Identification of factors that dominate size effect in micro-machining

https://doi.org/10.1016/j.ijmachtools.2011.01.004Get rights and content

Abstract

In micro-machining the so called “size effect” is identified as critical in defining process performance. Size effects refers to the phenomenon whereby the reduction of the undeformed chip thickness to levels below the cutting edge radius, or gain size of the workpiece material begins to influence workpiece material deformation mechanisms, chip formation and flow. However, there is no clear agreement on factors that drive this size effect phenomenon. To explore the significance of cutting variables on the size effect, micro-milling tests were conducted on Inconel 718 nickel alloy using 500 μm diameter carbide end mill. The experimental design was based on an L9 Taguchi orthogonal array. Fast Fourier transform (FFT) and wavelet transform (WT) were applied to acoustic emission (AE) signals to identify frequency/energy bands and hence size effect specific process mechanism. The dominant cutting parameters for size effect characteristics were determined by analysis of variance (ANOVA). These findings show that despite most literature focussing on chip thickness as the dominant parameter on size effect, the cutting velocity is also a dominant factor. This suggests that manipulating the cutting speed is also an effective strategy in reducing burr thickness, optimising surface finish and in breaking the lower limit of micro-machining.

Introduction

In micro-machining the “size effect” modifies the mechanism of material removal compared to conventional (macro-scale) machining [1], [2], [3], [4]. This can arise because the thickness of material to be removed is of the same order of magnitude as the tool edge radius or grain size of the workpiece material. The size effect is typically characterised in machining by a non-linear increase in the energy consumed per unit volume of material removed as the undeformed chip thickness decreases. Experimental observation of this unbounded increase in specific energy in machining of ferrous and non-ferrous materials has been reported. With respect to investigated workpiece materials, difficult to cut materials (Ni, Ti based alloys) have not been reported. Most of the reported research work was undertaken over a limited cutting velocity range and the material thickness removed was in the sub-micron length scale. For a selected cutting speed, the size effect was attributed to tool edge radius effect [5], [6], [7], [8], material micro-structure effect i.e. dislocation density/availability [9], [10], material strengthening effect due to strain, strain rate, strain gradient [11], [12], [13], [14], [15], [16], subsurface plastic deformation [17] and material separation effect [18].

Nickel-based alloys are extensively used in aerospace components because of their high specific strength (strength to weight ratio) which is maintained over a wide temperature range. Approximately, half of the total materials used for a gas turbine engine are nickel alloys [19]. The most widely used nickel alloy, Inconel 718 accounts for as much as 45% of wrought nickel based alloy production and 25% of cast nickel based products [20]. However, nickel based alloys pose a serious machinability challenge to the industry due to their work hardening, high temperature strength, low thermal conductivity and hard abrasive particles [19]. Moreover, a high adhesion tendency with the tool material, leads to galling and welding of chips on the workpiece surface and short tool life.

Most recently, Klocke et al. [21] presented an assessment of feasibility of micro-milling process on single crystal nickel based super-alloy (Rene´ N5). All cutting tests were performed at fixed cutting velocity (50 m/min) and chip load (15 μm/tooth) using 800 μm diameter micro-end mills. In another study, Weinert and Petzoldt [22] demonstrated micro-milling as an alternative production process for NiTi shape memory alloy micro-parts. This was done for a narrow cutting range using 400 μm diameter micro-tool. The selected cutting velocity was 33 m/min while chip load was varied from 12 to 20 μm/tooth.

Acoustic emission (AE) refers to the elastic stress waves emitted as a result of the rapid release of strain energy within a material due to rearrangement of its internal structure [23]. In machining, deformation can be considered as the main AE producing source [24]. This implies that AE from machining processes such as turning, grinding and milling represent the unique characteristics of the mechanics of each process. A milling operation is a discontinuous cutting process with varying chip load performed by multiple cutting teeth. AE associated with this process consists of continuous and transient signals. The continuous AE signals stem from plastic deformation of workpiece material in the primary shear zone, at the chip/tool interface (secondary shear zone) and at the tool flank/workpiece surface interface (tertiary shear zone). While the transient signals are associated with pulse shock loading, such as initiation or breakage of chips or by tool fracture. However, in micro-scale milling for chip thickness at or below few microns in length, AE generation is attributed to the interaction of the tool edge radius with the micro-structural features of the workpiece material [25].

Uehara and Kanda [26] reported that the amplitude of an AE signal from the cutting tool is reduced during AE transfer from the tool to the workpiece possibly by damping or reflection at the primary interface. As a result, by monitoring the AE signal from the workpiece side, the primary and tertiary deformation zones can be regarded as most accessible sources of AE generation in micro-milling. In ultra-precision cutting, Carpenter et al. [27] demonstrated that the tertiary deformation zone becomes a more significant source of AE due to the increased energy spent on sliding friction between the tool flank face and newly machined surface.

In literature it is suggested that, sporadic AE signal collected at the sensor contains many frequencies due to the several AE producing sources being active at the cutting zones [25]. Thus, the frequency of these different frequency bands can be used for identifying the dominant sources of AE in the micro-machining process. Fig. 1 shows identified frequency range according to the material removal length scale and signal source mechanisms. It is also clear from Fig. 1 that AE signals are better suited to characterising machining at very low length scales typically in the nanometre range as encountered close to the minimum chip thickness in micro-machining. Again it can be inferred from Fig. 1 that decomposing the AE signal into the frequency domain will enable identification of cutting mechanisms such as dislocation mechanics, inter-granular micro-fracture, cleavage and shearing material machining modes.

In relation to the cutting zones, it is reported that the primary deformation zone, is the largest AE source and generates in excess of 75% of the total AE signal [28]. Brittle and ductile fracture (as shown in Fig. 2) may occur in the deformation of ductile materials that can emit AE from the primary shear zone depending on the type of material being machined and the operating conditions. It is well known that brittle fracture occur in a transgranular (cleavage) manner while ductile fracture takes place in inter-granular manner where the grain boundaries are the fracture path within the material [29]. Most of the metals undergo a transition from transgranular fracture to inter-granular fracture when deformation occurs at temperatures greater than 0.5–0.7 times that of the melting temperature. Details of these fracture mechanisms are covered elsewhere [30].

In a case of continuous chip formation, Shaw [31] proposed that ductile materials deformed as a result of nucleation, growth and coalescence of microscopic voids (micro-crack initiation) that originated at inclusions and from second phase particle separation. However, due to compressive plastic deformation, the micro-cracks experience closure and this effectively postpones crack propagation until conditions are reached where plastic deformation does not occur. Barry and Byrne [32] observed a lamellae form as a result of cleavage during the non-overlapping cutting of a low alloy steel of 45.6 HRC with cutting speed of 100 m/min and for a depth of cut slightly greater than the minimum chip thickness during continuous chip formation.

With respect to a saw tooth formation, Komanduri and Schroeder noted thermoplastic instability leading to adiabatic shear in the chip formation process of Inconel 718 which was very similar to the mechanisms reported for titanium as well as hardened AISI 4340 steels, when machining at higher cutting speeds [33]. Furthermore, Barry et al. [34] elaborated on this and presented two distinct mechanisms for catastrophic shear failure when machining Ti–6Al–4V at 80 μm undeformed chip thickness. Under high cutting velocity (180 m/min), material failure was attributed to ductile fracture due to increase in ductility as a result of thermal softening. However, it was also noted that the upper region of the primary shear zone (segment tip) sheared off due to cleavage. Under low cutting velocity (15 m/min), failure was said to have occurred due to cleavage. Zhou et al. argued in favour of the material losing plasticity at high cutting speeds [35]. This observation is also in conformance with stress–strain curves reported for various temperatures for Inconel 718 alloy [33].

Under micro-cutting conditions, tool edge radius [5], [8], chip load [17], cutting speed [15] and chip aspect ratio [36] have been reported as some factors which affect the starting point of the nonlinearity in specific cutting energy. Additionally, chip thickness influences increase in surface roughness [37], [38], burr formation [39], [40] and rate of tool wear. The relative contribution and dominance of cutting factors on size effect have recently been experimentally explored for hardened tool steel [41]. However, the application of acoustic emission signals to identify process mechanisms and hence clarify the dominance of cutting variables on the size effect has rarely been reported. Neither is there any related work focussed on nickel alloys. The motivation for this study was to assess the dominant factors in the control of plastic deformation in micro-scale cutting. Understanding the size effect dominant process parameters should enable the manipulation of the minimum chip thickness (e.g. lowering minimum chip thickness) and creating flexibility for generation of optimum conditions for best surface roughness.

Section snippets

Experimental setup

Micro-machining tests were conducted on a wrought Inconel 718 nickel alloy workpiece material that was annealed by heating for one hour at 954–982 °C followed by cooling in air. The tests were conducted on a Mikron 400 high speed machining centre. All cutting tests were performed under dry conditions, which provided the opportunity to capture AE signals without coolant noise. Acoustic emission was measured with a Kistler Pieztron® acoustic emission sensor Type 8152B121 and the signal was

Signal processing technique

In relation to process monitoring approaches, a wide range of signal processing techniques such as fast Fourier transformations (FFT) [42], short time Fourier transformations (STFT) [43] and wavelet transformation (WT) [44] have been reported. These methods were mainly employed in monitoring ultra-precision machining. However, these methodologies have rarely been studied especially in micro-mechanical machining.

AE signals in micro-milling are composed of continuous and spontaneously released

Results and discussions

Table 3 presents the cutting trial and partial data collected against the characteristics of interest on repeated run (output function). For accuracy, all outcomes were measured at two identical locations and averages of those values were used in the analysis. The data shows that size effect indicators (e.g. increases in specific energy, surface roughness or burr root thickness) intensify with decreasing undeformed chip thickness.

Optimum process parameters and confirmation tests

Signal to noise ratio (SNR) corresponding to each trial for AE specific energy, surface roughness, and burr root thickness was calculated to reveal optimum process parameters. It may be mentioned that SNR analysis takes into consideration the mean value at a given level as well as its repeatability in terms of standard deviation from the mean level. The evaluated optimal micro-milling process settings for the selected outcomes are shown in Table 10. The optimal condition to obtain best surface

Conclusions

  • The specific energy, burr root thickness and surface roughness of micro-machined parts can be used as relevant measures of the size effect in micro-machining.

  • The wavelet transformation has a good resolution to extract energy bands related to the deformation mechanisms involved in the micro-milling process.

  • High frequency bandwidths in the AE signal as well as burr root thickness can be exploited to identify size effect phenomenon.

  • More than 50% of the AE signal was composed of micro-fracture

Acknowledgement

The authors would like to thank the support offered by the Engineering and Physical Sciences Research Council (EPSRC) UK under the ELMACT Grant DT/E010512/1.

References (50)

  • K. Uehara et al.

    Identification of chip formation mechanism through acoustic Emission measurements

    Annals CIRP

    (1984)
  • D.P. Saini et al.

    A quantitative model of acoustic emissions in orthogonal cutting operations

    Journal of Materials Processing Technology

    (1996)
  • M.C. Shaw

    A new mechanism of plastic flow

    International Journal of Mechanical Sciences

    (1980)
  • J. Barry et al.

    Observations on chip formation and acoustic emission in machining Ti–6Al–4V alloy

    International Journal of Machine Tools and Manufacture

    (2001)
  • L. Zhou et al.

    Material removal mechanism beyond plastic wave propagation rate

    Precision Engineering

    (2003)
  • Y. Furukawa et al.

    Effect of material properties on ultra-precise cutting processes

    Annals CIRP

    (1988)
  • H. Weule et al.

    Micro-cutting of steel to meet new requirements in miniaturization

    Annals CIRP

    (2001)
  • K. Lee et al.

    Micro-burr formation and minimization through process control

    Precision Engineering

    (2005)
  • R. Teti et al.

    Advanced monitoring of machining operations

    Annals CIRP

    (2010)
  • H. Schulz et al.

    High speed machining

    Annals CIRP

    (1992)
  • X. Liu et al.

    The mechanics of machining at the microscale: Assessment of the current state of the science

    ASME Journal of Manufacturing Science and Engineering

    (2004)
  • E.J.A. Armarego et al.

    On the size effect in metal cutting

    International Journal of Production Research

    (1962)
  • J.D. Kim et al.

    On the size effect of micro-cutting force in ultraprecision machining

    JSME International Journal, series C

    (1996)
  • W.R. Backer et al.

    Size effect in metal cutting

    Transactions of the ASME

    (1952)
  • M.C. Shaw

    The size effect in metal cutting

    Sadhana

    (2003)
  • Cited by (117)

    • Size effect during dynamic shear tests with hat-shaped specimens

      2023, Journal of Materials Research and Technology
    View all citing articles on Scopus
    View full text