M.Ing. / The aim of this investigation was to determine the N1 value in the relationship between pressure and flow for different types of water reticulation pipes with different forms of fractures. The relationship is defined as: Once these N1 values are established, they can be used as an indication for better pressure management in water reticulation networks. This investigation was limited to three types of pipe with diameter 110 mm and wall thickness of 4 mm: • uPVC • Cast iron • Steel And three different forms of failures: • Round holes • Longitudinal splits • Circular cracks Pressure step testing was used to obtain experimental data. During analysis of the data, Cd was calculated from the initial area of a fracture and kept constant thereafter. Conclusions could be drawn about similar type pipes with similar fractures and comparisons could be made between all the samples as well as previous studies. Generally it seems that longitudinal splits lead to the most excessive leaking, followed by round holes and then the circular cracks. A summary of typical N1 exponents found is shown below: • uPVC pipes with round holes 0.48 - 0.49 • uPVC pipes with longitudinal splits 0.89 - 1.26 • uPVC pipes with circular cracks 0.31 – 0.48 • Cast Iron pipes with round holes 0.43 – 0.44 • Cast Iron pipes with longitudinal splits 0.42 – 0.46 • Cast Iron pipes with circular cracks 0.41 – 0.43 • Steel pipes with round holes 0.42 – 0.44 • Steel pipes with longitudinal splits 0.39 – 0.45 • Steel pipes with circular cracks 0.38 – 0.48

M.Ing. / With an ever-increasing population and a decrease in availability of fresh water, the sustainable and secure supply of fresh water is a growing concern worldwide, and particularly so in South Africa. Plumbing systems have a significant impact on both the efficient use of water and water losses because high quality plumbing components can reduce wastage and leakage, while inferior components increase the risk of failures and leakage and can even have health impacts in case of leaching of toxic elements or contamination of drinking water. In recent years, South Africa has seen a plethora of imported and pirated plumbing components introduced onto the local market. While South African legislation requires that only plumbing components approved by the South African Bureau of Standards or Water Services Authorities (municipalities) shall be installed, it does not explicitly prohibit the importation and sale of non-compliant components. The availability of non-compliant components on the market combined with insufficient enforcement of legislation, have created a situation where many non-compliant components are installed, although the extent of this problem was not known at the start of the project. The study aimed to quantify the extent of non-compliant plumbing components installed in South Africa, as well as the impact this will have on future water demands and losses. Various research methods were used including literature reviews, surveys and interviews with role players, search of available plumbing components for sale in South Africa, on-site visits and analysis of some case studies. The study found that about 50% of plumbing components installed in South Africa do not comply with legal requirements. Because these components are of poor quality, they hurt the local industry whose components cost more in order to comply with SABS standards. Therefore, they suffer from reduced market share and lower profit margins, and South African job losses occur in the sector. In general, compliant Summary components are up to 135% more expensive than non-compliant components. This study also found that even plumbers who are IOPSA members, and therefore bound by a code of conduct to use only compliant components, install non-compliant components. The site visits to some government low-cost housing developments showed that less than 10% of the plumbing components installed were compliant, and a major problem with non-compliant components was leakage. Therefore if the situation does not change, on-site water leaks will remain a great concern in South Africa. In order to address this situation the main suggestions are better enforcement of legislation, better training and control of plumbers and restrictions on the importation of non-compliant components.

M.Tech. / The Water Services Authorities in South africa are currently responsible for the management and operation of rural water supply in the country.Many of these comunity water supply schemes are in poor conditions due to lack of proper operation, maintenance, poor design and/or poor construction with the result that they experience serious water losses through physical losses(leakage)and apparent losses, and often fail to meet the consumer's demand during peak periods.Indications are that substantial quantities of water are lost in the systems due to leakage, wasteful use, illegal connections and meter errors. As a result of these problems, the levels of service delivery are often very poor leading to low cost recovery resulting in water services institutions becoming ineffective and inefficient businesses. The investigation attempted to identify effective techniques and/or strategies for water loss reduction in Dinokana Village (Central District Municipality). The study identified and defined various procedures/methodologies for managing Non Revenue Water in community water supply schemes. Some of the strategies developed from the findings of this research have been implemented with a view of quantitatively verifying their effectiveness.

Water leakages and subsequent water shortages is an occurrence that causes inconveniences to many people around the world. Expanding population groups have the potential to stimulate urban development rates which in turn leads to an increase in water demand. This places more pressure on water suppliers and the infrastructure involved in water distribution systems causing pipe bursts and water leakages to occur more frequently. Water resources are thus wasted, leading to costs associated with the impact of the pipe burst and water shortage for both water suppliers and consumers. This study focuses largely on the identification, causes, impacts and quantification of losses through water leakages and the subsequent cost associated with the leakage. The study area that was chosen is Vorna Valley, which is in the region of Midrand and falls within the Gauteng province. Midrand is rapidly expanding spatially and population rates are on the rise. To achieve the objectives of this study, residents, water suppliers and key informants were consulted through questionnaires and the results were statistically analysed. The Economic Model for Leakage Management, that is specifically aimed at determining when a water supplier should invest in active leakage control for a specific, zoned metered area, was applied. The model established the costs of leakages and pipe bursts and the appropriate interval for active leakage control by the municipality for the study area. Results from the Economic Model for Leakage Management suggests that an active leakage control interval of 6 months is best as the cost due to water loss is at its lowest. In the study region, 83,1% of residents experience burst pipes over 6 months. Findings of this study could benefit water suppliers when choosing the best mitigation method to apply during water leakages and water shortages. / Dissertation (MSc (Environment and Society))--University of Pretoria, 2020. / National Research Foundation / Geography, Geoinformatics and Meteorology / MSc (Environment and Society) / Restricted

Made available in DSpace on 2015-06-17T19:33:44Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-12-09. Added 1 bitstream(s) on 2015-06-18T12:48:07Z : No. of bitstreams: 1 000830129.pdf: 1160653 bytes, checksum: 208eaf9a5e0179c424cd0aaf5d1ee860 (MD5) / Vazamento em tubulações é uma das principais preocupações de empresas fornecedoras de água, devido à escassez e disponibilidade de fontes de água potável, principalmente em grandes cidades. A técnica de correlação cruzada tem sido bastante utilizada nos últimos 30 anos na localização de vazamentos em tubulações metálicas. Entretanto, o método de correlação cruzada não tem a mesma eficácia na detecção de vazamentos em tubulações plásticas, devido ao elevado nível de atenuação entre a parede da tubulação e o fluido, que rapidamente suprimem o sinal de vazamento. Por isso, o sinal de vazamento não consegue percorrer longas distâncias em tais tipos de tubulações. Sabe-se que, a faixa de frequência na qual a energia do vazamento está localizada é baixa, consequentemente o sinal de vazamento é facilmente afetado pelo ruído de fundo. Para evitar esta interferência, o filtro passa-banda é utilizado para suprimir o ruído indesejado antes de aplicar a técnica de correlação cruzada. A escolha das frequências apropriadas a serem utilizadas no filtro passa-baixa são fundamentais para obtenção de uma estimativa de tempo de atraso confiável entre dois sinais de vazamento, este tempo de atraso é utilizado para calcular a posição do vazamento. Este trabalho tem como objetivo à investigação e criação de um simulador de vazamento, aqui chamado de Simulador Virtual de Vazamento, através da utilização de um modelo analítico de vazamento em conjunto com atuadores mecânicos (caixas de som). Este simulador de vazamento poderá ser utilizado para avaliar correlacionadores comerciais e para fins de treinamento. A principal vantagem deste método é o controle de todo o experimento, de modo que a característica do sinal de vazamento simulado é conhecida previamente / Leak in pipes is one of the main concerns of water companies due to the scarcity and availability of potable water sources, especially in big cities. The cross-correlation technique has been used in the past 30 years to locate leaks in metallic pipes. Although the cross- correlation method works very well for metallic pipes, it does not have the same effectiveness in plastic pipes. This is mainly due to the high attenuation between the pipe wall and the fluid, which suppress rapidly the leak noise. Hence, leak noise does not travel long distance in such pipes. Moreover, the frequency range over which the leak energy is located, is at low frequency. Hence, the leak signal is easily affected by background noise. To avoid these problems, a band-pass filter is used to suppress undesirable noise before conducting the correlation. The selection of the band-pass filter limits is the key to have a reliable time delay estimate between two leak signals, which is used to calculate the position of a suspected leak. This work concerns the investigation and design of a leak simulator, here named Virtual Pipe Rig, using an analytical leak model together with mechanical actuators (shakers). The analytical model is responsible for generating leak signals used to drive the mechanical actuators, which are responsible for simulating the measure positions. This virtual pipe rig can be used to assess commercial correlators and training purposes. The main advantage of this method is the entire control of the experiment, so that the characteristic of the simulated leak signal is known beforehand

Orientador: Michael John Brennan / Co-orientador: Fabricio César Lobato de Almeida / Banca: Gustavo Luiz Chagas Manhães de Abreu / Banca: Paulo José Paupitz Gonçalves / Resumo: Vazamento em tubulações é uma das principais preocupações de empresas fornecedoras de água, devido à escassez e disponibilidade de fontes de água potável, principalmente em grandes cidades. A técnica de correlação cruzada tem sido bastante utilizada nos últimos 30 anos na localização de vazamentos em tubulações metálicas. Entretanto, o método de correlação cruzada não tem a mesma eficácia na detecção de vazamentos em tubulações plásticas, devido ao elevado nível de atenuação entre a parede da tubulação e o fluido, que rapidamente suprimem o sinal de vazamento. Por isso, o sinal de vazamento não consegue percorrer longas distâncias em tais tipos de tubulações. Sabe-se que, a faixa de frequência na qual a energia do vazamento está localizada é baixa, consequentemente o sinal de vazamento é facilmente afetado pelo ruído de fundo. Para evitar esta interferência, o filtro passa-banda é utilizado para suprimir o ruído indesejado antes de aplicar a técnica de correlação cruzada. A escolha das frequências apropriadas a serem utilizadas no filtro passa-baixa são fundamentais para obtenção de uma estimativa de tempo de atraso confiável entre dois sinais de vazamento, este tempo de atraso é utilizado para calcular a posição do vazamento. Este trabalho tem como objetivo à investigação e criação de um simulador de vazamento, aqui chamado de Simulador Virtual de Vazamento, através da utilização de um modelo analítico de vazamento em conjunto com atuadores mecânicos (caixas de som). Este simulador de vazamento poderá ser utilizado para avaliar correlacionadores comerciais e para fins de treinamento. A principal vantagem deste método é o controle de todo o experimento, de modo que a característica do sinal de vazamento simulado é conhecida previamente / Abstract: Leak in pipes is one of the main concerns of water companies due to the scarcity and availability of potable water sources, especially in big cities. The cross-correlation technique has been used in the past 30 years to locate leaks in metallic pipes. Although the cross- correlation method works very well for metallic pipes, it does not have the same effectiveness in plastic pipes. This is mainly due to the high attenuation between the pipe wall and the fluid, which suppress rapidly the leak noise. Hence, leak noise does not travel long distance in such pipes. Moreover, the frequency range over which the leak energy is located, is at low frequency. Hence, the leak signal is easily affected by background noise. To avoid these problems, a band-pass filter is used to suppress undesirable noise before conducting the correlation. The selection of the band-pass filter limits is the key to have a reliable time delay estimate between two leak signals, which is used to calculate the position of a suspected leak. This work concerns the investigation and design of a leak simulator, here named Virtual Pipe Rig, using an analytical leak model together with mechanical actuators (shakers). The analytical model is responsible for generating leak signals used to drive the mechanical actuators, which are responsible for simulating the measure positions. This virtual pipe rig can be used to assess commercial correlators and training purposes. The main advantage of this method is the entire control of the experiment, so that the characteristic of the simulated leak signal is known beforehand / Mestre

M.Ing. / An investigation was conducted on on-site leakage in selected suburbs of Johannesburg to determine the average leakage flow rate of water lost per property, as well as the distribution patterns of leakage flow rates for two categories of properties: Residential (single domestic houses on stands) and Other (non-domestic and large domestic users). The incentive of this research project, funded by the Water Research Commission (WRC) of South Africa, was to provide figures and to gain insight into the magnitude of water lost at the consumer level, also defined as "on-site leakage". Little was published on on-site losses compared to water losses in municipal distribution systems. The key information for this study was collected through physical inspection of a group of randomly selected properties and by means of measurements taken from municipal water meters used for billing purpose. The project was assisted by Johannesburg Water (Pty) Ltd who provided a list of 233 properties having water meters of less than five years old. These properties were chosen at random and included properties whose meters were replaced due to meter failure or customer complaints. The results of the research show that overall 64 % of investigated properties had measurable on-site leakage at an average rate of 22.9 I/h per property, equivalent to a monthly volume loss of 16.5 kl per property. The median rate was 5.7 I/h per property, translated into a volume loss of 4.1 kl/month per property. An analysis of the data revealed that the general extreme value distribution and its derivatives (Frechet, Wakeby, Exponential and Gamma) provide good descriptions of the distribution of on-site leakage rates. The apparent losses resulting from metering errors were also estimated. The results of this study should be of interest to water engineers and to decision-makers involved in water demand management. It could form a basis for further investigation in the analysis of on-site leakage on a large scale in South Africa by systematically monitoring all segregated categories of users throughout different cities and towns in South Africa.

Due to increasing water treatment costs and conservation needs, traditional water loss analysis and acoustic leak detection methods are becoming heavily scrutinized by water utilities. This study explores water loss in Johnson City, Tennessee and how geospatial data analysis techniques improve water loss mitigation. This project uses sample water system pressure data and ordinary kriging spatial interpolation methods to identify leakage areas for further investigation. Analysis of existing geographic information system (GIS) water utility datasets with interpolated hydraulic grade values at sample water pressure points produce manageable survey areas that pinpoint areas with possible water leakage. Field detection methods, including ground-penetrating radar (GPR) and traditional acoustic methods, are employed to verify leakage predictions. Ten leakage areas are identified and verified using traditional acoustic detection methods, work order research, and GPR. The resulting data show that spatial analysis coupled with geospatial analysis of field pressure information improves water loss mitigation.

The overall objective of this study was to determine the influence of intermittent water application via overhead irrigation on water and N leaching, substrate water distribution, and plant growth in a pine bark substrate. Intermittent irrigation was defined as the application of a plants' daily water allotment in multiple applications with prescribed intervals between applications. Specifically, experiments were conducted to determine: 1) the influence of water application rate and pre-irrigation substrate moisture content on water application efficiency, 2) the influence of the number of intermittent applications and interval duration between applications on application efficiency, 3) substrate water distribution after intermittent irrigation, 4) efficiency of intermittent irrigation in unamended and sand-amended pine bark, and 5) influence of intermittent irrigation on plant growth. In all experiments, Marigold (<u>Tagetes erecta</u> L. 'Apollo') was grown in unamended or sand-amended pine bark in 3.8-liter containers. After saturation, bark was allowed to dry via evapotranspiration to targeted moisture deficits after which irrigation treatments were applied via a simulated overhead irrigation system. In experiments evaluating intermittent irrigation, water was applied intermittently (multiple applications with intervals between applications) or continuously (single application). After 1 h drainage, collected leachate was measured and analyzed (in two experiments) for electrical conductivity, N0₃-N and NH₄-N and water application efficiency (percentage of applied water that is retained in the substrate) determined. There was an inverse linear relationship between water application rate and efficiency. Leachate N (total) increased linearly as application rate increased. Efficiency increased curvilinearly and leachate N (total and concentration) increased linearly as pre-irrigation substrate moisture content decreased. Efficiency of intermittent irrigation was greater than continuous irrigation over a range of pre-irrigation substrate moisture contents, and in both unamended and sand-amended pine bark. Intermittent efficiency increased linearly as interval duration between applications increased. Efficiency of five 60 ml applications was greater than three 100 ml applications. Gravimetric moisture content of the bottom third of the substrate was greater with intermittent than continuous irrigation. Intermittent efficiency was greater in pine bark than in sand-amended pine bark when the water deficit from container capacity was the same for both substrates. Repeated intermittent irrigations resulted in less water and N leaching compared to continuous irrigation. Intermittent irrigation did not result in substrate nutrient accumulation and had no adverse effect on plant growth. Fresh and dry root weights and shoot and root N concentrations were greater with intermittent irrigation than continuous irrigation. / Ph. D.

Submitted in fulfilment of the requirements for the degree of Master of Engineering, Department of Civil Engineering and Surveying, Faculty of Engineering and the Built Environment, Durban University of Technology, Durban, South Africa, 2016. / Water is a critical resource in Southern Africa. The region thus needs to protect both the quality and the quantity of its water resources through robust water conservation and demand management (WC/DM) measures. Water demand management encompasses activities that aim to decrease water demand, improve the efficiency of water use and prevent the deterioration of water resources. Water conservation refers to policies, measures or consumer practices that promote the conservation of water resources. Water resources should be used wisely to secure a water supply that is of good quality and enough for South Africa’s people and its natural environment, which provides the ecosystem that supports all forms of life. When a water utility systems experience water losses, the amount of water available to consumers is reduced, making it difficult to satisfy demand. Water losses also occur as a result of inaccuracies in customer meters, data errors in the billing system and unauthorised consumption. Such losses result in non-revenue water (NRW), which is a serious threat to the water supply sector. NRW refers to the water that is produced and lost without generating revenue for the utility. This research study investigated strategies that could be used to address the challenge of water losses, by developing a more suitable business model that could be incorporated into Ugu District Municipality (DM)’s existing NRW reduction strategies. The study was carried out in Amandawe and Umzinto zones of the District Municipality and it covered the period 01 March 2014 to June 2015. The study objectives were made up of four components. The first was to identify and prioritise the implementation of NRW reduction strategies. This was achieved by identifying the pipes to be closed off, which were supplying a significant number of consumers. For those pipes that were not closed off, flow meters were installed to measure the flow into and out of a zone. The system was then tested for zero pressure by isolating all closed valves to ensure that there were no potential feed-backs into the zone. Pressure gauges were set up on standpipes for routine pressure monitoring. The test was run at night (between 01.00 and 05.00 hours) when the system was under pressure. When the pressure dropped consistently, this meant that there was no feedback into a zone. Leaks were detected by logging the system in order to obtain night flows, which were analysed to determine the system behaviour. The results for Amandawe Zone after implementation of the pressure management programme, indicated that the average zone’s night pressure (AZNP) decreased from 7.38 bars to 5.95 bars. For Umzinto Zone, the AZNP dropped from 5.5 bars to 3.3 bars. The minimum night flows (MNFs) dropped from 34.80 m3/hr to 15.20 m3/hr in Amandawe Zone and from 6.4 m3/hr to 1.70 m3/hr in Umzinto Zone. The daily cost of excess night flow due to bursts was reduced from R2276.17/day to R862.61/day in Amandawe Zone and from R361.24/day to R40.46/day in Umzinto Zone, which provided huge savings. The second objective was to identify the sources and causes of water losses in the study area by conducting field measurements and observations. This was achieved by physically inspecting the infrastructure using visual observation, mechanical listening sticks, correlators, ground microphones and system loggers. The following indicators were used to physically identify underground leaks: unusually wet surfaces in landscaped areas, pools of water on the ground surface, noticeably green, soft and mouldy areas surrounded by drier surfaces, a notable drop in water pressure or flow volume, unexplained sudden increase in water demand or water use at a fairly steady rate for several billing cycles, cracks in paved surfaces, potholes or sink holes and the sudden appearance of dirty water in the main distribution system. For this study, the water losses in the system were found to be as a result of various causes including leaks, aging infrastructure, high pressure in the system, damaged pipes and illegal connections, among others. The third objective was to construct a water balance in order to determine the key performance indicators for the NRW reduction strategies. This was achieved by determining the system input volume (SIV), billed authorized consumption (BAC), unbilled metered consumption (UMC), unbilled unmetered consumption (UUC), real losses (RL), apparent losses (AL) and IWA Key Performance Indicators. Bulk and domestic meter readings were used to calculate the components of the water balance. The results of the water balance indicated that there was a decrease in the SIV from 904 kL/day to 523 kL/day in Amandawe Zone and from 382 kL/day to 221 kL/day in Umzinto Zone. The physical water losses were reduced from 611 kL/day to 377 kL/day in Amandawe Zone and from 93.8 kL/day to 45.8 kL/day in Umzinto Zone. The NRW was reduced from 659 kL/day to 395 kL/day in Amandawe Zone and from 94.2 kL/day to 46.2 kL/day in Umzinto Zone. The fourth objective was to develop the most suitable business model for Ugu DM based on the results arising from the first three objectives. Ugu DM needs to ensure both operational and financial efficiency. Operational efficiency could be achieved by minimising real water losses through reviewing water services standards, developing district metering areas, pressure management, leak detection and repair, reservoir control to stop overflows and pipe replacement programs. Financial efficiency could be achieved by carrying out regular meter testing and calibration, securing database integrity, managing illegal connections, ensuring that all customer connections have meters and ensuring that the tariff structures were cost reflective in order for the municipality to cover costs and generate revenue. Findings of this study could assist other water utilities operating under similar conditions. The implementation of this study’s results could have positive economic, social and environmental effects on Ugu DM. It was concluded that rezoning, pressure management and leak detection were the most critical NRW reduction strategies as they had a positive impact on the system. The main causes of leaks in the system were identified as aging infrastructure, high pressures in the system, and illegal connections. All the critical KPIs of IWA water balance responded positively after the implementation of the strategies by reducing. The operational and financial efficiencies were identified as critical for a WSA to develop a business model that could sustain itself.