98 



HYDROnYN WIK S I\ SHIP DESIGN 



Sec. 45.6 



because of tlio nugmeiit of velocity +At" to Ijc 

 expecteil itt the potential flow abreast the wide 

 portion of the body or ship. 



This lack of essential data makes it impossible, 

 more often than not, to determine the boundary- 

 layer tiiifkncss by the delta-velocity method 

 described in Sec. (5.5 of \"olumc I. An example of 

 this situation is found in the upper right-lmnil 

 diagram of Fig. 45. D, giving three partial velocity 

 profiles for a large liner, taken from unpublished 

 data kindly furni.'shed by J. P. Comstock and 

 C. H. Hancock of the Newport News Shipbuilding 

 and Dry Dock Company. The data in all cases 

 are from observations with a Pitometer speed log. 

 Unfortunately, the log-tube extension was in this 

 case only suflicient to obtain readings of local 

 relative velocity U ecjual to the ship speed V at 

 23. G kt, with clean bottom and in deep water. 



The observations at 23.G kt, with dirty bottom 

 in deep water, indicate greater local velocities U 

 in the inner portion of the l>oundary laj'cr and a 

 considerabl}' thicker laj'er, as diagrammed for a 

 typical foul-bottom boundar3--layer velocity pro- 

 file in Fig. 22.H of Sec. 22.13. Again the full 

 boundarj'-laycr thickness could not be explored 

 because of the limited log-tube extension. 



For the test at 18.85 kt, with a dirty bottom 

 and in a depth of water of 2.24 times the draft, 

 the local velocities U are less than for the higher 

 speed with clean bottom in deep water. Considera- 

 tion of backflow in the shallow water, plus surface 

 roughness, but taking the reduced speed into 

 account, indicates that the local velocities should 

 possibly be higher rather than lower. In anj' case, 

 the boundary layer is thicker, as it should be. 



The boundary-laj'er velocity profiles for the 

 small destroj-er (the Swedish ]Vrangcl), shown 

 inverted in Fig. 45. D, are taken from tho.se pub- 

 lished by II. F. Nordstrom [SSPA Rep. 27, 1953, 

 Fig. 39, p. 85], with the horizontal scale modified 

 to suit the features being illustrated. These 

 velocity measurements were, like those on the 

 large liner, made below the ship but fortunately 

 the pitot-tube extension was greatly in excess of 

 the j/-distance where U ~ U^ = V. With the 

 relatively small submergence of the pitot-tube 

 orifice in the case of the Wrangcl, it is possible that 

 there is some wave wake being measured with 

 the friction wake. Indeed, it is possible that the 

 wave action in the ca.se of the large liner makes 

 itself felt not only down the side but under the 

 bottom U.H well. 



Both high-speetl deep-water profiles on Fig. 

 45. D show a characteristic flatness — almost a 

 hollowness — at a ratio U/Um of about 0.90. The 

 Laute and Gruschwitz profiles of Fig. 22.C in 

 Sec. 22.(5 are actually hollow. 



(Jthcr ships upon which velocity-profile observa- 

 tions have been made, including the Wrangcl, 

 are the: 



(1) Hindcnburg, German merchant vessel, early 1920's. 



The (lata, as reported by W. Dnhlmann, II. Hoppe, 

 and O. Schafor |\\TIH, 7 Sep 1920, pp. 415-4191, 

 were taken witli a resistance log towed abeam from 

 a boom, and are rather sketchy. 



(2) Hamilton, U. S. destroyer (DD 141), 1933-19.34. The 



TMB data are unpublished except for one model- 

 ship velocity profile comparison by E. A. Wright 

 (S\.\ME, 1940, Fig. 24, p. 393). 



(3) Ctairton, U. S. merchant vessel, 1933-1934. The 



TMB data are unpublished. 



(4) Tannenberg, German merchant ship, 1938. Some 



boundury-laycr profiles were measured on this ship 

 but they were not published |WRH, 1.5 Jun 1939, 

 pp. 107-174). 



(5) Sanln Elena, merchant ship, 1951. V'^elocity profiles 



for both rough and smooth hull surfaces were 

 measured amidships, 72 meters (236.2 ft) from the 

 stem and 4 meters (13.1 ft) below the at-rest 

 wat«rlinc [Kempf, G., and Karhan, IC, HSV.\ Rep. 

 200, 1952, Fig. 7A; TMB transl. available; STG, 

 1951, Vol. 45, pp. 228-243). 

 (0) Snaifell and Ashworth, British cross<-hannel steamers. 

 Observations by G. S. Baker [NECI, 1929-1930. 

 Vol. XLVI, pp. 8:5-100 and Pis. III-V; .also pp. 

 141-140). 



(7) Wrangcl, Swedish destroyer. Observations by II. F, 



Xordstrom and assistants ["Full-Scale Tests with 

 the Wrangcl and Comparative Model Testa," 

 SSPA Rep. 27, 1953). 



(8) Victory ship, APS, Tcrvacle. Observations by G. 



Aertsscn and his assistants [IN.V, 1953, pp. J21-J5G]. 

 Fig. 3 on p. J25 and Fig. 22 on p. J52 (a revision of 

 Fig. 3) show velocity profiles made with a rodmetor 

 speed log under various conditions. 



(9) Lucy Ashlon. Pilot traverses were made by a Pitometer 



log (INA, 1955, Vol. 97, pp. 543-545 and Figs. 13, 

 14). No velocity profiles show U > Ua, . 



I'ur llie sake of completciics,s, the references of 

 Sec. 22. G of Volume I are repeated here, along 

 with a few others. These, however, apply to 

 tests on models only: 



(a) Laute, \V., STG, 1933, pp. 402-460; T.MB Transl. 53, 



Mar 1939. This paper covers flow testj<, velocity 

 measurements, and pressure observations on a 

 cargo-ship model. 



(b) Hamilton, \V. S., "The Velocity Pattern Arnuml a 



Ship Model Fixed in Moving Water," Doctorate 

 Diss., IIIIR, Dec 1943. Dcscrilx-s flow measure- 

 mi'iils made on a mmlel of the (icrnian M. 8. 

 Sun Frnncixco. 



