A Novel Technique to Assess the Effect of Machining and Subsurface Microstructure on the Fatigue Performance of Ti-6Al-2Sn-4Zr-6Mo

4Zr-6Mo Daniel Suárez-Fernández1, *, Bradley P. Wynne1, Pete Crawforth2, Katharine Fox3, and Martin Jackson1 *DSuarezFernandez1@sheffield.ac.uk 1) Department of Materials Science and Engineering, The University of Sheffield, Sir Robert Hadfield Building, Mappin St, Sheffield, S1 3JD, UK 2) Advanced Manufacturing Research Centre, Advanced Manufacturing Park, Catcliffe, Rotherham, S60 5TZ, UK 3) Rolls-Royce plc, PO Box 31,DE24 8BJ, UK ABSTRACT Aerospace titanium components are manufactured under the strictest standards in order to ensure the highest quality. To develop highly efficient machining processes, extensive research and investment is necessary. For the specific case of rotating titanium critical components, large quantities of forged workpieces are machined to determine the effects of the different machining parameters on tool wear characteristics and component structural integrity and performance. However, testing the different permutations of metal removal parameters and tool combinations is expensive and time consuming at the development stage. The novel approach developed and presented here, enables the machining of smaller titanium parts that can be compared 1-to-1 to parts extracted from industrial machined disc components. This approach not only reduces cost, but ultimately accelerates the research and development process due to more rapid feedback between different iterations of the machining parameters. The proposed technique specifically replicates the face turning operation performed in rotational critical titanium components, such as compressor discs, using small coupons machined in a standard CNC machining centre. The machined coupons can be fatigue tested through a 4-point bending and microstructural analysis can be performed on the tested coupons to directly study the effects of the machining process on the surface and underlying microstructure.


Abbreviations AMRC
Advanced

Introduction
This paper presents a novel approach that replicates expensive and time-consuming face turning machining operation of disc profiles. The proposed technique developed at The University of Sheffield uses small coupons on a standard CNC milling machine. This enables the rapid assessment of machining and tooling parameters on subsurface damage and fatigue performance. The implementation of this technique will reduce research and development costs and provide the ability to test a wider range of development titanium alloys, machining parameters and development tools at an early stage. To generate confidence that the machining processes applied to these components will result in the desired mechanical properties, extensive testing must be performed. Titanium alloys are widely regarded as difficult to machine, due to their excellent mechanical properties at high temperatures, low thermal conductivity and high chemical reactivity [1]. This leads to very expensive machining costs, which can account for up to 60% of the total cost of the component and disproportional tool wear. Cost reduction is a key point in manufacturing titanium alloys, particularly with the increase in titanium usage in aircraft, coupled with the project increase in aircraft orders ( Figure 1). For example, it is estimated that more than 37000 new commercial planes (with more than 100 seats) will be required by 2037. Moreover, titanium alloy components have to be extensively machined, when in some cases the machined volume to final component ratio (buy-to-fly ratio) can be about 10:1 [4] or 20:1 [5], meaning that only 5% to 10% of the original titanium forging is used in the final component, with the remainder of the billet being turned into swarf. This is supported by Seong et al. [4] where it is pointed out that improved machining techniques for titanium alloys will show the greatest reduction in cost for the finished component in the near and mid-term future. Lowering costs in machining technologies will drastically reduce the component unit price, making them more attractive in a very competitive sector. For example, aerospace companies typically spend of the order of > £100 million on milling operations per year. As machining a is key manufacturing step, a better understanding of the effect of metal removal rates on performance is essential. The proposed method in this paper will enable companies such as Rolls-Royce make a more rapid assessment of the effects of the face turning operation.
The face turning operation is generally carried out on both sides of the disc concurrently, to avoid any vibrations that might affect the quality of the component or the machining process. A schematic of the operation can be seen in Figure 2. The tool selected for this machining operation was an uncoated T-Max® P Insert for turning DNMG 15 06 08-23 H13A made by Sandvik Coromant. The geometry and the dimensions of this tool are shown in Figure 2 and Table 1. As the finishing pass is a light cut that has to be within the machinability range for titanium alloys, the machining conditions selected for this face turning operation are 0.5 mm of depth of cut (ap), 0.15 mm/rev of feed per revolution (f rev ) and a cutting speed (Vc) of 50 m/min. The material removal rate (MRR) for this set of parameters is calculated in Equation (1). (1) It is also important to analyse the surface roughness that is obtained when machining a surface using this set of parameters. This is very important from the point of view of replicating a machining process, because the topography of the surface can affect the mechanical behaviour of a component when it is under operating loads. The machining tracks can act as stress concentrator points that ultimately could lead to cracking, which is detrimental for component's performance. With the selection of the tool and the processing parameters, it is possible to estimate also the surface topographic conditions such as Ra, based on the estimation formulas by P. De Vos [6] as shown in Equation (2). (2)

Coupons to be machined
The coupons where the machining process is designed to be applied must be of similar dimensions to coupons extracted from the disc's diaphragm after the final face turning finishing operation. The extracted coupons have dimensions of 3.6 x 19 x 180 mm. Due to the simple coupon geometry, using a turning lathe was not a viable option to replicate the disc finish face turning process, hence, a milling CNC centre was considered as a low cost and more widely available option.
Standard face milling using an indexable head -Iteration 1: One of the options that can be considered to replicate a face turning operation in a milling CNC machining centre, is using an indexable head with different tool inserts. This is a versatile option due to the vast range of suitable inserts that can be used. And therefore, the probability of matching a tool with similar characteristics to the turning insert is much higher. The selected indexable head was a CoroMill ® 415 face milling cutter 415-32EH25-05H with an external diameter of 23.6 mm combined with a CoroMill ® 415 insert 415N-05 02 06M-M30 H13A.
The optimum machining parameters for this tool are shown in Table 2. However, in order to avoid the machining marks and difference of surface topography generated by face milling: The process for this iteration uses the total cutting edge of the tool insert, for every longitudinal pass, instead the total circumference for machining the top surface. A schematic of the process is shown in Figure 3. In this case, the variable a width is set as the maximum cutting length provided by the insert (3.8 mm). As the total width of the "non-chamfered" machining surface is 11 mm, therefore, 3 passes are needed to machine the surface (Equation (3)). (

3)
This must be performed with an exit angle β (see Figure 3) close to 45 degrees, in order to avoid material being removed perpendicularly to the longitudinal direction of the coupon. This can be calculated using Equation (4). In this case this angle is 46 degrees.
Therefore, the machining conditions to replicate the face turning operation can be calculated for this specific case, such as the spindle speed, N max table feed v f and the MRR.
Based on the calculations in Equations (5) (6) and (7), it is possible to conclude that this process will use a lower cutting speed and therefore a lower MRR than the face turning process it is replicating.
-Iteration 2: The second iteration proposed was to align the rotation axis of the indexable head at the centre of the longitudinal axis. The aim of this was to reduce the machining time and increase the MRR, however, the machining marks on the top of the surface, Figure 4 (bottom), are not representative of a finish face turning operation. This surface topography occurs because the inserts cut through the whole rotation of the head.
Tilting the head can be a solution to this problem, specifically to the machining marks problem. A schematic of the tilted head is shown in Figure 4. The main issue in this case was that the machined surface is not flat, due to the tilting of the axis. Moreover, the cross section of the machined surface would be curved, see Figure 4 (right). As one of the mechanical tests that will assess the machining process is a 4-point bend test, the modification of the coupon cross section with stress concentrators is likely to significantly reduce the fatigue life, rendering the results not be representative of finish machined surface. Therefore, this 2 nd iterative machining process was deemed not suitable for replicating the conditions of a face turning operation.

Solution Proposed
The characteristics that the perfect replicating machining process are: Achieve the same: Cutting speed V c

Depth of cut a p
Feed per revolution f rev Same material removal rate (MRR) Use same or identical turning insert

Same material/chip removal technique to face turning
The solution proposed fixes the coupon at the table and rotates the main spindle of the CNC milling centre with a boring head attached with a similar radius that the face turning operation to be replicated. Boring heads can be equipped with off-the-shelf adaptors that are specifically designed for turning inserts.
For this specific case, a Sandvik CoroBore ® 825 (2) combined with S20-R825SL32 020 and a head for turning T-Max P ® SL-PDJNR-32-15HP (3) (or the left handed equivalent). All this is assembled to a BT 40 / MAS-BT403 taper to connect the tool to the main spindle of the machine. The tool in this case is the same insert used for the turning operation, T-Max ® P Insert for turning DNMG 15 06 08-23 H13A (4). A numbered schematic and a picture form the experiment is shown in Figure 5. The turning radius was set to maximum for this configuration (97.86 mm). Comparison of the machining marks that will be produced during a (1) face turning operation and (2) machining of a coupons with the boring head, the main difference will be the radius. However, if we consider the machine marks in the central testing area of the machined coupon, then the machined surface can be considered equivalent in both the large-scale finish face turning and small-scale coupon machining using the boring head, as schematically depicted in Figure 6. This approach enables the large-scale tool dimensions and finish machining parameters to be replicated, by calculating the table feed, depth of cut and spindle speed, which are inputted into the CNC milling controller. This is shown in Equations (8) and (9). The depth of cut ap is set at 0.5 mm by modifying the table-spindle distance. (8) Using these machining parameters provides the equivalent MRR compared to the face turning operation and the same material removal technique in a milling CNC machining centre. An image of a machined coupon is shown in Figure 7, images at different magnifications are also shown. On the top, it is shown the equivalent face turned surface machined in an industrial facility.

Conclusions
A novel technique has been designed and implemented in order to replicate costly face turning operations of large disc profiles using a standard CNC machining centre. The same machining conditions have been achieved on small titanium alloy coupons, using the same machining parameters and material removal rate.
The new approach presented in this paper has been designed with the goal of reducing resources necessary to assess the effect of different machining conditions for face turning operations applied to titanium alloy aerospace components, such as compressor discs. This technique enables more rapid feedback of the effect of machining conditions through the bore milling of small coupons, as opposed to large and expensive rectilinear forgings. In addition, the technique provides coupons ready for 4-point bend testing in order to evaluate the effects of the machining processes in components performance, by studying crack initiation and propagation mechanisms and fatigue life.