PSI - Issue 19

Kiyotaka Masaki / Procedia Structural Integrity 19 (2019) 168–174 Kiyotaka MASAKI/ Structural Integrity Procedia 00 (2019) 000–000

172

5

Nagata 2009, Shiozawa et al. 2009, Shiozawa et al. 2013). Other than that, Morita et al. reported that the fatigue limits of AZ31 and AZ61 are equal to each compressive 0.2 % proof stress (Morita et al. 2009). Therefore, the correlation between fatigue limit of rotating bending load and static strength was investigated by examining more than 200 past literature researches on the fatigue property of AZ series Mg alloys (For example, Hirukawa and Furuya 2010, Kato and Tozawa 1981, Zhang and Lindemann 2005, Asahina et al. 1994, Nan et al. 2004, Kitahara et al. 2006 etc. ). Fig. 4(a) shows the correlation with the tensile static strength obtained by the investigation and Fig. 4(b) shows the correlation with the compressive ones. The definition of fatigue limit was defined in each paper. The following relations were obtained as a result of linear approximation of each data.

. 0 2 0 54 

w  

(1)

Tensile:

. T

. 

0 38 



(2)

w

BT

. 0 2 0 87 

w  

(3)

Compressive:

. C

BC (4) Especially, a direct proportional relationship between compressive 0.2 % proof stress and fatigue limit is shown in Fig.4 (b). It can be seen that the compressive 0.2 % proof stress is almost directly proportional to the fatigue limit except for data where compressive 0.2 % stress exceeds 200 MPa. These data exceed 200 MPa were the data for T5 heat treated AZ80 alloys. w .   0 37  In this study, fatigue properties of rotating bending loading had been obtained as shown in Fig.3. From Fig.3, three fatigue strength values which were the first horizontal stress in the two-step bending S-N curve (  wh ), the fatigue strength at 10 8 cycles (  w10 8 ) and the minimum stress at which test specimen fatigued (  wL ) were decided. And then, the fatigue limits of each materials were estimated from tensile/compressive static strength property shown in Table 2 with prediction relation indicated in section 4.1. Table 3 summarizes each experimental fatigue limits and estimated fatigue limits and the ratio of them. The best correlation of the AZ31 material was the relationships between the minimum stress at which test specimen fatigued (  wL ) and estimated fatigue limit using 4.2. Evaluation of experimental results

Table 3. Correlation of estimation value and experimental value of the fatigue strength. (a) AZ31

Fatigue strength at 10 8 cycles

Static strength [MPa] 

Estimated fatigue limit  w(est) [MPa]

First horizontal stress in the two-step bending S-N curve

Minimum stress at which test specimen fatigued

8  90 [MPa]

120 [MPa]

90 [MPa]

 wh 

 w10

 

wL 

 w(est) /  w10 8

 w(est) /  wh

w(est) /  wL

196 0.54 x  0.2T

106

0.88 0.65 0.75 0.73

1.18 0.87 1.00 0.98

1.18 0.87 1.00 0.98

 0.2T  0.2C

90

0.87 x  0.2C 1.0 x  0.2C

78 90 88

231 0.38 x  BT

 BT

(b) AZ61

Fatigue strength at 10 8 cycles

Static strength [MPa] 

Estimated fatigue limit  w(est) [MPa]

First horizontal stress in the two-step bending S-N curve

Minimum stress at which test specimen fatigued

8  150 [MPa]

190 [MPa]

130 [MPa]

 wh 

 w10

 

wL 

 w(est) /  w10 8

 w(est) /  wh

w(est) /  wL

221 0.54 x  0.2T 134 0.87 x  0.2C

119 117 134 117

0.63 0.61 0.71 0.62

080 0.78 0.89 0.78

0.92 0.80 1.03 0.90

 0.2T  0.2C

1.0 x  0.2C

309 0.38 x  BT

 BT