M19-M24 Progress

Project progress, April 2017 (M24)

WP2 Sample preparation for test scenarios, sample characterization by reference methods and mechanical testing

All the samples were investigated via conventional NDT inspection by ADS using two different probes and two different settings respectively. Three different kinds of defects were observed:

1. Manufacturing defect with low impact on the use of sample for ENDT evaluation.

2. Deviation from reference without a clear defect signature (in case of MO and FP samples). Potentially due to contamination

3. “Contamination” induced defects such as disbonding or delamination (in case of FC and TD). They might have a detrimental effect on the future ENDT measurements. Conjointly with mechanical testing results, status of weak bonds can be confirmed for almost all the samples.

Finally, all the mechanical testing, namely mode-I, mode-II and centrifuge tests, performed by LTSM-UPAT, were completed and the experimental results reveal the negative effect of each contamination scenario in the fracture toughness and adhesive strength of the composite bonds.

Besides test coupons, a second geometry studied in this project is pilot sample which has a more complex geometry derived from real geometries present in the fields of application and with high relevance for aerospace applications. This means e.g. a curved (manufacturing) or a scarfed (repair) sample. The test scenarios with contaminations relevant for the respective field of application from manufacturing and repair comprise of samples with increasing geometric complexity and are presented in the figure below.

Approach of ComBoNDT illustrating the different types of test scenarios addressing the fields of application for manufacturing and in-service bonded repair

  Approach of ComBoNDT illustrating the different types of test scenarios addressing the fields
              of application for manufacturing and in-service bonded repair

The pilot samples were provided by industrial partners (ADS, ANN) and will be used to evaluate the efficiency and check the applicability of ENDT methods on more complex/curved geometries with multiple contaminations and also with a clean reference condition. With these samples, ENDT methods will be adapted in order to overcome limitations arising from measurements on non-flat surfaces. Mechanical tests on pilot samples were conducted, wherever possible, by LTSM-UPAT in order to evaluate the influence of the surface state on the mechanical properties of the scarfed pilot samples.

The manufacturing of the panels was performed considering the specifications and the surface quality requirements determined, and thus, ensure the achievement of future scheduled tasks for the following work packages. To verify that all pilot samples comply with the requirements established for the project, an ultrasonic inspection was realized over each one of them and the results not denoted any reportable attenuation, evidencing the high quality of the specimens manufactured. To summarize, the manufacture of the specimens within WP2 was performed successfully, reaching a high level of surface quality and guaranteeing the compliance of the quality standards imposed by the project for the achievement of the rest of scheduled activities.

Specifically, after cutting the plates to their final shape they were placed on the curing tool. The tool employed comprises a flat surface of 14 m2 approximately which accomplishes aeronautical regulation regarding smoothness material. The plates were distributed in different curing tools complying with specific requirements.

In order to allow a safe demoulding process and a smooth surface on the final plates, a water-based, silicon-containing release agent was applied to the tool. Then, the plates were distributed in order to minimize the number of tools and autoclave cycles needed to cure the whole set of plates, with a minimum separation between them to allow a proper compaction and curing. In order to ensure the uniform distribution of the pressure along the surface a ply of air breathers was employed to cover all the flat surface of the tool. Once the bag was finished and the vacuum was drawn, the curing tool was placed inside the autoclave and went through a curing cycle described below:

Diagram of the curing cycle

                                       Diagram of the curing cycle

After the cycle and once the mould was cold, the plates were removed from the surface and visually inspected before deburring to avoid cuts during the handling of the plates. Finally, all the plates were automatically inspected to ensure the quality of the parts manufactured. The inspection was carried out with a pulse echo equipment using water as coupling medium.

Representative results of the ultrasonic inspection of the specimens

                                 Representative results of the ultrasonic inspection of the specimens

The following materials and suppliers were used in the manufacture of the pilot samples

- CFRP Prepreg material supplied by Aernnova (ANN-4)

- Film adhesive supplied by: Airbus Operations

- Mould tool

- Water jet cutting contractor

Laminate Manufacturing Sequence

1) Clean and Prepare Mould Tool

Representative results of the ultrasonic inspection of the specimens

2) A circumferential 8 ply outer skin (shown in green below) was laid up and cured for the WP4 bonded samples, with peel ply on inner face.

Representative results of the ultrasonic inspection of the specimens

A second similar laminate was subsequently produced to provide the WP3 single skin samples.

Representative results of the ultrasonic inspection of the specimens

                                                     Manufacturing procedure

3) The skin was slid to the smaller diameter end of the mould tool and located on sacrificial strips. The laminate was then trimmed into rings approx. 120mm wide using a diamond oscillating saw.

Representative results of the ultrasonic inspection of the specimens

4) From each outer skin ring, the laminate was trimmed. Identification of original tool positions (1-5) was maintained.

Representative results of the ultrasonic inspection of the specimens

5) WP4 samples were shipped to IFAM at this stage for controlled contamination with the test substances.

6) While waiting for return of the prepared WP4 outer skin samples from IFAM, a circumferential 8 ply inner skin (shown in blue) was laid up and cured with peel ply on the outer face. This laminate was offset by 30mm from the outer skin tool position in order to match the radius of the outer skin after assembly as bonded samples.

Representative results of the ultrasonic inspection of the specimens

7) After return of the outer skin samples to ADS, the peel ply was removed from the inner skin. A layer of Cytec FM 300K film adhesive was laid up onto the skin and vacuum-consolidated in position. The outer skin samples were located to the inner skin with sections positioned to ensure matched radii. A vac bag was constructed around the assembled laminate stack, which was then cured.

Representative results of the ultrasonic inspection of the specimens

8) The inner skin (with bonded outer skin sections) was repositioned at the smaller diameter end of the mould tool and located on sacrificial strips. The laminate was then cut into rings 100 mm wide using a diamond oscillating saw.

Representative results of the ultrasonic inspection of the specimens

9) Each inner skin ring was trimmed to produce test coupons using a water jet cutting method.

Representative results of the ultrasonic inspection of the specimens

                               Bonded samples prior to final trimming

- CFRP for test coupon level was manufactured by ANN and delivered to IFAM. The CFRP plates were made according to the parameters specified within WP1. The samples have been cut using diamond dry cutting device.

- The scarfed samples have been contaminated as defined in WP1.

- Procedure for sample-manufacturing:

o First step: For all WP2, WP3 and WP4 samples in the project, the cutted CFRP samples were stored for 2h at elevated temperature in the oven.

o Second step: After thermal impact in the oven the samples have been scarfed by milling. After milling the scarfed surfaces have been manually grinded and cleaned in order to get of handling contamination and to remove residues of thermoplastic residues from milling.

Milling device

                                                               Milling device

Overview of scarfed sample for repair scenario

                                         Overview of scarfed sample for repair scenario

o Third step: The scarfed thermally treated (TD) samples are dip-coated using a 2 % De-icer-water mixture in the scarfed area. The resulted K concentration, measured by XPS has been reached.

- The samples have been delivered by ADS to IFAM. The curved samples have been contaminated as defined in WP1.

- Procedure for sample-manufacturing:

o First step: The release agent was applied to the surfaces by dip coating. First tests to yield the desired amounts of Si on the surface were conducted. Polymerization was allowed to take place with drying for 30 minutes at ambient conditions followed by a heat treatment for 60 min in the oven at 80°C.

o Second step: Samples for production scenario with fingerprint contamination were prepared using a standardized salty fingerprint solution (artificial hand perspiration solution) according to DIN ISO 9022-12. Samples were prepared by applying this solution with the size of a fingerprint to the samples. For WP3 the curved samples have been contaminated on the convex and concave side in the same way.

- All samples for WP2 and WP4 for repair scenarios were bonded using adhesive with the following curing cycle in the autoclave

Autoclave cycle for bonding samples with adhesive FM 300-2 from Cytec

                            Autoclave cycle for bonding samples with adhesive FM 300-2 from Cytec

The following figures show the preparation of samples for bonding in the autoclave at IFAM.

Preparation of samples for bonding in the autoclave

                                                             Preparation of samples for bonding in the autoclave, vacuum bag

For WP4, plates with a size of 12 cm x 12 cm were bonded with an overlap, scarfed area of about 6 cm x 12 cm. For WP2 the bonded samples have been cut into the desired sizes for tensile tests.

Overview of bonded pilot sample for repair scenario

                                    Overview of bonded pilot sample for repair scenario

In order to evaluate the influence of the surface state on the mechanical properties of the scarfed pilot samples, tensile mechanical tests were conducted by LTSM-UPAT. Tensile tests were conducted according to the ASTM D 3039 standard at room temperature using an MTS universal testing machine until the final failure (separation) of the two scarfed adherents. Four specimens were delivered for each sample set. The average dimensions of the specimens were 193 mm length, 25 mm width and 3.5 mm thickness. Aluminium metallic tabs (30 mm x 25 mm x 2mm) were adhered at the end of the specimens in order to avoid damage of the specimen at the grip points and to obtain a more uniform stress distribution to the specimen. The load applied to the specimen and the cross head displacement of the test machine were recorded continuously during the test.

Tensile test of scarf specimens

                                                Tensile test of scarf specimens

The results showed that the combined contamination with thermal degradation and de-icing fluid has a negative effect on the mechanical performance of the scarfed repair joints, reducing the failure load.

Additionally, the failure surfaces of the joints were examined after the tensile tests in order to characterize the failure modes and compare them with the tensile test results. The ASTM D5573 (Standard Practice for Classifying Failure Modes in Fiber-Reinforced-Plastic Joints) was followed. In the following photos of the fracture surfaces, showing the main failure modes observed in the tensile specimens are presented.

Representative fracture surfaces a) II-R-REF, b) II-R-TD1+DI1, c) II-R-TD1+DI2 specimens

           Representative fracture surfaces a) II-R-REF, b) II-R-TD1+DI1, c) II-R-TD1+DI2 specimens

The percentages of the different failure modes were compared for the different sample sets. In the reference samples, a mixed-mode failure was observed (Figure a above) with the dominant failure being the fiber-tear failure at a percentage of 63%. In Figure b (above) the fiber-tear failure remained the dominant one at an increasing percentage of 77% indicating that the combined contamination has a deleterious impact mainly on the behavior or CFRP adherents under tensile loading. Finally as shown in Figure c (above), the fiber-tear failure showed a slight reduction as it was observed at 71% of the surface area, while the adhesive failure was increased at 29%.

WP3 ENDT for surface quality assurance

WP3 focuses on the testing, maturation and improvement by adaptation regarding their sensitivity to detect physico-chemical properties of the adherend surfaces with multiple contaminations and the possibility to quantify the measuring results.

The samples characteristics requested for each ENDT technology were taken into account and test repair and production test scenarios were considered for the evaluation of test coupons, pilot samples with single and multiple contaminations according the peculiarities of the investigated scenario. For instance, for the production scenario multiple contaminations comprise release agents (RA) and finger print (FP). The ENDT tests performed: Laser Induced Breakdown Spectroscopy (LIBS), Optically Stimulated Electron Emission (OSEE), Full Field Vibrometry, Electronic Nose (E-NOSE), Aerosol-Wetting-Test BoNDTinspect (AWT), FTIR-Spectroscopy.

The results from the ENDT techniques investigated allowed evaluating which technique can better detect the types of contaminations and determine the optimum condition for the measurements. Thus, approaches for the improvements requested to adapt the techniques for realistic geometries and the automation of the ENDT are being undertaken by the partners based on the previous results.

For example, via AWT the differentiation of contaminations levels in the production scenario for moisture and release agents is clearly detected. Using OSEE and LIBS also under several contaminations conditions a very clear differentiation of contaminations levels was obtained.

WP3 focuses on the testing, maturation and improvement by adaptation regarding their sensitivity to detect physico-chemical properties of the adherend surfaces with multiple contaminations and the possibility to quantify the measuring results.

From M19 to M24 the work in WP3 was focused in tasks T3.2 Adaptation of ENDT techniques for surface analysis of realistic geometries and T3.3 Approaches to automation and industrialization of surface ENDT techniques. Mainly, the pilot samples were tested via the ENDT techniques and the partners are working in optimizing the techniques to investigate realistic parts as well as to turn the ENDT automated. Some of the ENDT as LIBS, OSEE and AWT in the frame of WP3 were developed to be handled directly by robot.

TOSEE system adapted to be used with robot

WP4 ENDT for adhesive bondline quality assurance

Results of ENDT Techniques

In the reporting period WP4 partners investigated flat coupons with single and mixed contaminations, curved bonded samples with mixed contaminations, and bonded scarf samples with mixed contaminations.>/p>

The μ-Computed Tomography (μ-CT) is used in this project to evaluate the bond quality by evaluating the degree of layer porosity at the bond line and gathering 3D information. This information is provided to the partners for comparison purposes.

μ-Computed Tomography System for CFRP

                                               μ-Computed Tomography System for CFRP

An example of the 3D data obtained is displayed in the figure below. The images are showing the stack of the three samples and a cutting plane from the top (blue) of the samples, from the side (red), and from the front (green) representing the three possible cuttings. In addition a 3D image is calculated indicating the position of the three cutting planes.

CT resulting images: cut from top (blue); cut from side (red); cut from front (green); 3D image indication the three cutting planes                                      

            CT resulting images: cut from top (blue); cut from side (red); cut from front (green);
                                      3D image indication the three cutting planes

It was possible to evaluate the layer porosity of the bond lines of the repair and production cases.

In the reporting period the non-linear local resonance and non-linear guided waves methods were being developed. The non-linear local resonance method uses commercial transducer and laser vibrometer. IT can be applied to natural surface of composite specimens without special surface preparation. In the case of non-linear guided waves method the generation technique was designed. A new transducer was designed to generate high amplitude non-linear effects. Works was also focused on improvement in data processing for dispersion curve measuring.

Laser scanning result for local resonance method

          Laser scanning result for local
                  resonance method

Non-linear guided wave method overview

                              >                                                                       Non-linear guided wave method overview

The non-linear local resonance method provides promising result to flat samples prepared with following modifications: thermal degradation, finger print contamination and moisture contamination.

The strain sensing principle of this technique is based on the special magnetic properties of the magnetostrictive ribbons. Initially, the inspection is performed when the test coupon is free and no load is applied to it. Then the test coupon is subjected to four point bending and inspected again. The data from the two states (loaded – unloaded) are then compared, by subtracting the measured values.

Sensing patch bonding to the CFRP samples

                                   Sensing patch bonding to the CFRP samples

It was observed that in some cases after the bonding of the sensing patch there are residual stresses on the samples which induce a residual strain field. The average value of the obtained data demonstrates that there is good repeatability of this method, especially on the repair test coupons for both the unloaded and the loaded states. In the cases of faulty cured and thermally degraded test coupons the etection was succesful. In the case multiple contaminations there is a good agreement between the measurements retrieved from the test coupons and the pilot samples.

EMI utilizes measurements of electrical parameters of piezoelectric transducer bonded to a host structure. Due to electromechanical coupling of the transducer and the structure, mechanical condition has its influence on electrical characteristics of the piezoelectric transducer. In the reporting period the research was focused on single and mixed contaminations as well as curved samples.

Schematic drawing of the set-up used for assessing of the adhesive bonds

                           Schematic drawing of the set-up used for assessing of the adhesive bonds

RMS values obtained with EMI method

                                          RMS values obtained with EMI method

A new signal processing approach was tested that compensate the influence of the individual characteristics of free transducers by subtracting the EMI spectra. The results of EMI were compared against GIC and GIIC values obtained in Work Package 2.

The next step if to analyse the EMI response obtained for bonded scarf samples.

The LASAT technology uses a laser (power density : 1-5 GW/cm2 and pulse duration in ns range) to create a tensile stress within a structure. Upon reaching the surface of the tested structure, the laser vaporizes a small part of the sample. Follows an interaction between the laser and the vaporized material, which leads in the creation of a high pressure plasma (GPa). It creates a shock wave inside the material.

The LASAT technique was supported by ultrasonic testing and photomicrographs to study the impacted samples.

The principle of LASAT technique

                                                                      The principle of LASAT technique

The bond threshold for the di-icer (DI) contaminated plates was decreasing with the increase in contamination level. It means that the technique could detect a loss of the mechanical properties of the bond. The same behaviour was noticed for the release agent (RA) and thermal degradation (TD) samples. The results for the finger print (FP) sample were unique. Indeed, no bond threshold could be found for the first two levels of contamination. Moreover, the composite threshold was much higher than for the other cases (almost twice as high). Further testing will have to be done before any conclusion can be made.

WP5 Demonstration of ENDT technologies in realistic environment

Objectives

- Provide a demonstration in real conditions of the technology maturity

- Assess the technique performances on real structures

The demonstration will take place in November – December 2017 on the IFAM premises (Bremen). In particular, the following activities will take place:

- Repair process (scarfing, surface treatment patching, bonding, etc.)

- Production process (secondary bonding of stringers above skin, autoclave, etc.)

- WP3 technique demonstration before bonding

- WP4 technique demonstration after bonding

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 636494.

This project is endorsed by the European Aeronautics Science Network - EASN.

Go to top