LOADS ACTING ON A SHIP AND THE ELASTIC RESPONSE OF A SHIP 



279 



Table 6 Main Particulars ol^ Modei 1723 (Destroyer) 

 (in feet) 



Length, overall 5.821 



Length between perpendiculars 5.710 



Load waterline length 5,717 



Beam niaximmn . 608 



Drafts: 



Forwarc.l , 107 



,\lt 0,220 



Displacement, 11) (F\V). .-.:■.■ 24.50_ 



Center of gravity, abaft amidships 0, 175 



Longitudinal radius of gyration 1 ,o4'J 



Natural period of vertical two-node 



vibration, cps 10,9 



Natural periods of oscillation: 



Heave" 02 



Pitch- 0,00 



" In seconds by manual oscillation in calm water. 



sagging bending moment is ob.sei'ved at a .ship speed 

 exceeding 20 l<;nots. In the pre.sent case, however, a 

 corresponding decrease of the hogging bending moment is 

 shown. The measurement of the bending moment in 

 smooth water demonstrated that this change is caused by 

 the formation of the ship-made wave. Fig. 33 shows 

 the bending moments with still water moment subtracted. 

 The curves are now seen to be smooth and show a gradual 

 increase of the sagging bending moment with speed. 

 It appears to be established that the total bending 

 moment can be considered as the sum of the Ijending 

 moment developed in smooth water and the moment 

 developed by waves and ship motions. 



Fig. 35 shows a sample of an oscillograph record. A 

 conspicuous feature of it is the jjersistence of oscillations 

 of the bending moment about a mean (appro.ximately 

 sinusoidal) value. These oscillations were disregarded 

 and the mean values were shown iit Figs. 32 and 33. 

 The oscillations, however, appreciably increase the 

 maxima of the bending moment. They have the same 

 period as the two-node period of the ship's vibration. 

 It will be recalled that Sato also commented on the 

 occurrence of such oscillations. 



Three conspicuous features of the bending-moment 

 behavior are present in a very fast ship. These are the 

 gradual increase of the sagging bending moment with 

 speed, a sharp hump in the bending-moment curve 

 apparently caused by the flare impact, and the e.s.sentially 

 continuous state of vibration. These featiu'es are ab- 

 sent in a typical cargo ship. The vibration caused by 

 slamming will be considered later in Section 5.5. 



Lewis and Dalzell's (1958) work on a destroyer model 

 included also a comprehensive series of tests in irregular 

 waves and the statistical analyses connected with it. 

 The data on the model's behavior in irregular seas are 

 summarized in Table 7 and in Fig. 36. Comparing this 

 latter with Fig. 34, it is observed that the variation of 

 bending moments with speed is reduced. For the speed 

 which is practically attainable under given wave condi- 

 tions, probably not over 15 knots, the bending moments 

 can be considered approximately as independent of 

 speed. 



20 19 IB n 16 15 14 



13 12 II 10 9 

 Ordina+e 



7 6 5 



Fig. 28 Calculated and experimentally measured model de- 

 flections for three positions of a 100-kg loading weight (from 

 Sato, 1951) 



A comparison was made between the average bending 

 moments which have been directly measured in irregular 

 seas and the moments which had been computed (jn the 

 basis of wave spectra and ship response operators. The 

 degree of agreement shown is given in Table 8. 



At the ship's speed of 12.2 knots agreement within 

 12 per cent can be considered to be good. Increase of 

 the error to 22 per cent at zero speed corresponds to the 

 difference of the ship's-speed effect in regular and ir- 

 regular seas which already has been commented upon. 

 This error casts some doubt on the accuracy of the zero- 

 speed data. It is generally known that towing-tank 

 data are often affected at zero and low speeds by the 

 waves reflected from tank sides. Examination of Fig. 

 32 shows in fact a pronounced local dip of several curves 

 at zero speed. It appears that the authors placed too 

 much reliance on the individual zero-speed measure- 

 ments and could have achieved better results by using 

 instead a smooth extrapolation of several low-speed 

 measurements to zero speed. 



