Transverse Cracks Research Articles

Automated detection of pavement distress based on enhanced YOLOv8 and synthetic data with textured background modeling

Regular inspections of pavement distress are essential for accident prevention, and deep learning based algorithms have been developed to ensure high accuracy of inspections. However, the lack of available data is a critical challenge for existing algorithms. To address this issue, an elaborate image synthesis strategy is proposed. By combining textured background modeling for real pavement and Unreal Engine-based distress block stitching technique, high-resolution virtual images that are indistinguishable from real images are generated, including five types of pavement distresses. In addition, an enhanced YOLOv8 network utilizing synthetic data is designed in this paper. The enhanced YOLOv8 network is embedded with the Squeeze and Excitation attention module, and the Swin Transformer module, which are designed to distinguish different types of pavement distress accurately and suppress the interference of complex backgrounds (noise, shadow, blur). The results show that the performance of the algorithms trained based on appropriately ratios of synthetic data (2:1 ratio of virtual to real) improves by more than 10% compared with no virtual data. The enhanced YOLOv8 network achieves a Mean Average Precision (MAP) of 94.8% for transverse cracks, longitudinal cracks, cross-cracks, alligator cracks, and potholes, which is better than seven existing object detection models. The proposed image synthesis method can contribute to improving the accuracy and reliability of pavement inspection and alleviate the reliance on large number of distress samples collection.

Fuzzy Rule Based System of Optimal Pavement Maintenance and Rehabilitation Strategies for Yangon-Mandalay Expressway

Purpose: Systematic pavement management plays a vital role in the better Road Asset Management system. The lifespan of pavement can be preserved and prolonged by adequate maintenance measures in proper time. By early detection and repair of defects at initial stages with predictive maintenance reduces the rapid deterioration of the pavement. Moreover, inaccuracy in pavement deterioration prediction leads to an improper maintenance with wastage of time, cost and labor strength. To overcome all of these obstacles, this paper attempts an approach on optimal pavement maintenance strategy by means of fuzzy rule-based system. Materials and Methods: In this research, four pavement indicators of PCI, IRI, PSR and PSI have been taken into account as antecedents and five different types of pavement maintenance as consequents. In this research, ASTM D-6433, Road Lab Pro software, AASHTO guidelines and Fuzzy Logic (Mamdani) model have been used to evaluate the current pavement condition and optimal pavement maintenance strategies. Findings: According to research area, 130 total pavement sample units were undertaken to study for the mean pavement condition of PCI value 54.93 (poor), IRI 7.01 (fair), PSR 3.33 (fair) and PSI 3.35 (fair). Due to 13 years-old life span of traffic volume, temperature and other external factors, this expressway is suffering from potholes, bleeding, longitudinal and transverse cracking, weathering, joint reflection cracking for Asphalt Concrete (AC) overlay and linear cracking, joint seal failure, scaling, faulting, depression, lane/shoulder drop off for existing concrete pavement. According to the fuzzy logic model results, 42 numbers of fuzzy If-Then maintenance rules were verified as the most relevant pavement maintenance and rehabilitation strategies for Yangon-Mandalay Expressway. Implications to Theory, Practice and Policy: This research provide implications to the study and contributions to theory, practice and policy. This research paper insights into effective infrastructure maintenance and investment, emphasizing the role of advanced computational techniques in transportation economics. The key advantage of this research paper is to be effective, supportive, easy and accessible decision-making tool for transportation and pavement engineers, expressway planners and highways department authorities especially from developing countries.

Ballistic impact wave and back bulge propagation mechanism and blunt trauma assessment of ceramic/UHMWPE composite body armor

With increasing local conflicts worldwide, studying the impact response of body armor is crucial for enhancing soldier survival rates. However, there is still insufficient understanding of the back bulge expansion patterns, the impact waves transmission features, and blunt injury assessing method of body armor. To investigate these problems deeply, the three-dimensional digital image correlation method and numerical models were employed to reveal the ballistic response mechanism of ceramic/ultra-high-molecular-weight polyethylene (UHMWPE) composite body armor impacted by 7.62 mm rifle bullet. This study built theoretical function for the bulge expansion, revealed the wave transmission mechanism and fractures in both materials and their interaction. Firstly, the experiment revealed that the back-face bulge transverse expansion velocity exhibits a double exponential decrease. Secondly, the contour of back-face bulges consistently demonstrated excellent hyperbolic characteristics at various moments. Thirdly, the average Viscous Criterion (VC) and Blunt Criterion (BC) values were 5.37 and 0.95, corresponding to Abbreviated Injury Scale (AIS) values of 6 and 4. Fourthly, the simulation revealed that the ceramics transverse and radial cracks seems primarily induced by shear stress xy; ceramics conical cracks occurring at the boundaries of equivalent stress seems primarily induced by stress in x and y directions. Fifthly, numerical results revealed that the velocity of compression waves within the laminate increases over time in the impact direction and consistent with ceramic velocity impacted on the laminate. These findings lead to experimental and theoretical advances in the impact response mechanism of composite body armor and provide index for improving the protective performance.

Cracking detection of unidirectionally reinforced SiC/SiC composite by X-ray Talbot–Lau interferometry

We developed a Talbot–Lau interferometry-based X-ray imaging system with a novel material testing system and studied its feasibility for monitoring the damage evolution process in SiC/SiC composites directly. We tensile-tested a polymer-derived unidirectional mini-composite, the specimen was repeatedly loaded and unloaded, the maximum load was successively increased, and images were captured when the specimen was tensed at the maximum load and when released. Unique X-ray images were obtained, and by comparing them with conventional X-ray computed tomography absorption images, we ascertained the advantage of Talbot–Lau interferometry-based X-ray in terms of low-magnification crack detection over very short periods. The results demonstrate that the proposed X-ray interferometry can detect the damage evolution of the composite in the form of randomly initiated transverse cracks in the matrix, even after unloading. These results indicate that Talbot–Lau interferometry-based X-ray imaging is viable as a nondestructive inspection tool for SiC/SiC composites.

Electroosmotic strengthening of soft clay under different electrification modes

AbstractElectrification mode is one of the important factors affecting the energy consumption and drainage effect of the electroosmosis method. In this work, a series of laboratory tests were performed to explore the impact of continuous changes in the electric field on the migration of ions and electrons, as well as the migration of electroosmotic flow. The results show that different electrification modes had a significant impact on the current and water discharge. An electrode reversed from anode to cathode, from one circuit to another, is beneficial to drainage, while an electrode reversed from cathode to anode, from one circuit to another, is unfavorable to drainage. The drainage effect of the specific cyclic and progressive electroosmosis (CPE) is not as good as the conventional electroosmosis method. However, combining the two electrification modes can further enhance the shear strength of the soil near a designated electrode. The generation of transverse cracks in soil is mainly due to the drainage difference caused by the electroosmotic flow, which is caused by the adjacent electric field at that location. When the electroosmotic flow is strong, longitudinal cracks will occur in the soil along the strongest electroosmotic flow path. In practical in situ engineering applications, the electrification mode of CPE will be more energy‐efficient than conventional electrification modes with the same processing time.

Transverse cracking signal characterization in CFRP laminates using modal acoustic emission and digital image correlation techniques

The process of formation and subsequent propagation of transverse cracks in 90° plies of carbon-fiber laminated composites was studied using modal acoustic emission approach and digital image correlation techniques. The results from modal acoustic emission approach, which included a newly developed processing tool for acoustic emission waveforms, provided information for identification and subsequent characterization or localization of signals originating from transverse cracking by analysis of the separated flexural and extensional Lamb wave modes in terms of their modal parameters. The digital image correlation method served as a verification tool of the acoustic emission data outputs in the terms of activity of significant localized events originating from the formation of the transverse crack in the 90oply. This made it possible to specify more locally the accompanying activity belonging to the formation or propagation of the magistral transverse crack. The manuscript also presents results related to the evolution of flexural/extensional wave modal parameters as the function of loading force for both [0/0/0/90]S and [90/0/0/0]S panels. It can be concluded that the detection of transverse cracks requires the need for applying a more complex acoustic emission data analysis methodology, while the standard parametric analysis, including the waveform peak frequency, is not sufficient. The presented methodology may serve as a basis for development of robust analysis tool capable of detecting the investigated phenomena.

Crack Detection of Curved Surface Structure Based on Multi-Image Stitching Method

The crack detection method based on image processing has been a new achievement in the field of civil engineering inspection in recent years. Column piers are generally used in bridge structures. When a digital camera collects cracks on the pier surface, the loss of crack dimension information leads to errors in crack detection results. In this paper, an image stitching method based on Speed-Up Robust Features (SURFs) is adopted to stitch the surface crack images captured from different angles into a complete crack image to improve the accuracy of the crack detection method based on image processing in curved structures. Based on the proposed method, simulated crack tests of vertical, inclined, and transverse cracks on five different structural surfaces were conducted. The results showed that the influence of structural curvature on the measurement results of vertical cracks is very small and can be ignored. Nevertheless, the loss of depth information at both ends of curved cracks will lead to errors in crack measurement outcomes, and the factors that affect the precision of crack detection include the curvature of the surface and the length of the crack. Compared with inclined cracks, the structural curvature significantly influences the measurement results of transverse cracks, especially the length measurement results of transverse cracks. The image stitching method can effectively reduce the errors caused by the structural curved surface, and the stitching effect of three images is better than that of two images.

Automatic Road Pavement Distress Recognition Using Deep Learning Networks from Unmanned Aerial Imagery

Detecting and recognizing distress types on road pavement is crucial for selecting the most appropriate methods to repair, maintain, prevent further damage, and ensure the smooth functioning of daily activities. However, this task presents challenges, such as dealing with crowded backgrounds, the presence of multiple distress types in images, and their small sizes. In this study, the YOLOv8 network, a cutting-edge single-stage model, is employed to recognize seven common pavement distress types, including transverse cracks, longitudinal cracks, alligator cracks, oblique cracks, potholes, repairs, and delamination, using a dataset comprising 5796 terrestrial and unmanned aerial images. The network’s optimized architecture and multiple convolutional layers facilitate the extraction of high-level semantic features, enhancing algorithm accuracy, speed, and robustness. By combining high and low semantic features, the network achieves improved accuracy in addressing challenges and distinguishing between different distress types. The implemented Convolutional Neural Network demonstrates a recognition precision of 77%, accuracy of 81%, mAP of 79%, f1-score of 74%, and recall of 75%, underscoring the model’s effectiveness in recognizing various pavement distress forms in both aerial and terrestrial images. These results highlight the model’s satisfactory performance and its potential for effectively recognizing and categorizing pavement distress for efficient infrastructure maintenance and management.

Consideration of the effects of frost heave force and train loads on the cracks and propagation pattern of ballastless track slabs in cold regions

The construction of ballastless railway tracks in cold regions is susceptible to frost heave forces, which may particularly impact the stability of track slab cracks and lead to crack propagation. In this paper, the frost heave tests are conducted on the concrete specimens with prefabricated cracks to observe the evolution of the frost heave force within the cracks. The frost heave force is subsequently applied to the finite element model of the concrete specimens to calculate the stress intensity factors (SIFs). In addition, the CRTS III ballastless track model is established to investigate the crack propagation patterns under frost heave and train loads. The results show that the frost heave force in concrete cracks during freezing and thawing is divided into five stages, which produces two peaks of frost heave force. Train loading caused the SIFs of the crack tips to open and close alternately and to rise significantly under the coupling effect of frost heave load. The SIF increases approximately linearly with crack width and is higher for transverse cracks than for longitudinal cracks. Small size cracks do not possess sufficient SIF values to exceed fracture toughness, implying no crack propagation. However, as the crack width is constantly increasing, the SIF will eventually exceed the fracture toughness. Therefore, strict control over track slab's crack width is necessary for CRTS III ballastless tracks.

Field Validation of Deep-Learning-Based Ground Penetrating Radar Image Analysis for Advancing Subsurface Distress Detection

This research introduces an innovative method for detecting subsurface cracks within pavements by leveraging ground penetrating radar (GPR) technology in conjunction with advanced deep learning techniques. Its primary aim is to significantly improve the accuracy and efficiency of pavement assessment, particularly for operational and maintenance purposes. The proposed model, GPR-YOLOR (You Only Learn One Representation), extends the YOLOR framework and incorporates a region of interest within the top pavement layer to detect subsurface cracks. While the model can be trained with annotated data, the main challenge lies in validating results in the field because of the inability to visually inspect subsurface conditions and the high cost associated with direct coring. To overcome this challenge, we propose an alternative approach that utilizes the co-occurrence of surface cracks as pseudo labels, allowing for easy verification. To ensure that surface cracks correspond to subsurface cracks, the focus is exclusively on transverse cracks that develop in a bottom-up manner, such as fatigue and reflective cracks. Through this methodology, our GPR-YOLOR model achieves an F1 score of 0.72, with a precision of 0.76 and a recall of 0.68. The results from field validation underscore the effectiveness of the GPR-YOLOR model in accurately identifying subsurface cracks, highlighting its practical significance in conducting field condition assessments.

Experimental investigation of dowel action in reinforcing bars using refined measurements

AbstractIn typical reinforced concrete design, reinforcement is designed to carry axial forces, but it can also resist transversal forces by dowel action. This is usually neglected for simplicity's sake in the design phase, but it can be accounted for either explicitly in mechanical models or implicitly in empirical relationships. Furthermore, there are cases where the connection between various concrete elements explicitly depends on dowel action, as for example, in connections between precast elements or between two concrete parts cast at different times. On the other side, dowel action can have a negative impact on the fatigue resistance of reinforcing bars subjected to cyclic loading, because of the local stress concentrations near interfaces due to relative movements, either in sliding or in opening of cracks not perpendicular to the bar. For the assessment of the remaining capacity of existing structures, improved models of the behavior are needed, including realistic models of the behavior of concrete, steel and their interfaces. The aim of the present paper is to provide a contribution to a better understanding of dowel action by two test series. The first series focused on the behavior of the dowel: the concrete specimens with the embedded bars were placed in a custom‐made test setup and subjected to monotonic or low stress‐level cyclic actions with a longitudinal and a transversal crack opening component, up to developing the full plastic capacity of the dowel and rupture at the peak of catenary action. The measurement system included tracking the displacement field at the surface of the concrete and the strains in the dowel by optical fibers glued on its surface. The latter measurements allow to derive the internal forces in the reinforcing bar and deformed shape of the bar as well as the contact pressure between the bar and the surrounding concrete. The results show a strong dependency on the test variables: diameter of the bar, imposed crack kinematics and angle between the bar and the crack. The second test series looked more closely at the behavior of concrete underneath the bar, in the presence of a point load introduced at various locations into concrete through a reinforcing bar. A comparison of the test results with existing models shows a general good agreement and some aspects that deserve to be improved.

Multiscale structural understanding of plasma spraying anti-ablation coating: an example of Ta-Hf-W-C ultrahigh temperature ceramics

The coating formation and structural design of ultra-high temperature ceramics (UHTCs) for aerospace vehicle anti-ablation coatings through plasma spraying remain uncertain despite extensive engineering applications. In this study, (Ta,Hf,W)C/HfC micro-composite, (Ta,Hf,W)C single-phase, and (Ta,Hf,C)/W nanocomposite coatings were successfully prepared using induction plasma spheroidization (IPS), vacuum plasma spraying (VPS), and spark plasma sintering (SPS). Density functional theory (DFT) was employed to capture the electronic structure of (Ta,Hf,W)Cy (y = 1, 0.875, 0.75, 0.5), confirming the influence of carbon vacancies and ceramic metallicity on the plastic forming behavior of sprayed coatings. Additionally, the contribution of HfC to structural stability and WC to the metallicity of carbon-based rock salt-structured UHTCs was elucidated. The pivotal role of carbon vacancies in inducing metastable phase formation and metal phase precipitation for (Ta,Hf,W)C was identified. Ablation behavior variations resulted from differences in multi-scale coating structures, where transverse cracks exhibited more detrimental effects than vertical cracks. Notably, a scale effect of oxidation related to the uniformity of metal elements in the multicomponent ceramic coating was uncovered. It was determined that components capable of forming high melting point oxides should not be uniformly dissolved in the matrix.

Failure mechanisms for Gd2O3–Yb2O3 co-doped YSZ thermal barrier coatings under high-temperature gradient

In this study, Gd2O3–Yb2O3 co-doped YSZ thermal barrier coatings with a top coat thickness of 400 μm were prepared on a nickel-based superalloy by atmospheric plasma spraying. The samples underwent multiple short-term and single long-term thermal exposures through burner rig tests under high-temperature gradients. The short-term thermal exposure duration of the coating was 25 s and the long-term thermal exposure duration was 1200 s. The microstructures and phase compositions of the samples were characterized, and their failure mechanism was discussed. The results demonstrate that during the burner rig tests, the coating surface temperature reached 2600K, resulting in coating sintering and sintered zone formation comprising coarse equiaxed grains. The short-term thermal exposure life of the coating is 3 cycles, and the long-term life is more than 1200s. After the single 1200s thermal exposure, the coating exhibited a stable cubic phase. However, under multiple 25s thermal exposures, the depth of the sintered zone increased with increasing number of cycles, leading to the formation of various vertical and transverse cracks within the coating. Coating failure was due to a combination of sintering and thermal-mismatch-induced stress.The early-stage coating delamination was attributed to sintering-induced transverse cracks, while the overall spallation was primarily ascribed to thermal-mismatch-induced transverse cracks.

Characterizing damage evolution in fiber reinforced composites using in-situ X-ray computed tomography, deep machine learning and digital volume correlation (DVC)

In-situ X-ray computed tomography (CT) techniques enable to examine microstructural evolution of composite materials continuously and nondestructively under loading for better understanding the initiation and propagation of microscopic damages. This study explores the tensile damage behavior of notched woven carbon/epoxy composites under successive loading process. Since the image contrast of different composite material phases is normally low and conventional image segmentation techniques are hard to identify microstructures and defects in different directions distinctively, a U-Net image segmentation model based on deep machine learning algorithms is employed to characterize microstructures and material damages precisely and consistently. Digital volume correlation (DVC) approach is proposed here tocharacterize the three-dimensional (3D) deformation fields of carbon/epoxy composites at different loading steps, and the DVC results are used to predict location of damage initiation and propagation. It is found that the damage modes of notched carbon/epoxy composites mainly include fiber breakage, longitudinal cracks in warp fiber tows, and transverse cracks in weft fiber tows. The individual fiber breakage occurs in warp fiber tows at the intermediate loading level, and longitudinal and transverse cracks appear in further loading levels. Longitudinal cracks and fiber breakage events take place in a region tangent to the notch tip and exhibit a strong interaction. The development of longitudinal cracks promotes the initiation and propagation of transverse cracks. In addition, the strain concentration regions in 3D deformation fields match well with the microscopic cracks segmented by the U-Net deep learning model. This study demonstrated that the integration of X-ray computed tomography, digital volume correlation and image semantic segmentation enables to identify damage initiation, evolution, extent and mechanism.

Formation of transverse cracks in cross ply laminates in an environment of nonuniformly distributed fibers

This work is concerned with extracting characteristic features of transverse crack formation in cross ply laminates under axial tension considering nonuniform distribution of fibers in the ply cross section. The fiber-matrix debond crack initiation sites are determined first by the dilatation induced cavitation criterion for an epoxy matrix. The growth of the debond cracks is analyzed by calculating the energy release rates using the virtual crack closure technique. The kink-out of the debond cracks into the matrix is determined in short incremental steps by the maximum energy release rate criterion. Different scenarios are considered for the linking up of the kinked-out cracks to form continuous transverse cracks. By studying two different degrees of fiber distribution nonuniformity, the interactive effects due to the fiber distribution on the transverse crack formation are clarified.

Explicit modelling of meso-scale damage in laminated composites – Comparison between finite fracture mechanics and cohesive zone model

The ability of carbon fibre reinforced polymers to respect functional requirements such as permeability is strongly related to their damage state. In order to identify the fracture properties on which transverse cracking kinetics depend, we combine experimental and virtual tests by explicitly modelling all transverse cracks. Two modelling methods are compared, the cohesive zone model (CZM) and finite fracture mechanics (FFM). Both models are based on a double criterion of energy and strength. Each potential crack is assigned a couple of material properties (fracture energy and strength) derived from a probability distribution. This study focuses on how modelling choice implies assumptions on energy dissipation and the cracking behaviour represented. The relevance of both modelling methods is also discussed. A methodology is provided for identifying fracture properties and predicting damage accumulation in an isolated ply.

Impact of cement content in cement bound materials on the reflection cracking performance of asphalt pavements

This study investigates the impact of cement contents in cement bound material (CBM) on asphalt pavement reflection cracking, considering two levels: 4.5 % (CBM1) and 5.4 % (CBM2). Mechanical properties of CBMs, including elastic modulus, rupture modulus, and thermal expansion, were evaluated. Simultaneously, properties of asphalt pavement layers (wearing course and base course) underwent assessment through dynamic modulus, flow number, and overlay tests. Resilient modulus tests were conducted for subbase and subgrade layers. Laboratory test outcomes served as inputs for a AASHTOWare Pavement ME Design program. The results indicated that despite CBM2's higher elastic modulus and rupture modulus, CBM1 outperformed in pavement design. Based on the outcomes from pavement design program CBM1 exhibited lower total transverse cracking (400.19 m/km) compared to CBM2 (403.38 m/km) after 6 years of service. The higher cement content in CBM2 was identified as a potential contributor to increased cracking, attributed to shrinkage during cement hydration process. Moreover, the higher cement content showed a higher coefficient of thermal expansion (CTE), where CBM1 displayed a CTE of 70 µm/°C, while CBM2 material had a CTE of 40 µm/°C. These findings emphasize the intricate relationship between cement content in CBM and pavement performance, crucial for effective design considerations.

Performance of fly ash and CNI as corrosion prevention methods for steel reinforcement embedded in cracked HPC concrete exposed to a natural marine environment

In this study, the effectiveness of fly ash and a calcium nitrite-based corrosion inhibitor in cracked high-performance concrete was investigated. Concrete prisms were fabricated and the specimens featured artificially transverse cracks 0.25 and 0.5 mm width. These specimens were exposed to a natural marine environment for seven years and the corrosive activity was monitored annually using electrochemical techniques. At the end of the exposure period, chloride ion content at the steel-concrete interface, mass loss, pitting depth and tensile parameters on the steel reinforcement were estimated. It was found that cracking exacerbated corrosion and the prevention methods assessed or their interactions were ineffective. Ultimate strength and ductility losses of the steel reinforcement ranged between 7% and 11% and 44–60% for 0.25 mm crack, while between 11% and 14% and 48–66% for 0.50 mm crack. The results from this research are original because they contribute to elucidating the effect of the interactions between the investigated corrosion prevention methods.

Stress and energy release rate analysis of cross-ply carbon fiber-reinforced plastic laminate with transverse cracks subjected to ultrasonic vibration using a variational approach

Application of carbon fiber reinforced plastics (CFRP) to structures that are subjected to increasingly high numbers of cyclic loads is expected to expand, so evaluation of fatigue properties up to the very-high cycle regime is essential to prove their long-term reliability. Conventional fatigue testing is not practical for such conditions owing to the excessive time required; ultrasonic fatigue testing, which can be performed at a frequency of 20 kHz, is an effective alternative. Stress distribution in ultrasonic fatigue testing differs from that in conventional fatigue testing because the specimen is resonated. The stress distribution of cross-ply laminates with transverse cracks subjected to resonance was calculated using a variational approach. The energy release rate was calculated based on the results of a previous study on transverse crack initiation. The results of the stress analysis are in good agreement with those of finite-element analysis. Quantitative evaluation of transverse crack initiation was also conducted.

Numerical Simulation of Transverse Crack on Composite Structure Using Cohesive Element

Due to their anisotropic behavior, composite structures are weak in transverse direction loading. produces transverse cracks, which for a laminated composite, may lead to delamination and total failure. The transition from transverse crack to delamination failure is important and the subject of recent studies. In this paper, a simulation of transverse crack and its transition to delamination on cross-ply laminate was studied extensively using a cohesive element Finite Element Method (FEM). A pre-cracked [0/90] composite laminate made of bamboo was modeled using ABAQUS/CAE. The specimen was in a three-point bending configuration. Cohesive elements were inserted in the middle of the 90° layer and in the interface between the 0° and 90° layer to simulate transverse crack propagation and its transition to delamination. A load–displacement graph was extracted from the simulation and analyzed. As the loading was given to the specimen, stress occurred in the laminates, concentrating near the pre-cracked region. When the stress reached the tensile transverse strength of the bamboo, transverse crack propagation initiated, indicated by the failure of transverse cohesive elements. The crack then propagated towards the interface of the [0/90] laminates. Soon after the crack reached the interface, delamination propagated along the interface, represented by the failure of the longitudinal cohesive elements. The result of the numerical study in the form of load–displacement graph shows a consistent pattern compared with the data found in the literature. The graph showed a linear path as the load increased and the crack propagated until a point where there was a load-drop in the graph, which showed that the crack was unstable and propagated quickly before it turned into delamination between the 0o and 90° plies.