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

Can Heat Be Stored in a Sand Bed Beneath the House?

Saving summer's thermal energy surplus is an attractive idea, but opinions vary on how effective these designs really are

A Solar Flywheel By supplementing an efficient gas water heater, the sun provides more than 75% of this home’s total heat and domestic hot water. The combination of active solar collection and passive distribution provides all but exceptional hot-water needs in summer. The winter sun is too weak to heat the domestic hot water, but it supplies a boost to the heating system. The sand is heated to 75°F.
Image Credit: Fine Homebuilding

David Meiland was intrigued by something he’d read in a Breaktime post at Fine Homebuilding magazine in which a Minnesota builder discussed plans for a heat-storing layer of sand 4 ft. thick below the house slab.

PEX tubing would dump heat gathered by solar collectors into the insulated layer of sand in summer, and extract it during the winter. Although sand isn’t an ideal material for this type of system, it’s cheap and easy to work with.

Anyone care to comment? Meiland asks in a recent Q&A.

Sand is not the best option

This idea isn’t new, writes J Chesnut. While Chestnut has heard architects discuss it many times, and despite the FHB poster’s experiences, he finds the sand-bed approach too complicated.

“There are too many variables that can’t be modeled and analyzed as a system (as far as I know),” Chesnut says, “daily weather patterns, heat storage capacity of the bed, efficiency of capturing solar heat, the rate of thermal transfer, the heating design load, heat distribution, etc.

“This strikes me as a ‘shooting from the hip’ approach.”

Given the vagaries of a sand-bed design, the house would probably need a back-up heat source capable of meeting the entire design heat load, a redundant second system that only adds to construction costs.

“Intriguing idea,” adds GBA advisor Martin Holladay, “but the data don’t back up the theory. In short, when winter comes around, the sand isn’t warm enough to provide useful heat.

“Building an insulated sand bed costs money — and the yield on the investment is nil or very low. Storing useful amounts of heat for more than three or four days is hard.”

“We did this about 25 years ago,” writes Tom of American Solartechnics.…

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. Kevin Dickson | | #1

    Drake Landing
    The Drake Landing project seems to be well designed. Does it have a lot of pumps and stuff that need repair and maintenance? Yes. Would I buy a house there? Never. HOAs are terrible at managing systems like this.

  2. GBA Editor
    Martin Holladay | | #2

    Drake Landing subsidy
    While the Drake Landing community has received a lot of admiring press, here's the number nobody wants to discuss: the equipment was made possible by a $7 million government grant, equivalent to $134,000 per home. Heck, with that kind of money a lot of things are possible.

  3. mike | | #3

    thermal tank solar storage
    is something i've been pondering for a while. after learning about it via sobek's hausR128, i've been trying to get my head around it...

    this super passivhaus in galway (IE) is attempting near 100% DHW and space heating from a small (10x10x8) superinsulated storage tank.

  4. Jesse Thompson | | #4

    That's a 6,000 gallon


    That's a 6,000 gallon storage tank! How many PV panels can you buy for the cost of one of those?

  5. mike | | #5

    i've seen a few


    i've seen a few underground 6,000 gal conc. water tanks for around $4k. insulation, manifold, pex tubing and earthwork maybe another 5-6k. the storage medium (water) is nearly free...

    i'm guessing for less than a 1.5 kW PV array, you could conceivably be near 100% space heating, 100% DHW.

  6. GBA Editor
    Martin Holladay | | #6

    Response to Mike
    If you really want to size the system for 100% space heating, you forgot a lot of equipment, including, say, 18 to 30 solar thermal collectors.

  7. Jan Juran | | #7

    Some/most tanks leak, at least eventually. Does anyone have any practical experience or knowledge of how long an interior sited very large tank might perform before a leak might occur? This could pose a significant risk for a homeowner, and fixing it would likely be more onerous than most maintenance tasks for alternative systems such as PV systems or wood/biomass systems.

  8. GBA Editor
    Martin Holladay | | #8

    Response to Jan
    They all leak. Such leaks take between two weeks and ten years to develop.

    The tank in Eric Doub's famous house in Boulder leaked soon after it was filled. It was fixed, but it was a big pain.

    The copper tank in the famous "Hanover House" designed by Marc Rosenbaum leaked after a few years. When I spoke to the homeowner a few years ago, the tank had been removed and he hadn't yet installed a replacement tank.

  9. user-788447 | | #9

    centralized vs. decentralized
    I'm a fan of hydronic heating and want to make it work for super-insulated buildings. One concern with hydronic heating on a PassivHaus design (i.e. a design with very low heat loads) I worked on was response time and cost. When the highly insulated envelope needs heat, electric resistance heating is an easy, cheap and quick way to supply the heat. (Note part of the assumption that hydronic would have lower response time is that we did not want any combustion within the envelope so we would be converting electric resistance to hot water anyway. While electricity is a dirtier energy source than the available natural gas the project included enough PV to make up for the calculated carbon footrprint of the annual electric use).

    I'm still interested in hydronic space heating systems that leverage the heat stored in solar hot water tanks that could makeup for several cloudy days without direct solar heat gain through windows. I'm thinking when you can get your design heat load under 40kBTU with your building envelope design plus passive gains you can use 2-3 solar thermal panels, a larger solar thermal storage tank w/ a backup supplementary heat source to cost effectively cover both DHW and space heating. A question I have is whether it is more efficient to use on on-demand as the backup or a heating element in the storage tank.

    I appreciate the attempt by the Drake development to find efficiencies with a centralized/ district heat source. It seems to me however that their configuration of discrete individual family homes, even though spaced closely, undermines much of the potential efficiencies. How much heat is lost from the storage tank into the ground when being transferred to the homes? How much cost is invested in insulating the long runs of water pipe? In this configuration I would expect the PassivHaus approach best capture usable solar heat in the winter and potentially be more cost efficient.

    In general I suspect centralized systems work best with commercial buildings that include internal heat gains that can be transferred to supplement the heating needs of other parts of the system.

    Are there any examples on GBA of apartments, condos, cooperative living setups that cut down heating/cooling loads by minimizing the loss/gains through the thermal envelope?

  10. GBA Editor
  11. user-788447 | | #11

    Thanks Martin,
    Thanks for point these out. I hadn't seen them.

  12. Mark Klein Gimme Shelter | | #12

    more on solar thermal
    The advantage of high mass systems is their simplicity which makes them accessible to owner builders and low tech folks, the disadvantage is their simplicity which means that they are less easily controlled. Fortunately as is the case with other radiant systems they are somewhat self regulating and are generally quite comfortable to live with. As the Drake Landing project illustrates seasonal heat storage is quite challenging ( and expensive) Our experience has been that there are diminishing returns on investments in the envelope and at some point in that calculation a somewhat expanded solar thermal system (either high mass or tank based) that is sized to provide space heating in the shoulders of the heating season and lots of domestic hot water can be a part of a successful and comfortable home that has little if any need for daily fossil fuel input.

  13. Tom Gocze-American Solartechnics | | #13

    Thermal storage tanks
    I think you are painting with a pretty broad brush when stating that all tanks will leak within ten years.
    Many tanks will last a lot longer than ten years before having any issues. Of course, there are always Murphy's Law cases, but this is apt to occur with any mechanical installation.
    We are removing a tank next week that has been in service for 23 years. We are installing one of our newer tanks. The only reason for replacement at this time is for us to install a better insulated tank and also to satiate my curiosity as to how well the liner has stood up.

    I am extremely skeptical of anyone who wants to bury a tank and expects it to not ever have a problem. Everything eventually needs some service. I think it makes more sense to take a pragmatic approach to all mechanical systems whether they are renewables or not. Something might need attention. This is why everything in our product line is modular in its design. The tanks can be refurbished without total replacement at a cost much lower than replacement.

  14. GBA Editor
    Martin Holladay | | #14

    Response to Tom
    You're right, of course, about commercial 80 to 200 gallon tanks. I was talking about the 1,000 to 8,000 gallon tanks -- many of which are site-built and trouble-prone.

  15. Coyo | | #15

    Some thoughts...
    Seems like someone always tries to lure me back to GBA...and after reading through this I just can't help myself voicing my opinion even knowing that my view is contrary to what most here seem to believe.
    My take is based on building and designing high energy efficient homes in very cold climates (14,000HDD). Increasing internal mass in the passive solar design helps tremendous to minimize heating demand in the shoulder season - especially in super insulted and tight homes with proper glassing with high SHGC. And they minimize overheating at the same token and help to control a even indoor climate. A insulated sand bed under a standard 4" slab can work great and is cheap. Besides adding thermal mass it also makes it possible to contain all plumbing within conditioned space under the slab. Traditional FPSFs slabs have many plumbing penetrations which connect the insulated slab with the cold ground. The better the slab is insulated the more this becomes apparent - and a infrared camera can reveal quite clearly how many weak points are build in - which are constant thermal bridges. IMO we cannot afford any thermal bridges in very cold climates if we want to build truly high efficient buildings...and it is fairly simple to place all the plumbing into a insulated sand bed to avoid these - which will leave only sewer and water main to penetrate the insulation layer. A secondary run of 3/4"PEX toward the bottom of the layer can be used to actively move heat into that storage. On an average home size your easily talking 120-140 tons of (cheap) internal mass. Loading this mass requires a substantial solar thermal system and time. One however also should not forget that this mass comes with additional upward vapor drive per delta T - which can cause havoc on flooring installed above the slab.
    Annual seasonal heat storage is difficult and we have plenty of examples of things gone bad. I hear the argument all the time especially from folks who have been building solar homes for a long time. I have seen and studied many good and bad attempts which have been explored decades ago. But over the last few years I have also seen very successful projects in Europe - especially Germany. By now plenty of homes have been build and operate 100% on solar and annual heat storage. Jenni Energietechnik has been building annual solar heat storage tanks since 1976 - and a lot of good information can be found at their webpage: The founder Josef Jenni has done a lot of groundbreaking work in this field over the last decades - his book "Das Sonnenhaus" is in it's 3rd addition. He build a multifamily home in 2006 with a 205 000 liter = 54 000 gallon seasonal storage tank which provides 100% heating and DHW year around without any backup heat. Good examples of homes build can be found at - unfortunately as is the case with anything Passiv Haus a lot of the information on the net is in German only. There are several commercial companies which build tank based systems and have perfected the idea of storing summer heat for use in the winter. EPDM lined rectangular "swimming pools" heated with a copper coil isn't the best way to go about seasonal storage. A lot of research has been done the last few years - especially on large scale storage and the most efficient ways to stratify tanks. Whilst I agree that early attempts might have been utterly unfeasible by now this is not the case anymore and there are many successful developments in Europe. It takes good tank design and most importantly proper tank stratification to make it work well.
    I personally believe that for our far north regions - especially our remote villages annual heat storage will become a necessity to be able to sustain themselves. Barging, shipping and flying out heating oil is already formidable expensive and we will need to find answers not just in our building systems as far as efficiency levels go - but also in energy production and storage. Without pursuing concepts and ideas in this direction we will stay stuck in the general believe we are in right now - that seasonal storage is nonsense and unfeasible. However sustainable buildings of the future will need to sustain there own energy demand...annually. And we have an abundance of sunlight 24/7 Far North in the summer and nill in the winter.
    We incorporated a 12,000 gal seasonal storage tank in a (passiv haus) project we finished this spring which is of grid and powered solely by a micro-hybrid system of solar thermal and PV, 3.5KW wind turbine and a masonry heater - which all can dump heat into the seasonal storage tank. And I am about to start construction of another "unfeasible" Solar home right now…
    Point is, it is easy to dismiss ideas of sandboxes and storage tanks but I argue to think outside the box and look into the future and be willing to learn, move forward and not be stuck with the idea that we know it all. Enough rambling, Happy Building...TC

  16. Tom Kelsey | | #16

    Cordwood home with solar-heated sand bed
    Check out the comprehensive journal/history that documents the cordwood home that Alan built in Minnesota. Featuring a large sand bed with solar heating, it provides a good example of what can be done using basic materials in an owner-built home.

  17. Drew | | #17

    PV grid costs
    Although in concept I agree that an integrated PV system tied into the grid for net-metering is ideal, in high cost states (such as NY) the additional monthly fees & taxes make getting off the grid more desirable & shorten the payback period.

  18. Foraker | | #18

    Solar Thermal Storage system
    Check out for some interesting work done in Montana (northern US) on an inexpensive (under $1,000) solar thermal water heating system. How much would such a system have to be expanded for building heating?

    Build-It-Solar is a good site that includes both the author's experimental results and a compilation of solar and related ideas on the web. A good reference site when looking for a solution to a particular problem.

  19. Richard Fox | | #19

    solar thermal
    we included a 9 panel solar thermal panels in our new home. thermal storage uses a 1000 gallon fibreglass tank with approx 8 in urethane foam and a sprayed rubber coating on the outer surface, buried under the garage surrounded with pea-gravel, and a structural slab over it on pilings. I have a manhole cover in the garage for access- which is appealing to my eccentric nature.
    I fabricated heat exchange coils with 6 3/8" annealed copper coils (25' lengths ) attached in parallel to a 3/4 inch header to deliver heat from the collectors. Heat is extracted from the tank through 2similar coil arrays. One array acts as a pre-heat circuit in the supply line to the domestic hot water, and a the second provides a pre-heat line to the return side of the radiant heat system. I ran a heat-pex line under the driveway slab as heat dump for an over-temp circuit. I places sensors all over this system to be able to analyse its performance and learn of its strengths and weaknesses.I am quite impressed so far with this system. from may- end of oct, the system can maintain the tank at 150-160 F/ The over-temp circuit is definately necessary in summer- My tank is rated to 180F, and the supply temps from the roof typically are 215F at the current flow rates.I installed a fill/drain system to lower the tank volumes in winter to keep the tank temperatures up.
    For example, the supply temp to DHW is 55F from the city, and returns from the storage tank coils at 110-112F with the tank temp at 120F, which is a modest operating temperature for the shoulder seasons. Thats an impressive deltaT! The key with solar thermal is to use the heat at a point in the system where the available temperature is useful. solar panels will be most efficient when supplied with relatively low temperature glycol.
    more on the system later. Acknowlegements to Mr Shea Long of EcoMechanical solutions, Devon, AB. Shea and I designed and fabricated the system. RF

  20. GBA Editor
    Martin Holladay | | #20

    Response to Richard Fox
    It sounds like you have installed a well-designed system. I'm sure the system works -- and it certainly makes more sense to store heat in a 1,000-gallon water tank than a sand bed.

    Over the long term, your system will require maintenance, of course. Eventually pumps and controls will fail; let's hope your fiberglass tank doesn't overheat when that happens. You'll have to monitor the pH of your glycol solution and replace the glycol periodically.

    After 20 or 30 years, you'll have enough maintenance and performance data to calculate the total system costs, including the cost of maintenance, so that you can compare those costs with the system's thermal benefits. The fact that you need to dump heat into your garage slab during the summer points to an often-neglected point: you can't use every BTU collected by a solar thermal system.

    With a grid-conntected PV system, however, you would be able to get credit for every kWh of energy your system produced. That's why the cost-effectiveness of grid-connected PV systems is usually better than solar thermal systems like yours.

  21. Richard Fox | | #21

    indeed, PV is the

    indeed, PV is the way. our cost was about about 6 $ per watt whhen i looked into this 2 yrs ago. what is your estimate of materials cost per watt for pv panels currently?

  22. GBA Editor
    Martin Holladay | | #22

    Installed cost of PV
    I've heard people estimate $6 or $7 per watt (installed cost) for PV systems, but the price is still dropping. You can buy PV modules now for $3 a watt.

  23. George Lawrence | | #23

    Sand bed storage
    Andy Cay did a presentation on two houses he designed and built in southern VT at the Better Buildings by Design Conference in Burlington VT in 2008. Here is a link to where his presentation can be found.
    It is called Whole house Energy design - Cay. In his presentation he states that the heat stored in sand during the warmer months can last into January as long as you have a very good shell.

  24. David Price | | #24

    Phase-change storage
    Storing "cold" in the winter, by freezing an insulated underground reservoir, takes advantage of the phase change from water-to-ice. Installations of such systems are not uncommon.

    Storage of heat should take advantage of phase-change materials, if available. The only examples I've seen are diurnal systems; storing daytime heat, for subsequent release at night. The expense of the phase-change crystalline materials used in these applications limits the size of the system.

    There is, however, a cheap, readily-available material that phase-changes at "customizable" temperatures from 120-150 degrees (F).

    "Paraffin wax is an excellent material to store heat, having a specific heat capacity of 2.14–2.9 J/g/K (joule per gram per kelvin) and a heat of fusion of 200–220 J/g." [Wikipedia]

    The advantage is in the latent-heat phase, where BTUs can be extracted at a constant temperature, until all liquid paraffin has been converted back into a solid state.

    Are you aware of any large-scale paraffin heat-storage systems?

  25. mike eliason | | #25

    are presently being researched by IEA TF 42 - looking at phase change materials and thermochemical materials for thermal energy storage.

  26. Jan Juran | | #26

    Response to Foraker
    Hi Foraker: Gary Reysa at has built a large solar thermal system for heating his house in addition to his separate solar DHW system. It utilizes a 420 (originally 500) gallon tank, radiant floor distribution, and 240 SF of solar collectors. Gary estimates it saves 330 gallons of propane per annum. Check out "The Solar Shed" on his website. While cost effective using DIY components, the system is quite large and Gary still maintains a backup heating system for extended cloudy/cold periods.

  27. Carl Mezoff | | #27

    Seasonal heat storage in sand?
    A sand bed has a relatively low heat capacity, relatively low conductivity, little stratification tendency, and poses severe control issues. Because of these factors it is unlikely that sand offers much in the way of opportunities as a medium for seasonal thermal storage. Some of these scheme sound promising, but often the cold, cruel calculations are not very promising.

  28. Ed Whitaker | | #28

    Response to Thorston Clupp
    Many Thanks to Mr Clupp for his thoughts on solar thermal storage. I believe everyone should reread his thoughts very closely as this is the most thoughtful and accurate entry.

    Many opinions are based partially or completely on historical projects of the past 30 years. Nearly all of this body of work was developed without the use of proper and current modeling, modeling that tracks the moment by moment relationship between collection, storage and load. Thermal storage can be used to completely heat nearly any structure, the trick is to understand the cost /benefit equation to provide a cost effective solution whether it be a 30% or a 98% fossil fuel replacement. The analogy regarding cost is similar to using a sledge hammer to drive a finish nail - without understanding the cost/benefit, one could build a highly effective solution that would never have a payback ! I urge everyone to renew your search for thermal storage solutions.
    As Thorston correctly wrote "Point is, it is easy to dismiss ideas of sandboxes and storage tanks but I argue to think outside the box and look into the future and be willing to learn, move forward and not be stuck with the idea that we know it all.". We are all headed for higher fossil fuel costs and even more of our national wealth will find it's way into the pockets of a few. The time for action is NOW !

    As far as PV and heating goes it's currently a bad idea. PV and the grid combine very well since most utilities are daylight peaking and PV can relieve the peak generation requirements. We all have a 'plug-in load" and it makes perfect sense to use PV to offset this load. Heating is very different because we have options for heating unlike our "plug-in load" . If we use the grid as a "battery" for our optional items such as heating , we will require the utilities to be sized, both in distribution and generation for at least double the current loads of traditional (non green) homes during the non daylight hours. Just imagine what would happen if large scale adoption of such a strategy occured, but it never would because it's a poor strategy. PV heating will be a poor choice until personal battery storage is cost effective.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |