Home > Research > Publications & Outputs > Computational and experimental analysis of the ...

Associated organisational unit

Electronic data

  • 2017karlsendaviesphd

    Final published version, 12.6 MB, PDF document

    Available under license: CC BY-NC-ND: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

Text available via DOI:

View graph of relations

Computational and experimental analysis of the effects of manufacturing tolerances on the performance of a regenerative liquid ring pump

Research output: Thesis › Doctoral Thesis

Published
Publication date2017
Number of pages231
QualificationPhD
Awarding Institution
Supervisors/Advisors
Thesis sponsors
  • Gilbert Gikes & Gordon Ltd
Publisher
  • Lancaster University
<mark>Original language</mark>English

Abstract

The regenerative liquid ring (RLR) pump is a type of rotodynamic pump which combines the mechanical impulses of the impeller with centrifugal force. The head is increased through recirculation, or regeneration, of the fluid in the impeller blades. The design of the regenerative machine has evolved over many decades and improvements have been made in the design, particularly in the shape and number of impeller blades. However, few publications exist on other design aspects such as the radial and axial clearances between the impeller and the casing, and more crucially most researchers tend to investigate these elements in isolation. This highlights the key driver and novelty of this current research, which applies modern Computational Fluid Dynamics (CFD) tools to systematically assess the effects on pump performance over a large range of impeller-casing clearance parameters and understand their operational behaviour. The sizes of the clearances are dictated by the manufacturer’s drawing tolerances on critical pump components.

The results of the parametric clearance analysis showed that there could be a potential variation in head performance by as much as 25%. Furthermore, it highlighted the need to experimentally test a number of pumps within a representative tolerance range. Three test pumps were therefore manufactured and tested. A discrepancy between the CFD and experimental results was observed but the experimental results confirmed that variations in clearances have an impact on the performance, particularly on the developed head. This was more noticeable at higher operational speeds where the pressure differential varied by as much as 19%.

A one-way Fluid-Structure Interaction analysis of the three test impellers was carried out to shed some light on the operational behaviour of the pumps and the effects on the clearances due to non-uniform loading by importing pressure profiles from the corresponding CFD simulations. Assessing the deformation against the computational and experimental pressure, the results indicate that the maximum allowable radial design clearance should be reduced by 35%. Furthermore, with this radial design tolerance the magnitudes of the drive end and non-drive end clearances have a lower impact on the performance. The RLR pump is unique and requires further comparison studies between experimental testing and computational analysis.