Bringing peace to troubled waters
The designers of a modification for hopper-bottomed clarifiers believe they have found a way to make significant improvements in the equipment's performance. The simple modification, developed by BHR Group Ltd, transforms the flow patterns within the clarifier, producing a much more stable and uniform sludge blanket. In a controlled trial conducted by Anglian Water, the modified clarifier produced on average, a 45% lower settled water turbidity than the unmodified clarifier.The problem of unstable sludge blankets is relatively common in hopper- bottomed clarifiers, particularly as operators try to squeeze more from their assets and raise through-flow rates to the design limits. During its efforts to increase the reliability of the clarifiers at Grafham WTW and produce a more consistent settled water quality, Anglian Water contacted BHR Group. BHR has extensive experience of flow modelling and hydraulic analysis in the water industry, and was able to offer a number of approaches to help solve Anglian's problem. These included site tracer studies, computational modelling using computational fluid dynamics (CFD) and physical modelling. Each approach has its strengths and weaknesses, but a fully validated CFD model was finally chosen.
The results from the converged CFD solutions were studied to determine the flow patterns and residence time distribution (RTD) through the clarifier. RTD is the result of a stimulus-response test whereby a 'pulse' of tracer is added to the inlet and its concentration monitored at the outlet over a period of time. The shape of the resulting plot, concentration vs. time, can provide valuable information on the type of flow pattern within the vessel.
In parallel to the computational modelling, salt tracer tests were carried out on site. A pulse of NaCl solution was added at the inlet of a single clarifier and the concentration, at various locations throughout the clarifier, was monitored over a period of time. Tracer tests were carried out at a number of flowrates and then repeated for increased confidence.
The close correlation between the site test and computational prediction confirmed the CFD model was producing representative results and could be used, with confidence, to develop possible solutions.
The flow patterns in the CFD model showed the problems were caused by streaming or jetting flow in the corners of the clarifier. This strong streaming flow set up large recirculations which tended to pull flow downwards in the centre of the clarifier. At low flowrates the jets in the corners do not reach the upper levels of the sludge blanket and therefore operation appears satisfactory. As the flow rate is increased, however, the jetting in the corners penetrates further through the sludge blanket until it reaches the upper levels. Here it not only creates a boiling in the corners but, in the centre the recirculating flow pulls solids downwards. It was thought this flow pattern leads to the instabilities in the blanket and to a blanket of varying thickness.
By studying these results, BHR Group was able to propose a modification to optimise the flow patterns throughout the clarifier. By eliminating the jetting and recirculation, it was hoped a more uniform, upward flow pattern would be produced. A second CFD model was used to assess the proposed modification. They showed a marked increase in uniformity with no recirculatory flows in the top two thirds of the clarifier. This improvement is also demonstrated by the change in shape of the RTD. The new RTD shows a much later initial trace and a later peak, which indicates improved plug flow characteristics.
Having produced a modified design, the next stage was to prove it worked by using a controlled performance trial. Anglian Water modified one of the clarifiers at Grafham WTW, and initiated a long-term trial comparing the modified unit with a standard one adjacent. Samples taken on an hourly basis from the outlet launders were pooled to create a sample covering 24 hours.
Three measurements of turbidity were taken on each composite sample. The three turbidity readings taken from each sample have been averaged to give a single reading for each day a sample was taken. The results from the trial show, on average, a 45% reduction in turbidity, a significant improvement. BHR Group has applied for a patent on the modification and is working with Anglian to assess its suitability for other sites.