Large-crack Data Research Articles

An investigation of small-crack effects in various aircraft engine rotor materials

Multiple sets of growth rate data for small fatigue cracks in seven different aircraft gas turbine engine rotor materials, including both titanium alloys and nickel-based superalloys, have been collected and are critically compared against corresponding large-crack data. The database includes a wide range of microstructures as well as multiple stress ratios and temperatures. The ability of the simple El Haddad small-crack model and its length parameter a0 to correlate the small-crack and large-crack data by adjusting the small-crack driving force is critically evaluated. Different methods of estimating a0, including the traditional calculation from large-crack threshold and endurance limit properties, as well as a new approach based on empirical scaling from microstructural dimensions, are explored and compared. Strengths and limitations of the simple El Haddad approach for engineering applications are discussed, as well as the practical significance of small-crack effects for life prediction.

Fatigue and crack-growth analyses of riveted lap-joints in a retired aircraft

For aluminum alloys, research has shown that fatigue behavior of notched and unnotched specimens can be characterized by fatigue–crack growth from micro-structural features, such as inclusion-particle sizes. A crack-growth model using small- and large-crack data was used to calculate fatigue lives and crack growth in lap-joints in laboratory specimens. The tightness of the rivets dictated the type of crack configurations that occurred in the joint and equivalent initial flaw sizes (EIFS) were selected to fit the fatigue test data. These crack configurations and EIFS values were then used to predict fatigue lives in fuselage lap-joints in a retired passenger aircraft and for curved panels cut from the retired aircraft. The paper demonstrates that fuselage lap-joint fatigue-life-prediction methods based on fatigue–crack growth alone (i.e. with accounting for any time spent in crack nucleation) are very adequate to model the lifetimes in fuselage riveted lap joints.

On the fatigue behavior of γ-based titanium aluminides: role of small cracks

Gamma-TiAl based alloys have recently received attention for potential elevated temperature applications in gas-turbine engines. However, although expected critical crack sizes for some targeted applications (e.g. gas-turbine engine blades) may be less than ∼500 μm, most fatigue-crack growth studies to date have focused on the behavior of large (on the order of a few millimeters) through-thickness cracks. Since successful implementation of damage-tolerant life-prediction methodologies requires that the fatigue properties be understood for crack sizes representative of those seen in service conditions, the present work is focused on characterizing the initiation and growth behavior of small ( a∼25–300 μm) fatigue cracks in a γ-TiAl based alloy, of composition Ti–47Al–2Nb–2Cr–0.2B (at.%), with both duplex (average grain size of ∼17 μm) and refined lamellar (average colony size of ∼145 μm) microstructures. Results are compared to the behavior of large ( a>5 mm), through-thickness cracks from a previous study. Superior crack initiation resistance is observed in the duplex microstructure, with no cracks nucleating after up to 500 000 cycles at maximum stress levels ( R=0.1) in excess of the monotonic yield stress, σ y. Comparatively, in the lamellar microstructure cracks nucleated readily at applied maximum stresses below the yield stress (85% σ y) after as few as 500 cycles. In terms of crack growth, measurements for small fatigue cracks in the duplex and lamellar microstructures showed that both microstructures have comparable intrinsic fatigue-crack growth resistance in the presence of small flaws. This observation contrasts previous comparisons of large-crack data, where the lamellar structure showed far superior fatigue-crack growth resistance than the duplex structure. Such “small-crack effects” are examined both in terms of similitude (i.e. crack tip shielding) and continuum (i.e. biased microstructural sampling) limitations of traditional linear elastic fracture mechanics.

SMALL CRACK GROWTH AND FATIGUE LIFE PREDICTIONS FOR HIGH‐STRENGTH ALUMINIUM ALLOYS: PART I—EXPERIMENTAL AND FRACTURE MECHANICS ANALYSIS

The small crack effect was investigated in two high‐strength aluminium alloys: 7075‐T6 bare and LC9cs clad alloy. Both experimental and analytical investigations were conducted to study crack initiation and growth of small cracks. In the experimental program, fatigue tests, small crack and large crack tests were conducted under constant amplitude and Mini‐TWIST spectrum loading conditions. A pronounced small crack effect was observed in both materials, especially for the negative stress ratios. For all loading conditions, most of the fatigue life of the SENT specimens was shown to be crack propagation from initial material defects or from the cladding layer. In the analysis program, three‐dimensional finite element and weight function methods were used to determine stress intensity factors and to develop SIF equations for surface and corner cracks at the notch in the SENT specimens. A plasticity‐induced crack‐closure model was used to correlate small and large crack data, and to make fatigue life predictions. Predicted crack‐growth rates and fatigue lives agreed well with experiments. A total fatigue life prediction method for the aluminium alloys was developed and demonstrated using the crack‐closure model.

The effect of grain size on small fatigue crack growth in pure titanium

The growth behaviour of small fatigue cracks has been studied in both fine- and coarse-grained versions of a pure titanium under axial loading at stress ratio, R, of −1. The growth behaviour and its statistical properties in a coarse-grained version of a different pure titanium have also been investigated under rotating bending ( R = −1), and the results obtained were compared with those of a fine-grained version of this titanium in a previous report. Under both loading conditions, small cracks grew faster than large cracks. As the growth data were plotted in terms of the effective stress intensity factor range Δ K eff (after allowing for crack closure, the growth rates could be well correlated with large-crack data in a large-crack regime. In a small-crack regime, however, small cracks still grew faster than large cracks. Small cracks in coarse-grained material showed higher growth rates than those in fine-grained material owing to a much smaller effect of microstructure such as grain boundaries and crack deflection. Stage I facets were observed in all the specimens tested, and their depths were less than the maximum grain size estimated by the statistics of the extreme values, but the distribution of stage I facet depths approximately corresponded to the maximum value distributions of grain size of the materials. The growth rates of small cracks followed log-normal distributions independent of grain size. The coefficients of variation, η, of growth rate in coarse-grained material were smaller than those in fine-grained material. The η values were significantly large at a/ d ⩽ 3 ( a = crack depth, d = grain size), indicating that the relative size of microstructurally small cracks was not dependent on grain size.

Microstructure and crack-shape effects on the growth behavior of small fatigue cracks in Ti24Al11Nb

The effects of four different microstructures in the titanium aluminide alloy Ti24Al11Nb (at.%) on the fatigue crack growth behavior of small surface cracks and large cracks have been investigated. The four microstructures were a Widmanstätten basketweave, a Widmanstätten aligned colony, an equiaxed primary α 2 in a Widmanstätten matrix and a completely equiaxed α 2 structure. Small cracks were found to develop arbitrary shapes owing to the effects of microstructure. The crack shapes (aspect ratios) were measured using a laser interferometric and photomicroscopic system, and these measurements allowed accurate calculation of crack growth rates at the surface position as well as at the depth position for the surface cracks. After accounting for the continuous variation in crack shape in crack growth rate calculations, the trends in small-crack growth rates agreed reasonably well with the corresponding large-crack growth rates. While the crack growth rates at depth positions for small cracks correlated well with large-crack data in the basketweave microstructure, crack growth rates at surface positions correlated well with the corresponding large-crack data in the other microstructures. The microstructural factors that may be responsible for this behavior are discussed.

The Effect of Grain Size on Small Fatigue Crack Growth in Pure Titanium.

Abstract The growth behaviour of small fatigue cracks has been studied in both fine- and coarse-grained versions of a pure titanium under axial loading at stress ratio, R, of −1. The growth behaviour and its statistical properties in a coarse-grained version of a different pure titanium have also been investigated under rotating bending (R = −1), and the results obtained were compared with those of a fine-grained version of this titanium in a previous report. Under both loading conditions, small cracks grew faster than large cracks. As the growth data were plotted in terms of the effective stress intensity factor range ΔKeff (after allowing for crack closure, the growth rates could be well correlated with large-crack data in a large-crack regime. In a small-crack regime, however, small cracks still grew faster than large cracks. Small cracks in coarse-grained material showed higher growth rates than those in fine-grained material owing to a much smaller effect of microstructure such as grain boundaries and crack deflection. Stage I facets were observed in all the specimens tested, and their depths were less than the maximum grain size estimated by the statistics of the extreme values, but the distribution of stage I facet depths approximately corresponded to the maximum value distributions of grain size of the materials. The growth rates of small cracks followed log-normal distributions independent of grain size. The coefficients of variation, η, of growth rate in coarse-grained material were smaller than those in fine-grained material. The η values were significantly large at a/d ⩽ 3 (a = crack depth, d = grain size), indicating that the relative size of microstructurally small cracks was not dependent on grain size.

Impact of small-crack effects on designlife calculations

An assessment has been made of the potential reduction in calculated fatigue-crack-growth lives when the small-crack effect is considered. The assessment was based on small-crack and large-crack grwoth-rate data of 2024-T3 aluminum alloy forR=−1 andR=0 constant-amplitude loading and for a fighter-wing spectrum loading (FALSTAFF). The potential impact of the small-crack effect was assessed by comparing lives computed using only large-crack data to those computed using a combined small- and, large-crack data based. Based on life calculations, the small-crack effect would have: (1) no impact on life analyses that assume an initial crack depth large than about 0.3 mm (such as in current airframe damage-tolerance analyses); (2) only a small impact on life analyses that assume an inital depth of 0.1 mm (such as im some current durability analyses) if the large-crack thresholds are ignored; but (3) a large impact on life analyses that assume an initial depth of about 0.01 mm (such as in a total-life analysis).

Relevance of the small crack problem to lifetime prediction in gas turbines

The anomalous behaviour of ‘small’ fatigue cracks is summarized, and the current bases for understanding this phenomenon are discussed. The limited literature pertaining to small cracks in nickel-base superalloys is reviewed, and the consequences of this behaviour upon the lifetime prediction of gas turbines are evaluated for a typical case. Evaluation of lifetimes using large crack data is found to be non-conservative, especially at lower stress (longer lifetimes). It is shown that improvement in the resolution of non-destructive inspection techniques leads to even greater discrepancy between lifetime estimates based on small crack or large crack growth rate data.