104 SOME EXPERIMENTS ON PROPELLER POSITION, ETC. 
37, and it will be observed that it is of rather a full type. It is hoped that at some future time 
if an opportunity permits, the stern lines may be further fined and the experiments repeated. 
Fig. 2, Plate 38, shows the three model propellers used. With the possible exception of 
571, they were smaller than would be used for the actual ship in practice. This permits a 
maximum vertical change of position and insures that the propellers work at a high real 
slip and cover a wide variation of slip. Fig. 3, Plate 39, supplemented by Table I, shows the 
test positions for the propellers, three positions at the upper level, three at the lower level, 
and three propellers used in each position. It was then necessary to make eighteen sets of 
runs at each speed of model. Figs. 4, 5, and 6, Plates 40, 41, and 42, show the usual character- 
istic curves of the three propellers as tested apart from the model. In making the tests with 
the propellers attached, the model was self-propelled at the desired speed, the torque and 
thrust being determined. From this there were deduced the two quantities w and ¢. The 
average speed of the wake over the propeller area is the speed of the ship multiplied by w, the 
wake fraction. The thrust deduction coefficient ¢ is a quantity such that the actual thrust 
multiplied by the factor (1 —t) is the resistance of the vessel without the propeller. The hull 
efficiency, as usual, is expressed by 1 — t/1 — w. 
Tests were made at a number of speeds extending from 2 knots speed of the model to 
2.8 knots speed of the model, but the variations in the results with speed were not sufficiently 
marked with this model to be significant. 
Fig. 7, Plate 43, shows the reduced results in the small tables, each located with its center 
in the same position relative to the ship (indicated by the stern outline) as the propeller 
centers. As is to be expected from a single set of experiments on such a small scale, there 
are some inconsistencies in the results. It would require other experiments enabling cross- 
fairing to eliminate them. Broadly speaking, however, Fig. 7, which is confirmed by the 
results at other speeds, shows that the dimensions and proportions of the screw had a minor 
effect upon the hull efficiency. The very wide variation of slip resulting from the variation 
in screw dimensions had comparatively little effect. At each level, as the screw was placed 
further and further aft, the wake fraction and thrust deduction coefficient fell off steadily. 
These changes rather neutralize one another, so that in this case it does not appear there was 
any material gain by variation in the fore and aft position of the propeller. When, however, 
we consider the vertical variation, it is another story. For the upper locations the hull effi- 
ciency is consistently higher than for the lower, the gains being somewhere between 15 and 20 
per cent. This is in accordance with what might be expected from theoretical considerations. 
It would seem that the naval architect is confronted by another case of conflicting ten- 
dencies where a compromise is necessary. For the majority of vessels virtually all consid- 
erations, except that of hull efficiency, dictate the lowest practicable position of the propeller. 
Doubtless in sea-going vessels with reciprocating engines the considerations dictating maxi- 
mum submergence must preponderate, but for vessels with types of propelling machinery not 
subject to racing in heavy weather, and for vessels primarily for smooth-water service, we 
cannot ignore the fact that the higher the propeller location the greater the hull efficiency 
and hence possible efficiency of propulsion. 
I say “possible” efficiency of propulsion because to také full advantage of the greater 
wake associated with high propeller locations, the high propeller should be of somewhat 
greater diameter than the low propeller for such a vessel as Model 2441. 
In conclusion, I should like to record my appreciation of the assistance in the prepara- 
tion of this paper given me by the Model Basin Staff, headed by Captain Eggert (C. C.), 
