PSI - Issue 50

Alexander Eremin et al. / Procedia Structural Integrity 50 (2023) 73–82 Alexander Eremin / Structural Integrity Procedia 00 (2019) 000 – 000

74 2

epoxy laminates. However the intrinsic drawbacks of such materials are low impact toughness and delamination susceptibility leading to reduction of strength especially under compression loading. Laminates are tested in the research stage utilizing different standards. For the evaluation of impact properties ASTM proposes a two-step procedure. The first step involves a drop-weight impact damaging of the specimen and is governed by D7136 standard. It provides the absorbed energy and loading curve while a formed damage can be studied using NDT. The second one uses an impacted specimen to study its residual compressive strength (D7137). Joint implementation of these testing procedures widens the data on mechanical properties, particularly compression after impact (CAI) that describes the susceptibility of the composite to impact damaging. In order to correctly analyze the deformation of the materials the strain data should be precisely collected during the mechanical tests. The traditional instrumentation includes strain gages and extensometers. There are some non conventional methods described by Luyckx et al. (2011) like speckle interferometry, Bragg cells, etc. which might be too complex and expensive, so they have better performance in non-destructive testing rather than in laboratory strain measurements. The other example presented by Ružek et al. (2015) is the use of optical sensing of special shaped markers to derive the strain. Strain acquisition using digital image correlation (DIC) is a powerful alternative that becomes more and more popular. It does not demand expensive hardware and specimen preparation providing both local and full-field strain measurement. Ostré et al. (2016 ) and Bogenfeld et al. (2022) were utilized DIC to demonstrate the possibilities of the method for evaluation of mechanical behavior of various impacted carbon fiber reinforced polymers. 2. Manufacturing of carbon fiber reinforced polymer and testing techniques 2.1. Hot pressing of pregregs Laminates were manufactured using hot pressing of carbon fiber epoxy prepregs. The binder was DT190 by Deltapreg and the fiber was Toray T700. Stacking sequence was orthotropic [0/90] 5S . Moulding was performed at 120  C and a pressure of 0.7 MPa using Gotech 7014 thermal press. After the one hour dwell time the stack was postcured at 100  C for 24 h in a heated oven. The specimens for testing were cut on a CNC milling machine using solid carbide mills designed for CFRP cutting. Final sizes of the specimens were 150 х 100 х 2.5 mm 3 (Fig. 1a). 2.2. Drop weight impact testing technique Drop weight testing is governed by ASTM D7136 “ Standard Test Method for Measuring the Damage Resistance of a Fiber-Reinforced Polymer Matrix Composite to a Drop-Weight Impact Event ”. During the test the impactor velocity, kinetic energy, and displacement are plotted versus time. The ASTM standard defines the energy-to thickness ratio of 6.7 J/mm. Thus the nominal impact energy of 17.1 J was calculated. In order to study the sensitivity of material to different impacts the low and high energy impacts were added. The former corresponded to 0.5 of nominal energy (8.55 J) while the latter was 1.5 of nominal (25.65 J). Drop-weight tests were carried out using Instron Ceast 9350. After the testing the sizes of the obtained damage were estimated using digital shearography. This NDT method, as stated by Kim et al. (2012), allows detecting different flaws including delamination, cracks, and barely visible impact damages. Then the specimen was tested for compression after impact. 2.3. Compression after impact and Digital image correlation After the impact the specimens were tested according to ASTM D7137 “ Standard Test Method for Compressive Residual Strength Properties of Damaged Polymer Matrix Composite Plates ”. Loading was performed using BiSS UTM 150 kN servohydraulic testing machine at the rate of 1.25 mm/min. Image acquisition was performed via two Canon 700D DSLR cameras at a rate of 1 Hz providing three-dimensional DIC. Image processing was carried out using self-developed DIC technique. It allows obtaining in-plane strain components ( ε xx , ε xy , ε yy ) as well as a surface profile in z direction and out-of-plane displacements ( z and w ). During the post-processing, five virtual