Optimum blade number and splitter blade length of a mixed-flow impeller based on mean streamline loss model

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In view of slip factor and splitter blade effects on internal losses of mixed-flow compressor impellers, a modified loss model is presented based on mean streamline aerodynamic analysis. In allusion to a mixed-flow impeller with a high inlet hub/tip ratio and a low aspect ratio, optimum blade number and splitter blade length for minimum head loss are investigated by modified model and CFD verification. A new empirical formula for picking blade number is proposed. Research results indicate that modified loss model can predict the mixed-flow impeller performance. At design flow rate point, recommended blade number scope is 40 to 42 and the splitter length scope is 0 to 20% of main blade length. Peak efficiency mainly depends on minimum loss superposition of blade loading, skin friction and tip clearance. In low blade solidity cases, blade loading loss and clearance loss dominate in internal losses. With increasing blade number and splitter length, skin friction loss becomes the major loss. In addition, blade number and splitter length have complex coupling which is inappropriate to be simply described by equivalent solidity or effective blade number.

Original languageEnglish
Title of host publication2018 Joint Propulsion Conference
PublisherAmerican Institute of Aeronautics and Astronautics Inc, AIAA
ISBN (Print)9781624105708
DOIs
StatePublished - 2018
Event54th AIAA/SAE/ASEE Joint Propulsion Conference, 2018 - Cincinnati, United States
Duration: 9 Jul 201811 Jul 2018

Publication series

Name2018 Joint Propulsion Conference

Conference

Conference54th AIAA/SAE/ASEE Joint Propulsion Conference, 2018
Country/TerritoryUnited States
CityCincinnati
Period9/07/1811/07/18

Fingerprint

Dive into the research topics of 'Optimum blade number and splitter blade length of a mixed-flow impeller based on mean streamline loss model'. Together they form a unique fingerprint.

Cite this