located in a homeowner's basement.

  23. DrDanger | | #23

    electric showerheads
    Your article brings back memories of staying in very cheap "hotels" that I stayed while backpacking in Peru, Educador, Bolivia. You could see the electric wire glowing in the showerhead, and hear the sizzling as the water passed over it. The scary part was turning over the switch , which was right next to the shower with dripping wet hands...scarey! On the other hand I never did run into other travelers having bad experiences or close calls. I bet the stuff was wired to at least a 240V line. It did however provide hot water, instantly and at a descent flow. Some of these were augmented by black painted 55 gallon drums sitting atop their roofs.

  24. Billy | | #24

    Nice Article; Recirc Controller

    That's a nicely written article. I'm glad you pointed out how inefficient recirc pumps can be. It's amazing how may people don't consider this at all and waste huge amounts of energy.

    I found a really nice little recirc pump controller that the engineers put a lot of thought into when they designed it. It is called the Redytemp and you can see it here:

    They have other models as well.

    What I like about this unit is that you can control the recirc pump with three different controls, all at the same time. First it has a timer option which by itself is nothing special. Second, it has a switch input so you can run a wire pair to anywhere and activate the pump. Third, and most important, it has a thermostat control You install a temperature sensor in your water line, say at the end of your run, and adjust a thermostat so the recirc pump runs until the end of the line reaches a present temperature that you can control, and then it shuts off the recirc pump. This is great because for hand washing a temperature of 100 degrees may be plenty and you don't even try to keep the water line heated to the higher temperature of the of the hot water heater. Make sure the hot water line is insulated from the hot water heater to just past the temperature sensor location.

    Then you use the timer so the recirc pump doesn't run at all during times when you're asleep or away. Even when you are home the recirc pump runs for only 20 or 30 seconds until the temperature sensor reaches a sufficient temperature.

    For those who really want a recirc pump this controller makes a lot more economic sense. Smart controls can save a lot of energy over dumb controls and on/off switches.


  25. user-984364 | | #25

    Low-flow sinks
    One thing I noticed after installing low-flow aerators (1.0/1.5gpm) on my sinks is that it's almost impossible to get hot water to the tap in a hand-washing minute. The lower the flow, of course, the longer it takes to get there.

    I'm aware of this, yet I still habitually turn on the hot tap. Some psychological block or something...

  26. user-1087436 | | #26

    What an interesting discussion, and an amazing conundrum, all caused by our desire to be virtuous as we pursue the luxury of hot water. Personally, I started washing my hands with cold water about fifteen years ago, not out of virtue but because the hot water I wanted never arrived anyway! At David Butler's website,, he advocates using a small 120-v. point-of-use electric tank ("minimal heat loss," he claims), plumbed in line with the shower. By the time the POU tank runs out of water for a shower, the water from the main tank has arrived. Whatever you do, don't install the gas-fired flash water heaters we used to have in Turkey! Installed in bathrooms, they were responsible for many deaths by asphyxiation: they simply used up all the oxygen in the room. The typical solution: a series of 2-inch holes bored in the bottom of the door.

  27. Marc Rosenbaum | | #27

    In order of appearance in the article and comments:

    - I reported that I have been disappointed with my add-on heat pump water heater, the Geyser by Nyle, in that it appears to me that the efficiency does not match the manufacturer's claims. It's difficult to measure efficiency due to the uncertainty of the size of the standby losses that the heater must make up in addition to the energy needed to actually heat water used. The lower the water usage, the lower the system efficiency of the heat pump water heater. This is nicely documented in the Steven Winter Associates ACI paper on HPWHs. I am not in any way condemning HPWHs and in fact my company is installing a couple of integrated HPWHs with some metering to further understand their performance. I haven't claimed that they are noisy, expensive, or mechanically complicated, just that my own unit has not performed as I had expected. Please draw no further conclusions :-)

    - I datalogged run time on my Buderus G115 oil boiler while it was only making DHW for about four weeks. It had a 0.5 gallon per hour (gph) nozzle and rated oil pump pressure of 145 psi which yields an oil flow rate in my understanding of about 0.6 gph (nozzles are rated at 100 psi pressure). The burner ran for just under one hour per day, in roughly twenty minute time periods, so I concluded that the boiler was using just under 0.6 gallon of fuel daily. It is possible that this is not an accurate number though no one has given me a good reason to suspect it is wildly off. At the time I hadn't done any plumbing mods so didn't have a water meter on the inlet of the water heater tank. I have had a water meter in place since some time in June 2011. From July 2011 through January 2012 (seven months) we've used 2,820 gallons of hot water, an average of slightly over 13.1 gpd. Heating that amount of water from 50F to 120F is a load of about 7,650 BTU/day. 0.6 gpd of fuel oil has an input value of about 82,800 BTU, so my overall system efficiency appears to be just under 10%. This Buderus was set up and controlled as a cold start boiler and therefore did not stay at boiler temperature except when it was firing to make DHW. I have made no claim about whether it was operating optimally yet I saw no signs during the months it was operating before we took it out that something was grossly out of whack.

    - Since I noticed this boiler/indirect tank issue, I have looked for it when our potential solar clients bring us their energy bills and I've seen a pattern of surprisingly high summer oil or propane usage with boilers heating indirect DHW. Everett Barber, in his excellent book Convert Your Home To Solar Energy, notes that these systems have much lower efficiencies making DHW in the warm season. I imagine that the state-of-the-art condensing low mass boilers are much better than the boiler I had and many other people have. If Mr. Gustafson's household is using five times as much DHW as mine and uses 0.45 gpd of oil to heat it then that system efficiency is about 60%, way better than mine. It would be great if more folks measured the DHW usage as well as the energy input, we'd have more real data to work with.

    . Using the HPWH for almost all of this time period has resulted in an electrical usage of about 520 kWh. I've been conservative in my energy budget estimate of 1,000 kWh to heat water for one year in my quest to take the house to zero annual net energy (much more info on my blog Thriving On Low Carbon). If this was all done with resistance heat I expect it would be in the range of 1,200 - 1,400 kWh/year, so the HPWH is definitely better than straight resistance heating even at our low DHW usage.

    - The electric water heater I installed is an 85 gallon Marathon because I wanted the option to use either a HPWH or solar water heating, so I knew I'd want extra gallons of storage. If my plan was to use only electric resistance I would have used the smaller Marathon (50 gallons). The two homes in South Mountain's Eliakim's Way project that achieved net zero averaged 20.6 gpd per household and in the 50 gallon Marathon averaged 0.233 kW/gallon of DHW compared to our 0.184 kWh/gallon with (mostly) the HPWH.

    - Mr. Gustafson, I hadn't quite viewed it this way, but I guess I do love to buy stuff. I learn best by doing, so I seem to have to try things out for myself, and I like to measure performance of buildings and systems. To this end, I consider myself fortunate to own a blower door, a small gaggle of Hobo dataloggers, four digital watt meters, a Fluke clamp-on multimeter, and a two channel digital thermocouple. I've been even more fortunate to have been a member of a collaborative community of practitioners, mostly through the Northeast Sustainable Energy Association, that for over thirty years has shared our discoveries, successes, and failures in a non-judgmental, mutually supportive way. Perhaps someday the online community will adopt that level of civility and caring.

  28. Larry Weingarten | | #28

    Martin, you covered a lot of ground and mentioned points that usually get left out. There are still a few things I'd add. First is simple solar. A black 55 gallon drum in the roof works in some climates, but there is work being done on freeze resistant simple solar that will cost one to two thousand dollars and can do a meaningful fraction of the DHW. We might rethink the idea of doing 75% or more with solar as it usually makes the equipment complex and delicate. There will be more on this at the upcoming Hot Water Forum in Berkeley CA.

    I'll throw in mention of maintenance on tank type heaters as a cost effective thing to do, particularly with the advent of relatively expensive FVIR units. If maintenance can help heaters to last fifty years rather than nine to twelve year average, there is a lot of potential for savings. Of course, let's only maintain well insulated tanks!

    Lastly, codes get in the way of efficient distribution. Depending on water pressure, gpm needed and length of run, 3/8" or even 1/4" tube can provide adequate water much faster with much less loss. When will plumbing codes catch up to the present?

    Thanks for a very nice article!

    Yours, Larry

  29. wjrobinson | | #29

    Keith, what exact heater do
    Keith, what exact heater do you have? Condensing, oil fired?

  30. GBA Editor
    Martin Holladay | | #30

    Response to Larry Weingarten
    Thanks very much for your comments. The black 55-gallon drum is, indeed, a time-tested solar water heater in hot climates -- but not one that is likely to gain favor for most U.S. homeowners.

    DOE researchers have been hoping to discover or develop "freeze-resistant simple solar that will cost one to two thousand dollars" for years. They've imagined plastic solar collectors or some not-yet-discovered breakthrough. I'm a skeptic. Solar hot water systems have been around for a hundred years. We understand the technology, which is simple. I don't see any reason why the technology will suddenly get cheaper.

    Many solar hot water proponents promote existing systems (available technology) by underestimating what the systems cost to install. Homeowners who call up a contractor to get a bid usually discover that their costs will be significantly higher than the estimate they read about in the DOE brochure.

    Finally, thanks for your reminder about the importance of water heater maintenance. GBA readers should definitely check out Larry Weingarten's website, Water Heater Rescue. The site contains many extremely useful tips on ways to extend the life of your water heater.

  31. GBA Editor
    Martin Holladay | | #31

    Response to Marc Rosenbaum
    Thanks for your detailed comments, which are much appreciated. And thanks also for your data monitoring activities, and your willingness to share your data and experience with the larger geek community at events such as the recent Better Buildings by Design conference in Burlington, Vermont.

    The listed disadvantages of heat-pump water heaters -- including my characterization of the devices as "noisy, mechanically complicated ... and expensive" -- were, of course, my own (as were all of the disadvantages listed for other types of water heaters). I never intended my reference to your Geyser monitoring data to imply that you shared my own views on these devices.

    However, just because a device has disadvantages, doesn't mean it shouldn't be specified. Even if a HPWH is noisy, it might still be appropriate for a basement, although not necessarily for a mechanical closet near a bedroom. Even if it is expensive, it might still be appropriate if its advantages justify the investment.

    That's why, in the final paragraph of my article, I noted that a heat-pump water heater might be a good choice, especially for homeowners in hot climates.

    Again, Marc, thanks for sharing your knowledge. I look forward to further reports on your work to achieve net zero energy.

  32. gusfhb | | #32

    re: Marc
    Thanks for replying

    I do not doubt your methodology, but three comments

    you did not close the loop on fuel usage, and over a small time period it could be significant, but not out of control

    You say your system ran three times a day to maintain water temp. While as I mentioned I have only had such systems with a much higher usage than yours, the system would almost never run when there was not a call for water. It has always made sense in relation to amtrol's specs of around .4 deg heatloss per hour. If one were not home it ought to run maybe once a day. Which brings me to my third point.

    That aint how we measure efficiency, is it? What I mean is that to be fair you need to run the same test with a regular tank hot water heater. You know 'everyone' is going to say 'well my gas tank hwh is 67 percent efficient, that is much better than the indirect' when that is not what you have proven.

    In the end with your hot water usage any storage hot water unit is going to look bad, especially in the summer when the various heat losses are of no help to you.

    Now I have to dig through that Brookhaven study because that looks to be pretty interesting.

    RE: AJ My boiler is a buderus gb125be, the middle sized one, cannot remember the suffix.

  33. gusfhb | | #33

    Brookhaven study
    Had a chance to to read through the study linked above in Kohta Ueno's post, and it is very cool. If you have hydronic heat it is a must read.

  34. vi8ZocrBu9 | | #34

    point of use electric for shower
    Hi, I wish for that too, but unfortunately it would require too much electricity:
    7.6(L/min) * 4180(J/L/degC) * 30(degC) / 60(s/min) = 15884 Watts
    or about 66A at 240V. Okay that's possible with some really big wires...
    I assumed raising the water temperature by 30 degrees, but I'm not sure how hot the water actually needs to be. You could try numbers you like better, but I think you still would need several tens of amps.

    I wonder if, combined with a very efficient drain water heat recovery device, you could make the electricity use reasonable?

  35. Larry Weingarten | | #35

    electric showers, etc
    Hello: In the UK they have electric showers that are far more powerful than the Caribbean shower heaters, running at 240 VAC and drawing from 7.5 to 10.5 Kw. These are designed to heat from cold to a useable temperature. Water pressure can be low as well, so they also have "power showers" which use a pump to move water through faster and give a stronger shower. The switch for these things in in the shower ceiling so one is less likely to get electrocuted.

    Responding to Richard Patterman, it used to be that when electric heaters were new in the world, tempering tanks were suggested as a pre-heater. Electric heaters had 800 or 1000 watt elements and needed something to help them perform. Any warm place in or around a house is potentially useful for getting water closer to a useable temperature. The lines between solar and pre-heating begin to blur.

    Yours, Larry

  36. user-980774 | | #36

    Photo of "electric
    Photo of "electric showerhead" reminded me of a trip to Guatemala many years ago when my wife took cold showers for a week because the "sizzle and glow" scared the hell out of her.

    Great article, good job of summarizing the options and I appreciate that you acknowledge that
    with low hot water usage the cheapest, simplest system (electric tank) might be the best decision.

    I'm facing a decision on a small cottege with no natural gas between small electric tank or electric tankless. I know the tank is the simple, cheap solution but my lifestyle points me toward tankless.
    I'm gone alot and shower at the gym a couple times a week so my usage is extremely low.
    The design has the bath and kitchen sink within 5' of the dhw heater so distribution losses will be minimal.

    Another thing rarely discussed about DHW is a preheat system or buffer before the DHW heater to raise the temperature of the incoming water and decrease the energy load. Here in Colorado the incoming water is 40 to 45 degrees. If that could be passively raised to house temperature (70deg) the heating load would be decreased by more than a 1/3. This could be accomplished by putting a pressurized tank or pipe loop in the warmest room of the house (utility room or sunroom).

  37. GBA Editor
    Martin Holladay | | #37

    Electric showers from the U.K.
    Thanks, Larry. If anyone is interested in more information on "electric showers" from the U.K., here is a relevant web page:

    Using an uninsulated metal water tank to pre-heat cold water before the water is heated by an electric element will work, of course. But in winter, such tanks will rob space heat from the house, and in summer such tanks will sweat.

  38. Marc Rosenbaum | | #38

    Mr. Gustafson

    My oil system ran three times daily presumably because the losses off of the Aero indirect tank were such that they exceeded 0.4 F per hour. This is not uncommon because the standby losses aren't only off of the tank, but from the piping connected to the tank as well.

    The efficiency I reported is the overall system efficiency, not the boiler efficiency. Since I'm concerned ultimately with actual fuel usage, this is what I set out to measure. The claim I'm making is that the system efficiency is about 10% with my low usage.

    It was not my intent to run a study of other systems (though my measurements on our 8 Eliakim's Way homes, and my own, furthers that aim) so I had no need to be "fair". We need better data on all of these systems in actual usage, to determine system efficiency. Indeed, please put a water meter on your inlet line and tell us what your system efficiency is, with your system, at your usage.

  39. gusfhb | | #39

    Mr Rosenbaum

    My sole real critique is that in rating your system at 10 percent efficient you have failed to give any comparison to any other system. If you had an electric water heater what would its efficiency have been?

    If you had been gone for a week, what would your system efficiency be?


    No matter what method of heating stored hot water you used

    Not really a useful number

    If you have read the link Brookhaven report, they did a pretty interesting comparison. While they do not list any brand names, it is pretty apparent that they used the non condensing version of my boiler for one of their test subjects[well insulated cast iron]. If I were to measure my water usage an it came out statistically different, who would you trust?

    Another interesting piece of data is that while many of the systems suffered in summertime efficiency, when looking at it as a full year they are not so bad, and in fact superior in some instances.

    I keep saying the same thing in different ways, that there is no storage hot water system that would have performed well in your test setup, thus you have not told us anything in relation to the efficiency of indirect hot water heaters, but only of storage hot water heaters for households with very low hot water usage.

    The brookhaven study tells us that a boiler with a well setup control strategy can give superior performance to a separate gas hot water heater for 'average' hot water usage. It does not tell us what would be the case at your usage level, although using your data point one could make an educated guess.

    People will walk away after reading your numbers thinking, 'indirects stink'. Well, no that is not really so. Storage hot water heaters stink.

    Your comments do not make that distinction, and I think it is an important one.

    I am sorry if you think my comments are rude or somehow personally biased against you. If I have come off that way, I apologize.

  40. MICHAEL CHANDLER | | #40

    $2,100 condensing tanks & E Star problem w split H2O heaters
    Just back from a week at the International Builders Show and catching up on e-mail etc but wanted to point out a few thoughts.

    First A.O. Smith has a very good condensing tank style water heater (Vertex) in NG for $2,067 or LP for $2,118. These provide 96% eff - 100 kBtu/h direct vent sealed combustion utilizing 2" PVC flue. So the days of paying $4,000 to $6,000 for a condensing gas water heater are over. A.O. Smith also offers the NEXT Hybrid for $2,000 which marries a 100 kBtu/h 90% eff tankless water heater with a tank to eliminate the cold water sandwich and pressure drop issues common with tankless water heaters which typically restrict flow by 7 psi at 5 GPM and 20 psi at 10 GPM.

    A. O. Smith also offers a split heat pump water heater with an outdoor mounted heat pump that heats an indoor tank at 35 to 171 kBtu/h for water heating (and 27 –132 kBtu/h cooling capacity.) at COP ranging between 3.9 and 4.2. I've not been able to get pricing on this model

    On Ted Clifton's suggestion we used a $4,600 UniChiller split heat pump water heater on a recent combined heat and hot water project with grid tie 8KW PV array. It provides up to 135 degree F water at 36 kBtu/h with COP ranging from 2.3 at 10 degrees F to 3.4 at 55 degrees F. We have an eMonitor on it so we'll be watching to see how it measures up to it's claimed efficiency.

    These units can be set with an out-door temp lock-out to shut them down and switch to back-up when the out door temp falls below 10 degrees F and can be switched to provide cooling to fan-coils in the summer if you aren't also using them for domestic hot water or have solar thermal HW for summer use.

    Rheem offers a split heat pump water heater sized for domestic hot water in Australia but has not brought it to the US due the Energy Star exclusion that Martin mentioned w/ indirect heaters and desuperheaters with separate storage tanks and heating elements. Until Energy Star changes this exclusion of two-part systems it is unlikely that there will be a split heat pump water heater offered in the US market to avoid heating water by cooling the indoor air provided by the heating system.

    Air Tap does offer a ducting kit for their retro-fit system to allow it to draw and exhaust outside air and many of the tank-top water heaters seem likely to be able to be modified to run with a concentric vent hood as would be used with a closet mounted PTHP hotel room heat pump. Some of the tank top systems circulate water through the upper unit and would be at risk of freezing damage if installed with outdoor air circulating through the heat exchanger. Seems like a product category waiting to happen if Energy Star doesn't change their policy soon.

    I understand that Gary Klein is working on code revisions to deal with the smaller pipe sizes used in manifold and "branch and twig" hot water distribution schemes he's been working on to address the modern low-flow WaterSense fixtures. He is also working with an organization called "Green Plumbers" to train American plumbing companies on the new piping strategies.

    That's all for now

  41. drzentner | | #41

    Any comments on using smaller diameter PEX to replace 3/4" copper hot water pipes? Are the heat losses less with PEX. I have a 40' run from the garage to the bathrooms.

  42. Marc Rosenbaum | | #42

    Do storage water heaters stink?
    The 8 home data set I mentioned above had two homes that averaged 20.6 gpd of DHW and made that with 0.233 kWh/gallon in a 50 gallon Marathon electric water heater. At a temperature rise of 70F, that's over 70% system efficiency. So that's a fairly low DHW usage rate with a decent system efficiency.

  43. D876j6g9Y7 | | #43

    Great article
    Thanks, Martin, for another great article. Your last invocation for thoughts about POU HW brings to mind an obvious idea: why not have a POU unit for each faucet and showerhead? Each is programmed/controlled a bit differently. A pushbutton for "Handwash" and another for "Very Hot" and so on.

    Regarding the power needed to heat POU water at reasonable flow and temperature rise, the key to the solution is the energy needed for the job. A hot shower, even for teenagers, does not continue indefinitely (as far as I know). If the energy needed for can be effectively stored ahead of time, then an appropriate power supply can deliver it at a higher power for a shorter time. (Think of a UPS.) (Think of a storage water heater used for the second teenager's shower.)

    This fits into my armchair-engineer's interest in handling high electrical loads. I have long dreampt about geeky ideas like battery-operated microwave ovens that can be bought and used in old kitchens like mine that have inadequate wiring for modern small appliances. The daily energy use of a microwave is rather small, yet the thing blows breakers if started when the coffee's brewing. The technology is swiftly dropping in cost, and is actually already well-handled by off-grid and yacht power supplies.

    Thanks again Martin, GBA, and commenters.

  44. GBA Editor
    Martin Holladay | | #44

    Response to Michael Chandler
    Many thanks for your detailed post, with lots of useful information. Thanks for reminding me about the Vertex, which I reviewed in the August 2006 issue of Energy Design Update. I have corrected the pricing information on condensing water heaters in light of your point about the cost of the Vertex.

  45. GBA Editor
    Martin Holladay | | #45

    Response to David Zentner
    If you have 3/4-inch copper tubing to a remote bathroom, you probably have a truck-and-branch system. If you switch to a PEX manifold system, you should be able to use 1/2-inch (or even 3/8-inch) tubing for your bathroom lavatory, which will reduce the amount of water in the line (and therefore reduce hot-water waste).

    You will probably be able to reduce the line to the shower to a 1/2-inch line, depending on your water pressure, although if you enjoy baths rather than showers, a 1/2-inch line will increase the time it takes to fill the tub compared to a 3/4-inch line.

    You're right that copper is more conductive than PEX, but the difference in conductivity isn't the most important factor. When it comes to heat loss from the tubing, the most important factor is whether the line is insulated -- so insulate your pipes.

  46. GBA Editor
    Martin Holladay | | #46

    Response to Steven Hunyady
    It's possible, of course, to have a separate point-of-use electric water heater for every single faucet, shower head, and appliance in your house. The main downside to this approach is the cost of all of those water heaters.

    I'm skeptical that your idea of operating high-amp electric resistance water heaters from a battery. The battery in my off-grid house weighs 1,800 pounds and cost thousands of dollars, and I still don't like to run our microwave oven from the battery system at night, because the microwave oven is such a large draw that it pulls down the house voltage. (The rule in our house: you only get to use the microwave oven when the sun is shining or the Honda generator is running.)

  47. MICHAEL CHANDLER | | #47

    David's 40' run
    David, assuming you have water sense listed 1.5 gpm faucet aerators and 1.75 gpm shower head in that remote bath you are taking 2.7 times as long to draw hot water from that 3/4" line than you would through a 1/2" line and leaving about 16 cups in the line to cool as opposed to 6 cups if you switched to 1/2".

    Elbows in the line also add to the wait and waste by mixing the water so the hot water doesn't just push the cold water ahead of itself (plug flow) but eddies at the outflows of the elbows (laminar flow) this makes the water warm up gradually as it arrives at the faucet, delaying the time it takes to get to a usable hot temperature. One more reason to use PEX as opposed to copper or CPVC.

    I've attached a sketch of a piping system I used to explain this to a plumber working on a LEED-h / NGBS gold house we designed in another city.

    The idea is to place a manifold (here it's five 3/4 x 3/4 x 1/2 PEX tees grouped two inches apart) as close to the water heater as possible with the 3/4 out at the end going directly to the garden tub with no side branches and each 1/2" branch feeding a separate fixture group such as a single bath room or kitchen.

    in this house it meant that the 3/4" to the big tub is in the same joist bay as the 1/2" to the kitchen and the 1/2" to the master bath shower and vanity group. The plumber took some convincing but this is what it took to keep each fixture within six cups of net displacement from the water heater. 1/2" PEX is so flexible that there generally is no need for elbows between the manifold and the first tee in the fixture group, where you have more than one tee it is good practice to group them close together if possible with the end outlet going to the highest flow fixture. I buy 100' rolls of red PEX so I can cut the exact length I need and not have any couplings in the lines.

    In this situation we also had to feed the island sink in the kitchen with 3/8" from the wall sink and dishwasher to meet the 6-cup goal. The cold continues to run in conventional 3/4" trunk and 1/2" branch as there is no worry about waiting for cold water. I've found that this system is a good value and easy sell to the inspectors.

    I wouldn't use a single fixture home run manifold system unless every home run was 3/8" except the garden tub. You can go 93' with 3/8" for the same six cups but using 1/2" individual fixture manifold piping means that you could waste six cups at the shower and then another six at your vanity and another six at your partners vanity . in actual practice 1/2" single fixture manifolds don't make much sense compared to 1/2" fixture group piping.

  48. Thyo2pdy6V | | #48

    installed a remote switch on
    installed a remote switch on our hot water recirc pump when upgrading the boiler to a condensing one this year and got the family trained to use it . love it . used to be two minutes to have hot water in the master bath. now instant on as long as you plan ahead two minutes. hot water savings and a lot of water i am thinking over the course of a year.

  49. CKQVPn5YhE | | #49

    One cold water supply line + pobuy showerhead heaters on Amazon
    I remember 30 years ago, in the UK, when people used those old fashioned tankless gas heaters; they were in the bathroom! With another often inside the kitchen wall mounted above the kitchen sink. Many houses had just one cold water supply line around the house which fed the point of use heaters. Why double the use of water pipe?

    In the UK you can also buy electric tankless water heaters that go under the sink. Most houses are wall mounted hot water radiator heated and have hot water through the high efficiency gas boiler so the point of demand heaters haven't really taken off.

  50. user-1004247 | | #50

    Manifold distribution system
    Great article. After studying all the options for several years I had come to the conclusion that for a small house and small family that only washes clothes on cold water--only a simple electric water heater made sense. However, in my last house we put in a Manibloc system and used 3/8" PEX insulated lines for all but the bathtubs which had direct 1/2" insulated PEX lines. The shortened time to get hot water is amazing and I will not put anyting else in a new home now.

  51. user-928968 | | #51

    Timer switches on electric water heater
    Are there savings to be had by not keeping an electric water heater on all the time? For example, if all showers & dishwashing happens in the evenings, manually turn the water heater on only from 6pm to 9pm. Or install a timer switch to accomplish the same thing.

  52. wjrobinson | | #52

    Bob, I always wanted to try
    Bob, I always wanted to try 3/8" instead of 1/2". No loss of flow? Good to hear. Will try such soon.

  53. user-1092886 | | #53

    POU Electric WH - Take a look at this.
    An interesting article and I couldn't agree more. I think this unit is pretty much what you are advocating regarding electric tankless and POU units. It is a commercially available unit, called an "Electronic Tankless Water Heater" by Norton Research, (409)783-9931, They have several models of varying output, but they are all the same price - $350.00. Take a look. What do you think?

  54. GBA Editor
    Martin Holladay | | #54

    Response to Barrie Scott
    Yes, I know that a lot of manufacturers make electric on-demand water heaters, in almost any size and capacity you want. (Many need high-amp service and therefore oversized electrical service panels, of course.)

    The main issue is not the availability of the equipment -- it's the relatively high cost of multiple units and the need for heavy-gauge wiring.

  55. watercop | | #55

    ASHP desuperheaters and more
    I perform deep energy retrofits in north Florida and I routinely install heat recovery desuperheaters on air source heat pumps. This performs the same function as the factory option desuperheater on watersource heat pumps (which I also install). These provide free hot water all summer and cheap hot water in (what passes for) winter down here.

    I commissioned one on August 20, 2011. I remember it well because it was part of a deep energy retrofit preceded by gathering a month or so worth of data using a 4 channel TED system. So I knew going in how much the client was spending on HVAC, clothes drying, and hot water.

    That system has 2@80 gallon tanks, side by side. One stores water from the desuper recovery box and it feeds the other, conventionally wired. On August 20, the conventional storage heater operated for about two hours at 4500 Watts, warming itself at a cost of a little over a dollar.

    At the same time, the HVAC / desuper combo went to work on the upstream tank. The main tank did not operate again until September 3, the first day with an outdoor high temp below 85. In other words, following initial warmup, the desuper provided ALL hot water for a family of five adults for two weeks straight. On Sept 3, the upstream buffer tank dropped below 100*F for the first time since startup and the main tank element ran for 15 minutes or so.

    My own system also has two 80s. The first is heated by the desuper option of my WaterFurnace 3 ton watersource heat pump. Its temperature ranges from 70 to 130, depending on HVAC intensity. The downstream tank was resistively heated from May 2008 until July 2009. During that time I logged hot water energy use via an hourmeter on the lower element (upper element disabled)

    In July 2009 I added an early model Geyser HPWH. I monitor its power use with a P3 Kill-a-Watt. The Geyser uses right at 50% of the power that the resistance element used to use, in other words, an effective COP around 2.0. It is noisy, unsuitable for use other than a basement or garage, but it usefully cools and dehumidifies my basement mechanical room. The cooling is anywhere from 0-3 degrees, and the dehu anywhere from 0-10%

    For 2-3 person families, our technology of choice is the GE 50 gallon HPWH. I like its design over the competing Rheem unit since it avoids the separate heat exchanger and pump. I recently read of a test by a western utility which deliberately installed a GE HPWH in a room 1/4 the recommended volume, essentially setting up a stress test. The GE passed with flying colors, losing only 0.1-0.2 EF despite the tiny room.

    For 4+ people, the add on desuper starts to make sense.

    In some new construction projects we configure a geo or air source desuper feeding an 80 gallon storage tank that then feeds 2-3 tankless electric water heaters located close to kitchen and baths, minimizing both energy consumption and hot water wait times

    My one regret is failing to specify a PEX manifold system for my own home. I like Martin's sketch, though if anything, I don't think it goes far enough...why can't modern lav sinks be supplied by 1/4" PEX? The fixtures are down around 1 GPM.

    The flow restrictor in a typical 2.5 GPM showerhead is 1/8", so it seems reasonable that even a shower could be fed with its own 1/4" line.

    Six cups? how about just 1-3?

  56. user-1091869 | | #56

    Responses to several posts
    A passive pre-heat tank for a PoU heater does not _use_ heat from the house - the "heat" that the tank collects is retained in the house and acts as a thermal mass.

    Gas tanks - as an alternative to atmospherically-vented tanks and power-vented tanks, there are direct-vented gas tanks that use no power - eg
    They are no more efficient than a regular gas tank, but they solve the problem of unbalanced air supply / back-drafting in an air-tight home. Relatively affordable at about $700 + installation. Also available for propane - and I think I've seen oil?

  57. badgerboilerMN | | #57

    indirect water heaters
    Dear Martin;

    Please don't tell the hundreds of customers I have here in Minneapolis that enjoy the benefits of 95% condensing gas boilers, coupled with indirect-fired water heaters, that indirects are not efficient. You will spoil the charade! They may even think that the 50% cut in summer time fuel usage is just a fluke!

    And please don't tell them about the hard-wired shower head water'll put me out of business!

  58. vZNXFmZfGh | | #58

    Why is a Water Heater burner worse than a gas stove burner?
    I live in an old house with the water heater installed in an unfinished portion of the basement. You mention the hazards of a back draft situation if a strong exhaust fan is running. Why is this more hazardous than having a gas stove or oven on in the kitchen?

  59. d296FrvJPL | | #59

    Domestic Hot Water Pipe Sizing
    An arbitrary amount of water in a plumbing line doesn't mean a whole lot. (6 cups?) Line sizes are based on the amount of friction loss that they incur based on the demand of the fixture. So if you arbitrarily size a line 3/8" to a remote shower on the second floor of a building you may not have any pressure once the water arrives.

    You have a finite amount of pressure on the supply. The farthest fixture has to have enough pressure to operate. This is usually a minimum of 15psi and can be much more for pressure assisted water closets. So the line sizes are based on the flow (gpm) at a given pressure. The gpm is determined by the number and type of fixtures. (Fixtures have a prescribed value known as fixture units.) If you have multiple stories you also have to account for static head loss as well.

    Listing the amount of time it takes water at a given pressure to reach a fixture is putting the cart before the horse. Making sure the water arrives at the fixture with the required pressure is the key to the design. Undersizing the lines to either save money or increase velocity doesn't make good sense and violates the code.

    The plumbing code limits the velocity of the flow in domestic systems to around 6 feet per second. Going above 8 fps creates a noisy system and adds additional wear to the piping. Many designers limit hot water flow to around 4 fps. The code also has supply line minimums for piping that is enclosed (i.e. in walls or floors or ceilings), which is 1/2" min. to showers, lavs and water closets. 3/8" is allowed only from the angle stop to the fixture. Showers must be 1/2".

    Higher velocity can increase water hammer frequency and magnitude. If you've ever noticed the sound of water flowing in the walls of someone's house or a sharp thud when you turn off a faucet it is probably due to improperly sized piping. Elbows and tees will literally blow out in systems that are undersized, especially PVC. Pex may fare better.

    The best (only) way to deal with slow hot water is to either recirc it, which I don't believe is feasible in most homes or install a point of use water heater at the remote fixture. The water heater should be in close proximity to the master bath. If it's at the other end of the house it's a poor design and should be retrofitted. Velocity of flow should always be minimized not increased. Fixture units, available pressure and length of run should dictate pipe sizes.

  60. GBA Editor
    Martin Holladay | | #60

    Response to Vincent Alvarez
    Q. "Why is a water heater burner worse than a gas stove burner?"

    A. In many cases, it isn't. But sometimes it is.

    Perfectly adjusted gas burners don't emit carbon monoxide (although they do emit large amounts of water vapor and small amounts of combustion byproducts that you don't really want to breathe). It's not a great idea to leave a gas oven on all day in a tight house; that's one reason that building codes require the installation of a range-hood exhaust fan.

    Poorly adjusted gas water heater burners sometimes emit carbon monoxide. When that type of burner backdrafts, the results can be deadly.

  61. GBA Editor
    Martin Holladay | | #61

    Response to Peter James
    You wrote, “A passive pre-heat tank for a point-of-use heater does not use heat from the house -- the ‘heat’ that the tank collects is retained in the house and acts as a thermal mass.”

    I'm afraid I disagree. Right now, the temperature of my incoming cold water is 40°F, while my house is at 70°F. If the cold water goes straight to an instantaneous water heater that raises the temperature to 120°F, then the instantaneous water heater needs to provide an 80 F° temperature rise.

    On the other hand, if I have an uninsulated metal tank in my living room that brings the cold water to room temperature before sending the water to the instantaneous heater, then the instantaneous heater needs to provide only a 50 F° temperature rise.

    In the second case, the instantaneous water heater used less energy. So where did the energy come from? It came from my home's space heating system, of course. Because it took a lot of heat to raise the temperature of the cold water in the uninsulated tank to 70°F, my space heating system had to work a little harder.

    It's not clear whether I saved any energy with this method of tempering the incoming cold water. Perhaps I did save some energy -- but only if my space heating fuel is cheaper than the fuel used for my instantaneous heater, or if the appliance that supplies space heat is more efficient.

    You wrote that “the ‘heat’ that the tank collects is retained in the house and acts as a thermal mass.” That would only be true if the tank were sealed, with no flow through the tank. In fact the heat collected by the tank is used to make hot water for my shower, and most of that heat goes down the drain and leaves the house.

  62. GBA Editor
    Martin Holladay | | #62

    About the direct-vent GSW water heater
    Thanks for providing the link to the GSW direct-vent water heater that operates without electricity. I wasn't aware of this appliance -- it sounds like a sensible solution for many homes.

    I have edited my article to include a reference to the GSW appliance.

  63. GBA Editor
    Martin Holladay | | #63

    Response to Patty McDaniel
    Q. "Are there savings to be had by not keeping an electric water heater on all the time? For example, if all showers & dishwashing happens in the evenings, manually turn the water heater on only from 6 pm to 9 pm. Or install a timer switch to accomplish the same thing."

    A. Yes, there are potential savings -- as long as you adjust the timer so that the water heater has time to recover before you need the hot water.

    Before you go out and buy a timer, make sure that your water heater is well insulated. (That means you need to buy a water heater with thick insulation -- ideally foam insulation -- or you need to retrofit a water heater blanket around the tank.) Then feel free to experiment with the timer. If your water heater is off, it isn't using any energy.

  64. Darin Zurliene | | #64

    Hot Water Heater
    We have had luck with Marathon Electric hot water heaters which we hook up to geothermal with a hot water assist. The water heater is very efficient and offers a generous no leak warranty to homeowners. When the geo system is in cooling stage we can turn of the breaker to the hot water heater where it is solely heated by the condensing heat from the geo. As for the lines definitely insulate them correctly so they maintain their desired temp for as long as possible. For adequate volume a balanced loop can help with that low pressure at the end user.

  65. Lizzieplants | | #65

    Just the Data Please
    Being the the process of planning a new home I eagerly read all the articles and posts about hot water heating and came away with some new information but also a number of questions. So I went web hunting.

    Question one was should we put in solar thermal since we are doing a slab with radiant heat. That reference to the Steve Winter report was eye opening on the real payback on solar thermal. Well I found the report and also another on the NYSERDA (NYS Energy Research) site (see attached) that basically told me the annual savings would be only around $100 per year. Even at the net costs I heard at the Home Show this weekend of as low as $2,500 that is a 25 year payback.

    I also found a table from the American Council for an Energy Efficiency Economy that I think will help a lot of people make the water heating decision (see attached screen clip) that shows a heat pump system is the way to go.

    My second question was about putting in PV to help pay for the electricity the hot water system and other uses. For New Yorkers, NYSERDA, has a great calculator which showed the payback would be, you guessed it, 25 years. They are talking about increasing the PV incentives in NY so maybe the time line will shorten.

    I'm 60 years old and a payback of 25 years means I will be 85 before this pays off. Maybe great for the young family but too long for me. Bottom line - those solar thermal and PV dollars will be going to better insulation and better windows.

  66. wolfworks | | #66

    Using an existing well
    There has not been any mention of an idea that was recently proposed for a project we are working on (working on meaning trying to wrestle this whole DHW issue to "that makes sense" status).
    The idea is to use the existing well as an open-loop for a ground source heat pump solely for DHW with the pump and backup(tankless electric?) load addressed with PV. I'd love to hear from anyone who has practical experience with this application.

    By the way, all these discussions remind me of that classic sign seen in a computer repair shop:

    "We have not succeeded in answering all of your problems. The answers we have found only serve to raise a whole set of new questions. In some ways we feel we are as confused as ever, but we believe we are confused on a higher level and about more important things."

    Thanks for bringing this essential domain to the table Martin.

  67. toddoskin | | #67

    Electric tankless for 'whole house'?
    Any info, or user experience from the Stiebel Eltron Tempra or Tempra Plus?
    (Tankless Electric Whole House Water Heater)

    Would this make sense instead of an electric tank DHW heater? (in all electric house/apartment)

    Takes up less space too... i wonder what the operating costs are ...and longevity of the unit.. also the initial cost..

    I'm building a 3 unit super insulated building, and was thinking electric tank HWH would be easiest/cheapest overall. (for all electric building).


  68. GBA Editor
    Martin Holladay | | #68

    Response to Todd Oskin
    Yes, you can use an instantaneous electric water heater for your entire house. Be sure to size the unit properly, however, and be sure that your electrical service panel can handle the enormous load.

    If you anticipate three simultaneous uses of hot water -- perhaps the washing machine and two simultaneous showers -- you might want 6 gallons per minute of flow. If your incoming water is cold, you might need a 70° or 80° temperature rise. You can't do that with the Stiebel Eltron unit you linked to.

    The Tempra 36 -- the most powerful model -- provides only 3.5 gallons per minute at a 70° temperature rise. That might be enough for many families. The unit requires 220 volt power and draws 36,000 watts (three 60 amp circuits). Yowser!

    Holy Frankenstein's monster, Batman -- why are the lightbulbs dimming?

    The advantage of a tank-style heater, of course, is that the appliance doesn't draw such a tremendous electrical load, and you can often take advantage of off-peak rates.

  69. GBA Editor
    Martin Holladay | | #69

    Response to Jamie Wolf
    You wrote, "The idea is to use the existing well as an open-loop for a ground source heat pump solely for DHW with the pump and backup (tankless electric?) load addressed with PV."

    Of course it's possible to do what you suggest, but the equipment costs are high -- you need a water-to-water heat pump, an instantaneous electric water heater, and a PV array. If your goal is to simplify your equipment or lower your capital costs, your proposed solution doesn't do it.

    Whenever you put a PV array on the roof of a grid-connected house, you're feeding electricity into the grid when the sun is shining -- but you can't tell the electrons where to go. It's not as if any particular load in the house is "addressed with PV."

  70. 7MZdikxaiR | | #70

    Zero Energy Water Heaters
    ZeroEnergy produces free hot water from normally wasted heat produced by air conditioning.
    This is an alternative that should be reviewed.

  71. GBA Editor
    Martin Holladay | | #71

    Response to Karl Moser
    ZeroEnergy is simply a brand name for a desuperheater. If you read my article, you'll see that I mentioned desuperheaters.

    Curt Kinder's long comment (above) gives more information on desuperheaters.

  72. Mike Eliason | | #72

    @elizabeth kormos didn't you
    @elizabeth kormos didn't you claim in the window performance posting to be planning a passivhaus? but now planning slab w/ radiant? what the what?!?

  73. kevin_in_denver | | #73

    A comprehensive water heater discussion should at least mention this bacteria that thrives in potable water at 105F and can kill you if you inhale it in your shower.

    The quick lesson is, don't set your water heater to below 120 or 140F (depending on what school of thought you follow). Also don't ever use the "vacation" setting. Completely off for a long vacation is probably OK.

    Heating professionals except for Radiantec warn against using your potable water in a radiant or baseboard type distribution system. So if your water heater is also your heating boiler, they say to isolate the heating system with a heat exchanger.

    Don't ever risk your family's health, but I haven't heard of a fatality from Legionella in the residential sector due to Radiantec style system design. Code allows it in many jurisdictions, but not in others.

  74. Brian Knight | | #74

    Electric tankless, plumbing manifolds
    Perhaps the biggest disadvantage of electric tankless for whole house use is the strain they put on the electrical grid. Utilities hate them. Ive heard that they can cause lights to dim to on neighboring homes (probably with older infrastructure). They often use their extreme power draws during "peak demand hours" which drives the need to build more power plants and causes excessive strain on an already stressed power grid. Of course there are exceptions but for an average american home they are usually a bad choice. POU for a remote fixture is a different story.

    An electric tank on time of use metering set up with a timer (and smart occupant behavior) is a good way of making this simple technology more affordable in terms of electric use and easier on the grid and thus environment.

    The article states "in most cases, home-run manifold systems make more sense". This is probably true regarding lag time but not necessarily payback or cost effectiveness. I dont think this applies as much to a design with a smart plumbing footprint. The last time I priced a manifold system, which the plumber objected to based on past experiences, it was pretty high apparently due to the extra materials and labor (dont forget all that pipe insulation). I think a trunk and branch system can be designed to be almost as effective at much reduced costs. I really like Michael's drawing and comments and think that a hybrid system is the way to go if the home design has a less than desirable plumbing layout.

    This is a great article and discussion on the 2nd biggest energy user in american homes.

  75. Mikey | | #75

    Solar performance; on-demand hot water
    A couple of comments from a DIYer in central Florida:

    First, my simple solar system (open-loop, single 10x4 panel, 80 gallon tank) produces 160°+ water year-round. It was installed several years ago in November; I inadvertantly left the power off the WH after installation and didn't notice the water getting cooler until a cloudy spell in mid-January. One advantage of the (affordable) high-temperature storage is (as mentioned by Kevin Dickson earlier) that it's in the disinfection range for Legionella, whereas the "green" recommended 120° setting is in the growth range.

    Second, point-of-use circulation pumps eliminate water waste and provide more-or-less instant hot water. If designed in at initial construction, it's pretty straightforward. As a retrofit, you can use the cold water line as a return, at the cost of (temporary) hot water coming from both sides. I used an existing trunk-and-branch for both hot and cold.

    All this is very satisfying, although in my case DWH is apparently a small fraction of my overall electricity consumption; I saw no significant difference in consumption attributable to the solar system, even during the period when the WH was powered off.

  76. GBA Editor
    Martin Holladay | | #76

    Response to Michael Armstrong
    You Floridians are lucky -- you have warm, sunny weather year round, and you don't have to worry about the water in your solar collectors freezing. That certainly makes your solar thermal systems simpler, less expensive, and better-performing than solar thermal systems in New England.

  77. johnmwalker | | #77

    Line sizing
    Based on valid comments by Les Dell and others, is there a straightforward method to size a home run manifold PEX system to account for friction/static head loss, fluid velocity limits (and any outdated Code requirements) if you know the incoming pressure to the system, line length and flow rate of a given fixture? Are manifolds designed to connect to and supply differing line sizes based on individual demand? To minimize head loss on the cold water side, should the largest line possible always be used or is an imbalance created? Thanks in advance.

  78. GBA Editor
    Martin Holladay | | #78

    Response to John Walker
    Here is a good JLC article that explains how to size water pipes to meet code requirements: Sizing Domestic Water Pipes.

    Assuming that there are no liability or code violation concerns, some plumbers and homeowners have experimented with PEX tubing sizes that are one size smaller than recommended by standard sizing methods. With less water sitting in the pipes, less hot water is wasted.

  79. bzockoff | | #79

    Heat pump water heater efficiency
    Just 1 year ago I installed a 50 gallon GE heatpump water heater. I've been monitoring my energy usage for many years and have the data to compare. We are a 2 person household that is "efficient" with water and energy and have public water that is 55F in winter in Massachusetts. We use about 40 gallons per day for hot water.

    Prior to the install of the new HPWH we used 2600KWH /year for an electric 10 year old 80 gal tank. In the past 12 months we have used 935KWH/year. Now I did leave the old 80 gallon tank in place as a tempering tank in the basement (ambient at 60F most of the time) which helps limit the time the HPWH runs. I can say that he HPWH runs longer since it has a lower recovery rate, but at a fraction of the electric usage (i.e. about 10%). I think of it like Prius, slower than my BMW but gets there just the same.

    I find that it is less noisy than the oil furnace and it does recover the heat loss from the furnace to produce the DHW. The aditional benefit is that it does dehumidify the basement in summer when I would likely run the seperate dehunidifier anyway. It's a win-win for me. I'm very happy with it as long as there are no early life failures.

    Martin, I think that you can tell from all the comments that this was a great article that stimulates many of us to reply. Thanks.

  80. GBA Editor
    Martin Holladay | | #80

    Response to Robert Zockoff
    Thanks very much for sharing your monitoring data. I assume that you installed a separate electric meter for your hot water circuit?

    Although I've taken a "wait-and-see" attitude toward heat-pump water heaters, I'm hearing more and more stories like yours. If these units prove to be durable -- without any expensive maintenance -- I think that an increasing number of homeowners will install them.

  81. bzockoff | | #81

    Multiple heat pumps
    BTW, I use multiple heatpumps in my own home. I have the HPWH, a mini-split air source heat pump for a seperate family room, and a central 5T air source heat pump on my oil hot air furnace. They all work well and have greatly reduced my operating costs, by approximately 40% in total.

  82. B_Carr | | #82

    Heat pump water heaters and hot/dry climates
    After reading the insight in this article and comments I'm thinking more an more that a HPWH is possibly the best hot water solution for hot/dry climates. Here in Phoenix, many people have their water heaters in the unconditioned garage that can easily see 120 degree temps during summer. Park your car in there after driving 20 miles home tfrom work during the hottest part of the day and you got even more efficiency from a HPWH. One thing I may try when my conventional electric water heater craps out is to go the HPWH route, but duct the inlet air from the attic to the unit. Unconditioned attics can easily reach 140 degrees or more on summer days and pulling from that air has got to make the efficiency of a HPWH soar.

  83. yivfw2nHxK | | #83

    Point of Use
    Question on the point of use applications; is there any reason that you could not have a two stage system for colder climates so as to preheat the ground water and then the second stage prior to use? Funky, but good application for vacation cabin or such.

  84. GBA Editor
    Martin Holladay | | #84

    Response to Scott Strodel
    No reason at all.

    Pre-heating methods include an uninsulated steel tank inside the thermal envelope of the home, or a solar collector on the roof.

  85. user-755514 | | #85

    Response to Scott Strodel
    There may be one reason one should not preheat the ground water. The electric point of use water heater units I installed under each of my kitchen and bath sinks came with a Warning: Do not preheat the incoming water. To prevent the generation of steam, the units include an over temperature limit which trips off the electric supply. In the winter when the incoming ground water from our water utility company is below the expected 55 degrees F ground temperature, the units usually do not trip; however, in the summer when incoming water is above 55 degrees F, we have these units tripping more frequently.

  86. [email protected] | | #86

    Hot Water -
    If you are considering a water heater be aware of its limitations, especially if you are thinking about an instantenous one, keep in mind that these heaters have very limited flow rates, best way to proceed, do a fixture count, then armed with that information look at the specs of whatever heater you are considering. Many of the manufacturers offer sizing software on their websites like Rinnai. This is a good way to ensure you have ample of hot water. If you are working on a commercial or industrial project call a mechanical engineer.

  87. tommyreno | | #87

    Heat Pump HW in Seattle -- and Does a Circulating System Save $?
    Thanks for article and all the great information and the comments. I also read your Apr 2012 post on heat pump water heaters.

    I live in Seattle, WA and need to replace my electric hot water heater in the coming months.

    I have two questions:

    1. What's your opinion about a heat pump water heater in a moderate northwest climate like Seattle? You recommend a heat pump water heater if you live in a hot climate. I would consider Seattle neither a cold nor hot climate, but a moderate one. I'm contemplating installing one in my unconditioned garage, which adjoins the basement of my house within the building envelope. I was all set to replace my standard electric heater in kind until I saw info and rebates on heat pump water heaters. A northwestern energy non-profit and the local public utility combined to offer a $1250 rebate on an Air Generate 66 gallon heater, which is a great incentive. If they’re not great products for the maritime northwest, why are utilities here promoting them? (BTW, I briefly considered solar and electric tankless as alternatives, but both seem to have more flaws than the heat pump for my situation. I do not have access to gas unless I pay $30K to bring it from a nearby street.)

    2. Would a hot water recirculation system make a standard electric water heater more efficient and reduce electricty use? When we remodeled our house, our plumber built in this capability, but we did not install the pump. When we install our new water heater, we could also install the pump if it would make sense to.

    Thanks for your advice.

  88. GBA Editor
    Martin Holladay | | #88

    Response to Thomas Iurino
    Q. "What's your opinion about a heat pump water heater in a moderate northwest climate like Seattle? ... I'm contemplating installing one in my unconditioned garage."

    A. If you live in a climate where your garage stays warm enough in the winter to prevent your pipes from freezing -- and you apparently do -- then installing a heat-pump water heater in your garage makes a lot of sense. Go for it.

    Q. "Would a hot water recirculation system make a standard electric water heater more efficient and reduce electricity use?"

    A. Such a system will definitely NOT make a water heater more efficient. In most cases, it will increase electricity use. However, a hot water recirculation system controlled by a switch in the remote bathroom (in other words, one that is usually off) may reduce wait times for hot water and may save water.

  89. Eric Johnson | | #89

    Do Not Use Marey Tankless Heaters
    We specified and had a couple of Marey tankless heaters put in a home. Do NOT I repeat not use them. Try calling their toll free number any time. You'll always get voicemail, and your call will never be returned. We ended up ripping out those units and throwing them back onto our distributor's doorstep. Here's what we went with instead. The homeowner is experiencing excellent results.

  90. ianfin | | #90

    Important opportunity to change the Federal testing standards
    Martin and all on this list.
    US DOE is currently asking for public comments on their proposal to update the testing methods for residential and small commercial water heaters of all kinds. If you are not already engaged in this national standard setting process, then I hope you can find the time in the next month to share your wisdom on this topic with DOE.
    The link to the Proposed rule is here:!documentDetail;D=EERE_FRDOC_0001-0672

    Titled (in case a search is needed to pull it up):
    Energy Conservation Program for Consumer Products and Certain Commercial and Industrial Equipment: Test Procedures for Residential and Commercial Water Heaters

    The deadline for comments is : January 21, 2014

    Here is a quote from DOE summarizing the proposed rule:
    "DOE proposes to modify the current test procedures for residential water heaters and certain commercial water heaters. The proposed amendments would modify the test procedure to be more representative of conditions encountered in the field (including modifications to both the test conditions and the draw patterns) and expand the scope of the test procedure to apply to certain commercial water heaters and certain residential water heaters that are currently not covered by the test procedure."

    Many thanks for your work in this field.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |