The purpose of this paper is to examine the side-wall effects on the ship motions among waves in an experimental tank. We can see the extent of the side-wall effects from the results of numerical examples in this paper.

View full abstract

In this final report, the measured total resistance of eight floating models S-101·102·201·202 (strictly corresponding models) and A-101·102·201·202 (approximately corresponding models) is given together with the calculated total resistance for the former four models (S-series models). Through these comparisons the following conclusions are obtained :
(1) For the purpose of quantitative application of the wave-making theory, we have to treat with the boundary condition on the surface of a model more exact than the so-called Michell's approximation.
(2) The viscosity of water has a great importance to the wave-making of after hull body. Its effect can be simply represented by (β, δ) correction which is to be applied to the asymptotic characteristics of the stern free waves in the far rear of the model.
(3) Besides the above, the following corrections are also found as indispensable :
(3 a) The sheltering or the self-interference correction factor α' due to the finite breadth of the hull which is to be applied to the pre-caused bow waves.
(3b) The finite amplitude correction factor γ which is to be applied to the full forms with large angle of entrance.
(4) Thus the calculated wave-making resistance coefficient is finally given as :
C_ω= η₁C_ω (1) +η₂C_ω (2),
where C_ω (1) = the steadily increasing or fundamental term in C_ω
C_ω (2) = the oscillating or interference term in C_ω, in whose argument the shifting correction factor δ being introduced
η₁=1/2 (γ²+β²) = γ²/2 (1+β'²)
η₂ =α'·γβ=α'·γ²β'
with β' = β/γ
(5) The correction factors which have been obtained from wave resistance comparison show a reasonable relation with the fineness of the models as well as with Froude number.
They also show good coincidence both with the observed wave profiles and with the photographic observations of the stern wave separation.

View full abstract

In measuring the heel angle of a ship under her inclining experiment, she is generally inevitable to oscillate more or less owing to the external forces such as wind and wave.
Even in such a case the measuring method by a long pendulum which has its own oscillation is generally still in use. Therefore, when accuracy is especially required, it may be said that, if the above method being taken, it will be very difficult to distinguish exactly the small heel angle induced by the moving weight.
Now the authors have lighted upon the optical method by a camera in place of a pendulum to measure the heel angle of a ship in such a oscillating case. This optical method will be very simple and accurate.
In this paper, general investigations in camera are described together with some test results with four actual ships obtained by our optical method.
From these test results you may also perceive that the exact rolling period of a ship can easily be obtained, if necessary, without the rolling test.

View full abstract

Recently, many successful trials for the statistical analysis of irregular ocean waves have been made by oceanographers, such as Pierson, Longuet-Higgins, Neumann and others, while the statistical investigations of the ship motion in irregular seaway have been promoted concurrently under the cooperation of oceanographers and naval architects.
Many brilliant results were produced by this cooperation, for instance the papers by Pierson and St. Denis, Lewis and others.
The authors have been working on the same problem by model experiments at a seashore, where we built a tower to get a stable platform.
From the platform, we measured the waves by a wave recorder, the ship's rolling angle by a pantagraph and the wind pressure by a wind pressure gauge simultaneously, changing the ship's encounter angle and it's natural rolling period.
Having applied the statistical analysis to the records obtained, we found that the statistical method given by Longuet-Higgins, Pierson and others gave the outstanding results.
Being encouraged by these results, we have tried to extend the same method to a case where irregular wind pressure as well as irregular wave moment affect a ship.

View full abstract

In order to investigate the conditions for temperature and stress which govern the propagation of brittle crack in mild steels, the authors performed special double tension tests using specimens made of 12mm thick rimmed steel. The specimens were composed of two parts, initial crack was started by statical tension in small attached part so as to reach at the edge of main part, where this initial brittle crack was tested whether it would propagate through the test piece or not under various combinations of tensile stress and temperature.
In the other part, the mechanism of propagation of brittle crack through mild steel plates was discussed. Namely, assuming a slip formed at the root of a notch by concentration of stress may finally develope to a small crack and calculating the variation of stored energy in the plate during this change, it was found that a brittle crack once started will propagate indefinitely or in other word the energy stored in cracked plate may diminish with the increase of crack length at low temperature and high stress, while at hihger temperature and lower stress there is a minimum energy at certain crack length and the initial crack will stop at this length. At certain temperature and stress, the energy curve may have a inflexion point with horizontal tangent at a definite crack length, and under this condition the critical stress and temperature for propagation of brittle crack may be defined. The relation thus obtained between the critical stress, critical length and temperature were compared with experimental results and it was found that there were good coincidence between these two.

View full abstract

In the welding of structure, when a pneumatic air hammer happened to be in use (for instance, rivetting, calking, chipping) in places nearby of where the welding was going on, in some instances, a crack was produced in the weld.
The authors, by using the apparatus which is illustrated in the Fig. 1, have reproduced its phenomenon in a laboratory and at the same time investigated the main factor which produces the crack, as well as the effect that is given by the vibrations to the toughness of welded joint and obtained the following summaries.
(1) Generally speaking, when vibration is transmitted to the joint during its welding, the weldment would be more susceptible to cracks when the air pressure of pneumatic tool used is more higher, and the gap between the steel member is more larger, and the length of the tack weld is more smaller.
(2) Crack sensitivity of the joint during its welding becomes larger under vibration caused by rivetting hammer than by chipping hammer and so becomes when fillet weld is done than when butt weld is carried out.
(3) Because of such facts it is absolutely necessary to give a full attention to the vibration of joint members produced by a pneumatic tool in the case of a fillet welding. Even in a butt welding it is advisable to avoid such vibration for the first layer of weld, but from the second layer vibration will bring no unfavorable result to the weldment.
(4) The cause of the development of cracks due to the vibration is not based on any metallurgical change, but it is conceivable to be a hot crack due to the mutual change that took place between joint members.

View full abstract

Following to the preceding report, the author has solved numerically the wave pressure problem acting on parabolic waterline ships with infinite draft fixed amongst waves.
Firstly, the steady drifting force has been obtained when the wave propagates at right angle to the ship's logitudinal center line. (Fig. 4, Table 3.)
Secondly, the surging force has been obtained in these cases, that is, when the wave propagates parallel to and at right angle with the ship's longitudinal center line. (Fig. 8)
For this calculation the wave height around the ship surface is necessary and calculated. (Fig. 5, 6)
When the wave length tends to infinity we are to remind a virtual mass effect of the ship, remembering Rayleigh's approximation in such case of other wave problem.
It is observed that the consideration for this connection leads us to the appropriate result of the surging force in the range of fairly large wave length.

View full abstract

The main problems treated in this paper are as follows
(1) Comparison of the measured value in NACA tank with the calculated by the author's, method.
(2) A calculating method for characteristics of the submerged hydrofoil of a dihedral angle. In §2, the calculated lift and resistance are compared with the measured in NACA tank, from which it can be safely said that the accuracy of the calculating method proposed is reasonably good for the ordinary aspect ratio of the submerged hydrofoil
In §3, a calculating method for characteristics of the submerged hydrofoil of a dihedral angle is proposed from the standpoint of the liftingline theory and wave resistance theory. Numerical calculations are made and it is shown that the numerical results are in good accordance with the American's empirical formula and Tinney's results.
In §4, the characteristics of the submerged hydrofoil of vertical finite span are examined in connection with the ship rudder which can be obtained as a special case.
In §5, the surface waves caused by the submerged hydrofoil of a dihedral angle are examined.

View full abstract

This paper sets forth the results of tank experiments on the hydrodynamical force upon a circular cylindor in uniform flow, which was measured directly and dynamically by the use of a specially designed strain gauge-type dynamometer. As the aim of experiments was to study the dynamical behavior of tall stack under the action of wind, they were carried out over the range covering the critical Reynolds number, R=0.86×10⁵
According to the test results, the drag force is stable in spite of the fluctuation of the lateral force and the Strouhal number referring to the lateral force coincides with the Relf's results which were obtained from the fluctuation in the wake flow. The drag coefficient C_D, the ratio of single amplitude of fluctuating lateral force to drag force, C_R. and the Strouhal number S versus the Reynolds number R are shown by the curve with mark TUET in Fig. 1.
From these results it is supposed that beyond the critical Reynolds number R≈2×10⁵ a realistic value of C_L is in the order of 0.1. This value, as compared with the value C_L≈1.0 suggested by Den Hartog and Fung, is considerably smaller, and is of the same order of C_L_??_0.20 by Ozker and Smith.

View full abstract

The authors have reported in 1951, on a system of indicating shaft-horsepower with a wattmeter to which an electric type torsionmeter and an electric tachometer are connected. Since then, further researches have been made with the intention of consummating the shaft-horsepower meter, especially we made our efforts to the simplification of its control, the maintenance of its accuracy, and the diminution of its cost.
In this paper, the progress of researches made on improving the shaft-horsepower meter is reported.

View full abstract

This paper deals with the resistance and self-propulsion tests in still and rough waters on a series of three geometrically similar models for a super tanker.
The results of resistance tests in still water show that Froude's method of calculating a ship's resistance has in principle lost none of its applicability if a proper friction line is chosen.
According to the propeller open tests and self-propulsion tests there will exist a slight scale effect on the smaller model propeller.
The results of tests in waves show that virtual mass coefficient is almost constant independently of length of models, and that there are little scale and wall effect in such a range of model length, model speed and wave conditions as in these experiments.

View full abstract

In order to clarify the effect of transverse curvature of the body surface on its frictional resistance, flow along the outer surface of a circular cylinder is studied. Applying the modified momentum transfer theory developed previously by the authors, the velocity distribution is found with curvature effect (Fig. 1). Substituting it into the momentum equation, the differential equation in terms of η₁, ratio of the sublayer and the boundary layer thicknesses, position-and radius-Reynolds numbers is obtained. After numerical integration, the frictional resistance is calculated in network of the two Reynolds numbers. The results, compared with the experimental data of Hughes, are shown in Figs. 3 and 4. It is interesting that the good agreement between the theoretical and the experimental results can be got in this old fashioned, as it were, mixture length theory.

View full abstract

The paper records the result of experiments with a systematic series of five-bladed aerofoil propeller models, designed in accordance with present-day practice.
Charts of series are shown by so called √B_p-δ design diagrams.

View full abstract

The resistance to rolling is of great importance to investigate the safety of vessels in rough seas. The passive resistance due to the action of the water is mainly composed of three kinds, i.e., the frictional resistance, the eddy making resistance and the wave making resistance Of these resistances only the wave making one was shown in the author's previous paper to follow the law of comparison for similar ships. It is therefore necessary to establish formulae for the frictional and eddy making resistances in order to extend the results of model experiments to actual ships.
In the present paper new formulae for the frictional resistance are obtained after analysing the results of experiments on the oscillations of suspended cylinders wholly immersed in the water. The frictional coefficient C_f for oscillating cylinders is found to be expressed in the form exactly same as that given by H. Blasius for laminar flow, if Reynolds number R_n be taken as 3.22r²θm²/T_v
The decrement of roll per swing due to friction for a ship model without bilge keels in fresh water is given by the formula
dθ_f=2.11Sr_s²θ_m/WmT^1.5 at 15°C in kg, m, sec, deg, units : and the decrement for a ship model with bilge keels is written as
dθ_f=2.11Sr_s²θ_m/WmT^1.5 {1+0.143 (r_sθ_m) ^0.772/T^0.386} at 15°C.in which the frictional coefficient for turbulent flow is used from the following formula given by G. Hughes
C_f= 1.328R_n^-0.5 + 0.014R_n^-0.114. The value of δθ_f for a ship in salt water is shown as
dθ_f=2.22Sr_s²θ_m/WmT^1.5 {1+0.141 (r_sθ_m) ^0.772/T^0.386} at 15°C.
Though the mean radius rs for the wetted surface may be calculated by using the formula r_s=√1/S∫∫r²sin²αdldx, it is easily and in close approximation obtained from the following formula
r_s=1/π {(0. 877+ 0.145C_b) (1.7 d+ C_b·B) -2×OG},
where C_b is the block coefficient, d the mean draught, B the breadth of ship and OG the value of (d-KG).
The ratio of δθ_f to the total decrement δθ at the mean angle of roll of 5 degrees for ordinary ship models 2 m in length is ranging from about 7 to 16 per cent, while the ratio at the mean angle of roll of 20 degrees is from 3 to 7 per cent. This ratio for actual ships is found to be between one tenth and one fifth of that of ship models.

View full abstract

The author has analysed in the previous paper the fact that statistical distributions of heights of waves are very complicated and various owing to the sea states and weather conditions, and that by a parameter n, where n>2 represents “regular swells”, n<2 “violent seas”; and for the case of n=2 is “confused seas and swells” which had already pointed out by Drs Longuet-Higgins. Statistical amplitude of rolling of a ship on such sea surfaces are :
θ=F (βs) ·Φ (n) ×θ0,
where θ0 in perfect resonance amplitude due to regular waves, F (βs) and Φ (n) is factor of correction due to irregularity of waves, namely
[F (βs)] n=βs∫xs0 [I (x) /I (O) ] ndx, I (x) ≅x2e-βsx2/√ (x2-1) 2+2Nsγsδmx2e-βsx2, Φ (n)=q [Γ (1+1/n) -n√lnq∞j=1∑ (-1) j (lnq) j/j! (jn+1)] + (1-q) n√lnq (average value)
Γ (1+1/n) [1-Nj=0∑ (-1) jNCj/n√j]
or n√lnN [1+1/n∞j=1 (-1) jj-1S=0Π (1-sn) ·1/j (lnN) j] ∫∞0e-z (lnz) jdz+etc, (maximum value)
q=3 corresponds to Drs. Sverdrup-Munk's significant mean value, N is numbers of swings of rolling. Ns is damping coefficient, γs is effective slope coefficient, βs wave age for reasonance ; and xs= Tw/Ts, Tw is maximum or most frequently observed periods of waves, Ts is ship's natural period of rolling. δm=0.140 is the maximum value of wave steepness. Here the author assumes that density of wave spectrum between β and β+dβ are
1/2ρgr2 (β), r (β) =π/g (βu) 2δ.
and spectral steepness δ (β) is the same that of wave steepness. In the estimation I (x) the author used Dr. Neumann's steepness curve δ=0. 140 e-β2. Cumulative wave energy is then expressed by the following formula
E=∫βm01/2ρgr2 (β) dβ, βm=gTw/2πu where βm can be analysed by Drs, Sverdrup-Munk's theory. u is wind velocity

View full abstract

A thin elastic shell in form of a circular cylinder is assumed to have small initial deformation_ The shell is assumed to suffer the following loads in succession : (a) At a state of static equilibrium, it is under the action of a normal pressure q₁, together with an axial thrust P. (b) During the period 0<t<t₀, the shell is suddenly acted on by an additional normal pressure q₂. (c) At time t₀<t onwards, the pressure q₂ ceases to act, and the static pressure q₁ only acts upon the shell.
In Part I of this report, the author has shown a theoretical formula which gives the amount of deformation (vibratory) of the shell at three stages (a), (b) and (c) as mentioned above. In this Part II, the author takes up a special case of a shell whose diam. is equal to 10m, the thickness 6cm, and the length 15m. The numerical values concerning the deformation of this specific shell is obtained, based on theoretical formula as mentioned above.

View full abstract

For the purpose of obtainning the approximate formulae of estimating the natural circular frequency a_n, the logarithmic decrement δ_n, and the amplitude A_n caused by the exciting force of known magnitude F, the author has carried out some investigations, the results of which are as follows.
(1) a_n=√EI₀g (m⁴_n+X_n) /w₀l⁴ (1+Z_n) (1+r_n)
(2) δ_n=πa_n {1/1+r_nα+r_n/1+r_nβ+1/a²_n (1+Z_n) (1+r_nγ)}
(3) A_n=Fl³/EI₀πu_n (x) u_n (l₁) /δ_n (m⁴_n+X_n)
The results of calculations by the use of these approximate formulae to the actual cargo ships on which the author had carried out the vibration tests by means of vibration exciter are compared with the data experimentally obtained, and it is shown that if the magnitude and the location of exciting force on board a vessel are given, it is possible to estimate not only the natural frequency but also the order of amplitude of her vibration.

View full abstract

An analysis is made of damping factors in the higher modes of ship vibrations. The solutions of the fundamental equation of the vibration of the uniform beam taking into account the effects of shear deflection, rotational inertia, internal damping force due to both of normal and tangential viscosities of a ship structure and the external damping force are obtained. The damping factors represented by logarithmic decrement in ship vibrations including higher modes are calculated from the solutions under the assumption of some numerical values of the damping coefficients. Some of the damping factors measured from the actual ships are compared with those obtained by the present theory. The empirical formula for estimating the damping factors in the higher modes of the hull vibration suggested by Dr. J. Lockwood Taylor is verified by the present theory.

View full abstract

The statical longitudinal bending moment acting on a ship is affected by many factors, which relate mutually. In spite of existing this relation, the functional expression which describes sufficiently how the bending moment depends upon all these factors supposed to concern it seems to have not yet been accepted.
The author studied this functional relation using two non-dimensional quantities : “load distribution coefficient κ” and “weight-displacement ratio ω”, as stated in the technical report No. 3 of the Nippon Kaiji Kyokai.
In Section II, the calculation formulae for the bending moment at midship are expressed using κ and ω due to each affecting component.
In Section III, based on the relations obtained in Section II, the effects of the various factors which mainly affect to the bending moment at midship are examined, such as fineness, hull and outfit weight, weight in machinery space, weight in pump space, amount of fuel and fresh water, amount of cargo weight, position and length of machinery space, position and length of cargo hold in superstructures.
In Section IV, the characteristics of the bending moment at midship which are inherent to the kind and type of ship are clarified.
In Section V, by considering its preferable characteristics, the optimum general arrangements of ships are discussed from the point of view of the statical longitudinal bending moment.

View full abstract

Using Pierson-Neumann's wave spectrum theory and the theoretical response function of bending moments of ships in regular oblique waves, the spectrum of bending moments of ships in a seaway was calculated by St. Denis Pierson's ship motion theory and its detail was studied. The obtained deck stress was compared with the several observed stress on actual ship in an ocean.
The results showed that the bending moment of ships moving in the direction rectangular to the wind is about 60% of that moving against wind. (Fig. 5)
The “equivalent wave height” was introduced, which produces the same amount of bending moment by conventional bending moment calculation as that of the statistical values. For shorter ships (ship length is less than 150 m), the wave height of 1/20 of wave length corresponds to the “equivalent wave height” of average of about 1/4 highest value in the 50 knot fully developed sea.
For the increase of ship length, the “equivalent wave height” increases. However its rate of increase decreases. Its amount lies between 1.1√L (ft) and L/20. (Fig. 8) An approximate formula of the “equivalent wave height” is given as L/20+0.13 (L/100) ³ where L is ship length in meter.

View full abstract

In the previous report, the authors investigated the interaction between deckhouse and main structure, and in this paper the elastic behaviour of superstructure was researched.
In the case of superstructure whose sides are continuous to the side shell, the interaction between superstructure and main structure depends on the tensile (or compressive) rigidity of their side plates in the vertical direction, while in the case of deckhouse the vertical reaction between house and hull is affected by the bending stiffness of the upper deck. Then an evaluating method of the stiffness factor for superstructure was presented.
The method for estimating the effective area of deckhouse shown in the previous report was examined to be applicable to that of superstructure by the comparison of the calculated values with the measured values which were obtained by the various model experiments. Using the stiffness factor and the effective area evaluated by these methods, the result of the bending test of aluminium superstructure was analysed.
The simplified formula for the bending calculation of the superstructure was presented and its numerical values were compared with results calculated by the successive approximate method and the validity of the formula was discussed.

View full abstract

Many results of the researches on structures containing brackets have been published by various researchers, while there have been scarcely any experimental data which justify these results. The author was able to conduct the tests on a perfectly fixed beam by using an acrylic material. According to the test results, it was found that the portion at the brackets did not follow the results predicted by the beam theory and the calculation currently used with a changing section resulted in an overestimation of the rigidity at the brackets. Hence, in this case, it is desirable to replace the brackets by the assumed effective brackets, as described in the present paper, and calculate with a changing section of the new brackets.
In the present paper, an approximate formula which had been obtained by introducing the idea of a span point was used to simplify the calculation on a fixed beam with brackets.
It was also found, as a result of the tests, that a lightening hole in the bracket had small effect on the flexural rigidity of the bracket, while a change in a shape of the flange on the bracket had a remarkable effect.

View full abstract

Commencing about the middle of the 20 th century, considerable attention was given by researchers to the problem of obtaining the ultimate strength of metal columns with residual stresses. In this paper, the ultimate strength of a column with residual stresses is described. A “modified Shanley model” is also studied to estimate approximately the maximum load of a column and to show the influence of the initial deflection on the column behavior.
The test results were referred to show the experimental verification.
Conclusively the following is found, under the existence of considerable residual stresses like cooling residuals or welding residual stresses, the ultimate strength of a column may be considerably reduced, but might be much higher than the tangent modulus load calculated by means of a cross-section test (stub column test).

View full abstract

The second series of slow notch bend test was performed to investigate the effect of residual welding stress on brittle fracture of steel. Chemical composition and mechanical properties of steel tested are shown in Table 1. The following 6 types of test specimens were prepared (Fig. 1).
(1) parent metal, as rolled
(2) parent metal, stress annealed at 650°C, 1 hour
(3) welded specimen, as welded
(4) welded specimen, stress annealed
(5) welded specimen, peened at internal passes
(6) welded specimen, low temperature stress relieved
On each one specimen among welded specimens (3) - (6), residual stress measurement was-performed by resistance wire strain gauges. Longitudinal stress are shown in Fig. 8 and Table 5. Fracture test procedure is the same as Van der Veen test. Fracture appearance and shear fracture percentage of deposit metals are shown in Figs. 9 and 10. As seen in Figs. 13-15, the results of V-Charpy tests on as-welded and annealed deposit metals found little difference, so we can conclude that stress annealing of welded joint increases notch toughness. Peening at internal passes showed little advantage. Low temperature stress relieving rather decreased notch toughness in our case.

View full abstract

Experimental and theoretical studies on the distribution of residual stresses in butt-welded joint were performed in this research. The details of distribution of strain change due to welding and also that of elastic strain release occured by cutting after welding were measured by Gunnert's apparatus for residual stress measurement. General view of specimen and measuring points are shown in Fig. 1. Experimental results are shown in Figs. 3 to 9.
Theoretical investigations upon the distribution of residual stress were also tried, and a new approximate method to calculate the distribution of residual stress was suggested. Elastic dislocation occured by transverse shrinkage and plastic deformation locally produced arround the welded zone are the two main origins of residual stresses in butt-welded joint. Distribution of residual stress due to elastic dislocation can be calculated as a problem of elasticity, however, it is very difficult to estimate the distribution of residual stresses caused by plastic deformation occurred in the region near the weld zone, as these deformations are caused by complicated change of stress during heat cycle due to welding. An attempt was, therefore, made to estimate the stress distribution due to plastic deformation locally concentrated near the weld. The method of analysis attempted in this study is quite resembled to that treated in the “boundary layer” theory in aerodynamics.
The calculated stresses were compared with experimental ones, as shown in Fig. 8, and it was proved that this theory is quite acceptable for the estimation of stress distribution in butt-welded joint.

View full abstract

Recently, wooden ships with outer boards of double diagonal system were often constructed, and it was proved that they showed higher rigidity in comparison with ordinary wooden ships.
In order to investigate the rigidity of a board of double diagonal system, a series of model experiments were carried out, and the results were analysed with the following method.
The authors calculated the displacement of an elemental plank of the board by the “minimum energy theory” under the assumption that the elemental plank of the board were perfectly rigid and then the working energy was conserved in rivets alone, as the wooden structure with rivets shows a large slip between their constructional members.
In this paper, the authors deal with the case under a tensile force. The calculated values of the displacement are compared with the experimental values with good results, as shown in Fig. 4. The distribution of shearing forces acting on the rivets is tabulate in Table 1.

View full abstract

The strength and rigidity of joints in wooden ships depend, first of all, on the bearing strength of wood under compression.
In this paper, the bearing strength of wood under compression and also the apparent modulus of compression E' are investigated considering the fine structure of wood. Structural calculations are made on the assumed models of fine structures of wood and show satisfactory coincidence with the anisotropic behavior of Young's modulus and the bearing strength of wood.
The “tracheid” simulated models, which are made up of laminated paper, are tested under compression and fairy explains the state of contact surface. From these model experiments, it is explained that due to the local collapse of fine structures which are ruined about 1-3 mm from the surface by sawing and reduced to 1/10-1/15 in the rigidity under compression, the apparent modulus of compression E' become smaller than Young's modulus of compression E. And the anisotropic behavior of Young's modulus are also well explainable with the above assumption of fine structure of wood.

View full abstract

The experiments of protection with magnesium anodes were carried out using three oil tankers. In these experiments, it was found that booster anode was necessary in cathodic protection. Thereafter we have studied the effect of booster anode in small experimental oil tanks. By use of booster anode the current requirements for achieving polarization become 70% as compared with non use of booster anode.

View full abstract

Fatigue properties of as-rolled and shot-blasted mild steel plates were investigated in air, in sea water and under impressed current cathodic protection. By shot-blasting the fatigue limit in air rises about 20% over that of as-rolled plate. The reasons why the S-log N diagrams of two surface conditions are close to each other in the corrosion-fatigue test and the fatigue limits under cathodic protection are lower than those in air were considered.

View full abstract

Recently, many damages by corrosion have occured on tailshaft sleeves of ships built after the War.
The corrosion damage is confined to segmental area on sleeve and numbers of these corroded grooving have some relations with propeller blade number.
This fact suggestes that these corrosion damages are caused by the vibration of tailshaft or hull excited by propeller.
As one of causes of these corrosion damages, we can suppose cavitation errosion grown up by a vibratory hydrodynamic pressure set up in cooling water when a tailshaft or a sterntube vibrates.
So as to ascertain this phenomenon, we have tried experimental research using a model with full scale shaft diameter. And we recognized that the cavitation occurs at frequency of about 30/s. This experimental result suggestes that the cavitation will grow up in sterntube when a tailshaft vibrates impulsively on actual ship.

View full abstract

One of the authors, S. Yamada, as a member of the 8 th Research Committee of the Shipbuilding Research Association of Japan, studied the statical strength of four throw solid crankshaft on actual bed plate. But precise investigation into the effect of web dimensions could not be found by the above study, because the results were considerably affected by bearing conditions as well as other unknown factors.
In this investigation, therefore, single throw crankshafts having same web section modulus bt²=0.417 d_c³ (d_c= crank diameter) have been examined by means of the apparatus so contrived as to eliminate the above influences.
From the results of the examinations, following conclusions have been deduced :
(1) In load bending test, maximum fillet stress of crankshaft having web thickness t=0.56d_c, is greater by about 10% than that having web thickness t =0.45 d_c.
(2) In web deflection test, however, maximum fillet stress is nearly equal in all crankshafts in spite of the change of web thicknesses.
(3) The increment of web distance in both load bending and web deflection tests may be calculated quite accurately by correcting the length of crank radius (r) by a factor q (q <1).
It is adequate to take q = 0.50 for the crankshafts used in this investigation.

View full abstract