A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering of the Asian Institute of Technology, Bangkok, Thailand.

Thesis (M.Eng.) - Asian Institute of Technology, 1971

An attempt was made on a laboratory scale to find a new method for treating surface water that would be technologically and economically applicable to small communities in Asia. The most promising solution found was the use of local materials in a series filtration system incorporating both a roughing or primary filter followed by a secondary or polishing filter of the slow sand filtration design. Since there were no basis for the design and filtration rates to be used for each media under study, the study was geared toward evaluating an optimum filtration rate and influent turbidity limit for the filter to function efficiently and for an extended period of time to minimize the frequency of cleaning. Materials that are ordinarily available in abundance in rural Asian villages were investigated as filter media and compared in terms of turbidity removal, length of filter run and head loss development. The following were the media investigated pea gravel, burnt rice husk, raw rice husk, coconut husk fiber, charcoal and sand. No one medium proved to be superior to all the others at this stage of comparison but the burnt rice husk and shredded coconut husk fiber proved to be more efficient than sand in terms of length of run and turbidity removal. Burnt rice husk appears to be a potential substitute for sand in slow- sand filters while coconut husk fiber could be substituted for coarse sand in the roughing filter. A detailed economic comparison with conventional water treatment systems was not possible. However, a general comparison of the possible savings from the new system was made and it was found that financial and labor requirements and the level of training needed by the operator could be substantially reduced by using the series filtration treatment system.