Breaking down new features of upcoming AFGROW release 5.2

The large amount of crack growth rate data available to AFGROW Users is extremely valuable, but it must be curve fit before it is useable for life prediction purposes. The curve fitting process can be too complicated and time consuming for it to be a viable option for many users. Therefore, LexTech has started work to fit data for several commonly used materials. This presentation will show the results of curve fits completed to date, how the fits will be added to the database, and the proposed plan to continue fitting data in the future.

This presentation summarizes the findings of a multi-year study on fatigue crack growth from holes. Tests were performed in simple coupons made from a single alloy (7075 T6). The coupons had single holes, and cracks grew from a single-sided starter notch under either applied uniform tensile stress (open hole coupons, OH) or pin loading (loaded hole coupons, LH). Coupons were in one of two conditions, either as-machined (AM, hole drilled and reamed) or cold expanded (CX, hole drilled, 3% cold-expanded (FTI process) and reamed). Test stress levels were designed to have a crack growth life of 200,000 cycles in all coupon conditions. OH coupons were tested in two thicknesses, one set of coupons termed thin (cut from 0.08 inch thick sheet) and a second set termed thick (cut from 0.19 inch thick sheet). Thin coupons had one-dimensional (through) cracks and thick coupons had two-dimensional (corner) cracks. LH coupons were tested only in the thin, through-crack configuration. Companion models of crack growth behavior were developed to assess the ability of linear elastic fracture mechanics to predict the observed behavior. For CX coupons, the models incorporated residual stress fields measured in replicate coupons using the Contour method, which enabled predictions for the average residual stress field, and the effects of residual stress variability.

Accurate quantification of crack tip stress intensity values is paramount in the analysis of damage tolerant structures. The present analytical investigation seeks to determine the stress intensity solutions for crack geometries outside the existing valid solution space and expand the analyst’s ability to capture representative crack growth behavior. The primary focus of this investigation is to better quantify the crack growth behavior of single quarter-elliptical corner cracks from centrally located holes in finite width plates, under various loading conditions (remote tension, bending, and pin loading). Some current finite width corrections, such as the Newman-Raju corrections, have been calculated from straight through cracks and universally applied to all locations along the quarter-elliptical crack front. Early investigations into the validity of this application seem to indicate that this correction procedure produces stress intensity values up to 30% high for relatively narrow plates (W/D<6). Furthermore, the crack depth to length ratio and depth to thickness ratio can significantly influence the applicability of the current finite width corrections. The analytical investigation utilizes the three dimensional virtual crack closure technique (3D VCCT) and a well-structured completely hexahedral element mesh. Stress intensity values are generated for a range of crack depth to crack length ratios (a/c = 0.01 to 10), crack depth to sheet thickness ratios (a/t = 0.01 to 0.99), hole radius to sheet thickness ratios (r/t = 0.1 to 10), and sheet width to hole diameter ratios (W/D = 1.1 to 20). This effort is being executed under a DoD Technical Corrosion Collaboration program.

A synthesized lower wing skin was modeled using USAF's life prediction code, AFGROW, to demonstrate the effect of initial crack length and stress on fatigue life. Prediction were run over bands of stresses and crack lengths covering a span from 0-300 MPa (0-43 ksi) and 0-25 mm (0-1") at increments of 0.1 mm, 1 mm, and 10 mm increments and 0.1 MPa, 1 MPa and 10 MPa increments. Multiple load ratios and material systems were also run to explore the sensitivity space. From these results, a clear mapping of the dominance of the threshold region has been clearly demonstrated. Furthermore, a practical grasp on error measurement can be derived rapidly from the investigation.

This research tested the fatigue life benefit from cold expansion in holes with preexisting cracks approximately 0.05 inches long. Tests were conducted under constant amplitude loading and A-10 wing spectrum loading at multiple max stresses. Test data was compared to crack growth models generated in AFGROW using standard United States Air Force specifications. The life improvement factor ranged from 90 to 60 for constant amplitude loading and from 22 to 1.3 for spectrum loading. The predictions from AFGROW using a 0.005" initial flaw size assumption to account for cold expansion, yielded a nearly constant life improvement factor of two. Therefore, predictions were conservative by a factor of more than 10 for some stresses and non conservative by a factor of 1.5 at other stresses.

The experiments performed in this research investigated the fatigue crack growth lives of short edge margin fastener holes that contained a crack prior to cold expansion. Three configurations were used – a baseline condition consisting of non cold expanded holes, holes that were cold expanded, and holes containing a crack when cold expanded. All configurations were investigated under constant and variable amplitude loading. The hypothesis was that the cold expansion of a short edge margin hole with a crack prior to cold expansion will provide a significant increase in fatigue crack growth life compared to a short edge margin hole that was not cold expanded. The fatigue crack growth life of a cracked then cold expanded hole was also compared to a hole that was not cracked prior to cold expansion. The United States Air Force (USAF) analytical approach used to account for the benefit due to cold expansion was compared to the experiment data and does not consistently provide conservative predictions.

ESRD will be presenting lessons learned and best practices for StressCheck fracture mechanics applications. This presentation will include how best to efficiently construct models with cracks, define appropriate mesh parameters, and extract stress intensity factors for use in Afgrow predictions.

At the turn of the century the fatigue crack growth analysis approach and tools were limited to “canned” stress intensity solutions mostly focused on corner cracks propagating from the edge of a hole in a flat plate. This paper will outline the events and their associated processes, technology, and methodology outcomes that have brought the United States Air Force’s (USAF) A-10 Aircraft Structural Integrity Program (ASIP) out of the “canned” solution paradigm of crack growth analysis to the point in which they can more accurately model and predict crack growth behavior in high residual stress fields and in a load redistribution environment. The development of coupled finite element modeling and fatigue crack growth analysis will be outlined along with an outlook of the future potential of crack growth analysis in the 21st century.

One necessary aspect of a robust fatigue crack growth model is the ability to account for load history influences. The Generalized Willenborg model is implemented in the A-10 Damage Tolerance Analysis to address load history influences. The Generalized Willenborg retardation model uses a parameter called the Shutoff Overload Ratio (SOLR) that the analyst can use to empirically tune the model to the appropriate amount of retardation. Because the amount of retardation that occurs during fatigue loading is generally a function of the material and the loading spectrum, the SOLR values used in AFGROW are specific to the materials and spectra associated with the aircraft structure and usage. A study was performed to determine the sensitivity of SOLR changes in each material property value, stress intensity input technique, and technique for fitting crack growth curves during SOLR correlation.

Calculating accurate stress intensity factors (SIFs) for countersunk holes can be challenging. SwRI has recently developed a set of SIF solutions that can can be used to quickly obtain accurate SIFs for the most common aerospace countersunk hole geometries. The current solution set covers a crack growing from the base or knee of the countersink with remote tension loading. Crack dimensions range from very small to very large with a/c aspect ratios ranging from 0.5 to 4. This presentation will discuss the methods used to obtain and validate the SIF solutions as well as the interpolation process that can be used to quickly obtain SIF solution from anywhere within the solution space.

Traditionally, stress intensity solution development and crack growth predictions are developed independently. For standard geometries and loading, this typically works quite well. However, for complex geometry and/or loading, varying crack aspect ratios, multiple cracking scenarios, etc., this classic approach doesn’t always fit. Synergies between the factors that affect the overall crack shape and growth are not necessarily captured, and thus can have a significant influence on the crack growth life.

The T-38 and A-10 analysis groups have developed a generic AFGROW plug-in that couples stress intensity development via StressCheck with AFGROW’s crack growth analysis capability. This new capability allows AFGROW to open, update, solve, and extract solutions from parameterized StressCheck models automatically. Solutions are imported into AFGROW, crack growth is calculated, and the new crack geometry is sent back to StressCheck. This process is repeated automatically until a defined failure or stop criteria is reached. This seamless integration allows for more accurate crack growth predictions in complex situations and eliminates many of the assumptions that are required with the traditional approach.

This paper describes the development of the code, keys to building a proper StressCheck model, limitations of the plug-in, as well as future applications and directions of this capability. It also presents comparisons to Classic and Advanced AFGROW solutions. Finally, test data will be presented to help focus the experimental validation process.

Information on the latest research and development efforts and plans beyound AFGROW Release 5.2.