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

Community and Q&A

Solar-pumping of slate roof

paddington_22 | Posted in General Questions on

Hello to the specialists !

My south-facing lounge & kitchen area has a 20-foot cathedral ceiling and PU foamed external insulation. The south-facing roof-slope is around 1000 square feet.

My question concerns the do’s and don’t of penetrating this ceiling in order to access the roof cavity and extract air from the 2-inch gap between the back of the slates and the insulation layer.

Normally as a basic DIY-er I would never even touch the roof, but in this case I feel myself getting tempted by the cost-benefit analysis of a large-scale transpired solar collector !

My mezzanine provides easy access and to boot has a serendipitously-situated roof light that makes things a lot easier: by removing the siding I can see inside the roof structure, including the area of interest (cavity) in the neighboring bay (height 4 ft x width 3 ft).

I propose to remove the existing strip-pine T&G, penetrate the underlying roof deck (= 22 mm particle board), then cut a pilot hole through the 70 mm of PU foam, and end up in the cavity. If all looks OK, I’d then enlarge the offtake port to full size.

As I see it, I have the choice of making a single opening the net size of the duct, or otherwise several openings, in which case I’d build a sealed offtake plenum over the full bay, with the aim of dropping the air pressure and getting an even flow.

The (horizontal) duct would run across my truss tie-beam, to my technical shaft that descends another 20 feet into my (concrete) crawlspace. A fan would blow air down the crawlspace (12 meters long) then out into the garden.

The crawl is already fitted with a whole-house fan that works fine (on fair days): encouraged by three winters of use, I now seek to augment this “solar hypocaust”, using massive roof air over the summer and tentatively get towards annualized geosolar storage…

In a few years we will be down-sizing and moving on; it would be nice to have an up-and-running green energy solution on the house, especially if it included an air-source heat pump capable of providing for our DHW needs (currently a quarter of our electric bill).

Thanks for reading and for any feedback !



(N. France)

GBA Prime

Join the leading community of building science experts

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


  1. Expert Member
    PETER Engle | | #1

    Many similar schemes were hatched in the early days of solar energy and few worked. Almost none are still in use. There are may reasons for this, but most come down to high cost of maintenance and poor performance. Your idea of seasonal storage in the crawl space soil won't work. If you ran some numbers using reasonable estimates of the mass and specific heat of the soil you are heating, you would find that, in order to carry any heat at all through the Winter, the starting temperatures in Fall would be very high. Even if you could achieve them, the house would dramatically overheat in Summer and Fall, and the low available heat remaining by the end of the season would mean that the house still needs its full complement of heating equipment. Even just using the roof cavity heat to boost winter performance on a cool, sunny day is somewhat problematic, as it requires bringing air of questionable quality (moisture, rodents, etc.) into the house. Even as a DIY, I suspect the costs will far exceed any benefits.

  2. paddington_22 | | #2

    Thanks, Peter, for your response.

    I already get the Crawl to warm up by a few degrees, using a whole-house fan. I wanted the roof to add a few more degrees, or more...

    A warmer crawl drains less heat from my basement...

    Just an aside, actually the roof air does not enter the "house", just the crawlspace, so nobody would be breathing it.

    If you could point me to some sources for those early failures, I'd appreciate it !

    For my part, I found some old stuff on the "Build It Solar" site, particularly:
    "A solar system that blows solar-heated air directly into a well-insulated Crawlspace or basement makes good use of low-grade solar heat".

    This was my main source of inspiration !

    Best regards,

  3. Expert Member


    One of the main proponents of solar thermal collection and storage was William Shurcliffe. He wrote the very influential book:

    Those houses, while ingenious, take the idea to it's most extreme. You are living in a solar collector. Shurcliife later repudiated this as a strategy when he realized that optimizing the building envelope produced the same results without the complications.

    I'd be a more worried about the possible effects on the roof than whether it proved effective or not. Introducing and inducing air-movement connected with the interior of the house may affect it in unintended or unexpected ways.

  4. paddington_22 | | #4

    Thanks, Malcom !

    I am starting to get the drift !

    I'd only be pulling modest and low-pressure quantities of air out of the roof.
    (in relation to the storms we get around here, this would be little effort on the roof...)

    The roof air does not enter the house, just the crawlspace, so the principle differs somewhat from the normal transpired solar facade.

    FWIW, here's some research on the subject that I came across:
    Transpired Solar Collector Installations in Wales and England

    Salient points:

    ''Building integrated solar thermal systems have been designed and developed around the world for over half a century.''

    ''In the UK TSCs are generally installed vertically on the façade and to date roof mounted systems do not exist commercially. The only non-vertical example is the TSC at (...) Jaguar Land Rover which is inclined on the façade at a 69° angle''
    To answer my own question, I guess to keep my sealed roof sealed, I just need to make sure that the extraction plenum is ...sealed !

    Have a great week-end,


    1. Expert Member
      MALCOLM TAYLOR | | #5


      Practicality aside, Shurcliff's solar thermal designs were ingenious and a lot of fun. His books are great reads.

  5. user-723121 | | #6

    The "envelope house" was an interesting concept in the day. Don't know if any were built, they required abundant south facing glass as I recall. Those were the early days of superinsulation, a building method that has stayed with us.

    1. paddington_22 | | #8

      Hi, Doug,

      Thanks for that !

      I admit to being a "fabric-first" person !
      My house was built in 1982, and it has external insulation, including a warm roof with PU foam. Loads of south-facing glass, two 10-inch RC floors...
      In last year's snow, this is the best roof around here, as far as I can see!
      I had a survey done, 2 yrs ago, the roofer said the roof is good for another 40 years,
      so I sure don't want to ruin it, LOL !



  6. walta100 | | #7

    From a return on investment point of view photovoltaic panels on your roof would likely be much better return if you can get a net metering agreement.

    Solar thermal is quaint idea from the 70s that has totally failed the real world.


  7. paddington_22 | | #9

    Hi, Walter,

    You are obviously entitled to your opinion about solar thermal.

    What is important is the envelope, first, and every time.
    I am a big fan of Larry Hartweg...

    The flip-side of my house was that the ground floor is on-grade and not insulated - this was obviously not a priority in 1982.

    I have been (slowly) renovating it by adding insulated floating floors, I still have some to go.
    I just want that solar-heated crawl... (-:

    Best regards,


    1. walta100 | | #13

      The way I see it is not just my opinion this set ideas has failed in the real life and the free market.

      The ideas are more or less unchanged 40 or 50 years now. If they could be made to work cost effectively they would have gone main stream and been incorporated into every new build long ago.

      Lots of very smart and determined tried very hard to make it work spending untold amounts of time and money.


  8. Expert Member
    NICK KEENAN | | #10

    I implore you: before making any holes in your house, do some engineering.

    First use the method in this article to calculate the heating load for your house:

    Next, estimate the contribution to heating that your proposed solar pump will make. The amount of heat (in BTU's) moved by a stream of air is given by the CFM times the temperature difference between the source and the end, times 1.08.

    Third, calculate the electricity usage for a fan to move that quantity of air.

    I think what you will find is that the quantity of heat is miniscule compared to heating needs of the house, but that the amount of electricity used is meaningful relative to the heat delivered.

    But I could be wrong, so do you own calculations.

  9. GBA Editor
    Martin Holladay | | #11

    I have written the following sentence many times on GBA, in response to a variety of proposed schemes: "The value of the heat that could be collected by this proposed scheme is too low to justify the high cost of the hardware required to collect it." The sentence applies here.

  10. Expert Member
    NICK KEENAN | | #12

    I want to analyze a little bit the idea of "annualized geosolar storage."

    For the sake of argument, let's say you can raise the temperature of the crawl space 10F above room temperature, to a depth of six inches. Let's postulate further that your crawl space is 1000 square feet, and the dirt in it has a density of 100 lbs per cubic foot and has the same heat capacity as concrete, about 0.5 BTU/lb/degree. So you've got 500 cubic feet, or 50,000 lbs of dirt. A 10-degree swing yields 250,000 BTU. A well-insulated house might use 25,000 BTU/hr on peak days, so that might be ten hours worth of heat. Depending on what you use for heat and what it costs where you are, that might be $10 worth of heat.

    To expand on what Martin just said, the reason that passive solar and thermal storage aren't popular isn't that nobody thought of it until you. And the reason the landscape is littered with failed passive solar projects isn't that the science isn't capable of predicting their performance. In fact, these systems perform exactly as the science would have predicted, had any analysis been done before they were built.

    Because Martin is exactly right. This approach just isn't cost-effective.

  11. paddington_22 | | #14

    Hello, DCContrarian and my other Advisors,

    Many thanks for all the feed_back,
    you have managed to send me back to my drawing board...

    I have not done as much math as this for a long time !

    Rest assured, I have NOT touched the roof (yet...)

    I DO hear you, and appreciate all the advice, I am currently working on the numbers as advised by DCC...

    Today is a very sunny day, it is not helping any (-:


    1. Expert Member
      NICK KEENAN | | #15

      This is the time of year when passive solar looks really good. The sun is starting to get really strong, it's cold out still but not as cold as a few weeks ago. A room with south-facing windows gets nice and toasty even with the heat off. I get it. It's crystal clear here today and even though it was in the 20's last night the sun is warming things right up.

  12. user-723121 | | #16

    Yesterday in MSP we had 43F in the afternoon, my furnace did not run for 5 hours. There is about 150 sf of south facing glass and my heat loss is 360 Btu per degree F. Free heat for part of the day. Robert Opaluch covers solar gain very nicely here on GBA.

    1. paddington_22 | | #17

      Hi, Doug,

      I had a couple of questions if you don't mind :

      "my heat loss is 360 Btu per degree F."

      Is this for the whole home, or per sq.ft, or per hour ?

      Also, what is the floor area of the house / or ratio of floor area to SF glass ?

      Please don't mind my asking, I am up to the neck at the moment, calculating my own heat-loss, (in metric...).

      Best regards,


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |