One or two points are worth mentioning in connection with this new formulation. 



In the first place, the Conference was careful to label the line a "model-ship correla- 

 tion line," thereby emphasising that the members did not consider it to be a line representing 

 the skin friction of the hull nor of an equivalent plank. It is, as the resolution states, "for 

 practical engineering purposes," and may be taken as including some allowance for form 

 effect. At the time of the 1957 Conference, a great deal of research was in progress on the 

 problem of extrapolation from model to ship, and it was generally felt that great developments 

 were likely in the not too distant future— hence the emphasis on an "interim solution." 

 These will probably take the shape of a three-dimensional system of extrapolation, allowing 

 for the effects of hull form and proportions upon the viscous resistance. Such methods have 

 been proposed, but the profession will no doubt wish to gain experience in their use before 

 making what will be, after all, a radical departure from the practice of nearly a hundred years. 



Secondly, the values of C p quoted above show that the 1957 ITTC line is everywhere 

 steeper than the 1947 ATTC line, and it is this slope which is important in the extrapolation 

 problem. Since the ITTC line is higher over the model range, the C^ values derived from 

 the model results will be smaller. When added to the ITTC C p values over the ship range, 

 which are nearly equal to or less than the ATTC values, those C^ values will result in a 

 lower prediction of the total "smooth" ship resistance and corresponding ehp. This will 

 apply whatever the model scale, but the effect will be larger with small models run at low 

 Reynolds numbers. 



In the third place, in adopting the new correlation line, the ITTC made no recommenda- 

 tion regarding the ship correlation allowance to be used in predicting ehp for the actual ship, 

 contenting itself merely with a general recommendation to continue work "to improve model 

 and ship correlation" and "to determine roughness allowances." In adopting its line in 1947, 

 the ATTC considered a number of model-ship correlations available at that time and while 

 recognizing the sparseness of the data and the possible dependence of the allowance on a 

 number of factors other than roughness, did finally recommend a "roughness" allowance of 

 + 0.0004 for all ships; this allowance has been used since in all published work based on the 

 ATTC 1947 line. For most merchant ships of the seagoing types, the resultant ship ehp did 

 not differ much from that obtained using the same model results and the Froude coefficients. 

 For the same model results and the same full-scale ship trial results, the use of the ITTC 

 line will call for a somewhat greater correlation allowance than would the use of the ATTC 

 line in order to obtain the same agreement. If a new three-dimensional method of extrapola- 

 tion is devised in the future, the value of the correlation allowance necessary to reconcile 

 the same model and ship results will have a still different value. It is very obvious, as pointed 

 out by the present author in 1957, that this factor is not just an allowance for the relative 

 roughnesses of model and ship but involves such things as the method of extrapolation used, 

 the relative sizes of model and ship, scale effect on wake, thrust deduction, and propeller 



E-2 



