Steve Harvey: "There is no pumice or scoria mine near me either, so I will be switching my fill to cellulose-vermiculite cement, I got this idea from my old high school art courses, we used to carve relief sculptures out of vermiculite cement, and I think this will make as good an insulation as any. I can always spray on vermiculite cement on the outside of the earthbags, it would probably provide a better scratch coat for lime plaster than cob. maybe mix in some poly fibre in this scratch coat as well."
Christopher Steen: "Sounds interesting. Would love to hear how tests and costs go. Ratio ideas? Maybe consider testing a denser weaved raschel mesh bags to let it dry if you go heavy on the cellulose (blown-in cellolose?). Or testing a 10% replacement of cement with borax to keep your fill more mold resistant. is a vermiculite-crete without cellulose fill to expensive? I like lightweight cretes, scoriacrete And pumicecrete and epscrete and papercrete have all found appropriate uses in my work but I have never used them in hard tamped applications. A lot of carving on the pumicecrete and epscrete though. If it doesn't test out for the impact and compressive tamping tests, and you want to see your experimental vermiculite-cellulose Crete though then you could always consider forming up for a pour in such a way that your roofing integrates with your formwork."
Steve: "I don't think those bags will work for me because the material will fall out the larger holes. Poly bags are fine and work well with lime plasters. Ratio mix will be a 50:50 cement to vermiculite, cellulose insulation will vary depending on desired performance. I don't think I will use borax, but may use some lime for preventing mold. Cost wise vermiculite is fairly cheap, cement is like 10 bucs for 50lb bag maybe cheaper in bulk, cellulose is cheap, and yes it will be blown in type. The mix should be on the dryer side, like a semi dry mortar, tamping should hydrate all material well enough to compact them and also leave enough air gaps inside for a insulation."
Christopher: "I'd like to suggest that you might try a dozen or more tests this winter before you go out by bulk quantities. I suggested *Denser weaved* raschel bags if you can find them, because I would not put papercrete in an earthbag--too long to dry. Lime in a papercrete plaster makes sense because there is
carbon dioxide exchange at those thicknesses. Borax replacing a 10% by volume of portland cement has been tested and shown in engineering papers to not reduce overall performance (compressive strengths mainly noted) of concretes by much. It seems to me that more borax (already included in blown in cellulose as a fire and mold retardant) would be more appropriate mold retardant when mixed with hydrophillic cement since lime won't carbonate at those thicknesses, and lime would be best used in more effective and effecient thicknesses and applications. However, concrete aggregates will produce a completely different animal than a 50:50 cement:vermiculite (dare I say near foamed cement mix at typical mixing rates and application), wherein your mix while insulative, easily molded (and pumpable at typical usage), I foresee this mix better suited to pouring and bulking up application to a ferrocement than a tamped bag fill. There is one caveat in my thoughts: if you were to have all your form work ready, and have enough hands on deck so that you could mix, fill, lay, and tamp all your bagwork in one day, wherein tamping upper rows does not excessively destroy the tender budding newly growing cement microcrystalline dendrite growths (the stuff that makes cement hard), then this could work with a wonderful result if mixed with a cement set retarder additive (don't skimp on the ratio due to cost) and mixed with a LARGE high torque tow mixer (this should probably be stationary for your logistics) to blend the cellulose and portland and borax and additive first then utilizing the vermiculite to dry up the mix enough to a point that when the bags are laid there isn't an excessive squish factor anywhere on the dome (mainly the bottom rows, around door and window openings, and the beginning of the real corbel rows to watch out for). Thankfully the small size of this dome is not structurally demanding or time consuming to bag up, but it is nice to be able to take time to check that everything is proper in relation to your compass and that all your window and door openings are true with a nice reveal and that all your roof cleats are where you want them. Please meditate on how tamping on uncured portland permantly damages possible dendrite growth, and that even tamping on fully cured rows of vermiculite-cellulose-crete will smash bags and dendrite growth and insulative values. If you are getting bulk yards of vermiculite (or cubic meters, trailer loads, dump truck, etc. from a mine or landscaping yard or whatever) that is cheaper than bagged vermiculite and is cost effective, then why not just carefully stack them and wrap a few layers of stucco netting before your stucco? Logistics, cost, environmental impact, not having decomposable organic matter in hydrophillic cement paste, insulation value all seem to favor that, even if it requires formwork inside of the dome. That form work could be temporary and reused outside as your roof framing, or if designed aesthetically, left in place to attach your roof framing to from the outside. Strength wise, a solid reinforced plaster skin on both sides of even shape-shifting vermiculite-bags, especially with a strong dome shape at your small size, is nothing to sneeze at structurally.
As far as your remark about "The mix should be on the dryer side, like a semi dry mortar, tamping should hydrate all material well enough to compact them and also leave enough air gaps inside for a insulation"--you understand the dry mix importance, but don't think about relying on tamping to hydrate or blend your binder. If you want your bag fill compacted as well as mechanically bound by binder it should all be mixed. As far as insulative air gaps once you mix in that rich (depending on your cellulose ratio) binder content and tamp it, don't think that your insulation value will be close to pure vermiculite or fluffed up cellulouse. Thankfully you have the thickness of earthbags working for you. Loose vermiculite will be your best r-value by far. Is it worth the cost and effort of mixing in portland just to put vermiculite and cellulose it in bags? To me it seems the bags are enough and best for a simple dome with a structural skin, and that mixing a funky cool insulative crete is best if you want to do an interesting shape like a groin vault...
All this isn't to say your 10' diameter dome won't be strong if you just do a row or few a day like I assume is your original plan thus far, but please know that tamping on yesterdays or last weeks work at those stated mix ratios will crush dendrite growth and insulation value? Or put another way, if you are going to take more than a day to tamp on bags with those types of insulative fill, why add portland inside if it robs your insulation value for questionable structural value? That same portland cement can be used much better as a structural plaster skin, wether its traditional stucco, ferrocement, or vermiculite-crete plaster.
Steve: "There are no issues with the cleat idea, in fact they will act as rebar between the bags since they grip into the bags with several large nails. Being located 1 km away from Lake Huron means heavy winds I want the roof to be well secured to the earthbag house. I believe the more cleats securing the roof to the dome the better. The roof is meant to just protect the plaster from excessive rain, but also not fly off with wind, which should not be as big of an issue if placing the earthbag behind the primary structure, as it will block the wind."
Christopher (EDITED for future readers wanting more info): "My first thought was difficulty in clean carpentry lines and possibly ending up with a funky askewed roof aesthetic. Without any plumbing inside or
shower generating excessive indoor humidity, you could also consider an elastomeric roof paint, synthetic stucco, or real ferrocement coating maybe topped with waterglass (.4 water : 1 portland : 2 sand) which when paired with an indoor clay content of earthen plaster/floor, would keep your bag fill very dry. i know, blasphemy on this forum to speak of synethic or non-breathable wall assemblies, but in your rainy and freeze thaw climate and with a hydrophillic vermiculite-crete exterior plaster, this could be a decent solution for someone if they chose an exposed plaster roof aesthetic. I am not against your roof idea at all, it was funky enough that I'm responding instead of clocking into work. Just giving my thoughts since you welcomed solicited advice."
Steve: "My plaster is water proof, however, in order to avoid having to reapply waterproofing agents like wax over the plaster every year, A roof will protect it from rain and snow. Absolutely no bandaid roof paint or synthetic anythings on this one."
Christopher: " If your exterior plaster is 50:50 vermiculite:cement, well it is pretty rich but not gonna be waterproof as is with the vermiculite aggregates. 2 sand: 1 portland (ferrocement) is considered pretty waterproof when water content is .4 to portland, but cement will always be hydrophillic. What happens with freeze thaw cycles. Wax is like an exterior bandaid, it melts and wears off, while something like waterglass is like getting stitches; a great waterproofer that's breathable and permanent (by far my preference). Quality elastomerics and other coatings like graco are just bomber in the durability department, and although not breathable can be a component in a well designed wall system like the above stated example with interior earthen plaster with all those clay platelets moderating indoor humidity and wall assembly moisture. For someone in a wetter climate wanting a dome (or someone wanting to cover an exterior dome, vault, wall, slab) an appropriate and quality coating is no more a band aid than a second roof system in order to push a dome comfortably into a wetter freeze-thaw climate than where domes originally excelled. For example, a quality ferrocement vault work topped with waterglass or elastomeric like graco should last way longer than the highest quality galvanized metal quonset (let alone shingles) in the face of many different destructive forces. Portland cement is synthetic, just like that paint on the metal roof, asphalt shingles, and misprint rice bags. It's about how well something is used."
1
?
In my case I am thinking - bathroom in one building (outhouse may be separate again) - cookhouse in another and a sleep house for sleep only... we'll see.
