The design and construction of the kennel was conducted with repetitive feedbacks between computational design and practical production.
Initially, I conceived the idea of a mono-pitched roof design, but I have decided to think about an alternative design after finding out that the initial design would require an excessive use of steel bars and waterproofing layers which is not sustainable.
After considering several design options, I decided to build a dome-shaped kennel due to the fact that both walls and a roof can be built as one uniformed structure being structurally stable. The dome design required a careful plan for fabric formwork. Since the wall is tilted gradually from the bottom to the top creating a curvature, I have planned to have three phases of fabric-formwork construction. However, I have altered this plan to have two phases after realising that fabric formwork consumed a lot of time to fabricate.
I designed the shape of Bodrami house with Rhino, and laid out the 3-dimensional geometry of the bottom as a planar figure. Then I attached the printed planar figure to a 2-metre wide Polyethylene (PE) Sheet. (The geotextile, PE or PP fabrics, is actually widely used in civil engineering and fabric-formed concrete construction due to its durability. Interestingly, my father has been using them as weed barrier fabric for his farming, I was able to get them for free.)
Using a sewing machine, I sew the folded edges of the PE sheet, so that the fabric would become a hexagonal shape eventually. Since the cut PE fabric frayed, the edges were sewn too.
When it comes to designing fabric formwork, I measured the dimension of available timber, and make a 3d modelling of fabric formwork based the information. The 3d modelling process was very useful to carefully consider how to assemble and dismantle fabric formwork, and where timber pieces should be located and how many pieces should be installed to withstand the tensile force that is caused during the ramming course.
Since the fabric formwork is a temporary structure, it is also very important to plan an easy disassembly. The outcome of the careful consideration is the joint of timber scaffolding of fabric formwork. Instead of nailing directly the vertical timber scaffolding to the bottom wood, two small wooden pieces were attached with the vertical timber and they were nailed to the bottom wood, so that it was easy for me to dismantle them afterwards (See the image below).
The vertical timbers functioned not just controlling the bulging of fabrics during the ramming course but also embedding the curvature on the façade of rammed earth.
After the fabric formwork was completed, I laid it on the top of poured-earth-foundation adjusting the entrance towards the south. Before starting a ramming course, I nailed the edge of the formwork to the foundation in order to avoid any movement of the formwork during the ramming course.
In the upcoming post, I will share the story of the ramming course and the outcome of completed rammed earth.
Hi Hab,
At my first test, I have added a very small amount of plasticiser and water into earthen mixture, so it was not fluid and there was no problem of ramming the model. The key lies in the amount of water. As you mentioned, if plasticiser is added to poured earth or concrete, no ramming is required since the mixture are in a liquid state so gravity creates its form. These construction process are different with rammed earth of which the forms is influenced by the shape of a rammer and the power of builders.
I haven't posted yet the outcome of the doghouse, but I am going to show it in advance in my talk at the Clayfest…
Thnaks for your reply,
i have another question, if you're using plasticizer, lime and cement, how do you rammed ? plasticizer fludifying the mix concrete, it's more like normal concrete, that you're left for curing , no ramming or i'm missing something :) even rammed can be difficult in fabric formwork ( i saw your post on dog house with modified tamper :) )