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
Q&A Spotlight

Using Sand to Store Solar Energy

Assessing the controversial claim that solar thermal heat gathered in summer can be stored in sand for winter use

Pumping solar-heated fluid through tubing embedded in sand. GBA experts ponder whether the heat gained from all this solar thermal equipment is worth the investment.
Image Credit: Fine Homebuilding

John Klingel’s question was simple enough: what’s the best way of heating up a thick bed of sand beneath a concrete slab with PEX tubing? But the underlying issue — whether a sand bed is a good idea in the first place — quickly takes center stage in this Q&A post at GreenBuildingAdvisor.

Klingel plans to include a 2-ft. thick bed of sand between his concrete slab and a layer of rigid foam insulation. The sand is a heat sink, but Klingel isn’t sure where the PEX tubing should be located for the best result. Nor is he sure what diameter the tubing should be, or what the spacing of tubing in the sand will work best.

Some writers think a sand bed is a waste of time. Others report they’ve had good luck with them, even in extreme climates. That discussion, similar to an exchange on the Q&A forum last year, is the subject of this week’s Q&A Spotlight.

Forget the idea — it won’t work

Count GBA senior editor Martin Holladay among those who think that an insulated sand bed doesn’t add much to solar design. “Here’s my opinion — subject to revision when someone gives me good monitoring data to contradict my statement: you can put the PEX wherever you want, because these systems don’t really work,” Holladay tells Klingel.

To get a useful amount of heat from the sand during the coldest months of the year, he says, it must be hot enough to get water in a hydronic heat distribution system to at least 100°F. And that, he adds, just isn’t going to happen.

“The sand doesn’t get that hot — or if it does, it doesn’t stay…

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. user-659915 | | #1

    Seasonal home?
    "I walk away from my house at 40° below and come back in the spring and don't worry about it" says Chlupp. I'm not sure I understand what this means. If he's saying this is a seasonally occupied building in a very cold climate then I totally get the value of the passive thermal mass to help keep the house from freezing up. Enough mass and Delta-T to keep a home above 32° is one thing - to maintain it year-round in a steady-state condition at occupancy comfort levels is quite another.

  2. user-723121 | | #2

    We use a lot of energy
    Hello to Mark Sevier!

    The thermal mass discussion is important because it shows just how much energy we use and how much stored energy is needed to get through a day, week, month or year. According to Larry Kinney, 1 million Btu's (10 therms of natural gas) is equivalent to the work output of a human for 1 year, think about that. A very well planned and insulated building shell is paramount, solar gain should be high on the list. Thermal mass to stabilize daily temperature swings make sense to me, beyond that I would question the cost effectiveness. Seasonal thermal storage has been attempted, the results and information on the costs are hard to come by.

  3. wjrobinson | | #3

    Mark, welcome. Would be
    Mark, welcome. Would be great if we could hear more about your home and net zero generally.

  4. delfstrom | | #4

    Modeling results of sand-bed solar thermal storage
    In the January-February 2011 issue of Solar Today magazine, David Sets, James T. McLeskey Jr. and Marshall Sweet report on the modeling and optimization of this system using TRNSYS. In their model, the storage was decoupled from the home (insulated on all sides). They will continue to monitor the actual project with thermocouples in the sand bed. Properly sized for a home, the modeling predicts 70% to 80% of the heat load can be provided by this.

    Unfortunately the heating load of the home isn't described in the article, though a chart shows monthly demand peaking at 1800 kJ/m2/day (about 0.5 kWh/m2/day) during the coldest month.

    I have collected quite a bit of information about these systems and I keep returning to the solution of super insulation first. However I think it might be quite good for something like a warehouse or big box store. Not necessarily economical though!

  5. jklingel | | #5

    James Morgan: What Thorsten
    James Morgan: What Thorsten meant by that is that he left his house sitting, unoccupied, for about 2 or 3 weeks in the dead of winter, and it stayed warm from the daily sunshine. I don't think he ever conceived of his system storing summer heat for use in the winter. Storing for use over a couple of days is doable, though. In mid-Feb, as he reported above, he shut off his masonry heater and has not lit (lighted?) a fire since. It was still going down into the minus 20's at night in mid Feb, too. The amount of wood he runs in the masonry heater is pretty small, too. When I asked him how much per day, he said "One fire with this much (a bear hug) wood." A man told me that in a presentation Thorsten gave at a home show (2-3 weeks ago?) he referred to needing another water tank to dump excess heat into already. We are still freezing solid at night, have a few feet of snow left, and there's not really a lot of melting on the back roads; so, it is yet cool. What will be interesting to see is whether or not he can manage the heat in the summer. I am hoping that he needs to hire me to plant some large shade trees. I think he can work the bugs out on this deal, though. In the meantime, I like the idea of the PV system, if it is not too spendy. That would sure eliminate the potential heat control issues.

  6. jklingel | | #6

    lighter is cheaper?
    On a related note, I will have to politely question the science of this statement. ""Bringing tons of sand up to temperature, and maintaining the sand at an elevated temperature, takes heat." What bothers me is the "maintaining it" part. My gut feeling says "Roger that", but then I don't see mass in any heat loss equations. Given a set amount of insulation, IS IT really cheaper to keep (not get) a lighter house warm than a heavier one? Is the issue that you'll have more radiant loss in the heavier house? Martin, if you answered this before and I forgot the answer, well, it won't be the first time I've forgotten.... Thanks to anyone who can explain that one to me.

  7. Coyo | | #7

    No tress needed
    What did you get me into here?

    You will not have to plant trees for me to avoid overheating in my house that was accounted for in the design. It is correct that the solar thermal system is currently melting out my yard through my heat dump – the 400ft ground loop I use to preheat ventilation air (Solar panels sitting idle = lousy investment). Initial I concluded my screw up on the system sizing which was based on information from European systems but by now, I see it as a great opportunity instead. I will pursue some new ideas on storage that this extra heat will be used for. As far as heating demand goes I am still relaying solely on passive solar gain since 02/16…

  8. Coyo | | #8

    The Bigger Picture
    Ok, I did not want to get further in to this but feel obligated to respond now. My apologies up front for the long rant, unfortunately not a subject wrapped up in two sentences.
    I am neither a sales person nor do I care about story telling. I however care about building science and renewable energy systems. We all have our believes and ideas and a discussion on heat storage as it had evolved from John K. original Q+A question is baseless without supporting data. After a completely fruitless exchange with Martin I regretted getting into it then and even more so now.
    The SunRise home is a prototype building (which did by far not have an unlimited budget and which at the end of the day is still my home and not my expensive lab) with a multitude of unconventional ideas. The diffusion open wall assembly, the foundation with its insulated internal mass and the solar thermal heating system are independently monitored by the Cold Climate Housing Research Center via many sensors, data loggers and BTU meters. Once all the tinkering and fine tuning is done the system will be life streaming online for everyone interested to follow its performance. By next year I will have a real opinion on how everything performs and we will have datasets to support discussion like this one. With lessons learned so far I already have redesigned many things and pursue new ideas which I am testing out this year. I never claimed to have the answer figured out and we made a lot of mistakes over the years and learned very valuable lessons – and I am looking for better solutions every day. In North America seasonal heat storage is considered an impractical solution which does not work – but there are many very successful projects throughout Europe. The tune at building conferences I attend in Europe every year is very different then what I experience here in the States.

    Since GBA continuously reports on PV systems as the best option with Martin frequently stating that everything else does not make sense (and my favorite: PV versus insulation) I also like to point out a few things to consider for everyone who is interested. If GBA wants to consider itself a site to educate and promote green building practices their contributions should IMO not only look at energy without looking at the bigger pictures surrounding it. Carbon emissions and embodied energy is something everyone always seems to conveniently forget about.

    “What really works is a grid-tied photovoltaic system”
    Technically speaking I very much agree that it is a lot easier to utilize grid tied PV systems to meet a home’s power and heating requirements. By their nature PV installed systems are much simpler and easier to incorporate into a home. Gordon Howell and Peter Amerongen from Canada’s EQuilibrium Building Program concluded after the first project that solar thermal systems make no sense for zero energy buildings and since then use exclusively PV systems on their projects. Most of the recent US based zero energy homes follow the same trend. I am very aware of this and certainly do not disagree with the general consent about this. However unfortunately in my opinion this is not quite as clear cut as everyone always makes it to be. If we are only concerning ourselves with a simple question of energy and economics this all holds fairly true. The classical definition for a ZEB – zero energy building is however is:

    “ZERO net energy consumption and zero carbon emissions annually”.

    Besides energy we therefor also need to consider carbon emissions. Now how do we account for this on a grid tied PV system? In some cases the grid tied electricity is coming from a fairly clean source – but most places surely are not there yet.

    Where does our electricity come from?

    Three major fuel groups are used by electric power plants throughout the United States to generate electricity: coal, at over 48 percent, or more than 2 billion watt-hours; natural gas and propane, at over 21 percent, or nearly 900 million watt-hours; and nuclear energy, at 19 percent, or more than 800 million watt-hours. Renewable sources of energy generate a total of about 291 million watt-hours, or 7 percent of total electricity, and biomass generated 64 million watt-hours, or 1.5 percent, to the net electricity generation in the United States. The total carbon dioxide emissions by electric power exceeded 2.5 billion metric tons in 2007 and the largest fuel groups—coal, natural gas, and propane—dominated carbon dioxide emissions with, over 94 percent of the total. Coal alone represents 78 percent of all CO2 emissions by electric power plants, or nearly 2 billion metric tons.

    Our grid tied electricity in perspective to CO2:

    In its lifetime in the atmosphere each molecule of CO2 traps 100,000 times more heat then was released during the combustion at the moment of its formation. For this reason in reality the use of a 800W hairdryer powered by average grid tied electricity contributes to planetary heating as much as two 747 airplanes at take-off at full power.

    We need to be careful on how we account for all of this and in most areas we really need to come up with better ideas. Especially in colder regions most electricity from these systems will be produced in the summer with not much production in the winter – when the usage is in contraire much higher. Home power loads are seriously out of sync with supply. The peak loads are at night and in mid-winter. At noon in the summer the needed on power is the lowest. For this reason it cannot work on a bigger scale as there is no storage means of PV grid tied electricity. A single gallon of gasoline contains as much energy as one ton of lead-acid storage batteries. Grid-connected solar systems can only work if you have a backup of fossil fuel energy plants. We might feel good about producing green electricity but will be using the same dirty power in the winter as everyone else.

    Besides that a lot of the northern regions simply have no grid. Energy prices in remotes parts of Alaska are over $8 for heating oil, which is also used to generate power in generators. Finding alternative means to power these remote communities is a matter of their basic survival as it is already today. We are not talking about some doomsday tomorrow…that is a reality many communities in the Far North are faced with for some time already. Solar PV will not be the answer to their questions. There is no feasible storage. Solar thermal energy can be stored, figuring better and cheaper ways to do so is the challenge. Working on this for now over two years I am more convinced than ever that there is tremendous potential and feasible solutions.

    We should ask ourselves carefully what our goals and motivation are in Zero Energy Buildings.

    The baseline should always be first and foremost efficiency and energy reduction within the building before we look at any form of renewable energy. Keeping what we have is the key and we should not just look into heat loss in a conventional manner but also through our wastewater. Second in a heating dominated climate, we should try to offset heating and production of hot water which have a big combined impact on the overall energy consumption with the most efficient renewable energy source. Then we can look into offsetting our remaining electrical power. Our we can simply just don’t care about any of these issues and simply turn up the heat – or provide enough PV to make it all work out on paper.

    That all said I leave it to everyone else to decide what makes sense and what doesn’t, to me there is no easy answer and it is about choices we make. I believe that we should not forget to look at the whole picture and not just pick out parts which we like best because they are easy to achieve. With the problems I see every day around me I am gladly following my lonesome path to try to figure out functional seasonal storage systems integrated in buildings. And Martin and everyone else is more than welcome to keep on telling me that it doesn’t make sense and that it cannot work. I still see the light at the end of the tunnel…

    Finally yet importantly, let us also not forget that there is so much more which needs to go into a truly functional and sustainable building. We need to design and build “Living Buildings” which will address more aspects then simply looking at primary energy. These living buildings will create micro ecological and biological environments that collect and store their own energy, harvest and recycle their own water, provide natural clean air and function as an extension of our own human bodies and allow for a connection between the inside and the outside.

    That at least is my vision. Thorsten Chlupp, no PE or PHD.

  9. GBA Editor
    Martin Holladay | | #9

    Response to Thorsten
    Thanks for taking the time to write. I love experimental buildings, and your house is an exciting one. I'm glad you built it.

    You asked about the embodied energy of photovoltaic modules. Here's the scoop, courtesy of information published in a Home Power magazine article by Justine Sanchez. Sanchez cites a 2006 study conducted by CrystalClear that calculated the estimated energy payback time (EPBT) for grid-connected roof-mounted PV systems. According to Sanchez, the study looked at the EPBT for balance of system (BOS) components (racks, inverters, wires, etc.) and assumed a system efficiency of 75%. “The study shows the EPBT for standard, single-crystalline module PV systems to be two years,” Sanchez writes. Systems using polycrystalline modules (produced with a casting method) had an even shorter EPBT at 1.7 years; modules produced with a ribbon method had an EPBT of 1.5 years. The study used average solar data for southern Europe, estimated at 1,700 kWh/m2 [160 kWh/ft2] per year; average solar production in the U.S. is higher (1,800 kWh/m2 [170 kWh/ft2] per year), meaning the EPBT for U.S. installations is shorter.

    Needless to say, solar thermal collectors and their associated hardware (pumps and tubing) also have embodied energy costs.

    Concerning PV, you wrote, "Especially in colder regions most electricity from these systems will be produced in the summer with not much production in the winter." Of course, solar thermal systems have exactly the same problem; they just lack a storage solution as convenient as the grid.

    Since I've lived off-grid for 36 years, I've pondered these questions a lot. The off-grid solution always comes back to biomass, in my opinion -- what we used to call firewood. Obviously, you use firewood for your masonry heater. And you've also found that once the sun returns in mid-February, it's much easier to keep a house warm than it is from Nov. 1 to Feb. 14. My own experience mirrors yours.

    Keep up the experiments and good work.

  10. GBA Editor
    Martin Holladay | | #10

    Response to John Klingel
    When I was an editor at EDU, I reported on a sand-bed house in southern Vermont. The builder made a house with a conventional basement. The basement was very well insulated with foam and then completely filled with sand. The sand had several PEX loops, at different levels, to add heat to the sand and to (in theory) pull heat from the sand. What happened?

    1. The homeowner lost the basement. Instead of having useful space, she had a basement full of sand.

    2. After a summer of putting heat into the slab from a solar thermal system, the sand never got hot enough to provide heat that could be extracted with the PEX tubing loops.

    In an example like this, it definitely takes heat energy to raise the temperature of the sand. The area of the home's thermal envelope has been increased compared to a slab-on-grade home, so now the house loses heat over a greater surface area. Heating up that sand, and accounting for the larger surface area, means that the house uses more heat than a house without a basement full of sand.

  11. user-723121 | | #11

    Thank you, Thorsten

    I have read with interest the posts about your home and there is no question it is very well built and very efficient. As an early adopter of superinsulation myself I really appreciate the amount of effort you have put into the project and your detailed analysis of the building. I live in MN with just under 8,000 hdd and I tip my hat to those of you building in Alaska and Canada, it is a true test of any building system.

    Please keep us updated here at GBA, the monitoring of this house will be exciting and rewarding. The data from this project is invaluable to the building community, keep it going.

  12. user-973254 | | #12

    sand vs soil
    if pursing this concept, why use sand in the first place? I alway hear about using sand in these types of things. A quick search says that soil has a higher (~50%) heat capacity than sand. Im sure im missing something. - after all it was only a 3 search minute effort - :)

  13. kevin_in_denver | | #13

    Thank Radiantec for Sand Beds
    Robert Starr's system from 1983 might have been what started this controversy. The New Shelter article is worth a read for some history:

    His main thrust was to simplify the the space heat delivery of an active solar system. In that it was successful, and it also helped to promote heated floors. We've since learned that heated floors in a superinsulated home are unnecessary, expensive, and overkill.

    “The only monitored-to-work seasonal storage system that I've ever heard of was MIT Solar 1, built in the 1930's"

    Actually, there is a recent, very successful system at Drake Landing, but it doesn't really count because it's a hugely complicated "district" heating system that has no future in one-off single family homebuilding.

  14. jklingel | | #14

    Thorsten, Martin, Kevin, and Lance
    Thorsten: I was kidding about the trees! I am confident that you already thought of that obvious detail.
    Lance: Check the density of soil, and I think you'll see why sand is preferred to soil, if I recall the numbers. Also, you have the problem of compacting the soil adequately, which may not be possible. Sand nearly self-compacts. Once it is wetted and compacted, it is good to go.
    Martin: OK, but anecdotal, so I ain't lettin' you off the hook yet. Ha ha. The area that Thorsten's system adds to the house is not huge, but measurable. It would be interesting to know what the solar data is on the house you mentioned; maybe there was insufficient solar to heat even 4" of sand. That is one example, and extreme, that was or was not engineered by someone with experience. Thus, my questions still float in my head. Thorsten's system is working for him, and I think he has taken some bold steps. As I mentioned earlier, I think he is going to work the bugs out of this deal.
    Kevin: Thanks for the links. I hear this "heated floors are expensive" phrase often, but am puzzled. How else does one distribute heat efficiently? Air ducts? Somehow heat has to get from source to socks on the floor, and at least in my area, I don't see a better way to deliver it. (Barring passive solar or a masonry heater.) PEX is not terribly expensive. Thoughts?

  15. GBA Editor
    Martin Holladay | | #15

    Response to John Klingel
    You wrote, "Thorsten's system is working for him." That's true, but we have no idea whether the house would have performed just as well (or possibly even better) without the sand bed. That's why careful research always beats anecdotes.

  16. kevin_in_denver | | #16

    Response to John Klingel
    In a well-insulated home with high performance windows, the heat doesn't need to be distributed throughout the house with a system of ductwork or pipes in floors. Just a small heater of any type on the lower floor. The current darling of the green building industry is a ductless minisplit heat pump.

    If you have solar heated water available for space heating, a fan coil is simpler, cheaper & just as efficient as a radiant system. But remember, once you have sized a solar system for most of the domestic hot water load, the marginal cost of space heating is usually too high.

    You only have one chance to install PEX in your slab, though, so, it's not a big deal even if you never hook it up. But trust me, if you go for radiant floors you'll be disappointed that the slab feels cold 95% of the time.

  17. jklingel | | #17

    Kevin: I see what you are
    Kevin: I see what you are saying, but "eliminating" a heating system (as most of us know it) is a spooky concept. I will need to study and digest this more. I've heard a lot of mini-splits, but have not inquired about their validity here, esp w/ a ranch style house. I'm open to anything that is efficient, easy to install/maintain, etc, and heats the house. I ran through Siegenthaler's procedure for calc'ing water and floor temps, and you are right about the "cool" floors; somewhere in the low 70's is what the spread sheet said. That beats the (probably) 60 degrees we now have in the basement when it is very cold outside, but it won't be the warm-fuzzy feeling. Thanks for the ideas.

  18. Coyo | | #18

    The 50 Below Mini-Splitt?
    Kevin, send me that mini-splitt which will work in 50 below and 14,000 HDD.
    Perfect solution in the right climate, but not in Fairbanks with today’s technology. TC

  19. user-935581 | | #19

    Can we just agree on some basics?
    As an interested amateur in the building science field, I must say that I find all these debates dispiriting. Can't we as a community agree on anything? I see the same incredibly wide range of opinions on just about every topic. Cellulose vs. foam. Passive solar is great. Passive solar is unworkable. Insulate the floor of the attic. No, insulate the roof to get the AC unit within the building envelope, etc.

    As someone who is active locally on energy conservation and renewable energy initiatives (I used to administer a local gov't fund for retrofitting gov't buildings, and now serve on a municipal utility board), I just want some clarity. I want to know what makes sense for my own house, and what I should be telling others for residential and commercial buildings, both new and retrofit (mainly retrofit, in my community).

    I know this stuff is complicated, but it's not that complicated! I just want a decision tree that takes into account e.g. location, orientation, shading, heat loss (e.g. therms/year), current heating system, etc., and spits out a roughly ordered list of what makes sense. I know the answer is ultimately "it depends," but how about "it depends, but you're probably best off airsealing/insulating" or "just go for PV" or "in your situation, a cooler roof would really help."

    I know it's naive to expect the herd of cats to reach consensus on this stuff, but I just want to register my concern that discussions like these, though very thought-provoking and informative, mainly leave people like me scratching their heads. So... what's the answer?

    Rant over. Cheers.

  20. user-917907 | | #20

    heat gain and controlling heat loss

    Did you bring your heated sand mass up to temperature primarily with solar heat, or primarily with wood heat? As I understand it, you are now solely heating your house with stored heat during the night, and with solar heat and/or stored heat during the day. Do you have any idea whether your sand bed temperatures are dropping since you quit using wood heat, which would indicate that the solar heat stored during the day is not sufficient to heat your house during the night? If your sand bed temperatures are dropping, when do you expect to have a net gain?

    If you are planning on storing solar heat during the summer in your sand bed, it's not clear to me how you will keep your house from overheating? Is there insulation between the sand bed and the floor? Will you use an active heat distribution system at various times during your heating season?

  21. user-917907 | | #21

    re: sand vs soil
    Lance W asked: "If pursing this concept, why use sand in the first place? I alway hear about using sand in these types of things. A quick search says that soil has a higher (~50%) heat capacity than sand. Im sure im missing something. - after all it was only a 3 search minute effort - :)"

    Lance, according to the table found at this web site, , the specific heat of dry soil and of dry sand are the same (0.19 btu/lb*F), while the specific heat of wet soil is almost twice the specific heat of dry soil (0.35). I think it is the addition of water to the soil, which has a very high specific heat (1.0), which increases the soil's specific heat.

  22. user-974256 | | #22

    experimental vs replicable
    Thorsten -

    We're more on the same page than a different one, as it relates to house-shaped experiments, and you are on the right side for discussing what hasn't worked, and setting up sensors to understand how things went.

    My points are to highlight the easily replicable ideas versus the experiments, as often I see people running with experimental ideas thinking (or being sold) they have proven their worthiness.

    I have to say I think there are some places on earth we shouldn't expect to be maintaining ASHRAE interior comfort for humans without the use of 'fuels', and maybe shouldn't be expecting to live in a 'carbon-neutral' world. Providing warm housing, food, and other goods and services in remote cold locations just isn't practical with our current standards of living if carbon neutrality is really the goal.

    There will be a point at which we realize the limits of our technology as it relates to current capabilities and future generations, and nature will corral us back to locations / environments where we can exist for longer terms within the means of more easily available resources. Currently we exist in a world where vast energy resources are being used to expand the range of mankind's reach, but this situation is temporary at the current scale and comfort levels.. Looking at the big picture, that is what I see.

    At some point during construction of our ZE house, it occurred to me that maintaining our PV inverters would require support from the fossil fueled system our house grew from, and therefore the project was and likely would always be dependent on that fossil fueled system. Same goes for PV cells drawn from tightly controlled electric furnaces, window glass floated on a bed of molten tin, materials shipped / trucked long distances to reach the site, foam plastic insulation, PEX tubing, SHW collectors, materials assembled into a refrigerator etc., etc.

    Projects built on the current fossil fuel infrastructure rely on that infrastructure, to varying degrees; winter electric grid support may be a greater form than replacing a broken window, but a broken window in a passive solar super-insulated house (with minimal heating system) cannot go unfixed in winter (something to consider when leaving your house un-occupied in winter) nor can that window replacement be made on-site of locally available materials (ie have no 'carbon impact').

    As the next logical step after realizing where we stood with our ZE project, I have considered trying my project over again, disconnected from the fossil fueled system we depend on, but even defining that experiment is difficult, and completing it would require a level of commitment that isn't currently within my means. Our 'First' world will have a different look to it when someone is working on that problem (the 'Third' world is still familiar with it in places).

    Still, like you, I attempt to work on the interesting problems attempting to reduce impact, while also realizing that overwhelmingly American customers want simple and fairly inexpensive (energy being so 'cheap') 'plug and play' solutions, as opposed to experiments with the potential discovery of drawbacks partway through the process.


  23. user-974256 | | #23

    Agree on basics? It depends on who you ask
    Brian -

    You really need to define a starting point and a budget to get constructive consistent advice, but from where I sit (Boston area), existing building retrofits would involve:

    1. Air sealing anything relating to the air barrier (envelope, ducts)
    2. Insulating existing envelope surfaces / cavities, if not already, R-value subject to practicality & cost (don't go crazy, insulation has diminishing returns)
    3. Providing at least 2 layers of glass at windows, ideally at least 1 low E coating
    4. Review mechanical equipment efficiency, replace with upgrade if at/near end of life or a 10%+ AFUE / SEER improvement is available
    5. Consider solar panels, type dependent on load and incentives

    New buildings: R-10 to 20 in the ground, R-25 to 30 above ground exterior surfaces, at least dbl LowE windows, triples with selective coatings if you can afford them / are doing passive solar. Air source heat pumps are the cheapest option to go all electric efficiently while getting both heating and cooling. DHW subject to fuel options and possible solar. Add solar to taste and budget.

    Compact forms with less glass are better energy-wise than sprawling forms with lots of glass, but remember its a house first and energy consumer second - people are usually willing to pay for energy to get what they want in form / floorplan.

    Simple enough, but debatable on any number of counts...

  24. user-859817 | | #24

    Solar storage on a larger scale
    It seems that soil is a great heat sink, (sand/dirt+water=soil, so it should hold more heat than dry sand). And this project is showing that what is missing in many seasonal heat storage schemes is scale: The Drake Landing site in Okotoks AB Canada is expected to take 5 years just to fully charge the heat sink! in 2010, the heat field provided 80% of the required heat for the dozens of houses that contribute to and draw from it. I can't say it is economical yet (thanks to Federal funding to the tune of $2 million, this one may work well for the residents none the less). Oh and one extra benefit - when you get really huge volumes, your surface area to volume ratio is very small so heat losses are reduced. In this case, there is no insulation around the heat sink at all. (other than more dirt, which does have a small but important R value... but there is lots of it!) see for more info

  25. Coyo | | #25

    Sand, Dirt and lots of confussion
    Dirt vs Sand
    The original question and discussion was about insulated thermal mass in form of sand within conditioned space. You cannot lay your insulation out and place and compact soil and then pour concrete and build your building on top. No compaction = big problem. Sand, gravel or pea gravel are your choices. There are other concepts of heat storage in smaller scale or as mentioned bigger like Drakes landings system – many more in Europe. This is however a completely different approach.

    I absolutely and whole heartily agree…this whole discussion is counterproductive and confusing and I wish I had never answered John K.’s to my minds simple question. For anyone reading through this discussion (and many more on GBA) it must leave you simply confused and frustrated. There is always more than one answer to certain problems but one should assume under “experts” we at least could agree on a general direction…but that is just not happening. My time is extremely limited and wasting it with effectively confusing folks is the last thing I like to do. I contemplated long and hard to post again for this very same reason – but reading through the questions I just can’t let it sit and by now at least it can’t get anymore confusing then it already is:

    Function of the added sand bed in my SunRise home:
    I explained the function of the design in my original answer to John K. but for reason beyond my comprehension Martin more or less ignored what I actually said and kept on defending his opinion that this all simply doesn’t work and is counterproductive and will in reality require more heating instead of saving energy. Somehow and somewhere in this initial exchange there is just a total disconnect…so let’s just forget about this and I try again for the sake of at least maybe Jack and some other folks who got this whole concept backwards to explain it again.

    1. Forget about anything else which was discussed about this so far
    2. Forget about any source of active heating – renewable or fossil fuel based.
    3. Understand the basics of Passive Solar Heat Gain
    4. Understand the fundamentals of thermal energy and heat flow – and this is crucial if you want to ever grasp anything related to building physics.
    5. Thermal energy is always in motion and cannot be stored, we commonly refer to this phenomena as heat loss
    6. Lets forget about using sand – let’s just call it thermal mass
    7. Be willing to think outside the box

    The added INSULATED mass in a building allows the temporarily storage of thermal energy in form of heat being absorbed by convection into the mass by the laws of physic from warm to cold. The amount of mass and the temperature differential available within this mass allows small or large amounts of heat transfer depending on the available mass to do so. Energy never stays still and is in constant movement by the laws of thermal dynamics. During the day we gain energy through the solar radiation and the appropriate glazed windows with a high SHGC from the free power of the sun. The thermal energy will heat up the building…depending on the design it overheats and we open the windows and let the energy back out or we incorporated some method of temporary heat storage in form of internal mass. If we design this system well we can storage large amounts of free passive solar energy which allows us to not have to rely on active heating in the home at night or cloudy days as the temperature differential will make the thermal energy flow supply the needed BTUs to heat the home from the temporary heat storage. Thermal energy is always in motion – and always travels from warm to cold. Any building no matter how well insulated will have heat loss if the outside temperature is colder than the inside of the building.

    The SunRise home has 180 tons of passive storage capacity in the superinsulated foundation. I chose sand as it is very cheap, easy to work and has 95% compaction when placed. A thickened slab could be used but is expensive, has very high embodied energy and is impractical in such a scale. Without any active heat since February 16th the home maintained anywhere from 68 to 72F with constant freezing outside temperatures up to -28F at night. Look up the last two months of weather records for Fairbanks. No anecdotes, no sales pitch – but think about how well the thermal fly wheel in the building needs to function to make this happen. I find this pretty amazing not because I designed and build it, that is not the point – the point is that it actually can work that well. And yes absolutely no question that it would be impossible to maintain the inside temperature for two months without the mass loaded foundation. This is certainly also not anything new and has been done successfully many times by others. Still here we are debating about it. Business as usual. Passive solar heat gain is the ONLY free energy we can capture. And to me at least that has potential and appeal…

    The other functions of the design moves all the grey water plumbing within conditioned and insulated space and eliminates many unaccounted thermal bridges through these pipe connections which usually establish a physical connection between the cold ground and the insulated space. This also acts as one huge drain water heat recovery sink. These are both tremendous advantages in my book. And I don’t even like to add this because it started this confusion – I can actively divert heat from my solar thermal collectors into the sand bed in the fall to bank additional heat before the home goes into winter. As the main function of the system is passive solar heat gain there is no control of the thermal energy flow when it is banked active. Heat in – heat out. That is why I don’t heat it now or in the summer as I would simply overheat and be uncomfortable.
    Now I said this before in my closing comment to Martin – sand in a foundation can either be a disaster or it can add tremendous value. It only makes sense if a good amount of passive solar heat is available – that is essential. It is counterproductive to heat a mass foundation like this with wood or a boiler. And it needs to be insulated and within conditioned space to function.

    Martin – your turn to rant – I will however not respond to this any further. Once enough data is available from CCHRC to make a proper case I will write up a report with details and make it available to whoever is interested.

    PS: Mark, I believe that our days of cheap energy are numbered, so that typical “turn up the heat” attitude will regulated itself soon enough. With what’s ahead of us at least in the far north regions we better come up with some solutions soon. Unfortunately that requires R+D and willingness to step outside the norm, my believe at least.

  26. GBA Editor
    Martin Holladay | | #26

    Response to Thorsten Chlupp
    I'm sorry that you have concluded that GBA's attempt at answering John Klingel's question left readers confused and frustrated. Obviously, that's not GBA's aim; we want to clarify technical questions by providing authoritative answers.

    I don't want to get into a debate with you, Thorsten, especially since you have announced that you are bowing out of the debate. But I'd like to remind our readers that the origin of the discussion was a question from John Klingel: "For you folks who put a sand box under your slab and heat it, where do you put the PEX?" According to my reading of Klingel's question, he planned to heat the sand with his wood gasification boiler. Klingel's proposed system was different from yours.

  27. user-917907 | | #27

    Thanks for taking

    Thanks for taking the time to once again explain your system. Some of us take longer to catch on than others do. As I understand it your sand bed is well insulated on its periphery, but has no insulation between it and the living space of your house. You use active means to transfer heat from solar panels or wood stove to the sand bed, but rely on passive conduction to transfer heat from the sand bed to the house.

    At this time of year, the Spring, with longer sunny day but still cold nights, I can see that your passive heat transfer system could work quite well. You are able to collect and store a lot of heat during the day, and then that heat passively transfers out of the sand and into your house during the night.

    What I'm having trouble understanding is what will happen in the Autumn and Winter. In Autumn you will also have long sunny days, but the nights will still be warmish and you won't need much nighttime heat. Since the sand is not insulated between the bed and the house, if you put much heat into the sand bed you will quickly overheat the house. By the time Winter comes in earnest, with its cold days and colder nights, it seems to me that you will not have been able to store up very much of that free summer/autumn heat because if you tried to do so your house would have become uncomfortably hot. Perhaps in the dead of winter you expect your primary heat source to be your wood stove, and the solar panels will just contribute what they can when they can?

    I hope that I am as confused as you suggest I am, and that this time next year, after a whole year of experience with your system, you can provide us with actual data that will clarify how your system works in the full range of yearly conditions.

  28. user-723121 | | #28

    Thermal storage
    If my fuzzy math is correct, 180 tons ( 360,000 lbs. ) of sand storage at .19 Btu per lb. per degree F yields 68,400 Btu's of thermal storage per degree F. This amount of thermal mass (180 tons) is a lot but is in no way sufficient to store a season's worth of heat or even a large fraction thereof. This is not a critique of Thorsten or his project, I am fascinated by it and the information he has provided. What it does show is how much energy we use and how energy dense fossil fuels are. It helps to make Mark Sevier's point, Net Zero homes have a direct connection to the grid and fossil energy.

    With Thorsten's house, if available wood is the main heat source and solar is providing some daytime gains, this is a very interesting project. He certainly has enough thermal mass to prevent daily temperature swings and the ability to store a considerable amount of low grade heat. Thorsten has generously shared the details and his time here and we should learn from it.

  29. homedesign | | #29

    "Some writers think a sand bed is a waste of time."
    After first reading this blog....
    I had the impression that "some writers" think ALL sand beds are a waste of time.

    Are "some writers" now saying that perhaps NOT ALL sand beds are a waste of time?

  30. Coyo | | #30

    Thermal flywheel vs seasonal storage vs the sum of all its parts
    The mass loaded foundation is the thermal flywheel of the home, not the seasonal storage. The 5000 Gal internal thermal storage tank stores up to about 5 Mio BTUs to bridge longer gaps. Mark Sevier kindly did the math to show how this also won't get me far. Many engineers have calculated this for me on other projects and shown me on paper how this cannot work. However experience from prior projects has shown me that in reality things work out a tad different then on paper. The building and the integrated renewable energy systems are the sum of all of its parts and the constant changing outside conditions. It is impossible to model and predict how the interaction between these variables flow in reality. And it is never a constant and changes every season. No winter is ever the same; there can be more or less sun which is the governing factor of the energy flow.
    If I didn't see viable potential I obviously would have abandoned the concept two years ago. Next spring will know how much wood I burned to make up for the difference. I don't think it will be more than two cords, most likely less. That isn't too bad to make it through a winter in Fairbanks, Alaska for heating and supplying hot water to a 2300 SF home for a family of four.

    Jack Wolfe,
    it takes some time to load up the mass foundation in the fall and it will have us open windows once in a while. The stored heat will be one part of the overall system. The main thermal battery is the superinsulated seasonal storage tank which sits within the home and is the main heat source and provides DHW through the tank in a tank design. The thermal collectors prioritize heating demand in the house, if the T-stat is calling for heat the collectors will heat the home directly first if there is enough solar heat. We only need 70F for this. This winter after we finally commissioned the system I had 38 below at the collector sensor in the morning and pulled 136F of heat by 11:30 AM at 24F below outside temps. If the house is not calling for heat it loads the tank if there are enough temperature differentials to do so. Since the tank can stratify very well we can harvest lower temps and still collect useful BTUs. If there is no demand in the seasonal tank it will start loading heat into the foundation (starting in the fall).

  31. user-716970 | | #31

    But can a high mass floor be too thick?
    Just to stir the pot a little more, here is what Robert Riversong wrote in his Home Power Magazine article

    "Dense materials, like concrete, which have a specific heat of 28 Btu per cubic foot per degree F (about half that of water), tend to allow heat diffusion at a rate of about 1 inch per hour. So the heat of the noontime sun will penetrate to the bottom of a four inch thick slab by about 4 p.m. and all that heat will have returned to the interior by about 8 p.m. --making a 4-inch-thick slab ideal for solar thermal mass. This is called the "diurnal heat constant" and it shows a diminishing return with a slab thicker than 4 or 5 inches."

  32. user-935581 | | #32


    See? That wasn't so hard- provided others don't chime in and criticize your list!

    By the way, I will mention that I'm halfway to local micro-CHP with the boiler half of a Climate Energy system. I'll do the generator at some point, but I've got other more immediate priorities. And I'm sure we can have a hearty debate on the merits of micro-CHP vs. air-source heat pumps, etc.

  33. user-917907 | | #33

    "ah ha" moment
    Thanks for the explanation. Understanding how your hot-water storage tank works in conjunction with the sand bed makes things much clearer. Flywheel vs storage battery. Got it! Neat system. :)

    Are you using flat-panel or evacuated-tube solar panels? Are you orienting them more towards the summer sun or the winter sun -- or can you adjust their pitch seasonally?

    What is the best way to build a large insulated hot water storage tank? Rather than buying a tank would it make any sense to create some sort of insulated "swimming pool" in a basement, and line with a waterproof membrane perhaps like EDPM?

  34. user-797100 | | #34

    Use Mark's example,
    Use Mark's example, 14.4mbtu/yr. Let's say 20m2 south window, SNCC 0.5, avg 3kwh/m2-d, now you have 30kwh/day passive gain, or about 100kbtu a day. If 5 month winter season, that's 150day, that's about 15mbtu. So in general sense, passive gain already meets the load. So the storage only needs to take care of the fluctuation. It's all about design objective. Say we want enough storage for 10 very cold days with no solar gain, let's assume avg 0F so deltaT 68F so 6800btu/h, or 2kwh/h, 48kwh/d, 480 kwh for these 10 days. That's about 1.6mbtu. So I think John's system may actually work. of'course this is very simplified calculation just to show we should consider overall design (heat load, solar gain, and storage all together). The solar gain can be window, or solar panel, or combination. I'm looking forwad to see John's data next year.

    Harry Zhou, a car engineer

  35. jklingel | | #35

    I'm glad I asked this question, as I am learning things here and there. Confusing (which I didn't really see) or not, I am glad there are various opinions; it seems everybody knows something, or at least challenges someone else to explain their position better. FWIW, if I go this route, I will be copying Thorsten as best I can. First, however, he needs to come out with his Gee Whiz box and see if he thinks I have enough solar to function properly; I fear I don't, at my site. This is all new to me, too, and I am very tentative about its viability at my site, hence the gasification boiler in case my solar is insufficient. That opens a can of worms, so there is much to think about. My personal opinion, based only on my gut, is that the PEX, if used to heat the sand for 1-3 days use, should be roughly in the middle of the sand to heat it uniformly. How important that is, I have no clue, as I just see the heat eventually going to cooler locations, that being into the house. This is a great concept that Thorsten is bring here to air out. Cheers. john

  36. user-716970 | | #36

    For John Klingel
    Here is a thread you might find particular attention to Robert R's comments...

  37. user-917907 | | #37

    Don't heat uniformly

    My personal opinion, based only on my gut, is that the PEX, if used to heat the sand for 1-3 days use, should be roughly in the middle of the sand to heat it uniformly.

    In another thread someone mentioned that a radiant slab that is heating the house to a comfortable temperature will feel cold to walk on -- because 70-degree concrete quickly conducts heat away from our 98-degree feet. So if you heat up the concrete to a temperature comfortable to walk barefoot on it will overheat the house. To get around that problem it was suggested to heat certain areas to 80 or 90 degrees, but leave the rest of the slab unheated, or less heated.

    So maybe you want the sand under the bathroom, or at entrance ways, or bedroom walkways to be warmer. In those areas perhaps put the PEX close to the slab, or put one layer in the middle of the sand and another layer close to the slab just in the critical areas.

    Another possibility would be to place insulation between slab and sand in places where heat wasn't needed -- utility room, under beds, under living room furniture, under cabinets and closets, etc.

  38. user-723121 | | #38

    Floor temperature
    A floor at room temperature is not uncomfortable, no more than 70 degree air is uncomfortable. Sub-slab insulation bears this out. With insulation beneath the concrete floor, the slab temperature is very close to room temperature, without insulation the slab is near the seasonal soil temperature. Insulated slabs feel warm, uninsulated slabs feel cold, the radiant heat loss (from your body) is very noticeable, like standing next to a dual pane window in winter.

  39. user-917907 | | #39

    Floor temperature
    I agree that an insulated slab will be warmer than an uninsulated slab, given that the soil temperature is colder than the indoor temperature. However, I walk around in my house either barefoot (in summer) or in stocking feet (in winter) and I can assure you that concrete and tile feel much colder than wood or carpet, in spite of their being, or not being, insulated.

    If you think that all materials which are 70 degrees will feel the same as 70 degree air, I suggest you try taking a shower with 70 degree water. What temperature a material feels is dependent on how thermally conductive it is, as well as its specific heat. Standing on a 70 degree slab of copper will feel colder than a 70 degree slab of steel, which will feel colder than a 70 degree slab of concrete, which will feel colder than a 70 degree wood floor, which will feel colder than a 70 degree carpet, which will feel colder than a 70 degree block of insulation.

  40. user-917907 | | #40

    temp of concrete floor
    I'm not sure what temperature concrete would need to be heated to feel comfortably warm, because radiant-heated concrete floors are generally not heated to that temperature. If they were heated to, say, 80 degrees, then the room temperature would be uncomfortably warm for most of us.

    I should reiterate that I typically walk around in my house in bare feet or stocking feet, so I am much more sensitive to floor temperatures than someone who keeps their shoes on. Shoes insulate your feet from the temperature of the floor, so a 70 degree concrete floor would likely be comfortable to most people who wear shoes indoors. (I'm speaking about temperature, not softness)

    I understand that some folks with radiant heated concrete floors cover the concrete with a thin layer of vinyl or wood flooring. This covering needs to be thin so as not to impede the heat transfer from the concrete to the room, but a thin layer of either of these over a concrete floor will feel warmer, as well as softer, than the bare concrete floor will.

  41. user-723121 | | #41

    Human comfort

    Are you saying for floors (concrete) to be comfortable they should be heated to 98.6?

    I agree 70F concrete floors on bare feet will feel cool, cooler than carpet certainly. I advise customers against heated floors, with concrete on the warm side of the insulation, the space will be comfortable. I had a townhouse project where the customer was considering radiant heat in the lower level (concrete floor). In Minneapolis forced air heat with central AC is the standard. To add a condensing boiler and tubing in the floor would have cost 13K, instead we added a (forced air) zone for the lower level for $2,500 and the customers could not be happier with the result. The basement floor has 2" rigid beneath and is isolated from exterior foundation walls and footings. This is a dry basement and the bulk of it is carpeted but the bath is tile over the concrete and they have no comfort complaints. When you walk from the main level to the lower level there is no perceived temperature change. all surfaces are 70F.

  42. user-973254 | | #42

    reply: comfortable floor

    because 70-degree concrete quickly conducts heat away from our 98-degree feet. So if you heat up the concrete to a temperature comfortable to walk barefoot on it will overheat the house.

    human feet are more like around 80º

    even so, an 80º floor would only make the house 80º (ie overheat) if there werent any thermal losses. And who has a zero-heat loss house? :)

    I'm not sure what temperature concrete would need to be heated to feel comfortably warm, because radiant-heated concrete floors are generally not heated to that temperature.

    Again it depends on the heat loss of that space, the floor could very well be 80º...

  43. user-723121 | | #43

    I had a 2nd floor addition project that included a master bath, all tile floors and shower. We debated radiant mats under the tile but decided against it. I created a dedicated chaseway for the supply and return to the 2nd floor via a closet on the main level. The connecting point for the heat supplies was right at the edge of the bathroom so I intentionally routed my heat supplies for that side of the addition under the bathroom tile floor. When the furnace starts in the morning after the nightly setback the long initial run time heats the tile floor nicely before the customer uses the shower. There is also a large south facing window in the bath area so direct beam energy is absorbed by the floor during sunny days.

    Jack's earlier point about heating living spaces differentially is a good one. Why could we not set the heat for 50F for the building and have radiant panels where the occupants spend most of their time. The panels could warm small spaces for people while leaving the bulk of the building much cooler with far lower heat loss.

  44. vCWJKZSiuP | | #44

    thermal mass and storage
    As usual I've enjoyed the discussion and the thoughts of all involved.It can be frustrating,for sure,when crunch time comes and a choice is necessary.As we know,even the most simple structure involves hundreds of choices,and each one is made with the usual combination of unquestioned assumptions,actual research,cost factors,availability,and on.For someone like Thorsten to assemble most of this into a very workable solution for his circumstances is admirable,and it's worth thanking all who make such efforts for us to study and analyze,
    My only contribution here is the strained sense of deja-vu that makes a high-pitched whine in my head.Anyone ever heard of Norm Saunders? In the late 70's,early 80's he was designing and building houses which claimed 100% solar heat for his locations in Massachusetts.He used a more thoughtful version of the solar collector-rock bed idea. In the mid-80's i was a volunteer guide through a building which was the headquarters for a state conservation organization.Designed to test and demonstrate various solar building strategies of the time,it won awards for it's designers.The most successful part employed extra insulation(R-45 roof !) and a fan-driven loop that pulled air through an insulated crawl space if the space above became too warm.Data on temperatures in all parts of the building was collected.
    So,after seeing all this effort more or less discarded for another 30 years of blissful indulgence in fossil-fuel gluttony,only to become the subject of more debate now , makes me tired and cynical.Yes, I believe this is important.No I don't believe it will make any difference whatsoever.As they say,you can't dig yourself out of a hole,but as far as I can see we are still living in a society that is digging furiously. I come here for info I can use in my own work and it keeps me going.But I no longer bother to think that it will change anything . So,lets not get too caught up in criticism.

  45. Ericsundial | | #45

    Sub Slab Thermal Storage
    Thanks to everyone for the great reading on this subject.

    I am in the process of building our family garage with some living space above. The foundation design has led me to use this opportunity to attempt to implement some thermal storage capacity. I am in western Manitoba. The building will be thoroughly insulated and air sealed (still seeking a solution for the overhead garage door & threshold) and rather than calling it a garage, it might be better considered as a passive house that you park your cars in.

    I have included some drawings and photos of the progress so far but I am now nearing the point of needing to commit to the piping/insulation configuration in and below the slab. I am very appreciative of any advice on where sub slab insulation should be placed (and how much), as well as the location for sub slab hydronic piping.

    My goals are to take better advantage of passive solar heating from the south facing windows and to make some use of the low grade heat produced by my evacuated tube collector (currently providing most of my DHW and some space heating for my house directly adjacent to this new building). Although I do not plan for my house/garage to ever be off grid, I expect to design some off grid buildings in the future and feel that thermal storage will be part of the equation. I would like to gain some insight from this project with those future projects in mind. My vision is for such off grid buildings to achieve their space heating and DHW requirements without the need for ongoing wood or fossil fuel heating.

    I have attached some drawings to show my progress so far and to illustrate two configurations I have been considering. Although I would like to keep the system as passive as possible, I am expecting to use a (very small) water to water heat pump to pull/push heat from my sub slab storage when required. I would expect at times to push/pull heat from the slab to the sub slab piping by pumping the heat transfer fluid (water) directly and at other times to use the heat pump. In many ways I believe what I have here is a small sub slab geothermal field that may have some better than typical capacity for thermal storage. I look forward to anyone's thoughts and advice.

  46. GBA Editor
    Martin Holladay | | #46

    Response to Eric Bjornson
    My opinion hasn't changed since this article was written. Attempts to store heat in soil or sand under a building are a waste of the equipment needed to transfer heat. The amount of useful heat that can be stored and recovered is too low to justify the capital cost of the equipment.

  47. Ericsundial | | #47

    Response to Martin Holladay
    I've spent the last couple of days reading the posts & comments on GBA related to thermal storage/sand beds/seasonal storage etc and was bracing for this response. I will not disagree with you however in terms of my proposed concept not being the lowest lifecycle cost method for heating a grid tied building. However if considering solutions for cold climate buildings/communities not connected to the electrical grid I feel there is some credibility in exploring reasonable cost options for sub-grade thermal storage. It could be argued that the marginal cost of adding the thermal storage to a foundation as shown in my drawings is only that of the installation of the sub slab pex piping and any related pumps and controls. Although pricey now, I would expect/hope that there is a lot of room for small W-W heat pumps and simple controls to come down in price.

  48. GBA Editor
    Martin Holladay | | #48

    Response to Eric Bjornson
    Be careful with any plan to use a heat pump to remove heat from soil under your foundation. It's possible to freeze the soil under your house if you remove enough heat, and that's not good for your foundation.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |