Building a snake kiln in a damp and confined area.

Challenges and solutions in building  a snake kiln to fire Nanban ware in a damp and small area.


In many countries space and positioning of a kiln present very little problem as people generally have ample room so both issues don’t come into play. In Japan it is often the case the land one is working with has been in the family for years and so isn’t easily expanded or changed. A different situation and the one that I found myself in is since land is fairly expensive the lots are small. The land I bought 10 years ago after 3 years of searching is 400 tsubo. One tsubo is 3.3 square meters so 1320 square meters or 14,203 square feet, about 1/3 of an acre. Of that about 1/3 is more or less unusable because of steep inclines. The remaining is divided into the area for my house with a garden and the remainder is my workshop. I have a noborigama and a workshop that take up a 10 meter by 10 meter square. I wanted to build a snake kiln and the only place available was a space 11 meters long by 4 meters wide. Taking into account clearances for stoking on one end and a natural barrier on the other side I decided on a design of 8 meters long by 120 cm. wide outside dimensions. The area inside the kiln that is available for firing is 6 meters from the front baffle wall to the back exit wall.


Cost of materials, excess of water in the sub-grade, limited working space, no access to motorized digging equipment because of cost and site terrain.

I didn’t really have much of a choice on situating the kiln for reasons laid out above so what I ended up with is a site that is very close to the foot of a tall wooded slope on one side and that has slopes on two other sides. The water table is extremely high. The water level in my well for our drinking water is only about 1 meter below the floor level of the finished floor and drainage has been a major consideration in designing this kiln. If I just set everything flat on the sub grade I would have a situation where after rains the water that had been adsorbed from the side slope would come out of the cut out areas of the kiln and literally pour water from sub-grade into the kiln area. One of the reasons I am building this kiln is my noborigama is situated in a similar area and is very damp. I lose sometimes up to 70% from breakage that looks like early temperature cracks. Although a watery kiln seems to be a romantic ideal of Japanese that fire snake kilns it is in practice the cause of major amounts of loss in ware that cracks before the firing is finished. The idea of a watery kiln goes back to the Tanegashima roots of Nanban ware. One of the main researchers and firers of modern Nanban, Fujio Koyama is said to have had sluices in his kiln to facilitate the introduction of water during firing. So the story goes. I have never seen proof of these sluices but everyone seems to want a damp kiln except those who fire Nanban.


My main concerns were to get a floor that would act as a moisture barrier, be strong enough to prevent any movement in the floor and to do this in as economically as possible.

It would be easy enough to excavate out an extra vertical and horizontal 50 cm. of earth  and put in heavy plastic to act as a barrier. Easy if I had access to a backhoe. I dug the kiln out by hand so I tried to keep all excess excavation to a minimum. I also wanted to avoid shiftage in any foundation I put in. This would be easy enough to accomplish if steel reinforced netting was used in the foundation layer. This being Japan most construction materials are prohibitively expensive. I could easily add 50 % to the cost of the kiln by doing things “properly”. In lieu of “proper” I have opted for using materials at hand and extra elbow grease.


I settled on the concept of a triple floor. This took care of my first two concerns of moisture barrier and strength. To do this economically I decided to use local material and concrete mix as a foundational material. I also found cheap used soft fire brick that I crushed to make a thick mortar to set the floor bricks on and further stop the wicking of water up through the floor. These two provided me with a cheap strong material that I could mix up in bulk and lay down as liberally as needed.

Design and grading.

I first considered the finished floor elevation of the kiln as it would relate to natural grade. My main concern was to minimize the amount of digging I would have to do and at the same time not create any need for fill areas. I did this by using an arbitrary “zero” elevation close to the top of where I knew the terminal point, i.e., the chimney, of the kiln would be.   I made sure the floor elevation I would need above that point would be easily available with minimal fill. I then checked that the natural slope from the zero point down hill was as close to what I would need. It wasn’t possible to avoid all excavation but I tried to get my zero point as close to an area as I could to minimize it. After deciding where I would put my “natural grade equals finished floor grade” I drove a small stake in to the ground to preserve that as a vertical control point. I actually used it as a horizontal control point too by making sure it was in line with the alignment I wanted for the long axis of the kiln. I then set a 4cm by 4cm. board running the length of the kiln on a set distance out from my control point on both sides of the kiln. I first set the side that was lower elevation wise, setting it so it was a few centimeters off natural grade but so it has a natural slope. That is to say it didn’t run in waves, the board basically started at zero and flowed naturally but as far as elevation goes arbitrarily down to the bottom of the kiln. I then set the other side runner the same distance from the center control stake trying to keep it as level as possible with the runner on the other side. I didn’t choose an arbitrary distance from the control stake. I set the runner so between the runners there was the width of the kiln including arch brick width plus an extra 15 centimeters on either side. Because I was hand digging each 10 cm. extra width was a big consideration. These two runners became my main vertical and horizontal control points. I took a considerable amount of time to get alignment correct. They weren’t exactly level across at all points but I tried to make them so as much as possible. I did set them very accurately on the long axis as I wanted to use them as control.

To excavate I used two more boards set perpendicular to my long runners. The lower of the two was set to be level with the upper. First I set the upper board flat on the two runners, leveling it if necessary. I then set the lower of the perpendiculars level with the upper, using two stakes to get the height I needed. I set the lower at a distance that was the width of the step in the floor plus about 5 cm. I then set a board between these two levels.  This board acted as a kind of screed that I would pull across to get the sub grade correct. Since this board wasn’t set at actual grade I attached verticals that dropped down to grade. I then proceeded to grade each level by moving my two perpendiculars. 

For the main moisture barrier I used a mixture of local clay type soil and concrete. I used a mixture of 5 clay/sand mix, 2 concrete and 2 water. I had to adjust the water depending on the moisture content of the clay. If I was mixing after a rain I left out some of the water, if I had dug out the clay a few days before I used it and it had dried out I used the full complement of water. I poured out each level of the kiln in this mixture to a thickness of roughly 4 cm.   I poured a level and then poured the next step down. This left the step face unfinished, that is dirt was visible beneath the upper step and the lower step. I found the best way to seal the dirt was to use the same mixture and apply it liberally to seal the dirt in. I tried to use brick as a facing material but found the concrete mix cracked if I did.  After finishing the base of local clay and concrete I used a mixture of crushed soft brick and regular mortar to mix up large batches of a fireproof type of material that then applied liberally to use as a setting for the bricks I used for the floor. I also added a lot of rice hulls to the mixture.   The hulls will burn out and leave dead air spaces that will act as a barrier to heat transfer. I then set the floor bricks. They are roughly 33cm. square by 13cm. thick. The floor of the kiln now consists of a 4 cm. thick concrete base, another base of firebrick mortar and sitting on that large bricks. 


Working in tight confines proved difficult. As I graded the steps I found I wanted to get the steps as close to finished as possible, especially on the long axis. I tried to get the step in the grade as close to the point where the brick would need to step up. By doing this I minimized my fill in or cut out work later on. The problem with trying to get things so close is the movement of the screed apparatus was restricted by the tightness of the two perpendicular boards it sat on. It would have been better to try to get a little more play in the screed board but by doing so I would have made more finish work later on.

The method of having two runners as control for both vertical and horizontal worked out well over all. As I moved the excavation down toward the firebox the depth I had to dig out got progressively deeper.  Having the screed board hanging on the perpendiculars and the actual finished floor elevation down 80 cm. or so caused some problems. I would start a new excavation level and have to dig down 50-80 cm. before I got to desired finished floor. I can think of no better way to control vertical but it was a little difficult to use my method.

The triple floor looks like it will keep water from seeping up into the kiln.


Tags: , , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: