|The Cape Peninsula University of Technology (CPUT) Electronic Theses and Dissertations (ETD) repository holds full-text theses and dissertations submitted for higher degrees at the University (including submissions from former Cape Technikon and Peninsula Technikon).|
Evaluation of centrifugal pump performance derating procedures for non-Newtonian slurries
The performance of a centrifugal pump is altered for slurry or viscous materials (Stepanoff, 1969) and this needs to be accounted for. Usually, the suitable selection and evaluation of centrifugal pumps is based only on water pump performance curves supplied by the pump manufacturer (Wilson, Addie, Sellgren & Clift, 1997). In 1984 Walker and Goulas conducted a number of pump performance tests with kaolin clay slurries and coal slurries on a Warman 4/3 AH horizontal slurry pump and a Hazleton 3-inch B CTL horizontal pump (Walker and Goulas, 1984). Walker and Goulas have analysed the test data and correlated the performance derating both at the best efficiency flow rate (BEP) and at 10% of the best efficiency flow rate (0.1 BEP) to the modified pump Reynolds number (NRep). They have noticed that the head and the efficiency reduction ratio decreased for the pump Reynolds number less then 10⁶. Furthermore, Walker and Goulas obtained a reasonably good agreement (± 5%) between pump test data for non-Newtonian materials and pump performance prediction using the Hydraulics Institute chart. Sery and Slatter (2002) have investigated pump deration for non-Newtonian yield pseudoplastic materials. The NRep was calculated using the Bingham plastic viscosity (µp). Results have shown good agreement with regard to head and efficiency reduction ratios in comparison with previous work. However, Sery and Slatter's pump performance correlation using the HI chart did not reach the same conclusion. Error margin of ± 20% and ± 10% were found for head and efficiency respectively. This study is an attempt to reconcile the differences between Walker and Goulas (1984) and Sery and Slatter (2002) and extend the evaluation of these derating methods to pseudoplastic materials. The test work was conducted in the Flow Process Research Centre laboratory of the Cape Peninsula University of Technology using two centrifugal pumps; a Warman 6/4 and a GrW 4/3. The materials used were water, CMC solution bentonite and kaolin suspension at different concentrations (7% and 9% by weight for bentonite; 5%, 6% and 7% by weight for CMC; 17%, 19% and 21% by volume for kaolin).