PSI - Issue 23

Tomáš Vojtek et al. / Procedia Structural Integrity 23 (2019) 481–486 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

483

3

This dependence results in a hyperbola-shaped function shifted by a constant value f PICC , which becomes the asymptote.

3. Results

Crack growth experiments were done using the compact tension (CT) specimens with the thickness of 10 mm. The material was ferritic-pearlitic steel (0.46% C, 1.20% Mn, 0.92% Si, 0.014% P, 0.025% S, 0.10% Cr, 0.11% Ni, 0.03% Mo, 0.25% V, 0.14% Cu) with the yield strength of 700 MPa and the ultimate strength of 1000 MPa, which is a perspective material for automotive industry. The fatigue crack growth rates d a /d N were experimentally determined as a function of the SIF range Δ K for the stress ratio R = 0.1. The crack increments (lengths) were measured optically using CCD cameras with digital micro-meter measurement of their position. The effective crack growth rate curve d a /d N = f ( Δ K eff ) was measured using the load-shedding procedure at the stress ratio R = 0.8 and fitted by the Klesnil- Lukáš formula ( 6) (Klesnil (1992)):   m eff, th m eff d d A K K N a    , (6) where A = 1.8·10 – 8 for d a /d N in [mm/cycle], m = 3 and Δ K 1/2 are the fitting constants. For the purpose of calculation of K cl (Eq.(2 )) the values of Δ K eff were expressed from Eq. (6): eff,th = 2.7 MPa· m

1

N a

A d 1 d

  

  

m

m eff, th

K   eff

K



(7)

1

N a

d 1 d

  

  

m

m eff, th

K K cl   

K





  1

(8)

R A

1

  K A R d 1 1 1 d    N a

  

m

m eff, th

max cl     

K

K R K cl



.

(9)

Since the experimental points are measured at different load levels, it is necessary to express the effective crack growth rate curve by continuous function. This enables calculation of K cl for each point. Fig. 1(a) shows the experimental points and the fits according to Eq. (5). Variation of R cl near threshold is caused by various levels of OICC occurring at various air humidity levels. Therefore, two fits are shown, one for the lowest humidity and OICC level and the second one for a high level of OICC. The lowest points show how large the maximum of RICC can be. The constants were f PICC = 0.25, K RICC = 2.5 MPa· m 1/2 , K OICC = 0 for low air humidity and K OICC = 2 MPa· m 1/2 for high air humidity. The d a /d N vs. Δ K curves were reconstructed for the applied stress ratio R < R cl according to Eq. (9) or the combination of Eqs. (2) and (6):

   

   

m

    K K cl

  

N a

d

m eff, th

  1 R A K

  



.

(10)

d

They are shown in Fig. 1(b), where the original experimental data can be seen.