94. COMPARATIVE TESTS OF BILGE KEELS AND A GYRO-STABILIZER 
i 
were two and one-third times the quenching moments was the gyro unable to stabilize. 
We have known since the experiments on the Worden in 1912 that the gyro would 
quench rolling when the rolling moments were at least one-third greater than the quench- 
ing moments, but we never knew that it could still be depended upon when the rolling 
moments were twice as great. 
Very great interest centers in the last of the paper, starting at paragraph 9, where 
Commander McEntee confirms our calculations of a year ago that the power required 
to move ships through the sea under conditions of rolling was about 1 per cent per 
degree of heel, added to which are the other losses referred to above. 
With reference to Plate 32, during the year we have found ways to still further 
reduce the very small power required by the stabilizer, and that power is only on when 
there is a sea, and only in proportion to the sea. You will notice here in our latest 
stabilizer how very small the motor is as compared with the wheel, and this motor, you 
understand, is yet 100 per cent too large, made oversize for the purpose of quickly 
spinning up the wheel whenever there is a sea in prospect. This contrast in size of motor 
and wheel gives a very good idea of the little power required. 
THe CHAIRMAN :—Is there any further discussion? If not, I will ask Commander 
McEntee if he desires to reply. 
CoMMANDER McENTEE:—With reference to the point brought up by Mr. Smith, 
I understand that he desires to distinguish between the losses due to the rolling of a 
single-screw ship as compared with those of a twin-screw ship. Although we have made 
no experiments along that line, I should imagine that the drop in propulsive efficiency 
would be greater with the twin-screw ship than with the single-screw ship. 
With regard to the general proposition of losses in the speed at sea compared with 
the speed obtained in the model test, my experience agrees with that stated by Professor 
Sadler, namely, for a slow-speed ship or for a medium-speed ship the actual sea speed 
over a period of six months or a year will be about 10 per cent less than that estimated 
from the model test. 
Of course I do not understand from that that this is a criticism of the model-basin 
results, because the sea conditions are so variable, the condition of the ship’s bottom is 
so changeable, that there is no very real ideal standard to compare it with except that 
which is absolutely normal and where the conditions are ideal, and that is in the model 
basin. In other words, you get the model-basin results and know they are the best you 
can hope to get. Your experience must indicate how much of a percentage you will 
allow to cover the average sea conditions. 
I might remark, with regard to the query of Mr. Smith, that we made experiments 
in the model basin a number of years ago which showed the decided influence that 
fouling has on the resistance of the ship. These experiments were not absolutely con- 
clusive, in that they indicated the upper limit, but the increase in resistance of the ship, 
even a month or two months after being painted, is really startling. I think Mr. Fer- 
guson can confirm me in the case of the Pennsylvania, which had been out of dock only 
two months, and the builders’ trial showed a considerable power increment over that 
required when the vessel was tried immediately after she had been painted. 
So far as determining the actual increase in resistance of power required, due to 
yawing, rolling and all the miscellaneous items which go to make up the increase of 
