Final Report Summary - NESMONIC (NET SHAPE MANUFACTURE OF Ni SUPERALLOY ENGINE CASING)
Executive Summary:
The final result of Nesmonic project is a validated cost effective NSHIP manufacturing route for IN718 parts for aero-engine application. To obtain this result it has been necessary to optimize all steps of powder manufacturing route. Initially, a complete characterization of IN718 powders has been done and in close collaboration with powder producers a powder with the optimum properties has been designed, produced in industrial atomisers and bought. It is mandatory to work with a material with the adequate tap density, interstitials content, chemical composition, morphology and price. After that, it was investigated an outgassing procedure to encapsulate the material without any modification of as-atomised powder properties. In addition, the optimisation of HIP and HT window was essential to successfully achieve the microstructural and mechanical requirements. Nesmonic project has been able to analyse and optimize all these steps and using the optimum experimental conditions, the OEM requirements initially stablished have been successfully achieved although toughness, measured by Charpy tests is smaller than for wrought IN718.
Moreover, the other main advantage of Nesmonic project is the cost benefit of the NSHIP approach. For this reason, many different tooling methods have been studied in the project couple with the use of computational modelling to ensure that net shape or near net shape parts were produced, reducing machining operations. After the analysis of the different tooling methods a complex mid-scale and large demonstrator component has been manufactured and characterised using the optimum tooling methods to demonstrate the ability of Nesmonic project to apply the manufacturing route proposed in the fabrication of a real engine part (Low Pressure Turbine casing).
Finally, additive manufacturing technique (AM) has been also used in the project to fabricate the canister needed for HIPping the IN718 powder and produce the small demonstrator component. This part has been an interesting opportunity to introduce AM technology in the fabrication of engine parts, knowing the current advantages and limitations of this technology and their solutions.
Therefore, the main impact of Nesmonic is in the area of manufacturing. The impact in manufacturing is in both the reduced cost and improved eco-efficiency of the manufacturing process which requires less material and energy input than current manufacturing methods based on forging and machining. The improved Inconel 718 powder, tooling methods, which have been developed in the project, coupled with the use of computational modelling to ensure that net shape parts are produced in the minimum number of iterations, have significant bearing on the overall cost benefit of the NSHIP approach.
In conclusion, the main impacts of this project are:
• Development of a new manufacturing route based on the Hot Isostatic Pressing (HIP) of In718 gas atomized powders able to successfully achieve the mechanical and microstructural requirements of the OEM.
• Significant reduction in raw material and energy consumption during manufacture and performance of high performance IN718 parts.
• Ability to manufacture complex components, with an optimum accuracy due to computational modelling used.
Project Context and Objectives:
Nesmonic Clean Sky Project aims to manufacture static engine components from IN718 using Net Shape Hot Isostatic Pressing of powder (NSHIP).
Net Shape HIPping (NSHIP) of powder is being increasingly used to produce components for a number of industrial applications, including aerospace. This process has the clear potential to vastly improve the ‘buy-to-fly’ ratio of large aero-engine components, when compared with traditional forging and machining. By using NSHIP ‘buy-to-fly’ ratios can be reduced from 10:1 to close to 1.5:1 leading to a significant reduction in material waste, lead time and associated costs.
However, there are several concerns that need attention before Ni superalloys components can be produced using NSHIP.
The NSHIPping is a process used to manufacture near net shape components by HIPping powders (most of the time gas atomized) in an accurately machined can. The mold tooling (can) is sacrificial (only one use) so that its cost has a large impact on the final cost of the process. Furthermore, NSHIP is a batch process, which limits its applicability to components where only hundreds are required per year and this limited number of components means that the cost of iterations is not spread over thousands of components. In addition, NSHIP requires computer-modelling of the tooling since the powder packing density is typically 60-65 % and the response of the powder and of the tooling during the compaction of the powder to 100% density must be modelled if a component of the desired geometry is to be produced. The net shape component is obtained after removing the tooling through machining/pickling and a truly net shape processing is obtained only if minimal subsequent machining is required. In addition to the accuracy in dimensions, the interaction between tooling and powder during HIPping has to be minimized.
In terms of the microstructure of the final powder based component, prior particle boundaries (PPBs), which are mainly decorated with stable oxide and brittle metal carbide precipitates in Powder Metallurgy HIP-processed superalloys, have been reported and compromise some critical mechanical properties. These have been attributed to oxygen-induced surface contamination of pre-alloyed powders. It should be noted that PPB networks are hard to eliminate once they have been formed. The presence of PPBs is particularly difficult to avoid in PM IN718 material due to its alloy chemistry and in addition to oxides and carbides, the formation of detrimental phases like Laves and delta (δ) have been reported in HIPped powder samples. These phases are detrimental to the hot ductility, fatigue and stress rupture properties. To control the PPB networks and the formation of detrimental phases, it is vital to optimise the HIPping route process parameters of pre-alloyed powder and to use powder with minimal interstitial content.
In this context of potential benefits and concerns, the main objective of the project was to develop and validate a cost-effective NSHIP manufacturing route for IN718 parts for aero-engine application.
To reach that objective the overall approach used in the project was to gather key OEM requirements, undertake a comprehensive experimental programme covering HIP processing of IN718 and can- production, supported by modelling, together with the final manufacture of demonstration parts.
The project activity was split into 5 Work Packages (WPs):
WP1 Project Management
WP2. Current limits of NSHIP technology and OEM requirements
WP3 Development of NHIP parameters for IN718
WP4 Can design, manufacture and testing
WP5 Demonstration component manufacture and assessment
WP1 provided the framework for effective management of the programme, including establishing the management team for the project (programme, technical and quality), and the reporting schedule and mechanism.
In WP2 information on the component performance requirements (target material properties, geometrical accuracy, surface finish etc.) was gathered from the OEM. The current limitations of the NSHIP approach was defined based on the available literature and results of previous work undertaken by the partners in the programme. The material usage was predicted based on both the tooling and finished components. The final objective of this WP was to select the demonstration components to be produced within WP5.
In WP 3 the response of IN718 powders (gas atomised, plasma atomised and PREP) to consolidation by HIPping was explored. The acceptable processing window for IN718 powder was established, based on the density/micrographs of simple cylindrical samples produced by CEIT. Further refinement and optimisation of the processing conditions was undertaken using larger samples produced by UoB which will be machined by MTC to provide the key machinability data. Samples were subjected to intense mechanical and microstructural characterization (CEIT/UoB) in as-HIPped and after heat treatment condition.
As soon as the optimum processing conditions were established samples were produced and supplied to NADCAP approved labs to allow sufficient time for a comprehensive range of material testing, including tensile testing, low cycle fatigue, crack propagation, fracture toughness, stress rupture and creep strain.
The experimental programme conducted in WP3 was focused on identifying the optimum HIP conditions so that samples of 100% density were produced. IN718 is one of the most difficult alloys to process by NSHIP. As with many Ni super alloys it is very difficult to get crystal growth to occur across Prior Particle Boundaries (PPBs) because these tend to contain high interstitial contents (commonly in the form of carbides and oxides). Work carried out in the project has been able to solve the problem induced by PPBs by a right combination of powder and HIP window.
The development of lower cost tooling by MTC in WP4 commenced early on in the programme. Test tooling manufactured in WP4 was used in the HIP trials conducted in WP3 and this permitted the particular characteristics of the new tooling methods to be understood. The data generated (in particular powder consolidation together with compressive flow stress data obtained on partially HIPped samples) was used to help perfect the computation model being developed for accurate prediction of the required tool geometry. The computer model was used to calculate the required can geometry to produce the two demonstration components specified by the OEM in WP2. Using this information MTC and UoB designed and manufactured the cans based on the laminated and AM-HIP methods developed in WP4.
In WP5, MTC and UoB produced the demonstration parts, selected in WP2, using the most appropriate powder (size and morphology), optimum processing conditions and can manufacturing methods, developed in WP3 and WP4. The demonstration parts were subjected to extensive evaluation including NDT, geometrical measurement, microstructure, and were also reviewed by the OEM.
In summary, the project has established the HIP processing window for IN718 powder and has explored new low cost canning manufacturing methods development. Then, using optimised parameters, components have been produced and assessed to see if they meet the OEM requirements in terms of accuracy, integrity and mechanical properties.
Project Results:
Main S&T results are described below organized by Work Packages.
WP1 Project management, coordination and dissemination of results
The objective of the project management Work Package (WP1) has been to manage the project ensuring that deliverable deadlines were met, that progress reports were submitted on time and that periodic meetings and discussions were held between the partners. In addition, management of exploitation (including IP issues) and dissemination of the results (including review of publications) resides within this WP.
The project has been relatively simple to coordinate and manage since there was a clear beneficiary of final product (Topic Manager) and executors of work (RTD partners) and the proposed management structure looks as follows:
There was a single point of contact between the Project and Topic Manager – the Programme Manager. He was responsible for all interactions with the EU Topic Manager and communicating issues both to and from the project team. The internal project was controlled by a Project Manager who oversees the day-to-day running of the technical aspects of the project and reports to the Programme Manager.
The Project Manager was supported by a senior technical advisor responsible for Technical Oversight of the project. In addition the Quality Manager was responsible for all aspects of QA in the project. As the Project Manager is at CEIT the teams at UoB and MTC had a local manager to ensure smooth running of the project at these organisations. Finally, depending on the local manager every RTD partner had a project team responsible for the execution of the project.
In terms of quality assurance, the MTC Business Management System (BMS) is a customised, web-based system used by the organisation to control and monitor core business processes. The BMS is MTC’s Quality Assurance program for projects and has been specifically designed for MTC using best-practice learnt across many organisations and industries and to meet the needs of MTC members, research partners and funding organisations. Project managers at MTC are supported by both an Administration Team (responsible for financial tracking, management etc.) and the Project Management Office (PMO). The PMO is responsible for defining the project management processes, producing tool and templates as well as supporting the project managers with help and advice on project server and any other issues.
An Exploitation Manager has been appointed who took responsibility for drafting an exploitation plan, within the first 6 months of the project and revised this plan in the last 6 months of the project to reflect the increased knowledge which has arisen from the project. In addition, this increased knowledge has allowed publishing and participating in a wide variety of congress.
• J. Cortés, M. Aristizabal, M.H. Loretto, M.M. Attallah, I. Iturriza, “Effect of HIP parameters and particle size distribution in the CSL and Twin boundaries of In718” October 2016 PM World Congress 2016, Hamburg.
• M. Aristizabal, J. Cortés, M.H. Loretto, M.M. Attallah, I. Iturriza, R.H.U Khan. “Study of the influence of outgassing parameters, powder particle size and HIP temperature in the mechanical properties of IN718”, 4-7 October 2015.EURO PM2015, Reims, France.
• I. Iturriza, K. Essa, J. Cortés, F. Castro R. Trepleton, C. Carpenter, R.H.U Khan, M.M. Attallah, M.H. Loretto “Net Shape HIPping components of IN718” 9-13 June 2014. HIP’14, the 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden.
• R.H.U Khan, M.H. Loretto, M.M. Attallah, J. Cortés, I. Iturriza, F. Castro. “Microstructure and properties of HIPped Inconel 718” 9-13 June 2014. HIP’14, the 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden.
Web site of Nesmonic was created at the beginning of the project. This share point has been used to share different documents between partners and topic manager, such as deliverables, minutes of the meetings, drawings, reports...
During the three years of the project meeting have been held at least once by month by WebEx and each three months, approximately, face to face meeting have been taken place to update the progress carried out during this period of time. Minutes of each of these meeting have been written, sent to each partner for any modification and submitted to Web site share point of Nesmonic.
Webex dates were:
22nd July 2013; 27th October 2013; 12th December 2013; 12th December 2013; 29th April 2014; 5th November 2014; 4th December 2014; 26th February 2015; 26th March 2015; 29th May 2015; 24th June 2015;2nd October 2015; 15th December 2015; 26th April 2016
And face to face meetings were:
Quick off meeting was held the 29th May 2013 at CEIT; 15th October 2013, UoB; 13rd March 2014, MTC; 4th July 2014, ITP; 7th October 2014, CEIT; 27th January 2015, UoB; 22nd April 2015, MTC; 23rd July 2015, CEIT; 27th October 2015, UoB; 23rd February 2016, MTC; end of project meeting was held the 31st May 2016 at ITP Zamudio (Spain).
WP2 Current limits of NSHIP technology and OEM requirements
The objectives of the WP2 were:
• To highlight the current limits of NSHIP.
• To define the geometry and properties of the components which will be manufactured in this project.
• To predict the material usage for the new approach.
At the beginning of the project the NSHIP limitations were explored. Net Shape Hot isostatic Pressing (NSHIP) is a technology for compacting metal powder to full density to a near net shape component in one step, by simultaneously applying a very large isostatic pressure and high temperature. NSHIP requires computer-modelling of the tooling since the powder packing density is typically 65 % and the response of the powder and of the tooling during the compaction of the powder to 100 % density must be modelled if a component of the desired geometry is to be produced. The net shape component is obtained after removing the tooling. Prior work in this area are in the literature but especially in the case of some Ni alloys specific developments are required if NSHIP is to be a cost-effective process for routine production of reliable aerospace components.
The most obvious limitation of current NSHIP technology is the fact that tooling, which has to be produced to the same accuracy as is required for the final component, is sacrificial. A second limitation of current NSHIP processing is the accuracy of the computer-modelling, since if the shape required is not predicted at the required accuracy a second or even third iteration or post-HIP machining will be required. It is important therefore that computer-modelling is improved, so that numerous iterations and extensive post-HIP machining are not required. In addition, in order to make the process more cost-effective it is desirable that either low cost tooling or re-usable tooling is developed.
Other limitations of NSHIP are more concerned with the material than with the process. Thus the feedstock is powder and this raises several difficulties in developing a process which has the same reliability as the conventional forging/machining route. In the case of IN718 in particular the surfaces of the powder particles are known to be decorated by carbide particles which form on oxide particles and these particles tend to inhibit bonding between the particles during HIPping. There are therefore two conflicting requirements; finer powder size fractions are preferred in order to exclude any large foreign particles in the feedstock, but larger size fractions are preferred to reduce oxidation and thus in the case of IN718 improving interparticle bonding during NSHIP.
In conclusion, NSHIP is a well-investigated process-route for production of components up to about 1 m diameter, but currently there are several limitations which are addressed in the current project. These are:
(i) The cost of the sacrificial tooling.
(ii) The accuracy of the modelling of NSHIP.
(iii) The role of particle size and oxides or other particles in limiting bonding during NSHIP.
The second goal of this WP was to define the geometry and properties of the components which will be manufactured in this project. Demonstration parts to be manufactured in the project were LPT casing and Tail bearing housing. In addition, components requirements, microstructure and mechanical properties have been also supplied by the OEM.
The third and final objective of this WP2 was to define material usage in the project on both tooling and finished parts. Raw material in the project was IN718 powder and it represented a large expenditure. It was then necessary to make an estimation of the powder utilization over the project. Powder was mainly used in: 1.- HIP parameter development, 2.- low cost tooling development, 3.- machined tooling development, 4.- Demonstrators production, 5.- Machinability study and 6.- NADCAP tests. In addition, mild steel powder has been needed and has been bought during the third year for the production of the canister by SLM for the SDC.
Powder bought during the Nesmonic project was as high as 1,104 kg. It was produced by gas atomization, plasma atomization and PREP. 7 different supplier were tested. Interaction to increase powder characteristics was more intense with two of them.
Deliverables related with this WP2 have been successfully completed during the project. T There was also a Milestone associated with WP2.
WP3 Development of NHIP parameters for IN718
The primary objective of this WP was to determine the optimum HIP processing parameters for IN718. The work involved an assessment of the interstitial content and microstructure of powders, assessment of their flowability, composition, powder-filling and response to HIPping using a range of HIP conditions for the production of samples for mechanical testing, machining trials and to identify HIP conditions for the large demonstrator. In addition, the WP contained several other objectives; assessment of the machinability of HIPped samples, test specimen design and NADCAP tests.
• Powder characterisation and response to HIPping
Initially, a complete characterization of different IN718 powders was made. Six different powder suppliers were analysed in this work to evaluate all commercial available IN718. In addition, three diverse types of atomized pre-alloyed IN718 powders were down selected, so an analysis of difference between these processing technologies took place (plasma atomization – PA, gas atomization – GA, and plasma rotating electrode process – PREP). Finally, a wide range of fraction sizes were employed to determine the best fraction powder of work. This wide variety of powder parameters analysed allows guaranteeing the selection of optimum powder to achieve the mechanical properties required.
A wide variety of studies of the diverse powders were carried out. This activity was really important due to the properties of the powder have a strong influence on the mechanical properties of the final material. Therefore, it was completely necessary to successfully achieve the mechanical properties required that the powder selected had the following properties:
▪ Flow rate: If the powder has a high flow rate, the filling of the canister will be easier and all cavities will be filled with powder.
▪ Tap density: High tap density improves the filling of the canister and especially allows better control the shrinkage of the canister during HIP cycle. The accuracy of the modeling of the densification will be also improved for high tap density. Intensive activity was carried out during the project to obtain a powder with a high tap density.
▪ Oxygen and carbon content: Low contents of these elements decrease the presence of oxides and carbides in PPBs and thus, improve the mechanical properties of the material.
▪ Morphology: The presence of flakes, satellites and aggregates in the powder is really bad to achieve the required flowability, tap density and mechanical properties. To avoid this behaviour, manufacturing process has to be optimised to obtain clean powders.
▪ Fraction size: Particle size distribution has been defined in order to get a compromise between tap density and flowability. In addition, the manufacturing process of the powder was also optimized to obtain the highest amount as possible of the fraction of interest (yield), so the price of the powder will be cheaper
To obtain a powder with these optimum properties, analysis of morphology, interstitials, chemical composition and physical properties have been carried out. After these studies, the optimum powder for Nesmonic project has been selected.
Cross sections of particles show that GA powder particles commonly contain pores associated with trapped argon gas. Some pores are also visible in the PREP powder.
Density measurements, thermal analysis and phases determinations by X-ray diffraction have been also done to complete the characterization and ensuring the quality of IN718 powders.
Finally, heat treatments on powders have been carried out trying to decrease oxygen content to promote a reduction of oxides and carbides at PPBs and thus, improving the mechanical properties.
Differential scanning calorimetry (DSC) of the GA powder indicated several reactions during heating/cooling. These are related to the dissolution/ evolution of γ´´, γ´ precipitates as well as niobium carbide (NbC) particles, solidus and liquidus temperature.
All these were used to select HIPping and heat treatment cycles.
ENCAPSULATION:
The encapsulation process of the powder in a canister is really important in this new developed technology. It has a strong influence on the mechanical properties of the final material. During canning and degassing it is mandatory to evacuate gas trapped in between powder particles and avoid interstitials, especially oxygen, increasing in the powder. For this reason, a new outgassing procedure has been developed during Nesmonic project which is able to successfully keep interstitials as the level of as-atomized powder. In addition, a purging system to decrease the oxygen content during degassing step has been investigated. This new technique is able to reduce the oxygen content of as-atomised powders and its influence on mechanical properties was evaluated.
HIP and HT:
Finally, a wide variety of parameters has been studied to optimize the HIP and HT window, such as, HIP temperature, HIP pressure, HIP time, fraction size of powder, interstitials content, effect of oxygen reduction during degassing, chemical composition, manufacturing technology of the powder, different post-HIP HT, HT and different powders. After this complete study, the optimum experimental conditions have been selected to analyze the mechanical and microstructural properties of the material. Microstructure of the material without the presence of PPBs has been observed. Mechanical and microstructural OEM requirements have been successfully achieved, except for non-initially considered Charpy toughness measurements, the reason of the low value of toughness was found and a solution was proposed.
In addition, an interaction between the canister and IN718 during HIPping was detected. Diffusion from the canister increases with increase in temperature. This diffusion layer (~200-300 μm) has to be removed as it can affect the mechanical properties of net shaped components. The thickness of this interaction layer was decreased by using diffusion barriers in the internal wall of the canisters.
Finally, the effects of HIP temperature and post-HIP heat treatments on the % of Σ3 twin boundaries have been studied. In both cases, as-HIP and HIP plus HT, the fraction of Σ3 boundaries was related to the HIP temperature. Thus, the fraction of Coincidence Site Lattice (CLS) was higher at higher HIP temperatures and increased further on heat treatment. In addition, it was observed that the increase of Σ3 boundaries improved the ductility of the material. However, at the first stage of melting (T4) the fraction of Σ3 boundaries and the ductility decreased.
Conclusions
The powder production and raw material used are quite critical for IN718 and interstitials must be kept low to limit PPB formation. The degassing strategy helped in restricting oxides on powder particle surfaces. Finally, an optimum HIP and HT window has been obtained and allowed successfully achieving the mechanical and microstructural OEM requirements, except toughness.
• Machining of HIPped IN718
For the study, a roughing and finishing insert was used at a variety of conditions in order to find the optimum cutting parameters. This was done by looking at the subsurface microstructure, subsurface microhardness, surface finish, tool wear, and analysis of the swarf. For the majority of samples, microhardness did not change between 50 and 500 µm below the surface, however the hardness of the roughing HIPped sample decreased slightly (from 370 HV to 240 HV) between 50 and 250 µm below the surface.
Comparison of optimally HIPped IN718 against forged material:
This section of WP3 was aimed at determining the best generic turning parameters whilst comparing the machinability of forged and HIPped samples in terms of the most appropriate tool material; surface finish achievable and tool wear.
Roughing inserts:
Forged material – very little change in tool wear was evident over the time the inserts were used for (2.3 minutes). Swarf analysis revealed that cutting efficiency reduced with higher feed rates and cut depths.
HIPped material – ‘crater wear’ was consistently observed on most of the tool tips. The swarf for all cuts were all of similar size.
For both materials, no correlation with cutting speed was observed.
Finishing inserts,
Forged material – slight correlation between surface finish and cutting speed was found. The best surface finish (0.50 µm) was achieved at the highest cutting speed. No correlation between tool wear and cutting speed was observed.
HIPped material – no change in surface finish was observed as cutting speed was increased. Tool life appeared to decrease for higher cutting speeds
Overall it appears that the forged material has better machinability than HIPped material. Less tool wear was evident, and reduced work hardening. A better surface finish could also be achieved on the forged material, with an average of 0.95 µm in comparison to 1.27 µm for the HIPped material (finishing passes).
Subsurface Microstructural Features and Mechanical Damages
The grain structure of the HIPped sample was more homogeneous than the forged sample For all passes on the forged samples, no mechanical damage was found in the subsurface at magnifications of 50X, 100X and 200X.
Subsurface microhardness
Work hardening of the deformed layer beneath the machined surface causes a higher superficial hardness than the average hardness of the bulk material. The HIPped material experiences work hardening in the first 0-150 µm. The forged roughing material, however, experiences the opposite, with a lower microhardness in the first 0-150 µm before increasing after 150 µm.
Tool wear
There are many factors which affect the cutting tool wear, including the tool geometry (flank angle, rake angle, cutting edge angle), geometry of workpiece, cutting parameters (speed, feed, depth of cut). The lack of work holding of the HIPped roughing rod noted above had a dramatic effect on the tool life due to the vibration of the rod. The HIPped finishing rod’s work-holding was modified to have the same percentage work holding as the forged finishing rod. As the speed of the tool increases, the cutting temperature inevitably increases, which leads to a slight increase in flank wear on the insert. The increase in feed rate, depth of cut and the tool speed of the roughing cut did not show signs of an increase in tool wear for the forged material.
Surface Finish
Surface roughness plays an important role in a part’s accuracy and service life. Many factors have an impact on the machined surface roughness, namely cutting parameters, tool parameters, and processing parameters. In this study, the tool parameters and processing parameters were kept constant and only the cutting parameters were varied. Only the finishing passes were analysed.
The roughing HIPped billet yielded the highest surface roughness results, with an average value 81% higher than that of the roughing forged billet. It also presents the widest range of values, with a minimum Ra of 1.39 µm and a maximum Ra of 6.13 µm. For the finishing operation, the forged billet has the smallest average Ra value at 0.95 µm which is 30% smaller than the average value for the HIPped billet.
Swarf Analysis
The increase in feed rate and cut depth gave smaller chip sizes for the roughing passes of the forged rod, indicating than a lower feed rate and a lower depth of cut gives better cutting efficiencies. The same correlation was also present within the roughing pass of the HIPped rod, however it was less apparent.
Overall it appears that the forged material has better machinability than HIPped material. Less tool wear was evident and less work hardening of the material. A better surface finish could also be achieved on the forged material, with an average of 0.95 µm in comparison to 1.27 µm for the HIPped material.
Machining of Representative Features for Full Size Casing
Representative features for full size casing were primary machined in the subscale casing.
Features to be machined
Four features were chosen to be machined. These were: A. A top flange undercut, B. A thin wall, C. A rib and D. A bottom flange undercut. To machine the bottom of the casing, the top of the casing was clamped to the spindle. A spider fixture was also attached to the inside of the casing in order to stop the casing from warping.
In summary the most important factor affecting the tool wear in this study was the work-holding. For the HIPped roughing sample, the smaller work holding used caused the rod to experience a much higher vibration, which resulted in many inserts chipping and breaking.
The four features identified, were successfully machined into a sub-scale casing. Due to the nature of the HIPping process, a uniform wall thickness is unachievable and hence casings will inevitably have a run out. In order to reduce the amount of rubbing of the insert against the material, an increased feed rate should be used in order to decrease this time.
• NADCAP tests
Design of test samples–engineering drawings defining the specimens for NADCAP tests were supplier by the OEM.
NADCAP tests performed by Westmoreland
Results:
(i) Charpy tests. Using the optimum conditions for HIPping the Charpy values have been determined at RT and ET for HIPped and forged materials after post-HIP heat treatments. The results obtained are indicating higher Charpy values for forged IN718 than for HIPped HT grade. At both temperatures, RT and 650 °C.
A wide range of post-HIP heat treatments has been carried out in which the Charpy values of solution-treated and of samples aged at different temperatures have been assessed. The most important observation made in these extensive tests is that the Charpy value of as-HIPped samples exceeds that of forged sample and that most post-HIP heat treatments downgrade the Charpy values.
ii) Stress rupture. Using the optimum experimental conditions developed in the project (powder + HIP + HT) the stress rupture at elevated temperature has been analysed. The results obtained successfully achieve the OEM requirements and are better than typical forged material and clearly improve on the values reported in the literature for HIPped samples.
iii) High cycle fatigue (HCF). The values obtained in this tests successfully achieved the OEM requirements. They are even better than typical forged material and clearly improve on the values reported in the literature by this material developed by HIP.
Conclusions from mechanical testing work
NADCAP tests carried out have demonstrated that the OEM requirements are achieved for stress rupture and high cycle fatigue properties of IN718 material developed by the new HIP powder processing route. In addition, this material was analyzed and also achieved the requirements for grain size, microstructure, hardness and tensile properties. However, Charpy test results obtained with HIP + HT samples are below the OEM requirements.
• Overall conclusions for WP3
The objective of this WP was to determine the HIP processing parameters so that optimally processed samples were produced which were then available for mechanical testing, machining trials and production of a large demonstrator.
It has been shown that HIPped samples could be produced which had room temperature and elevated temperature tensile and stress rupture properties, which reached the specifications for forged samples. However, the post HIP heat treatments which were used to achieve these properties very significantly downgraded the excellent Charpy values of the as-HIPped samples. Further detailed microstructural work, involving transmission electron microscopy was carried out in order to understand the microstructural changes occurring during heat treatments which lead to this unacceptable decrease in Charpy values. The conclusion of this characterization is going to be tested out of the active period of the Nesmonic project. The machinability of as-HIPped samples was carried out successfully. The optimum HIP conditions identified were used in the WP5 aimed at producing a large demonstrator.
WP4. Can design, manufacture and testing
The primary objective of this work package (WP4) was to optimise cost-effective can design using novel low cost manufacture methods and to investigate a hybrid AM-HIP method to produce tooling for smaller components. This work also encompassed development of a computational model which enable the geometry of tools to be calculated to enable net shape components to be produced. There were six deliverables in this work package:
1. Tools for WP3 trials
2. Laminated test tools and report on performance
3. AM-HIP test tools and report on performance
4. Validated compaction model
5. Cost evaluation of tooling options
6. Design of demonstration component tools
• Tools for WP3 trials
The major obstacle with the current NSHIP approach is the cost of the tooling. The partnership worked to lower tooling production cost by developing two cost effective canning solutions one for large structural components and the second one for the production of small components.
• Low cost tooling
This report assessed three routes to produce low-cost tooling to produce simplified HIPped In718 casing component to 100% density and appropriate form. The three routes were:
• Machined HIP Tooling
• Sheet metal spinning HIP Tooling
• Laminate tooling
The Machined HIP Tooling subscale component, consolidated symmetrically and in line with the UoB compaction model and showed no indication of failure to consolidate. The component wall thickness was found to be uniform and no local geometric artefacts were observed.
The Sheet metal spun HIP canister passed He leak detection and was also successfully HIPped. However, post HIP, it was found to have a significant reduction in diameter and height. Further iterations involved the use of laminate and solid mandrels. These iterations were able to hold the shrinkage in diameter. Nevertheless, they were unable to hold the shrinkage in height. With the modelling able to model the shrinkage in height, future iterations should be designed to take the height shrinkage into account.
For the laminate tooling, it was found that by the use of discreet fixturing between layers, the required canister wall thickness and cost of laminate HIP canisters could be significantly reduced. This concept of the Laminate tooling failed during the HIP process due to failure to contain powder between the laminates. However, coatings were used in the powder cavity for subsequent iterations of tooling design to prevent powder egress within laminate tools during vibration filling.
The analysis of the three methods conclude that machined canister is the more robust for the production of Large Demonstrator Component. A real LPT casing.
• AM-HIP test tools and report on performance
A novel hybrid route based on AM-HIP for the production of a mount lug has been developed. The tooling or shell was manufactured by SLM powder bed using mild steel powder with optimised parameters. It was found that the quality of the powder was a key factor in obtaining a canister without porosity or cracks.
The As-built shell parts has to be heat treated for stress relief before removing the support structure and the substrate to avoid any distortion or bending. In addition, the original shell design was modified to ensure the lack of distortions.
The obtained near net shape canister was filled by IN718 powder, outgassed, crimped and HIPped. 3D shape measurement, using GOM system, was carried out at the MTC before and after HIPping to evaluate the shape change and the extent of deformation due to powder consolidation during HIPping. The mount-lug reduced its size homogenously in the whole area during the HIP cycle.
The Finite Element computational model of powder compaction and consolidation during HIPping, which was developed for the large demonstration, was modified from 2D to 3D (in a single run) for the small demonstrator or mount lug. The work shown in this section was use to fix the procedure of manufacturing an IN718 mount-lug by mean of hybrid SLM/HIP process. The fixed procedure and the final mount-lug properties are described in WP5.
• Validated compaction model
An important stage of the proposed approach was the comparison of the predicted and the measured geometry, which was extracted by optical 3D scanning system (GOM scanner). The comparison between the measured tooling geometry and that predicted by the FEA model concluded that the axial shrinkage of the powder is larger than the radial shrinkage. The maximum reduction in height and maximum reduction in diameter were measured and considered acceptable. The axial shrinkage is uniform along the radial direction while the radial shrinkage was less uniform along the axial direction. Less radial shrinkage was seen at the upper and lower corners of the inner tooling. This discrepancy may be due to a number of factors, such as the directional stiffness variation of the mild steel tooling (the stiffness in the axial direction is different from that in the radial direction) and the non-uniform pressure distribution during HIPping. The reduction in the internal diameter is larger than that in the external diameter. This is due to the fact the thickness of the internal tooling is much smaller than that of the external tooling thus the pressure is transmitted more easily through the internal surface.
In general, there was a very good agreement between the geometry predicted by the FEA model and the measured one. The maximum deviation between the FE profile and the experimental one was about 0.5mm which is found at the upper and lower corners of the outer tooling at which maximum radial shrinkage is taking place. Although the material model that has been used is relatively simple, very good prediction of the final tooling geometry is achieved. The simulation time of this model is 10 min using a single processor, which represents quite fast computational time as compared with other complicated models.
• Cost evaluation of tooling options
The Casing Cost Evaluation Tool captures all the costs associated with manufacturing a HIP tooling through three manufacture routes: Machining, Spinning, and Laminates.
The Casing Cost Evaluation Tool also contains a Generic Casing Cost Calculator, which is based on the costings of the LPT Casing. This calculator gives the cost of manufacturing a generic casing based on five inputs:
1. Desired % overstock of cavity
2. Desired wall thickness of the final casing
3. Number of complex features
4. Quantity of canisters produced per year
5. Diameter of final casing
Having a greater overstocked canister has two advantages. Firstly it reduces the non-conforming geometry failure risk which occurs during HIP, and secondly, the overstocked design can reduce the complexity of the canister design and hence it would be cheaper to manufacture and the lead time would be reduced. The setback of having a greater overstocked canister is that more powder is needed and the cost of post machining will increase.
The cheapest route to manufacture a LPT casing has been analysed in between different methods and identified taken into account the entire production cycle.
• Design of demonstration component tools
The target component was a full height engine casing of a 1.5 m maximum diameter In718 engine casing. The objective of this casing was to demonstrate the ability to manufacture both net shape and near net shape features. The target desired geometry has four NS and four NNS features.
Design work was carried out to design a machined HIP demonstrator with net shape (NS) and near net shape (NNS) features. Simulated modelling of the HIPping process was also carried out on the design iterations by the University of Birmingham in order to optimise the design. The casing was geometrically assessed by the ATOS Triple Scan and the profile sketches of the casing was analysed against the desired geometry and also the inner tooling pre-HIP.
The finished casing is shown in the attached pdf.
Conclusions of WP4
• Low cost tooling was validated
• A novel hybrid route based on AM-HIP for the production of a mount lug has been developed
• Prediction of the compaction model was confirmed
• Full complexity 0.727 m demo accurate to high % value in both height and diameter
WP5. Component manufacture and validation
The objective of this work package (WP5) was to manufacture demonstration components using the most appropriate powder, HIP parameters, tooling methods and finish machining strategy, as determined from the results of the preceding work packages. The quality of these demonstration components has been assessed by the partners (NDT, metrology and metallurgy) and then they have been supplied to ITP for long term evaluation and to promote the use of the NSHIP approach.
Eight deliverables in this work package have been written:
1. Supply most appropriate powder for demonstration part based on the results of WPs 3 & 4;
2. Tool for small demonstration part;
3. Tool for large demonstration part;
4. Small demonstration part;
5. Large demonstration part;
6. NDT/Metrology reports for demonstration components;
7. Delivery of completed small NSHIP demonstration part to ITP for assessment;
8. Delivery of completed large NSHIP demonstration part to ITP for assessment.
• Most appropriate powder for demonstration part
Previous studies by different researchers show that the raw material can influence the net shape component properties, hence affecting the integrity of the final part. Therefore, it is vital to assess the pre-alloyed powders through an intensive characterisation work.
Tensile properties required by the OEM were only achieved with one of the gas atomised powder, PSD and optimum experimental conditions (PSD, HIP conditions, HT...). Many different effects and experimental conditions were analysed to achieve these properties and understand the behaviour of the material.
These results were obtained because in HIPped specimens, formation of PPBs, which are detrimental for mechanical properties, were avoided. Besides, it is also important that the powder has a high tap density to avoid distortions in the final demonstrator part.
• Small demonstration part
The original mount lug sample geometry was simplified by removing holes. The CAD model was scaled down to 40 %.
The tooling or shell for the small demonstrator component was manufactured by SLM powder bed using mild steel powder.
The near net shape canister was filled by IN718 selected powder, outgassed, crimped and HIPped. 3D shape measurement, using GOM system, was carried out at the MTC before and after HIPping to evaluate the shape change and the extent of deformation due to powder consolidation during HIPping. The mount-lug reduced its size uniformly in the whole area.
Two small demonstrator components (SDC) were manufactured during the project. Due to slight differences in the manufacturing route between the two SDCs, the first one showed better properties and it was characterised in terms of microstructure and XCT tomography. The first SDC was found to be fully dense but with crack especially in areas with stronger curvatures. The microstructure showed a large amount of Al and Ti oxides at the prior particle boundaries and NbC precipitation inside the particles because the outgassing procedure developed in the project was not appropriated carried to the SDC. GOM was used to assess the extent of the shrinkage during HIP. Both SDCs show similar shrinkage in the whole body.
The Finite Element computational model of powder compaction and consolidation during HIPping, which was developed for the large demonstrator, was modified from 2D to 3D (in a single run) for the small demonstrator or mount lug.
• Large demonstration part
It was decided by the consortium that for the final full scale demonstrator, a large 1.5 m IN718 engine casing should be manufacturing through the NSHIP process.
The encapsulation process encompasses everything from the assembly of the canister to the state where the canister is ready for HIPping. This process has the following steps:
• Canister assembly
• Welding
• Weld inspection
• Powder filling
• Evacuation
• Crimping
Canister was machined in two parts and welded. Weld inspection was performed on the canister assembly using a helium leak test. The canister was evacuated for 20 minutes prior to the He leak. The test identified that there was a leak in the canister, however not at the welds. Holes were identified and then covered with 5 mm thick weld-on-patches.
After appropriated filling with optimum IN718 powder and long outgassing, the Canister was HIPped at Bodycote’s Lego HIP in Surahammar Sweden. However porosity was find in the HIPped component. The level of porosity was measured using image analysis software, giving a value of 9.3 % porosity.
A surface comparison of the GOM data of the post-HIP, pre-etched canister against the CAD was carried out. From this data, the following conclusions can be drawn:
• Overall the canister has shrunk 6.4% in volume
• The inner tool has moved inwards
• The outer tool has moved inwards
Cracks appeared around the top flange of the casing during the etching.
The geometrical analysis of the casing against the desired geometry was done. There was not enough shrinkage in the bottom of the casing whilst in the middle and the top, there was too much shrinkage.
The failure during HIP was most likely caused by porosity in outer (thin) canister material. Increasing the thickness of the outer casing would avoid the problem.
• NDT Results for Ø0.74 m Casing
An initial study was carried out to examine the potential NDT routes for the fully dense mid-sized demonstrator component. That component was HIPped with optimum powder at optimum processing window. Dye penetrant inspection, eddy current, magnetic particle inspection, radiography, ultrasound, and visual inspection were considered for NDT characterization. After a downselection, digital radiography was chosen as the most suitable method for detecting surface and subsurface defects, in combination with visual and/or dye penetrant inspection.
Digital radiography was used to assess the part integrity of the NS&NNS casing. The inspection showed no subsurface defects, however did show surface indications around one of the sectors. This, however, was confirmed by visual inspection to be residue canister, which was not completely taken off at the etching stage.
GOM was used to assess the geometrical accuracy of the casing and the scan was compared to the desired geometry. The scans showed that the casing shrunk more than expected, however it shrunk uniformly. The inner tool, which was designed not to move, horizontally also shrunk.
The surface roughness was also measured on 5 parts of the canister, with an average surface roughness of the as-HIPped surface being Ra: 2.41 µm.
Conclusions of WP5
Small and Large Demonstrator components have been HIPped with the optimum powder developed in the project. SDC was fully dense with some cracks from strong curvature areas. Large demonstrator with 1.5 mm in diameter was also HIPped but its density was compromised because of the interconnected porosity of the thin outer canister. However, mid-scale canister has proved that the encapsulated HIPping of IN718 gas atomized powder can produce large PM components free of defects (fully dense, no cracks, no PPBs) with net shape features.
Potential Impact:
Potential impact.
The final result is a validated cost effective NSHIP manufacturing route for IN718 parts for aero-engine application. Therefore, the main impact is in the area of manufacturing. The impact in manufacturing is in both the reduced cost and improved eco-efficiency of the manufacturing process which requires less material and energy input than current manufacturing methods based on forging and machining. The improved tooling methods, which have been developed in the project, coupled with the use of computational modelling to ensure that net shape parts are produced in the minimum number of iterations, have significant bearing on the overall cost benefit of the NSHIP approach.
It is also expected that this project will have an impact on the fuel consumption. In-service benefits will arise from the use of efficiently designed NSHIP components, which will perform better and weigh less, thus reducing fuel consumption. In addition, there will be significant fuel efficiency benefits to be gained from the increased use of high performance nickel alloys in aero engine construction which will be made possible if part manufacturing costs are reduced using NSHIP and final properties are as high as promised.
The potential impact of the project is quantified below;
• Significant reduction in raw material and energy consumption during manufacture and performance of high performance IN718 parts.
▪ 80% reduction in aerospace material consumption (nickel alloys) and elimination of swarf disposal / recycling costs. The reduction in the amount of material required would also lessen the effect of international material supply fluctuations.
▪ 75% reduction in energy-consumption during manufacture by reduction of energy-intensive machining operations.
• Ability to manufacture complex components which will provide the essential high pressure and temperature capability critical to achieve the required reductions in fuel burn in future large civil aero-engines. There is also a huge improvement in engine emissions to be gained with the use of hotter engine cycles enabled by employing high performance nickel alloys that will help the aerospace industry achieve tough ACARE targets such as a 50% reduction in CO2 emissions per passenger-kilometer by 2020.
In addition to the impact directly related with Nesmonic objectives there is also an extra benefice by the diffusion of the capacities of the global PM route NSHIP. The extensive use of NSHIP in aeronautic, energy, oil and gas and other industries will increase the impact of the project in terms of material usage, energy needed for production and even performance. Finally, the successful application of hybrid AM-HIP open the possibility of imaginative combination of technology in order to increase the possibility of producing complicated high performance component.
Dissemination.
Dissemination of the project has been taken place mainly in International Congress but also in some aeronautic forums and webs.
• NESMONIC webpage was released. The webpage can be accessed at [http://www.tecnun.es/nesmonic/]. It describes the project, main goals, methods as well as some issues of relevance for the technology itself.
• News release through CEIT’s webpage. Through CEIT’s electronic newsletter and bulletin, all interested companies and individuals have been informed about the project.
Oral presentations in International Congress have been:
• “Net Shape HIPping components of IN718”. HIP’14. The 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden. 9-13 June 2014
• “Microstructure and properties of HIPped Inconel 718”. HIP’14. The 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden. 9-13 June 2014
• “Study of the influence of outgassing parameters, powder particle size and HIP temperature in the mechanical properties of IN718”. EuroPM´15 Congress held in Reims, France, 4 – 7 October 2015.
• One oral presentation will be carried out in the WorldPM’16 Congress to be held in Hamburg, Gemany, 9 – 13 October, 2016. The title of the paper is: “Effect of HIP parameters and particle size distribution in the CSL and Twin boundaries of In718”.
• A poster was presented in ILAS 2015 congress “Microstructure and properties of IN718 in samples manufactured using selective laser melting fabrication”.
New dissemination will take place using the final information obtained in the last months of the project. A paper is scheduled for HIP’17 to be held in Australia in December 2017. Two papers are under preparation to be published in Journals. The PhD Thesis of Mr. Javier Cortes is under preparation. Public defense is scheduled at the end of 2016. Finally, dissemination on the media is also going to be carried out pointing out the most important achievements.
Moreover Nesmonic project was presented as invited conference in the “XI TECHNOLOGY CONFERENCE Gas Turbines Technologies for the next decade”. The title of the presentation was “Net Shape Hot Isostatic Pressing for Ni Superalloys; toward an improved material usage”. Madrid 2014.
Nesmonic project was presented in a Workshop held in Bilbao (Spain, October 2013) at Aerotrend.
First Work shop ITP_Technology Centres. Nesmonic project was presented and discussed. Lerma (Spain) April 2015.
Exploitation of results
According to the objective of the NESMONIC Project, during the Project, a Single Products (SP) has been already developed and identified as NSHIP.
This SP, in turn, has been divided in several Intermediate Single products (ISP) qualified as Joint works B (JWB) that means that contributions of each Party are interdependent that is, when contributions although separately identifiable and exploitable, are intended to go together in a Single Product. The ISP are composed of several contributions of the Parties divided into independent works (IW) and Joint works A that means that Contributions of Parties are inseparable that is, they are not separately identifiable.
According to the Consortium Agreement section 8, contributions (IW) and (JWB) shall be the property of the Party carrying out the work generating that Foreground.
Concerning contributions JWA, they are identified in section 8.1 and especially in section 8.1.2 of the Consortium Agreement that provides for them the following default joint ownership regime:
(a) Each of the joint owners shall be entitled to Use their jointly owned Foreground on a royalty-free basis, and without requiring the prior consent of the other joint owner(s), and
(b) Each of the joint owners shall be entitled to grant non-exclusive licenses to third parties, without any right to sub-license, subject to the following conditions: at least 45 days prior notice must be given to the other joint owner(s) concerning licenses identification and direct or indirect compensation fees given in return (if any); and fair and reasonable compensation must be provided to the other joint owner(s).
Considering the nature of the Contributions identified, Patents applications are recommended.
Exploitable foreground (EF) obtained in the project has been identified and reported in Deliverable 1.7. Final report on Intellectual Property and Foreground. It has been stablished: the type of exploitable foreground, a description of the EF, the exploitable product, sectors and applications, timetable for exploitation, commercial or any other uses, patents or other IPR exploitation (licences) and finally owners or other beneficiary(s) involved.
The identification of IPR Qualifications of the exploitable foreground has been carried out together with the parties ownership percentages to components/contribution in ISPs. Those ISPs are related with:
• Atomization
• Encapsulation
• HIP modelling
• HIP processing window
Also the parties ownership percentages to FSP have been identified.
Related with the market, there are a number of competing technologies. It is possible to forge and machining the HIP canisters, however, these have a longer fabrication time, are of higher cost, have geometric limitations, use more energy and produce more waste. Additionally, the component could itself be forged and machined. In this case there is a poor fly-to-buy ratio, machining is greatly increased leading to a resource and energy inefficient process resulting in high cost.
About patents, the consortium is unaware of any existing patents that would negate freedom to operate.
Furthermore, the research partners are concerned with gaining further knowledge and understanding into the technically challenging process of producing high performance aero Inconel 718 parts by powder HIP (PHIP). The objectives of UoB and CEIT was to develop understanding and control of the PHIP processing conditions for Inconel 718. In addition the UoB developed a model which may be used to predict compaction of metal powders during the HIP process. The objectives of the MTC are to develop and exploit new high value manufacturing processes in close collaboration with industrial partners. The MTC is extremely keen to widen the application of powder HIP which is restricted due to the high cost of the sacrificial tooling. In NESMONIC the MTC is given the opportunity to develop a new generation of low cost tooling for powder HIP.
The overall market size according with Boeing estimations for aircraft casings is 62,644 from 2015 to 2034. Assuming an increasing penetration of the casings market by HIP technology from 16% to 95% from 2017 to 2034 and a consortium market share of 25% of the total HIP casings and a prize of 50.3K€ per casing the income will rise to 842.2 M€ that with a profit of 15% gives 126.3 M£. More complete figures are available in D1.8 Nesmonic exploitation plan.
It is expected that the NESMONIC process would benefit the manufacture of high-value components enabling a wider uptake of HIP across a range of industries including oil and gas, power generation, and civil nuclear.
Based on an investment of 1.88 M€, the 5 and 10 year average ROI are respectivwely of 64 % and 167%.
To exploit this market, the collaborators in this project have worked together to overcome the significant technical challenges highlighted in version 1 of the exploitation plan (Smith, 2013). The technical risks highlighted in version 1 and the work packages they are addressed in are:
• Failure to establish processing conditions for IN718 which yields components with acceptable mechanical properties (WP3)
• The tooling methods investigated in the project are unable to provide acceptable performance in a HIP cycle (WP4)
• The NSHIP process cannot be made cost effective (WP4)
List of Websites:
The website of Nesmonic project is:
http://www.tecnun.es/nesmonic/
It has been intensively used during the project between the partners mainly to share the documents and results generated.
Relevant contact people in Nesmonic project are:
Coordinator / Project Manager
Is the responsible of the management of the project and will be the highest authority in the Project and shall perform all task assigned to it as described in the EC-GA and in the CA.
✓ Iñigo Iturriza: iiturriza@ceit.es
Intellectual Property Officer (IPO)
IPO will assist the Steering Committee, the Scientific Committee and the Coordinator on all questions relating to IPR.
✓ Isabel Hernando: isabel.hernando@ehu.es
Project management Office (PMO)
It is mainly in charge of the Day-to-day administrative decisions.
✓ Iratxe Irizar: iiotaegi@ceit.es
Steering Committee
The steering Committee is constituted by one member per partner. Members are:
• Iñigo Iturriza (Ceit-IK4) iiturriza@ceit.es
• Moataz Attallah (UoB) m.m.attallah@bham.ac.uk
• Ross Trepleton (MTC) ross.trepleton@the-mtc.org
In addition, important people related to the project development to be contacted are:
• Miren Aristizabal (UoB) m.aristizabalsegarra@bham.ac.uk
• Charley Carpenter (MTC) Charley.Carpenter@the-mtc.org
• Javier Cortes (Ceit-IK4) jcortes@ceit.es
The final result of Nesmonic project is a validated cost effective NSHIP manufacturing route for IN718 parts for aero-engine application. To obtain this result it has been necessary to optimize all steps of powder manufacturing route. Initially, a complete characterization of IN718 powders has been done and in close collaboration with powder producers a powder with the optimum properties has been designed, produced in industrial atomisers and bought. It is mandatory to work with a material with the adequate tap density, interstitials content, chemical composition, morphology and price. After that, it was investigated an outgassing procedure to encapsulate the material without any modification of as-atomised powder properties. In addition, the optimisation of HIP and HT window was essential to successfully achieve the microstructural and mechanical requirements. Nesmonic project has been able to analyse and optimize all these steps and using the optimum experimental conditions, the OEM requirements initially stablished have been successfully achieved although toughness, measured by Charpy tests is smaller than for wrought IN718.
Moreover, the other main advantage of Nesmonic project is the cost benefit of the NSHIP approach. For this reason, many different tooling methods have been studied in the project couple with the use of computational modelling to ensure that net shape or near net shape parts were produced, reducing machining operations. After the analysis of the different tooling methods a complex mid-scale and large demonstrator component has been manufactured and characterised using the optimum tooling methods to demonstrate the ability of Nesmonic project to apply the manufacturing route proposed in the fabrication of a real engine part (Low Pressure Turbine casing).
Finally, additive manufacturing technique (AM) has been also used in the project to fabricate the canister needed for HIPping the IN718 powder and produce the small demonstrator component. This part has been an interesting opportunity to introduce AM technology in the fabrication of engine parts, knowing the current advantages and limitations of this technology and their solutions.
Therefore, the main impact of Nesmonic is in the area of manufacturing. The impact in manufacturing is in both the reduced cost and improved eco-efficiency of the manufacturing process which requires less material and energy input than current manufacturing methods based on forging and machining. The improved Inconel 718 powder, tooling methods, which have been developed in the project, coupled with the use of computational modelling to ensure that net shape parts are produced in the minimum number of iterations, have significant bearing on the overall cost benefit of the NSHIP approach.
In conclusion, the main impacts of this project are:
• Development of a new manufacturing route based on the Hot Isostatic Pressing (HIP) of In718 gas atomized powders able to successfully achieve the mechanical and microstructural requirements of the OEM.
• Significant reduction in raw material and energy consumption during manufacture and performance of high performance IN718 parts.
• Ability to manufacture complex components, with an optimum accuracy due to computational modelling used.
Project Context and Objectives:
Nesmonic Clean Sky Project aims to manufacture static engine components from IN718 using Net Shape Hot Isostatic Pressing of powder (NSHIP).
Net Shape HIPping (NSHIP) of powder is being increasingly used to produce components for a number of industrial applications, including aerospace. This process has the clear potential to vastly improve the ‘buy-to-fly’ ratio of large aero-engine components, when compared with traditional forging and machining. By using NSHIP ‘buy-to-fly’ ratios can be reduced from 10:1 to close to 1.5:1 leading to a significant reduction in material waste, lead time and associated costs.
However, there are several concerns that need attention before Ni superalloys components can be produced using NSHIP.
The NSHIPping is a process used to manufacture near net shape components by HIPping powders (most of the time gas atomized) in an accurately machined can. The mold tooling (can) is sacrificial (only one use) so that its cost has a large impact on the final cost of the process. Furthermore, NSHIP is a batch process, which limits its applicability to components where only hundreds are required per year and this limited number of components means that the cost of iterations is not spread over thousands of components. In addition, NSHIP requires computer-modelling of the tooling since the powder packing density is typically 60-65 % and the response of the powder and of the tooling during the compaction of the powder to 100% density must be modelled if a component of the desired geometry is to be produced. The net shape component is obtained after removing the tooling through machining/pickling and a truly net shape processing is obtained only if minimal subsequent machining is required. In addition to the accuracy in dimensions, the interaction between tooling and powder during HIPping has to be minimized.
In terms of the microstructure of the final powder based component, prior particle boundaries (PPBs), which are mainly decorated with stable oxide and brittle metal carbide precipitates in Powder Metallurgy HIP-processed superalloys, have been reported and compromise some critical mechanical properties. These have been attributed to oxygen-induced surface contamination of pre-alloyed powders. It should be noted that PPB networks are hard to eliminate once they have been formed. The presence of PPBs is particularly difficult to avoid in PM IN718 material due to its alloy chemistry and in addition to oxides and carbides, the formation of detrimental phases like Laves and delta (δ) have been reported in HIPped powder samples. These phases are detrimental to the hot ductility, fatigue and stress rupture properties. To control the PPB networks and the formation of detrimental phases, it is vital to optimise the HIPping route process parameters of pre-alloyed powder and to use powder with minimal interstitial content.
In this context of potential benefits and concerns, the main objective of the project was to develop and validate a cost-effective NSHIP manufacturing route for IN718 parts for aero-engine application.
To reach that objective the overall approach used in the project was to gather key OEM requirements, undertake a comprehensive experimental programme covering HIP processing of IN718 and can- production, supported by modelling, together with the final manufacture of demonstration parts.
The project activity was split into 5 Work Packages (WPs):
WP1 Project Management
WP2. Current limits of NSHIP technology and OEM requirements
WP3 Development of NHIP parameters for IN718
WP4 Can design, manufacture and testing
WP5 Demonstration component manufacture and assessment
WP1 provided the framework for effective management of the programme, including establishing the management team for the project (programme, technical and quality), and the reporting schedule and mechanism.
In WP2 information on the component performance requirements (target material properties, geometrical accuracy, surface finish etc.) was gathered from the OEM. The current limitations of the NSHIP approach was defined based on the available literature and results of previous work undertaken by the partners in the programme. The material usage was predicted based on both the tooling and finished components. The final objective of this WP was to select the demonstration components to be produced within WP5.
In WP 3 the response of IN718 powders (gas atomised, plasma atomised and PREP) to consolidation by HIPping was explored. The acceptable processing window for IN718 powder was established, based on the density/micrographs of simple cylindrical samples produced by CEIT. Further refinement and optimisation of the processing conditions was undertaken using larger samples produced by UoB which will be machined by MTC to provide the key machinability data. Samples were subjected to intense mechanical and microstructural characterization (CEIT/UoB) in as-HIPped and after heat treatment condition.
As soon as the optimum processing conditions were established samples were produced and supplied to NADCAP approved labs to allow sufficient time for a comprehensive range of material testing, including tensile testing, low cycle fatigue, crack propagation, fracture toughness, stress rupture and creep strain.
The experimental programme conducted in WP3 was focused on identifying the optimum HIP conditions so that samples of 100% density were produced. IN718 is one of the most difficult alloys to process by NSHIP. As with many Ni super alloys it is very difficult to get crystal growth to occur across Prior Particle Boundaries (PPBs) because these tend to contain high interstitial contents (commonly in the form of carbides and oxides). Work carried out in the project has been able to solve the problem induced by PPBs by a right combination of powder and HIP window.
The development of lower cost tooling by MTC in WP4 commenced early on in the programme. Test tooling manufactured in WP4 was used in the HIP trials conducted in WP3 and this permitted the particular characteristics of the new tooling methods to be understood. The data generated (in particular powder consolidation together with compressive flow stress data obtained on partially HIPped samples) was used to help perfect the computation model being developed for accurate prediction of the required tool geometry. The computer model was used to calculate the required can geometry to produce the two demonstration components specified by the OEM in WP2. Using this information MTC and UoB designed and manufactured the cans based on the laminated and AM-HIP methods developed in WP4.
In WP5, MTC and UoB produced the demonstration parts, selected in WP2, using the most appropriate powder (size and morphology), optimum processing conditions and can manufacturing methods, developed in WP3 and WP4. The demonstration parts were subjected to extensive evaluation including NDT, geometrical measurement, microstructure, and were also reviewed by the OEM.
In summary, the project has established the HIP processing window for IN718 powder and has explored new low cost canning manufacturing methods development. Then, using optimised parameters, components have been produced and assessed to see if they meet the OEM requirements in terms of accuracy, integrity and mechanical properties.
Project Results:
Main S&T results are described below organized by Work Packages.
WP1 Project management, coordination and dissemination of results
The objective of the project management Work Package (WP1) has been to manage the project ensuring that deliverable deadlines were met, that progress reports were submitted on time and that periodic meetings and discussions were held between the partners. In addition, management of exploitation (including IP issues) and dissemination of the results (including review of publications) resides within this WP.
The project has been relatively simple to coordinate and manage since there was a clear beneficiary of final product (Topic Manager) and executors of work (RTD partners) and the proposed management structure looks as follows:
There was a single point of contact between the Project and Topic Manager – the Programme Manager. He was responsible for all interactions with the EU Topic Manager and communicating issues both to and from the project team. The internal project was controlled by a Project Manager who oversees the day-to-day running of the technical aspects of the project and reports to the Programme Manager.
The Project Manager was supported by a senior technical advisor responsible for Technical Oversight of the project. In addition the Quality Manager was responsible for all aspects of QA in the project. As the Project Manager is at CEIT the teams at UoB and MTC had a local manager to ensure smooth running of the project at these organisations. Finally, depending on the local manager every RTD partner had a project team responsible for the execution of the project.
In terms of quality assurance, the MTC Business Management System (BMS) is a customised, web-based system used by the organisation to control and monitor core business processes. The BMS is MTC’s Quality Assurance program for projects and has been specifically designed for MTC using best-practice learnt across many organisations and industries and to meet the needs of MTC members, research partners and funding organisations. Project managers at MTC are supported by both an Administration Team (responsible for financial tracking, management etc.) and the Project Management Office (PMO). The PMO is responsible for defining the project management processes, producing tool and templates as well as supporting the project managers with help and advice on project server and any other issues.
An Exploitation Manager has been appointed who took responsibility for drafting an exploitation plan, within the first 6 months of the project and revised this plan in the last 6 months of the project to reflect the increased knowledge which has arisen from the project. In addition, this increased knowledge has allowed publishing and participating in a wide variety of congress.
• J. Cortés, M. Aristizabal, M.H. Loretto, M.M. Attallah, I. Iturriza, “Effect of HIP parameters and particle size distribution in the CSL and Twin boundaries of In718” October 2016 PM World Congress 2016, Hamburg.
• M. Aristizabal, J. Cortés, M.H. Loretto, M.M. Attallah, I. Iturriza, R.H.U Khan. “Study of the influence of outgassing parameters, powder particle size and HIP temperature in the mechanical properties of IN718”, 4-7 October 2015.EURO PM2015, Reims, France.
• I. Iturriza, K. Essa, J. Cortés, F. Castro R. Trepleton, C. Carpenter, R.H.U Khan, M.M. Attallah, M.H. Loretto “Net Shape HIPping components of IN718” 9-13 June 2014. HIP’14, the 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden.
• R.H.U Khan, M.H. Loretto, M.M. Attallah, J. Cortés, I. Iturriza, F. Castro. “Microstructure and properties of HIPped Inconel 718” 9-13 June 2014. HIP’14, the 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden.
Web site of Nesmonic was created at the beginning of the project. This share point has been used to share different documents between partners and topic manager, such as deliverables, minutes of the meetings, drawings, reports...
During the three years of the project meeting have been held at least once by month by WebEx and each three months, approximately, face to face meeting have been taken place to update the progress carried out during this period of time. Minutes of each of these meeting have been written, sent to each partner for any modification and submitted to Web site share point of Nesmonic.
Webex dates were:
22nd July 2013; 27th October 2013; 12th December 2013; 12th December 2013; 29th April 2014; 5th November 2014; 4th December 2014; 26th February 2015; 26th March 2015; 29th May 2015; 24th June 2015;2nd October 2015; 15th December 2015; 26th April 2016
And face to face meetings were:
Quick off meeting was held the 29th May 2013 at CEIT; 15th October 2013, UoB; 13rd March 2014, MTC; 4th July 2014, ITP; 7th October 2014, CEIT; 27th January 2015, UoB; 22nd April 2015, MTC; 23rd July 2015, CEIT; 27th October 2015, UoB; 23rd February 2016, MTC; end of project meeting was held the 31st May 2016 at ITP Zamudio (Spain).
WP2 Current limits of NSHIP technology and OEM requirements
The objectives of the WP2 were:
• To highlight the current limits of NSHIP.
• To define the geometry and properties of the components which will be manufactured in this project.
• To predict the material usage for the new approach.
At the beginning of the project the NSHIP limitations were explored. Net Shape Hot isostatic Pressing (NSHIP) is a technology for compacting metal powder to full density to a near net shape component in one step, by simultaneously applying a very large isostatic pressure and high temperature. NSHIP requires computer-modelling of the tooling since the powder packing density is typically 65 % and the response of the powder and of the tooling during the compaction of the powder to 100 % density must be modelled if a component of the desired geometry is to be produced. The net shape component is obtained after removing the tooling. Prior work in this area are in the literature but especially in the case of some Ni alloys specific developments are required if NSHIP is to be a cost-effective process for routine production of reliable aerospace components.
The most obvious limitation of current NSHIP technology is the fact that tooling, which has to be produced to the same accuracy as is required for the final component, is sacrificial. A second limitation of current NSHIP processing is the accuracy of the computer-modelling, since if the shape required is not predicted at the required accuracy a second or even third iteration or post-HIP machining will be required. It is important therefore that computer-modelling is improved, so that numerous iterations and extensive post-HIP machining are not required. In addition, in order to make the process more cost-effective it is desirable that either low cost tooling or re-usable tooling is developed.
Other limitations of NSHIP are more concerned with the material than with the process. Thus the feedstock is powder and this raises several difficulties in developing a process which has the same reliability as the conventional forging/machining route. In the case of IN718 in particular the surfaces of the powder particles are known to be decorated by carbide particles which form on oxide particles and these particles tend to inhibit bonding between the particles during HIPping. There are therefore two conflicting requirements; finer powder size fractions are preferred in order to exclude any large foreign particles in the feedstock, but larger size fractions are preferred to reduce oxidation and thus in the case of IN718 improving interparticle bonding during NSHIP.
In conclusion, NSHIP is a well-investigated process-route for production of components up to about 1 m diameter, but currently there are several limitations which are addressed in the current project. These are:
(i) The cost of the sacrificial tooling.
(ii) The accuracy of the modelling of NSHIP.
(iii) The role of particle size and oxides or other particles in limiting bonding during NSHIP.
The second goal of this WP was to define the geometry and properties of the components which will be manufactured in this project. Demonstration parts to be manufactured in the project were LPT casing and Tail bearing housing. In addition, components requirements, microstructure and mechanical properties have been also supplied by the OEM.
The third and final objective of this WP2 was to define material usage in the project on both tooling and finished parts. Raw material in the project was IN718 powder and it represented a large expenditure. It was then necessary to make an estimation of the powder utilization over the project. Powder was mainly used in: 1.- HIP parameter development, 2.- low cost tooling development, 3.- machined tooling development, 4.- Demonstrators production, 5.- Machinability study and 6.- NADCAP tests. In addition, mild steel powder has been needed and has been bought during the third year for the production of the canister by SLM for the SDC.
Powder bought during the Nesmonic project was as high as 1,104 kg. It was produced by gas atomization, plasma atomization and PREP. 7 different supplier were tested. Interaction to increase powder characteristics was more intense with two of them.
Deliverables related with this WP2 have been successfully completed during the project. T There was also a Milestone associated with WP2.
WP3 Development of NHIP parameters for IN718
The primary objective of this WP was to determine the optimum HIP processing parameters for IN718. The work involved an assessment of the interstitial content and microstructure of powders, assessment of their flowability, composition, powder-filling and response to HIPping using a range of HIP conditions for the production of samples for mechanical testing, machining trials and to identify HIP conditions for the large demonstrator. In addition, the WP contained several other objectives; assessment of the machinability of HIPped samples, test specimen design and NADCAP tests.
• Powder characterisation and response to HIPping
Initially, a complete characterization of different IN718 powders was made. Six different powder suppliers were analysed in this work to evaluate all commercial available IN718. In addition, three diverse types of atomized pre-alloyed IN718 powders were down selected, so an analysis of difference between these processing technologies took place (plasma atomization – PA, gas atomization – GA, and plasma rotating electrode process – PREP). Finally, a wide range of fraction sizes were employed to determine the best fraction powder of work. This wide variety of powder parameters analysed allows guaranteeing the selection of optimum powder to achieve the mechanical properties required.
A wide variety of studies of the diverse powders were carried out. This activity was really important due to the properties of the powder have a strong influence on the mechanical properties of the final material. Therefore, it was completely necessary to successfully achieve the mechanical properties required that the powder selected had the following properties:
▪ Flow rate: If the powder has a high flow rate, the filling of the canister will be easier and all cavities will be filled with powder.
▪ Tap density: High tap density improves the filling of the canister and especially allows better control the shrinkage of the canister during HIP cycle. The accuracy of the modeling of the densification will be also improved for high tap density. Intensive activity was carried out during the project to obtain a powder with a high tap density.
▪ Oxygen and carbon content: Low contents of these elements decrease the presence of oxides and carbides in PPBs and thus, improve the mechanical properties of the material.
▪ Morphology: The presence of flakes, satellites and aggregates in the powder is really bad to achieve the required flowability, tap density and mechanical properties. To avoid this behaviour, manufacturing process has to be optimised to obtain clean powders.
▪ Fraction size: Particle size distribution has been defined in order to get a compromise between tap density and flowability. In addition, the manufacturing process of the powder was also optimized to obtain the highest amount as possible of the fraction of interest (yield), so the price of the powder will be cheaper
To obtain a powder with these optimum properties, analysis of morphology, interstitials, chemical composition and physical properties have been carried out. After these studies, the optimum powder for Nesmonic project has been selected.
Cross sections of particles show that GA powder particles commonly contain pores associated with trapped argon gas. Some pores are also visible in the PREP powder.
Density measurements, thermal analysis and phases determinations by X-ray diffraction have been also done to complete the characterization and ensuring the quality of IN718 powders.
Finally, heat treatments on powders have been carried out trying to decrease oxygen content to promote a reduction of oxides and carbides at PPBs and thus, improving the mechanical properties.
Differential scanning calorimetry (DSC) of the GA powder indicated several reactions during heating/cooling. These are related to the dissolution/ evolution of γ´´, γ´ precipitates as well as niobium carbide (NbC) particles, solidus and liquidus temperature.
All these were used to select HIPping and heat treatment cycles.
ENCAPSULATION:
The encapsulation process of the powder in a canister is really important in this new developed technology. It has a strong influence on the mechanical properties of the final material. During canning and degassing it is mandatory to evacuate gas trapped in between powder particles and avoid interstitials, especially oxygen, increasing in the powder. For this reason, a new outgassing procedure has been developed during Nesmonic project which is able to successfully keep interstitials as the level of as-atomized powder. In addition, a purging system to decrease the oxygen content during degassing step has been investigated. This new technique is able to reduce the oxygen content of as-atomised powders and its influence on mechanical properties was evaluated.
HIP and HT:
Finally, a wide variety of parameters has been studied to optimize the HIP and HT window, such as, HIP temperature, HIP pressure, HIP time, fraction size of powder, interstitials content, effect of oxygen reduction during degassing, chemical composition, manufacturing technology of the powder, different post-HIP HT, HT and different powders. After this complete study, the optimum experimental conditions have been selected to analyze the mechanical and microstructural properties of the material. Microstructure of the material without the presence of PPBs has been observed. Mechanical and microstructural OEM requirements have been successfully achieved, except for non-initially considered Charpy toughness measurements, the reason of the low value of toughness was found and a solution was proposed.
In addition, an interaction between the canister and IN718 during HIPping was detected. Diffusion from the canister increases with increase in temperature. This diffusion layer (~200-300 μm) has to be removed as it can affect the mechanical properties of net shaped components. The thickness of this interaction layer was decreased by using diffusion barriers in the internal wall of the canisters.
Finally, the effects of HIP temperature and post-HIP heat treatments on the % of Σ3 twin boundaries have been studied. In both cases, as-HIP and HIP plus HT, the fraction of Σ3 boundaries was related to the HIP temperature. Thus, the fraction of Coincidence Site Lattice (CLS) was higher at higher HIP temperatures and increased further on heat treatment. In addition, it was observed that the increase of Σ3 boundaries improved the ductility of the material. However, at the first stage of melting (T4) the fraction of Σ3 boundaries and the ductility decreased.
Conclusions
The powder production and raw material used are quite critical for IN718 and interstitials must be kept low to limit PPB formation. The degassing strategy helped in restricting oxides on powder particle surfaces. Finally, an optimum HIP and HT window has been obtained and allowed successfully achieving the mechanical and microstructural OEM requirements, except toughness.
• Machining of HIPped IN718
For the study, a roughing and finishing insert was used at a variety of conditions in order to find the optimum cutting parameters. This was done by looking at the subsurface microstructure, subsurface microhardness, surface finish, tool wear, and analysis of the swarf. For the majority of samples, microhardness did not change between 50 and 500 µm below the surface, however the hardness of the roughing HIPped sample decreased slightly (from 370 HV to 240 HV) between 50 and 250 µm below the surface.
Comparison of optimally HIPped IN718 against forged material:
This section of WP3 was aimed at determining the best generic turning parameters whilst comparing the machinability of forged and HIPped samples in terms of the most appropriate tool material; surface finish achievable and tool wear.
Roughing inserts:
Forged material – very little change in tool wear was evident over the time the inserts were used for (2.3 minutes). Swarf analysis revealed that cutting efficiency reduced with higher feed rates and cut depths.
HIPped material – ‘crater wear’ was consistently observed on most of the tool tips. The swarf for all cuts were all of similar size.
For both materials, no correlation with cutting speed was observed.
Finishing inserts,
Forged material – slight correlation between surface finish and cutting speed was found. The best surface finish (0.50 µm) was achieved at the highest cutting speed. No correlation between tool wear and cutting speed was observed.
HIPped material – no change in surface finish was observed as cutting speed was increased. Tool life appeared to decrease for higher cutting speeds
Overall it appears that the forged material has better machinability than HIPped material. Less tool wear was evident, and reduced work hardening. A better surface finish could also be achieved on the forged material, with an average of 0.95 µm in comparison to 1.27 µm for the HIPped material (finishing passes).
Subsurface Microstructural Features and Mechanical Damages
The grain structure of the HIPped sample was more homogeneous than the forged sample For all passes on the forged samples, no mechanical damage was found in the subsurface at magnifications of 50X, 100X and 200X.
Subsurface microhardness
Work hardening of the deformed layer beneath the machined surface causes a higher superficial hardness than the average hardness of the bulk material. The HIPped material experiences work hardening in the first 0-150 µm. The forged roughing material, however, experiences the opposite, with a lower microhardness in the first 0-150 µm before increasing after 150 µm.
Tool wear
There are many factors which affect the cutting tool wear, including the tool geometry (flank angle, rake angle, cutting edge angle), geometry of workpiece, cutting parameters (speed, feed, depth of cut). The lack of work holding of the HIPped roughing rod noted above had a dramatic effect on the tool life due to the vibration of the rod. The HIPped finishing rod’s work-holding was modified to have the same percentage work holding as the forged finishing rod. As the speed of the tool increases, the cutting temperature inevitably increases, which leads to a slight increase in flank wear on the insert. The increase in feed rate, depth of cut and the tool speed of the roughing cut did not show signs of an increase in tool wear for the forged material.
Surface Finish
Surface roughness plays an important role in a part’s accuracy and service life. Many factors have an impact on the machined surface roughness, namely cutting parameters, tool parameters, and processing parameters. In this study, the tool parameters and processing parameters were kept constant and only the cutting parameters were varied. Only the finishing passes were analysed.
The roughing HIPped billet yielded the highest surface roughness results, with an average value 81% higher than that of the roughing forged billet. It also presents the widest range of values, with a minimum Ra of 1.39 µm and a maximum Ra of 6.13 µm. For the finishing operation, the forged billet has the smallest average Ra value at 0.95 µm which is 30% smaller than the average value for the HIPped billet.
Swarf Analysis
The increase in feed rate and cut depth gave smaller chip sizes for the roughing passes of the forged rod, indicating than a lower feed rate and a lower depth of cut gives better cutting efficiencies. The same correlation was also present within the roughing pass of the HIPped rod, however it was less apparent.
Overall it appears that the forged material has better machinability than HIPped material. Less tool wear was evident and less work hardening of the material. A better surface finish could also be achieved on the forged material, with an average of 0.95 µm in comparison to 1.27 µm for the HIPped material.
Machining of Representative Features for Full Size Casing
Representative features for full size casing were primary machined in the subscale casing.
Features to be machined
Four features were chosen to be machined. These were: A. A top flange undercut, B. A thin wall, C. A rib and D. A bottom flange undercut. To machine the bottom of the casing, the top of the casing was clamped to the spindle. A spider fixture was also attached to the inside of the casing in order to stop the casing from warping.
In summary the most important factor affecting the tool wear in this study was the work-holding. For the HIPped roughing sample, the smaller work holding used caused the rod to experience a much higher vibration, which resulted in many inserts chipping and breaking.
The four features identified, were successfully machined into a sub-scale casing. Due to the nature of the HIPping process, a uniform wall thickness is unachievable and hence casings will inevitably have a run out. In order to reduce the amount of rubbing of the insert against the material, an increased feed rate should be used in order to decrease this time.
• NADCAP tests
Design of test samples–engineering drawings defining the specimens for NADCAP tests were supplier by the OEM.
NADCAP tests performed by Westmoreland
Results:
(i) Charpy tests. Using the optimum conditions for HIPping the Charpy values have been determined at RT and ET for HIPped and forged materials after post-HIP heat treatments. The results obtained are indicating higher Charpy values for forged IN718 than for HIPped HT grade. At both temperatures, RT and 650 °C.
A wide range of post-HIP heat treatments has been carried out in which the Charpy values of solution-treated and of samples aged at different temperatures have been assessed. The most important observation made in these extensive tests is that the Charpy value of as-HIPped samples exceeds that of forged sample and that most post-HIP heat treatments downgrade the Charpy values.
ii) Stress rupture. Using the optimum experimental conditions developed in the project (powder + HIP + HT) the stress rupture at elevated temperature has been analysed. The results obtained successfully achieve the OEM requirements and are better than typical forged material and clearly improve on the values reported in the literature for HIPped samples.
iii) High cycle fatigue (HCF). The values obtained in this tests successfully achieved the OEM requirements. They are even better than typical forged material and clearly improve on the values reported in the literature by this material developed by HIP.
Conclusions from mechanical testing work
NADCAP tests carried out have demonstrated that the OEM requirements are achieved for stress rupture and high cycle fatigue properties of IN718 material developed by the new HIP powder processing route. In addition, this material was analyzed and also achieved the requirements for grain size, microstructure, hardness and tensile properties. However, Charpy test results obtained with HIP + HT samples are below the OEM requirements.
• Overall conclusions for WP3
The objective of this WP was to determine the HIP processing parameters so that optimally processed samples were produced which were then available for mechanical testing, machining trials and production of a large demonstrator.
It has been shown that HIPped samples could be produced which had room temperature and elevated temperature tensile and stress rupture properties, which reached the specifications for forged samples. However, the post HIP heat treatments which were used to achieve these properties very significantly downgraded the excellent Charpy values of the as-HIPped samples. Further detailed microstructural work, involving transmission electron microscopy was carried out in order to understand the microstructural changes occurring during heat treatments which lead to this unacceptable decrease in Charpy values. The conclusion of this characterization is going to be tested out of the active period of the Nesmonic project. The machinability of as-HIPped samples was carried out successfully. The optimum HIP conditions identified were used in the WP5 aimed at producing a large demonstrator.
WP4. Can design, manufacture and testing
The primary objective of this work package (WP4) was to optimise cost-effective can design using novel low cost manufacture methods and to investigate a hybrid AM-HIP method to produce tooling for smaller components. This work also encompassed development of a computational model which enable the geometry of tools to be calculated to enable net shape components to be produced. There were six deliverables in this work package:
1. Tools for WP3 trials
2. Laminated test tools and report on performance
3. AM-HIP test tools and report on performance
4. Validated compaction model
5. Cost evaluation of tooling options
6. Design of demonstration component tools
• Tools for WP3 trials
The major obstacle with the current NSHIP approach is the cost of the tooling. The partnership worked to lower tooling production cost by developing two cost effective canning solutions one for large structural components and the second one for the production of small components.
• Low cost tooling
This report assessed three routes to produce low-cost tooling to produce simplified HIPped In718 casing component to 100% density and appropriate form. The three routes were:
• Machined HIP Tooling
• Sheet metal spinning HIP Tooling
• Laminate tooling
The Machined HIP Tooling subscale component, consolidated symmetrically and in line with the UoB compaction model and showed no indication of failure to consolidate. The component wall thickness was found to be uniform and no local geometric artefacts were observed.
The Sheet metal spun HIP canister passed He leak detection and was also successfully HIPped. However, post HIP, it was found to have a significant reduction in diameter and height. Further iterations involved the use of laminate and solid mandrels. These iterations were able to hold the shrinkage in diameter. Nevertheless, they were unable to hold the shrinkage in height. With the modelling able to model the shrinkage in height, future iterations should be designed to take the height shrinkage into account.
For the laminate tooling, it was found that by the use of discreet fixturing between layers, the required canister wall thickness and cost of laminate HIP canisters could be significantly reduced. This concept of the Laminate tooling failed during the HIP process due to failure to contain powder between the laminates. However, coatings were used in the powder cavity for subsequent iterations of tooling design to prevent powder egress within laminate tools during vibration filling.
The analysis of the three methods conclude that machined canister is the more robust for the production of Large Demonstrator Component. A real LPT casing.
• AM-HIP test tools and report on performance
A novel hybrid route based on AM-HIP for the production of a mount lug has been developed. The tooling or shell was manufactured by SLM powder bed using mild steel powder with optimised parameters. It was found that the quality of the powder was a key factor in obtaining a canister without porosity or cracks.
The As-built shell parts has to be heat treated for stress relief before removing the support structure and the substrate to avoid any distortion or bending. In addition, the original shell design was modified to ensure the lack of distortions.
The obtained near net shape canister was filled by IN718 powder, outgassed, crimped and HIPped. 3D shape measurement, using GOM system, was carried out at the MTC before and after HIPping to evaluate the shape change and the extent of deformation due to powder consolidation during HIPping. The mount-lug reduced its size homogenously in the whole area during the HIP cycle.
The Finite Element computational model of powder compaction and consolidation during HIPping, which was developed for the large demonstration, was modified from 2D to 3D (in a single run) for the small demonstrator or mount lug. The work shown in this section was use to fix the procedure of manufacturing an IN718 mount-lug by mean of hybrid SLM/HIP process. The fixed procedure and the final mount-lug properties are described in WP5.
• Validated compaction model
An important stage of the proposed approach was the comparison of the predicted and the measured geometry, which was extracted by optical 3D scanning system (GOM scanner). The comparison between the measured tooling geometry and that predicted by the FEA model concluded that the axial shrinkage of the powder is larger than the radial shrinkage. The maximum reduction in height and maximum reduction in diameter were measured and considered acceptable. The axial shrinkage is uniform along the radial direction while the radial shrinkage was less uniform along the axial direction. Less radial shrinkage was seen at the upper and lower corners of the inner tooling. This discrepancy may be due to a number of factors, such as the directional stiffness variation of the mild steel tooling (the stiffness in the axial direction is different from that in the radial direction) and the non-uniform pressure distribution during HIPping. The reduction in the internal diameter is larger than that in the external diameter. This is due to the fact the thickness of the internal tooling is much smaller than that of the external tooling thus the pressure is transmitted more easily through the internal surface.
In general, there was a very good agreement between the geometry predicted by the FEA model and the measured one. The maximum deviation between the FE profile and the experimental one was about 0.5mm which is found at the upper and lower corners of the outer tooling at which maximum radial shrinkage is taking place. Although the material model that has been used is relatively simple, very good prediction of the final tooling geometry is achieved. The simulation time of this model is 10 min using a single processor, which represents quite fast computational time as compared with other complicated models.
• Cost evaluation of tooling options
The Casing Cost Evaluation Tool captures all the costs associated with manufacturing a HIP tooling through three manufacture routes: Machining, Spinning, and Laminates.
The Casing Cost Evaluation Tool also contains a Generic Casing Cost Calculator, which is based on the costings of the LPT Casing. This calculator gives the cost of manufacturing a generic casing based on five inputs:
1. Desired % overstock of cavity
2. Desired wall thickness of the final casing
3. Number of complex features
4. Quantity of canisters produced per year
5. Diameter of final casing
Having a greater overstocked canister has two advantages. Firstly it reduces the non-conforming geometry failure risk which occurs during HIP, and secondly, the overstocked design can reduce the complexity of the canister design and hence it would be cheaper to manufacture and the lead time would be reduced. The setback of having a greater overstocked canister is that more powder is needed and the cost of post machining will increase.
The cheapest route to manufacture a LPT casing has been analysed in between different methods and identified taken into account the entire production cycle.
• Design of demonstration component tools
The target component was a full height engine casing of a 1.5 m maximum diameter In718 engine casing. The objective of this casing was to demonstrate the ability to manufacture both net shape and near net shape features. The target desired geometry has four NS and four NNS features.
Design work was carried out to design a machined HIP demonstrator with net shape (NS) and near net shape (NNS) features. Simulated modelling of the HIPping process was also carried out on the design iterations by the University of Birmingham in order to optimise the design. The casing was geometrically assessed by the ATOS Triple Scan and the profile sketches of the casing was analysed against the desired geometry and also the inner tooling pre-HIP.
The finished casing is shown in the attached pdf.
Conclusions of WP4
• Low cost tooling was validated
• A novel hybrid route based on AM-HIP for the production of a mount lug has been developed
• Prediction of the compaction model was confirmed
• Full complexity 0.727 m demo accurate to high % value in both height and diameter
WP5. Component manufacture and validation
The objective of this work package (WP5) was to manufacture demonstration components using the most appropriate powder, HIP parameters, tooling methods and finish machining strategy, as determined from the results of the preceding work packages. The quality of these demonstration components has been assessed by the partners (NDT, metrology and metallurgy) and then they have been supplied to ITP for long term evaluation and to promote the use of the NSHIP approach.
Eight deliverables in this work package have been written:
1. Supply most appropriate powder for demonstration part based on the results of WPs 3 & 4;
2. Tool for small demonstration part;
3. Tool for large demonstration part;
4. Small demonstration part;
5. Large demonstration part;
6. NDT/Metrology reports for demonstration components;
7. Delivery of completed small NSHIP demonstration part to ITP for assessment;
8. Delivery of completed large NSHIP demonstration part to ITP for assessment.
• Most appropriate powder for demonstration part
Previous studies by different researchers show that the raw material can influence the net shape component properties, hence affecting the integrity of the final part. Therefore, it is vital to assess the pre-alloyed powders through an intensive characterisation work.
Tensile properties required by the OEM were only achieved with one of the gas atomised powder, PSD and optimum experimental conditions (PSD, HIP conditions, HT...). Many different effects and experimental conditions were analysed to achieve these properties and understand the behaviour of the material.
These results were obtained because in HIPped specimens, formation of PPBs, which are detrimental for mechanical properties, were avoided. Besides, it is also important that the powder has a high tap density to avoid distortions in the final demonstrator part.
• Small demonstration part
The original mount lug sample geometry was simplified by removing holes. The CAD model was scaled down to 40 %.
The tooling or shell for the small demonstrator component was manufactured by SLM powder bed using mild steel powder.
The near net shape canister was filled by IN718 selected powder, outgassed, crimped and HIPped. 3D shape measurement, using GOM system, was carried out at the MTC before and after HIPping to evaluate the shape change and the extent of deformation due to powder consolidation during HIPping. The mount-lug reduced its size uniformly in the whole area.
Two small demonstrator components (SDC) were manufactured during the project. Due to slight differences in the manufacturing route between the two SDCs, the first one showed better properties and it was characterised in terms of microstructure and XCT tomography. The first SDC was found to be fully dense but with crack especially in areas with stronger curvatures. The microstructure showed a large amount of Al and Ti oxides at the prior particle boundaries and NbC precipitation inside the particles because the outgassing procedure developed in the project was not appropriated carried to the SDC. GOM was used to assess the extent of the shrinkage during HIP. Both SDCs show similar shrinkage in the whole body.
The Finite Element computational model of powder compaction and consolidation during HIPping, which was developed for the large demonstrator, was modified from 2D to 3D (in a single run) for the small demonstrator or mount lug.
• Large demonstration part
It was decided by the consortium that for the final full scale demonstrator, a large 1.5 m IN718 engine casing should be manufacturing through the NSHIP process.
The encapsulation process encompasses everything from the assembly of the canister to the state where the canister is ready for HIPping. This process has the following steps:
• Canister assembly
• Welding
• Weld inspection
• Powder filling
• Evacuation
• Crimping
Canister was machined in two parts and welded. Weld inspection was performed on the canister assembly using a helium leak test. The canister was evacuated for 20 minutes prior to the He leak. The test identified that there was a leak in the canister, however not at the welds. Holes were identified and then covered with 5 mm thick weld-on-patches.
After appropriated filling with optimum IN718 powder and long outgassing, the Canister was HIPped at Bodycote’s Lego HIP in Surahammar Sweden. However porosity was find in the HIPped component. The level of porosity was measured using image analysis software, giving a value of 9.3 % porosity.
A surface comparison of the GOM data of the post-HIP, pre-etched canister against the CAD was carried out. From this data, the following conclusions can be drawn:
• Overall the canister has shrunk 6.4% in volume
• The inner tool has moved inwards
• The outer tool has moved inwards
Cracks appeared around the top flange of the casing during the etching.
The geometrical analysis of the casing against the desired geometry was done. There was not enough shrinkage in the bottom of the casing whilst in the middle and the top, there was too much shrinkage.
The failure during HIP was most likely caused by porosity in outer (thin) canister material. Increasing the thickness of the outer casing would avoid the problem.
• NDT Results for Ø0.74 m Casing
An initial study was carried out to examine the potential NDT routes for the fully dense mid-sized demonstrator component. That component was HIPped with optimum powder at optimum processing window. Dye penetrant inspection, eddy current, magnetic particle inspection, radiography, ultrasound, and visual inspection were considered for NDT characterization. After a downselection, digital radiography was chosen as the most suitable method for detecting surface and subsurface defects, in combination with visual and/or dye penetrant inspection.
Digital radiography was used to assess the part integrity of the NS&NNS casing. The inspection showed no subsurface defects, however did show surface indications around one of the sectors. This, however, was confirmed by visual inspection to be residue canister, which was not completely taken off at the etching stage.
GOM was used to assess the geometrical accuracy of the casing and the scan was compared to the desired geometry. The scans showed that the casing shrunk more than expected, however it shrunk uniformly. The inner tool, which was designed not to move, horizontally also shrunk.
The surface roughness was also measured on 5 parts of the canister, with an average surface roughness of the as-HIPped surface being Ra: 2.41 µm.
Conclusions of WP5
Small and Large Demonstrator components have been HIPped with the optimum powder developed in the project. SDC was fully dense with some cracks from strong curvature areas. Large demonstrator with 1.5 mm in diameter was also HIPped but its density was compromised because of the interconnected porosity of the thin outer canister. However, mid-scale canister has proved that the encapsulated HIPping of IN718 gas atomized powder can produce large PM components free of defects (fully dense, no cracks, no PPBs) with net shape features.
Potential Impact:
Potential impact.
The final result is a validated cost effective NSHIP manufacturing route for IN718 parts for aero-engine application. Therefore, the main impact is in the area of manufacturing. The impact in manufacturing is in both the reduced cost and improved eco-efficiency of the manufacturing process which requires less material and energy input than current manufacturing methods based on forging and machining. The improved tooling methods, which have been developed in the project, coupled with the use of computational modelling to ensure that net shape parts are produced in the minimum number of iterations, have significant bearing on the overall cost benefit of the NSHIP approach.
It is also expected that this project will have an impact on the fuel consumption. In-service benefits will arise from the use of efficiently designed NSHIP components, which will perform better and weigh less, thus reducing fuel consumption. In addition, there will be significant fuel efficiency benefits to be gained from the increased use of high performance nickel alloys in aero engine construction which will be made possible if part manufacturing costs are reduced using NSHIP and final properties are as high as promised.
The potential impact of the project is quantified below;
• Significant reduction in raw material and energy consumption during manufacture and performance of high performance IN718 parts.
▪ 80% reduction in aerospace material consumption (nickel alloys) and elimination of swarf disposal / recycling costs. The reduction in the amount of material required would also lessen the effect of international material supply fluctuations.
▪ 75% reduction in energy-consumption during manufacture by reduction of energy-intensive machining operations.
• Ability to manufacture complex components which will provide the essential high pressure and temperature capability critical to achieve the required reductions in fuel burn in future large civil aero-engines. There is also a huge improvement in engine emissions to be gained with the use of hotter engine cycles enabled by employing high performance nickel alloys that will help the aerospace industry achieve tough ACARE targets such as a 50% reduction in CO2 emissions per passenger-kilometer by 2020.
In addition to the impact directly related with Nesmonic objectives there is also an extra benefice by the diffusion of the capacities of the global PM route NSHIP. The extensive use of NSHIP in aeronautic, energy, oil and gas and other industries will increase the impact of the project in terms of material usage, energy needed for production and even performance. Finally, the successful application of hybrid AM-HIP open the possibility of imaginative combination of technology in order to increase the possibility of producing complicated high performance component.
Dissemination.
Dissemination of the project has been taken place mainly in International Congress but also in some aeronautic forums and webs.
• NESMONIC webpage was released. The webpage can be accessed at [http://www.tecnun.es/nesmonic/]. It describes the project, main goals, methods as well as some issues of relevance for the technology itself.
• News release through CEIT’s webpage. Through CEIT’s electronic newsletter and bulletin, all interested companies and individuals have been informed about the project.
Oral presentations in International Congress have been:
• “Net Shape HIPping components of IN718”. HIP’14. The 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden. 9-13 June 2014
• “Microstructure and properties of HIPped Inconel 718”. HIP’14. The 11th International Conference of Hot Isostatic Pressing, Stockholm, Sweden. 9-13 June 2014
• “Study of the influence of outgassing parameters, powder particle size and HIP temperature in the mechanical properties of IN718”. EuroPM´15 Congress held in Reims, France, 4 – 7 October 2015.
• One oral presentation will be carried out in the WorldPM’16 Congress to be held in Hamburg, Gemany, 9 – 13 October, 2016. The title of the paper is: “Effect of HIP parameters and particle size distribution in the CSL and Twin boundaries of In718”.
• A poster was presented in ILAS 2015 congress “Microstructure and properties of IN718 in samples manufactured using selective laser melting fabrication”.
New dissemination will take place using the final information obtained in the last months of the project. A paper is scheduled for HIP’17 to be held in Australia in December 2017. Two papers are under preparation to be published in Journals. The PhD Thesis of Mr. Javier Cortes is under preparation. Public defense is scheduled at the end of 2016. Finally, dissemination on the media is also going to be carried out pointing out the most important achievements.
Moreover Nesmonic project was presented as invited conference in the “XI TECHNOLOGY CONFERENCE Gas Turbines Technologies for the next decade”. The title of the presentation was “Net Shape Hot Isostatic Pressing for Ni Superalloys; toward an improved material usage”. Madrid 2014.
Nesmonic project was presented in a Workshop held in Bilbao (Spain, October 2013) at Aerotrend.
First Work shop ITP_Technology Centres. Nesmonic project was presented and discussed. Lerma (Spain) April 2015.
Exploitation of results
According to the objective of the NESMONIC Project, during the Project, a Single Products (SP) has been already developed and identified as NSHIP.
This SP, in turn, has been divided in several Intermediate Single products (ISP) qualified as Joint works B (JWB) that means that contributions of each Party are interdependent that is, when contributions although separately identifiable and exploitable, are intended to go together in a Single Product. The ISP are composed of several contributions of the Parties divided into independent works (IW) and Joint works A that means that Contributions of Parties are inseparable that is, they are not separately identifiable.
According to the Consortium Agreement section 8, contributions (IW) and (JWB) shall be the property of the Party carrying out the work generating that Foreground.
Concerning contributions JWA, they are identified in section 8.1 and especially in section 8.1.2 of the Consortium Agreement that provides for them the following default joint ownership regime:
(a) Each of the joint owners shall be entitled to Use their jointly owned Foreground on a royalty-free basis, and without requiring the prior consent of the other joint owner(s), and
(b) Each of the joint owners shall be entitled to grant non-exclusive licenses to third parties, without any right to sub-license, subject to the following conditions: at least 45 days prior notice must be given to the other joint owner(s) concerning licenses identification and direct or indirect compensation fees given in return (if any); and fair and reasonable compensation must be provided to the other joint owner(s).
Considering the nature of the Contributions identified, Patents applications are recommended.
Exploitable foreground (EF) obtained in the project has been identified and reported in Deliverable 1.7. Final report on Intellectual Property and Foreground. It has been stablished: the type of exploitable foreground, a description of the EF, the exploitable product, sectors and applications, timetable for exploitation, commercial or any other uses, patents or other IPR exploitation (licences) and finally owners or other beneficiary(s) involved.
The identification of IPR Qualifications of the exploitable foreground has been carried out together with the parties ownership percentages to components/contribution in ISPs. Those ISPs are related with:
• Atomization
• Encapsulation
• HIP modelling
• HIP processing window
Also the parties ownership percentages to FSP have been identified.
Related with the market, there are a number of competing technologies. It is possible to forge and machining the HIP canisters, however, these have a longer fabrication time, are of higher cost, have geometric limitations, use more energy and produce more waste. Additionally, the component could itself be forged and machined. In this case there is a poor fly-to-buy ratio, machining is greatly increased leading to a resource and energy inefficient process resulting in high cost.
About patents, the consortium is unaware of any existing patents that would negate freedom to operate.
Furthermore, the research partners are concerned with gaining further knowledge and understanding into the technically challenging process of producing high performance aero Inconel 718 parts by powder HIP (PHIP). The objectives of UoB and CEIT was to develop understanding and control of the PHIP processing conditions for Inconel 718. In addition the UoB developed a model which may be used to predict compaction of metal powders during the HIP process. The objectives of the MTC are to develop and exploit new high value manufacturing processes in close collaboration with industrial partners. The MTC is extremely keen to widen the application of powder HIP which is restricted due to the high cost of the sacrificial tooling. In NESMONIC the MTC is given the opportunity to develop a new generation of low cost tooling for powder HIP.
The overall market size according with Boeing estimations for aircraft casings is 62,644 from 2015 to 2034. Assuming an increasing penetration of the casings market by HIP technology from 16% to 95% from 2017 to 2034 and a consortium market share of 25% of the total HIP casings and a prize of 50.3K€ per casing the income will rise to 842.2 M€ that with a profit of 15% gives 126.3 M£. More complete figures are available in D1.8 Nesmonic exploitation plan.
It is expected that the NESMONIC process would benefit the manufacture of high-value components enabling a wider uptake of HIP across a range of industries including oil and gas, power generation, and civil nuclear.
Based on an investment of 1.88 M€, the 5 and 10 year average ROI are respectivwely of 64 % and 167%.
To exploit this market, the collaborators in this project have worked together to overcome the significant technical challenges highlighted in version 1 of the exploitation plan (Smith, 2013). The technical risks highlighted in version 1 and the work packages they are addressed in are:
• Failure to establish processing conditions for IN718 which yields components with acceptable mechanical properties (WP3)
• The tooling methods investigated in the project are unable to provide acceptable performance in a HIP cycle (WP4)
• The NSHIP process cannot be made cost effective (WP4)
List of Websites:
The website of Nesmonic project is:
http://www.tecnun.es/nesmonic/
It has been intensively used during the project between the partners mainly to share the documents and results generated.
Relevant contact people in Nesmonic project are:
Coordinator / Project Manager
Is the responsible of the management of the project and will be the highest authority in the Project and shall perform all task assigned to it as described in the EC-GA and in the CA.
✓ Iñigo Iturriza: iiturriza@ceit.es
Intellectual Property Officer (IPO)
IPO will assist the Steering Committee, the Scientific Committee and the Coordinator on all questions relating to IPR.
✓ Isabel Hernando: isabel.hernando@ehu.es
Project management Office (PMO)
It is mainly in charge of the Day-to-day administrative decisions.
✓ Iratxe Irizar: iiotaegi@ceit.es
Steering Committee
The steering Committee is constituted by one member per partner. Members are:
• Iñigo Iturriza (Ceit-IK4) iiturriza@ceit.es
• Moataz Attallah (UoB) m.m.attallah@bham.ac.uk
• Ross Trepleton (MTC) ross.trepleton@the-mtc.org
In addition, important people related to the project development to be contacted are:
• Miren Aristizabal (UoB) m.aristizabalsegarra@bham.ac.uk
• Charley Carpenter (MTC) Charley.Carpenter@the-mtc.org
• Javier Cortes (Ceit-IK4) jcortes@ceit.es