Strategies for static failure analysis on aerospace structures

Javier S. Millán; Iñaki Armendáriz; Juan García-Martínez; Roberto González    * Materials and Structures Department, Instituto Nacional de Técnica Aeroespacial (INTA), Torrejón de Ardoz, Madrid Spain

Abstract

Primary objective of the failure analysis of a structure is the prediction of damage onset and the determination of its causes. As damage onset does not mean the final failure of the structure, the failure analysis also comprises the subsequent damage growth analysis that occurs when the structure is loaded; this is called progressive failure analysis (PFA). Performing accurate predictions of damage onset and PFA allows optimizing the structural design and improving its reliability. In this chapter, several methodologies developed by Instituto Nacional de Técnica Aeroespacial (INTA), the Spanish research center for Aerospace, in order to achieve reliable finite element model simulation of damaged structures including PFA are presented. These include delamination growth in composites, debonding onset and growth, and crack growth in thin metallic structures. Besides accuracy, INTA focuses on developing computationally efficient techniques and correcting mesh size effects.

Keywords

Progressive failure analysis

Finite element model

Virtual crack closure technique

Cohesive zone

Crack tip opening angle

Chapter outline

1 Introduction   4

2 Delamination Growth in Composites   4

2.1 VCCT Fundamentals   5

2.2 Experimental Benchmark and FEM Simulation   7

2.3 FEMs Comparison   8

2.4 Delamination Growth Tool   9

2.5 Correlation Between FEM Simulations and Tests   9

2.6 Mesh Size Effects   10

2.7 Comparison of Mixed-Mode Failure Criteria   10

2.8 Conclusion and Further Work in Delamination Growth Analysis   10

3 Debonding Onset and Growth   12

3.1 DCB Coupon: Mode I Interlaminar Fracture Toughness Test   13

3.2 FE Modeling   13

3.3 CZ Fundamentals   14

3.4 Mesh Dependency   14

3.5 Experimental Results   16

3.6 Correlation FEM Simulation-Tests   17

3.7 Conclusion and Future Work in Debonding Analysis   20

4 Crack Growth in Metallic Structures   20

4.1 CTOA Criterion-Experimental Obtaining of CTOAC   21

4.2 Crack Growth Tool   23

4.3 Benchmarks Description   23

4.4 FEM Modeling   24

4.5 Correlation Simulations-Tests   24

4.6 Crack Growth in Metallic Structures-Conclusion and Future Work   26

References   26

1 Introduction

Failure analysis comprises the prediction of damage onset on a structure when subjected to loads and environmental conditions. Damages may consist in permanent structural deformations (plastic strains for instance); local damages as cracks for instance, or in general any deterioration of the structure, or lack of functionality. It should be noted that damage onset does not mean the final or catastrophic failure of the structure. Failure analysis also comprises the subsequent progressive failure analysis (PFA) that occurs when the structure is loaded in static or fatigue environments.

In PFA, two features are commonly studied. The first is damage progression before it may reach a critical size producing the final failure. The second is the so-called residual strength, the remaining capability of the damaged structure to withstand loads. In the aircraft sector, it is common to refer the concept of damage tolerance, which means that a structure in presence of undetected damages, either produced by manufacturing defects, fatigue, ambient conditions, or accidental, is still able to withstand the loads produced during its service life. The fail-safe concept is employed and is defined as damage that must not lead to failure before it is detectable by means of inspections. Properly understanding of failure causes and PFA, allows the engineer improving and optimizing the structural design, additionally improving structural reliability.

Instituto Nacional de Técnica Aeroespacial (INTA) is currently involved in developing reliable simulation techniques for damaged structures including PFA. The methodologies are based in the finite element model (FEM) technique and have been applied to typical aerospace structures made in metallic or composite materials, monolithic or sandwich, etc. Some examples are shown below:

 Composite structures with interlaminar delaminations: prediction of delamination growth under static loads.

 Debonding analysis: prediction of debonding onset and growth under static loads.

 Crack growth in thin metallic structures (structures with high plasticity).

Besides accuracy, INTA focuses on developing efficient techniques from a computational point of view (reasonable computation and postprocessing time) as well as understanding and correcting the mesh size effects (FE results dependency on mesh size).

2 Delamination Growth in Composites

The general trend in modern aircraft structures is the progressive replacement of metallic materials with composites. Composites exhibit superior structural properties such as higher stress allowable, better behavior in fatigue and damage tolerance, less sensitivity to corrosion phenomena, etc. Both the new Airbus A350 and Boeing B787 each with over 50% of their structure made up of composites are illustrations of this tendency. The fuel consumption of these aircrafts is reduced around 20%.

One of the typical failure modes of composite materials are interlaminar delaminations, which means a lack of cohesion between adjacent plies in the laminate. Delaminations can be originated by design features prone to develop interlaminar stresses (curved sections, drop-offs, free edges, etc.), manufacturing defects (shrinkage of the matrix during curing, formation of resin-rich areas, etc.), or accidental causes such as tool impacts. Fatigue loading may create fiber-matrix debonding within the material (interfacial failure) or microscale matrix damage, which eventually leads to interlaminar delaminations.

Delaminations degrade material structural properties and reduce its structural load capacity. Moreover, they are prone to grow when compression and/or out of plane loads (static or fatigue) are applied to the structure. An additional inconvenience of internal damages of this type is that they require complicated and expensive inspections in order to be detected.

The lack of knowledge of the composites damage mechanics regarding delamination for instance, joined with the high dependency of analysis results from material parameters difficult to characterize experimentally, which has traditionally led to rather conservative designs. Moreover, Airworthiness certification requirements are generally more restrictive for composites than for metallic materials. Therefore, when developing a new aircraft model, a more extensive test campaign is often required. Currently, aircraft developers use a strain design approach for composites in order to cover impact damage and to avoid delamination growth. For monolithic laminates, this limit is typically 3500-4000 µ?, while for other applications, such as honeycomb panels, lower limits are quoted.

Substantial research has been carried out recently to find accurate and reliable simulation methodologies for damaged composite structures, in either static or fatigue load environments. Many authors have characterized delamination growth in composites and one of the tools used is the virtual crack closure technique (VCCT) [1] that requires the calculation of the strain energy release rates (SERR) according to the pure modes of fracture. VCCT requires a predamaged structure, and it is able to predict delamination (or debonding) growth.

In the following sections, a methodology developed at INTA based in VCCT theory, with the ability to predict delamination growth under static loads...