U. S. Navy Ship-Model Powering Correlation 1982 to 1995

The idea of an adjustment to model test procedures in order to improve ship powering predictions was initially introduced many years ago as a roughness allowance because it was recognized that the basic methodology involved in predicting power from model tests yields powering estimates for ships whose hulls are smooth. Thus an increment to the smooth ship resistance was added to account for the ships’ roughness and this increment was called the roughness allowance. Later on it was recognized that the difference between the predicted and measured power was due to roughness as well as other factors and the concept of a correlation allowance coefficient designated by the symbol CA was developed for engineering and design purposes in order to obtain the most accurate ship powering predictions. The value of the correlation allowance for a specific ship is determined from the analysis of ship and model powering data. The correlation allowances for specific ships forms a database or experience factor that helps guide the powering prediction for a proposed new design. In the years following WWII, the U.S. Navy ships still had riveted hull plating or a combination of riveted and welded plating and many ships used a hot plastic finish which resulted in a relatively rough ship hull. Paints with smoother finishes were just being introduced. Hadler et al [1] discussed the impact of paint type and construction method on CA and showed that in the 20 years leading up to 1960 there was a general decrease in the CA. It was common practice at the time to assume a constant value of CA equal to 0.0005 for all ship sizes. By the early 1980' s the common ship construction practice was an all welded ship with a relatively smooth paint finish such as the Navy Milspec vinyl. The Navy ship-model correlation data was reassessed and summarized by Hagen et al [2]. The earlier data with hot plastic 205 were purposefully disregarded and a correlation allowance instruction which is a function of ship length, see Fig. 1 , was issued by Naval Sea Systems Command (NAVSEA) [3] and defined for U.S. Navy powering predictions according to the standard David Taylor Model Basin (DTMB) powering prediction method documented by Grant and Wilson [4] . The primary purpose of this paper is to examine the ship-model correlation data gathered since the early 1980' s and to see if there is a need to modify the NAVSEA CA curve. Other purposes are to examine the accuracy of the RPM prediction using the David Taylor Powering Prediction method, and to discover if there are any ship or model characteristics in addition to length that could influence the selection of CA.