Final Report Summary - TIALCHARGER (Titanium Aluminide Turbochargers – Improved Fuel Economy, Reduced Emissions)
Executive Summary:
Turbocharger usage within passenger vehicles has become an accepted method for automobile manufacturers to greatly reduce emissions and to improve efficiency. When compared with cars of similar power output, those with turbocharged smaller engines can reduce emissions by 20-40% and can increase fuel economy by 15-20%. Although simple in construction, turbochargers operate at high temperatures (>800°C) and at extreme rotational speeds (>100,000 rpm). Current production versions are limited in performance primarily due to materials and manufacturing technology.
The TiAlCharger project set out to overcome these limitations through the creation of an innovative lightweight, cost-effective, mass producible, low inertia turbocharger assembly. This was achieved through advancements in both materials and manufacturing method. TiAl has a lower density compared to currently used nickel super alloys and retains its strength at high temperatures, making it both lightweight and suitable to withstand the operating conditions within a broad range of engine types. Fabricating the rotor wheel in layers from powder using the electron beam melting (EBM) process offers the opportunity to produce a hollow rotor wheel, further contributing to the weight reduction.
To assess the potential of this design/manufacturing concept, a number of challenges had to be overcome:
• Design/optimisation of a hollow rotor wheel.
• Development of a TiAl alloy suitable for turbocharger application, which can be processed successfully using the EBM process.
• Building of rotor wheels from the TiAl alloy using the EBM process.
• Development of a heat treatment protocol to give the required microstructure and material mechanical properties.
• Development of a procedure for joining the TiAl rotors to standard steel shafts, with the required joint strength.
• Development of a joint inspection procedure.
• Investigation of potential surface finishing techniques for EBM rotor wheels (extra to the original scope).
• Manufacture of a wheel/shaft assembly and integration into a turbocharger unit.
• Performance testing of a wheel/shaft assembly.
• Generation of a production time/cost model to demonstrate viability.
A TiAl alloy based on RNT650 has been modified to make it suitable for EBM processing and its properties have been deemed to be fit-for-purpose. A hollow wheel has been designed based on turbocharger performance requirements and practical EBM limitations. Finite element modelling of the hollow wheel design predicts a reduction in mass of 21.5%, when compared to a solid wheel, and an associated reduction in moment of inertia about the centre of mass of approximately 8%, which is an attractive improvement. Rotor wheels have successfully been fabricated from the modified TiAl by the EBM process, heat treated to give the desired microstructure and joined to steel shafts using an EB brazing process. The fabricated wheel/shaft assemblies were machined and balanced and integrated into a turbocharger unit for performance testing. Performance test data showed a higher than expected drop in efficiency and failure of the turbine at moderate testing conditions. This is attributed to the quality of the tested assembly which requires improvements before any further turbocharger tests can be done.
In summary, the TiAlCharger project has proved the concept of producing a turbocharger assembly with a hollow EBM TiAl rotor wheel. Further testing, validation and productionisation work is required to demonstrate fitness-for-purpose, before commercial viability can be fully established. Primarily, it is necessary to:
• Optimise the EBM processing parameters to improve the productivity of the EBM build of rotor wheels of the correct dimensions to allow for post-processing.
• Demonstrate the potential of a cost-effective surface finishing process to produce the required finish whilst achieving the demanding tolerances.
The potential for improved response characteristics with γ-TiAl material for turbocharger turbine wheels warrants this additional development work and the consortium intend to continue to pursue this application.
Project Context and Objectives:
The TiAlCharger project aimed to create a cost-effective, mass producible, low inertia TiAl turbocharger assembly providing:
• Weight savings of 60% and a reduction in mass moment of inertia of 36% (due to the lower density of TiAl compared to currently used nickel super alloys and the hollow configuration made possible by the EBM manufacturing route).
• Expansion in the application of turbochargers to a broad range of engine types (TiAl retains its strength at high temperatures of >950°C).
• Improvement in vehicle efficiency and reduction in CO2 emissions (due to the increased fuel to air ratios achievable as a result of the lightweight rotor).
The technologies behind this innovation are electron beam melting (EBM) and electron beam brazing (EB brazing). The EBM process has the potential to fabricate a turbocharger wheel from successive layers of powder allowing a hollow, lightweight, low-inertia rotor wheel to be formed. The TiAl wheel was joined to the steel shaft using the EB brazing process; the challenge was to create a joint between dissimilar materials that was robust enough to withstand vibrations, high temperatures and rotational speeds present in a turbocharger unit. This fabrication method provides the possibility to manufacture turbocharger wheels from TiAl, which (if of the required quality) retains its strength at high temperatures, expanding the usage of turbochargers to a broad range of engine types.
The TiAlCharger project objectives were:
• Development of consolidated TiAl powders capable of meeting >950°C turbocharger operating temperatures that can be melted and formed using EBM machinery.
• Optimisation of the design of a hollow lightweight TiAl rotor.
• Development and demonstration of post-processing methodology to create γ-TiAl rotor wheels.
• Fabrication by EBM of a low mass turbine wheel with 36% less rotating inertia.
• Development of a process to join the TiAl alloy to a standard steel shaft with lower than a 0.5% failure rate under normal anticipated operating conditions.
• Build a batch quantity of wheel-shaft turbine assemblies using SME supplied/adapted equipment and integrate into an existing turbocharger.
• Demonstrate the ability to produce a γ-TiAl rotor wheel-shaft assembly for <€22 equivalent to a build rate for an example 50g, 30mm x 30mm sized rotor wheel of <1 hour.
• Successfully test a turbine wheel assembly at 950°C and 12% higher operating boost pressure than current generation products.
During the course of the project it became clear that the surface finish of the EBM rotor wheels was a major issue. As a result, an additional objective was added:
• To assess methods available to improve the surface finish of the as EBM TiAl rotors to a level acceptable for turbocharger usage (surface roughness of <4µm).
Project Results:
Optimisation of the design of lightweight rotor
A design requirements specification was produced, against which the developed wheels could be assessed.
A solid wheel FEA model (based on an existing Ni alloy wheel design) was meshed to allow manufacture of a solid wheel using the EBM method and TiAl alloy (via a CAD file for transfer of the design to the EBM machine). FE modelling was used to predict the maximum stress from centrifugal loading at the service operating temperature for both solid and hollow rotor wheels. Several designs of the hollow wheel were analysed and a design that appears to satisfy the stress requirements was generated in a CAD file format appropriate for conversion to a EBM manufacturing program. The model for the hollow wheel design predicted a reduction in mass of 21.5%, when compared to a solid wheel, and an associated reduction in moment of inertia about the centre of mass of approximately 8%. Although the latter falls short of the desired 10% reduction (of moment of inertia), this is considered still to be an attractive improvement.
Selection of TiAl alloy for EBM turbocharger application
A literature review was carried out on powder composition, to understand the effect of alloying elements on the properties of TiAl. This led to the selection of a suitable powder for this application, based on RNT650. This powder choice had the added benefit of allowing a direct comparison with wheels made by casting (which have been produced from this alloy in the past). Based on initial trials, the powder was modified with the aim of achieving the desired properties in the EBM processed components. The grain size distribution was adjusted to improve the flowability of the powder, and in order to counteract an Al loss which is known to be typical during EBM process, the powder was enriched in Al.
EBM trials to make TiAl blanks and hollow wheels
In order to develop the EBM processing parameters for the selected TiAl powder in a controlled way, initially cylindrical blanks were built. These cylindrical blanks were subsequently used to analyse the material properties of the EBM processed material, for joining trials and for mechanical testing.
In order to build the rotor wheel, a packing arrangement had to be developed and support structures had to be designed and inserted. The support structures are very thin additional parts, which on one hand stabilize the part and on the other hand ensure the transfer of heat and charge from the part. The CAD data for the wheel (with the support structures) was used to programme the EBM machine, and then further process development was required to enable the build of rotor wheels. The task of building rotor wheels from the RNT650 based TiAl alloy using the EBM process proved to be more complex than originally thought, with the powder required a higher processing temperature than other TiAl alloys, It became apparent that pre-heating was a critical stage in the processing of the optimised powder and packing samples were used at this stage to maintain the homogeneity of the temperature during building for process stability. This proved particularly important during the early stages of the build job.
The developed conditions and processing parameters were used to build batches of prototype wheels in both the solid and hollow design. The support structures were then removed and a combination of compressed air and ultrasonic treatment was employed to remove the excess powder from the hollow rotor wheels.
Performance and metallurgical properties of TiAl
The optimised powder was thoroughly characterised in terms of grain size distribution, powder morphology and defects, flowability, apparent density, bulk density and chemical composition. It was considered to be fit-for-purpose.
The EBM processed TiAl specimens were characterised in terms of residual porosity, density, roughness, thermal expansion coefficient, heat capacity, thermal conductivity, chemical composition and as EBM microstructure and were also deemed to be acceptable.
A series of heat treatment procedures were examined in an iterative way, resulting in a final heat treatment procedure which gives a lamellar microstructure with a small number of equiassic grains along the lamellar grain boundaries, very similar to the microstructure of a reference cast wheel. The heat treatment was not found to have a significant effect the dimensions of the rotor wheels.
Basic mechanical properties of the heat treated EBM TiAl material were evaluated. There was an issue with porosity in the later cylindrical blanks (affecting the tensile results) which requires further investigation. The creep behaviour at 300MPa compares favourably with data in the public domain for a similar material. The creep data obtained at 200MPa was slightly lower than the reported data. However the material displayed sufficient creep strength for the rotor wheels to be performance tested safely.
Development of a process to join TiAl rotors to steel shafts
Joining a Ni alloy wheel to a steel shaft is quite straight forward. The use of TiAl as a wheel material presents a greater challenge as both Ti and Al form brittle intermetallic compounds when alloyed with iron, potentially leading to a risk of welds being produced with little ductility and severe cracking. Following a review of potential strategies for joining EBM TiAl rotors to steel shafts and a series of preliminary trials, it was concluded that EB brazing was the most appropriate joining method.
The principle was that brazing would be achieved through the use of a localised, diffuse, heat source provided by a defocussed EB positioned either side of the braze line thereby locally heating the steel and TiAl parts above the braze liquidus, melting the braze foil by conduction and forming a brazed joint at the planar interface at the end of the rotor wheel. In order to understand the heating requirements and preferred heat distribution to achieve satisfactory brazing, an FE model of the joint was produced. This allowed simulation of the temperature rise and the EB tailored heat pattern required to achieve a successful braze in different component geometries. The use of this simulation in conjunction with a series of experimental trials led to the development of an EB brazing procedure for the joining of the rotor wheel/shaft assemblies. The quality of the brazed joints was assessed by visual examination, x-ray CT scanning of the brazed region, tensile testing and examination of transverse sections.
It can be concluded that it is a practical possibility to join EBM produced TiAl rotor wheels to near finished steel shafts for application in the serial manufacture of high performance turbochargers. A brazing procedure has been developed, although improvements in process reproducibility are necessary. A nickel based braze alloy proved effective in wetting both the TiAl and steel component parts and the joint strength achieved was sufficient to promote base metal failure in tensile testing on the TiAl side. It was concluded that this was in large part a consequence of the particularly low room temperature fracture toughness of the EBM RNT650 based material in the as-manufactured condition and did not reflect the brazed joint strength, as failures were not at the braze.
In production it is recommended that a shaft specifically designed for brazing is used rather than standard production shafts designed for welding. Details associated with the interim machining tasks (ie the joint detail) and the brazing procedure requires refinement to suit the final volume production (product design and equipment). Issues which should receive particular attention include:
• Ensuring consistent assembly length after brazing.
• Ensuring the required run-out requirements are met.
These points would most likely be addressed through minor product specific joint design detail changes and use of production standard tooling.
Development of a joint inspection procedure
Both visual inspection and x-ray CT techniques have been shown to be viable approaches for inspection of the brazed joints. The two techniques have been shown to be complimentary in that they are able to detect surface and buried flaws respectively. Verification in accordance with the two standards, BS EN 12799:2000 and BS EN ISO 18279:2003, or their international counterparts is appropriate for this application; the highest integrity level (Stringent B) should always be applied in this application. In a pre-production or production environment (where batch testing at an agreed level between customer and supplier is expected) the visual inspection should be applied in accordance with the standard requirements. The x-ray CT should be applied in accordance with the above standards as far as practically possible using the defined procedural parameters above as guidance to setting the actual equipment to be deployed.
Investigation of potential surface finishing technique for EBM rotor wheels
The surface finish of the EBM processed rotor wheels proved to be inadequate (in the as-EBM state) for the turbocharger application. Initial EBM trials produced surface finish figures ~300µm RA, which was later improved to ~30µm RA. Whilst it is acknowledged that planned advancements in EBM machines and further optimisation of EBM processing parameters may improve the surface finish, it is not thought that the surface roughness required for a turbocharger rotor wheel (3.2µm RA) will be reached. Potential surface finishing concepts were assessed, in order to determine preferred methods which could achieve the required surface condition in a uniform manner, whilst being low-cost and suitable for practical implementation in production.
A combination of automated local grit/bead/shot blasting of blade roots (the most critical region for stress raisers) after heat treatment, with CNC machining of the joint detail, references and blade tips/OD appears to be the most effective and low cost option. It should be noted that parts need to be designed and EBM manufactured oversize to accommodate the loss of material associated with the surface finishing process.
Production of prototype wheel assemblies
The innovations described above (hollow rotor wheel design, modified TiAl alloy, EBM procedure, heat treatment protocol, joining procedure) were brought together to produce a series of wheel/shaft assemblies. The wheel shaft assemblies were measured and underwent standard grinding and balancing operations, prior to being checked for condition, accuracy and balance. Rotor wheel/shaft assemblies meeting all the stringent requirements were integrated into a turbocharger unit in readiness for performance testing.
Performance testing of wheel/shaft assemblies
A turbocharger unit containing a solid EBM TiAl rotor/shaft assembly was mounted in the hot gas test bench for testing. Performance test lines were carried out with closed variable geometry at 80, 100 and 120krpm. The variable geometry was changed to fully open position and one further line was measured at 100krpm. During the second line at fully open position with a target of 140krpm, the turbocharger started acoustic striking even at a low speed. The performance test was aborted and the turbocharger was removed from the test bench, disassembled and examined for damage.
There was evidence of some material missing from the back disk between the blades. Additionally there were indications of strong impact marks (from acoustic strike) on the outlet edges of all the nozzles of the VGS and there were small particles left at the heat shield. All other turbocharger parts were in very good condition. With no single mark of rubbing at the back disc or the heat shield it can be assumed that the burst of the pieces at the back disk is the root cause for the turbocharger failure.
The initial performance test data for the turbine with closed variable geometry showed that the solid EBM TiAl turbine has about 3% less efficiency compared to an investment cast γ-TiAl turbine of the same design. It is anticipated that this reduced efficiency and the failure of the rotor back disk at moderate operating conditions after a short test duration, may be attributed to imperfections in the wheel/shaft assembly. Contributory factors include deviations in size and accuracy (particularly the back disk thickness), the influence of the machining operation, and the high surface roughness of the rotor. All these issues could be resolved with further iterative process development.
The joining connection was without noticeable problems. Considering the test bench operation condition and the premature failure of the turbine the joining connection could not be fully tested.
Exploitation and dissemination
A production time/cost model has been designed to generate an estimate of the cost and possible price of an EBM TiAl rotor wheel assembly under serial production conditions. The calculated production times, production costs and sale prices are a good basis for:
• Comparison of different machine set-ups
• Improvements in cost efficiency of the developed technologies
• Comparison to alternative production technologies
Using realistic estimates of the input parameters, a predicted sale price for a wheel shaft assembly of 50-60 Euro seems possible. This compares to the current price of around 20 Euro for an investment cast nickel alloy wheel shaft assembly. Indications suggest that the potential advantages of the hollow, lightweight EBM TiAl rotor wheel shaft assembly would warrant the higher price, especially in the premium/performance markets. It was acknowledged that there were still technological advancements (particularly in the EBM processing) necessary before an accurate processing time/cost could be calculated.
It was concluded that the production of EBM TiAl wheel shaft assemblies could be carried out profitably, providing solutions could be sought to the remaining technological challenges.
Potential Impact:
We have successfully built EBM rotor wheels from a turbocharger grade TiAl (modified RNT650) powder, joined them to a production steel shaft and generated performance test data. The wheels are built to the hollow design, one of the significant innovations within the TiAlCharger project and display considerable weight savings over the current state-of-the-art nickel alloy rotor wheels. However, the project aims (in terms of the reduction in mass moment of inertia, operating temperature, and vehicle efficiency improvements) proved very ambitious and a considerable amount of further development work is required before these can be fully proven and serial production is viable.
The project has removed many of the barriers for the SME consortium to:
• Create a new IP protectable product giving greater security from ever decreasing margins.
• Direct increased sales and revenues worth an estimated €58 million, creating/safeguarding ~176 new jobs p.a by year 5 post-project (EU market).
• Supply into a market technology that is expected to expand from 20% of US cars to over 85%, representing the potential to increase European exports by over >€200m by 2020.
However, further testing, validation and productionisation work is required to demonstrate fitness-for-purpose before commercial viability can be established. Specifically, it is necessary to:
• Optimise the EBM processing parameters to improve the productivity of the EBM build of rotor wheels.
• Demonstrate the potential of a cost-effective surface finishing process to produce the required surface finish whilst achieving the demanding tolerances (including incorporating excess material into the EBM part to allow for post-processing.
The potential for improved response characteristics with γ-TiAl material for turbocharger turbine wheels warrants this additional development work. The consortium intend to continue to work together to pursue this application and are actively looking for suitable funding sources.
List of Websites:
The project website (www.TiAlCharger.com) was established at the start of the project. The website consists of a public area to promote the project and a member’s only area (accessed via a user password) for the storage of project documentation.
Turbocharger usage within passenger vehicles has become an accepted method for automobile manufacturers to greatly reduce emissions and to improve efficiency. When compared with cars of similar power output, those with turbocharged smaller engines can reduce emissions by 20-40% and can increase fuel economy by 15-20%. Although simple in construction, turbochargers operate at high temperatures (>800°C) and at extreme rotational speeds (>100,000 rpm). Current production versions are limited in performance primarily due to materials and manufacturing technology.
The TiAlCharger project set out to overcome these limitations through the creation of an innovative lightweight, cost-effective, mass producible, low inertia turbocharger assembly. This was achieved through advancements in both materials and manufacturing method. TiAl has a lower density compared to currently used nickel super alloys and retains its strength at high temperatures, making it both lightweight and suitable to withstand the operating conditions within a broad range of engine types. Fabricating the rotor wheel in layers from powder using the electron beam melting (EBM) process offers the opportunity to produce a hollow rotor wheel, further contributing to the weight reduction.
To assess the potential of this design/manufacturing concept, a number of challenges had to be overcome:
• Design/optimisation of a hollow rotor wheel.
• Development of a TiAl alloy suitable for turbocharger application, which can be processed successfully using the EBM process.
• Building of rotor wheels from the TiAl alloy using the EBM process.
• Development of a heat treatment protocol to give the required microstructure and material mechanical properties.
• Development of a procedure for joining the TiAl rotors to standard steel shafts, with the required joint strength.
• Development of a joint inspection procedure.
• Investigation of potential surface finishing techniques for EBM rotor wheels (extra to the original scope).
• Manufacture of a wheel/shaft assembly and integration into a turbocharger unit.
• Performance testing of a wheel/shaft assembly.
• Generation of a production time/cost model to demonstrate viability.
A TiAl alloy based on RNT650 has been modified to make it suitable for EBM processing and its properties have been deemed to be fit-for-purpose. A hollow wheel has been designed based on turbocharger performance requirements and practical EBM limitations. Finite element modelling of the hollow wheel design predicts a reduction in mass of 21.5%, when compared to a solid wheel, and an associated reduction in moment of inertia about the centre of mass of approximately 8%, which is an attractive improvement. Rotor wheels have successfully been fabricated from the modified TiAl by the EBM process, heat treated to give the desired microstructure and joined to steel shafts using an EB brazing process. The fabricated wheel/shaft assemblies were machined and balanced and integrated into a turbocharger unit for performance testing. Performance test data showed a higher than expected drop in efficiency and failure of the turbine at moderate testing conditions. This is attributed to the quality of the tested assembly which requires improvements before any further turbocharger tests can be done.
In summary, the TiAlCharger project has proved the concept of producing a turbocharger assembly with a hollow EBM TiAl rotor wheel. Further testing, validation and productionisation work is required to demonstrate fitness-for-purpose, before commercial viability can be fully established. Primarily, it is necessary to:
• Optimise the EBM processing parameters to improve the productivity of the EBM build of rotor wheels of the correct dimensions to allow for post-processing.
• Demonstrate the potential of a cost-effective surface finishing process to produce the required finish whilst achieving the demanding tolerances.
The potential for improved response characteristics with γ-TiAl material for turbocharger turbine wheels warrants this additional development work and the consortium intend to continue to pursue this application.
Project Context and Objectives:
The TiAlCharger project aimed to create a cost-effective, mass producible, low inertia TiAl turbocharger assembly providing:
• Weight savings of 60% and a reduction in mass moment of inertia of 36% (due to the lower density of TiAl compared to currently used nickel super alloys and the hollow configuration made possible by the EBM manufacturing route).
• Expansion in the application of turbochargers to a broad range of engine types (TiAl retains its strength at high temperatures of >950°C).
• Improvement in vehicle efficiency and reduction in CO2 emissions (due to the increased fuel to air ratios achievable as a result of the lightweight rotor).
The technologies behind this innovation are electron beam melting (EBM) and electron beam brazing (EB brazing). The EBM process has the potential to fabricate a turbocharger wheel from successive layers of powder allowing a hollow, lightweight, low-inertia rotor wheel to be formed. The TiAl wheel was joined to the steel shaft using the EB brazing process; the challenge was to create a joint between dissimilar materials that was robust enough to withstand vibrations, high temperatures and rotational speeds present in a turbocharger unit. This fabrication method provides the possibility to manufacture turbocharger wheels from TiAl, which (if of the required quality) retains its strength at high temperatures, expanding the usage of turbochargers to a broad range of engine types.
The TiAlCharger project objectives were:
• Development of consolidated TiAl powders capable of meeting >950°C turbocharger operating temperatures that can be melted and formed using EBM machinery.
• Optimisation of the design of a hollow lightweight TiAl rotor.
• Development and demonstration of post-processing methodology to create γ-TiAl rotor wheels.
• Fabrication by EBM of a low mass turbine wheel with 36% less rotating inertia.
• Development of a process to join the TiAl alloy to a standard steel shaft with lower than a 0.5% failure rate under normal anticipated operating conditions.
• Build a batch quantity of wheel-shaft turbine assemblies using SME supplied/adapted equipment and integrate into an existing turbocharger.
• Demonstrate the ability to produce a γ-TiAl rotor wheel-shaft assembly for <€22 equivalent to a build rate for an example 50g, 30mm x 30mm sized rotor wheel of <1 hour.
• Successfully test a turbine wheel assembly at 950°C and 12% higher operating boost pressure than current generation products.
During the course of the project it became clear that the surface finish of the EBM rotor wheels was a major issue. As a result, an additional objective was added:
• To assess methods available to improve the surface finish of the as EBM TiAl rotors to a level acceptable for turbocharger usage (surface roughness of <4µm).
Project Results:
Optimisation of the design of lightweight rotor
A design requirements specification was produced, against which the developed wheels could be assessed.
A solid wheel FEA model (based on an existing Ni alloy wheel design) was meshed to allow manufacture of a solid wheel using the EBM method and TiAl alloy (via a CAD file for transfer of the design to the EBM machine). FE modelling was used to predict the maximum stress from centrifugal loading at the service operating temperature for both solid and hollow rotor wheels. Several designs of the hollow wheel were analysed and a design that appears to satisfy the stress requirements was generated in a CAD file format appropriate for conversion to a EBM manufacturing program. The model for the hollow wheel design predicted a reduction in mass of 21.5%, when compared to a solid wheel, and an associated reduction in moment of inertia about the centre of mass of approximately 8%. Although the latter falls short of the desired 10% reduction (of moment of inertia), this is considered still to be an attractive improvement.
Selection of TiAl alloy for EBM turbocharger application
A literature review was carried out on powder composition, to understand the effect of alloying elements on the properties of TiAl. This led to the selection of a suitable powder for this application, based on RNT650. This powder choice had the added benefit of allowing a direct comparison with wheels made by casting (which have been produced from this alloy in the past). Based on initial trials, the powder was modified with the aim of achieving the desired properties in the EBM processed components. The grain size distribution was adjusted to improve the flowability of the powder, and in order to counteract an Al loss which is known to be typical during EBM process, the powder was enriched in Al.
EBM trials to make TiAl blanks and hollow wheels
In order to develop the EBM processing parameters for the selected TiAl powder in a controlled way, initially cylindrical blanks were built. These cylindrical blanks were subsequently used to analyse the material properties of the EBM processed material, for joining trials and for mechanical testing.
In order to build the rotor wheel, a packing arrangement had to be developed and support structures had to be designed and inserted. The support structures are very thin additional parts, which on one hand stabilize the part and on the other hand ensure the transfer of heat and charge from the part. The CAD data for the wheel (with the support structures) was used to programme the EBM machine, and then further process development was required to enable the build of rotor wheels. The task of building rotor wheels from the RNT650 based TiAl alloy using the EBM process proved to be more complex than originally thought, with the powder required a higher processing temperature than other TiAl alloys, It became apparent that pre-heating was a critical stage in the processing of the optimised powder and packing samples were used at this stage to maintain the homogeneity of the temperature during building for process stability. This proved particularly important during the early stages of the build job.
The developed conditions and processing parameters were used to build batches of prototype wheels in both the solid and hollow design. The support structures were then removed and a combination of compressed air and ultrasonic treatment was employed to remove the excess powder from the hollow rotor wheels.
Performance and metallurgical properties of TiAl
The optimised powder was thoroughly characterised in terms of grain size distribution, powder morphology and defects, flowability, apparent density, bulk density and chemical composition. It was considered to be fit-for-purpose.
The EBM processed TiAl specimens were characterised in terms of residual porosity, density, roughness, thermal expansion coefficient, heat capacity, thermal conductivity, chemical composition and as EBM microstructure and were also deemed to be acceptable.
A series of heat treatment procedures were examined in an iterative way, resulting in a final heat treatment procedure which gives a lamellar microstructure with a small number of equiassic grains along the lamellar grain boundaries, very similar to the microstructure of a reference cast wheel. The heat treatment was not found to have a significant effect the dimensions of the rotor wheels.
Basic mechanical properties of the heat treated EBM TiAl material were evaluated. There was an issue with porosity in the later cylindrical blanks (affecting the tensile results) which requires further investigation. The creep behaviour at 300MPa compares favourably with data in the public domain for a similar material. The creep data obtained at 200MPa was slightly lower than the reported data. However the material displayed sufficient creep strength for the rotor wheels to be performance tested safely.
Development of a process to join TiAl rotors to steel shafts
Joining a Ni alloy wheel to a steel shaft is quite straight forward. The use of TiAl as a wheel material presents a greater challenge as both Ti and Al form brittle intermetallic compounds when alloyed with iron, potentially leading to a risk of welds being produced with little ductility and severe cracking. Following a review of potential strategies for joining EBM TiAl rotors to steel shafts and a series of preliminary trials, it was concluded that EB brazing was the most appropriate joining method.
The principle was that brazing would be achieved through the use of a localised, diffuse, heat source provided by a defocussed EB positioned either side of the braze line thereby locally heating the steel and TiAl parts above the braze liquidus, melting the braze foil by conduction and forming a brazed joint at the planar interface at the end of the rotor wheel. In order to understand the heating requirements and preferred heat distribution to achieve satisfactory brazing, an FE model of the joint was produced. This allowed simulation of the temperature rise and the EB tailored heat pattern required to achieve a successful braze in different component geometries. The use of this simulation in conjunction with a series of experimental trials led to the development of an EB brazing procedure for the joining of the rotor wheel/shaft assemblies. The quality of the brazed joints was assessed by visual examination, x-ray CT scanning of the brazed region, tensile testing and examination of transverse sections.
It can be concluded that it is a practical possibility to join EBM produced TiAl rotor wheels to near finished steel shafts for application in the serial manufacture of high performance turbochargers. A brazing procedure has been developed, although improvements in process reproducibility are necessary. A nickel based braze alloy proved effective in wetting both the TiAl and steel component parts and the joint strength achieved was sufficient to promote base metal failure in tensile testing on the TiAl side. It was concluded that this was in large part a consequence of the particularly low room temperature fracture toughness of the EBM RNT650 based material in the as-manufactured condition and did not reflect the brazed joint strength, as failures were not at the braze.
In production it is recommended that a shaft specifically designed for brazing is used rather than standard production shafts designed for welding. Details associated with the interim machining tasks (ie the joint detail) and the brazing procedure requires refinement to suit the final volume production (product design and equipment). Issues which should receive particular attention include:
• Ensuring consistent assembly length after brazing.
• Ensuring the required run-out requirements are met.
These points would most likely be addressed through minor product specific joint design detail changes and use of production standard tooling.
Development of a joint inspection procedure
Both visual inspection and x-ray CT techniques have been shown to be viable approaches for inspection of the brazed joints. The two techniques have been shown to be complimentary in that they are able to detect surface and buried flaws respectively. Verification in accordance with the two standards, BS EN 12799:2000 and BS EN ISO 18279:2003, or their international counterparts is appropriate for this application; the highest integrity level (Stringent B) should always be applied in this application. In a pre-production or production environment (where batch testing at an agreed level between customer and supplier is expected) the visual inspection should be applied in accordance with the standard requirements. The x-ray CT should be applied in accordance with the above standards as far as practically possible using the defined procedural parameters above as guidance to setting the actual equipment to be deployed.
Investigation of potential surface finishing technique for EBM rotor wheels
The surface finish of the EBM processed rotor wheels proved to be inadequate (in the as-EBM state) for the turbocharger application. Initial EBM trials produced surface finish figures ~300µm RA, which was later improved to ~30µm RA. Whilst it is acknowledged that planned advancements in EBM machines and further optimisation of EBM processing parameters may improve the surface finish, it is not thought that the surface roughness required for a turbocharger rotor wheel (3.2µm RA) will be reached. Potential surface finishing concepts were assessed, in order to determine preferred methods which could achieve the required surface condition in a uniform manner, whilst being low-cost and suitable for practical implementation in production.
A combination of automated local grit/bead/shot blasting of blade roots (the most critical region for stress raisers) after heat treatment, with CNC machining of the joint detail, references and blade tips/OD appears to be the most effective and low cost option. It should be noted that parts need to be designed and EBM manufactured oversize to accommodate the loss of material associated with the surface finishing process.
Production of prototype wheel assemblies
The innovations described above (hollow rotor wheel design, modified TiAl alloy, EBM procedure, heat treatment protocol, joining procedure) were brought together to produce a series of wheel/shaft assemblies. The wheel shaft assemblies were measured and underwent standard grinding and balancing operations, prior to being checked for condition, accuracy and balance. Rotor wheel/shaft assemblies meeting all the stringent requirements were integrated into a turbocharger unit in readiness for performance testing.
Performance testing of wheel/shaft assemblies
A turbocharger unit containing a solid EBM TiAl rotor/shaft assembly was mounted in the hot gas test bench for testing. Performance test lines were carried out with closed variable geometry at 80, 100 and 120krpm. The variable geometry was changed to fully open position and one further line was measured at 100krpm. During the second line at fully open position with a target of 140krpm, the turbocharger started acoustic striking even at a low speed. The performance test was aborted and the turbocharger was removed from the test bench, disassembled and examined for damage.
There was evidence of some material missing from the back disk between the blades. Additionally there were indications of strong impact marks (from acoustic strike) on the outlet edges of all the nozzles of the VGS and there were small particles left at the heat shield. All other turbocharger parts were in very good condition. With no single mark of rubbing at the back disc or the heat shield it can be assumed that the burst of the pieces at the back disk is the root cause for the turbocharger failure.
The initial performance test data for the turbine with closed variable geometry showed that the solid EBM TiAl turbine has about 3% less efficiency compared to an investment cast γ-TiAl turbine of the same design. It is anticipated that this reduced efficiency and the failure of the rotor back disk at moderate operating conditions after a short test duration, may be attributed to imperfections in the wheel/shaft assembly. Contributory factors include deviations in size and accuracy (particularly the back disk thickness), the influence of the machining operation, and the high surface roughness of the rotor. All these issues could be resolved with further iterative process development.
The joining connection was without noticeable problems. Considering the test bench operation condition and the premature failure of the turbine the joining connection could not be fully tested.
Exploitation and dissemination
A production time/cost model has been designed to generate an estimate of the cost and possible price of an EBM TiAl rotor wheel assembly under serial production conditions. The calculated production times, production costs and sale prices are a good basis for:
• Comparison of different machine set-ups
• Improvements in cost efficiency of the developed technologies
• Comparison to alternative production technologies
Using realistic estimates of the input parameters, a predicted sale price for a wheel shaft assembly of 50-60 Euro seems possible. This compares to the current price of around 20 Euro for an investment cast nickel alloy wheel shaft assembly. Indications suggest that the potential advantages of the hollow, lightweight EBM TiAl rotor wheel shaft assembly would warrant the higher price, especially in the premium/performance markets. It was acknowledged that there were still technological advancements (particularly in the EBM processing) necessary before an accurate processing time/cost could be calculated.
It was concluded that the production of EBM TiAl wheel shaft assemblies could be carried out profitably, providing solutions could be sought to the remaining technological challenges.
Potential Impact:
We have successfully built EBM rotor wheels from a turbocharger grade TiAl (modified RNT650) powder, joined them to a production steel shaft and generated performance test data. The wheels are built to the hollow design, one of the significant innovations within the TiAlCharger project and display considerable weight savings over the current state-of-the-art nickel alloy rotor wheels. However, the project aims (in terms of the reduction in mass moment of inertia, operating temperature, and vehicle efficiency improvements) proved very ambitious and a considerable amount of further development work is required before these can be fully proven and serial production is viable.
The project has removed many of the barriers for the SME consortium to:
• Create a new IP protectable product giving greater security from ever decreasing margins.
• Direct increased sales and revenues worth an estimated €58 million, creating/safeguarding ~176 new jobs p.a by year 5 post-project (EU market).
• Supply into a market technology that is expected to expand from 20% of US cars to over 85%, representing the potential to increase European exports by over >€200m by 2020.
However, further testing, validation and productionisation work is required to demonstrate fitness-for-purpose before commercial viability can be established. Specifically, it is necessary to:
• Optimise the EBM processing parameters to improve the productivity of the EBM build of rotor wheels.
• Demonstrate the potential of a cost-effective surface finishing process to produce the required surface finish whilst achieving the demanding tolerances (including incorporating excess material into the EBM part to allow for post-processing.
The potential for improved response characteristics with γ-TiAl material for turbocharger turbine wheels warrants this additional development work. The consortium intend to continue to work together to pursue this application and are actively looking for suitable funding sources.
List of Websites:
The project website (www.TiAlCharger.com) was established at the start of the project. The website consists of a public area to promote the project and a member’s only area (accessed via a user password) for the storage of project documentation.