Final Report Summary - SATCAS (SIMULATION OF THE ASSEMBLY TOLERANCES FOR COMPOSITE AIRCRAFT STRUCTURES)
Executive Summary:
The project SATCAS (Simulation of the Assembly Tolerances for Composite Aircraft Structures) has been defined to support the design of the assembly process for the wings of the Low Drag Demonstrator within the Smart Fixed Wing Aircraft ITD (SFWA - Integrated Technology Demonstrator of the Clean Sky JU) through the development of a numerical simulation technique/strategy. The wings proposed in the mentioned program are being designed for operating under natural-laminar-flow (NLF) conditions, which is expected to lead to significant improvements in terms of energetic efficiency. One of the challenges that currently exist for the design of large aircraft structures for NLF regimes is the extremely tight requirements that must be achieved in terms of tolerances with the aim of ensuring that the flow will be sufficiently stable. In systems that are composed of many parts, the assembly processes acquire a special relevance.
The main objective of the project was to develop a methodology for the analysis of the deviations/deformations that might be produced during assembly processes of aircraft structures caused by: deviations in the constituent parts, deviations in the assembly jigs and/or deformations introduced directly by the assembly operations. This will allow checking the fulfillment or not of the tolerances specified for the ensembles and determining, for example, the influence of considering different assembly sequences.
The method proposed in SATCAS is based on a finite element method (FEM) strategy for the simulation of the whole assembly process on which all the parts are considered as flexible bodies. For this, the approach considered in the project covers from the detailed/local analysis of the fastening techniques (WP1 and WP2) to the simulation of each of the main operations expected to be used within the assembly processes object of study (WP3). In addition, the methodology developed also covers a proposal for the definition of the cases for the cumulative tolerances analysis and the development of an application/code to perform the treatment of the FE results (WP3).
In order to validate the FEM strategy proposed, it has been applied to a small assembly (lab scale) representative of the real ones, analyzing some of the basic operations both numerically and experimentally (WP4). Finally, the method has been applied to the analysis of two large assemblies (two wing boxes concepts, WP5).
As result, the methodology developed has demonstrated to respond satisfactorily to the objectives defined in the project. The FE numerical strategy proposed for the simulation of the assembly processes has shown its feasibility, adequacy and potential for this type of studies. Moreover, it has proven that it might be helpful to improve the processes (through recommendations for the fixation systems, the assembly/fastening sequences, etc) or even to support the design of the assembly jigs.
The project has been performed by ITAINNOVA under the supervision of Aernnova Engineering Division as Topic Manager. SATCAS is a sub-project associated to the activity of Assembly Simulation defined in the BLADE project – SFWA - Clean Sky programme, led by Airbus.
Project Context and Objectives:
Independently of the manufacturing processes and their sophistication, parts always present deviations and imperfections. The wear of the tools and moulds, misalignments in the fixation systems and jigs, errors that the operators may introduce, deformations caused by the own fabrication processes and many other factors; will inevitably generate defects and variations in the components. One of the challenges that currently exist for the design of large aircraft structures under NLF regime is the extremely tight tolerances that must be achieved with the aim of ensuring that the flow will be sufficiently stable. In general, the deviations in such systems can be generated: during the constituent parts manufacture, during the wings assembly, or in service. In systems constituted by many parts (as the Smart Fixed Wing Aircraft torsion boxes object of study in the project), the assembly processes acquire a special relevance, taking into account that depending on how they are executed, the tolerances of each part might accumulate and generate deviations much higher than those of the parts taken individually.
The conventional tolerances analysis techniques focus on the evaluation of the impact that geometric and/or dimensional deviations in the constituent parts may have in the dimensions of the resultant ensembles. For the aeronautic and automotive industries, the study of accumulative tolerances has gained a significant relevance in the last years since its influence in the manufacturing costs is generally important. These studies allow predicting deviations in assemblies, understand how they are produced, identify the main sources and, finally, obtaining recommendations to reduce the manufacturing non-conformities.
In most of the conventional tolerances analysis techniques the constituent parts are considered as rigid bodies. In many applications, this hypothesis is inadequate because:
- The assemblies are strongly affected by the part compliances (the hypothesis of parts as rigid bodies is not acceptable).
- The assemblies have such tolerance requirements that, despite the fact that the parts behave almost rigidly, the deformations generated during the assembly processes cannot be dismissed.
In those cases, issues come into play such as the interaction between parts as a consequence of the deviations generated by the tools or the assembly jigs, the distortions generated in the joints, or the spring-back effects in the parts. From a methodological point of view, the incorporation of the part compliances in the tolerance analysis of assemblies can be carried out through the integration of mechanical/structural calculation techniques (as the finite element model) with conventional approaches for tolerances analysis.
In this sense, the methodology presented in SATCAS is based on the Finite Element Method (FEM) as calculation tool to analyse the accumulation and propagation of deviations considering the parts compliance, necessary taking into account the tight tolerances that apply for the NLF Wings and the flexibility and size of these assemblies.
The methodology developed in the project covers the following main points:
- Detailed/local analysis of the fastening techniques
- Definition of the FEM strategy for the simulation of the whole assembly process (global) and the representation of the different fastening elements (bolts, temporary fasteners, etc).
- Definition of the approach for the cumulative analysis of tolerances (selection of cases to be analysed) and for the introduction of the deviations in the individual parts.
- Development of a code for the automatic treatment of results (in order to characterize the ‘waviness’ in the Upper Cover surface).
Taking into account the complexity of the processes used in aeronautics, their simulation becomes a major challenge. In this sense, it must be considered that the deformations in the final ensembles can be caused by the deviations in the constituent parts, deviations in the assembly jigs or by the assembly operations itself (for example deformations introduced by the temporary fasteners or the bolts). All these possible sources of deviations should be taken into account in the numerical simulations. The FEM strategy proposed in SATCAS covers all these aspects using different techniques to simulate the operations and/or steps that may have an effect in the final deviations of the resultant component.
The main objective of SATCAS is the development of a methodology for the analysis of assembly tolerances considering the propagation/accumulation of deviations coming from multiple sources and the parts compliance.
From the point of view of its application to a real assembly, the methodology has as objective to allow:
- identify key driving parameters of the assembly processes,
- analyse the possible influence of the joining techniques (type of bolts, nuts, washers...)
- determine how certain deviations in the components can affect to the final assembly tolerances,
- evaluate the possible influence of deviations coming from the jig setup,
- analyse different fastening strategies/sequences,
- support the design of the assembly jigs/tools.
The method has been applied to the analysis of a NLF Wing Boxes defined within the framework of the BLADE- SFWA project previously mentioned.
Project Results:
The main S & T results obtained within the SATCAS project are resumed below for each work-package:
WP1 – LOCAL ANALYSIS OF BOLTED/RIVETED JOINTS
WP1 has been defined with two specific objectives:
- Evaluate the local stress/strain fields that are generated near the bolted joints (considering that, accumulated, may produce global distortions).
- Analyse the joints stiffness, which may also affect the final geometry/deviations of the assemblies.
In order to achieve these objectives, the proposed approach for WP1 was based on:
- The execution of detailed numerical studies using the finite element method (FEM) as mechanical/structural analysis tool. These simulations cover the different joints configurations that might be used or might be present in the NLF wings assemblies.
- An experimental analysis carried out over a specific joint configuration (called the basic joint configuration). These tests have been used as reference for the FE models adjustment/definition.
The specific activities carried out within this work-package were:
1) Definition of the assembly processes and joints to be analysed. Definition of the basic joint configuration.
2) Execution of an experimental test program for the basic joint configuration.
3) Development of detailed FE models for the basic joint configuration.
4) Experimental vs. Numerical correlation of the previous results. FE models adjustment.
5) Development of FE models for the analysis of other joint configurations or certain variations of interest.
Some of the results obtained from all the analyses carried out can be summarized as follows:
-The response of the joints is affected by the utilization of sealant and shim, reducing notably their initial global stiffness (comparing with the same joint without sealant and shim).
-The bolts diameters affect the joint stiffness both under shear and peel loads if sealant is not used.
-When sealant is applied, the initial global stiffness is not significantly affected by factors as the applied torque, the bolt/hole clearances, losses of perpendicularity in the holes, the addition of washers or even by the utilization of different bolt diameters. These results have been observed either under peel or shear loads. In the range of low applied loads, these influences are hidden by the sealant.
-In terms of local deformations, the values of the applied torques and the bolts diameters have demonstrated to have a relevant impact in the strain fields that are produced. In this sense, these parameters should be taken into account in the in-plane strains that can be generated locally and that may be accumulated in a joint with multiple bolts.
- The global stiffness of the joints is strongly affected by the substrates thicknesses.
Then, the stiffness of the joints for various configurations of interest and the local stress/strain fields that are generated near the bolts/rivets have been successfully assessed in this work-package.
WP2 – SIMPLIFIED FEM TECHNIQUE FOR THE SIMULATION OF BOLTED JOINTS
Considering the high computing power that would be implicit in using the detailed FE models developed in WP1 for the study of assemblies with multiple joints, it becomes clear the need of having available simplified techniques for approximating the real behaviour of the joints with reasonable precision at an affordable computational cost. This constitutes precisely the objective of the work-package WP2 of SATCAS.
The activities performed within this work-package were:
1) Study of preliminary concepts for simplification. Selection of the best simulation technique.
2) Definition/adjustment of parameters for the FE technique selected.
3) Parameters definition/adjustment.
4) In-plane strains fields generated depending on the bolt diameter.
As main result of the activities performed for WP2, a numerical strategy for the simulation of bolted/riveted joints in large FE models has been defined. More specifically, the strategy covers de following aspects: meshes, connector responses, fastener interactions, shim and sealant properties and the way to introduce the in-plane strains fields that are generate locally in each bolted joint. Moreover, the adjustment/definition of the parameters associated to the models was carried out taking into account the results obtained in WP1.
The approach proposed has demonstrated to adequately reproduce the joints stiffness for the range of loads expected in assembly operations, as it is evidenced from the correlations obtained with the results of the detailed FE models for different substrates thicknesses. On the other hand, has been verified that the generation of local thermal expansions constitutes an effective way to introduce the in-plane strains produced by the bolting operations in large FE models.
WP3 – ANALYSIS OF ASSEMBLY TOLERANCES - METHODOLOGY
The third work-package proposed for the project was referred to the development of the methodology for analysing the tolerances of assemblies, which is planned as a combination of techniques of cumulative tolerances and the finite element method for the analysis of assembled deformable parts.
The activities associated to this work-package were:
1) Definition of a numerical strategy to simulate the assembly processes (techniques for the different assembly operations).
2) Introduction of deviations.
3) Approach selected to define the deviation scenarios to be analysed (definition of experiments/simulations).
4) Development of a computational tool for the results treatment.
5) Evaluation of the computational costs associated to the assemblies simulations.
From the technical point of view, these are the main results obtained from the work performed in WP3:
- As part of the methodology developed, different numerical strategies for the simulation of the assembly processes have been established:
- General definition of the FE models in terms of type of elements, meshes, contacts, material properties, and boundary conditions.
- Application of loads as ‘connector motions’ for the connectors between the parts and the tools, representing these fixations, the connectors representing the temporary fasteners, and the connectors representing the bolts.
- Mesh requirements.
- Definition of contacts.
- Activation and deactivation of parts and contacts.
- Strategy for the simulation of the assembly jigs and the fixation of the parts.
- Strategy for the simulation of the initial fixation through temporary fasteners.
- Strategy for the simulation of shim and sealant addition.
- Strategy for the simulation of the bolting step.
- A strategy for the introduction of deviations has been defined.
- A code for the automatic treatment of the FE results in terms of ‘waviness’ has been developed.
- An approach to define the deviation scenarios to be analysed (cases for the cumulative tolerance analysis) has been proposed.
- Preliminary evaluations in terms of computational costs have been performed.
The operation of the methodology applied to real assemblies has been evaluated in work-packages WP4 and WP5.
WP4 – EXPERIMENTAL VALIDATION OF THE METHODOLOGY
In order to validate the methodology for the simulation of assembly processes, the analysis of a real demonstrator assembly (laboratory level) has been proposed in WP4.
The specific objectives defined for this work-package were mainly the following:
- Design of the validation assembly and the tests to be performed.
- Design and construction of the tool to be used for validation.
- Definition/adjustment of some of the parameters associated to the numerical technique selected for the simple representation of the bolted joints in the FE simulations of the assembly processes (those not defined in WP2).
- Execution of the experimental and numerical analyses for the validation work-package.
- Validation of the methodology. Numerical vs. experimental correlation.
The specific tasks performed were:
1) Selection of the part to be used for the validation activities.
2) Definition/design of the validation tool.
3) Fabrication of the validation tool.
4) Experimental analysis of the tolerances of a real assembly at laboratory scale.
5) Numerical analyses with the validation assembly and correlation.
Summarizing, considering the work performed for WP4, the following results have been obtained:
- The validation tool has been designed and fabricated.
- The experimental tests and the simulations planned for the validation phase (WP4) have been completed.
- The FE model required for the validation phase has been prepared.
- The simulations planned for the validation phase has been performed.
Finally, the analysis of the experimental and the numerical results (correlation) have evidenced that:
- The FE simulations reproduce in all cases the general shape of the deformations generated in the Upper Cover demonstrator surface for the different steps and deviations scenarios analyzed.
- Considering the absolute values of the deformations, different correlation levels have been observed. Although in many cases the simulations have reproduced properly what happen during the assembly operations, in some cases significant divergences have been obtained. However, it has been checked that these differences were produced mainly by unwanted effects in the experimental tests.
- Part of the tests performed allowed checking that the shape of the deformation initially generated by the initial fixation through temporary fasteners is maintained during the subsequent assembly steps. This has been observed in both the experimental and the virtual (FE) tests.
- On the other hand, it has been checked that the simulation strategy proposed allows also to adequately reproduce the resistance to the ‘squeeze’ that the shim and the sealant may offer during some of the assembly operations, affecting to the clearances that are finally obtained between the parts.
- Finally, it has been verified that the numerical strategy proposed is sensible to the clamping sequences defined for the assembly operations.
WP5 – EXPERIMENTAL VALIDATION OF THE METHODOLOGY
The main objective of this work-package was to apply the methodology developed in previous phases of the project for the analysis of the assembly of torsion wing boxes.
The specific objectives of WP5 can be summarised as follows:
- Analyse deviations that might be produced by the assembly jigs.
- Study the effect that different deviations in the constituent parts may have in the final assembly.
- Analyse alternative fastening sequences.
- Obtaining recommendations for the assembly.
The activities performed are listed below.
1) Preparation of geometries for both RH (Right Hand) and LH (Left Hand) Wings.
2) Preparation of MEF models for the assembly simulations (RH and LH Wings).
3) Carrying out the simulations related with tooling effects.
4) Carrying out the simulations considering different deviations scenarios for the constituent parts.
5) Carrying out the simulations with alternative fastening sequences.
The technical results obtained in this work-package (mainly for the RH Wing) are:
- The waviness tolerances have been checked for all the deviation cases studied, analysing the results obtained from the simulations against the tolerances defined in the PKC (Product Key Characteristics) of the wings. The most critical values have been identified always in chord-wise direction.
- The profile tolerances have been also analysed for each case of study.
- Regarding the deformations that might be introduced by the vacuum vents that support the Upper Cover and the Leading Edge, it has been concluded that the control of the vacuum level in these systems (and consequently, the forces that might be introduced by them) should be adjusted depending on the state of the Upper Cover. In addition, a change in the position of some of the vents has been recommended.
- The analysis of the waviness have evidenced that the deviations introduced in the Front Spar, the Rear Spar and in the Lower Covers, for the magnitudes considered, have a minor effect in the deviations that are produced in the surface of interest.
- Regarding the analysis of the factors of influence, the results evidence that, among the deviations analysed, the deviations considered for the Upper Covers are the most critical ones.
- Finally, regarding the analysis of other fastening sequences, it has been observed that the alternative sequence analysed could reduced the waviness generated in the surface of interest.
GENERAL CONCLUSIONS AND LESSONS LEARNT
- A numerical methodology for the study of aeronautic assembly processes in terms of deviations and tolerances has been developed. The methodology has been applied to the assembly of aircraft torsion wing boxes.
- The methodology developed has demonstrated to be useful for the analysis of deviations that may arise during assembly processes caused by deviations in the constituent parts, deviations in the assembly jigs and/or effects associated to the assembly operations itself. Moreover, it has proven that it might be helpful to improve the processes (through recommendations for the fixation systems, the assembly/fastening sequences, etc) or even the design of the assembly jigs (for example, distribution of vacuum vents and cradles). Nevertheless, from the point of view of the prediction of the deviations (shape and value) and the virtual verification of the tolerances, a higher effort it is required to characterise adequately each of the fixation and positioning systems used during the assemblies in order to improve the accuracy of the results and the confidence of the method.
- The simulation strategies proposed have demonstrated to be feasible to simulate assembly processes as complex as the ones used in aeronautics. The FE techniques proposed for the simulation of these processes has demonstrated its feasibility, adequacy and potential for this type of studies. However, further improvements must be done in order to reduce the computation time required for each simulation. This will allow performing studies with a higher number of casuistries and deeper statistical analysis. In this sense, an improvement in the definition of the deviations of the individual parts will be also needed.
- The extremely high complexity of the FE models required to simulate aeronautical assembly processes need the development of specific tools to control the models definition and checking some control parameters during the calculations, in order to avoid errors that might be difficult to be detected a priori in large FE models and that could invalidate calculations that last several days.
- The methodology and the numerical strategies developed are broad in their possible uses, being of application also for other type of components/assemblies or even for other type of analysis.
Potential Impact:
The expected impact of the project may be resumed in the following points:
- The design of natural-laminar-flow (NLF) aircraft wings requires extremely high precision tolerances that must be controlled over the whole manufacture/construction processes. Controlling the assembly tolerances represents a challenge of great relevance, taking into account the complexity associated if all the possible sources of deviations need to be considered. In this sense, the methodology developed in this project offers clearly significant contributions, allowing to predict possible non-conformities and to analyse corrective or improvement measures.
- From the point of view of the design and development of natural-laminar-flow wings for commercial aircrafts, the method could be useful to analyse new assembly strategies or improving the techniques used nowadays, avoiding the large amount of demonstrators and tests that are usually needed.
- The methodology and the numerical strategies developed are broad in their possible uses, being of application also for other type of components/assemblies. For example, they could be applied to other aircraft structures in standard programmes, focused in the optimization of the tolerances and processes.
- The simplified techniques defined for the representation of different type of fasteners and joints in MEF analysis might be also used in other studies different from the defined for the present project. As an example, the techniques might be also applied in common structural studies or for the analysis of part tolerances under service conditions. Moreover, these techniques can be also applicable in other sectors (automotive, railway, etc).
- Additionally, the project has demonstrated that the numerical techniques used can be helpful also to support the design of the assembly tools and jigs (for example, the fixation systems).
- Finally, the project has covered several activities that are inevitably required to address the problem of dimensional and geometric tolerances of aircraft assemblies designed to operate under natural laminar flow (NLF) regime, considered a key technology to reduce drag and, thus, to significantly improve the airplanes performance and efficiency. In this sense, the project is an important step for the possible future implementation of this technology, representing a significant contribution to the European competitiveness in aeronautics.
Regarding the diffusion/dissemination of the SATCAS project, the activities already performed and the ones planned for the next months cover various presentations and publications in different events and media. They are listed below:
1) ITAINNOVA WEBPAGE - Entry in the web of the Technological Institute of Aragon.
2) MATERPLAT Annual Meeting - Poster presented at a meeting of the Materials Spanish platform MATERPLAT (Oct – 2012).
3) AERONAUTICAL FAIR – SEVILLE (Aerospace and Defence Meetings Seville – June 2014)
4) Presentations to BLADE partners - Presentations that were given to some of the partners of the BLADE project in the Quarterly Review Meetings.
5) Presentations to ITAINNOVA’s customers - Presentations that were given to some costumers of ITAINNOVA.
6) Newsletter Clean-Sky - September 2014 Newsletter – eNews – Clean Sky Webpage.
7) International Workshop on Aircraft System (Hamburg, February 2015). Article already accepted.
8) 2015 SIMULIA Community Conference (Berlin, May 2015). Abstract initially accepted.
9) Publication of part of the work developed in a peer-review journal (not defined yet, to be done in 2015).
The activities have been performed in all cases with the permission of the Topic Manager, the BLADE project coordinator and the Project Officer.
List of Websites:
Website address: not applicable.
contact: achiminelli@itainnova.es
The project SATCAS (Simulation of the Assembly Tolerances for Composite Aircraft Structures) has been defined to support the design of the assembly process for the wings of the Low Drag Demonstrator within the Smart Fixed Wing Aircraft ITD (SFWA - Integrated Technology Demonstrator of the Clean Sky JU) through the development of a numerical simulation technique/strategy. The wings proposed in the mentioned program are being designed for operating under natural-laminar-flow (NLF) conditions, which is expected to lead to significant improvements in terms of energetic efficiency. One of the challenges that currently exist for the design of large aircraft structures for NLF regimes is the extremely tight requirements that must be achieved in terms of tolerances with the aim of ensuring that the flow will be sufficiently stable. In systems that are composed of many parts, the assembly processes acquire a special relevance.
The main objective of the project was to develop a methodology for the analysis of the deviations/deformations that might be produced during assembly processes of aircraft structures caused by: deviations in the constituent parts, deviations in the assembly jigs and/or deformations introduced directly by the assembly operations. This will allow checking the fulfillment or not of the tolerances specified for the ensembles and determining, for example, the influence of considering different assembly sequences.
The method proposed in SATCAS is based on a finite element method (FEM) strategy for the simulation of the whole assembly process on which all the parts are considered as flexible bodies. For this, the approach considered in the project covers from the detailed/local analysis of the fastening techniques (WP1 and WP2) to the simulation of each of the main operations expected to be used within the assembly processes object of study (WP3). In addition, the methodology developed also covers a proposal for the definition of the cases for the cumulative tolerances analysis and the development of an application/code to perform the treatment of the FE results (WP3).
In order to validate the FEM strategy proposed, it has been applied to a small assembly (lab scale) representative of the real ones, analyzing some of the basic operations both numerically and experimentally (WP4). Finally, the method has been applied to the analysis of two large assemblies (two wing boxes concepts, WP5).
As result, the methodology developed has demonstrated to respond satisfactorily to the objectives defined in the project. The FE numerical strategy proposed for the simulation of the assembly processes has shown its feasibility, adequacy and potential for this type of studies. Moreover, it has proven that it might be helpful to improve the processes (through recommendations for the fixation systems, the assembly/fastening sequences, etc) or even to support the design of the assembly jigs.
The project has been performed by ITAINNOVA under the supervision of Aernnova Engineering Division as Topic Manager. SATCAS is a sub-project associated to the activity of Assembly Simulation defined in the BLADE project – SFWA - Clean Sky programme, led by Airbus.
Project Context and Objectives:
Independently of the manufacturing processes and their sophistication, parts always present deviations and imperfections. The wear of the tools and moulds, misalignments in the fixation systems and jigs, errors that the operators may introduce, deformations caused by the own fabrication processes and many other factors; will inevitably generate defects and variations in the components. One of the challenges that currently exist for the design of large aircraft structures under NLF regime is the extremely tight tolerances that must be achieved with the aim of ensuring that the flow will be sufficiently stable. In general, the deviations in such systems can be generated: during the constituent parts manufacture, during the wings assembly, or in service. In systems constituted by many parts (as the Smart Fixed Wing Aircraft torsion boxes object of study in the project), the assembly processes acquire a special relevance, taking into account that depending on how they are executed, the tolerances of each part might accumulate and generate deviations much higher than those of the parts taken individually.
The conventional tolerances analysis techniques focus on the evaluation of the impact that geometric and/or dimensional deviations in the constituent parts may have in the dimensions of the resultant ensembles. For the aeronautic and automotive industries, the study of accumulative tolerances has gained a significant relevance in the last years since its influence in the manufacturing costs is generally important. These studies allow predicting deviations in assemblies, understand how they are produced, identify the main sources and, finally, obtaining recommendations to reduce the manufacturing non-conformities.
In most of the conventional tolerances analysis techniques the constituent parts are considered as rigid bodies. In many applications, this hypothesis is inadequate because:
- The assemblies are strongly affected by the part compliances (the hypothesis of parts as rigid bodies is not acceptable).
- The assemblies have such tolerance requirements that, despite the fact that the parts behave almost rigidly, the deformations generated during the assembly processes cannot be dismissed.
In those cases, issues come into play such as the interaction between parts as a consequence of the deviations generated by the tools or the assembly jigs, the distortions generated in the joints, or the spring-back effects in the parts. From a methodological point of view, the incorporation of the part compliances in the tolerance analysis of assemblies can be carried out through the integration of mechanical/structural calculation techniques (as the finite element model) with conventional approaches for tolerances analysis.
In this sense, the methodology presented in SATCAS is based on the Finite Element Method (FEM) as calculation tool to analyse the accumulation and propagation of deviations considering the parts compliance, necessary taking into account the tight tolerances that apply for the NLF Wings and the flexibility and size of these assemblies.
The methodology developed in the project covers the following main points:
- Detailed/local analysis of the fastening techniques
- Definition of the FEM strategy for the simulation of the whole assembly process (global) and the representation of the different fastening elements (bolts, temporary fasteners, etc).
- Definition of the approach for the cumulative analysis of tolerances (selection of cases to be analysed) and for the introduction of the deviations in the individual parts.
- Development of a code for the automatic treatment of results (in order to characterize the ‘waviness’ in the Upper Cover surface).
Taking into account the complexity of the processes used in aeronautics, their simulation becomes a major challenge. In this sense, it must be considered that the deformations in the final ensembles can be caused by the deviations in the constituent parts, deviations in the assembly jigs or by the assembly operations itself (for example deformations introduced by the temporary fasteners or the bolts). All these possible sources of deviations should be taken into account in the numerical simulations. The FEM strategy proposed in SATCAS covers all these aspects using different techniques to simulate the operations and/or steps that may have an effect in the final deviations of the resultant component.
The main objective of SATCAS is the development of a methodology for the analysis of assembly tolerances considering the propagation/accumulation of deviations coming from multiple sources and the parts compliance.
From the point of view of its application to a real assembly, the methodology has as objective to allow:
- identify key driving parameters of the assembly processes,
- analyse the possible influence of the joining techniques (type of bolts, nuts, washers...)
- determine how certain deviations in the components can affect to the final assembly tolerances,
- evaluate the possible influence of deviations coming from the jig setup,
- analyse different fastening strategies/sequences,
- support the design of the assembly jigs/tools.
The method has been applied to the analysis of a NLF Wing Boxes defined within the framework of the BLADE- SFWA project previously mentioned.
Project Results:
The main S & T results obtained within the SATCAS project are resumed below for each work-package:
WP1 – LOCAL ANALYSIS OF BOLTED/RIVETED JOINTS
WP1 has been defined with two specific objectives:
- Evaluate the local stress/strain fields that are generated near the bolted joints (considering that, accumulated, may produce global distortions).
- Analyse the joints stiffness, which may also affect the final geometry/deviations of the assemblies.
In order to achieve these objectives, the proposed approach for WP1 was based on:
- The execution of detailed numerical studies using the finite element method (FEM) as mechanical/structural analysis tool. These simulations cover the different joints configurations that might be used or might be present in the NLF wings assemblies.
- An experimental analysis carried out over a specific joint configuration (called the basic joint configuration). These tests have been used as reference for the FE models adjustment/definition.
The specific activities carried out within this work-package were:
1) Definition of the assembly processes and joints to be analysed. Definition of the basic joint configuration.
2) Execution of an experimental test program for the basic joint configuration.
3) Development of detailed FE models for the basic joint configuration.
4) Experimental vs. Numerical correlation of the previous results. FE models adjustment.
5) Development of FE models for the analysis of other joint configurations or certain variations of interest.
Some of the results obtained from all the analyses carried out can be summarized as follows:
-The response of the joints is affected by the utilization of sealant and shim, reducing notably their initial global stiffness (comparing with the same joint without sealant and shim).
-The bolts diameters affect the joint stiffness both under shear and peel loads if sealant is not used.
-When sealant is applied, the initial global stiffness is not significantly affected by factors as the applied torque, the bolt/hole clearances, losses of perpendicularity in the holes, the addition of washers or even by the utilization of different bolt diameters. These results have been observed either under peel or shear loads. In the range of low applied loads, these influences are hidden by the sealant.
-In terms of local deformations, the values of the applied torques and the bolts diameters have demonstrated to have a relevant impact in the strain fields that are produced. In this sense, these parameters should be taken into account in the in-plane strains that can be generated locally and that may be accumulated in a joint with multiple bolts.
- The global stiffness of the joints is strongly affected by the substrates thicknesses.
Then, the stiffness of the joints for various configurations of interest and the local stress/strain fields that are generated near the bolts/rivets have been successfully assessed in this work-package.
WP2 – SIMPLIFIED FEM TECHNIQUE FOR THE SIMULATION OF BOLTED JOINTS
Considering the high computing power that would be implicit in using the detailed FE models developed in WP1 for the study of assemblies with multiple joints, it becomes clear the need of having available simplified techniques for approximating the real behaviour of the joints with reasonable precision at an affordable computational cost. This constitutes precisely the objective of the work-package WP2 of SATCAS.
The activities performed within this work-package were:
1) Study of preliminary concepts for simplification. Selection of the best simulation technique.
2) Definition/adjustment of parameters for the FE technique selected.
3) Parameters definition/adjustment.
4) In-plane strains fields generated depending on the bolt diameter.
As main result of the activities performed for WP2, a numerical strategy for the simulation of bolted/riveted joints in large FE models has been defined. More specifically, the strategy covers de following aspects: meshes, connector responses, fastener interactions, shim and sealant properties and the way to introduce the in-plane strains fields that are generate locally in each bolted joint. Moreover, the adjustment/definition of the parameters associated to the models was carried out taking into account the results obtained in WP1.
The approach proposed has demonstrated to adequately reproduce the joints stiffness for the range of loads expected in assembly operations, as it is evidenced from the correlations obtained with the results of the detailed FE models for different substrates thicknesses. On the other hand, has been verified that the generation of local thermal expansions constitutes an effective way to introduce the in-plane strains produced by the bolting operations in large FE models.
WP3 – ANALYSIS OF ASSEMBLY TOLERANCES - METHODOLOGY
The third work-package proposed for the project was referred to the development of the methodology for analysing the tolerances of assemblies, which is planned as a combination of techniques of cumulative tolerances and the finite element method for the analysis of assembled deformable parts.
The activities associated to this work-package were:
1) Definition of a numerical strategy to simulate the assembly processes (techniques for the different assembly operations).
2) Introduction of deviations.
3) Approach selected to define the deviation scenarios to be analysed (definition of experiments/simulations).
4) Development of a computational tool for the results treatment.
5) Evaluation of the computational costs associated to the assemblies simulations.
From the technical point of view, these are the main results obtained from the work performed in WP3:
- As part of the methodology developed, different numerical strategies for the simulation of the assembly processes have been established:
- General definition of the FE models in terms of type of elements, meshes, contacts, material properties, and boundary conditions.
- Application of loads as ‘connector motions’ for the connectors between the parts and the tools, representing these fixations, the connectors representing the temporary fasteners, and the connectors representing the bolts.
- Mesh requirements.
- Definition of contacts.
- Activation and deactivation of parts and contacts.
- Strategy for the simulation of the assembly jigs and the fixation of the parts.
- Strategy for the simulation of the initial fixation through temporary fasteners.
- Strategy for the simulation of shim and sealant addition.
- Strategy for the simulation of the bolting step.
- A strategy for the introduction of deviations has been defined.
- A code for the automatic treatment of the FE results in terms of ‘waviness’ has been developed.
- An approach to define the deviation scenarios to be analysed (cases for the cumulative tolerance analysis) has been proposed.
- Preliminary evaluations in terms of computational costs have been performed.
The operation of the methodology applied to real assemblies has been evaluated in work-packages WP4 and WP5.
WP4 – EXPERIMENTAL VALIDATION OF THE METHODOLOGY
In order to validate the methodology for the simulation of assembly processes, the analysis of a real demonstrator assembly (laboratory level) has been proposed in WP4.
The specific objectives defined for this work-package were mainly the following:
- Design of the validation assembly and the tests to be performed.
- Design and construction of the tool to be used for validation.
- Definition/adjustment of some of the parameters associated to the numerical technique selected for the simple representation of the bolted joints in the FE simulations of the assembly processes (those not defined in WP2).
- Execution of the experimental and numerical analyses for the validation work-package.
- Validation of the methodology. Numerical vs. experimental correlation.
The specific tasks performed were:
1) Selection of the part to be used for the validation activities.
2) Definition/design of the validation tool.
3) Fabrication of the validation tool.
4) Experimental analysis of the tolerances of a real assembly at laboratory scale.
5) Numerical analyses with the validation assembly and correlation.
Summarizing, considering the work performed for WP4, the following results have been obtained:
- The validation tool has been designed and fabricated.
- The experimental tests and the simulations planned for the validation phase (WP4) have been completed.
- The FE model required for the validation phase has been prepared.
- The simulations planned for the validation phase has been performed.
Finally, the analysis of the experimental and the numerical results (correlation) have evidenced that:
- The FE simulations reproduce in all cases the general shape of the deformations generated in the Upper Cover demonstrator surface for the different steps and deviations scenarios analyzed.
- Considering the absolute values of the deformations, different correlation levels have been observed. Although in many cases the simulations have reproduced properly what happen during the assembly operations, in some cases significant divergences have been obtained. However, it has been checked that these differences were produced mainly by unwanted effects in the experimental tests.
- Part of the tests performed allowed checking that the shape of the deformation initially generated by the initial fixation through temporary fasteners is maintained during the subsequent assembly steps. This has been observed in both the experimental and the virtual (FE) tests.
- On the other hand, it has been checked that the simulation strategy proposed allows also to adequately reproduce the resistance to the ‘squeeze’ that the shim and the sealant may offer during some of the assembly operations, affecting to the clearances that are finally obtained between the parts.
- Finally, it has been verified that the numerical strategy proposed is sensible to the clamping sequences defined for the assembly operations.
WP5 – EXPERIMENTAL VALIDATION OF THE METHODOLOGY
The main objective of this work-package was to apply the methodology developed in previous phases of the project for the analysis of the assembly of torsion wing boxes.
The specific objectives of WP5 can be summarised as follows:
- Analyse deviations that might be produced by the assembly jigs.
- Study the effect that different deviations in the constituent parts may have in the final assembly.
- Analyse alternative fastening sequences.
- Obtaining recommendations for the assembly.
The activities performed are listed below.
1) Preparation of geometries for both RH (Right Hand) and LH (Left Hand) Wings.
2) Preparation of MEF models for the assembly simulations (RH and LH Wings).
3) Carrying out the simulations related with tooling effects.
4) Carrying out the simulations considering different deviations scenarios for the constituent parts.
5) Carrying out the simulations with alternative fastening sequences.
The technical results obtained in this work-package (mainly for the RH Wing) are:
- The waviness tolerances have been checked for all the deviation cases studied, analysing the results obtained from the simulations against the tolerances defined in the PKC (Product Key Characteristics) of the wings. The most critical values have been identified always in chord-wise direction.
- The profile tolerances have been also analysed for each case of study.
- Regarding the deformations that might be introduced by the vacuum vents that support the Upper Cover and the Leading Edge, it has been concluded that the control of the vacuum level in these systems (and consequently, the forces that might be introduced by them) should be adjusted depending on the state of the Upper Cover. In addition, a change in the position of some of the vents has been recommended.
- The analysis of the waviness have evidenced that the deviations introduced in the Front Spar, the Rear Spar and in the Lower Covers, for the magnitudes considered, have a minor effect in the deviations that are produced in the surface of interest.
- Regarding the analysis of the factors of influence, the results evidence that, among the deviations analysed, the deviations considered for the Upper Covers are the most critical ones.
- Finally, regarding the analysis of other fastening sequences, it has been observed that the alternative sequence analysed could reduced the waviness generated in the surface of interest.
GENERAL CONCLUSIONS AND LESSONS LEARNT
- A numerical methodology for the study of aeronautic assembly processes in terms of deviations and tolerances has been developed. The methodology has been applied to the assembly of aircraft torsion wing boxes.
- The methodology developed has demonstrated to be useful for the analysis of deviations that may arise during assembly processes caused by deviations in the constituent parts, deviations in the assembly jigs and/or effects associated to the assembly operations itself. Moreover, it has proven that it might be helpful to improve the processes (through recommendations for the fixation systems, the assembly/fastening sequences, etc) or even the design of the assembly jigs (for example, distribution of vacuum vents and cradles). Nevertheless, from the point of view of the prediction of the deviations (shape and value) and the virtual verification of the tolerances, a higher effort it is required to characterise adequately each of the fixation and positioning systems used during the assemblies in order to improve the accuracy of the results and the confidence of the method.
- The simulation strategies proposed have demonstrated to be feasible to simulate assembly processes as complex as the ones used in aeronautics. The FE techniques proposed for the simulation of these processes has demonstrated its feasibility, adequacy and potential for this type of studies. However, further improvements must be done in order to reduce the computation time required for each simulation. This will allow performing studies with a higher number of casuistries and deeper statistical analysis. In this sense, an improvement in the definition of the deviations of the individual parts will be also needed.
- The extremely high complexity of the FE models required to simulate aeronautical assembly processes need the development of specific tools to control the models definition and checking some control parameters during the calculations, in order to avoid errors that might be difficult to be detected a priori in large FE models and that could invalidate calculations that last several days.
- The methodology and the numerical strategies developed are broad in their possible uses, being of application also for other type of components/assemblies or even for other type of analysis.
Potential Impact:
The expected impact of the project may be resumed in the following points:
- The design of natural-laminar-flow (NLF) aircraft wings requires extremely high precision tolerances that must be controlled over the whole manufacture/construction processes. Controlling the assembly tolerances represents a challenge of great relevance, taking into account the complexity associated if all the possible sources of deviations need to be considered. In this sense, the methodology developed in this project offers clearly significant contributions, allowing to predict possible non-conformities and to analyse corrective or improvement measures.
- From the point of view of the design and development of natural-laminar-flow wings for commercial aircrafts, the method could be useful to analyse new assembly strategies or improving the techniques used nowadays, avoiding the large amount of demonstrators and tests that are usually needed.
- The methodology and the numerical strategies developed are broad in their possible uses, being of application also for other type of components/assemblies. For example, they could be applied to other aircraft structures in standard programmes, focused in the optimization of the tolerances and processes.
- The simplified techniques defined for the representation of different type of fasteners and joints in MEF analysis might be also used in other studies different from the defined for the present project. As an example, the techniques might be also applied in common structural studies or for the analysis of part tolerances under service conditions. Moreover, these techniques can be also applicable in other sectors (automotive, railway, etc).
- Additionally, the project has demonstrated that the numerical techniques used can be helpful also to support the design of the assembly tools and jigs (for example, the fixation systems).
- Finally, the project has covered several activities that are inevitably required to address the problem of dimensional and geometric tolerances of aircraft assemblies designed to operate under natural laminar flow (NLF) regime, considered a key technology to reduce drag and, thus, to significantly improve the airplanes performance and efficiency. In this sense, the project is an important step for the possible future implementation of this technology, representing a significant contribution to the European competitiveness in aeronautics.
Regarding the diffusion/dissemination of the SATCAS project, the activities already performed and the ones planned for the next months cover various presentations and publications in different events and media. They are listed below:
1) ITAINNOVA WEBPAGE - Entry in the web of the Technological Institute of Aragon.
2) MATERPLAT Annual Meeting - Poster presented at a meeting of the Materials Spanish platform MATERPLAT (Oct – 2012).
3) AERONAUTICAL FAIR – SEVILLE (Aerospace and Defence Meetings Seville – June 2014)
4) Presentations to BLADE partners - Presentations that were given to some of the partners of the BLADE project in the Quarterly Review Meetings.
5) Presentations to ITAINNOVA’s customers - Presentations that were given to some costumers of ITAINNOVA.
6) Newsletter Clean-Sky - September 2014 Newsletter – eNews – Clean Sky Webpage.
7) International Workshop on Aircraft System (Hamburg, February 2015). Article already accepted.
8) 2015 SIMULIA Community Conference (Berlin, May 2015). Abstract initially accepted.
9) Publication of part of the work developed in a peer-review journal (not defined yet, to be done in 2015).
The activities have been performed in all cases with the permission of the Topic Manager, the BLADE project coordinator and the Project Officer.
List of Websites:
Website address: not applicable.
contact: achiminelli@itainnova.es
