Modeling the effects of cutting parameters in MQL-employed finish hard-milling process using D-optimal method

https://doi.org/10.1016/j.jmatprotec.2007.08.029Get rights and content

Abstract

In any finish milling process the minimization of surface roughness is the prime objective, and if the workpiece belongs to the category of ultra-hard steels then the objective of maximizing tool life also gains considerable importance. In this research work, the effects of four parameters, namely, hardened steel's microstructure, workpiece inclination angle, cutting speed, and radial depth of cut were studied upon tool life and surface roughness (in directions of feed and pick-feed). The milling was performed under environment of minimum quantity of lubrication (MQL), using coated carbide ball-nose end mills. The quantification of the aforementioned effects was done using a new response surface methodology known as the D-optimal method. For tool life, workpiece material was found as the most influential parameter followed by the rotational speed of tool. High values of tool's rotational speed proved unfavorable for tool life but favorable for surface finish. In addition, the effects of workpiece inclination angle and radial depth of cut were analyzed upon effective cutting speed and cusp height and, subsequently, upon surface roughness. SEM and EDS analyses of the worn-out tools were carried out in order to study the effects of different levels of parameters selected upon the severity of different tool damage modes. The major tool damage mechanisms detected were notch wear, adhesion, and chipping. The severity of chipping was relatively smaller as compared to that of adhesion and of notch wear because of reduced effective cutting speeds and feed rate employed.

Introduction

Hard milling is the name given to the process in which the hardened steels (hardness: 35–70 HRc) are machined in their respective high-speed range (cutting speed: 30–500 m/min) using an end mill cutter (Zeren, 2002). The quoted advantages of hard milling consist of increased fatigue life of the workpiece, compressive residual stresses, and minimal alterations of micro-hardness and microstructure (Axinte and Dewes, 2001, Axinte and Dewes, 2002), along with the general benefits offered by high-speed machining (Schulz, 2004). The drastic reduction in tool life is considered as the only significant drawback of the hard-milling process. Lot of research work has been and is being done in order to find the optimal combination of tooling, cutting, and environment parameters for enhancement of tool life, without compromising the high values of material removal rate (MRR) offered by the high-speed milling process. In addition, if the process is finish milling then the requirement of minimizing the workpiece's surface roughness gains prime importance.

The issues of enhancing tool life and improving surface finish will be addressed in this research work by investigating the effects of steel's microstructure, workpiece inclination angle, cutting speed, and radial depth of cut in MQL-employed finish milling of cold work tool steels (AISI D2 and X210 Cr12) using coated carbide ball-nose end mills. AISI D2 and X210 Cr12 (nearest equivalent of AISI D3) are widely used in manufacturing of plastic injection molds; manufacturing gages; blanking, drawing and piercing dies; shears, forming and banding rolls; lathe centers, mandrels, broaches, reamers, taps, and threading dies. Table 1, Table 2 present the chemical composition of AISI D2 and X210 Cr12, respectively, and Fig. 1 present the microstructures of both the steels.

AISI D2 is an air quenching, deep hardening, highly wear resistant cold work tool steel, possessing poor machinability characteristics. Table 1 shows very high contents of chromium and carbon in the composition of AISI D2. These elements are responsible for generating large chromium carbide inclusions in the microstructure when the material is hardened using heat treatment process (Coldwell et al., 2003). The large volume of highly hard undissolved chromium carbide particles in the microstructure of D2 (clearly visible from Fig. 1(a)), which is responsible for imparting excellent wear resisting characteristics to the material, also promotes poor machinability characteristics. The microstructure of X210 Cr12 (Fig. 1(b)) is almost similar to that of AISI D2, consisting mainly of primary carbides and martensite. The X210 Cr12 steel composition consists of higher percentage of carbon as compared to that of AISI D2, and also it contains tungsten instead of molybdenum.

It has been stated that the main goal of high-speed milling (HSM), when applied to finishing operations, is the reduction of Rmax down to 10 μm or even less (Fallbohmer, 1998). Ra values ranging from 2.1 to 3.8 μm for sharp TiAlN coated carbide tools, and from 3.4 to 4.4 μm for worn tools, have been reported in (Koshy et al., 2002). Different observations have been reported regarding effects of workpiece's inclination angle upon performance measures. It has been reported that tool life in the case in which workpiece surface was inclined at 30° was about three times more than that obtained when workpiece surface was kept normal to axis of cutter (Kita et al., 2001). Moreover, the tool failure in the first case was caused by chipping on the rake face while it was caused in the second case because of generation of extremely rough surface. In other paper, the authors reported the contrary observation, i.e., longer tool life values could be achieved when operating with cutter axis oriented normal to the workpiece surface rather than oblique one (Urbanski et al., 2000). The Ra values obtained varied from 0.5 to 2.5 μm at speed of 350 m/min and from 0.5 to 3.5 μm at speed of 250 m/min, for TiAlN coated carbide ball-nose cutters. It has also been reported that non-inclined milling provided 1.5 times longer tool life values as compared to those provided by 60° inclined milling (Axinte and Dewes, 2001, Axinte and Dewes, 2002). The reasons reported were the possibility of chatter and/or higher effective speeds of cutter in the case of inclined milling. On the other hand, the workpiece surface roughness improved at higher level of inclination angle due to avoidance of cutting in the tool center and corresponding rubbing of workpiece material.

Application of flood coolant to HSM has not been very promising. Majority of the coolants, especially the water-based ones, have negative effect upon tool life. Also the matters related to fluids cost, handling, recycling, and health hazards have restricted the utilization of coolants in HSM. Machining with minimum quantity of lubrication (MQL) can cut down cost and improve both tool life and surface finish (Quaile, in press). MQL is the name given to the process in which very small amount of oil (less than 30 ml/h) is pulverized into the flow of compressed air (Braga et al., 2002). MQL helps in reducing cutting temperature and also averts thermal shocks, experienced by flood coolant. The air/oil aerosol mixture is then fed to the cutting area through the ducts (normally two in number).

In the current research work the effects of four predictor variables, namely, workpiece material's microstructure, workpiece inclination angle (α), tool's rotational cutting speed (N), and radial depth of cut (also called as ‘step-over’ in ball-nose end milling) (ae) will be studied upon following three response variables, in MQL-employed finish hard-milling process, using coated carbide ball-nose end mill cutters:

  • 1.

    Tool life—to be measured in mm2 of area (length of cut {No.} × radial depth of cut {ae}) of material removed until tool failure criteria is attained.

  • 2.

    Arithmetic roughness of workpiece's surface, measured along the direction of feed, and averaged throughout the life of tool (Ra_avg(along))—to be measured in μm.

  • 3.

    Arithmetic roughness of workpiece's surface, measured across the direction of feed (or along pick feed direction), and averaged throughout the life of tool (Ra_avg(across))—to be measured in μm.

A new statistical methodology called as D-optimal method will be employed for design of experiments, analysis of variance (ANOVA), empirical modeling, and numerical optimization. Afterwards, the SEM and EDS analyses of the worn-out tools will be carried out in order to study the effects of different levels of predictor variables upon the severity of different types of tool wear modes.

Section snippets

Effects of inclination angle and radial depth of cut upon effective cutting speed and cusp height

Consider the inclined milling process, employing single direction raster horizontal downward orientation (−βfn), as shown in Fig. 2. Consult Ref. (Toh, 2004) for nomenclature details.

In Fig. 2(a) the formation of cusp has been shown, whose height ‘C’ depends upon value of ae selected. The higher the value of cusp height the rougher is the workpiece's surface. Fig. 2(b) presents more details regarding ball-nose inclined milling. The effective diameter Deff (≤D) of the end mill can be visualized

The D-optimal method

The D-optimal method is relatively a new technique, related to response surface methodology, used for carrying out the design of experiments, the analysis of variance, and the empirical modeling. The D-optimal criterion was developed to select design points in a way that minimizes the variance associated with the estimates of specified model coefficients (Myers and Montgomery, 2002). In a sense this method is more useful than central composite design (a conventional response surface method)

Experimental results, ANOVA, regression, and optimization

Following sub-sections describe ANOVA, regression, and optimization applied to the experimental results. All the statistical analyses were performed using statistical computing package Design-Expert 7.0®, which has been developed by Stat-Ease®.

Progress of tool wear and surface roughness

Fig. 9 displays the progress of maximum width of flank wear land along the area of material removed for seven tests. Two groups can easily be identified. The three plots found at the left hand side of the graph, representing the rapid progress of tool wear, belong to the milling of AISI D2, while the other four plots appearing at the right hand side, representing medium to slow progress of tool wear, belong to milling of X210 Cr12 tool steel. This observation suggests that the effect of

Tool wear mechanisms

Fig. 12 shows the SEM photographs of the worn-out tools used in tests number 2, 3, 8, 9–10. Images A and B belong to the tool used in test number 2. Image A shows the flank face, adjacent to one of the central cutting edges, a portion of which is densely covered by the adhesive layer of workpiece material. The adhesion has also been verified by EDS. Image B shows a massive scaled notch wear (depth of cut wear) occurring at one of the central cutting edges of the tool. The adjacent portion of

Conclusions

This article provided the profound analysis of ball-nose end finish milling of hardened steels using coated carbide cutters. The D-optimal technique was utilized to investigate the influence of workpiece material, workpiece's inclination angle, tool's rotational speed, and radial depth of cut upon tool life and workpiece surface roughness. For tool life, the workpiece material (chemical composition + hardness) was found most influential parameter followed by the rotational speed of tool. It was

Cited by (50)

  • Research trends in high speed milling of metal alloys: A short review

    2019, Materials Today: Proceedings
    Citation Excerpt :

    The stability of the milling process is complicated due to the rotating tool, multi-cutting teeth, peripheral machining forces, and chip formation. Therefore, cutting force models can be used to improve manufacturing productivity [47]. Cryogenic machining conditions using liquid nitrogen in end milling process is experimented by Norcahyo and Soepangkat [17].

  • Optimal efficient biodiesel synthesis from used oil employing low-cost ram bone supported Cr catalyst: Engine performance and exhaust assessment

    2018, Energy
    Citation Excerpt :

    As a means of process parameter optimization, D-optimal design (a relatively new type of response surface method) was used to evaluate optimal parametric combination and interactions among the three process parameters using only 13 experimental runs as per Design Expert Software (Stat-Ease, Inc.; USA) [28]. D-optimal design requires lesser number of experiments in comparison with central composite design (a traditional response surface method) for evaluation of the relative effects of predictor parameters in governing the process response [29]. From the prior works, it is evident that Cr impregnated BiHAp (ram bone derived) supported solid acid catalyst has not been reported before.

  • Experimental study on the effect of nanoparticle concentration on the lubricating property of nanofluids for MQL grinding of Ni-based alloy

    2016, Journal of Materials Processing Technology
    Citation Excerpt :

    Nevertheless, MQL achieves better lubrication effect. Iqbal et al. (2008) cited the numerous reports on MQL. They emphasized that compressed gas input into the grinding zone failed to achieve the expected cooling effect.

View all citing articles on Scopus
1

Tel.: +86 25 84892551; fax: +86 25 84891501.

2

Tel.: +92 300 850976.

3

Tel.: +86 1300340957; fax: +86 25 84891501.

4

Tel.: +92 300 8544016.

View full text