Title

Session

Authors

Abstract

There exists uncertainty about the accuracy with which the fan scaling laws can predict the performance characteristics of a full scale prototype fan from model testing. This uncertainty stems from the large difference in the value for Reynolds number that typically exists between the prototype and model fan. Increasing the turbulence intensity of the inlet flow during testing of the model fan is a suggested method for improving Reynolds number similarity between the two scenarios.
The M-fan was designed as a low pressure rise, high volume flow rate fan, for application in a specific Concentrated Solar Power (CSP) plant. It has been shown to exhibit two dimensional flow across its blades far from the end walls. Assuming that the M-fan’s solidity is low enough so that the blades can act as isolated aerofoils, allows for a comparison between the effect of increased free stream turbulence intensity on the performance of the M-fan and the lift and drag characteristics of a two dimensional aerofoil. The similarity between increasing the Reynolds number of the flow and increasing the turbulence intensity of the flow is demonstrated on a NASA LS 0413 aerofoil, used in the blading of the M-fan, through two dimensional aerofoil testing in a wind tunnel. Increased free stream turbulence intensity is shown to result in a clear delay in stall whilst increasing the maximum coefficient of lift, with the effect becoming more prominent as the turbulence intensity increases.
Results from model testing of the M-fan at its design, and moderately increased, blade setting angles is shown to be independent from any increase in inlet flow turbulence intensity. Thus, the fan test results are shown to be Reynolds number independent. Model testing at a very high blade setting angle showed a steady increase in fan static efficiency with an increase in turbulence intensity. This is similar to increasing the Reynolds number at which the fan operates, thus validating the experiment. These results indicate that, at the scale test of 1.542 m diameter and 720 rpm, the M-fan operates independent of the Reynolds number. The use of the fan scaling laws on results obtained from model testing of the M-fan under these conditions will yield accurate results when scaled to represent the fan characteristics of the full scale prototype fan.