PSI - Issue 13

Marijana Milković et al. / Procedia Structural Integrity 13 (2018) 1861 – 1866 Marijana Milkovi ć / Structural Integrity Procedia 00 (2018) 000–000

1863

3

Dimensions of the specimen laser-cut from the same sheet plane of Al7075-T6 are given on figure 1 b). For the experiment were used 3 specimens, of which 2 were electropolished at “ESI” (Erich Schmid Institute of Material Science) in Leoben (Austria) in order to get real fatigue properties of the material which are similar or almost equal thorough the whole section of the specimen. On the surface of the specimens were arbitrarily marked 9 points (example on Fig. 4.a)) for measurement of residual stresses and roughness. Residual stresses were measured with Pulstec µ X360 in longitudinal direction. Roughness was measured with TESA Rugosurf 10G in transverse and longitudinal direction. Measured values of residual stresses and roughness on untreated and untested specimen “1” are used as reference value in comparison with electropolished and fatigue tested specimens “2” and “3”. Uniaxial tensile fatigue test was conducted on a servo-hydraulic testing machine. In the tables 1 and 2 are given data of test conditions (N is number of cycles N , N t is total number of cycles, F max is values of maximum force and  max is maximum stress the specimen, and R is the load ratio and f is frequency of cyclic loading). Figure 4.b) shows the dimensions of 6 lines cut out on the electropolished part of specimen “3” after 1 000 cycles. On the specimen “2” was cut out the same shape after 10 997 cycles with the same dimensions. Lines were graved with the “FIB” (Focused ion beam) technique to see if there will be any visible change on the surface due to fatigue. Changes occurred on lines graved with FIB due to fatigue were monitored with SEM (Scanning electron microscope). 3. Results and discussion Statistical analysis of longitudinal residual stress in polished unloaded specimen revealed compressive residual stresses of average magnitude -9 MPa with standard deviation 32.08 MPa. However, after 319 990 load cycles statistical analysis of longitudinal residual stress revealed compressive residual stresses of average magnitude -70 MPa with standard deviation 19.55 MPa. The results of measured residual stresses before and after fatigue on surface polished with diamond paste are presented on figures 3 a) and 3 b). Figure 3.a) shows the positions on the surface of the specimen of measured residual stresses are marked as “x” and “y”. Loading was conducted in the direction of “y” axis (longitudinal direction). Residual stresses were measured in longitudinal direction. Figure 3.b) shows statistical frequency distribution of measured values of residual stresses. Residual stress initially remained in the material because of manufacturing. In table 1 and 2 are presented, besides data of test conditions, results of average values of roughness Ra (in both longitudinal and transversal direction) and residual stresses  x (with deviation V ) for specimens “2” and “3”. In the first row of the tables are given measured values of specimen “1”. In the second row are given measured values of specimens “2” or “3” after electropolishing.

0 0,005 0,01 0,015 0,02 0,025

Frequency

‐150

‐100

‐50

0

50

100

Residual stress, MPa

Before fatigue

After 319 990 cycles

a b Fig. 2 a) Measured residual stresses on polished surface before (cyan) and after (red) fatigue test, b) Gaussian distribution of measured values of residual stresses before and after fatigue test (319 990 cycles, max. stress 105 MPa, load ratio R=0,1)

Made with FlippingBook. PDF to flipbook with ease