A model design tool
Physical modelling is still a valid tool in design and construction of water industry assets. Dr Ray Baker explains why.
Physical hydraulic models have been used for many years to optimise the design of civil engineering hydraulic structures such as dam spillways, pumping stations, sewer drop shafts, weirs and river systems. Although computer simulation techniques have improved considerably in recent years there are still some features of water flow that cannot be properly mathematically modelled.
Even if the algorithms can be established, there are situations where water and air move in a three-dimensional multi-phase pattern. This would need such a fine mesh on a CFD model that the computational time would exceed the memory and processor capabilities of computers available in the normal engineering design office. It is in these circumstances that physical models still offer a valuable tool to the designer.
Pumping station wet wells are an area where physical models are still required if the designer wishes to be confident that the finished structure is free from problems of vorticity, pre-swirl, air entrainment and sedimentation. Complicated flow patterns can be set up by the geometrical inter-relationship between the flow inlet and the pump location, which in a multi-pump pumping station will often change with flow rate and the pumps that are in operation.
If the water starts to swirl, vortices can be established which may drag an air core from the water surface but more normally consist of a tight swirling core of water propagating either from the floor below, or from a wall adjacent to the pump suction inlet. Such vortices will cause inefficiency and vibration damage to the pump.
On a physical scale model they can be located visually using dye and eliminated empirically using baffles, splitters or cones near to the pumps or occasionally by more serious changes to the chamber geometry.
For sewer pumping stations and other sewerage structures where sedimentation could be a problem especially sized plastic granules can be used to simulate grit sediments. This allows a visual assessment of points where sediment may be deposited and changes to be made to the shape of benching and other features of the structure to keep sediment moving.
A physical model allows changes to be assessed for very little cost and with ultimate confidence that the final design will give a sediment free solution. Making changes directly on the finished structure is both difficult if it is in operation, expensive to perform and the designer’s ideas cannot be guaranteed to work.
In some structures such as sewer drop shafts the physical model may not only be used to evaluate the design and check that it adequately stills the flow and is sediment-free. It may also be used to measure pressures and hence gain an indication of forces on a chamber at the bottom of the shaft.
In addition to giving confidence in the design process physical models can save money on the final construction. This is especially common for large structures such as spillway design and evaluation where it has been shown over the years that on many schemes the cost of the model will be paid back in savings in the scope of construction work that is required on site.
Physical models in civil engineering are usually built at scales between 1:5 and 1:10 for sewerage structures, pumping station sumps and the like and at larger scales around 1:20 to 1:30 for spillways and bigger structures. Hydraulic factors are scaled using Froude Number similarity, which compensates for the fact that the water itself cannot be scaled.
There are various competent test laboratories that can carry out hydraulic modelling and most will manufacture pumping station sumps and sewerage structure models from clear plastic allowing full visualisation of the flow in the chamber. Spillway models may also be made with clear plastic or timber channels but will often contain representations of adjacent ground surfaces fabricated in timber, fibreglass or mortar.
Bigger models do take a few weeks to construct and the designer should commission the hydraulic model as early as possible in the design process to avoid testing becoming rushed. A recent trend of testing the model whilst the scheme is under construction is not recommended because often the model shows that features of the prototype that are already built could be better sited or sized.
This normally results in a compromise solution to any operational problems that could have been solved completely had the model been commissioned earlier.
If a hydraulic structure of some size or flow complexity is being designed then consider a physical hydraulic model once initial design ideas have been formulated.
The resulting tests will allow optimisation of the detail design of the structure, give confidence that it will work as intended and in some cases give sufficient savings in the cost of construction of the final scheme to pay for the cost of the model work.
Dr Ray Baker is principal engineer at CRM Rainwater Drainage Consultancy.
T: 01204 701934.