Abstract
The measured variation of losses in a mixed flow pump have been analysed to determine the fraction V of the losses that are dependent on Reynolds number and the exponent n that characterizes that dependence. The variation of the Reynolds number dependent fraction has been obtained for different operating points on the pump characteristic. It is shown that the experimentally determined points are fitted very well with several combinations of V and n within the experimental range and it is strongly suggested that the exponent n be chosen on the basis of a-priori theoretical grounds. The importance of specifying a model reference Reynolds number is pointed out. The present formulae used for efficiency scale-up are most useful in the region of rapidly changing efficiency below the knee in the efficiency–Reynolds number curve. They are not applicable in the region of most practical interest in Reynolds numbers above the knee where prototype pumps operate.
Get full access to this article
View all access options for this article.
