Rammed Earth Learning Curve

It started with an idea, then quite a bit of web research, a few books and finally several attempts to put all the theory into practice. But rammed earth is all about the soil, at least initially.

There are two distinct types of subsoil on the property - that soil in which organic matter is largely absent. One is red and has a high clay content. It is generally full of fine material and is not suitable without significant additives. The other, which is more yellow in colour, has a broader set of grain sizes, less clay and is far better suited to this application with a bit of stabilisation.

In an attempt to get a better understanding of rammed earth, we built a very simple, rough and ready wall using a 'Finnbuilder'. Using 8% Portland Cement, I was surprised how solid it was. I accept it looks pathetic, but structurally it was surprisingly strong.

The next attempt was using a small custom-built form that was 750mm long, 1250mm high and 300mm wide. This was a very basic system made from components exclusively available from the local hardware store. We thought the first finished wall using this formwork was stunning. Each rammed layer was distinct and the idea of this finish over a 6m wall was a dream. Again 8% Portland Cement was used and the wall was solid. As a second test, a smaller wall was built without any cement stabilisation. The soil used in these tests was not ideal for this application and when trying to hand compact them, the soil was tacky and stuck to the compactor resulting in a fairly ineffective tool that needed cleaning after every layer. It was messy, time consuming and frustrating. But at the time it did look good.

Both of these walls have since been through a very wet rainy season without any protection. The stabilised wall is showing slight signs of weathering but otherwise retains its shape, looks and hardness. The unstabilised wall has developed dramatic cracks (it has a high clay content) and is probably less effective now than a newly laid mud wall. It breaks off in clumps and is soft to touch. Also, termites would decimate it if it came to their attention and termites are prevalent on this site.

After this very early attempt at building the wall, a soil analysis was performed. Fortunately as part of the process, they did a site visit and found 'better' soil. Better was an understatement as it performed stunningly when compacted. It compacted to a clean, solid & even surface and did not at all attach to the compactor. Also the colour was completely different and made our original wall (that we thought was stunning) look rather ordinary. In fact it looked pitted and worn in comparison. As we did not have enough to fill the form, we began mixing it with some of the red soil and the effect was marvellous without detracting from the ramming properties at all. By mixing colours will allow a few interesting effects to be included in the finish of the final house.

An issue that has since developed is that the wall crumbles! By rubbing it at the corners or sharp edges causes a rapid degradation of the wall - it literally crumbles off in a fine powder. Apparently this is due to oxidation and the best way to deal with this is to seal the wall. Traditionally this was done with boiled linseed oil - I read that somewhere, and the next wall that we built using the imported soil (see Getting Started) was sealed using a modern commercial sealant that addressed this problem. The water just pools on the top surface and runs off without any hint of being absorbed. The only place where water ingress is evident is at the base (as can be seen in the photo on the left) and it is being sucked up from beneath. However, in the final build, the walls will be resting on a stem wall that will keep the rammed earth wall clear of any pools of water or damp surfaces. Also a damp course layer will assist in this regard.

One thing that does become very evident is that the quality of each wall seems to get significantly better than the previous one. Clearly this method of building needs a bit of experience added in to achieve the best results.

For the techies: hydrated lime otherwise known as road lime - Ca(OH)2 has been recommended for stabilisation instead of cement. All the test walls that have been built to date have used 8% Portland cement for stabilisation. Instead a 4% mix of hydrated lime has been recommended as the ICL (initial consumption of lime) from the lab tests was 2%.

The next ramming project is to build a final test wall using hydrated lime and then onto the real thing. The equipment is on order to sieve, mix and convey the soil in bulk quantities. So too is the wood for the formwork. The pipe clamps have arrived and the project is developing a head of steam. But before all of this, I need to sink a borehole and get water on the site. Man needs water before labour. I also have a food forest I need to finalise the planning and planting of...and to sort the water to irrigate the trees.