Know your sludge
If you want to make the right choice when selecting a reliable pump for sludge, you need to know your rheology. Dr Nigel Heywood and Dr Thanos Sotiriadis of BHR Group explain
Pump selection depends on sludge volume flow, pump head/discharge pressure, inlet pressure or net positive suction head, and the sludge rheological properties. The only way to make reliable estimates of the pump pressures involved is to measure the relevant sludge rheological properties under the right test conditions.
These pressure changes are the sum of the frictional pressure drop through the pipeline section and the various pipe fittings, and the pressure loss or gain due to the changes in pipe elevation. The sludge friction losses through pipework are estimated using a wide range of empirical or semi-empirical equations. Many of these equations depend on the rheological model used to characterise the sludge and on the flow regime: laminar or turbulent.
The relationship between the frictional pressure drop through a straight pipe section and the sludge volumetric flow throughput depends on this flow regime and the sludge rheology, through the rheological model used. Hence, estimates of the frictional pressure drop for laminar or turbulent sludge flow can be only made with confidence provided that reliable rheological raw data are available. Such an example is the laminar pressure drop model suggested by Chilton and Stainby (1998).
Sludges with a dry solids content of less than about 1% or 2% have similar hydraulic characteristics to water, i.e. they flow essentially like water. Water is a Newtonian fluid, so its viscosity is constant for a given temperature. This is shown in Figures 1 and 2 where the Newtonian rheological flow curve is a straight line passing through the origin. But as the solids content of the sludge increases, it becomes increasingly difficult to pump. This difficulty is accompanied by a change in the sludge's rheological properties and a shift to non-Newtonian behaviour (the shear stress is not proportional to shear rate). The non-Newtonian character of the vast majority of sludges can be quantified for process optimisation and design using appropriate viscometers. Sewage sludges are generally shear thinning fluids (Figure 2) i.e. the viscosity reduces with increasing shear rate. The sludge viscosity affects all the operational parameters of a pump: total head developed, throughput and efficiency.
A general relationship relating shear stress (t) to shear rate (g), which describes the rheological behaviour of more concentrated non-Newtonian sludges, is described in the Herschel-Bulkley model (sometimes known as the generalised Bingham model):
This model is widely used to characterise sludge flow behaviour because of its generality and there have been successful attempts in the literature that investigated pipe pressure loss dependence on the Herschel-Bulkley parameters (Chilton and Stainsby, 1998). Table 1 summarises the other common sludge rheological models and shows that they are just special cases of the Herschel-Bulkley model.
Both Herschel-Bulkley and Bingham fluids contain a yield stress parameter (t0), which represents a finite stress, required to achieve flow. This is an important parameter since essentially it determines the pressure that a pump has to generate during sludge start-up inside pipelines. K is the consistency coefficient and is usually expressed in Pa sn and n is the flow behaviour index, which is dimensionless.
Two main types of rheometer for measuring sludge flow properties exist - tube type and rotational type. Tube types are generally single-pass devices with the sludge passing through the system just once, although recirculating flow loops are also used. They are best used for time-independent fluids. Rotational types can be controlled shear rate, controlled shear stress and/or oscillatory. Time-dependency can be evaluated with rotational instruments. Alderman and Heywood (2004) give a good summary of rheological measurement instruments.
There is a clear lack of a reliable source of sludge rheological data in the literature for process design work. In addition, there are no reliable predictive correlations for sludge viscosity in terms of solids content and sludge type, particularly for new sludge types resulting from new processes adopted in the water industry. These new sludge types include hydrolysed and pasteurised sludges and chemical-dosed sludges. None of these newer sludge types is covered by the 1980s TR185 report issued by the WRc, and in widespread use throughout the industry.
BHR Group has responded to the need for a reliable source of sludge rheological data by developing a comprehensive database of sludge rheological properties (Figure 3). This database project forms part of the Water & Wastewater Mixing consortium project (WWM) funded by water companies and suppliers of equipment and services.
Ideally, of course, the rheological properties of specific sludges should always be measured, if possible, and BHR Group provides a comprehensive sludge rheology measurement service to industry based on well equipped laboratory as well as portable rheometers for on-site testing.
Sludge samples can be collected on site and couriered to BHR's laboratory.
Rheological measurements can be carried out at appropriate temperatures and shear rates using rotational viscometry. The sludge's dry-solids content can also be assessed. The measured rheology data can be provided raw or fitted to the appropriate mathematical models for subsequent use in engineering design.
However, direct test work on actual sludge samples is not always feasible. This can be because of lack of equipment or skilled staff to carry out such measurements, or because there are strict time constraints-deadlines to be met that prohibit such measurements. It is then that the sludge rheology database becomes a powerful tool in the hands of plant operators and engineers.
The new database is being established from a combination of existing data and sludge samples from the current WWM members and includes entries for:
- Sludge process type and origin
- Type of polymer and chemical dosed in the upstream processes, and the dosing flow rates
- Dry-solids content
- Rheometer type used, together with the subsequent associated correction factors of the measurement procedure
- Shear stress versus shear rate data (including the raw data of angular speed versus torque)
- The fitted rheological model parameters using different approaches in the literature and an assessment of the accuracy of each data set
- Energy savings through lower power and maintenance costs by improvements in equipment selection and sizing (e.g. pumps)
- Reduced risk of costly equipment failure
- Lower Capex through elimination of over design