Soil and Perlite Cement mixes


In the process of designing a garage and possibly a house, I have spent a bit of time looking into varous forms of natural building construction techniques. The most natural appears to be building with earth. The techniques I've looked at are rammed earth, cast earth, and soil cement.


The site for the building(s) is a sandy hillside in eastern Colorado. A simple sedimentation test suggests that the soil is roughly 85% sand and 15% silt/clay. This dirt is extremely hard on hands, suggesting sharp angular surfaces on the sand grains.


Rammed Earth: A process by which soil is compacted in forms to create walls (or building blocks). This often involves mixing the dirt with a binding agent, usually portland cement. This can be an effective low cost building solution if you have the right kind of dirt, a strong back and enough time on your hands.


Cast Earth: A process by which soil is mixed with gypsum (plaster of paris), water and a retarder in large batches and poured into forms. This is marketed as a licensee based distribution with a patent pending on the formulation. Not a low cost solution, but worth looking at for ideas.


Soil Cement: similar to rammed earth, but using a slurry type mix rather than beating on moist dirt.


Perlite Cement: Perlite is a volcanic rock that when heated pops like popcorn. It's used in several industries including gardening and construction. It makes an excellent aggregate that can be added to cement or plaster to form a lightweight insulating cement or plaster. Note that the resulting cement will be considerably weaker than either a soil cement mix or a standard concrete due to the relative weakness of perlite.


Rastra: A note here on Rastra. This is a styrofoam/cement mix used in insulated concrete forms. It sells for roughly $4.00/sq ft. and is available in 8 to 14 inch widths. Looking at this got me interested in forming my own, which got me interested in the perlite options.


Curing


All cement and plaster mixes need to be properly cured before they attain their maximum strength. Testing of cast material usually takes place at 7 and 28 days after casting. Even though plaster will set relatively quickly, it will need to cure properly. Don't be fooled because you can touch it sooner than cement. Don't try to stress your casts too quickly or you will find yourself with some pieces and not samples.


The following table shows my prelimiary test samples, along with comments based on the first few day's curing.


Test samples



Sample

Binder

Soil

Perlite

Comments

1

1 - cement


8

Very soft, flakes easily, long set time

1A

1 - plaster


5

Soft, some flaking looks better than 1

2

1 cement


5

Light, but perlite floats in a wet mix

2A

1 plaster

2

3

Light, and firm

not as firm as cement binder

3

1 cement

1

4

Light, and firm

3A

1 plaster

7


Heavy, sets very slowly

4

1 cement

2

3

Light, and firm

4A

½ cement, ½ plaster

2

3

Similar to but more fragile than 4 or 2A

5

1 cement

7


Heavy, hard

sets well


A concept I'm looking at is a two part wall. The outside 6 inches would be a lightweight insulating cement or plaster (2, 2A 3,4 or 4A) with an inside 6 inches of heavy solid soil cement (5 or 3A). There could either be a 1 inch styrofoam insert between the two (and/)or they could be tied together with a wire framework if the lighter side insulates enough. The forming could be tricky since it would involve two pours for each wall setting. Also it would be difficult to pour both mixes at the same time as the heavier mix would displace the lighter, setting beneath it rather than along side it.


So far, it appears to me that for the most part, the cement mixes are preferrable to the plaster ones. They set slower at first, but seem to be more solid after a few days. I have concerns over how well the plaster binds the soil mixes, although it does fairly well with just perlite.