Burundi earthbag school nearly complete

Burundi earthbag school nearly complete

The following information has been graciously provided by Dr. Johnny Anderton of Eternally Solar / EarthBagBuild. Their building system is described below. It’s obviously another huge step forward in earthbag construction, right up there with Hyperadobe, Reinforced Earthbag and Superadobe.
Eternally Solar's earthbag building method use tubes sewn into three parts

Eternally Solar's earthbag building method use tubes sewn into three parts

“The way the bag is stitched divides it into 3 tubes – only the outer 2 are filled, creating 2 sausages joined by the central flat web (the third central ‘virtual’ tube). In building a wall, earth is poured into the central channel created by these 2 tubes and with the next layer, another bag is draped over this ‘mountain range’ and the central web compacted down to flatten the earth below it.

[Note by Owen: The center area creates a spline that strengthens the wall more than just bags on top of each other.]

• The bags are all equally filled, no chance of great variation in filled volume as with a large bag.
• Each ‘lift’ is approximately the same height
• Very easy to lay with no previous experience, as the bag essentially takes its shape while it is being filled.
• The bags are not heavy to move and can be prefilled before construction commences
• Plastic on plastic between rows is limited to the area along the length of the tubes, but the compacted ‘fillet’ of earth in the channel, locks the upper bag to the one below it. So no slippage and no barbed wire needed. Yet the actual amount of polyprop bag material is more or less the same.
• The length of the bag in its filled state in the wall @ 750mm is long enough to create a strong overlapping running bond between rows.
• The channel can be filled with cement and rebar, to create very strong lintels, ring beams or other horizontal stiffening members (we’ve tested lintel spans to 3 metres, by using the cement and rebar in 4 successive layers, way surpasses the strength required by local building codes).

“What is special about our EarthBags, what are they made of and how long will they last?

The bags are a unique patented design that allow for interlocking of horizontal layers greatly adding to the stability of the wall. This design also allows for accurate and equal filling, relatively low mass at approximately 13kg, uniform horizontal layering in the wall, easy construction of integrated lintels and ring beams and laying of conduits both vertically and horizontally.

All of these advantages are reasons the EarthBag is made the way it is. Using a standard polypropylene fertiliser bag, it is possible to construct a wall, but the ease of build, the end result and the ultimate strength are all very adversely affected. For example, the bag will be very heavy and therefore needs to be filled in place on the wall, it is difficult to fill every bag to the same degree, and to get the bags to lie symmetrically takes experience.

Furthermore, with a standard bag, one bag lies directly on the one below, plastic on plastic, with minimal lateral stability, and increased chance of slippage. Barbed wire then has to be used to ‘velcro’ the layers together. Using sand alone with these bags is not a good option, so stabilising with cement or using a clay-earth mixture is required.

As an example of how the system will slash the cost of transport in a major construction project, sufficient EarthBags to build the walls for two hundred and fifty 40m2 houses, can be transported in a standard 12 metre shipping container!”


Eternally Solar / Earthbag Building System Follow-up — 28 Comments

  1. Ok. I know a lot of people are going to say this defeats the cost benifits, but I’m more focused on eco friendly and structure streangth. So my plan has been to go hyper adobe but use barbed wire anyway to be a little more secure. I have plenty of dirt on my 60 acres and its good for this as far as I know at this point. So my question is. Are these bags mesh like hyper adobe er are or like the earth bags. I’m thinkin it would be awesome if there hyper adobe. Then instead of loose dirt I could run hyperbags down the middle. Use barbed wire and wile it would make fer wide walls would also be more then secure anough fer my northern california mountian location. Wile it would make for wided walls I do have the room if there is a structual benifit. Not to mention I’m thinkin that would be strong anough to alow fer the 8 ta 10 foot walls I want without worring about there streangth. I still have a lot to learn about this stuff so any advice would be helpfull. Thanks

    • Their bags are made with polypropylene tubes — the same material sand bags are made of. You could make your own with raschel mesh. Search our blog for raschel mesh and hyperadobe for full details.

      You only need single wyth walls. It’s too much work to build double walls and it’s not necessary.

      The best system for you is probably 18″ wide raschel mesh tubes. Search Discount Mesh for the lowest cost supplier.

      Keep reading. We have over 1,000 pages of free information.

  2. Ultimately people deciding to choose this or other earthbag versions will have to weigh costs. Adding concrete posts and beams for ‘confined masonry’ like this has higher costs (in most cases) than a pure earth and bags and barbed wire building. But the type of soil available will have great impact. In many places trucking soil in is prohibitively expensive, and if very sandy soil is all that is available, this may be the most cost-effective system. The three-part construction should give it more stability against overturning than a wall built with a single bag width and filled with sandy soils.

    South Africa, and many parts of the world, are in low seismic risk regions where reinforcement can be modest for good buildings. So a system like this or hyperadobe that requires less metal overall can be helpful.

    We need dynamic tests to explore earthbag’s performance in earthquakes. My guess is that an earthbag bearing wall will be more flexible than a confined masonry wall with a stiff reinforced concrete frame (whether it confines earthbags or harder masonry). But we need to see the way these perform under quake conditions.

    Dr. Anderton’s mention of issues with metal in coastal areas is insightful. But metal encased fully in a high clay content soil is protected against humidity by clay’s naturally low internal humidity. So my opinion is that either rebar or barbed wire in both concrete and sandy soil might be at risk in coastal areas with high salt content.

    We need more testing! Then eventually we can develop a cost estimate program which people can use to discover which types of earthbag are cheapest for the needed strength and resources and hazards of your region…

  3. This is a great idea. It would be interesting to see some seismic testing on this system with and without the soil being stabilized. I would think with a section of rebar thru the center and perhaps some vertical bars every 4 ft, these bags could be stacked to some impressive heights. Especially if the center material was stabilized.

  4. Thanks to Owen and John Anderton for getting us so much information on this interesting earthbag technique.
    This system makes construction lighter, and can take advantage of sandy soils well. Unless tested well, it would be best in low seismic hazard areas because it lacks barbed wire.
    I like the interlocking technique of the central earth between the upper and lower side tubes. Pouring loose sand into the side tubes should be easy to do, then it acts as a form to hold larger amounts of soil containing clay. Wall construction should be quicker than filling ordinary bags. With barbed wire this will be a sturdy wall.

    I am not sure how it will function with compacted loose fill in the center (like sand). That should require a structural skin of reinforced plaster mesh, but might be much more stable than sand in ordinary bags.
    There is some possibility that sand bags like this will function better for seismic regions because the wall can vibrate and dampen earthquake shocks. But it would absolutely need a strong mesh to hold the walls together since the barbed wire would be missing (or ineffective if included).

    It will require 40% more bags than walls of a normal 18 x 30″ bag (45 x 76 cm) because although the bags end up longer, the lifts are only a little more than 3″ instead of the 5″+ of normal earthbags. For non-hazardous areas this might be offset by the cost and time savings of not including barbed wire.
    I think I’d be sure to use a central fill containing clay and also interlay barbed wire in portions of walls receiving more stress. My guess would be this would be needed at areas usually vulnerable to quake damage in earthen buildings, including above windows and doors and near corners.
    We need more testing, but this is a great new technique to add to the growing earthbag family!

    • Hi Patti

      The design of the bags with the central channel allow for a lot of variation in approach to structural elements in the wall, which can be adapted depending on strength requirements.

      We have used unstabilised sand only in the wall in our pilot and other projects, in both tubes and channel. The resultant structures are extremely sound and have been tested to destruction for lateral loading, way surpassing international Agrement requirements. So I have no doubt that the sand in the channel could be sustituted with vermiculite or similar, for extra insulation, without compromising the structure. [Ed.: Johnny and I have talked this over and concluded that more testing needs to be done due to the low shear value of vermiculite and perlite.] The highly compacted sand (created through the weight of the wall above) together with the plaster skins, forms a dense laminate on either side of the lightweight core just as in a structural insulated panel (SIP).

      Having said that, in a cold Cape winter, we’ve found that the sand acts as an excellent insulator due to the myriad air pockets. Also has a thermal mass effect.

      In seismic areas, I would use columns and horizontal stiffeners. How do you build these? Where the plan specifies a column, you end the rows of bags the specified distance apart (say 300mm). This gap in the wall is where the vertical rebar comes up from the foundation. This vertical rebar is tied to horizontal rebar that is cast into concrete in the bag channels every 7 bag layers or so as you build the wall. These horizontal stiffeners can run into the bag wall just a short distance either side of the column, or can run right around the structure, creating a bond beam at all these levels.

      Once one has built say 1.5 m of wall, you shutter on both sides of the wall, and pour concrete into the space created by the shuttering and between the wall ends. You therefore end up with a cast column and horizontal beam. These can be created anywhere in the wall, whether in a straight of a wall run or in the corners of the structure. You thereby create a very lightweight reinforced concrete ‘lattice’ integrally in the wall. This is a much more engineered approach than using barbed wire and I believe is much stronger.

      It would be a good idea to use structural foam on the inside of the shuttering, so as to separate the column both thermally and structurally from the plaster skins. This means that the stiff skeleton of columns and beams is not directly connected to the flexible element (bags and sand fill) nor to the plaster skins. It is desirable to avoid having elements in the wall of different rigidity in direct contact. And this would allow the mass of the wall to vibrate at a different frequency to the skeleton, absorbing the vibrational energy.

      With the appropriate engineering input, I believe that a massively strong but flexible wall can be created, which will withstand significant seismic force. And 2 storey buildings or more could be created (We’ve not done anything more than single level thus far.

      Having said that, in our non seismic area, we have used unstabilised sand and no vertical or horizontal stiffeners (until ring beam or lintel height). So the wall to that upper level is purely the bags and sand, no barbed wire, steel, concrete or any other material. We don’t even used mesh in the plaster. This is how our test wall was built. Easy to build, nothing to corrode, and plenty strong. The corrosion issue concerns me, as there is always an element of residual moisture in earth walls, and any metal is bound to corrode eventually. Is there any research to show how the barbed wire stands up over the long term in coastal areas?

      With regards to cost, given that the shell of a house is a fairly small percentage of the completed cost including all fittings, I would rather spend a little more on the structure with only a small increase in the total cost of the project.


  5. Are the outer tubes structural enough to fill the center channel with an insulating material (expanding foam, perlite, scoria, what have you)? There would be minimal thermal bridging that way- probably just at the lintels/bond beams.

    • It doesn’t seem like loose insulation would be adequate. Let’s see if Dr. Anderton chimes in. I think it would be way stronger with the insulation stabilized in some way — say pumicecrete. Even a small amount of cement or lime or even clay mixed with the scoria would prevent the bags from shifting. (It wouldn’t have to be nearly as strong as concrete to do this.)

      • I wonder, however, if you could not do something similar with larger exterior tubes that WOULD be sound enough to have the central area be non-structural instulation. I’m imagining something along the lines of 3 bags (or tubes) sewn together along their edges, with the two outers filled with tamples clay bearing soil, and the inner filled with perlite or whatever. The bags could be slightly smaller than typically used for a single row wall, I would think. The problem would be tying the inner and outer clay walls together structurally, but it just might be possible that the inner bags themselves might suffice for that. There would, after all, be very many such connections between the walls.


    • The next blog post on this subject will shed more light. We’re going to examine this subject thoroughly because I liken it to Hyperadobe in terms of potential benefits. Stay tuned.

    • We have the polyprop tube material specially woven to the width we have found to work the best for our EarthBags.

      The flat width (tube flattened therefore 2 layers) is 420 mm.
      The bag is 900 mm long and is stitched lengthwise into 3 equal compartments, each therefore is 140 mm wide. As a tube, this results in a diameter of about 90mm, which when in a wall and slightly flattened means that the lift of each row is about 80mm.

      The end of the bag is folded to retain the contents, approx a 150 mm fold resulting in a usable filled bag that is about 750mm long, by 320-340 wide. The plastered wall ends up being around 380 -400 mm wide.

      The fill in the tubes is compacted if loose earth or clay mix, or damp sand. Only if dry sand is this not necessary. Easy to compact while in the filling device, simply with a broom handle or similar. No need to be too energetic as the fill is compacted while building the wall.

  6. this is really cool. combine this with the mesh/hyperadobe bags, and you have something really nice.

    How do they sew the bags? by hand?
    I guess you could use one of those stitching machines that they use to close the feed sacks.

    Our local feed sacks are the 40kg size, so bigger than the normal earthbag. I could see how this method could make it easier to use those bigger bags.

    • I’ve already written an upcoming blog post on how to combine the various earthbag building methods. However, I hadn’t thought of sewing mesh tubes/bags. I’ll write them for more details.

    • Bags are machine stitched commercially, but one of the advantages of this system was the possibility of job creation in Africa where poverty is such a big issue. The very smallest industrial machine preferably sewing a chain stitch which is self locking.

  7. Pingback: Eternally Solar / Earthbag Building System Follow-up (via Earthbag Building Blog) « Seeds-2-Sow LIVE Online

  8. This isn’t related to this product but i just came across this while researching dome homes… Not sure if you’ve seen this or not….

    • I’m well aware of monolithic domes, but the roving is new to me. Thanks for the heads up. I’ve been working on a basalt fiber and mesh blog post and this will tie in perfectly.

  9. I have several questions:

    1.) Do you have any real pictures of this product?

    2.) How much does this cost?

    3.) I can see how this can be used for straight walls, but what about domes or curved walls? If you were trying to make a curved wall, the inner track would need to be shorter and the outer track would need to be longer…. Is there anyway to detach one of the tubes from the center webbing and then reattach it?

    • We currently sell the bags for R4.80 (currently approx 0.68 US cents) here in South Africa. This may seem expensive relative to a simple misprint feedbag, but we are currently trying to claw back some of the funds we spent in developing the system, as we built a house for family who were living in an iron shack as a pilot project.

      Also, the quantity of bags we sell at present is still small, with larger numbers, the price could possibly halve.

      The other thing to consider is the savings that can be made through constructing one’s own lintels, ring beams, no barbed wire or chicken mesh required, etc.

      With respect to curved walls, the first structure we ever built with this system was a curved low garden wall. we did not really alter the level of fill in the tubes as we found that there was more than enough ‘give’ in the bag to allow a reasonable curve. Certainly, if one wanted a really tight bend, then different fill levels could make sense.

      • So that’s at least two more huge pluses for your system (in addition to no need for barbed wire, ability fill the tubes with sand, no need to fill bags on the wall, more rigid walls due to the center ‘spline’, etc.):
        – savings on bond beams and lintels (this is a big, big plus)
        – suitability for round and curved structures

        What’s the smallest practical radius?

    • Sorry, forgot to say that we have not used the system for domes, only for sloping retaining walls. More suited to circular structures with vertical walls than domes if one is using unstabilised sand as we do for most structures, but if clay-earth or stablised fill material, there is no reason why not. Again, the virtual elimination of slippage between courses would be a significant strength advantage, without the need for barbed wire.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.