Robert W. Dies
Bachelor of Applied Science
University of British Columbia, Canada, 2001
Submitted to the Department of Civil and Environmental Engineering
In Partial Fulfillment of the Requirements for the Degree of
at the
June 2003

© 2003 Robert Dies.
All right reserved.
The author hereby grants to MIT permission to reproduce and distribute publicly paper and electronic copies of this thesis document in whole and in part.



Countries like Nepal face tough challenges in terms of providing safe, clean drinking water for their citizens. The World Health Organization estimates that nearly 5 million people in Nepal lack access to safe drinking water while globally, 1.1 billion lack access to improved water supplies. Point-of-use water treatment technologies, such as household ceramic water filters, offer an affordable and effective means of treating water to standards suitable for drinking. The fact that ceramic water filters can be manufactured and produced by local ceramists with local materials makes ceramic filters particularly attractive as a point-of-use treatment technology that is affordable, appropriate, and sustainable.

This thesis examines existing ceramic water filter technologies, production processes, and methods for bringing a low-cost ceramic water filter to market in Nepal. Three types of disk filters and five types of candle filters are evaluated in terms of microbiological removal efficiency and flow rate. A red-clay grog disk filter coated with colloidal silver and three of the five candle filters (Katadyn􀀀 Ceradyn, Katadyn􀀀 Gravidyn, and the Hari Govinda white-clay candle filter capped on both ends) also coated with colloidal silver, performed the best in terms of microbiological removal efficiency (>98%) and flow rate (ranging from 641 mL/hr/candle (Ceradyn) to 844 mL/hr/candle (Gravidyn)).

In addition to filter testing, a guideline for developing a ceramic water filter in preparation for bringing a product to market is presented, along with a discussion on the importance of laboratory and field testing to ensure overall product performance. A step-by-step summary of the production process is also presented along with a comparison of the theoretical flow rate through a candle filter versus a disk filter. Recommendations for future work include testing and modifying the current disk-filter prototype design and research on the most appropriate filter element for the proposed prototype.

Download (PDF, 3.1MB)


Kiln Building: The Mani Kiln

Ceramic Pot Water Filter Kiln Building Resources

Building the Mani Kiln

Kiln building: The Mani kiln is an improved design for a wood burning kiln with a capacity of 50 ceramic pot water filters. Designed and distributed by Manny Hernandez – Northern Illinois University.

Complete drawings are included in the following PDF

Download (PDF, 4.08MB)

Kiln Building: LPG Burner Calculations

Kiln Building: Calculating gas consumption : A rough rule-of-thumb ratio is one square inch of flue
area to 8,000 BTU’s of maximum gas input


There are several variables that come into play when choosing burners for a kiln or furnace. Listed below
are the facts you need to know before deciding the size (Btu’s per Hour) of your burner system.

1. Total inside volume of the kiln.
2. Type of wall construction.
3. Maximum temperature you will be reaching.

Calculating Kiln Volume
Kiln volume is usually expressed in Cubic Feet (CF). In a flat top kiln this figure is arrived at by multiplying
the interior height (H) by the interior width (W) by the depth or length (L).

Sprung or Roman arch: CF = W x L x (Side wall + 2/3 of the arch rise)
Catenary arch: CF = L x Arch area (4/3 H x 1/2 Base Width)
Barrel kiln: CF = H x Pi x R2 (R2 – Radius is 1/2 the diameter x itself) (Pi = 3.14)

If you have used inches in the above equations, divide the total by 1728 to convert to Cubic Feet.

Wall Construction & Temperature
The type of material and its’ insulating values determines how many Btu’s per Cubic Feet per Hour (Btu/
Cf/Hr), you will need to reach a desired temperature. Below is a simplified chart showing materials,
desired temp., and the corresponding Btu/Cf/Hr. There are a host of variables that can affect kiln
efficiency. This is a basic guide only.


9″ Hard Brick

9″ Insultating Brick

6″ Ceramic Fiber

This simple table gives you an idea of how many Btu’s per Cubic Feet per Hour you will need. Multiplying
this figure by the total Cubic Feet will give you Btu/Hr. Now divide Btu/Hr by the number of burners you
plan to use to determine what Btu/Hr rating each burner should have. The numbers above show a range
of BTU figures. The highest figure in each range produces a 6-7 hour firing. The lowest figure will produce
firings in the 14-18 hour range. I feel it is better to have extra Btu’s than not enough. The above is a guide
not a guarantee. If you would like us to verify your calculations, please feel free to call or write.

Cone 06




Cone 6




Cone 10




Raku Construction & Btu/Hr Values

Many folks don’t realize that Raku kilns have much higher Btu input rates than stoneware kilns. This is
because Raku is traditionally done very quickly. For this reason, it is very difficult to bisque fire in a Raku
kiln. If you plan on purchasing or making a Raku kiln, please note that you could have problems with
steam explosions of the ware if you attempt to use the kiln for bisque. Also, the structural nature of Raku
kilns make many of them impractical for use at stoneware temperatures. The chart below gives the basic
Btu input for Raku kilns of various materials. These input values are for a fast firing rate of around 20-30
minutes for the first load. Subsequent loads would be slightly faster.


4 1/2″ Hard Brick

2 1/2″ Insulating Brick

4 1/2″ Insulating Brick

1″ Ceramic Fiber

2″ Ceramic Fiber







Sample Calculations & Burner Options

Flat Top Kiln: 45″ H • 45″ W • 45″ L. Constructed of 9″ insulating brick.

45 x 45 x 45 = 91,125 cu/in. divided by 1728 = 53 cu/ft (aprox.)
For cone 10 firing of 6-8 hours – 53 x 16,000 = 848,000 Btu/HR

For 2 Burners – 848,000 ÷ 2 = 424,000 per burner

For 4 Burners – 848,000 ÷ 4 = 212,000 per burner

For 6 Burners – 848,000 ÷ 6 = 141,333 per burner

Sprung Arch Kiln: (30″H + 5″ RISE) • 30″ W • 30″ Hard Brick construction

(30 + [.66 x 5]) = 33.3 x 30 x 30 = 29,970 cu/in divided by 1728 = 17.5 cu/ft
For cone 10 firing of 6-8 hours – 17.5 x 20,000 = 350,000 Btu/HR

For 2 Burners – 350,000 ÷ 2 = 175,000 Btu per burner

For 4 Burners – 350,000 ÷ 4 = 87,500 Btu per burner

For 6 Burners – Not necessary

Catenary Arch Kiln: 40″ W • 48″ H • 60″ L Insulating brick w/2″ ceramic fiber

([4/3 x 48]=64) x ([1/2 x 40]=20) x 60 = 76,800 cu/in divided by 1728 = 44.5 cu/ft
For cone 10 firing of 6-8 hours – 44.5 x 12,500 = 556,250 Btu/HR

For 2 Burners – 556,250 ÷ 2 = 278,125 Btu per burner

For 4 Burners – 556,250 ÷ 4 = 139,062 Btu per burner

For 6 Burners – 556,250 ÷ 6 = 92,608 Btu per burner

55 Gallon Drum Kiln: 18″ D • 32″ H Lined with 2″ ceramic fiber

([18 ÷ 2] = 9{radius} squared (9×9) x Pi (3.14) x 32 = 8,139 cu/in divided by 1728 = 4.7 cu.ft..
For Raku firing of 20-30 minutes – 4.7 x 20,000 = 94,000 Btu/HR

1 Burner Adequate

Ward Burner Systems
PO Box 1086 • Dandridge, TN • 37725
(865) 397-2914 phone • (865) 397-1253 fax



Based on required capacity of 50MJ/m2 (4500 BTU/ft2) per hour for cubic kilns lined with 150 mm (6″)
of ceramic fibre. The area on which the calculations are based is the total internal surface area of the kin
space, not the stacking space only.
Example: 20 cubic foot (total volume) fibre kiln to 1300oC uses approx. 210,000 BTU/hour at top
temperature therefore use 2 x LK32 burners or 4 x LK25 burners for more even heat.


Download (DOC, 132KB)