Friday, October 10, 2008

Rammed Earth In Chad

Reader Mark David Heath sends this report, with colour commentary by REi4E:

This will offer information on a demonstration project of the use of Rammed Earth in Chad, Africa. It is believed that this is the first use of rammed earth in Chad, certainly in Southern Chad and in the recent past. No person, including persons with over 30 years of extensive construction experience in Chad, seeing the demonstration project described here, had ever seen any similar methods previously.

The use of earth in construction is the norm in Chad. Most buildings are built of earthen bricks, often fired with charcoal to increase strength and moisture resistance. (unfired bricks cost about 15 XFA/pc and fired bricks about 25 XFA/pc. They measure approximately 90mm x140mm x250mm or about 3.5” x 5.5” x 10”) The closest similar method, found to be already in the marketplace in Chad, was the use of a manually operated press to take a mixture of earth and Portland cement and press the mass into bricks under high pressure. This press was manufactured in Belgium and imported into Chad. While it produces an excellent product it required a high capital investment (more than $10,000) and delivers a finished product only slightly better, if at all better, than the well produced native fired-brick product. While it is a more uniformly shaped product than the native product, it is still used in hand-laid masonry walls and is dependent on the mortar and skill level of the mason for the overall quality of the final product.

Rammed earth, while perhaps new to Chad, is not a new technique or material. Rammed earth was used to construct ancient structures in Northern Africa, the Middle East, and Asia, with the Great Wall of China being the most dramatic demonstration of Rammed Earth. The ancient structures did not use Portland cement, given that Portland cement is a modern material having been created in the 1800’s. However, Rammed Earth can be well made with the correct soil blend, absent any Portland cement. For example, a known mix is the use of sandy soils with a binder of approximately 30% clay soils. The use of Portland cement has simply made the use of soil an overall easier technique since adding 5% to 10% Portland Cement of almost any non-vegetable soil will produce good Rammed Earth structures. Without the use of Portland cement a higher level of technical and sophisticated knowledge is required to ensure that the mix of sandy soil with the binder of clayey soils is correct to produce a strong and stable structure. In short, Portland cement just makes ‘getting the mix right’ a lot more simple and easy.

The underlying principle with Rammed Earth is similar to the underlying principle in conventional concrete. We are trying to get a mix of materials, of ever-smaller sizes, so that as we mix the materials together, the smaller particles can fit into the gaps between the larger particles. Many will be familiar with the demonstration of a bottle filled with small stones, all of a fairly similar, uniform, size. The bottle appears to be full and one could not get any more stones of the same size into the bottle. However, this same bottle that is so full that we cannot even fit one more stone into it, will receive a very large quantity of sand. We can add sand to the bottle, and by shaking or vibrating the bottle, we can get the sand to fill the voids between the stones. Again, with the bottle apparently now filled to capacity, with stones and sand, we can again add a large quantity of water to the bottle. The water will filter down through the bottle, filling the spaces between the particles of sand.

We are trying to do a similar thing with concrete. We are mixing gravel (small stones), with smaller gravel, with sand, and then with Portland cement and last, with water. The small stones fit within the larger ones, and the sand within the stones. Portland cement is finer than sand and fits within the sand. The addition of the water does two things: it activates the Portland cement, forming a paste that will soon harden, locking everything together, and the water is a surfactant or lubricant allowing the mass to be well blended and formed into the desired shape. In Rammed Earth we are doing the same with soil and clay, which is finer than common soil, or with soil and Portland cement. The clay, or the Portland cement, fills in the voids and makes a more dense mass. Both the clay and the Portland cement also act as binders; when wet they form a paste to coat everything in the mass and then they dry out to harden and lock everything together.

The ramming process in Rammed Earth accomplishes two tasks, one, it consolidates the material making it as dense and possible, eliminating all the air voids possible, so that the material is as well-locked together, physically, as is possible. Secondly, the ramming drives out as much water as possible so that there are not pockets of water that will later evaporate and leave voids and consequential weak spots in the structure. Such voids also allow for future water, such as rain, to return into the structure and soften the clay and allow the mixture to become soft and loose its’ strength, or its’ ability to resist the loads on it.

The accompanying videos show the ramming process used in Chad. In the US, for example, Rammed Earth is commonly carried out with the use of pneumatically powered tampers. These are very fast and powerful and produce excellent results. They do, however, require a power source – an air compressor. The air compressor requires a power source, electricity or hydrocarbon fuel. And all of it requires significant capital investment (Probably well over $30,000, all told). Then the operation requires both the fuel expense as well and the maintenance and repair costs. In Chad, electricity is not common, and it is expensive, as are hydrocarbon fuels. Large air compressors are expensive and rare, and pneumatically powered tampers are practically unknown. Instead, Chad has abundant and inexpensive labor (around $1/day) and an indigenous skill of grinding grains with mortars and pestles made of ironwood. Little girls can be seen grinding grains alongside their mothers as toddlers. By the time they are mature women they have muscles and skills at grinding that are ideally suited to ramming earthen structures. Further, it allows women to work in construction, which is not a common opportunity.

Rammed Earth construction has two primary needs: a good mixture of earth that will produce a dense and stable structure and a forming system that can resist the forces produced by the ramming. As discussed above, the mixture can be accomplished by the right mix of sandy soils with clay soils or, with sandy soils and Portland cement. The forms are another matter. The force produced by ramming is greater than the forces produced by pouring concrete into forms. Hence, common, conventional concrete forms are inadequate for use in Rammed Earth construction, although they can be reinforced to serve adequately.

For our experiment we found scrap steel ‘U’s. These are sheet steel, bent into the shape of a ‘U’, that were originally shipping containers used to ship and protect oil well equipment. These U’s were ideal for our Rammed Earth experiment. First because they were strong enough to resist the forces of the ramming and, second, they were scrap and the price was perfect- free. Last, we took a waste material and turned it into a new, useful product, with a long useful life, which will produce another product, the Rammed Earth structure, also of high quality and long life.

This first project was a wall, about 1.5m tall by almost 1km long, encompassing my employer's construction lay-down/staging yard. We needed some kind of fence or wall and I proposed Rammed Earth, first because my calculations showed that it would not cost any more than the proposed chain-link fence, second, because I hate chain-link fencing (I find it terribly ugly), and last, because if Rammed Earth could be shown to be economically viable it would offer a better and more appropriate construction system to Chad.

In our demonstration we used a soil material known as ‘laterite’, a reddish-brown material, with about 15% clay naturally occurring in it. We mixed 5% Portland cement with the laterite and enough water to pass the ‘ball test’ (see below). Before deciding on 5% being sufficient for our needs, we made test samples of 5%, 10% and 15% Portland cement. The video clip shows the results.

After just a few days we were able to bounce a sledge-hammer off all the samples and decided to go with the lowest-cost mixture. In the future we will attempt a demonstration of Rammed Earth using sand and clay soils. Sand is readily available in Chad and clay soils are commonly used for making the conventional adobe bricks. We made our mix by shovel because it was our decision that it was better to pay 9 men to mix by hand than to use a diesel-powered mixer, at approximately the same cost; better to pay local men to work than to pay foreign manufacturers and refiners.

Counting shovels full of material achieved the mixing ratios. Since we were seeking a 5% Portland cement to 95% laterite mix, we mixed 19 shovels of laterite with 1 shovel of Portland cement. This method of measuring is common worldwide in mixing masonry mortar, just as it is in Chad. The biggest change was to teach the greatly reduced water quantity needed for Rammed Earth vs. the higher quantity commonly used and known in mixing masonry mortar. The test for the correct amount of water is the previously referred to, “ball test”. Taking a sample of material in the palm of the hand and squeezing the material in the hand by closing the hand with muscle force performs this test. If the material will form a ‘ball’ and hold that form when the hand is opened, there is enough moisture. If it crumbles and falls apart when the hand is opened, more moisture is needed. To ensure there is not too much moisture, the ball is dropped from about waist height. The hand is closed on the ball, the hand inverted and then quickly opened, allowing the ball to free-fall to the ground, without a lot of spinning or rotating action. When the ball hits the ground, it should completely break apart. If it holds together, then there is too much moisture. To add moisture to the mixture, water is added with watering cans with ‘rose heads’, or, in other words, outlets with small holes, so that the water is better spread throughout the mix with less chance of mud pockets in the mix. If too much water is added, the mixture can be dried out by adding dry laterite and Portland cement in the same 5% / 95% ratios. A homogenously moist mixture is achieved by turning the pile of materials at least four times with shovels. This process and the ‘ball test’ are shown in the video clips.

We fabricated large U’s from the steel scrap and added screws fabricated from threaded rod to tighten the forms and hold them in position. This can be seen in the video clips. We then added the moistened laterite and Portland cement mix to the forms in layers of a maximum of 15 cm (6”) thickness. This maximum thickness of layers ensured thorough compacting of the material. The material was then compacted by the local women ramming the earth mixture with their pestle sticks. These sticks are made of ironwood and are quite hard. We found that the pestle sticks would grind off and become rounded after about a week, and that by taking a saw and trimming them off square again, we achieved significantly better compaction results. The women have been all their lives grinding grain with these pestle sticks and have well developed muscles and techniques.

Not the least of the techniques that assist in the ramming is chanting. The women chant in melodic rhythms and tell stories all day long, as they go. The chanting gives them a rhythm to pound to, increased camaraderie and reduces monotony with the tales they tell. Our experience showed us that few were the men that could pound all day long and keep up with the women, and none were able to do it for a week. The women would ram 8 hours per day, 6 days per week, for weeks. Half of the women were nursing mothers, and older siblings would bring the nursing children to the work site 4 or five times a day to nurse while the women were taking a break and/or eating lunch. They would nurse their babies before 7, as they were having breakfast of tea and cakes. Then they would ram until 10 and take a break of tea and cakes and nurse the babies. They would ram until noon when we would provide lunch and when they would again nurse the babies. They rammed until 14:00 when they had another tea break and nursed the babies, then finished ramming at 16:30 and went home to nurse the babies and fix dinner. When we first went to recruit the women, we were met with incredulity. The village chief and the husbands could not believe that we wanted to hire the women to "pound dirt". But, since we were paying them, they laughingly agreed.

The folks from the village came out to watch and spent the day laughing at the idea of "the crazy white man" to pay their women to pound soil into a wall. They were all very certain that it would fall down as soon as the forms came off. At the end of the day, when we stripped the forms, there was total silence. They were literally dumbfounded that the wall stood. After 4 days when we hit it with a sledge-hammer and the hammer just bounced off, there was no more laughing at the "crazy white man" and is "dirt wall". I am attaching a short video of about a dozen men trying to knock the wall over.
After they saw it would stand and was resilient, they then said that it could just be pushed over. So, after a couple of weeks, when it was good and hard, I let a dozen of them have a go at it. Then they said that the torrential rain of the rainy season would wash it away. By then, we had already had a chunk of the wall sitting baptized in a bucket of water for almost 2 months, so I knew that the rains would not be an issue. Now, three rainy seasons later, and no degradation to the walls, we have no more naysayers.

With this demonstration we have now introduced a high quality, faster, lower cost building system to Chad. The Rammed Earth uses less Portland cement than even traditional fired-clay bricks when laid with Portland cement mortar. It is faster than the brick method, and cost us about ½ of the market price for a brick wall. Further, it allowed women to work in construction. We were able to involve women without crossing traditional sexual boundaries by having the women use their ‘women’s tools’ of pestle sticks to do the ramming.

A hidden agenda for using the women is that they are simply better workers than the men. They come to work on time, they do not get drunk on the weekends. They do not give us trouble during the day and get into fights with one another. And, when we pay them, we know the money will all get to the house and not end up with much or most of it going to the local bartender or prostitute. The women use the money to feed and clothe their families and to pay for their children to go to school, since there is no free public education in Chad.

Our next steps will be to demonstrate Rammed Earth with clay soils and sand, since laterite is not widely available in Chad. We are also planning on demonstrating construction of a single-story building and, then, a two-storey building. These next steps will offer a broad enough demonstration to make the Rammed Earth system something broadly adaptable to Chad. We need to figure out an improved forming system from readily available materials, to complete the introduction of the system. We would also like to find a permanent and appropriate roofing system.

A quick note on insulation might be of interest. We rammed the walls about 200mm thick (8”) and that is just enough mass that the heat of the day cannot quite make it thorough. You can go to the walls in the heat of the afternoon and the Southern exposure will be very warm to the touch. But the Northern exposure, which has not seen the sun, will be obviously cool to the touch. We would like our future roofing system to compliment the excellent performance of the walls so that they are durable, appropriate, sustainable, inexpensive and well insulating.

With our first demonstration completed, we are well pleased with the results, and have received excellent reception by those who participated and those who have come to observe.


sokone said...

This is a great demonstration project. I am actually working on doing exactely the same type of demonstration in Senegal.
I would like to be able to contact Mr Mark David Eath for specific question.


Rammed Earth said...

Please end your email address to and it will be forwarded to Mr Heath.

Please keep us posted about your project!


Unknown said...

Wonderful. I am from The Gambia and I am currently doing my dissertation (University of Manchester) on Rammed Earth with the intention of introducing this building technique to my home country. Will be grateful to get in touch so I can learn from your experience and implement this in my country properly. My email address is