Elsevier

Composite Structures

Volume 133, 1 December 2015, Pages 911-920
Composite Structures

Compression after impact strength of repaired GFRP composite laminates under repeated impact loading

https://doi.org/10.1016/j.compstruct.2015.08.022Get rights and content

Abstract

During their service life, composite materials are prone to damage, which compromises their structural performance significantly. In this study, glass/epoxy composite specimens fabricated using hand layup method and further cured in a compression molding machine were cut from the laminates and subjected to low velocity impact damage in order to investigate the effects of repair. The impacted laminates were repaired by removing the damage area with a circular cutout and filled with a chopped short Kevlar/epoxy: the efficiency of the repair procedure and toughness of the repaired laminates were examined by repeated impacts on the repaired site. The residual strength of the post impacted repaired laminates is investigated by the conduction of compression after impact (CAI) loading with acoustic emission monitoring (AE) technique. The structural performance of both repaired and unrepaired laminates are compared and discussed.

Introduction

Glass fiber reinforced composite materials are widely used for structural applications such as aircraft, automotive, ships and space launch vehicle components due to their distinct properties such as light weight, relatively high specific properties and low maintenance cost [1]. During their service life, composites are likely to undergo various impact loading conditions. For example, a foreign object can strike during takeoff, a tool can be dropped onto composites during maintenance operations and impact of debris, fragments or projectiles can impact the composite structures. As a consequence of these accidental impacts, damages like matrix cracking, delamination, debonding and fiber breakage can occur in the composite structures.

During low velocity impacts (generally regarded to be at velocities lower than 10 m/s), the impactor contact time is deemed to be long enough for the entire FRP structure to be affected by impact loading [2]. Low-velocity impact of fiber-reinforced plastic composites has been the subject of many experimental and analytical investigations [3], [4], [5]. Some investigators have used compression-after-impact (CAI) and other static tests to estimate the damage tolerance and residual strength after low-velocity impact [6], [7], [8], [9]. Repeated impacts on composite structures are used to find out about damage progression and progressive decrement in residual strength. By subjecting specimens to repeated instrumented impacts, residual strength and damage progression can be monitored. In a study on the response of stitched and unstitched E-glass/epoxy laminates subjected to transverse impact, specimens were repeatedly impacted to examine the growth of damage, including delamination and transverse cracking [10]. Out-of-plane displacement generates flexural and shear stresses which can lead to matrix cracking as well as fracture of fibers. The resultant intra-ply micro cracking and inter-ply delamination can be significant even without the presence of visual surface evidence of the impact event itself, therefore even if energy producing barely visible impact damage (BVID) is not yet reached [11], [12], [13], [14]. In this sense, low-velocity impact may be insidious and pose a considerable challenge in its ability to significantly reduce mechanical performance without being readily detectable [15], [16].

The propensity of composite panels to external impact damage requires proper repair and maintenance operations. Sometimes functional requirements may lead to introduce cutouts in the structures for electrical wiring, installation of electronic instruments and access to hydraulic lines which induces high stresses around the cutout region. These necessary operations can cause buckling and post buckling behavior to be easily precipitating towards laminates collapse, with this evolution being fast hence not easily predictable.

Applying external patches offers a fast and effective repair, which can be mechanically fastened [17] or adhesively bonded [18]. To prevent the crack to propagate further, it is a normal practice to remove the damage area before the patch is fixed in that place. Laminates with thickness exceeding 2–3 mm generally carry too much load for external patch repairs and therefore scarf repairs are often recommended [19]. The key benefit of a scarf repair is the relatively uniform adhesive shear stress distribution along the bonded area, providing increased load-bearing capability. The damaged material is removed prior to replacement with virgin material that closely matches the properties of the parent one [20]. Tapered angles, usually from 20:1 to 60:1, are often required; therefore, a large volume of undamaged material must be removed, especially in thick laminates [10]. An alternate repair approach has been reported, whereby matrix cracks and delamination are infused with a resin via holes drilled in the damaged zone [21], [22]. Greszcuk [23] reported that the strength of heavily loaded wings and fuselage skins in aircraft structures for example could be reduced substantially by local discontinuities, such as presence of holes or impact damage. Recent studies involved in the investigation of composite repair have suggested overwrap repair as an alternative repair system for composite pipelines, which has been integrated into the ASME B31.4 [24], B31.8 [25] CSA and Z662 [26] pipeline codes. Recent work carried out by Goertzen and Kessler on carbon/epoxy composite has suggested that composite repair to be more economical than the other conventional repair methods used [27].

Hideki et al. [28] conducted a study on locations and shapes of crack and disbond fronts in aircraft structural panels repaired with bonded FRP composite patches. Engels et al. [29] have performed closed-form analysis of external patch repairs of laminates. In the paper, the problem of a laminate plate with an elliptical hole repaired by elliptical patches under in-plane and bending load was investigated using the classical laminate theory [29]. Baker et al. [32] concluded that, although bonded composite patches often offer a far more effective repair than conventional mechanically fastened patches, full credit cannot be given for their effectiveness in reducing crack growth when used to repair flight safety structure [30]. Liu and Wang [31], and Baker et al. [32] have performed the investigation on progressive failure analysis of bonded composite repairs. In their work, a 3-D progressive damage model was developed and verified by experimental study. Using this model, various repair parameters were studied and the failure initiation strength and ultimate strength of these bonded repaired structures were predicted. A significant non uniformity of shear and normal stress distribution in the adherends was observed and correlated with three-dimensional stress analysis predictions.

The residual strength of the repaired plate is reduced by mechanical fasteners because of stress concentration but adhesively bonded composite patch have been shown to provide high levels of bond durability under the operating conditions. The application of adhesive bonded patches normally allow recovering as much as possible the structural integrity: this includes stiffening of damaged regions, restoring strength or stiffness and reducing stress intensity factor (SIF) [32]. Two kinds of patch works are usually employed in composite repair: single sided (un-symmetrical) and double sided (symmetrical). Mostly double sided patch work is preferred as a higher reduction in SIF has been measured using this method [33]. Being a symmetric patch, for double sided patch work, the maximum strengthening effect of repair would take place around the defect area [34]. Composite laminates subjected to impact loading after patch repair suffers a significant reduction in post-impact compressive strength. This can be due to a number of reasons: this would include instability caused by delamination of plies, stress concentration effect caused by the reduced stiffness distribution inside the damage region and presence of an indentation inducing local out-of-plane deflections and bending under in-plane loads. Hence, compression after impact (CAI) performance remains an important design criterion in composite structures. Ghelli et al. [35] have reported that delamination produced by impact could change the buckling mode and/or decrease both critical load and ultimate strength. Compression loading on impacted laminates will lead to different failure modes such as matrix cracking, micro buckling, and delamination, fiber–matrix debonding, kinking and fiber failure. The mode of failure generally depends on material type, geometry and repair methodology [36], [37], [38].

To prevent catastrophic failure in composite patches, damage mechanisms like delamination, matrix cracking, and fiber pull out should be detected. Therefore, a suitable non-destructive testing system can be implemented for continuous online monitoring. The active patch material can also be used as sensors for predicting failures in the repaired system with slight adjustment utilizing their sensitivity to electro-mechanical coupling [39]. Acoustic emission (AE) is a non-destructive technique regularly used for structural health monitoring of materials in evaluation and propagation of damage. Many researchers have attempted to categorize the failure modes using two different approaches, namely parametric-based approach and signal-based approach. In signal-based approach, dominant frequency content of the different failure modes were investigated using Fast Fourier Transform (FFT) analysis of AE waveforms. The event primary frequency alone was enough to characterize each acoustic event, due to the fact that signals having similar frequency content may have also a similar source mechanism or damage [40]. Considering these facts, the present work aims at performing an experimental work on the repair of composite laminates subjected to multiple impact loadings. The effects of repair on the efficiency and the toughness of the repaired laminates are examined by repeated impacts on the repaired site. The residual strength of the post-impact repaired laminates is further investigated by conducting compression loading with acoustic emission monitoring (AE). The procedure involves removing the damage area with a circular cut out and later fills them with a chopped short fiber Kevlar/epoxy laminate of the same geometry and size.

Section snippets

Materials and fabrication of composite laminates

Glass/epoxy cross-ply laminates of stacking sequence [0°/90°/90°/0°]3 with size of 300 mm × 300 mm were prepared by hand lay-up method using unidirectional E-glass fiber. The nominal thickness of the laminate was measured to be 3 ± 0.045 mm. The matrix medium used was epoxy resin Araldite and hardener HY551 in the ratio of 10:1 by weight. The laminates were allowed to cure at room temperature under a pressure of 5 MPa in a 30 kN compression molding machine for 24 h. ASTM D7137M-12 standard CAI specimens

Results and discussion

Fig. 7 shows the contact force versus time for both repaired and unrepaired specimens subjected to repeated impacts. From the figure it was evident that peak contact force was found to be higher and the contact duration was found to be lower for repaired specimens in comparison with the unrepaired ones. This reveals the higher rigidity of the specimens repaired with chopped Kevlar short fibers, which might enhance their capacity to resist a higher number of impacts. Fig. 8(a)–(c) shows the

Conclusions

In the present work, an experimental investigation is carried out to understand the effect of repair procedure on the cross-ply glass/epoxy laminates subjected to post-impact compression loading using both destructive and non-destructive methods. The main results enabled observing that on the low velocity falling weight impact tests, the peak contact force was found to be at the higher side and the contact duration was found to be lower side for repaired specimens in comparison with the

Acknowledgement

The authors would like to express our sincere appreciation to Mr. Balaji for his excellent support in carrying out some of the experimental works.

References (45)

Cited by (108)

View all citing articles on Scopus
View full text