MANUFACTURING OPTIMIZATION OF A PLENUM WITH GFRP CYANATE ESTER-BASED PREPREG

Final Report Summary - PLENOPTIMUM (MANUFACTURING OPTIMIZATION OF A PLENUM WITH GFRP CYANATE ESTER-BASED PREPREG)

Executive Summary:
PLENOPTIMUM project dealt with the solution of a series of persistent manufacturing problems in the production of the plenum of the air cooling unit of aircrafts with a new glass/cyanate ester material system. Heterogeneities, pink coloration and porosity were the most important problems reported leading to an increasing disposal rate of manufactured plenums with all the associated costs (financial, environmental etc). A secondary objective was the feasibility of manufacturing the plenum in out-of-autoclave (OoA) conditions.
The PLENOPTIMUM consortium was formed by the Applied Mechanics Lab, University of Patras (UOP) and FUNDACIÓN TECNALIA RESEARCH AND INNOVATION (TECNALIA) both partners with a strong background on processing-characterization and manufacturing of cyanate ester based materials from their participation in various EC, ESA and national projects.
The team utilized a novel Design of Experiments approach in order to identify a number of possible influential factors which affect the manufacturing process (in autoclave and out-of-autoclave) and cause deviations and heterogeneities. Vacuum level, time to exhaust the vacuum and curing temperature were focused in the first level and an extensive test matrix was built in order to find the influential parameter(s). The quality assessment of the manufactured plates was conducted via fiber volume fraction (Vf), porosity, Tg and ILSS experimental measurements as well as color characterization. The results of the test matrix actually suggested that the manufacturing process is quite robust as far as the aforementioned factors is involved. In the meantime, some key observations from the topic manager coupled with knowledge acquired in the first stages of the project brought into light a factor missed in the first test matrix. The out-life and the storage conditions of the material. This factor was targeted in a second test matrix where the manufacturing trials and associated quality characterization were repeated for material left in room temperature controlled conditions. The results of the second test matrix were quite revealing offering solid experimental evidence that the storage of the material in room temperature up to 9 days actually solved the problem of non-repetitive heterogeneous manufacturing. Apparently, a major issue in cyanate-ester material systems is high resin flow due to low viscosity when a less than perfect vacuum bag seal is present. In parallel, a physical modeling of the curing cycle offered useful insights to the way viscosity changes during curing towards an optimized curing cycle. Regarding the autoclave process itself, since the factors addressed in the first test matrix proved non-influential, the importance of securing the vacuum in the bagging system via careful sealing was highlighted. Sound seal of the vacuum bag eliminates the need of reduced resin flow, and can ensure repeatable high quality results using a commercial glass/cyanate ester prepreg system.
In the last part of the project, efforts were directed towards the OoA process. The manufacturing trials early on showed that it would be very hard to acquire the desirable compaction and consequently a minimum Vf and low porosity with a material system designed and optimized for autoclave use. Trials in flat plates as well as generic elements with double curvature in an innovative self-heated mold cannot be characterized successful but gave the consortium the impression that with the proper material system it is quite feasible. Finally, manufacturing trials for the actual plenum part in an autoclave process took place taking into account the results of the executed test campaigns.
The main achievements of PLENOPTIMUM project concern the delineation of the factors that influence the manufacturing process of plenums resulting often in non-acceptable parts. The consortium came out with material as well as process related guidelines to the manufacturer and to the direction of securing the baseline autoclave process. These guidelines with a proper implementation can minimize the percentage of non-acceptable parts. A parallel outcome was the definition of an optimized curing cycle, a result that can lead to energy savings and together with the decrease of rejected plenums to a greening of the whole manufacturing process meeting some of the main goals of Clean Sky initiative. It was calculated based on scientific evidence throughout the project, that the ensuring a leakage free vacuum system can optimize the baseline manufacturing process so that only a 2% scrap is reached, then the cost of the plenum could be reduced 22%. Reducing the process time by means of adopting an optimized curing cycle could further decrease 3.85% the plenum costs. The total estimated possible manufacturing cost savings if these two actions are taken (ensuring a free leakage system and using the optimized cycle) is an impressive 25% of the actual cost.
The out-of-autoclave process proved a too ambitious goal at this stage. Extensive manufacturing trials did not produce an acceptable quality part but clearly suggest that this can be feasible on the condition that a specially designed material system for this process is designed. Apparently, a successful out-of-autoclave process could really have an important economic and environmental impact so it is a challenge that should be re-addressed in the near future.
Project Context and Objectives:
PLENOPTIMUM project dealt with the solution of a series of persistent manufacturing problems in the production of the plenum of the air cooling unit of aircrafts with a new glass/cyanate ester material system. Heterogeneities, pink coloration and porosity were the most important problems reported leading to an increasing disposal rate of manufactured plenums with all the associated costs (financial, environmental etc). A secondary objective was the feasibility of manufacturing the plenum in out-of-autoclave (OoA) conditions. The following list summarizes the objectives of the project:
The topic foresees the following activities:
• Identify the factors that lead to non-acceptable plenum parts with the baseline manufacturing process (autoclave)
• Establish deep knowledge of specific pre-preg behavior for autoclave process.
• Develop and optimize an out –of-autoclave (OoA) process for specific pre-preg.
• Provide a proof by producing a prototype plenum using autoclave or OoA process

The aforementioned objectives have been sought into 5 work packages according to their specific role within the project. WP2 covered all tasks required to set the engineering problem from current manufacturing and to deliver a specific roadmap for solving the problem. WP3 dealt with the implementation of designed roadmap via testing and material modelling to come up with conclusions about what and why goes wrong in current manufacturing process and provide specific suggestions for solving the reported problems. WP4 refered to all process developments and optimization tasks (experimental and simulation) based on recommendations of previous work-package. WP5 corresponded to the work planned for proofing the concept via prototype manufacturing and includes advanced tooling and curing control techniques. All management and dissemination activities took place in WP1.

Project Results:
The main scientific and technological results and developed foreground of knowledge are summarized below.

1. Procedure for the identification of influential factors on an autoclave manufacturing process with cyanate ester prepregs

The drive behind the whole PLENOPTIMUM project concept was to identify and eliminate the factors that led to defected (mainly high porosity, inhomogeneity and pink coloration) plenum components from cyanate este prepregs in an autoclave process. A significant goal at the beginning of the project was to develop a methodology which addresses all the possible factors which affect the manufacturing process and the behavior of the material, and assess their influence. This was succeeded via a specially designed test matrix and a number of manufacturing trials and targeted tests. The factors that were included in the first test matrix were related exclusively to the process. All factors were quantitative i.e. Vacuum level, Time to exhaust the vacuum, Curing temperature.
For each test, five quantities were measured with a goal to be analyzed statistically in order to define their influence on the quality and properties of the obtained products: colour, fiber volume fraction (Vf), porosity, glass transition temperature (Tg) and interlaminar shear strength (ILSS).
A statistical approach developed by the team based on the obtained results and analysis of variance suggested that the influence of the initially chosen set of parameters was rather minimal not allowing the reproduction of defects. Thus it was concluded that the defects reported by the plenum manufacturer should be related to other parameters missed in the 1st test matrix.

2. Curing cycle optimization process

A second significant outcome was obtained after targeted efforts to develop suitable curing and process models in order to optimize the process parameters involved in the baseline autoclave process. More specifically it involved the utilization of statistical experimental design of the results of the tests performed according to the previously defined test matrix. The aim was to estimate the influence of the processing variables on the different parameters of the manufacturing process of the plenum in order to improve it. Thermal (DSC) and viscosity tests were conducted in order to obtain the viscosity, Tg and degree of cure relationship with time and temperature. Three models/equations were applied for the simulation task:

1. Kamal-Sourour phenomenological model → in the case of the degree of cure
2. Dibenedetto equation→ in the results of the glass transition temperature, Tg
3. Arrhenius equation→ fit the results of the viscosity vs temperature.

After the aforementioned analysis was performed the recommended curing cycle was obtained. This curing cycle allows reaching an appropriate degree of cure for demoulding at the lower time/energy consumption.

3. Influential factors identification

After the inability of the first test matrix to identify the influential parameters, the design and realization of a new test matrix was necessary. To this direction the consortium resorted to some key past observations from the Topic Manager based on his experience with the material under study. During the early stages of the PLENOPTIMUM project, the TM reported that as the prepreg aged, the chances of a good quality part increased. Having a useful out – life of about 3 weeks, that meant that the prepreg had to spent more of its half-life in order to give rise to that increase in good quality probability. The pink coloration that has been systematically been observed and was assumed to originate from oxidation in accordance with the provider's documented observations. That assumption was critical, since it led to the decision to induce an artificial limited air flow through the vacuum bagging system during the curing of the laminate.
Taking into account the aforementioned observations, a decision was made, to carry out a new test matrix where the pursue of the low quality could continue, along with identifying the influential parameters and tempering with them. The prepreg out-life was proposed as the main parameter as it seemed to hold a significant role and was not controlled sufficiently.
Analyzing the second test matrix results, very important conclusions were drawn. It became apparent that the primary source of all reported problems by the TM was the fact that resin flow becomes more difficult as the material ages, and this acts as a barrier to the air channels that evolve into the laminate in the low viscosity state of the resin during curing. These air channels are the reason that oxidation and consequently pink coloration occurs, along with the porosity, which again is air presence (not resin starvation). Evidence of these air channels is that the red color is presented as longitudinal lines (in plate manufacturing), and they are parallel to the air flow direction (from the air leakage point to the vacuum port).
Another important observation is that Tg reduced versus time of out-life. This was an unexpected result, and it clearly indicates a chemical change in the material. Oxidation apparently occured in the prepreg material, which was evident from the reddish color that was enhanced as the days passed and the prepreg was left in atmospheric conditions.

4. Guidelines for the baseline autoclave manufacturing process optimization
The above observations gave rise to a hierarchical approach in trying to deal with the problem in an industrial level:

Process related
Ensure leakage free vacuum bagging system via:
a) Instrumented approach: vacuum proof test prior to curing
Advantages: High degree of confidence
Disadvantages: Time consuming, some initial investment cost

b) Proactive approach: better preparation of the mold to secure even better sealing and ensure a leakage free system

Advantages: Easy to follow, no new steps in production line
Disadvantages: Lower degree of confidence

Material related
a) Leave the fresh material out of the freezer inside its bag (not exposed in air) in controlled conditions for ~9 days.
Advantages: High degree of confidence
Disadvantages: FVF control problems, decrease of the useful life of the material

b) Change the material chemical's formulation with the provider towards reduced resin flow characteristics.

5. Out-of-Autoclave approach and manufacturing trials

The main goal during this task was to achieve an acceptable CPT (cured ply thickness) value in the absence of external pressure. The procedure attempted first utilized full vacuum and no overpressure. The composite presented high uniformity in terms of thickness, but also a high degree of porosity. This was expected, as the presence of vacuum without the aid of overpressure led to resin starvation (overbleed) during the low viscosity phase of the curing. A second trial involved changes in the used consumables and produced a highly irregular, in terms of thickness, plate. Even worse, the porosity was still quite evident. It was made quite clear that the differential pressure of vacuum (1 bar) was not fully applied. Further investigation was needed. The 3rd and 4th trials achieved better plates, but still with evident porosity. A no-bleed, aged prepreg material was decided to be used as a next trial. A high thickness plate was manufactured. The porosity levels dropped significantly, but not eliminated. Without having the OoA issues fully solved, University of Patras decided to proceed to a larger scale, multi thickness approach, in order to build up knowledge and push the process to the limits. A step wise plate of 4, 8, 12 and 16 layers was laminated. Trying to analyze the data and understand the mechanisms that control the behavior of the OoA procedure, some control points were identified:
✓ Good CPT values are achievable
✓ Resin bleed control failed completely using the glass tow system
✓ Differential pressure (1 bar vacuum) seemed not to be applicable – the compactness was very low
✓ Curing gases cannot escape, leading to irregularities and lost of pressure
A procedure that would compensate for all these, should:
✓ Provide full pressure
✓ Provide some level of resin bleed (controllable)
As seen before, these two were conflicting. A way of decoupling the pressure and the bleed (both having as source the vacuum applied) is essential. A two vacuum bag system with separate vacuum controls was decided to be used in the final trial. Furthermore, a self – heated mold was manufactured for that purpose, in order to both facilitate the process, and to optimize the energy consumption of the process. In that point, the curing monitoring using FBG sensors inside the laminate (embedded) was considered necessary in order to control the curing in the out of autoclave and out of conventional oven conditions. Finally, that mold had a complex shape (a curved c–section) in order to investigate the behavior of the material in the specific process in both flat and curved surfaces. The Out of Autoclave manufacturing proved to be very demanding, difficult process to realize with the given material. Poor quality composites were manufactured. Variation of process parameters failed to control adequately the behavior of the material. Obviously the material system was not designed or qualified for OoA manufacturing.
The team concluded after several manufacturing trials that the OoA process can be feasible with a material system specifically designed for OoA. Lower Glass reinforcement phase aerial weight would be beneficial, and higher viscosity resin system would improve the results without demand of very complex processes.

6. Autoclave manufacturing trials of Plenum

Based in all the obtained results during the project, the team came out with a set of realistic industrial recommendations towards the manufacturer for securing the baseline manufacturing of the plenum in the autoclave with the cyanate ester prepreg and minimize the reported problems.

Process related guidelines
Ensure leakage free vacuum bagging system via:
a) Instrumented approach: Vacuum leak check should be performed prior to curing. One strict test should not show more than 0,068 bar loss in 10 min. Other checking procedure less severe and much applied is not show more than 0,050 bar loss in 5 min. This second checking procedure is usually applied at TECNALIA for autoclave processing.
b) Proactive approach: better preparation of the mold to secure better sealing and ensure leakage free system.

Material related guidelines
a) Store the material sealed at RT in order to avoid moisture uptake and oxidation
b) Store the material in the freezer when possible, avoid storage at RT
c) Vacuum bag the plenum if lamination has to be stopped in order to protect the prepreg material.

Recommended process parameters
The recommended cure cycle as obtained by the simulation task.

During quality control, visual examination of the manufactured Plenum was carried out. The Plenum did not present any signs of pink coloration accompanied with porosity. Also, no abnormal thickness variations were evident. In general, the quality of the Plenum is assessed satisfactory based on detailed visual checking and thickness measurements. The Plenum was sent to LIEBHERR’s facilities for further Quality Control.

7. Techno-Economic Analysis

A last significant outcome of the project was an economic analysis on the impact of the suggested guidelines and the reported problems minimization to the production of plenum parts by the autoclave process. For this purpose, all technical and financial data from previous tasks have been collected in order to provide an economic analysis of the PLENUM prototype. This includes an assessment and quantification of recommendations in terms of process and material achievable through the optimization of manufacturing process.
The economic analysis is a tool that refers not only to the cost of materials and manufacturing but also includes the wider implications of optimized manufacturing process on the company performance of the TM Liebherr. While the cost of materials and manufacturing might be determined from direct contact with the supply chain and suppliers of raw materials and machines, the wider implications of the technology has been assessed along the project by the PLENOPTIMUM partners. Detailed economical evaluation of optimized manufacturing process was made by the team and concluded that, if ensuring a leakage free vacuum system can increase the quality so that only a 2% scrap is reached, the cost of the plenum could be reduced 22%. Reducing the process time by means of cure cycle optimization could further decrease 3.85% the plenum costs. The total estimated possible manufacturing cost savings if these two actions are taken (ensuring a free leakage system and using the optimized cycle) is an impressive 25% of the actual cost.

Potential Impact:
PLENOPTIMUM was an industrial project that had as a main objective the solution of a series of persistent manufacturing problems in the production of the plenum of the air cooling unit of aircrafts with a new glass/cyanate ester material system. Heterogeneities, pink coloration and porosity were the most important problems reported leading to an increasing disposal rate of manufactured plenums with all the associated costs (financial, environmental etc). A secondary objective was the feasibility of manufacturing the plenum in out-of-autoclave (OoA) conditions.
PLENOPTIMUM project addressed the topic (JTI-CS-2013-2-ECO-01-072) that belongs to the Area 01: EDA (Eco-Design for Airframe) of the Eco-Design Clean Sky Integrated Technology Demonstrator (ITD).
The proposed work for the specific topic is part of Clean Sky Work Package WP A.2.1.1 (Composites and surface treatments) that refers to the investigation on materials and surfaces (WP A2.1) in the framework of the development of new technologies (WP A2) fulfilling the requirements of the Eco-Design Approach for the airframe (WP A). The Airframe Application of the Eco-Design ITD focuses on the complete life cycle of an aircraft, from early design and raw materials to the phasing out of the aircraft. To this direction the results of PLENOPTIMUM project have a direct impact towards:
• Optimal use of raw materials, a decreased consumption of non-renewable
materials, natural resources and energy, and a lowered emission of noxious
effluents;
• Green manufacturing and optimized production processes (autoclave and out of autoclave)

It was calculated based on scientific evidence throughout the project, that the ensuring a leakage free vacuum system can optimize the baseline manufacturing process so that only a 2% scrap is reached, then the cost of the plenum could be reduced 22%. Reducing the process time by means of adopting an optimized curing cycle could further decrease 3.85% the plenum costs. The total estimated possible manufacturing cost savings if these two actions are taken (ensuring a free leakage system and using the optimized cycle) is an impressive 25% of the actual cost.

In a wider sense the results from PLENOPTIMUM will contribute also to the following objectives of Clean Sky:
• Accelerate the delivery of technologies for radically improving the environmental impact of air transport (by reducing the scrap/ reject rate of produced plenums).
• Increase the competitiveness of European industry, thus contributing to the Lisbon Strategy objectives.
• Encourage the rest of the aviation world to make greener products.

Taking into account the aforementioned framework, PLENOPTIMUM project provided CSJU optimized manufacturing guidelines for composite parts produced via Cyanate ester (CE) based pre-preg material. This will pave the way for wider use of Cyanate ester materials in aeronautical structures. Cyanate ester materials are considered as one of the best choices for replacement or aluminum alloys or epoxy based composites for parts used in aircraft and rotorcraft components subjected to elevated temperature and moisture atmosphere.
PLENOPTIMUM project results give CSJU the tools for resolving the technical challenge of substitution of metallic parts with parts made by Cyanate ester based composites for components such as the plenum. The deeper understanding of the processing characteristics of Cyanate ester based composites will help to optimize the autoclave manufacturing process while it provides valuable info towards the effort to establish an out of autoclave process.
Regarding the dissemination of the main project results, the project duration was too small for the academic partners to prepare and publish their scientific results. The most important dissemination activity was a poster dedicated to PLENOPTIMUM project in the stand of University of Patras in JEC exhibition - Paris 2015.
As far as the exploitation is concerned, the topic manager is determined to utilize directly the industrial guidelines that were the outcome of the project in order to reduce down to 2% the reject rate of the produced plenums and reduce the environmental CO2 footprint of the process through the reduction of the curing temperatures obtained in the optimized curing cycle. This last outcome will be readily adopted by the Topic Manager and it will be an intervention of positive financial and environmental impact. Further to this direction, an industrialization plan for the implementation and the exploitation of the most important results of the project was developed in the form of guidelines for the manufacturer of the plenum. These were two kind of guidelines, process and material related guidelines. The process related guidelines involve the better instrumentation of the vacuum bagging system and/or the better preparation of the sealing of the mold. Specific measures were discussed as to what measures can the manufacturer take to implement these proposals. The material related guidelines are simpler easy to adopt measures and concern the material handling and storage conditions.

List of Websites:
n/a

Final Report Summary - PLENOPTIMUM (MANUFACTURING OPTIMIZATION OF A PLENUM WITH GFRP CYANATE ESTER-BASED PREPREG)

Share this page

Download