Experimental investigation to determine equivalent sand grain for very rough hydraulic surfaces

Abstract

Considering the actual roughness shape, distribution and orientation found on hydraulics surfaces, obtaining a sand grain equivalent for loss evaluation remains a challenge. To investigate this problem, a variety of surfaces ranging from hydraulically smooth to a very rough surface representing a heavily rusted penstock were tested on an experimental cylinder rotating in water. The larger-scale roughness was replicated through rapid prototyping. The surface effect on the flow for an increasing Reynolds number was measured and the resulting equivalent sand grain was validated against experimental intake head loss measurements.