*Not being a member of the Society of Naval Engineers, the author makes reply through the Naval Institute.
In the May number of the Journal of the American Society of Naval Engineers an article appeared entitled "Screw Propellers of U. S. Naval Vessels," by Lieut. C. W. Dyson, U. S. N. In criticising a given propeller from its performance the term "critical speed" was used to denote the break in the curve of "wake coefficient," where it changed direction and turned rapidly upward near the top speed, the basis of reference being as to the sufficiency of propeller surface. M is considered the wake coefficient, as determined from V = M S where S is the speed of ship, and V the "speed of the water in which the propeller works." V is determined from the formulas by the aid of the logarithmic curves, which give the same results as Barnaby's tables; hence, M contains the resultant compounding of the 10% wake correction already inherent in the logarithmic curves, and Is in error by that amount. Furthermore the definitions determining M appear somewhat different from the accepted theories.
It is my purpose to prove that the curves of M are derived from erroneous assumptions, and that the conclusions drawn from them are likewise in error. The correct curve that indicates the value of the propeller is the curve of slip, as was originally pointed out by me in the reports of the trials of the Torpedo Boat Destroyers Paul Jones, Perry, and Preble, over a year ago; I characterized the break in the curve of slip, as the point indicating the "critical speed," due principally to insufficiency of propeller surface to attain a higher speed without cavitation. I note that Mr. Dyson applies the identical method of criticism to the M curves, and I desire to prove that my method is correct. For the sake of clearness the curves presented in Mr. Dyson's article are reproduced in Fig. 1. The curves of speed and power arc evidently reproduced from my original drawings.
The curves of apparent slip are shown in Fig. 2; they were originally prepared and used in locating cavitation, the cause of vibrations and lack of speed, on the Perry, as well as in correcting these errors on the Paul Jones.
Towards the top speed the curves of slip change direction and turn upward; that marks the true point of critical speed. If the vessel's displacement be decreased, the critical speed would increase, and vice versa; and if the propeller surface were increased, the critical speed would increase, and of this, also, the reverse is true. In a given vessel the critical speed marks the point where the revolutions and horsepower increase abnormally for a small increment of speed; and vibrations begin, the cause being cavitation.
In Fig. 3 I have reproduced the Perry's curves from Fig. 1, and have added to it the curve of slip for this vessel taken from Fig. 2, but reduced to the proper scale. The so-called wake-coefficient curve M would indicate the critical speed at 23 1/2 knots, but there is nothing abnormal in the performance of the vessel at that speed. The break in the curve of slip indicates 25 1/4 knots as the point where abnormal conditions begin, beyond which the hump in the power curve appears.
In order to emphasize the correctness of my contention, the Perry's curve of revolutions and speed has been reproduced in Fig. 4; there is nothing wrong at 23 1/2 knots, but at 25 1/4 knots the point of disturbance begins.
Recurring to the power and speed curve of Fig. 3, the hump that occurs beyond the 257 knot point has the appearance of being due to the wave resistance of the vessel, and I so judged It at first sight. The depth over the course averaged only 12 fathoms, which portended that this hump might also be due, in part, to a shallow water resistance.
It required considerable investigation to eliminate the two last-mentioned, and other, probable causes before it could be stated definitely that insufficient propeller surface was the cause of this hump. In designing the propellers of the Paul Jones I followed these conclusions solely, with the result that the high speed was attained without vibrations, and without reaching a critical speed with its attendant evils.
It is evident that the wake-coefficient curve M is only correct in so far as it bears a general resemblance to the curve of slip, and is in error by the amount it differs therefrom.