Final Report Summary - TIFAN (manufacturing by SLM of TItanium FAN wheels. Comparison with a conventional manufacturing process)
Executive Summary:
Titanium alloys are increasingly used because of their properties: High specific strength at elevated temperature, outstanding fracture resistance, excellent fatigue behavior, and exceptional corrosion resistance. However, the complexity of the primary extraction process, difficulty of melting (due to its high reactivity with oxygen) and problems during fabrication and machining of titanium alloys make them expensive compared to most of other alloys, limiting this way their application.
Thus, titanium alloys are especially appropriate in parts subjected to high centrifugal loads such as disks and blades, which have reduced flow diameters, as well as operate under severe fatigue conditions. Nevertheless most of titanium aero parts are manufactured by conventional machining methods, such as turning, milling, drilling, grinding, etc. In all of them, material is removed to achieve the final geometry, resulting in significant high-cost material wastage. In contrast to conventional subtractive methods, additive manufacturing (AM) is the creation of objects layer by layer. In recent years, Selective Laser Melting, an AM technology for metal parts, has gained in maturity. In theory, it has the advantage of being a cost effective and environmentally friendly and sustainable manufacturing process for producing one-off parts with high geometrical complexity structures.The manufacturing of disks, blades and in general complex geometry parts, arise as an excellent opportunity to exploit the advantages of SLM additive manufacturing technology. This was the case of the fan wheel of an air cooling unit manufactured within TIFAN project, which is currently made of stainless steel because of the high cost of producing this part in titanium by conventional machining.
In the framework of Eco-Design ITD for Airframe application, a new generation of lightweight environmental friendly fan wheels made of Titanium alloy TA6V were developed by means of an environmental friendly additive manufacturing technology: Selective Laser Melting (SLM). Reduction of component’s weight, improved material usage efficiency and reduction of material wastage (saving of raw material and reduction of scrap rates) were accomplished by changing the base material from stainless steel (density 8.0 g/cm3) to TA6V alloy (density 4.5 g/cm3) and by taking fully advantage of lightweight design possibilities and maximum material savings of SLM technology. It must be mentioned that SLM is considered as an environmental friendly technology in which the scraps can be reduced to the minimum and more than 95% of the remaining material (powder that has not melted) may be recycled. The mechanical properties are equivalent to conventionally manufactured parts.
Despite the promising expectations above mentioned, it was observed that TA6V fan wheels produced by SLM have greater environmental impact than those produced by conventional bar machining (4 times higher), contrary to what was expected. This non desirable result is mainly attributable to the impact of post-treatments and transport by air of TA6V powder necessary for the SLM process.
Project Context and Objectives:
The aim of TIFAN project was to develop an alternative green manufacturing process for a TA6V fan wheel of an air cooling unit fabrication by Selective Laser Melting (SLM), that is currently made of stainless steel and produced by bar machining. Furthermore, the project addressed the comparison between optimized SLM and conventional manufacturing process (bar machining) in terms of mechanical, fatigue and corrosion performance as well as environmental impact.
The optimized SLM process comprised the definition of the required manufacturing route including surface finishing and thermal treatments, process parameters, machine and powder characteristics. Additionally, methodologies for taking full advantage of SLM process based on the improvement of surface quality, powder usage efficiency and wheel design were validated.
Nevertheless, the objective of TIFAN project was not only to achieve the aimed goals (manufacturing of TA6V fan wheel demonstrators and comparative study with conventional bar machining manufacturing process), but to identify the critical factors that could give rise to a further manufacturing cost reduction, weight reduction and mechanical performance improvement, so as to meet with overall requirements of Eco-Design ITD program.
The specific objectives related to full life cycle of component (design and production, use and maintenance and withdrawal) are listed below:
• Reduction of component’s weight. A 50% weight reduction target in comparison with current fan wheels (made of stainless steel) and 10% in comparison with Titanium fan wheels manufactured by bar machining due to new SLM design possibilities.
• Improved corrosion resistance compared to conventional manufactured parts. Reduction of 10% of weight loss after exposure to corrosion conditions according to reference standards.
• Improved mechanical properties, both static and dynamic (fatigue). Mechanical properties will be at least equivalent to conventionally manufactured parts.
• Improved material usage efficiency by reducing buy-to-fly ratios (target value 1.25:1).
• Reduction of material wastage by applying SLM technology and optimizing powder recycling methodology: saving of raw material and reduction of scrap rates. A reduction of waste of raw material of at least 40% in comparison with bar machining.
• Optimization of SLM manufacturing costs. Definition and validation of different cost reduction strategies including optimization of powder usage efficiency, study of alternative SLM processing routes (with and without surface finishing) and working with low cost powders.
Other collateral objectives of TIFAN project, aligned with Eco-Design ITD and coming from the implementation of SLM, were:
• Minimal production of dangerous wastes (lubricants, waste oil, polluted scrap, contaminated rags and absorbents, etc.).
• Being SLM a green technology, reduction of CO2 emissions related to the energy consumption required for part manufacturing by at least 10% compared to conventional manufacturing processes.
• Energy consumption reduction in the air cooling system due to the design optimization.
• Reduction of noise output of manufacturing process. New manufacturing process will completely eliminate the noises related to current bar machining process of Titanium alloys.
Many of these specific goals should have been accomplished by changing base material from stainless steel (density 8.0 g/cm3) to TA6V alloy (density 4.5 g/cm3) and taking fully advanced of lightweight design possibilities and maximum material savings of SLM technology. It must be mentioned that SLM is considered as an environmental friendly technology in which the scraps can be reduced to the minimum and 95-98% of the remaining material (powder that has not melted) may be recycled.
Project Results:
The following are the main tasks carried out during the whole project as well as the results achieved:
1- Use of recycled powder (ACHIEVEMENT: Recycling methodology validated - Powder can be reused up to ten times)
The quality of recycled powder (sieved and dried) used in consecutive batched was determined following the same procedure as with fresh powder. In addition, oxygen content was measured. It was concluded that recycled powder can be used up to ten times without any relevant oxidation.
2- Selection of process parameters-manufacturing of simple samples (=> ACHIEVEMENTS: Almost fully dense parts produced with a surface roughness lower than 10 microns)
The SLM process parameters were optimized in order to obtain parts with optimal surface condition, roughness and relative density. For that a series of cubes of 10x10x10 mm were built being the starting point the material data master. Both contour and boundaries were optimized. With these optimized manufacturing conditions cubes with the following properties are obtained:
• Relative densities of almost 100 % .
• The microstructure of as-built samples is alpha prime martensite as expected.
• Roughness value (Ra) of 9.0 µm signifying a reduction of 20% comparing to non-optimized parts.
3- Manufacturing of control (testing samples)-thermal and post-processing operations (=> ACHIEVEMENTS are later explained within each sub-task)
Different control or testing samples were manufactured using optimised SLM parameters. Control samples included tensile, fatigue and corrosion testing samples.
3.1- Tensile samples
The manufactured SLM tensile samples were subjected to different thermal treatments, including conventional thermal treatment and HIP at different temperatures. Tensile properties, microhardness and microstructure were analysed in order to study the mechanical behaviour of the samples.The main results and conclusions of these tests are the following:
• The ductility of as-built samples is very low, it is absolutely necessary to subject the samples to a thermal treatment in order to change the microstructure and release residual stresses generated in the selective laser melting process.
• Conventionally treated samples exhibit lower strength but improved ductility due to the transformation of alpha prime to alpha + beta phases.
3.2- Corrosion samples
Corrosion samples were also heat and HIP treated. Afterwards, three different surface treatments were applied both to thermally treated samples and to samples in as-built state. These surface treatments gave rise to different surface roughnesses. The main conclusion was that smoother surfaces have better corrosion behaviour: Lower roughness postpones the beginning of corrosion.
3.3- Fatigue samples
In the first stage, the manufactured fatigue samples were subjected to the same thermal treatments performed in corrosion pieces. Then different surface treatments were applied. In principle, the diverse roughness values were enough to study its effect in fatigue performance, but the first poor fatigue results forced to machine the samples. The most relevant conclusions acquired from these tests are:
• With the selected surface treatments different roughness values have been achieved, needed for the study of the roughness in fatigue performance. Two surface treatments gave rise to Ra values below required threshold value (1 µm).
• Facet size is a critical factor for ensuring good fatigue performance. This is even more critical than apparent roughness (measured parallel to facets). This effect cannot be masked or removed with surface treatment.
• If the effect of facet size is removed, it can be concluded that HIPing is required in order to ensure good fatigue performance. This is linked to the possibility of removing porosity and other internal defects (lack of fusion). These defects have been identified as fracture initiation points. Subsurface defects are particularlly critical defects that affect fatigue performance. Nevertheless, conventionally treated samples with the effect of facet size removed, present promising mechanical properties (fatigue life close to 480 MPa).
4- Manufacturing of first prototypes (pizza-slice) (=> ACHIEVEMENT: Defined the best strategy to manufacture the fan wheels - optimized supporting structure)
Before starting with fan wheel demonstrator, first prototypes were manufactured with similar shape as the final fan wheel. These parts consisted of a quarter of a fan wheel named as pizza-slice. The objective of this task was to manufacture pizza-slices meeting dimensional specifications. For that, it was required the optimisation of supporting strategy. It must be mentioned that the supporting strategy should ensure dimensional stability (avoid collapse) and final dimensions (avoid distortions) while using the minimum number of supports. In addition, supporting structure should be easily removed. In addition, regarding the efficiency of metal powder usage, the need of supporting structures reduces this efficiency because powder required to build the supports cannot be recycled. Therefore, an optimization of the supports structure and volume is necessary in order to reduce the volume of supporting structures.
All the relevant results obtained from the manufacturing of pizza-slices were transferred directly to final fan wheels.
Several pizza-slices with different supporting strategies were manufactured. The main differences between of these prototypes relied on:
• Nature and shape of supports: both solid (conic, cylindrical or wall shape) and non-solid (thin reticular net) supports.
• Location of supports: edge supporting and blade surface supporting.
• Number and density of supports. This was quantified comparing the relative supporting volume vs total sample volume (pizza-slice + supports).
• Relative orientation of pizza slice vs wiper.
The manufactured pizza-slices were characterised in terms of dimensional accuracy, that is, non-destructive tests were applied consisting of tridimensional measurements. It should be pointed out that the validation of pizza-slices with non-solid supports was carried out validating the fan wheels with the same supporting strategy.
Besides non-destructive tests, several destructive tests were performed to pizza-slices. The transversal section of blades was analysed measuring the density, studying the defects present in the section and the microstructure, and measuring the microhardness. The destructive tests were realized to pizza-slices in as-built state and after applying thermal treatment.
The most relevant conclusions drawn from the manufacturing and testing/measuring of pizza slices were:
• Extended supporting structures are required to avoid SLM process induced distortions of blades.
• Supporting strategies that enable to meet dimensional specifications (control of distortions) have been determined. Supporting structure was optimised obtaining pizza slices and fan wheels within required tolerances. Both solid and non-solid strategies are suitable but they require high supporting volume.
• Both solid and non-solid supports are not easy to remove and they leave some marks on pizza slices that could not be removed by abrasive polishing.
• Difficulties in optimizing powder usage efficiency have been detected due to problems finding the equilibrium between low supports quantity and dimensional accuracy and stability.
• The surface quality of the bottom of the blade is worse than the top of the blade, due to the manufacturing on powder of the bottom side.
• Samples HIPed show a very fine microstructure without defects and high hardness together with optimised mechanical properties (tensile tests). This HIP temperature condition was selected for heat treating the final fan wheels.
5- Manufacturing of fan wheels (=> ACHIEVEMENTS: Manufactured 8 fan wheels and defined the best manufacturing route)
Several fan wheels were manufactured with the supporting strategy which met dimensional specifications and using the optimized manufacturing parameters. In the first approach, one of the fan wheels was subjected to a HIP cycle and dimensional measurements were performed before and after the thermal treatment. The numerous measured points that appeared out of dimensions after the HIP cycle indicated the need of subjecting the fan wheels to a stress relieving cycle before HIPing. It is worth noting that in as-built condition this fan wheel was considered valid (only two points were out of dimensions). Thus, as a second approach, another manufactured fan wheel was subjected to a stress relieving treatment. Comparing to the previous fan wheel which was directly HIPed, less points did not meet dimensional specifications. Afterwards, this fan wheel was HIPed at the same conditions.
The rest of the manufactured fan wheels were processed following these two different routes, half of them were stress relieved and the rest were directly HIPed.
6- Comparison between optimized SLM and bar machining (=> ACHIEVEMENTS: SLM samples properties comparable or superior to bar machined samples)
One of the objectives of this project was to compare the properties of SLM samples with properties of samples obtained from conventional manufacturing. Nowadays the fan wheels are machined instead of using additive manufacturing technologies. In this section a comparison between both types of samples is performed in terms of microstructure, tensile properties and fatigue properties. The most relevant conclusions are the following:
• The microstructure of SLM sample is alpha prime martensitic transforming to alpha + beta after the applied thermal treatments. The alpha phase appears as fine needles. However, the bar machined sample is alpha equiaxic and lamellar alpha + beta.
• Although having different microstructure, the microhardness of both types of samples is comparable, around 380 HV.
• SLM samples show superior strength but lower ductility even if they are subjected to a HIP cycle.
In general, it can be concluded that SLM samples properties are comparable or superior to bar machined samples. Nevertheless, in order to reach the results presented, it is compulsory to apply a thermal treatment and a surface treatment. The thermal treatment needs to be a HIP cycle to close the remaining porosity and thus, to have enhanced tensile and fatigue properties.
7- Environmental evaluation (=> ACHIEVEMENTS: Life-cycle impact assessment performed)
Methodology for LCA was defined by CTME. LCA gives the order of magnitude of CO2 emitted for the different manufacturing ways. It is possible to quantify the gains and its magnitude. It reveals too where to focus the effort to use SLM technology at best and to further minimize environmental impact. The goal of the comparative LCA was to compare the environmental behaviour of the combination of two manufacturing processes and two materials used to manufacture a fan wheel, with the aim of identifying which technology/material is better from an environmental point of view. The planned application of the results of this study is for internal use in order to implement improvements in the product ecodesign.
Five methodologies, included in the commercial software SimaPro 8.0.4.30 were used to model impacts: CML 2011, ReCiPe (H) endpoint and midpoint, Cumulative Energy Demand, Impact 2002+ and ILCD 2011 Midpoint +.
In view of the results of life cycle impact assessment, it was considered that all relevant information and data necessary for the interpretation were available and complete. When comparing the cradle-to-gate environmental profile of the three fan wheel under study, it was observed that the TA6V SLM fan wheel was the one with the greatest impact, followed by TA6V machining fan wheel; being the stainless steel machining fan wheel the more environmental friendly. However, when considering the entire life cycle, the stainless steel machining fan wheel was the worse one, mainly due to its greater weight, and the relation of the weight with the impact in the use phase (emissions and fuel combustion). The raw material consumption, the management of the process wastes and the manufacturing process are the items with greater impact in the three fan wheels. Nevertheless, the difference of impact between the three product systems is mainly attributable to the impact of post-treatments needed to ensure the right finish of the TA6V SLM fan wheel and to the impact of the transport by air of TA6V powder for the SLM process.
Moreover, SLM manufacturing process requires a great amount of raw material, 32 times the weight of the fan wheel vs. about 10 times the weight of the fan wheel in the case of the machined ones. Hence, the raw material impact in the titanium alloy SLM fan wheel is the highest; however, the benefit associated with the recycling of the waste generated during its manufacture is also the highest, which reduces its impact, since the used assessment model gives credits for recycling or re-use.
Taking into account the impact of product life cycle, eco-design efforts should focus on the use phase, lightening the fan wheel or increasing its energy efficiency.
Potential Impact:
Although research projects like TIFAN are mainly focused on technical developments it is also part of the work to identify exploitable foreground and present the results to a wider public, e.g. through conferences or fairs. Being a research work co-financed by the Clean sky Joint Undertaking (and therefore also indirectly by the European Commission), the results should be used within European industries directly or later on. The research should lead to a better knowledge of processes and/or technologies in Europe.
TIFAN is a level 3 project (see figure below), so it was working on advanced technology demonstration, specifically for the manufacturing of TA6V fan wheel demonstrators. It is therefore expected to have developments matured to a Technology Readiness Level (TRL) 6.
From an industrial point of view, comparison between two manufacturing process is aimed to detect advantages, optimising manufacturing route and cost-efficiency. And of course to reduce costs or to make the process more efficient is a priority. The industrial value chain relevant to this project would include the Powder supplier or manufacturer (for example, TLS Technik or LPW), the SLM technology supplier (the equipment proposed is SLM Solution GmbH´s latest model), surface finishing providers (such as Kennametal, BinC, Metal Improvement) and the Engineering or Technical services supplier (IK4 and others), all related to the final end-user that would be for example the ITD leaders. All these industrial profiles will benefit from TIFAN as improving the manufacturing process and its target market, and as a consequence improving the whole business, not only the end-users interests.
Environmentally speaking the impact is based on three advantages that SLM has:
• Raw material management (more efficient, 40% less waste and higher recyclability).
• Noise reduction. The noise output of the manufacturing process compared to bar machining will be drastically reduced.
• Being a “green technology”, CO2 emissions related to the energy consumption required for part manufacturing should be reduced by at least 10% compared to conventional manufacturing processes.
On the other hand, TIFAN´s outputs were expected to have essential societal impact. More than 3.0 million people are employed in the European Aircraft and Airlines industry, thus strengthening the competitiveness of this industry and related SME’s through the development of leading AM technologies is vital to Europe’s
economic future:
1. Due to saves and increased safety in manufacturing process and fuel consumption as a consequence of the replacement of several functionalities by a unique lighter multifunctional systems.
2. Technologies and tools that will result from this project are essential to increase the European market from the current level in the next 10 years, and they will provide opportunities for the employment of highly skilled professionals. This would contribute in solving of heavy societal problems interconnected with the high unemployment in Europe derived from the economic crisis.
3. Jobs will primarily be created at subcontractors and suppliers, due to additional constituents of the material used, and therefore suppliers will need to increase production.
4. The new technology has broad potential applications in many other industries (automotive and general transportation, medical, etc.) opening opportunities for further employment.
Dissemination activities and exploitation of results:
TIFAN has not been during the project too active disseminating the results, because it has been carried out in just 18 months and the time has been spent performing research and development activities. During the 1st period of the project the only dissemination activity performed was the uploading of the “Project fiche” (project description, information available in CORDIS) on LORTEK´s website. Both consortium and Topic Manager agreed that some dissemination activities must be carried out and after completion of project some dissemination activities are programmed, which are:
- Exhibition of one demo part in international fairs: Paris Air Show, Aerodays;
- Oral presentation to a scientific event: TRATERMAT 2015 (see below more detailed information in Table 2);
- Publication in a peer-reviewed journal => At least one publication , to be published probably in the first half of 2016. Some scientific journals are already identified and are the following: Materials Science and Engineering A, Journal of Materials processing technology, Science and Technology of Advanced Material, Materials Today, Journal of Materials Science Research, among others. The last three are open access journals;
- Press releases, e.g. in the IK4 research alliance newsletter or in ASERM (Spanish AM association) newsletter… Brief summary of what was done in the project.
Related to exploitation of results, no patent application is planned at the moment, despite of having identified a key exploitable result: "Method for manufacturing TA6V fan wheels by SLM". The final Plan for Use and Dissemination of Foreground (hereinafter referred as PUDF, deliverable D1.2) submitted within 2. period report, identifies the dissemination and exploitation activities that may arise from the project. PUDF was made up of two sections:
* Section A, related to scientific publications generated thanks to the attained foreground: This section describes the dissemination measures, including any scientific publications relating to foreground. Its content will be made available in the public domain thus demonstrating the added-value and positive impact of the project on the European Union; and
* Section B, that contains exploitation plans: This section specifies the exploitable foreground and provides the plans for exploitation. All these data can be public or confidential; the report must clearly mark non-publishable (confidential) parts that will be treated as such by the European Commission. Information under Section B that is not marked as confidential will be made available in the public domain.
List of Websites:
There is no public website.
Relevant contact details:
Dr. (Mr.) Pedro Álvarez, scientific in charge of the project
IK4-LORTEK, Responsible of Processes&Materials Dpt.
email: palvarez@lortek.es
T: +34943882303
Dr. (Mrs.) Ane Miren Mancisidor, Researcher working in SLM
IK4-LORTEK, Researcher of Laser-based manufacturing Research Unit
email: ammancisidor@lortek.es
Dr. (Mrs.) Yolanda Nuñez, Person in charge of Life-cycle assesment study
CENTRO TECNOLÓGICO DE MIRANDA DE EBRO, Researcher of Environment Dpt.
email: yolandanunez@ctme.es
T: +34 947331515
Titanium alloys are increasingly used because of their properties: High specific strength at elevated temperature, outstanding fracture resistance, excellent fatigue behavior, and exceptional corrosion resistance. However, the complexity of the primary extraction process, difficulty of melting (due to its high reactivity with oxygen) and problems during fabrication and machining of titanium alloys make them expensive compared to most of other alloys, limiting this way their application.
Thus, titanium alloys are especially appropriate in parts subjected to high centrifugal loads such as disks and blades, which have reduced flow diameters, as well as operate under severe fatigue conditions. Nevertheless most of titanium aero parts are manufactured by conventional machining methods, such as turning, milling, drilling, grinding, etc. In all of them, material is removed to achieve the final geometry, resulting in significant high-cost material wastage. In contrast to conventional subtractive methods, additive manufacturing (AM) is the creation of objects layer by layer. In recent years, Selective Laser Melting, an AM technology for metal parts, has gained in maturity. In theory, it has the advantage of being a cost effective and environmentally friendly and sustainable manufacturing process for producing one-off parts with high geometrical complexity structures.The manufacturing of disks, blades and in general complex geometry parts, arise as an excellent opportunity to exploit the advantages of SLM additive manufacturing technology. This was the case of the fan wheel of an air cooling unit manufactured within TIFAN project, which is currently made of stainless steel because of the high cost of producing this part in titanium by conventional machining.
In the framework of Eco-Design ITD for Airframe application, a new generation of lightweight environmental friendly fan wheels made of Titanium alloy TA6V were developed by means of an environmental friendly additive manufacturing technology: Selective Laser Melting (SLM). Reduction of component’s weight, improved material usage efficiency and reduction of material wastage (saving of raw material and reduction of scrap rates) were accomplished by changing the base material from stainless steel (density 8.0 g/cm3) to TA6V alloy (density 4.5 g/cm3) and by taking fully advantage of lightweight design possibilities and maximum material savings of SLM technology. It must be mentioned that SLM is considered as an environmental friendly technology in which the scraps can be reduced to the minimum and more than 95% of the remaining material (powder that has not melted) may be recycled. The mechanical properties are equivalent to conventionally manufactured parts.
Despite the promising expectations above mentioned, it was observed that TA6V fan wheels produced by SLM have greater environmental impact than those produced by conventional bar machining (4 times higher), contrary to what was expected. This non desirable result is mainly attributable to the impact of post-treatments and transport by air of TA6V powder necessary for the SLM process.
Project Context and Objectives:
The aim of TIFAN project was to develop an alternative green manufacturing process for a TA6V fan wheel of an air cooling unit fabrication by Selective Laser Melting (SLM), that is currently made of stainless steel and produced by bar machining. Furthermore, the project addressed the comparison between optimized SLM and conventional manufacturing process (bar machining) in terms of mechanical, fatigue and corrosion performance as well as environmental impact.
The optimized SLM process comprised the definition of the required manufacturing route including surface finishing and thermal treatments, process parameters, machine and powder characteristics. Additionally, methodologies for taking full advantage of SLM process based on the improvement of surface quality, powder usage efficiency and wheel design were validated.
Nevertheless, the objective of TIFAN project was not only to achieve the aimed goals (manufacturing of TA6V fan wheel demonstrators and comparative study with conventional bar machining manufacturing process), but to identify the critical factors that could give rise to a further manufacturing cost reduction, weight reduction and mechanical performance improvement, so as to meet with overall requirements of Eco-Design ITD program.
The specific objectives related to full life cycle of component (design and production, use and maintenance and withdrawal) are listed below:
• Reduction of component’s weight. A 50% weight reduction target in comparison with current fan wheels (made of stainless steel) and 10% in comparison with Titanium fan wheels manufactured by bar machining due to new SLM design possibilities.
• Improved corrosion resistance compared to conventional manufactured parts. Reduction of 10% of weight loss after exposure to corrosion conditions according to reference standards.
• Improved mechanical properties, both static and dynamic (fatigue). Mechanical properties will be at least equivalent to conventionally manufactured parts.
• Improved material usage efficiency by reducing buy-to-fly ratios (target value 1.25:1).
• Reduction of material wastage by applying SLM technology and optimizing powder recycling methodology: saving of raw material and reduction of scrap rates. A reduction of waste of raw material of at least 40% in comparison with bar machining.
• Optimization of SLM manufacturing costs. Definition and validation of different cost reduction strategies including optimization of powder usage efficiency, study of alternative SLM processing routes (with and without surface finishing) and working with low cost powders.
Other collateral objectives of TIFAN project, aligned with Eco-Design ITD and coming from the implementation of SLM, were:
• Minimal production of dangerous wastes (lubricants, waste oil, polluted scrap, contaminated rags and absorbents, etc.).
• Being SLM a green technology, reduction of CO2 emissions related to the energy consumption required for part manufacturing by at least 10% compared to conventional manufacturing processes.
• Energy consumption reduction in the air cooling system due to the design optimization.
• Reduction of noise output of manufacturing process. New manufacturing process will completely eliminate the noises related to current bar machining process of Titanium alloys.
Many of these specific goals should have been accomplished by changing base material from stainless steel (density 8.0 g/cm3) to TA6V alloy (density 4.5 g/cm3) and taking fully advanced of lightweight design possibilities and maximum material savings of SLM technology. It must be mentioned that SLM is considered as an environmental friendly technology in which the scraps can be reduced to the minimum and 95-98% of the remaining material (powder that has not melted) may be recycled.
Project Results:
The following are the main tasks carried out during the whole project as well as the results achieved:
1- Use of recycled powder (ACHIEVEMENT: Recycling methodology validated - Powder can be reused up to ten times)
The quality of recycled powder (sieved and dried) used in consecutive batched was determined following the same procedure as with fresh powder. In addition, oxygen content was measured. It was concluded that recycled powder can be used up to ten times without any relevant oxidation.
2- Selection of process parameters-manufacturing of simple samples (=> ACHIEVEMENTS: Almost fully dense parts produced with a surface roughness lower than 10 microns)
The SLM process parameters were optimized in order to obtain parts with optimal surface condition, roughness and relative density. For that a series of cubes of 10x10x10 mm were built being the starting point the material data master. Both contour and boundaries were optimized. With these optimized manufacturing conditions cubes with the following properties are obtained:
• Relative densities of almost 100 % .
• The microstructure of as-built samples is alpha prime martensite as expected.
• Roughness value (Ra) of 9.0 µm signifying a reduction of 20% comparing to non-optimized parts.
3- Manufacturing of control (testing samples)-thermal and post-processing operations (=> ACHIEVEMENTS are later explained within each sub-task)
Different control or testing samples were manufactured using optimised SLM parameters. Control samples included tensile, fatigue and corrosion testing samples.
3.1- Tensile samples
The manufactured SLM tensile samples were subjected to different thermal treatments, including conventional thermal treatment and HIP at different temperatures. Tensile properties, microhardness and microstructure were analysed in order to study the mechanical behaviour of the samples.The main results and conclusions of these tests are the following:
• The ductility of as-built samples is very low, it is absolutely necessary to subject the samples to a thermal treatment in order to change the microstructure and release residual stresses generated in the selective laser melting process.
• Conventionally treated samples exhibit lower strength but improved ductility due to the transformation of alpha prime to alpha + beta phases.
3.2- Corrosion samples
Corrosion samples were also heat and HIP treated. Afterwards, three different surface treatments were applied both to thermally treated samples and to samples in as-built state. These surface treatments gave rise to different surface roughnesses. The main conclusion was that smoother surfaces have better corrosion behaviour: Lower roughness postpones the beginning of corrosion.
3.3- Fatigue samples
In the first stage, the manufactured fatigue samples were subjected to the same thermal treatments performed in corrosion pieces. Then different surface treatments were applied. In principle, the diverse roughness values were enough to study its effect in fatigue performance, but the first poor fatigue results forced to machine the samples. The most relevant conclusions acquired from these tests are:
• With the selected surface treatments different roughness values have been achieved, needed for the study of the roughness in fatigue performance. Two surface treatments gave rise to Ra values below required threshold value (1 µm).
• Facet size is a critical factor for ensuring good fatigue performance. This is even more critical than apparent roughness (measured parallel to facets). This effect cannot be masked or removed with surface treatment.
• If the effect of facet size is removed, it can be concluded that HIPing is required in order to ensure good fatigue performance. This is linked to the possibility of removing porosity and other internal defects (lack of fusion). These defects have been identified as fracture initiation points. Subsurface defects are particularlly critical defects that affect fatigue performance. Nevertheless, conventionally treated samples with the effect of facet size removed, present promising mechanical properties (fatigue life close to 480 MPa).
4- Manufacturing of first prototypes (pizza-slice) (=> ACHIEVEMENT: Defined the best strategy to manufacture the fan wheels - optimized supporting structure)
Before starting with fan wheel demonstrator, first prototypes were manufactured with similar shape as the final fan wheel. These parts consisted of a quarter of a fan wheel named as pizza-slice. The objective of this task was to manufacture pizza-slices meeting dimensional specifications. For that, it was required the optimisation of supporting strategy. It must be mentioned that the supporting strategy should ensure dimensional stability (avoid collapse) and final dimensions (avoid distortions) while using the minimum number of supports. In addition, supporting structure should be easily removed. In addition, regarding the efficiency of metal powder usage, the need of supporting structures reduces this efficiency because powder required to build the supports cannot be recycled. Therefore, an optimization of the supports structure and volume is necessary in order to reduce the volume of supporting structures.
All the relevant results obtained from the manufacturing of pizza-slices were transferred directly to final fan wheels.
Several pizza-slices with different supporting strategies were manufactured. The main differences between of these prototypes relied on:
• Nature and shape of supports: both solid (conic, cylindrical or wall shape) and non-solid (thin reticular net) supports.
• Location of supports: edge supporting and blade surface supporting.
• Number and density of supports. This was quantified comparing the relative supporting volume vs total sample volume (pizza-slice + supports).
• Relative orientation of pizza slice vs wiper.
The manufactured pizza-slices were characterised in terms of dimensional accuracy, that is, non-destructive tests were applied consisting of tridimensional measurements. It should be pointed out that the validation of pizza-slices with non-solid supports was carried out validating the fan wheels with the same supporting strategy.
Besides non-destructive tests, several destructive tests were performed to pizza-slices. The transversal section of blades was analysed measuring the density, studying the defects present in the section and the microstructure, and measuring the microhardness. The destructive tests were realized to pizza-slices in as-built state and after applying thermal treatment.
The most relevant conclusions drawn from the manufacturing and testing/measuring of pizza slices were:
• Extended supporting structures are required to avoid SLM process induced distortions of blades.
• Supporting strategies that enable to meet dimensional specifications (control of distortions) have been determined. Supporting structure was optimised obtaining pizza slices and fan wheels within required tolerances. Both solid and non-solid strategies are suitable but they require high supporting volume.
• Both solid and non-solid supports are not easy to remove and they leave some marks on pizza slices that could not be removed by abrasive polishing.
• Difficulties in optimizing powder usage efficiency have been detected due to problems finding the equilibrium between low supports quantity and dimensional accuracy and stability.
• The surface quality of the bottom of the blade is worse than the top of the blade, due to the manufacturing on powder of the bottom side.
• Samples HIPed show a very fine microstructure without defects and high hardness together with optimised mechanical properties (tensile tests). This HIP temperature condition was selected for heat treating the final fan wheels.
5- Manufacturing of fan wheels (=> ACHIEVEMENTS: Manufactured 8 fan wheels and defined the best manufacturing route)
Several fan wheels were manufactured with the supporting strategy which met dimensional specifications and using the optimized manufacturing parameters. In the first approach, one of the fan wheels was subjected to a HIP cycle and dimensional measurements were performed before and after the thermal treatment. The numerous measured points that appeared out of dimensions after the HIP cycle indicated the need of subjecting the fan wheels to a stress relieving cycle before HIPing. It is worth noting that in as-built condition this fan wheel was considered valid (only two points were out of dimensions). Thus, as a second approach, another manufactured fan wheel was subjected to a stress relieving treatment. Comparing to the previous fan wheel which was directly HIPed, less points did not meet dimensional specifications. Afterwards, this fan wheel was HIPed at the same conditions.
The rest of the manufactured fan wheels were processed following these two different routes, half of them were stress relieved and the rest were directly HIPed.
6- Comparison between optimized SLM and bar machining (=> ACHIEVEMENTS: SLM samples properties comparable or superior to bar machined samples)
One of the objectives of this project was to compare the properties of SLM samples with properties of samples obtained from conventional manufacturing. Nowadays the fan wheels are machined instead of using additive manufacturing technologies. In this section a comparison between both types of samples is performed in terms of microstructure, tensile properties and fatigue properties. The most relevant conclusions are the following:
• The microstructure of SLM sample is alpha prime martensitic transforming to alpha + beta after the applied thermal treatments. The alpha phase appears as fine needles. However, the bar machined sample is alpha equiaxic and lamellar alpha + beta.
• Although having different microstructure, the microhardness of both types of samples is comparable, around 380 HV.
• SLM samples show superior strength but lower ductility even if they are subjected to a HIP cycle.
In general, it can be concluded that SLM samples properties are comparable or superior to bar machined samples. Nevertheless, in order to reach the results presented, it is compulsory to apply a thermal treatment and a surface treatment. The thermal treatment needs to be a HIP cycle to close the remaining porosity and thus, to have enhanced tensile and fatigue properties.
7- Environmental evaluation (=> ACHIEVEMENTS: Life-cycle impact assessment performed)
Methodology for LCA was defined by CTME. LCA gives the order of magnitude of CO2 emitted for the different manufacturing ways. It is possible to quantify the gains and its magnitude. It reveals too where to focus the effort to use SLM technology at best and to further minimize environmental impact. The goal of the comparative LCA was to compare the environmental behaviour of the combination of two manufacturing processes and two materials used to manufacture a fan wheel, with the aim of identifying which technology/material is better from an environmental point of view. The planned application of the results of this study is for internal use in order to implement improvements in the product ecodesign.
Five methodologies, included in the commercial software SimaPro 8.0.4.30 were used to model impacts: CML 2011, ReCiPe (H) endpoint and midpoint, Cumulative Energy Demand, Impact 2002+ and ILCD 2011 Midpoint +.
In view of the results of life cycle impact assessment, it was considered that all relevant information and data necessary for the interpretation were available and complete. When comparing the cradle-to-gate environmental profile of the three fan wheel under study, it was observed that the TA6V SLM fan wheel was the one with the greatest impact, followed by TA6V machining fan wheel; being the stainless steel machining fan wheel the more environmental friendly. However, when considering the entire life cycle, the stainless steel machining fan wheel was the worse one, mainly due to its greater weight, and the relation of the weight with the impact in the use phase (emissions and fuel combustion). The raw material consumption, the management of the process wastes and the manufacturing process are the items with greater impact in the three fan wheels. Nevertheless, the difference of impact between the three product systems is mainly attributable to the impact of post-treatments needed to ensure the right finish of the TA6V SLM fan wheel and to the impact of the transport by air of TA6V powder for the SLM process.
Moreover, SLM manufacturing process requires a great amount of raw material, 32 times the weight of the fan wheel vs. about 10 times the weight of the fan wheel in the case of the machined ones. Hence, the raw material impact in the titanium alloy SLM fan wheel is the highest; however, the benefit associated with the recycling of the waste generated during its manufacture is also the highest, which reduces its impact, since the used assessment model gives credits for recycling or re-use.
Taking into account the impact of product life cycle, eco-design efforts should focus on the use phase, lightening the fan wheel or increasing its energy efficiency.
Potential Impact:
Although research projects like TIFAN are mainly focused on technical developments it is also part of the work to identify exploitable foreground and present the results to a wider public, e.g. through conferences or fairs. Being a research work co-financed by the Clean sky Joint Undertaking (and therefore also indirectly by the European Commission), the results should be used within European industries directly or later on. The research should lead to a better knowledge of processes and/or technologies in Europe.
TIFAN is a level 3 project (see figure below), so it was working on advanced technology demonstration, specifically for the manufacturing of TA6V fan wheel demonstrators. It is therefore expected to have developments matured to a Technology Readiness Level (TRL) 6.
From an industrial point of view, comparison between two manufacturing process is aimed to detect advantages, optimising manufacturing route and cost-efficiency. And of course to reduce costs or to make the process more efficient is a priority. The industrial value chain relevant to this project would include the Powder supplier or manufacturer (for example, TLS Technik or LPW), the SLM technology supplier (the equipment proposed is SLM Solution GmbH´s latest model), surface finishing providers (such as Kennametal, BinC, Metal Improvement) and the Engineering or Technical services supplier (IK4 and others), all related to the final end-user that would be for example the ITD leaders. All these industrial profiles will benefit from TIFAN as improving the manufacturing process and its target market, and as a consequence improving the whole business, not only the end-users interests.
Environmentally speaking the impact is based on three advantages that SLM has:
• Raw material management (more efficient, 40% less waste and higher recyclability).
• Noise reduction. The noise output of the manufacturing process compared to bar machining will be drastically reduced.
• Being a “green technology”, CO2 emissions related to the energy consumption required for part manufacturing should be reduced by at least 10% compared to conventional manufacturing processes.
On the other hand, TIFAN´s outputs were expected to have essential societal impact. More than 3.0 million people are employed in the European Aircraft and Airlines industry, thus strengthening the competitiveness of this industry and related SME’s through the development of leading AM technologies is vital to Europe’s
economic future:
1. Due to saves and increased safety in manufacturing process and fuel consumption as a consequence of the replacement of several functionalities by a unique lighter multifunctional systems.
2. Technologies and tools that will result from this project are essential to increase the European market from the current level in the next 10 years, and they will provide opportunities for the employment of highly skilled professionals. This would contribute in solving of heavy societal problems interconnected with the high unemployment in Europe derived from the economic crisis.
3. Jobs will primarily be created at subcontractors and suppliers, due to additional constituents of the material used, and therefore suppliers will need to increase production.
4. The new technology has broad potential applications in many other industries (automotive and general transportation, medical, etc.) opening opportunities for further employment.
Dissemination activities and exploitation of results:
TIFAN has not been during the project too active disseminating the results, because it has been carried out in just 18 months and the time has been spent performing research and development activities. During the 1st period of the project the only dissemination activity performed was the uploading of the “Project fiche” (project description, information available in CORDIS) on LORTEK´s website. Both consortium and Topic Manager agreed that some dissemination activities must be carried out and after completion of project some dissemination activities are programmed, which are:
- Exhibition of one demo part in international fairs: Paris Air Show, Aerodays;
- Oral presentation to a scientific event: TRATERMAT 2015 (see below more detailed information in Table 2);
- Publication in a peer-reviewed journal => At least one publication , to be published probably in the first half of 2016. Some scientific journals are already identified and are the following: Materials Science and Engineering A, Journal of Materials processing technology, Science and Technology of Advanced Material, Materials Today, Journal of Materials Science Research, among others. The last three are open access journals;
- Press releases, e.g. in the IK4 research alliance newsletter or in ASERM (Spanish AM association) newsletter… Brief summary of what was done in the project.
Related to exploitation of results, no patent application is planned at the moment, despite of having identified a key exploitable result: "Method for manufacturing TA6V fan wheels by SLM". The final Plan for Use and Dissemination of Foreground (hereinafter referred as PUDF, deliverable D1.2) submitted within 2. period report, identifies the dissemination and exploitation activities that may arise from the project. PUDF was made up of two sections:
* Section A, related to scientific publications generated thanks to the attained foreground: This section describes the dissemination measures, including any scientific publications relating to foreground. Its content will be made available in the public domain thus demonstrating the added-value and positive impact of the project on the European Union; and
* Section B, that contains exploitation plans: This section specifies the exploitable foreground and provides the plans for exploitation. All these data can be public or confidential; the report must clearly mark non-publishable (confidential) parts that will be treated as such by the European Commission. Information under Section B that is not marked as confidential will be made available in the public domain.
List of Websites:
There is no public website.
Relevant contact details:
Dr. (Mr.) Pedro Álvarez, scientific in charge of the project
IK4-LORTEK, Responsible of Processes&Materials Dpt.
email: palvarez@lortek.es
T: +34943882303
Dr. (Mrs.) Ane Miren Mancisidor, Researcher working in SLM
IK4-LORTEK, Researcher of Laser-based manufacturing Research Unit
email: ammancisidor@lortek.es
Dr. (Mrs.) Yolanda Nuñez, Person in charge of Life-cycle assesment study
CENTRO TECNOLÓGICO DE MIRANDA DE EBRO, Researcher of Environment Dpt.
email: yolandanunez@ctme.es
T: +34 947331515
