Overview:
This 1911 report from the American Society of Civil Engineers details the findings and observations of the Washington, D.C. filtration plant over a period of five years. The plant, completed in 1905, consists of a pumping station, 29 slow sand filters, a filtered-water reservoir, and the necessary piping for water treatment. The report highlights the innovative sand-handling system and the effectiveness of raking the filter beds to prolong filter runs. The report also describes the results of experiments with preliminary treatment methods, including rapid filtration, a Maignen scrubber, and coagulation.
The report emphasizes the importance of efficient and economical operation, noting that sand handling is the most expensive component of filter maintenance. The report examines the effectiveness of various cleaning methods, including raking, scraping, and washing in place, and concludes that the most efficient and safe approach is to remove a small portion of the surface layer during cleaning. The report also discusses the results of experiments conducted at different filtration rates, concluding that increasing the filtration rate does not significantly affect the quality of the effluent, but does increase the cost of operation.
Key Findings:
- Raking is an effective and cost-saving method for extending filter runs. Raking the sand surface can double or triple the time between scrapings, with no adverse effects on filter performance.
- Hydraulic sand ejection is a highly effective and economical method for replacing filter sand. This method significantly reduces the cost of sand handling compared to traditional cart-based methods.
- Preliminary treatment is necessary for effectively and economically treating turbid water. Experiments with different preliminary treatment methods demonstrate that coagulation is the most reliable method for removing fine clay particles that cause residual turbidity.
- High filtration rates can be maintained with minimal impact on effluent quality. Experiments with different filtration rates indicate that doubling or even tripling the rate does not significantly affect bacterial counts.
Learning:
- Slow sand filtration is a complex process. The process relies on several factors, including straining, oxidation, and bacterial action.
- Raking filter beds can significantly improve efficiency and reduce costs. Raking allows for more water to pass through the filters between scrapings, saving time and labor.
- Hydraulic sand ejection is a highly effective and cost-saving innovation in filter maintenance. This method offers significant advantages over traditional methods for sand handling.
- Coagulation is an effective and reliable method for treating turbid water. Coagulation removes fine particles that can cause residual turbidity in the effluent.
- Higher filtration rates are possible with minimal impact on effluent quality. This finding suggests that the traditional emphasis on low filtration rates may not be necessary with certain types of water.
Historical Context: This report was written in 1911, a time when slow sand filtration was a well-established method of water treatment, but new technologies and techniques were being developed. The report reflects the growing understanding of the importance of efficient and economical operation in water treatment facilities, as well as the need for careful consideration of the impact of new technologies on public health.
Facts:
- The Washington, D.C. filtration plant was completed in 1905.
- The plant consists of 29 slow sand filters, each with an effective area of 1 acre.
- The Potomac River water is often turbid due to fine clay particles.
- The Washington Aqueduct system includes three settling reservoirs: Dalecarlia, Georgetown, and McMillan Park.
- The Dalecarlia Reservoir has a capacity of 141,000,000 gallons.
- The Georgetown Reservoir has a capacity of 140,000,000 gallons.
- The McMillan Park Reservoir has a capacity of 180,000,000 gallons.
- The gates at Great Falls are closed to exclude excessively muddy water, especially during the summer months.
- In the summer of 1907, the practice of shutting out water with a turbidity of 500 or more was established.
- A 10-inch hydraulic dredge removed more than 100,000 cubic yards of mud from Dalecarlia Reservoir in 1907 and 1908.
- The sand in the filters has an average effective size of 0.32 millimeters and a uniformity coefficient of 1.77.
- The average depth of sand in the filters is 40 inches.
- The average consumption of water in Washington, D.C. is 60 million gallons per day.
- The cost of operation of the filtration plant is about $82,000 per year.
- The cost of filter operations averages about $0.50 per million gallons of water filtered.
- The cost of pumping water for sand handling is $0.06 per cubic yard of sand ejected and washed and $0.03 per cubic yard for replacing.
- The average cost of labor for sand handling is about $0.26 per million gallons of water filtered or $0.34 per cubic yard of sand handled.
- The average turbidity of the raw water entering the filters is 106 parts per million.
- The average bacteria count in the raw water is 6,400 per cubic centimeter.
Statistics:
- The gates at Great Falls were closed for 10.50% of the time, excluding 40.06% of the suspended matter from the water supply.
- The average turbidity of the filtered water is 10 parts per million.
- The average bacteria count in the filtered water is 149 per cubic centimeter.
- The average typhoid fever death rate in Washington, D.C. for the six years prior to the installation of the filtration plant was 59 per 100,000 population.
- The average typhoid fever death rate in Washington, D.C. for the five years after the installation of the filtration plant was 37 per 100,000 population.
- The average length of a filter run is about 113 days.
- The average amount of sand removed from a filter during a scraping is about 213 cubic yards per acre.
- The average cost of sand handling is about $0.54 per million gallons of water filtered.
- The average cost of filtering is about $1.06 per million gallons of water filtered.
Terms:
- Slow sand filtration: A method of water treatment that involves passing water through a bed of sand. The sand acts as a filter to remove suspended matter, bacteria, and other impurities.
- Schmutzdecke: The layer of organic matter that forms on the surface of slow sand filters. It is important in the filtration process, but must be carefully managed to prevent clogging.
- Coagulation: The process of adding chemicals, such as aluminum sulfate, to water to cause small particles to clump together. These clumps are then easier to remove by sedimentation or filtration.
- Turbidity: A measure of the cloudiness of water caused by suspended particles.
- Bacteria: Microscopic organisms that can cause disease.
- Bacillus coli: A type of bacteria that is commonly found in the intestines of humans and animals. The presence of this bacteria in water is an indicator of fecal contamination.
- Hypochlorite of lime: A chemical that is used to disinfect water by killing bacteria.
- Ejector: A device that uses a jet of water to move sand or other materials.
- Under-drain: A system of pipes beneath the sand bed of a filter that collects the filtered water.
- Loss of head: The difference in pressure between the water entering a filter and the water leaving the filter.
Examples:
- Raking filter beds: The report describes how raking the sand surface of the filters at Washington, D.C. was successful in extending the time between scrapings and increasing the efficiency of the filters.
- Hydraulic sand ejection: The report details the implementation of a hydraulic sand ejection system at the Washington, D.C. filtration plant, which significantly reduced the cost of sand handling.
- Preliminary treatment experiments: The report describes experiments conducted with three different preliminary treatment methods, including rapid filtration, a Maignen scrubber, and coagulation. These experiments demonstrated that coagulation was the most effective method for removing turbidity.
- Filtration rate experiments: The report describes experiments conducted with six small filters operated at different filtration rates. These experiments showed that increasing the filtration rate did not significantly affect the quality of the effluent, but did increase the cost of operation.
Conclusion: The Washington, D.C. filtration plant has proven to be an efficient and economical method for treating the Potomac River water. The report highlights the success of innovative practices, such as raking and hydraulic sand ejection, in improving efficiency and reducing costs. However, the report also underscores the importance of careful consideration of the impact of new technologies on public health. The experiments with different preliminary treatment methods and filtration rates demonstrate that there is no single “best” method, and that the choice of treatment should be based on the specific characteristics of the water and the desired level of water quality.