JOINT DISCUSSION ON TWO PRECEDING PAPERS. 107 
3. Miscellaneous :——Struts and bossings; cutting away of deadwood; rake of shafting 
and trim of vessel; location of rudder behind screw; speed-length ratio, etc. 
In view of the large number of factors involved, it does not seem feasible to treat 
their application in elementary form. It would appear more practical to divide all vessels 
into six or eight main groups, preparing for each group suitable data and coefficients for tak- 
ing care of practical variations. 
On page 98 the author recommends the use of the vertical prismatic coefficient of the 
vessel as the basis for estimating the wake fraction. Mr. James Semple, in a paper read 
before the Institution of Naval Architects in 1919, used the prismatic coefficient of the after 
body for this purpose. As far as I can see, the forward and the parallel middle bodies will 
have no appreciable effect on the hull coefficients. It seems desirable therefore to consider 
only the after body. The prismatic coefficient of the after body alone will not be sufficient 
unless there is a standard relation between all of the elements given above. Otherwise it will 
be necessary to provide data to allow for such variations. 
Referring to the tests at various draughts, it would be of value if the author would give 
the prismatic coefficient of the after body for each draught. 
It would also be of interest to know how Prof. Bragg’s data compare with those given 
by Mr. Luke in 1910 and 1917, and by Mr. Semple in 1919. 
Referring to Plate 30, Fig. 4, it is noted that the block coefficients of these particular 
models have very little effect. This seems to indicate that block coefficient alone is not a 
very suitable basis, and that considerable errors can occur in certain cases if the other ele- 
ments, as noted above, are not allowed for. 
Referring to Plate 35, Fig. 23, it is noted that the wake curves for the wing positions do 
not agree with those for outward turning screws given by Luke in 1910. According to his 
curves, the wake increases with the diameter. For inward turning screws, Luke’s curves 
agree with the author’s. 
The large wake caused by a thin board in front of the wheel, as shown in Plate 36, Fig. 
27, is more than one would suppose. In this case the projected area of the board on the disc 
area of the wheel is one-quarter of the latter, and the wake is 20 per cent. This gives some 
idea of the considerable effect of struts, bossings, and deadwood in so far as they retard the 
flow into the wheel. 
These tests show that the effect of the position of the wheel is very great. Therefore, 
in preparing propeller estimates, I believe that more consideration must be given to allow- 
ing for this feature. 
The curves and data given in this and other papers, which cover wake and other hull 
coefficients, are very complex and, in fact, almost confusing. It would be most useful and 
desirable to reduce these data to some more simple and usable form for regular use in pre- 
paring estimates; in other words, to standard estimating data. 
This paper also emphasizes the difficulty of making estimates accurately from general 
data, and the importance of self-propulsion tests which represent exactly the conditions to be 
met with. However, self-propelled model tests will not be truly valuable until they are proved 
to be reliable. This will not be until a number of test and trial results are compared, co- 
ordinated and published. é 
Prof. Bragg has thrown a great deal of light on a subject which greatly needed it. 
However, the present knowledge of this subject is by no means complete, and it is hoped 
that research work of this nature will continue. I am sure that the Society is very apprecia- 
tive of Prof. Bragg’s work, 
