Results

WP-1 Analysis of existing state-of-the-art knowledge base

The literature review is performed to summarise and to analyse the present state of the art in analysis, testing and validation of honeycomb composite sandwich panels. A total of almost 200 articles, reports and test standards has been overviewed and analysed for best practice transfer. The review was conducted to address, among others, several manufacturing issues – composite face skin materials and lay-ups, honeycomb core types with different thickness, cell size and cell wall thickness proportions, adhesives used to assemble the panel. Secondly, testing methods were addressed starting from standard tension and compression for components up to full scale sandwich panels experimental setups. This has turned out to be a core chapter describing a variety of methods, their boundary conditions, pros and cons in order to proceed with most efficient methods in project phase. Furthermore, current state of the art in numerical modelling has been analysed as well. Where it is required to considering intrinsic and extrinsic variables. Furthermore, review on non-destructive testing techniques applicable to sandwich panels has been examined and is included in further research process. Finally, the attention was focused on previous funded research projects in order to map the key players and scientific achievements so far and to build the current project coherently on outlined findings and conclusion.

WP-2 Concept design, analytical and numerical analysis of sandwich specimens

A vital necessity before panel manufacturing and planning test matrix was to determine design variables to be assessed. Damage tolerance is affected by a multitude of factors from panel construction to boundary conditions. These factors could be divided in two groups: intrinsic and extrinsic. Intrinsic variables stem from panel construction: skin layup, skin thickness, honeycomb cell size, honeycomb thickness, honeycomb wall thickness. Extrinsic variables: indentation type, indenter diameter and damage size or impact energy.
These assumptions where integrated in numerical analyses by commercial finite element software code ANSYS and scripting APDL log files. The main focus was on barely visible damage (BVID) introduction by indentation and estimation of residual dent depth and accumulated strains. While numerical methods capture all variety of boundary conditions, layup and honeycomb orientation as well as geometrical and material nonlinearities, the analytical approach assumes symmetry and equivalent load distribution around dent. Both, initial verification and final validation was performed in line with selected physical experiments. This included a reverse identification of reference panel performance assessment by self-frequency tests and modal analyses. At the current stage, we have achieved very good correlation among dozens of currently tested indented specimens by assessing residual dent (barely visible damage). The results are available in section SIV.

D 2.1 Numerical and analytical sandwich structure analysis - technical note

The main emphasis of the interim report is focused on development steps towards numerical and analytical methods for simulation of CFRP/Al honeycomb sandwich structures under compressive loading with introduction of barely visible damage (BVID). As input parameters from specimen prototyping and physical testing was pending, initial effort was given for finite element model verification with physical tests obtained from non-destructive evaluation tests reported in WP-3. Therefore the report summarises results of sandwich equivalent stiffness identification by methodology described in ASTM E1876. Furthermore, verification of detailed FEM model developed in commercial code ANSYS have been performed. All those verification cases have been provided in line with actual prototyping and NDE effort, thus detailed BVID and CAI simulations were shifted to next reporting period when models could be validated with physical tests.

D 2.2 Numerical and analytical sandwich structure analysis

The report summarises efforts in development of numerical analysis procedures which has been verified analytically and validated with initial experimental tests. All numerical analysis are implemented in commercial finite element software code ANSYS by scripting APDL log files. The main emphasis was focused on barely visible damage (BVID) introduction by indentation and estimation of residual dent depth and accumulated strains. While numerical methods capture all variety of boundary conditions, layup and honeycomb orientation as well as geometrical and material nonlinearities the analytical approach assumes symmetry and equivalent load distribution around dent. Both, initial verification and final validation was performed in line with selected physical experiments. This included a reverse identification of reference panel performance assessment by self-frequency tests and modal analyses. At the current stage, we have achieved very good correlation among dozens of currently tested indented specimens by assessing residual dent (barely visible damage). The results are available in section SIV.

WP-3 Manufacturing and NDT tests of the coupon scale sandwich specimens

A total of fifty 480 x 220 mm panels each composing of six 150 x 100 mm specimens required for compression after impact tests have been produced and tested. Four different lay-ups along with core thickness and height were investigated. The specimens were manufactured of 100 g/m2 carbon fibre reinforced plastic pre-preg sheets. Besides a required amount of residual strength specimens, a dedicated series of coupons where delivered to assess the mechanical characteristics and adhesion, damage properties. Furthermore, all produced panels underwent a series of non-destructive evaluation steps. This involved assessment of composite plate and CFRP/Al-honeycomb sandwich panel manufacturing and prototyping practice as well as barely visible damage introduction to testing panels. As a first step, NDE is applied as ultrasound inspection to assure that CFRP face plates are produced with relatively small scatter for thickness distribution, by delivering a histogram of thickness distribution for each skin of the panel. After that, each panel underwent a self-frequency nature test and assessment by Modal Assurance Criteria robustly assessing the whole panel maturity/robustness before the mechanical testing. A panel assemblies composed of six test specimens were inspected before and right after introduction of barely visible damage. The indentation depth was measured using laser scanner comparing groove depth to the reference plate. Obtained surface scan output file, representing xyz point cloud, is analysed to extract and validate damage severity. All of those results have been incorporated in software delivered within WP-6.

D 3.1 Mechanical tests and NDE evaluation - technical note

The aim of the deliverable / technical note is to summarize initial work towards prototyping of CFRP/Al-honeycomb sandwich panel manufacturing and prototyping practice in line with NDT evaluation and characterisation of obtained specimens. The work has focused on thin wall CFRP plate production with aim to reduce thickness distribution variety and to obtain almost no porosity specimens as one relevant to autoclave production. The next step was to focus on honeycomb extrusion and adhesion with CFRP skins for panel production. Finally, NDE by means of ultrasound A, B and C scans have been performed and statistical distribution of obtained quality have been investigated. In addition, NDT by means of physical self-frequency measurements was performed to have a second means of quality assurance of both – CFRP skins and CFRP/Al honeycomb panels.

D 3.2 Mechanical tests and NDE evaluation

The deliverable summarises evaluation of non-destructive testing results of mechanically tested specimens. This involves assessment of composite plate and CFRP/Al-honeycomb sandwich panel manufacturing and prototyping practice as well as barely visible damage introduction to testing panels. Initially, NDE is applied as ultrasound inspection to assure that CFRP face plates are produced with relatively small scatter for thickness distribution; this involves histogram production. Each panel underwent a self-frequency nature test and assessment by Modal Assurance Criterion robustly assessing the whole panel maturity/robustness before the mechanical testing. A panel assemblies composed of six test specimens were inspected before and just after introduction of barely visible damage. The indentation depth was measured using laser scanner comparing groove depth to the reference plate.

WP-4 Characterization of the mechanical behaviour of sandwich specimens

An extensive testing programme has been accomplished within the project, it's more than a thousand of different mechanical tests including coupon tests for mechanical characteristics, impact caused damage assessment tests and, finally, compression after impact tests. A dedicated and detailed test matrix for edge-wise compression after impact tests was developed. Corresponding tests were recorded for strain mapping by digital image correlation system. This assured both the load introduction path and failure verification in all physical tests. Majority of the test results are available in sections SIV and CAIV.

D 4.1 Results of the mechanical tests

The report summarizes initial efforts on material characterization by means of mechanical tests. All relevant mechanical test procedures identified within the literature review have been produced and initially verified with dedicated prototype specimens. This included designing and machining of dedicated testing equipment such as climbing drum peel test setup, compression after impact testing rig, and BVID set up in low velocity impact tower. Aluminium honeycomb structure properties were extracted and verified by analytical approach for estimating the load carrying capacity of produced panels. Furthermore, a dedicated compression after impact test setup with all edges restrained has bed designed and tested while simply supported edge support testing jig have been optimized and improved for further tests. Moreover, initial impact induced damage tests outline the need for further update and reduction of drop weight for our testing equipment and procedure, this has been considered and redesigned and will be implemented in further tests. Therefore, within initial reporting period approximately 200 specimens were tested and evaluated serving both aims: verification of numerical model and for update of production procedure.

D 4.2 Results of mechanical testing

The report summarizes mechanical test efforts for assessment of both coupon and specimen scale samples. All relevant mechanical test procedures initially identified within the literature review have been produced and verified with dedicated prototype specimens. This allowed designing as well redesigning for machining of dedicated testing equipment such as climbing drum peel test setup, compression after impact testing rig and BVID setup in low velocity impact tower. Among other characterisation of CFRP and various aluminium honeycomb structure properties enabled both numerical FE based and analytical procedures for estimating the load carrying capacity of produced panels.
Main emphasis of the deliverable is devoted on summarising compression after impact tests. Produced 50 panels and casted approximately 250 specimens for dedicated tests due time consuming procedure and additional test requirement on produced specimens. A summary of compression behaviour is given by statistically comparing similar configuration sandwich panels. Some pattern could be identified, nevertheless some outliers require statistically more mature revisiting of several configurations. Finally, visual illustration for each specimen tested was added to the report with front / back / both side views of compressed sandwich panels. It should be noted that indentation introduced mechanically is reported in NDE report as numerical analysis was done in automated means by developed software therefore introduced damage level is reported separately in project report D 3.2 and D 5.4.

WP-5 Elaboration of design guidelines and fast design tool

This work package is essential for summarizing and delivering both software tools and conclusions of the projects lessons learned. All developed numerical procedures based on finite element method were verified analytically and validated with experimental tests. Lessons learned include the mesh sensitivity, cohesive element applicability, material properties and their sensitivity analysis, failure and buckling non-linear solution settings. Moreover, as summary of best practices concerning experimental investigations and specimen preparations are covered within developed guidelines. The delivered software provides a full set of required utilities to crosslink and to transfer imperfection and damages configuration to the numerical simulation.
Delivered tools include:

Automated Surface Image Analysis (ASIA) toolbox (developed in MATLAB). The toolbox is intended for detection, analysis, and reporting of the extent of damage caused by the low-velocity drop weight impact event or quasi-static indentation on honeycomb sandwich panels. The task is performed using xyz point cloud digital data from a laser-based xyz scan of the specimen surface. The software uses automated algorithms to detect and measure geometry of the damaged area – outline, depth, diameters, area, and volume. It also returns surface plots and cross-section plots of the area.
Analytical Indentation Model (AIM) functions (developed in MATLAB) estimate dent depth, diameter, area, and volume using an analytical model of quasi-static indentation. The second version of AIM extended applicability to both quasi-static indentation and impact loading cases while also applying minimization of the potential energy when estimating the residual dent depth in order to improve the accuracy of the dent depth prediction.
Surface Inspection Viewer (SIV) tool is intended for reviewing of surface inspection results after low-velocity drop weight impact or quasi-static indentation as well as for side-by side comparison of results from FEM model and analytical calculations. This software tool enables convenient viewing of results from the ASIA toolbox and AIM functions. The tools is developed with JavaScript and integrated as a part of the project's website.
The Compression After Impact Viewer (CAIV) software tool is intended for viewing videos of Compression After Impact tests together with the results of the tests for panels with various layups and indentation or impact damage. The tools is developed with JavaScript and integrated as a part of the project's website.
ANSYS Model Configuration Tool (AMCT) is intended for ANSYS APDL user configurable Finite Element Model (FEM) log file generation for user defined parametrically variable analyses of Indentation (IND) / Compression after impact (CAI) simulation. Finite element model considers the substitution of the actual honeycomb structure with the solid volume. Hill plasticity was considered for core material properties. MaxStress failure criteria was considered for composite face sheets with progressive damage propagation mechanism. Cohesive zone model with contact separation was considered for adhesive layer material. Finite element model consists of two load step non-linear analyses. Load step 1 for indentation analysis and Load step 2 for CAI analysis. Interaction between indenter and indentation frame is realized by the use of surface-to-surface contact. The user is given a form with the according input fields to define the parameters of the simulation and as a result is given a ready-to-use ANSYS APDL FEM log file for analyses of Indentation / Compression after impact simulation. The tools is developed with JavaScript and integrated as a part of the project's website.

D 5.1 Interim simulation and testing guidelines

The report summarises initial efforts in development of numerical and analytical analysis procedures/guidelines that were to be validated with experimental tests. All numerical analysis was implemented by commercial finite element software code ANSYS by scripting APDL log files. The main emphasis was focused on barely visible damage (BVID) introduction in composite sandwich structures by indentation and estimation of residual dent depth and accumulated plastic deformations of honeycomb structure and composite damage. While numerical methods capture all variety of boundary conditions, layup and honeycomb orientation as well as geometrical and material nonlinearities the analytical approach assumes symmetry and equivalent load distribution around dent. Initial verification was performed in line with selected physical experiments. This includes reverse identification of reference panel performance assessment by self-frequency tests and modal analyses. At the initial stage, very good correlation among a dozen of initially tested specimens was achieved by assessing residual dent (barely visible damage). The final simulation and testing guidelines are available in deliverable D5.2.

D 5.2 Final simulation and testing guidelines

The report summarises a set of guidelines derived from extensive effort on setting up the numerical and analytical analysis procedures. All developed numerical, analytical and finite element method-based methods were verified analytically and validated with experimental tests. Moreover, a summary of best practices concerning experimental investigations and specimen preparations is included in the report. All numerical analysis were implemented by commercial finite element software code ANSYS by scripting APDL log files. The main emphasis was focused on barely visible damage (BVID) introduction by indentation and estimation of residual dent depth and accumulated damage in honeycomb and composite structure. While numerical methods capture all variety of boundary conditions, lay-up and honeycomb orientation as well as geometrical and material nonlinearities, the analytical approach assumes symmetry and equivalent load distribution around dent. Both analytical tool verification and finite element model (FEM) validation was performed in line with series of physical experiments. Finally we have achieved good correlation among a dozen of currently tested specimens by assessing residual indentation (barely visible damage BVID) and prediction of load carrying capacity of compression after impact (CAI) tests that are available in sections SIV and CAIV.

D 5.3 Virtual simulation software

This deliverable is a set of software developed within the project: Automated Surface Image Analysis (ASIA) toolbox for MATLAB, two versions of Analytical Indentation Model (AIM) for MATLAB, Modal Assurance Criteria (MAC) script for MATLAB as well as ColorThick.

D 5.4 Virtual simulation tool documentation

The report presents documentation for all developed software during the project. This includes Automated Surface Image Analysis (ASIA) toolbox for MATLAB, two versions of Analytical Indentation Model (AIM) for MATLAB, Surface Inspection Viewer (SIV), Compression After Impact Viewer (CAIV), Modal Assurance Criteria (MAC) for MATLAB, ColorThick well as ANSYS Model Configuration Tool (AMCT).

WP-6/7 Project management, dissemination and exploitation

Besides technical work, a considerable focus was devoted within final year to dissemination activities. Since the project was one of first funded research projects within ESA/Latvia cooperation agreement PECS, the project has a major requirement and general acceptance to become a “good practice story” for national public media. A great effort has been done in attracting media attention and securing public broadcasts on appropriate main national TV and radio shows. As a first step, a video teaser was filmed by RTU public relation team to attract public media attention showing what is done within project. For more information, see section Communication.