Cyclic deformation and fatigue data for Ti–6Al–4V ELI under variable amplitude loading

This article presents the strain-based experimental data for Ti–6Al–4V ELI under non-constant amplitude cyclic loading. Uniaxial strain-controlled fatigue experiments were conducted under three different loading conditions, including two-level block loading (i.e. high-low and low-high), periodic overload, and variable amplitude loading. Tests were performed under fully-reversed, and mean strain/stress conditions. For each test conducted, two sets of data were collected; the cyclic stress–strain response (i.e. hysteresis loops) in log10 increments, and the peak and valley values of stress and strain for each cycle. Residual fatigue lives are reported for tests with two-level block loading, while for periodic overload and variable amplitude experiments, fatigue lives are reported in terms of number of blocks to failure.


a b s t r a c t
This article presents the strain-based experimental data for Ti-6Al-4V ELI under non-constant amplitude cyclic loading. Uniaxial strain-controlled fatigue experiments were conducted under three different loading conditions, including two-level block loading (i.e. high-low and low-high), periodic overload, and variable amplitude loading. Tests were performed under fully-reversed, and mean strain/stress conditions. For each test conducted, two sets of data were collected; the cyclic stress-strain response (i.e. hysteresis loops) in log 10 increments, and the peak and valley values of stress and strain for each cycle. Residual fatigue lives are reported for tests with two-level block loading, while for periodic overload and variable amplitude experiments, fatigue lives are reported in terms of number of blocks to failure.
& 2017 Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). The material used was mill-annealed wrought Ti-6Al-4V ELI bar, manufactured in compliance with ASTM standard F136-13 [1]. Cylindrical fatigue specimens with uniform gage section were designed following ASTM standard E606/E606M-12 [2]. The specimens were polished to achieve 0.5 μm surface finish in the gage section. Three coats of acrylic M-coat D were applied on the gage section to protect the specimens surface from the extensometer blades during testing.

Value of the data
Fatigue damage in most applications is commonly caused by variable and complex loadings. In some applications such as aerospace and biomedical, where Ti-6Al-4V ELI has been widely used as a structural material, understanding the fatigue behavior of the material is extremely important since majority of failures in structural components are attributed to fatigue damage.
The presented data offers a representation of Ti-6Al-4V ELI mechanical behavior in a controlled environment, thus contributing to the fundamental knowledge about this structural material. The data is also valuable as a baseline for other special applications (i.e. additive manufactured medical implants and aerospace components), or to compare with newly developed/improved materials.
The data presented in this article can be used for fatigue behavior and cyclic deformation related research on Ti-6Al-4V ELI under more realistic loading conditions. This data can be used to develop/calibrate constitutive models, cumulative fatigue damage models, and cycle counting methods.

Data
Strain-controlled block loading (i.e. high-low (H-L), low-high (L-H)), periodic overloading (PO), and variable amplitude (VA) fatigue data of Ti-6Al-4V ELI (a titanium alloy) is presented in this article. H-L, L-H, and PO experiments were conducted using fully-reversed (R ε ¼ À 1), and pulsating (R ε ¼ 0) strain loadings with various strain amplitudes, ε a . The strain ratio, R ε , is defined as R ε ¼ ε min / ε max and strain amplitude, ε a , is defined as ε a ¼ (ε max À ε min )/2, where ε min is the minimum strain and ε max is the maximum strain. The VA tests utilized a variable amplitude loading spectrum (i.e. load history) of various strain amplitudes, ε a , and strain ratios, R ε . For each test condition, two types of data were recorded. These include the cyclic (i.e. hysteresis loops) stress-strain responses recorded in log 10 increments, and the maximum (peak) and minimum (valley) values of stress and strain for each cycle. All relevant data has been made available in the Data in Brief (DiB) Dataverse: https://dataverse.harvard.edu/dataset.xhtml?persistentId ¼ doi:10.7910/DVN/SUCU5X.

Experimental design, materials and methods
Round fatigue specimens with uniform gage section and the dimensions and geometry, shown in Fig. 1, were designed according to ASTM standard E606/E606M-12 [2]. The specimens were fabricated from 12.7 mm diameter round bars of Ti-6Al-4V ELI, received in a mill-annealed condition (annealed for 1 h at 1300°F). All fatigue experiments were conducted under strain-controlled condition at room temperature, utilizing an MTS extensometer model 634.31F-25, a servohydraulic test frame with 100 kN load cell capacity, and a sinusoidal waveform input. Tests were performed using strain amplitudes that ranged from 0.0015 to 0.012 mm/mm. Influence of strain rate effects was minimized by adjusting the test frequency, ranging from 0.5 to 5 Hz, for each test condition to maintain a relatively consistent average strain rate for all tests. In addition, for each prescribed test condition, duplicate tests were conducted to validate the collected data and ensure the repeatability of experiments. Further details on the experimental program are presented in the following subsections according to the type of loading utilized.

High-low (H-L) and low-high (L-H) block loading
Constant strain amplitude (CA) fatigue tests were conducted using high-low (H-L) or low-high (L-H) block loading with various combinations of strain amplitudes, ε a , under fully-reversed (R ε ¼ À 1) and pulsating-tension (R ε ¼ 0) conditions. Fig. 2(a) and (b) displays schematics of the twolevel block loading using R ε ¼ À 1, where n 1 represents the number of cycles for the first loading block and n 2 denotes the number of cycles for the second loading block until the failure of specimens. Fig. 2(c), and (d) illustrates a schematic of the loading using R ε ¼ 0. Table 1 provides a summary of the straincontrolled H-L and L-H fatigue tests. The table includes the specimen ID, loading sequence (H-L or L-H), frequency of the first and second loading blocks, f 1 and f 2 , strain amplitude for the first and second loading blocks, ε a1 and ε a2 , and the number of cycles for first and second loading blocks, n 1 and n 2 .

Periodic overloading (PO)
Periodic overloading (PO) fatigue tests were conducted using a loading block with a predetermined number of cycles under selected constant strain amplitude, followed by 1 cycle of overloading. The PO experiments were performed using various combinations of strain amplitudes, ε a , under R ε ¼ À 1 and 0 conditions. Fig. 2(e) illustrates the schematic of the first loading combination, where the CA with R ε ¼ À 1 was applied for 100 cycles, followed by an overload strain amplitude, OL, with R ε ¼ À 1 for 1 cycle. The loading block was repeatedly applied until the specimen reached failure. Fig. 2(f) shows the schematic of the second loading combination, where the constant strain amplitude, CA, with R ε ¼ À 1 was applied for 100 cycles, followed by an overload, OL, with R ε ¼ 0 for 1 cycle. The collected data from strain-controlled PO fatigue tests are tabulated in Table 2. The experimental details presented in Table 2 include the specimen ID, the overload and constant strain amplitude strain ratios, R ε,OL and R ε,CA , the overload and constant strain amplitude frequency, f OL and f CA , the overload and constant strain amplitudes, ε a,OL and ε a,CA , and the number of blocks to failure, n B .

Variable amplitude (VA) loading
Variable amplitude (VA) fatigue tests were conducted using a random strain loading spectrum. Fig. 3(a) illustrates the loading spectrum A, which consists of multiple cycles within the maximum and minimum strains of 70.012 mm/mm. The loading block was repeatedly applied until the specimen reached failure. Fig. 3(b) shows spectrum B, which follows the same loading path as spectrum A, but within the maximum and minimum strain levels of 70.009 mm/mm (75% reduction as   (2) H-L 0. compared to spectrum A). Table 3 lists the specimen ID, loading spectrum (A or B), and the number of blocks to failure, n B for the strain-controlled VA fatigue tests.

Disclaimer
The corresponding author of this article is on the editorial board of Data in Brief. The editorial and peer review process for this article was not handled by Nima Shamsaei. Furthermore, the authors of this article do not have access to any confidential information related to its peerreview process.