the completed earth walls and pillars

Rammed Earth Walls

We had originally thought to use rammed earth blocks for the walls but concerns about their timely supply convinced us to think again.


We liked the earth but our earth building experience with the pump house (it was poured earth) convinced us that choice was not appealing.

Acquiring the forms and moving 80-some tonnes of earth, at times lifting some of it 3 metres in the air were both discouraging factors.

Getting Ready

We subcontracted locals John Oliver and Gabor Farkas, then owner-operators of Rammed Earth Constructions to build the walls who not only had all the gear, they had expertise as well.

The formwork is arriving and being piled up around the site.

the formwork for the rammed earht walls arrives


Work starts on the first section of the first wall
Work starts on the first section of the first wall

Moving up

With a team of 5, enthusiastically aided by Norman and I, the walls took a total of 13 days to complete.

We started at the northern corner and zigzagged our way around the walls - always working as far away from a freshly-done section as we could.

Some days we would finish 3 wall sections, other days fewer - it all depended on how much earth could we mix, move, form-up and ram before sundown.

The weather was very hot and dry so the walls started out drying rapidly and we had to keep gently spraying them to give the cement in them a chance to start forming crystals.



Layer by layer

Because the earth is rammed together it is only possible to do a layer at a time, even so you can see layers where the bottom is not as compressed as the top.

The earth is all shoveled in by hand to get an even distribution and a depth amenable to ramming solid.

A compressor-driven pogo-stick rammer is worked around the bottom of the space formed by the steel shutters compressing the contained earth.

the shutters are held together by bolts and 300mm spacers make them keep their proper distance apart.

John ramming a lower layer


John and James finishing the first wall section

The first section stands

The walls had been heightened to 3 metres because the shutters were 600mm high and it just didn't make sense to try and finish a wall half way up a shutter if we wanted anything resembling a flat top to the wall.

By the time the top of the wall was being rammed we were 3 metres up in the air.


Shoveling the earth into the forms became an adventure something akin to high-wire work in a circus.

Happily workplace health and safety did not ever choose to inspect us but scenarios like this were obviously begging to be improved.

A significant improvement

A pair of brickies trestles were summoned and a platform perched atop with a large shallow tray which served as an admirable raised reservoir for the earth.

This removed some of the adventure but speeded progress dramatically.

The raised earth reservoir  - a significant improvement


starting the form work for another section
starting the form work for another section

The cycle goes on

Each wall section started with getting the end shutters and corner forms in place and absolutely vertical.

These are made entirely from steel and were about as heavy as one person could lift - there was not much chance of fine-tuning the positioning alone.

Some sections would need 5 of these juggernauts precisely positioned before we could add the formwork to them.



The large mixer

The mixture used was a combination of 3 different earths, 5% cement and 10% moisture content.

Each load was about 6 bobcat buckets of earth, 3 bags of cement add water to moisten, mix well with bobcat.

The earth was on the dry side and the weather was very hot and dry so we were very grateful for the unlimited supply of springwater delivered at high pressure right on site.

a load is mixed by the bobcat
a load is mixed by the bobcat


The bobcat delivering
The bobcat delivering

And so it went

While the walls were being built Norman and I had our hands full preparing and locating the forms for the alcoves and electrical conduits.

These had to be rammed into the walls at the correct location and there was no going back if we missed getting a single one ready and into place before the earth had been rammed past that location.



and went

The RAC crew usually finished by 3pm but Norman and I would often be still at it after 9 at night - wire brushing the wall surfaces to eradicate the unsightly dents and lines left by the edges of the formwork.

on and on


a distant view of the work in progress

a distant view



of a work in progress



The alcoves debate

I was keen to add some character to the walls as we built them and had decided on alcoves - it seemed a good idea, make up some wooden shapes, put them in the formwork, ram the earth around them, remove them and, hopefully, a space would be left.

John and Gabor were skeptical and insisted it was infeasible, wouldn't work and was not to be considered by any sane builder.

the walls with alcove forming still in place
The first set of alcoves out the forms

the first set of alcoves

The alcoves are sane

However, being the man with the money persuaded them to “at least have a go”.

Time and engineering facilities were limited so we used more of the 12" X 2" oregon and a 45° circular saw to cut pieces which, joined together - would create 100mm deep forms with angled edges for easy removal.


Room for improvement

And this was the result - a set of alcoves in the first wall section.

This certainly looked good to me and a very welcome counterpoint to the bland featurelessness of some rammed earth walls.

The bottoms of the alcoves were quite uneven (from the ramming) and difficult to see what we could do about it other than pour cement in or lay a tile to fit.

We had, at the time, a heap of granite and sandstone pieces we had “rescued” from a landfill and thought they could be rammed in with the earth and form a nice flat bottom for the alcoves.

And so another task was added - cutting the bottoms of the alcoves with a grinder.

Another task - stone cutting


The alcoves mkII - with granite shelving
The alcoves mk II - with granite shelving

And the result

A set of alcoves with granite bottoms.

By this stage I had discovered that a semi-circular router bit made a lovely channel in the oregon formwork and........

it seemed obvious to me to use the faces of the formwork to make more “decorations” and so the alcoves mk II came into being - to add symbols to the surfaces.

Once again I was regaled with professional advice that it was infeasible, wouldn't work and was not to be considered by any sane builder.

Once again the financial argument won out and these were the result - I did treat the relief work to lashings of bondcrete as soon as I could get at them.

After 11 years they are showing no signs of deterioration.



Smothering the dust forever

The final step in completing the walls is a coat of earth-bonding emulsion which soaks into the surface down to about 3mm.

As its name suggests this glues the surface of the wall together and prevent it from forming dust.

It is a PVA compound - heavily diluted with water for spraying and simply sprayed on with a pump.

the coat of earth bonding emulsion
the coat of earth bonding emulsion


rain water covers the kitchen floor
rain water covers the kitchen floor

The wet arrives

The first storm of summer arrived and brought heavy rain - presaging a wet summer, little did we realize how wet.

This was the first rain since we had finished the dwarf walls and we soon discovered that we had made no allowance for drainage in the rooms with no plumbing.

 and most of the main room

and most of the main room



Gabor grinding a ousing for a lintel beam
Gabor grinding a housing for a lintel beam

The lintel beams

Each of the openings needed a header or lintel beam made from stable, strong material.

We had acquired some seasoned hardwood timbers which we felt would look in keeping with the massive thick earthen walls.

These had to be set into the corners of the walls and over the cyclone rods either side of each opening.

First the housings had to be ground and chiseled out of the walls.



Another “good” idea

It seemed sub optimal to let the lintel beam slop around in the housings so I had the bright idea of filling the housings with sand after the beam was in place.

A good seal could be made be rolling up some malthoid (the bituminised felt) and squeezing it into the gaps on either side.

No - seriously it was a good stable solution for about 5 years.

It seemed a good idea at the time
View from above ~ It seemed a good idea at the time


And then we noticed small piles of sand on some window sills - the seal had stopped working 100% - possibly with the help of some rats, but we have had to revisit these “seals”.

A lintel beam in place

The finished picture

One by one we went around the building measuring and gingerly inserting the lintel beams.

The earth walls - for all their solidity, were quite fragile enough to crack if we were to try to force a beam into the space against the wall's will.

Eventually all the beams were in place.



Another look at the finished product
Another look at the finished product


And now, a word or two about those pillars



return to top return to top of this page return to top