Design iteration of Phase 2 rammed earth doghouse
I have resumed Phase 2 rammed earth doghouse construction. Before conducting a construction, Myeong-rang house was created with a consistent design iteration. There were two major factors to consider in the design. Firstly, it was the design and material of the roof. Since I delved into a sustainable roof design and construction using locally available materials, it took a significant time and research to find a solution. Also, based on the roof design, the shape of rammed earth wall was changed accordingly. Another concern was to create a smooth join between brick arch and upper rammed earth wall. Although the iterative design process requires a long time, research and consideration, it was an interesting process for me to come up with diverse design solutions, and choose what works the best.
For the design of the upper rammed earth wall, a hexagonal plan of the wall is morphed into a square plan. I intentionally made the morphological change in the structure because of two reasons – to control the bulging of the fabric by the tension caused by a twist, and to install a HyPar roof (I will explain further about the roof in the next blog post).
Fabric formwork design and fabrication
The fabric formwork of the Phase 2 was different from the Phase 1’s fabric formwork. The phase 1’s fabric formwork was made of one single fabric creating the doghouse’s opening. On the other hand, the Phase 2’s fabric formwork was created with two fabric layers. Interior fabric formwork was installed on the completed rammed earth wall, but the exterior was installed on bricks.
The interior fabric formwork was carefully designed and fabricated to make a smooth join with bricks. If it was rigid formwork, it would require more complicated fabrication and longer time to make it perfectly fit to the brick opening. Or it would be the best to avoid the demanding work. However, the fabric formwork gave sufficient scope to improvise the fitting during the installation process due to its flexibility, thereby it was an easy installation. Using the tension of the fabric and sewing the fabric, I made the fabric formwork fit tightly to both curved rammed earth walls and rectangular bricks. Through the exercise, I thought that the tension and flexibility of fabric, which is the quality of fabric formwork that rigid formwork doesn’t have, is a very much interesting field to explore further.
Another consideration of designing the interior fabric formwork was removal. As the upper rammed earth wall has a narrow opening at the top, the fabric formwork had to be built as simple as possible for its easy removal after completion. Therefore, I minimised the number of the timber frames for the fabric formwork, and designed the timber frames are pushing each other to minimise any excessive joints.
While building the exterior fabric formwork, I added extra vertical timber frames after realising 7 vertical frames that I initially planned in the digital modelling are not sufficient to control the fabric’s bulging.
Throughout the whole design process, 3d computational design was very much useful for me to design and plan the fabric formwork, check any expected problems, and estimate the outcome based on the modelling. However, even if physical models are developed based on digital models, there are ALWAYS differences between a digital world and a real construction. This is the reason why I do not rely on a computational design but use it as a tool. As developing my hands-on experiences, I always learn something new from building physical models and get new insights from how materials perform.
Ramming is the easiest part in the whole process. Compared to the iterative design process, ramming earth is straightforward although a physical body is really exhausted.
The interior fabric formwork was installed first and it was filled with lime-stabilised earthen mixture up to the top level of a brick layer. During the ramming course, I found that the fabric is greatly bulging because only 6 vertical timber frames are installed, and the lateral earth pressure was greater as the wall was sloped inward. As a result, the interior space has become much smaller than I expected. As fabric bulges far more greatly during the construction of slanted walls, I expect it is very much important to pre-consider designing a little bigger interior space considering bulged fabric and also to install more timber frames or something to control fabric’s bulging. For the doghouse, I intentionally added just 6 vertical timber frames for an easy removal after the completion.
After the rammed earth became cured for a week, I laid a single brick as a lintel. Before applying cement mortar, I laid mesh to control a tension force.
Then, I laid the exterior fabric formwork on the brick layer and drove nails into them in order for them not to move during the ramming course. The earthen mixture was prepared with the ratio – lime (1): soil (4) : sand (8) which is the optimum ratio that I found through a test. The first layer was stabilised with lime, and the second layer was unstabilised. The next layers were mixed with red powder. According to the amount of the coloured power, a diverse colour of mixtures was made from vivid red to burgundy.
As I did not successfully control the moisture content in the previous rammed earth construction by relying on a ball-dropping test, I carefully judged the moisture content of earthen mixture this time. I think the test is reliable for clayey soil, not for sandy soil. Because the soil I have got was very sandy, even if the mixture reaches an appropriate moisture content for building rammed earth, the earthen ball was completely scattered into pieces. (On the other hand, when I did the test with a clayey soil in Edinburgh, the test was more accurate.) As long as the moisture content reaches up to the level that an earthen mixture can be formed into a ball not broken into pieces when I hold them with two hands, I used the earthen mixture for ramming. However, when red power was added, the earthen mixture becomes drier, I had to add slightly more water.
Before I compressed the first layer, I fixed 4 fully threaded bolts around the edge of the fabric formwork. These bolts will be used for tying a roof structure.
During the ramming course, I compressed the earthen mixture with a rounded rammer first, and pressed the edges harder with fingers because these are not compacted properly because of the sloped angle of the wall.
At the next blog post, I am going to show the fabric formwork’s removal process and the outcome of the fabric formed rammed earth wall. Follow my Facebook Page or Instagram to keep updated.