Corrosion Pillowing in Aircraft Fuselage Lap Joints Essay Example for Free

Corrosion Pillowing in Aircraft Fuselage Lap Joints Essay Nicholas C. Bellinger,âˆâ€" Jerzy P. Komorowski,†  and Ronald W. Gould†¡ National Research Council Canada, Institute for Aerospace Research, Ottawa, Ontario K1A 0R6, Canada DOI: 10.2514/1.18589 This paper presents the results of studies that have been carried out at the National Research Council Canada on the effect that corrosion pillowing has on the structural integrity of fuselage lap joints. Modeling of corrosion pillowing using ï ¬ nite element techniques showed that the stress near the rivet holes increased to the material (Al 2024T3) yield strength when the corrosion present was above 6% thickness loss. In addition, the analysis showed that pillowing resulted in a stress gradient through the skin thickness, which suggested that semi-elliptical cracks with high aspect ratios could form. During teardowns of service-exposed lap joints, these types of cracks were found at numerous holes and a closer examination of the fracture surfaces revealed the presence of fatigue striations. Therefore, a new source of multisite damage, other than fatigue, was identiï ¬ ed. I. Introduction N THE 1980s, it became apparent that commercial transports would remain in service well beyond their original design life, which raised concerns that corrosion combined with fatigue could lead to catastrophic failures of fuselage lap joints. Although there are multiple lap joint designs present in a single fuselage, for older aircraft the majority of them consist of an outer and inner skin fabricated from aluminum 2024-T3 joined together with multiple rows of countersunk rivets (Fig. 1), as well as an adhesive layer. During the operation of an aircraft, the adhesive layer can deteriorate and disbond allowing moisture to migrate between the skins. This moisture can, in turn, breakdown the material protective system resulting in the formation of crevice corrosion. As the corrosion forms, the skins between the rivets are forced apart due to the presence of the corrosion products resulting in a bulging or â€Å"pillowing† of the skins. This phenomenon is referred to as corrosion pillowing and is the feature used to detect corrosion in lap joints using visual (ï ¬â€šashlight) nondestructive inspection techniques. This paper reviews the results to-date of studies that were carried out to determine the effect that corrosion pillowing has on the structural integrity of fuselage lap joints. The results indicate that corrosion pillowing can cause the formation of semi-elliptical cracks with high aspect ratios, which were found in a number of serviceexposed corroded lap joints. These cracks formed at a number of different rivet holes and thus are considered to be a new source of multisite damage (other than fatigue), which could signiï ¬ cantly affect the residual life and strength of a lap joint. II. Corrosion Pillowing Analysis A. Mathematical Model Presented as Paper 2023 at the 46th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics Materials Conference, 13th AIAA/ ASME/AHS Adaptive Structures Conference, 7th AIAA Non-Deterministic Approaches Forum, 6th AIAA Gossamer Spacecraft Forum, and 1st AIAA Multidisciplinary Design Optimization Specialist Conference, Austin, Texas, 18–21 April 2005; received 5 July 2005; revision received 17 February 2006; accepted for publication 17 March 2006. Copyright  © 2007 by National Research Council of Canada. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00 per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923; include the code 0021-8669/07 $10.00 in correspondence with the CCC. âˆâ€" Structures Group Leader, Structures and Materials Performance Laboratory, 1200 Montreal Road, Building M14. †  Director General, Institute for Aerospace Research, 1200 Montreal Road, Building M3. †¡ Senior Technical Ofï ¬ cer, Structures and Materials Performance Laboratory, 1200 Montreal Road, Building M3. 758 The visual nondestructive inspections used to detect corrosion pillowing in fuselage lap joints are not capable of determining the level of corrosion that is present within a joint. Therefore, to determine if a correlation existed between the amplitude of the pillowing deformation of the outer skin of a lap joint to the degree of corrosion inside the joint, a mathematical model was developed. This model presumed that after the lap joint disbonds the aircraft skin between the rivets deforms perpendicularly to the lap joint surface to accommodate the additional volume required by the corrosion product. A chemical analysis on corrosion samples taken from service-exposed lap joints indicated that the insoluble product mainly consisted of aluminum oxide trihydrate (aluminum hydroxide), which has a molecular volume ratio of 6.454 times that of pure aluminum [1]. The model assumed that the corrosion product was distributed within the joint so as to exert a uniform lateral pressure on the fuselage skins. It was also assumed that the joint was symmetrical about its midplane and thus only the outer skin was modeled. The closed-form classical plate theory of Timoshenko and Krieger was used to calculate the deformation of the outer skin supported by equidistant rivets and subjected to a uniform lateral pressure [2]. This deformation was used to calculate the pillowing ratio given by the ratio of the central deï ¬â€šection to the volume under the deformed plate [3]. From this model, the central deï ¬â€šection was determined to be approximately 3.3 times the thickness loss for a rivet spacing ratio of one. To better understand the effect that the rivet spacing ratio had on corrosion pillowing, the deformed shapes of plates with various rivet spacing ratios were calculated and the results are plotted in Fig. 2. The results showed that as the rivet spacing increased, the relative deï ¬â€šections at the shorter edges decreased, whereas those at the longer edges increased, which can signiï ¬ cantly reduce the probability of detecting the corrosion visually. This suggests that the detection limit for joints with a high rivet spacing ratio, may be signiï ¬ cantly larger than the maximum allowed 10% thickness loss. B. Stress Analysis Because actual lap joints contain free edges and stiffeners, which cannot be modeled using the closed-form solution, ï ¬ nite element techniques were developed to model an actual aircraft lap joint (Fig. 1). This model included the effects of the hoop stress from the pressurization of the fuselage, the rivet prestress caused by the rivet installation as well as the corrosion pillowing stress. The fuselage curvature was ignored for all the analyses that were performed during this program.