PSI - Issue 2_A

J.A. Pascoe et al. / Procedia Structural Integrity 2 (2016) 080–087

83

4

J.A. Pascoe et al. / Structural Integrity Procedia 00 (2016) 000–000

P max

U= α N β dU/dN = αβ N ( β -1)

U

U tot = U cyc + U mono

P

P min

U cyc

U mono

d min

d max

d 0

N

d

Fig. 1. Definition of the various terms related to the strain energy and a schematic example of how d U / d N was determined. From the measured load and displacement, U was determined. A curve fit was then used to find a function relating U to N . d U / d N follows from the derivative of this function.

10 −1

10 −1

R = 0.036 R = 0.29 R = 0.61 R = 0.86

10 −2

10 −2

10 −3

10 −3

10 −4

10 −4

da/dN (mm/cycle)

10 −5

10 −5

10 −6

10 −6

0.01

0.1

1

0.2

0.4

0.6 0.8 1.0

1.6

∆√ G 2 =( √ G

2 (N/mm)

G max (N/mm)

max − √ G min )

Fig. 2. d a / d N as a function of G max (left) and ∆ √ G 2

(right).

∆ √ G

2 (or ∆ G ), and thus a decrease in d a / d N would also be expected, and is indeed seen here. Likewise, for a given ∆ √ G 2 an increase in R -ratio implies an increase in G max (or mean load), and therefore the increase in d a / d N that is seen should be expected. Figure 3 shows the crack growth rate as a function of the energy dissipation per cycle (d U / d N ). It is clear that there is a very strong correlation, and with the exception of one outlier, the curves for the di ff erent experiments appear to collapse onto one line. In fact there is still a small R -ratio e ff ect present, which will be discussed below. First however,