Answers are grounded in this paper and cite their sources.
THIS paper presents the results of one series of tests included in a comprehensive program of research on hydraulic friction and roughness being carried on at the University of California. An investigation was made of the friction coefficients of closed rectangular channels having the same absolute roughness. The walls of the channels were made of smooth brass artificially roughened by sixty degree V-grooves transverse to the direction of flow. The effect of relative roughness on skin friction is important in the extrapolation of experimental results on models to the prototype. At present Nikuradse's equation based on data of the pressure drop in pipes artificially roughened with sand is used for this purpose. The results of these tests provide data to justify this procedure. The friction coefficients of four channels having hydraulic radii ranging from 0.030 to 0.230 feet were determined for different rates of flow. The channels were constructed of two brass plates one-quarter inch by five and three-quarters inches by eight feet and spacers one-sixteenth, one-eighth, one-quarter, and one-half inch wide. Both plates and spacers were roughened their full length by the grooves machined 0.060 inches apart with an average depth of 0.0125 inches. Practically, the absolute roughness of the different channels was the same since the relative width of the plates, which were the same in all, was large compared to that of the spacers. The large aspect ratio of the channel also insured nearly two-dimensional flow. The plates were held in place by heavy iron castings running the full length of the channel. Fig. 1 is an end view of the assembled channel. Five piezometer holes were located nineteen and one-half inches apart along the center-line of one plate. Pressure drops were measured between adjacent holes with vertical or inclined, mercury or carbon-tetrachloride manometers. The results given are for the downstream pair of taps only, pressure drops farther upstream being appreciably affected by entrance conditions. Rates of flow were measured either by an orifice calibrated in place or by direct weighting. A second series of tests were made on the one-eighth, one-quarter, and onehalf inch set-ups after the original one had been dismantled, and the results checked with those of the first series. In the turbulent region where the friction coefficient for rough surfaces does not depend on Reynold's number, a function, containing two empirical constants, FIG. 1. End view of roughness channel showing brass roughness plates and cast iron supporting frame.
No takes yet. Share an insight, caveat, or question.
V.J. Skoglund (1936) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: