Umpium Report Notes
In building the first kiln many decisions were based on an already existing infrastructure. When we arrived the kiln site was already developed with a level area and a thatch roofed-open sided building housing a drum style smoking kiln. The site was built into a red clay hill, a 10' almost vertical bank was at the back of the building (kilnshed). There is a small generator building beside the site, and one dwelling. Above the site is the second reservoir.
The YDC staff (Ou-Moun especially) had already made a run of brick using the Grey Ou-Moun fish pond clay. The brick were of the small size 6"x2.5"x1.25" and about 300 were dry and ready. Having received a sample of the both the red clay which made up the hill the kiln would be standing on, and also the white gray clay the bricks were made of we had a good idea of the temperature range of the two clays. We decided that for a preliminary Bisque temperature kiln it would be safe to use the Grey clay in the high temperature areas such as the firebox and the chimney base and the red clay for the kiln chamber and the rest of the chimney.
We wanted to get our project underway as soon as possible as our time here was limited, we decided on a design that would use all local materials. If this kiln is washed away by the rain it can be rebuilt fairly easily.
The first day we arrived at the camp we started excavating the embankment behind the kilnshed in order to allow a chimney base to be built out from under the thatch roof (fire hazard). After the dimensions of the kiln were roughly decided upon we drew lines on the ground and leveled proportional areas for the firebox, chamber, and chimney base. We wove a bamboo framework to support the arch of the firebox and built it from Ou-Moun's gray brick, and used mortar made from the red clay we excavated from the hill. We brought in a few bags of rice husk and rice husk ash, later to find the husk intermittently available at the camp. Sand was dug from a place below the camp on a road and delivered by motor-trike.
We began to excavate more clay from the bank behind the site for use in making the special cone bricks for the kiln chamber. We decided to use the cone brick design as it was relatively easy to use them in forming an arch shape for the kiln, they were hollow for faster drying potential, and they were hand formed (we had lots of hands). The excavated clay was screened in a 1/2" mesh hardware cloth then mixed in 6clay 3sand 1rice husk proportions to achieve a light semi plastic body. We had a lot of help from friends and onlookers and were able to make over 300 bricks in two days.
We mixed a mortar of 6clay 4sand for the mortar and began constructing the chamber and chimney bases simultaneously, as we went up the bamboo door arch was installed and finally the key bricks were added.
The next day the kiln was lightly scraped to trim off the lumps and the chimney was built to 5' in height. We then lightly chinked the outside of the kiln, plugging all of the holes we could see, and filled the door using extra cones and a mortar mixture of 3clay 3sand 1husk. We then lit a small drying fire in the firebox, burning extruded coal for fuel. The fire was maintained for four or five days.
Since the first bisque firing we have made certain improvements to the kiln structure, which have not been tested. When we bisque fired the first time the chimney was only partially completed. Using the cone type brick we had we were not able to make a proper chimney of any height, as they would not be stacked in an interlocking fashion. Since then we tore the old chimney down to the ground, extended the base to allow for a small ventury area (originally intend for a damper although we did not have a damper shelf to spare) and rebuilt the base with a smaller internal diameter. New, near standard size (3"x6"x4") bricks were formed and dried, and the chimney was rebuilt to about 8í in height. The chimney leans on the hill for added stability.
Inside the chimney, on the floor and in the ventury a layer of bricks were added to further reduce the internal diameter. These bricks are only dry set and may be removed or added to depending on the necessary regulations.
One of the nice things about this particular kiln is that it is built in a modular fashion allowing for independent improvements to be made. Both the firebox and chimney have been built over-sized for the chamber for good reason, itís easier to reduce the size of each with addition of a few bricks than to tear down and rebuild. Also the kiln chamber may be torn down and built larger if required without disturbing the chimney or fire-box areas.
The fuel system is designed for a variety of fuels. The large firebox with removable grate allows for he burning of Bamboo, Wood, Rice Straw, Coal, Charcoal, Coconut Husk or Coconut Shell. Currently the cheapest and most readily available fuel seems to be coconut husks. The first part of the firing is achieved with solid fuel; the second part is adapted for burning waste oil. Waste oil is found cheaply and in appropriately large supply in Mae sot, we found that it was more likely to be purchased from small garages, as the larger stations have established contracts for its removal. The waste oil is stored above the kiln-site in a 200litre drum; a plastic hose runs in the rafters above the kiln and down to a black iron pipe resting on the firebox. A hole is located in the top of the firebox where a funnel is inserted. A similar gravity-fed system brings a water line to the same location, where two gate valves allow for the regulation of the water and oil into the funnel. Below the funnel an angle iron ladder burner is placed into the firebox. The fuel drips onto the top of the ladder and is burned off as it drips onto the lower rungs. The optimum water/oil mix seems to be 1:10 depending on the temperature in the firebox. (See section on firing).
Firing #1 (Bisque temp 850degC):
The process of firing begins with the preparation of the kiln setting. There are a number of ways to regulate the temperature dispersal within the kiln chamber. The bag wall at the exit of the firebox may be stacked differently allowing more or less flame to enter the lower part of the chamber; likewise the height should be regulated depending on the results of previous firings. The optimum setting (Before the new chimney was added) seemed to be about 12" tall, and openly stacked with 2" air spaces between bricks.
A second consideration is the opening to the chimney flue. Bricks may be stacked or removed from here to regulate the velocity of exhaust. This portion has not been tested since the addition of the new chimney.
The wares are stacked into the kiln on carbide shelves and the door is then filled with cone bricks and mudded over with a loose mixture of sand, husk and clay. (Extra should be made for lagging while the firing is in progress).
The most important part of the firing is the beginning where the wares are weakest and most susceptible to thermal shock. This being the case a slow warming fire is built from scraps of small pieces of dry coconut husks. The chimney temperature is tested often by holding ones hand over the top. The fire should be kept quite small (such as a cooking fire) until moisture is significantly reduced at the chimney (hand test). This fire may burn for as little as 2hours and as long as an entire day (I noticed that the base and floor of the kiln could get quite moist due to groundwater absorption). Ideally the entire firing process should begin in the early morning and end around 9pm.
The next temperature range up to 400degrees should happen at a rate not exceeding a steady 100degC per hour. During this time the firebox is partially blocked off with bricks somewhat restricting the draft. Chopped greenish coconut husks are added as fuel.
The firing is begun with the damper open and now may be regulated depending on the amount of draft in the firebox area.
After 400degC is reached the temperature may rise at a faster rate.
When the firebox has reached a dull red heat the oil burner may be installed into the firebox, and the firebox entrance blocked off even more so as to allow only small openings for the coconut husk to be added. It has been suggested that a cover be made to replace the brick, which would help reduce the draft even more. After the burner is heated for a while small test drips of oil may be added.
To monitor the efficient burning of the oil you can watch the chimney for smoke, if the fuel/water mix is working efficiently there should be little change in the exhaust. Billows of black smoke endanger air quality and may endanger the future of the program due to unhappy neighbors. Another way to assess the burner is to watch for oil dripping off the lowest rung of the ladder. While burning coal we found that oil lost from the end of the ladder tended to soak into the ash and pool at the bottom of the firebox. This oil continued to smolder days after the firing until we cleaned it out. The ladder should produce a steady crackling fan of flames. A steady stoking of coconut husk during the oil firing stage ensures the proper combustion temperatures. The oil burner did not appear to work well under its own heat.
Continue increasing the temperature until the wares all reach a dull red hue (or 850degC) at this point it up to you to decide if you want to higher temperatures. At the higher temperatures the draft is monitored more closely by watching smoke and listening to the fire, the damper settings (a carbide shelf over the chimney) are adjusted accordingly. Testing the temperature of the kiln is done in a number of ways. A digital pyrometer is a nice and expensive option, but the use of spy-holes (to see the color of the wares) and cones are invaluable resources. If you are firing to higher temperatures you may want to consider making test rings which may be extracted from the kiln during the firing, then cooled and examined to determine the ripeness of the wares.
Once the desired temperature is achieved the firebox is blocked off and the chimney damper is closed. The next morning (8hrslater) the damper may be opened and tested for heat, if its not too hot the firebox is opened, and then eventually the door is opened. Voila.
The kiln is built from two different clays. The high temperature areas from Ou-Moun light bodied clay, which when tested in a gas kiln, was able to withstand temperatures exceeding 1300degC. The lower temperature chamber and chimney from red kiln clay, which when tested to 1300deg, had developed a glaze and had appeared a little over-fired. This said the kiln would likely be able to withstand temperatures to 1150deg without problems, above this temperature hot spots may begin to glaze over, not a problem as long as nothing drips onto the wares, or collapses under its own weight.
The construction of a ceramic-fiber kiln was essential in being able to test the clay and glaze materials in a controlled high temperature environment.One of the continuing criteria in these projects was the use of materials available within the area of the camp, materials used in this kiln came mostly from a town two hours drive from Mae Sod. Since this kiln has a relatively long life it would be possible to bring in the materials once, and not risk further hassles at the camp borders. Using special materials and fuels we converted a 16gage 200litre steel oil drum into a oil/propane kiln capable of withstanding firings of over 1300deg centigrade. With this capability we were able to assess the temperature range of various local clays.
The ceramic fiber, propane burner, cup anchor assembly, and sodium silicate cement were all purchased on our trip to Lampang. Second hand carbide shelves and K32 hard firebrick were purchased for low cost (we only bought a few and they were well used, and brittle) (the bricks were free) at a large slip casting ceramics factory on the same trip.
The inside of the drum was first burned out with newspaper and twigs, then lightly sanded to ensure a clean surface bonding for the cement. Markings were drawn onto the barrel for the placement of anchors, burner port, thermocouple port, spy-hole, and chimney vent. The barrel was taken to a Karen machine shop for cutting. The sodium silicate cement was painted onto the inside surface of the drum and the fiber cut and fit as a lining, the cup block anchors were fitted into place and the kiln was ready to go.
The burner assembly for the first firing consisted of one 15kg propane tank, one high pressure regulator (screw down type) and a direct line to the burner. It was suggested that there would be a problem with freeze-up as we would be evacuating the small tank at quite a high velocity. We purchased a second tank of the same size, another regulator, and a T joint to run the tanks in tandem.
To warm the kiln we formed a tunnel of firebrick about 12" long to the burner port. The burner was ignited at its lowest setting at the entrance of this tunnel. After a couple of hours we started to increase the burner pressure and eventually removed the tunnel altogether, placing the burner at an angle so as to spiral the flames around the kiln chamber.
We did not use the tandem setup in the first firing, rather we switched tanks at 850degC when the first tank was half empty. There was significant condensation on the outside of the tank at this time and the kiln was not able to rise in temperature more than 1deg per minute. On the second tank the rise was a steady 1degC/Min until 933degC when it stalled and the burner went out. The pressure was lowered and the temperature increase became 1decC/5min the kiln stalled and we were not able to exceed 970degC.There was frost on the outside of the tank.
Upon unloading the kiln the next morning two solutions became apparent.
Primarily the settings in the kiln were arranged so that a large shelf was placed at the bottom of the kiln with three large firebricks near the sides of the chamber as stilts. These three bottom firebrick were surface glazed in such a way that would suggest that the temperature in this section of the kiln was higher than 1100degC. (The thermocouple was near the middle of the kiln). The firebrick being near the sides of the chamber, as well as the large size of the bottom shelf inhibited the swirling pattern of the flame thereby shortening the flame path and creating some back-pressure. The two shelves above the firebox level (where the wares were) were broken 3/4 sections allowing for a lower pressure area out to the chimney. Perhaps the kiln was hot in the firebox, but as the heat came around the first shelf it was whisked away out the exit vent.
Secondly, we had discussed the high costs of propane in the area and the possibility of supplementing the firing with an oil drip system similar to that of the bisque kiln. It is difficult to get a large propane tank because it would have to be taken out of the camp to be exchanged, the smaller tanks do not have to leave the camp confines.
On our second firing the following changes were made:
The firing went as before, we had trouble gaining temperature above 800degC. This firing was a bisque firing, higher temperature was not needed. The oil drip seemed to work OK at this temp; it was difficult to regulate the oil as it had a long distance to travel from the valve to the nozzle.
This time we had bisque wares to work with and we tried to gain 1100deg. We only achieved 1000degC and with much fuss. The oil drip system had not worked as planned due to two obvious problems. The first was a buildup of carbon in the nozzle, which was so hard we had to drive a nail into it to get it to flow again. The oil was not being burned completely due to lack of oxygen and lack of proper dispersion; this caused a lot of smoke and bad air.
The tank arrangement worked well at first; we used two partial tanks to achieve 850degC or so (a quarter full and a third full tank). We switched to a full tank in the hopes that its volume would overcome the pressure evacuation problems we were experiencing on the last firings. Unfortunately we found that pressure was the one thing we were lacking the entire way through. Once the oil drip system had failed we had to use burner pressure to increase the temperature, with obvious effect. The temperature would rise incrementally as we increased the burner pressure. After placing the tank in a tub of water to dissuade tank freezeup we eventually reached 1000degC and came to a few important conclusions.
Our recommendation was to change the layout of the studio slightly in order to economize the useful floor space. Kick-wheels were moved against the back wall and shelves were raised to head level. It would be useful eventually, to have enough shelf space to hold at least a full kiln-load of green wares at a time. Space is also needed for the storage of bisque wares ready for glazing and waiting for high temperature firing. A worktable would move hand-building work off the floor and onto a more accessible space.
Wedging Bench: In order to facilitate the processing of clays into useable batches we decided to design a small plaster filled wedging table. The staff at the YDC were able to put together a 2íx3í(x32"Height) frame of teakwood, the top of which was a 4" deep tray. After this table was secured to the wall we mixed two batches of pottery plaster (From Lampang) and filled the tray to the brim. After an hour Burt smoothes out the surface with a stainless burnishing scraper and we waited 24hours before trying it out. Although we started using the table right away it will not be completely dry for a few weeks.
Slab Cutter: Due to the slow process of learning to use a kick-wheel it may be necessary to focus on hand-building techniques as a way to produce wares which students may be proud of. In order to make slabs of consistent thickness we suggested a simple design for an adjustable slab cutting tool. This tool cuts slabs of different thickness, which have little plastic memory and have less tendency to crack or warp the way rolled slabs do. Slabs of clay may be used free form or with moulds to achieve consistent shapes and effects.
Scrapers: Sections of strapping steel came from the scrap yard for use in making different types of hand tools. Large scrapers were made for the fine-tuning of bricks in the bisque kiln, and also used in trimming the lumps of mortar from the inner and outer surfaces. Other scrapers were made for cleaning off the workbench. A knife was made for the cutting of asbestos board into sections of batt. Various footing and turning tools were made for both general and custom forming purposes. This steel is plentiful in the area, and is virtually free. It was mentioned that there might be a source of this strapping from within the camp.
New Cutting Line: A braided nylon cutting line may replace the braided steel line now in use in the studio. The problem with the steel line is that as it ages filaments of the steel will break and may cause injury and an inconstant cutting edge.
Batts: On our trip to Sukothai we visited a large pottery facility making flowerpots. The potters were using a drying system, which only used bats for moving the wares during the leather soft stage. The batts they were using were made of asbestos board, as it is both cheap and sufficiently absorbent so as to promote an even drying of wares. One of the negative effects of asbestos is the release of toxic micro-fibers mostly while it is being cut and some while it is in use. The cutting was done wearing masks; a further precaution would be to dampen down the boards while in use and being cut. Discussion about the use of this material was short; it was accepted due to its low-cost and relative danger as compared to other problems in the camp.
Drying Bowls: Low fired drying bowls are used in the quick drying of slips and cast offs of used clays in the studio. The clay scraps are collected in buckets of water around the studio and then put into the bowls until ready for wedging and re-use. The clay used to make these bowls can be of low plasticity and low temperature range. The bowls are hand formed into thick shapes, and then turned at the leather hard stage. The inside surface is then burnished with a scraper or plastic bag in order to resist adhesion of the clays.
Spray Bottles: Spray bottles will be used to wet down the cement floor in the studio, as well the fine mist can help pull clay particles down from the air. While sweeping this can significantly reduce the potentially harmful dust from blowing around.
There were 9 glaze tests done, two control samples, and one curious black powder. All samples were fired to 1000degC (The upper limit of the fiber test kiln when fired with 15kg propane tanks). If low-fired glazes are going to be developed they can be made from simple three part recipes of Frit, Clay, and Silica. (60-80% Frit, 5-10% Clay, 10-30% Silica (Flint)).
UMP SERIES #1
Frit 50% / Muslim Mkt. 17.5% / Flint 32.5% =100%
A= +5% Silica
B= +10% Slica
C= +15% Silica
Result: All four samples were dark gray, underfired.
UMP SERIES #2
Frit 35% / Kiln Clay 25% / Muslim Mkt. 25% / Coconut ash 15%
A= +5% Silica
B= +10% Silica
C= +15% Silica
Result: All four samples were gray, underfired.
UMP SERIES #3
Frit 60% / Muslim Mkt 10% / Silica 30%
A= +5% Coconut Ash
B= +10% Coconut Ash
C= +15% Coconut Ash
D= +20% Coconut Ash
Result: Five samples vary in lightness, A being the lightest, D is dark mottled gray, semi matt. Melted. Well adhered to body.
Frit 40% / Coconut Ash 30% / Flint 20% / Muslim Mkt. 10%
Result: Dark gray semi matt.
UMP SERIES #5
Frit 40% / Coconut Ash 6.5% / Feldspar 34.5% / Silica 16%
A= +5% Frit
B= +10% Frit
C= +15% Frit
Result: All four samples are gray, some melt, semi matt. B and C are somewhat dryer looking.
Frit 50% / Kiln clay 25% / Muslim Mkt. 25%
Result: Gray dryish matt
Frit 50% / Muslim Mkt. 50%
Result: Gray Matt, well adhered to body.
UMP SERIES #8
Frit 25% / Feldspar 25% / Muslim Mkt. 17.5% / Silica 32.5%
A= +5% Frit
B= +10% Frit
C= +15% Frit
Result: Light Gray Matt. All four samples are underfired.
Frit 25% / Feldspar 30% / Muslim Mkt. 15% / Silica 30%
Result: Underfired. Gray and powdery.
Control Sample #1:
Frit obtained in Lampang from ceramic supply store. (Ceramic Chemical Refactory Co. Ltd. 39/13-14 Thacrown-Noi Rd. Muang Lampang 52100 Thailand, 054-217463/054-222739)
Result: Raw Frit sample was melted completely.
Control Sample #2:
The feldspar used in these tests is naturally mined from the Lampang area and is a Potash/Soda spar. (TW Ceramics 561Moo2 Phaholthin Rd. Tambool Cfhompoo Amphur Muang Lampang 52100 Thailand, (6654) 226105 / (6654)323205)
Result: Raw Feldspar sample was unmelted.
Curious Black "Wonderpowder":
This sample was taken from black rock chunks of hard coal-colored material taken from the kiln clay excavations. It was easily crushed and screened.
Result: Light brownish as a stain.
Two samples of Terra Sigillata glaze were verigated. One sample came from the red kiln clay and another from the Mae Sot Quarry clay. Only the kiln clay sample was fired, to 850degC.
This sample was siphoned off the top of the settling bucket before it was dumped into the drying ponds. It was screened through a fine mesh and settled, the clear water on top poured off until it was of desired thickness (enough to dip coat a piece to 1/32").The Sigillata was applied to leather hard ware (of Ou-Moun clay body). When applied to dry wares it cracked away, probably due to not so fine particle size. After it was leather-hardd, it was burnished using an old plastic shopping bag and incised to bring out the body colors. Once fired it was buffed and waxed with shoe polish and came out very nicely.
All of the clays within the camp have limited plasticity when dug. All will greatly improve by aging. We would recommend that they be processed and mixed wetter than they would be used, a minimum of 6 months before they are needed. This will require that an assessment be made of the amount of clays required for programs. It is suggested that you make twice as much clay as you will need per week and store it in advance for at least 6 months. If it is stored in plastic bags it can be left outside. The amount of time necessary for processing is not an urgent factor except at the beginning of the program.
Lathun Brick Clay
From Mae Sod
Received in extruded brick form,
Grey when dry, Fairly plastic, has a lot of non-clay material in it.
Processed by slaking, big chunks were picked out
Accessible through Heh-Ehs friend by the truckload at no cost
Fired to 1000deg shrinkage of 9%, has a strong fired strength, buff colored, and no visible signs of vitrification.
Mae Sot Quarry Clay
From Mae Sod in direction of Burmese Border, This sample was taken from the banks of a small reservoir
Yellow and white in raw form, it contained a lot of non clay material
processes to bright yellow clay, screened wet and levigated from top
Fine grained and plastic
Shrinkage at 1000degC is 13%,strong-fired strength
Iron rich, dark, some signs of vitrification, absorption shows approx. 9%
Accessibility unknown (possibly large quantities)
Kiln area clays
Samples were taken from Handicap, kiln-site, and Zoa Ed center.
Screened dry through windowscreen25% of material was recovered, slaked overnight, A short clay that is barely throwable. Particle size is not particularly fine but will improve with aging.
Red in the raw, poor to good plasticity,
Shrinkage at 1000degC 12%, fires to terra cotta red, good to excellent fired strength, minor signs of vitrification. Estimated that this clay can be fired to at least 1100degC
Muslim Market area clay
Samples were taken from a site within the market area behind a clothing stall. Access is difficult but not impossible. The clay seam is located under a significant overburden. It must be transported a significant distance to the studio.
Grey to white in the raw. Plasticity, throwable but short clay, fine grained will improve greatly with aging
Processed dry screened and slaked. 95% material was recovered. A very clean clay.
Fires buff to white at 1000degC, shrinkage 8%, a sample of this was fired previous to arrival at 1300degC, good fired strength. No signs of vitrification at 1000deg. This is a semi porcelainesque stoneware.
A sample was taken from below the orphanage in Zone A
This clay is very similar in substance to the Muslim Market area clay. The sample that was obtained was very carefully chosen from the overburden. It may be that this clay can be obtained more easily if the site is excavated.
Screened wet, a very small sample. A short clay
Shrinkage is 10% at 1000degC, buff colored no visible signs of vitrification, fired strength is good
Ou Mounís Clay
Mined from a fish pond near Ou-Mounís House, it is dug from an 8í pit in the middle of a garden in a populated area. A gray clay when raw. Was slaked and screened wet, some non-clay material. Good fired strength but is very short and flabby. The buff colored brick in the firebox and chimney base of the bisque kiln is made from this clay. Accessibility is poor, transportation is required to bring it from the bottom of the camp.
Drying: Once it is established what clays are going to be used these clays must be dug and allowed to dry thoroughly.
Screening Dry: Once dry it may be dry screened. Dry screening seems to be faster in some cases. In the preliminary screening of mined clays 1/2" hardware cloth or window screen is used to mill the clay to a consistent size for drying or immediate slaking into water. While taking part in this process it was decided that when using hardware cloth, a smaller sized mesh (1/4" or 3/16th") may be more durable, and more suitable for both of these applications
Slaking/Blunging: The dry clay should be mixed with water enough to allow it to melt completely. Let it absorb as much water as it needs, this process can be accelerated by agitating or stirring.
Screening Wet: At this point you may want to pour the clay slip through a window screen mesh (unless you have already done this while it was dry) to remove the larger particles. Garbage Buckets: 160litre hard plastic garbage buckets are being used to slake large batches of milled dry clays. The slaking of the clay happens beside the drying ponds so the wet clay slip can be dumped into the ponds without too much work.
Settling: The slip is settled overnight and the water is poured or siphoned off the top.
Stiffening: The clay slip is stiffened by pouring it into cloth lined drying ponds. When it can be rolled off the cloth into wet balls. At that point you can further stiffen the balls of clay. Drying Ponds: Two large pits 7'x7'x8"(depth) were dug and lined with three lengths of muslin cloth. These pits are used for drying large quantities of slaked clay slip into a soft clay stage to be brought up to the workshop for wedging on the plaster wedging table.
Storing: Store in sealed plastic bags or covered buckets softer than for normal use. This will allow the clay to age. Due to the limited plasticity of the newly mined clays it is advisable to age these materials as long as possible to improve their plasticity. The preliminary slaking process adds water to much of the clay structure, however, in order for this hydration to happen on a molecular level the clay must be stored in a moist state to allow time for the further absorption of water.
Wedging: Pottery may not be formed without proper clay preparation, wedging takes the clay from the raw state and turns it into an even consistent mass. It is this process that allows the clay to be easily formed into shapes.