Drinking water quality and prediction of temperature distribution in domestic drinking water systems

Drinking water quality and prediction of temperature distribution in domestic drinking water systems

researcher: Ljiljana Zlatanovic


Research objectives
The aim of the research is to determine the key factors that affect microbial safety and quality of drinking water in DDWSs. In addition to this, the effects of plumbing extension for sprinkler accommodation on physico-chemical and biological drinking water parameters will be investigated. Furthermore, this research should determine the degree to which the domestic drinking water model outcomes (developed by Moerman, 2013.) are accurate representation of the real temperature dynamics in DDWSs.

According to the Dutch drinking water act the drinking water temperature should remain lower than 25°C at the customers’ tap. However, the drinking water temperature is difficult to control as it is a function of multiple factors, including: water temperature in the distribution network, climatic conditions, room temperature and residence time in domestic drinking water distribution systems (DDWSs). Accurate measuring of temperature at the tap point level is not only time consuming, but is often associated with high costs for drinking water companies. An accurate temperature model can, thus, be a useful tool to simulate and predict the levels and dynamics of temperature at each point in the DDWSs.

The impact of water residence time on the microbiological quality in DDWSs is yet not clear. In the distribution networks, bacterial growth is to a certain extent limited by short residence times. It is possible that the microbial quality of water can alter in DDWSs, where considerably longer residence times (under favourable conditions such as available nutrients and higher water temperatures) could promote the bacterial growth.

Last but not least, with a recent development of a fire sprinkler head that is efficient under low flow and pressure ranges, normally found in a Dutch household, it is now possible to integrate a fire sprinkler system into DDWSs. However, sprinkler system integration will influence DDWS’s layout such as piping diameters, total pipe length and thus might affect drinking water quality at the consumers’ tap.

Two full scale DDWSs, with and without the plumbing extension for sprinkler accommodation, were built in the water laboratory located in the building of Civil Engineering and Geosciences, TU Delft. The design of the experimental plumbing rigs was done according to a plan of a 2 storey house (Typical Dutch House). In order to simulate realistic drinking water consumption at the household level, the test rigs comprise 11 solenoid valves (tap points), which are configured to run automatically according to the one year demand patterns generated by SIMulation of water Demand, an End Use Model. Every tap point is fitted with a flow sensor and a temperature probe. The water temperature and flow are continuously measured, while the sampling campaigns for chemical and microbiological assessment are carried out twice a month.

For the sake of temperature modelling in DDWSs, the open source program EPANET is being used to perform hydraulic calculations, while the Multi-Species Extension (MSX) package of EPANET is being applied for the heat transfer calculations


Figure 1 Left: Number of intact cells per mL in fresh and stagnated water coming from the from kitchen tap in the DDWS without extension for fire sprinklers. Right: Number of intact cells per mL in fresh and stagnated water coming from the from kitchen in the DDWS with extension for fire sprinklers

The total number of cells (damaged cells and intact cells) is being distinguished by flow cytometer measurements. The total number of cells in fresh water samples is, so far, in the range from 2.7x105 to 3.5x105 cell/mL, which are typical values for the cell concentration in unchlorinated drinking water. Though water is supplied without residual disinfectant in the Netherlands, it was found that only ~40% of total cells in fresh water were alive/intact. As seen from Figure 1, the concentration of intact cells appears to be dependent on the temperature of the fresh water, as room temperature is generally stable ~20-21°C. Up till now, only negligible difference in water quality parameters was found between the two investigated DDWSs.

Figure 2 Modelled (blue series) vs. measured data (red series) for the cold (left) and hot (right) kitchen tap

As is evident from Figure 2, modelled temperature results agree well with the measured data, and thus, Epanet-MSX model might be enough accurate to be used for simulation of temperature in DDWSs.

Scientific relevance
In 2006 growth of opportunistic pathogens in DDWSs was recognized as a high priority research area by National Research Council, since the ability of pathogens to amplify is greatly favoured by increase in water temperature, available nutrients and long water stagnation time. Still, the scientific knowledge about the impact of DDWSs on water quality is very limited. This project will lead to a better understanding of the factors affecting the quality of drinking water within DDWSs.


Social relevance
High quality and microbially safe drinking water is essential for human health. Despite significant advances in drinking water treatment technologies, well treated water is not necessarily reaching consumer’s tap, as water quality might deteriorate in the distribution systems. A better understanding of the factors governing deterioration of water quality within DDWSs can help designing an effective control strategy that will ensure safe drinking water at the consumers’ tap.