PSI - Issue 25

Daniel J. Cruz et al. / Procedia Structural Integrity 25 (2020) 316–323

319

4

Daniel J. Cruz et al. / Structural Integrity Procedia 00 (2019) 000–000

Table 1: DIC image post-processing parameters

VIC - 2D Analysis Option

Aperture setting

f / 4

Extension

96 mm 200 mm

Approximate working distance

Acquisition interval

2 fps

Subset Size

29 Pixel

Gauge length of virtual extensometer

12 mm (SS-2X12) / 2 mm (SS-2X2) Normalized Squared Di ff erences

Correlation criterion Interpolation method

Optimized 8-tap

Subset weights

Gaussian

Step size

5 Pixel

Low-pass filter images Incremental correlation Maximum pixel confidence

Yes Yes

0.05 Pixel 0.5 Pixel

Epipolar threshold

Table 2: Results of tensile tests for AZ31B sheet alloy

R p 0 . 2 [MPa]

R m [MPa]

A gt [%]

A t [%]

Macro specimen (ASTM E8-04)

219.2 210.4 209.8

267.1 268.1 270.0

16.5 11.4 12.1

21.4 16.6 17.2

SS - 2X12 SS - 2X2

The results for the two geometries are represented in Fig. 6 and in Tab. 2. To ensure the usefulness of miniature test techniques, it is important to have comparable results of miniature and conventional tests. Therefore, a reference curve relative to a tensile test of AZ31B steel using a macro specimen, according to the ASTM E8-04 at the same temperature is also represented on the results. It is possible to note that both curves have the same shape and can be easily compared with the macro reference: strength parameters are similar, namely yield strength ( R p 0 . 2 ) and ultimate strength ( R m ) with an error for both geometries of approximately 4% and 1%, respectively; however, elongation is more sensible to miniaturisation, particularly the elongation corresponding to ultimate strength ( A gt ) and the total elongation ( A t ), with an error of approximately 30% and 20%, respectively. This results can be considered satisfactory and provide a very good evaluation of mechanical properties for this material. The strain field obtained by Digital Image Correlation is represented in Fig 5. That distribution is visible along the uniform gauge length, which already normally occurred in macro-scale tests. All these results allow us to conclude that the Digital Image Correlation technique is perfectly adequate for this type of scale as described by Dzˇugan et al. (9; 10).

(a) SS - 2X12

(b) SS - 2X2

Fig. 5: Strain field distribution during the tensile test