AFGROW User Workshop 2023

Damage tolerant structures require accurate fatigue crack growth rate models for life prediction. Central to these models are fracture mechanics similitude and empirically gathered growth rates. In this work, standard test methods failed to achieve similitude for 17-4PH stainless steel round-rod but succeeded for plate. Material product form differences are interrogated through constant dK tests, crack path analysis, residual stress (RS) characterization, and growth rate simulation. Analyses revealed that quench-induced RS in the round-rod promoted a non-planar, 3D crack path with closure effects. Strategies to mitigate the RS/path effects are evaluated by controlling constraint, closure, environment, and heat treatment.

The critical crack-tip-opening angle or displacement (CTOA/CTOD) fracture criterion is one of the oldest fracture criteria applied to metallic materials. Improved computer-aided photographic methods have been developed to measure CTOA during the fracture process; and elastic-plastic, finite-element analyses (ZIP2D) with a constant CTOA and a plane-strain core have been used to simulate fracture of laboratory specimens. The fracture criterion has been able to link the fracture of laboratory specimens to structural applications. This paper analyzes fracture of cracked thin-sheet 2219 aluminum alloy over an extremely wide range in width, crack-length-to-width ratio, and applied loading (tension, bending, and combined tension-bending loads). The results from the critical CTOA fracture analyses on the thin-sheet material showed that the stress-intensity factor at failure (KIe) was linearly related to the net-section stress (Sn), as predicted by the Two-Parameter Fracture Criterion (TPFC).

The A-10 program office performed an urgent action TCTO to inspect fittings that attach the aft cowl of the nacelle to the main center section. This was done in response to an aft cowl departure in flight and subsequent preliminary crack findings. The TCTO revealed an extensive cracking issue with this interface and an analysis team was tasked to determine the root cause and subsequent inspection requirements for cracking.

The presentation shows how organic Air Force capabilities were leveraged to perform failure analysis, Finite Element Modeling (FEM), and multi-point crack growth analysis to validate results and determine a way forward for fleet management. The materials lab at Hill AFB was able to perform striation counting at multiple points along the fracture surface to correlate crack growth rates to analysis. Siemens Simcenter was used to build a Nastran FEM of the area near the attach fittings. The legacy stress analysis was used to determine appropriate loads. The forces on the attach fitting calculated with the Nastran model were then transferred over to the ESRD StressCheck to model the cracking geometry found in the failure analysis and correlate stress intensities to what would have been required to achieve the measured crack growth rates. The loads were scaled to create the appropriate stress profiles to match failure analysis. Then a multi-point fracture mechanics model was created to grow multiple cracks within the fittings as was seen in the data. The resulting crack growth curves and the detectable flaw sizes from inspection were used to reduce risk and manage the fleet.

In 1978, The Code of Federal Registry, Title 14 § 25.571 - Damage-tolerance and fatigue evaluation of structure, was published and mandated damage tolerance based inspections to ensure that the principal structural elements (PSE) of the airplanes remain safe throughout the operational life of the airplane. The threat associated with this risk mitigation was associated with the fatigue stresses that the structures experience during the operation. To satisfy the requirements of § 25.571, damage tolerance evaluations are carried out for every PSE and maintenance actions are developed and included in the airworthiness limitation section. For the damage tolerance evaluation, fracture mechanics is the common means to set the maintenance program. As part of fracture mechanics, crack growth analysis is performed to determine how many flight cycles or flight hours it takes for a crack to grow from a certain length to the critical length. The assumption of the location and modes of the lead crack and the associated adjacent cracks in the detail design points of the PSE are the key to a meaningful damage tolerance evaluation and suitability of the maintenance program to detect the crack before it reaches the critical length. In this presentation, acceptable cracking scenarios and patterns that are based on test and service history are presented to help the analyst establish suitable inspections to detect a crack in a timely manner and to meet the safety requirements prescribed by the aforementioned regulations.

For more than two decades CAStLE has been facilitating the creation of high-fidelity stress intensity factors for millions of crack configuration through the work of Börje Anderson. While this work had been supported by various USAF ASIP organizations in the past, since before 2010 CAStLE has obtained sufficient discretionary funds from a variety of DoD sources to continue this work independent of any direct project funding. Additionally, CAStLE has facilitated the in-person participation of Börje in the annual AFGROW Users meeting as a way of sharing the new solution space created over the previous year and to help steer the priorities of the remaining solution space in the years to come. As a way of introduction to this year’s presentation by Börje, this presentation will briefly summarize the accomplishments of the recent previous years.

Novel high-accuracy K_I (φ)-solutions developed during 2022/2023 are reviewed.

• For straight-shank hole geometries, the most important developments are new crack solutions for 19 different plate widths. A recent benchmark 2020 emphasized an urgent need for these solutions.
• For countersunk hole geometries, two different hole depths are considered namely b/t = 0.50 and b/t=0.05 (almost knife-edge countersunk holes) for 19-30 different plate widths.
• A difficulty with the K_I(φ)-solutions is that K_I(φ) goes to infinity near crack-vertices located at the conical surface, or upper plate surface, at almost half of the crack shapes. A special filtering technique was therefore developed which is based on the mathematical vertex theory, a technique that is used at each crack vertex when generating all K_I (φ)-databases.
• The solution accuracy is very high for all solutions, in all databases, with a relative error in K_I (φ) less than 0.05% at every point φ on the crack fronts. That includes all the vertex regions where K_I → ∞, or, K_I → 0, where the abovementioned filtering technique have been used.

The lecture describes analysis techniques, filtering techniques, schemes for control of solution accuracy and contents of new K_I (φ)-databases.

With the recent publication of a USAF Structures Bulletin on including residual stresses in analysis (EZ-SB-17-001 Revision A), the A-10 team has developed ground rules for utilizing residual stresses in analysis as well as ground rules for utilizing multi-point analysis methods. The presentation provides an overview of these ground rules as well as a brief summary of test validation for some applications.

Compressive residual stresses (RS) imparted to structures via cold-expansion (Cx) of fastener holes are critical for fatigue life extension of airframes. Recently, there has been progress in predicting the fatigue life of Cx holes with RS measured using the contour method and multi-point stress intensity simulations. However, the Cx process has several factors that cause significant variation in the RS field and associated lifing predictions. One critical factor is the Cx applied expansion (I_a), but manufacturing tolerances (e.g., hole size, mandrel size, etc.) can result in Ia variations up to ~25% from nominal. The current work examines fatigue life prediction variations due to Ia differences in contour-based Cx hole RS fields. Over 50 simulations were performed representing both in- and slightly out-of-tolerance Cx processes. Predictions for in-tolerance Cx holes (I_a = 3.0-4.5%) showed a wide life improvement factor (LIF) ranging from 2.5-31. With the large number of unique RS fields analyzed, results from this work may serve as a benchmark for Cx predictions utilizing contour-method based RS fields.

This presentation will give an overview of the upcoming Spectrum Manager v 1.3.

Information on the latest research and development efforts and plans beyond AFGROW Release 5.4