PSI - Issue 13

Bruno Atzori et al. / Procedia Structural Integrity 13 (2018) 1961–1966 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

1964

4

Table 1. Fatigue characterisation of AISI 304L based on energy-life experimental data. Fatigue curves Material parameters Principal parameters*

Derived parameters**

Equations

a f =0.494

/ /

/ /

) f a

(

a f

fp W W'

2N = 

fp

f

W' fp

W' fp =130 MJ/m 3

a

/ /

a=-0.506

(7a) (7b)

) a

(

'

p p f W W 2N  =  

 W' p

 W' p =260 MJ/(m

3 cycle)

p p W  =   

 =0.666







n'

n'=0.336

/

/

n' 1



p 2 2K'     =

  

K'

K'=2328 MPa

/

/

b

/

b=-0.127

(6), (7a)



( ' f 2N

) b

=  

f

2

' f 

' f  = 1048 MPa

/

(6)

c

/

c=-0.379

(7a)



p

(

) c

' f

2N

=  

f

2

' f  =0.093

' f 

/

(6), (7a)

*: from best fitting of experimental results;**: from compatibility equations reported in the Table

0 100 200 300 400 500 600 700 0.000 0.005 0.010 0.015 0.020  /2 [m/m] Serie6 ep 2 [MPa] E=194700 MPa K'=2328 MPa n'=0.336 Experimental tensile static curve Cyclic curve  e /2 (b)

(a)

1.E-1

b= -0.127

E=194700 MPa

c= -0.379  ' f =1048 MPa  ' f =0.093

1.E-2

2 [ m/m ]

2 data

ep tot Serie3 AISI 304 L  /2  p /2  e /2

1.E-3

1.E-4

1.E+2 1.E+3 1.E+4 1.E+5 1.E+6 1.E+7

2N f , number of reversals to failure

Fig. 1. (a) experimental strain-controlled fatigue results and (b) stress-strain cyclic curve of AISI 304L plain material Fig. 1a shows the results of the strain controlled fatigue tests in terms of Manson-Coffin curves, and Fig. 1b shows the stabilised Cyclic Curve of the material, which was derived by applying the Ramberg-Osgood model to the same results at the half-life. From this curve (Fig. 1b), the results of Fig. 1a can be converted in terms of the Strain Energy Density under the cyclic stress-strain curve, as shown in Fig. 2a. The analysis of stabilised or half-life hysteresis loops showed that the Masing hypothesis was not satisfied. Therefore, the  W p values were derived by evaluating the area of the experimental hysteresis cycles and  was evaluated by fitting the  W p versus the  ·  p data at the half-life. It was found that  =0.666 instead of (1-n')/(1+n')=0.497, as evaluated according to Halford (1966). The fatigue curves of the different plastic strain energy densities are summarised in Fig. 2b compared to the experimental results. For the same life, the  W p /W SCp ratio is equal to 3.56 (290/73.1=3.56, from Fig. 2b). For each fatigue test, the Q parameter was evaluated according to Meneghetti (2007). Fig. 3a shows a comparison between this quantity and  W p measured during the same fatigue tests. It appears that, except for very low cycle tests, the values of the two energies are very similar, because of the very low value of the internal energy per cycle stored within the material (Meneghetti et al 2013).