The present paper is a continuation of the work described in the 4 th report (J. of Zosen Kiokai, vol. 63, 1938) on the effective height of coaming plates. In the first part of this paper an approximate solution of the problem of determining stresses in the perforated plate with an infinitely long coaming plate under tension, shear and bending in its plane respectively is presented. In the second part, it is stated that in order to verify the mathematical results, experiments with gelatine models were carried out with respect to some typical examples.

View full abstract

This paper describes the torsion test results and plastic characteristics of mild steel bars (C=0.053, Si=0.015, P=0.067, S=0.052%) with five different sections including a circle, three kinds of curved-sided regular triangle and a regular triangle. These triangles inscribe to a base circle of 10mm radius, the boundary shapes of which are obtained by computating from the following equation :
K (R_b³ cos 3θ-1) +1-R_b² =0,
where K is a profile constant.
Two sets of these test pieces were finished with accuracy only by the hands of skillful workman, using the special profile gauges, and were ground and lapped like a mirror surface. One set was broken by twisting. The other was only twisted until the elastic limit and further again twisted after thirty one months to ascertain the time effect upon the strength.
According to the test results, there were found two modes of a characteristic curve of torque-angle diagram during the yielding of material from an elastic range to a plastic. Along with the gradual change in section from a circle to a regular triangle, the curvature of the characteristic curve becomes larger, which means the gradual increase of the maximum shearing stress induced on the middle point of a curved-side. To a strange matter, thee happened some irregular plottings on the torque-angle diagrams, the reason of which will be necessary to seek for.
The results brought us the following linear relation :
Q_p= 0.78 Q',
Q'=0.342 (Q_max-500),
where Q_p= the moment in elastic limit in kg.cm,
Q'= the constant moment in plastic range in kg.cm,
Q_max=the maximum breaking moment in kg.cm.
Number of twisting until breaking varied also with the sectional area and the fractures were almost plane. The average values of modulus of rigidity for the same test pieces were as follows : -
The circular bar in the second test indicated the very high value of 11.98 × 10⁵ kg/cm², which was not included in the above figure. These facts will be taken as the time effect upon the strength of bars.
The stress ratio of zθ_max at elastic limit to zθ'_max (constant) in plastic range varies from 1.33 to 2.3 according to the gradual change of section from a circle to a regular triangle. As the true torque at elastic limit, however, was really lower than the constant torque in plastic range, the true stress ratio varies from 1 068 to 1 733. The torque formula used in this calculation was verified as reliable by these tests

View full abstract

In his previous paper, the author obtained the coefficient of vibration damping of a plate that is parallel to a water stream, and movable normal to the same stream, with a theory regarding the dynamic damping of the vibrations of a ship in motion-ahead. Although the vibration damping of any ship increases almost linearly with speed, since the vibrational force usually varies as the square of the same speed, the amplitudes of the ship vibrations are apparently augmented with speed increase. Since, however, no data on vibration test of an actual ship was given in the previous paper, there has remained some doubt as to whether or not the damping condition in a full-sized ship would obey the law that the author formulated.
In the present paper, with a view to confirming the results given in the previous paper, the author directs his attention to the results of the rolling experiments of the “Revenge” that was conducted about fifty years ago. Based on the data of his model, experiments with a plate vibrating in a circulation water tank that were published in his previous paper and also reexamined in this one, he predicted the damping coefficient of the rolling of H.M.S. “Revenge” in her motion-ahead (free of the actual coefficient of the vessel itself), from which he finds that, notwithstanding the large differences between the conditions of the vibrating model plate and the conditions that attended the rolling experiments of the “Revenge, ” in oscillatory frequency as well as in displacement amplitude, the damping coefficient of rolling of the “Revenge” predicted from the data of model experiments fairly agrees with the same coefficient that had been actually observed. The author thus concluded that the data of his model experiments can straightway be applied to the vibrations of a full-sized ship-a ship with the frequencies and amplitudes within the range of those model plates in the author's experiments. In extending his investigation to the case of the pitching motion of a ship, the reason for the pitching of a ship in her motion-ahead in a calm sea being damped more quickly than in her stand-still condition was ascertained. Since, as a matter of fact, the problem is merely qualitative, it was impossible for the author to give accurate values of the dynamic damping coefficients under consideration.

View full abstract

The writers, with the aid of mathematics, succeeded in obtaining some insight into the question whether, from the point of view of ship vibrations, the stiffness of the engine Beds should be rigid or elastic. It was found that, if the frequency of the natural vibration of the engine bed is higher than that of the ship's body, the amplitude under resonance that corresponds to the ship's fundamental vibration does not diminish, and that at higher engine speeds, another resonance that corresponds to the vibration of the engine system is likely to occur. If, on the other hand, the frequency of the natural vibration of the engine system were lower than that of the ship's body, the amplitude under resonance that corresponds to the ship's fundamental vibration will be greatly reduced and, moreover, that the resonance condition that corresponds to the engine system, will scarcely be encountered. Naturally, vibration damping should exist in the ship's body, and, possibly, also in the engine system. In the case of high speed engines, it does not follow that it is always safe to stiffen the engine bed as rigidly as possible.
It was also found that, in the case of ship vibration, coupled with vibration of the engine system, even should the position of the engine room be at the loop of the ship's vibrations, the amplitude at any point under resonance vibration of the ship's body, is not very marked unless the mass of the engine system is negligible, and also that the damping coefficients for ship vibration combined with the vibration of the engine system, are intermediate between the coefficient that is proper to the ship's body alone and the one that is proper to the engine system alone.

View full abstract

The present paper contains an empirical analysis of the equations of motion for the turning circle on the basis of the pivoting, point of a ship, and a comparison of the results of this analysis with those on a turning trial made with a self-propelling model and the ship itself.

View full abstract

In order to obtain the constant of a torsion-meter, it is reasonable to measure the relation between torque and reading by statical method in workshop. But, as mentioned by the Author in the prececeding paper, it is impossible to make the fixing end of shaft perfectly rigid by the twisting apparatus prevailed. This causes the displacement of fixing end and bending of shaft, so the torsion-meters of sleeve system are subjected to errors. The accurate value of the constant of a torsion-meter can be calculated, by using the modulus of rigidity, measured by the reliable strain-meters.
The accurate value of modulus of rigidity can be measured by the method of mechanical or optical lever. Since this method requires 4 to 5 m. clearance from both sides of shaft and four readings at different points, the Author has invented the present strain-meter, in order to avoid these defects.
This meter, which may be considered in principle as the combination of two sets of author's optical torsion-meters (2), is free from the errors and defects mentioned above.

View full abstract

Since the propulsive performance of a ship is predicted by the result of self-propulsion test, the high reliability is required for inboard propeller-dynamometer. As the construction is simple and the inertia is small, the present propeller-dynamometer gives very reliable results of self-propulsion tests in waves as well as in smooth water.
The following is specially noticeable among many features of this instrument. In general, the torque, caused by the friction of rotating system, which varies with load and revolutions, is present in the torque-reading of dynamometer, so it is necessary to carry out brake-test, which requires no little time and labour. When the present dynamometer is used, the daily brake-test can be replaced by the static test of the spring used, since the friction-torque has varied with revolutions only, as expected by its construction.
Though the method of recording torque electrically is of Author's invent, the method of recording thrust is under investigation.

View full abstract

Principal part of flooding calculation is divided into three parts. (1) To find the floodable volume so as to make the ship after flooding float at the trimmed water line which crosses the margin line only at one point, and the longitudinal position of its centre of volume. (2) With the given distribution of sectional area below that trimmed water line, to find the floodable length and the position of its middle point. (3) To find the end points of the floodable length curve. In this paper, the following methods are adopted to carry out these calculations.
As an extension of the displacement calculation, at several parallel water lines between the load water line and deep water line, the volume above L.W.L. (v), the distance of its centre of volume from midship (λ), the distance of the centre of flotation. from midship, and the corresponding longitudinal metacentric radius (R) are calculated, and the profile between the L.W.L. and margin line with the locus of centre of flotation of even keel water line (I) (scale of depth being taken as ten times that of length), sectional area curve below margin line (II), and the plan area curve of margin line (III) are prepared.
(1) Neglecting Leclert's correction, from the centre of flotation of an intermediate even keel water line between m.l. and L.W.L., two tangent lines (or lines through the end points of m.l.) are drawn towards fore part and aft part, and their inclinations α are measured.
By virtue of the relation
d=V₀+v/vR tan α+λ (V₀ : vol. below L.W.L.) the longitudinal distance d of the centre of floodable volume v from midship is calculated. (In this formula, d and α are assumed to be positive when referred to the fore part, λ being taken as positive for the centre of volume of layer v allocated after midship.) The draft d_e of corresponding even keel water line is also noted. With the base of d, curves of v (floodable volume curve) and of d_e (even keel draft curve) are drawn.
(2) As for a certain value of d, the corresponding v and d_e are read and with (I), through the centre of flotation corresponding to d_e, a tangent line to margin line is drawn. Now, d being assumed to be the first approximation to the position of the middle point of the floodable length, the sectional area A_m below margin line by (II), the height h_m between margin line and the tangent by (I), and the breadth b_m of the margin line plan area by (III) are taken at the position d, then the sectional area a₀ below the trimmed water line at d will be given by a₀=A_m-b_mh_m, and the first approximate value of the floodable length 2l will be v/a₀. By the same way, the sectional areas a_e and a_m at d±l can be obtained. Assuming the distribution of sectional area below the trimmed water line as a parabola of second order nearly within this range, the exact values of the floodable length L_F and the distance of its middle point from a₀, section Ej can be given by 2lL and 2le, in which L and e are taken from the diagram of Fig. 4 with the given values of μe=a_e/a₀ and μ_m= a_m/a₀.

View full abstract

During the repair of propeller shafts of a certain ship, which were docked in Nagasaki Works, Mitubisi Zyukogyo K. K. last year, many fine cracks were found out around the shaft surface coming into contact with the propeller boss in the vicinity of its fore end. As the results of investigation, these cracks were considered to be caused by neither ordinary corrosion nor fatigue, but by rubbing action with metal on metal, and it suggested the similar cracks hidden underneath the after end of the gun-metal sleeve. Therefore the after ends of the sleeve were removed for a few centimeters, and, as was expected, the similar cracks as above were discovered.
In this paper, the author explained that, judging from the peculiarity of these cracks, their cause should be attributed to the combined action of repeated rubbing and stresses, and proposed a method to prevent such cracks developing on the shaft.

View full abstract

In oil-tankers of Japan, it is very rare to have steam plants. “Kuroshio-Maru” which was recently completed has three main boilers of such a modern forced-circulating type as never been found formerly in our country. The steam condition of these boilers is 21 kg/cm² G and 350°C. Each boiler generates steam 17, 000 kg/h at normal and 22, 000 kg/h at maximum. They are of the Kawasaki La-Mont marine water tube type and each has heating surface of 333 square metres excluding air-preheater, and is equipped with five oil burners, La-Mont water wall, superheater, economizer and tubular air-preheater. These boilers were adopted for the purpose to save space and to improve the.whole plant efficiency. One of these boilers was tested on land and obtained the efficiency of 87.29%.

View full abstract

Hoyo Maru which was built by the Yokohama Dock Company as an up-to-date tanker was completed in November, 1936.
She was grounded on the beach of a coral reef island during a storm in January, 1937, with a full cargo of oil.
After her grounding, severe storms came along several times in close succession, and the ship rolled and rolled on the ground of the coral reef. Consequently, she was badly damaged and broken all over her bottom. Moreover, the bottom parts of the bow and stern sank into the crashed sand of coral reef, as they were shaped like wedges and were easy to get into the sand, while the bottom at the middle part was flat and remained on the sand.
Thus the storm deformed the ship into the state of “Hogging, ” and the deformation at midship was 24 inches upward at upper deck and 40 inches upward at keel. Ship's shell on both sides buckled and cracked vertically at midship. But the ship was salved successfully after two months' efforts, and she was brought into dry dock. Sufficient number of wooden blocks were built by divers on the dock bottom in water to support the ship on her irregularly damaged bottom and the heights of all the blocks were adjusted to touch the different parts of the bottom at the same time when the ship was lowered down to her position by pumping out of the water in the dock.
Special device was adopted for stretching out the contracted bottom lengthwise with gradual lowering down of the ship's body and also for sliding the bottom outward gradually at both ends. The repair job was completed in a dry dock and the ship was restored to perfect condition in 180 days after the job was commenced.

View full abstract