AT MICROFICHE REFERENCE LIBRARY A project of Volunteers in Asia Flow Water SvsteU$ f by : A. Scott Faiia Published by: A. Scott Faiia CARE Lombok Jalan Veteran 1 Mataram, Lombok Indonesia Available from: A. Scott Faii, CARE Lombok Jalan Veteran 1 Mataram, Lombok Indonesia Reproduction of this microfiche document in any form is subject to the same restrictions as those of the original document. PRACTICAL DESIGN NOTES FOR S IMPLE RURAL WATER SYSTEMS A.SCQTT FA IIA Saxaitary Engineer 1982 GRAVITY FLOW WATER SYSTEMS PRACTICAL DFSIGN NOTES FOR SIMPLE RURAL WATER SYSTEMS CONTENTS PAGES I. 2. 3. 4. 5. Introduction The Water Source General Syetem Design and Design Parameters Pipeline Design Summary of Suigested Design Guidelines l- 2 3-5 6 - IO 11 - 18 19 Appendices A. General Explanation of Flow and Head Losses in Closed Pipes Al - A4 13. The K w Hater Flow Calculator B1 - B2 C. Construction Notes CI - c8 D. Steps in Survey and Design Dl E. Sample Design for Desa Gembira El - E8 F. GlO86ary F1 - Fb -l- 1. INTRODUCTION 1.1. These design notes primarily cover Simple gravity flow Water systems. This is often the only feasible alternative for many rural areas at tk~ti presant time ,.nd in many countries a major portion of the funds available for water systems are allocated to gravity flow systems. Hovever , field iABpeCtiOA6 of completed projects and the literature indicate that there is a lack of capable design peraonuel and a substantial number of systems do not function properly due to poor design. Enphasis in the notes is placed OA those aspects most neg- lected or misunderstood. Certain topics such as maintenance, cornmu- nity participation and health education are beyond the scope of this manual but must be given due consideration in COABtrUCtiAg any type of water system. 1.2. These design notes present simple examples and explanations to illustrate some of the basic principles of gravity flow systems. Suggested guidelines l for design paraoeters are also presented. The theory of water system design is extensively covered in other publications. The emphasis here is placed OA practical methods that have been tested in the field and have given acceptable results. However, it must be emphasized that there are AO set or standard solutions for the design of a vater system. Attempts to implement vater supply programs using a limited number of standard designs can be expscted to produce poor results. Each community is unique and will require its own carefully prepared design by a person thoroughly fsmiliar with local conditions. 1.3. The notes are based on several years experience in Indonesia and have been used for training of field staff responsible for site selection, cleeign,and implementation. They are intended for use by persons respOABible for planning, designing,and implementing, rural water systems. While a technical background or prior fami- liarity vith water systems uouid obviously be useful it is intended that the material can be understood by persons vithout such a back- ground. Since the guidelines are based on IndOAesiWI conditions they should be applied with caution in other social, culttial and physical settings. l an asterisk denotes terms contained in the glossary in Appendix F. Such terms are so marked the first time they appear in the text. -2- 1.4. The type of systems discussed are simple branched systems as in the example in Appendix E. Several dozen systems have been cona- tructed using the design guidelinee and they generally servepopula- tions of 1,000 to 5,000 and are 2 to 8 kilometers in length. The largest system constructed using the suggested guidelines*contained 59 distribution reservoirs of 6 q 3 capacity, seven break pressiwe * tanks, and over 20 kilometers of distribution pipe. The flow in the system is approximately 15 l/s and this serves a 1982 population of approximately 12,000. -3- 2. THE WATER SOURCE 2.1. General Considerations 2.1.1. The water source should be free of fecal must supply continously s minimum amount contamination and of water to the sys- tem (purity and reliability). 50th of these aspects are \ difficult to measure with accuracy. Quality and quantity will fluctuate around some mean value under natural conditions and and the variations from the mean tend to increase as the na- tural ecological. balance is disturbed. A major change such as transformation from forest to agriculture, can be expected to significantly affect quantity and quality. Decreased yields and the total drying up of the springs are not uncommon. 2.1.2. The source also must be protectable and containable. In muddy areas and soft soil it is sometimes impossible to collect and protect the water. Areas of seepage rather than true springs are also difficult to deal with. Limestone areas must be thoroughly investigated because the spring could easily shift position altogether and the possibility of contamination is P greater.' $ 2.1.3. Another key consideration is the source's availability for use in the system. It is frequently not possible to use water previously used for irristion, &Mea&y allocated to other planned schemes or from sacred areas* 2.2. Estimation Of Quantity 2.2.7. Accurate measurement of quantity would require frequent monitoring over a period of several years and should pre- ferably include a drier than average year. However, this information is generally not obtainable and an estimation must be mede from point measurements+. As many point measure- ments as possible should be taken and they should be during the driest part of the year. Experience and judgement are critical, Note should be taken of the condition of the catchment tirea, vegetation, land use etc. Comments of long- time local residents regarding reliability and change over time are also useful. These however, must be veighed carefully because people- tend to overestimate flows during dry period6 and will sometimes falsely report that a source never drye up for fear of losing the project. -4- 2.3. Estimation Of Quality 2.3.1. Quality generally fluctuates much less than quantity but it would also require monitoring over a long period to establish an accurate estimation. Point measurments should include periods of both high and low flove. In general, the greater the fluctuations the greater the amount of data that muet be collected. The main considerations in assessing quality are bacterial quality, consumer acceptability and chemical quality. 2.3.2. %asuremeat of bacterial quality through testing for specific disease causing agents is not practicable. Instead an indica- tor organism is used ta assess the likelihood that the water is contaminated by harmful pathogens. The most suitable in- dicator organism at present is fecal coliform* as determined by the membrane filter technique?. The majority of fecal coliform organisms are ‘not harmful to man but their association with fecal matter indicates that organisms dangerous to health may be present. However, the link between presence of fecal coliform and contamination by fecal matter is not yet firmly established for tropical areaa and further research is needed. The establishment of guidelines for bacterial quality is com- plicated by several other factors. Firstly, most sources under consideration are unprotected and the simple act of cleaning up the area and protecting the source may often result in dramatic imProvementa in quality. However, for political and social reasons, it is often impassible to carry out such improvements without making a commitment to complete the water system regardless of changes in quality. Secondly, a rigid standard. awry result in the rejection of a source which is superior in quality to existing source& The replacement of a source containing several hundred thousand fecal8 per 100 ml by one with only 100 fecal6 per 100 ml can result in dramatic improvements in health. Thirdly, there is sufficient evidence to suggest that improving the quantity of wa- ter available without improving the quality will still result in significant health benefits. 2.3.4. In view of the above, only the most tentative bacterial quality guidelines can be proposed. Each situation must be judged on i tsOge:its. \ The obvious goal is no fecal coliforms pre- sent in any sample taken from the proposed source. If some contamination exists then it is best if the average of all samples is less than 50 fecals/lOO ml and if no single sample -5- exceeds I\)0 fecals/lOO ml. For levels greater than these it is best to provide some form of treatment to improve quality if the source is to be used. 2.3.>. All water sources should also be monitored several times a year after source improvement and pipe installation. This is useful to document changes brought about by the improvements and to guard again& contamination. 2.3.b. Consumer acceptability of the water is of prime importance. The be+ designed system in the world will be ureless if the con6umers do not accept and use the water. A simple visual inspection and sample testing of water along with a survey of the intended users will usually suffice to determine accep- tability. It is very important to avoid exce88 iron in the water that will turn tea or rice black. If a source has not previously been used for drinking then the water should be boiled and tea made to check for any undeeirable changes. Previously unusea sources are sometimes also objectionable for cultural reasorhs. 2.3.7. Extensive chemical testing is both costly and time consuming and is usually omitted if the water is acceptable to the consumer. However, any unusual circumstances should be noted and guarded against. For example, the presence of excess car- bon dioxide may not be noticed by consumers but it would ra- pidly corrode steel piping. -6- 3. GENERAL SYSTEM DESIGN AND DESIGN PARAMETERS 3.1. Type Of Service 3.1.1. The L~~:e’! of serv,cc proviaed in rural Indonesia is generally dietributlon at public taps for all types of domestic water uae i.e. drinking, bathing, laundry,and toilet. rroperly designed and maintained toilet facilities are noi a iltielth hazard. However, if there are serious doubts that the people will not use such facilities properly or sre totally unfamiliar with them then it is better not to include them in the design. Inslch cases a sample facility strategically located would be in or- aer, 3.1.2. The distribution points and facilities should be freely acceaa- able to all intended users, and should be located according to population density. Their design and placement should facilitate and encourage water use thus maximizing project benefits. As general guidelines,no more than 10 to 20% of intended uama should have to walk more than 100 meters to obtain water and the number of users per water faucet should be between 30 to 100. For example,a small ayatem serving 1,000 people could have 6 distribution points (either atand- posts, reservoirs or public baths) each with four faucets or approximately 41 persons per faucet. The lower range in the design figure should be used for multiple purpose systems while the higher figures are acceptable for systems with restricted use, e.g. drinking water only. 3.2. Water Usage - Design Figures 3.2.1. Multiple use systems should generally be designed for a per capita r16e of 60 liters/day. If sufficient water is available the figure could be 80 or 100 liters per day. If water is ex- tremely scarce and was to supplied for drinking purposes only the minimum acceptable figure would be 20 liters per person per day. These figures include an allowance for waSbge. 3.3. Storage Cepscity 3.3.1. Storage of water is often necessary and will be influenced by the nature of the water source and the design of the system. For example if the flow of the water source is just 1 l/a (%mj/day) and the.average daily ueaE* is also 86 m3/day then a certain amount of storage q uet be provided. -?- . This is because the flak ;f water is constant throughout the 24 hour period while the usage is not. Therefore, during periods of low usage, such as during the night ,part of the 86 m3 will flow from the source and not be available for use upless it is stored. If the flow is high enough ‘then the storage is not necessary provided that the pipe is large enouw to provide sufficient water at the time of peak usage+. How- ever, even with a auffioiently large source it is preferable to provide some storage at the point of uaew This aspect is discussed in more detail in paragraph 3.3.2. Figures IA to ID depict several possibilities for storage placement. 3.3.2. Fig 1A depicts the general schematic for placing storage at the point ob use. In moat cases this is the preferrad type of design for the following reasons : a. The inflow to each reservoir can be regulated so that each area receives a set allotment of water. If the people at that reservoir tend to waste water then they can only waste their allotment but nut that of others. In a atand- pipe system with--n--s .at file aouroe,wastage would be much greater if taps were inadvertantly left open. b, The small reservoir acts to break pressure in the system. This means that faucets at the point of use have only the he$‘of the reservoir itself on them and will last fur b loti;;cr,v Reduced head at the point of use also reduces wastage. c+. Storage at the point of use means that the main distribu- tion pipe is in use at all times. It can therefore be of a smaller diameter and thus reduce oosts. (Influence on total coat will depend on the flow of the source compared to average daily usage as this will influence storage costs). d. It is generally easier to obtain community support and cultivate feelings of ownership snd consequently improve maintenance through the construction of small scattered reservoirs as compared to one large distant reservoir and standpipes. Additionally, the construction of small re- servoirs allow’s each segment of the community to work at its own pbce during construction and a lack of community organization will be leas likely to impede the project. -PI- Figures IA - ID STORAGE PLACEMENT Fig. IA - Storage at the point of water usage. schematic for placemtnt of storage and r$,quires the smallest pipe diameter. Its advantages are outlined in paragraph 3.3.2. \ Fig. 33 - Storage far from source but still above distribution net- work, distribution from storage to public standpipes. The pipe line to the point of storage i6 the aame a6 in Fig. IA but after the point of storage is the same a6 Fig. IC and ID. This option should Only be used when Storage at the point of U6e is not Possible. Fig. AC - No storage provided, distribution direct to public stand- pipes. This type of system require6 a larger diameter pipeline. The cost is sometime6 less than that of the system in Fig. IA but u6e of etor8ge a6 in Fig. IA is preferable for reason6 outlined in paragraph 3.3.2. Storage provided at or near the water source then distribution direct to public standpipes. The pipeline diameter is the same as in Fig. IC and this is the most expeneive option. -9- e. In Indonesia the u6e of reservoirs can encourage water use and increase health’benefits from the 6y6tem. bith a few aic’itional walls the area can provide privacy for bathing and washing that is often explicity requested by the commu- nity. The walls of the reservoir can also be used for displaying health education messages. 3.3.3. The following example of storage calculation is applicable when storage is to be provided at the point of uee. (see Fig. IA). Under normal circumstances the neceseary storage is determined by comparing the supply curve ‘kith the consumption curve for the village. However, information on the consumption curve for Indonesian village6 is not available and 6ome other method is necessa.ry, Based on recent experience in Indonesia it is recommended that the 6torage capacity be fixed at. one half the average daily usage a6 determined by the population to be served and per capita consumption. For example, a system supplying 1,000 people with 60 liters/day would have an average L daily usage of 60 m3 and would thus require 30 m3 storage ca- pacity. If storage were provided at the point of u6e and there were 6 distribution points, a6 in the example in paragraph 3.1.2.. it could be accomplished with 6 distribution reservoirs each 5 m3 with four taps. This solution would be suitable if the population distribution were uniform. If it were not uni- form then the reservoirs would be of different size and could have different numbers of faucets. For example, the main heavily populated area could be served by a 10 q 3 reservoir with 8 fau- Three areas of medium density with reservoir6 of 5 m3 1 cets. ! and 4 faucets and three areas of low density with reeervoirs of 2,s q 3 and two faucets. The important point is that the total storage capacity and number of faucets remains the same (or approximately the same) but their placement is according to LiJa tonditions. In practice the numbers do not divide 60 neatly and 6ome adjustments sre necessary* For example, if the design population vere 961 the needed storage capacity would be 28,8 m3. This may be rounded up to 30 q 3 so that a etsndard design for tanks of 5 m3 or such could be used. The use of etsndard designs for certain eystem components can be very aonvenient and economical. The deeign figures are rough approx- lmationsonly, and osn be modified if necessary 3.3.4. There are situations where the distribution system is fed from a central storage reservoir above the village. (See Figs, 1B and 10) - 10 - This Is recommended primarily when the population density is too great to allow for placement of the storage reservoirs within the village. Because of their smaller size, standpipes can- be more easily placed at strategic locations. The storage capacity necessary for euch a system will be determined by the ratio of the estimated, minimum