886 



HYDRODYNAMICS IN SHIP DESIGN 



Sec. 78.17 



Direction of 

 RotQtio 



Plane of Propeller 

 Disc 



Fig. 78.L EMB Model Propeller 2294 



Fig. 78.0 is a plot of the observed total re- 

 sistances Rt for the transom-stern model, TMB 

 4505, both with and without turbulence stimula- 

 tion. Apparently, stimulation is not required on 

 this model, nor do the stimulating devices (short 

 studs) have any effect on the measured resistance. 

 Included in the diagram is a plot of the dimen- 

 sional ratio Rt/V^, where Er is in lb and V is 

 in kt. It is customary at the David Taylor Model 

 Basin to make this as a check whUe the resistance 

 testing is in progress. 



It was noted and commented on by the TMB 

 staff members responsible for testing these models 

 that their performance during the test runs was 

 uniformly excellent. When measuring resistance, 

 the models settled down quickly, allowing rapid 

 adjustment of the pan weights. Once the correct 

 weight was established there was Httle oscillation 

 of the resistance scriber from the zero Une. In the 

 self-propulsion tests the actual D/o weights used 

 on the pan agreed very closely with the calculated 

 Df values, resulting in remarkably low AD/ cor- 

 rections [C and R Bull. 7, 1933, pp. 37-38]. 



This consistency and reliabiUty, for both 



models, is believed to be due directly to the fact 

 that separation around the hull is limited to the 

 region abaft the transom and that the separation 

 points at the water surface always lie exactly at 

 the outer corners of the transom. On a form 

 having waterhne slopes both less and greater than 

 the critical slope for separation, the separation 

 points may and undoubtedly do lie at slightly 

 different positions for each run at the same speed. 

 The positions depend upon the "history" of the 

 viscous flow, on a distance basis, reckoned from 

 the bow of the model back to the points in ques- 

 tion. The separation points may not remam 

 fixed during the constant-speed portion of a run, 

 and they may not be opposite each other, on the 

 two sides of the model. 



The chemically traced flowlines on the surface 

 of the transom-stern model, depicted in Figs. 

 78.C and 78.D of Sec. 78.7, reveal definite flow 

 patterns. There appears to be no uncertainty 

 on the part of the water as to the paths that it is to 

 follow. There are no excessively wide, dark 

 traces indicating undesirable slowing up in any 

 region. 



