463 



300 



200 



100 

 70 



40 

 20 



10 

 20 40 70 100 



2 



Hv (kg/mm ) 



(a) Vickers Hardness 



300 

 200 



E 100 



-. 70 



en 

 •\ 



20 



300 



200 



100 

 1^ 70 



z 



5 40 



ai 

 ca 



S 20 



10 



o o 



o 



1 2 



7 10 



Sj,( kg/mm ) 



(b) Engineering Strain Energy 



° o 



o 



0,007 0.01 0.02 O.Oil 0.07 0.1 0. 



(cl ultimate Resilience 



Cavitation Number 



FIGURE 19. Illustration of MDDR peak characteristic 

 (test data given in Figure 11) . 



E^ « n {pa,-Pv) <5 BV 



napv3 (i^) l2 



(4) 



where 6 



displacement thickness of cavitation 

 bubbles , 

 foil span , 



cavity thickness at the cavity end, 

 velocity, 

 reference length. 

 Assuming that a cavity bubble grows according to 

 Knapp's similarity law, the volume, V, is: 



B 



<5e 



V 



L 



V a r3 o: ( T v/— ) 3 



where T = — , where A is the cavity length. 



V 

 The pressure difference, Ap, is assumed as 



(5) 



FIGURE 18. MDDR vs. various mechanical 

 properties of material (NACA 0015, C = 30 mm, 

 a = 4 deg. , V = 45 m/s) . 



Modelling of Cavitation Erosion and Scaling Factors 



As mentioned above, one of the authors developed 

 a model of the cavitation erosion mechanism. How- 

 ever it is limited to only constant cavitation 

 numbers and the effects of material properties were 

 derived empirically from accelerated tests. In the 

 present paper, this model is developed further to 

 treat differences in the cavitation number. The 

 effect of the material's mechanical properties is 

 also studied and a simple model is introduced. 



The total energy of collapsing bubbles per unit 

 is given as : 



n(p-pv)Q 



(3) 



where r\ : probability of bubble collapses on a 

 foil surface, 

 p-p.^^ : pressure difference at the collapse 



point , 

 Q : volumetric flow rate of cavitation 

 bubbles. 

 Equation (3) can be modified: 



4. 



0.3 0.4 



Cavitation Number 



0,5 



FIGURE 20. Impression of effect of roughened leading 

 edge [NACA 16021, tested by Ozaki and Kiuchi (1975)]. 



