jatinder kumar j.s. khamba s.k. mohapatra*libvolume6.xyz/.../ultrasonicmachiningusmnotes1.pdf ·...

Int. J. Machining and Machinabiliy of Materials, Vol. 3, Nos. 1/2, 2008 143

Copyright © 2008 Inderscience Enterprises Ltd.

An investigation into the machining characteristics of titanium using ultrasonic machining

Jatinder Kumar Department of Mechanical Engineering, Ajay Kumar Garg Engineering College, Ghaziabad 201009, India Fax: +91-0120-276-7384 E-mail: [email protected]

J.S. Khamba Department of Mechanical Engineering, University College of Engineering, Punjabi University, Patiala 147004, India Fax: +91-0175-304-6324 E-mail: [email protected]

S.K. Mohapatra* Department of Mechanical Engineering, Thapar Institute of Engineering and Technology, Patiala 147004, India Fax: +91-0175-239-3005 E-mail: [email protected] *Corresponding author

Abstract: This paper presents a review on the problems encountered in machining titanium and application of USM in machining titanium and its alloys. Experiments have been conducted to assess the effect of three factors-tool material, grit size of the abrasive slurry and power rating of ultrasonic machine on machining characteristics of titanium (ASTM Grade I) using full factorial approach for design and analysis of experiments. It has been concluded that all factors have significant effect on Material Removal Rate (MRR), Tool Wear Rate (TWR) and surface roughness of the machined surface. Two-way interactions having significant effect on MRR, TWR and surface roughness have also been identified using Minitab14 software. The levels for each factor that contribute the most to the variation in machining performance of USM of titanium have also been established. It has been concluded that titanium is fairly machinable with USM process. Moreover, the surface finish obtained is better than many of the other non-traditional processes.

Keywords: Ultrasonic Machining; USM; titanium; design of experiments; full factorial design; abrasive; grit size; power rating; work hardening; Material Removal Rate; MRR; Tool Wear Rate; TWR; surface roughness.

144 J. Kumar, J.S. Khamba and S.K. Mohapatra

Reference to this paper should be made as follows: Kumar, J., Khamba, J.S. and Mohapatra, S.K. (2008) ‘An investigation into the machining characteristics of titanium using ultrasonic machining’, Int. J. Machining and Machinabiliy of Materials, Vol. 3, Nos. 1/2, pp.143–161.

Biographical notes: Jatinder Kumar is working as a Senior Lecturer in the Department of Mechanical Engineering at Ajay Kumar Garg Engineering College, Ghaziabad, India. He received a BE in Production Engineering from G.N.E. College Ludhiana and an ME in Industrial Engineering with Honours from Thapar Institute (Deemed University), Patiala. He is also pursuing his PhD studies in the Department of Mechanical Engineering at TIET Patiala. He is a Life Member of ISTE. He has contributed about eight research papers in journals, conferences and workshops at national and international level. His areas of research include industrial engineering, non-traditional machining, quality engineering and manufacturing engineering.

J.S. Khamba is working as a Professor and the Head in the Department of Mechanical Engineering at University College of Engineering, Punjabi University Campus, Patiala, India. He received a BE in Mechanical Engineering and ME in Industrial Engineering from TIET, Patiala. He received a PhD in Flexible Management of New Technology from the Department of Mechanical Engineering, TIET, Patiala. He has contributed more than 40 research papers in journals and conferences at national and international level. He has supervised two PhD and more than 20 ME theses. He is supervising six PhD theses currently. His areas of interest include industrial engineering, non-traditional machining and technology management.

S.K. Mohapatra is working as Professor and Head in Mechanical Engineering Department at Thapar University, Patiala. He completed his BE in Mechanical Engineering from Utkal University, Orissa and MTech in Fuels and Combustion from Indian School of Mines, Dhanbad. He obtained PhD in Thermal Engineering from ISM, Dhanbad in 1997. He has published about 50 research papers in journals and conferences at national and international levels. He has completed six sponsored projects. He is guiding four PhD theses as well. He has organised three national conferences. His areas of interest include thermal engineering, manufacturing engineering and non-traditional machining.

1 Introduction

Ultrasonic Machining (USM) is a non-conventional mechanical material removal process used for machining both electrically conductive and non-metallic materials; preferably those with low ductility (Gilmore, 1989; Moreland, 1988) and a hardness above 40 HRC (Gilmore, 1990; Haslehurst, 1981) for example, inorganic glasses, ceramics, nickel alloys, etc. The process came into existence in 1945 when L. Balamuth was granted the fist patent for the process. USM has been variously termed ultrasonic drilling; ultrasonic cutting; ultrasonic abrasive machining and slurry drilling.

In USM, high frequency electrical energy is converted into mechanical vibrations via a transducer/booster combination, which are then transmitted to an energy focusing as well as amplifying device: horn/tool assembly. This causes the tool to vibrate along its longitudinal axis at high frequency; usually >20 kHz with an amplitude of 12–50 µm (Kennedy and Grieve, 1975; Kremer, 1991). The power ratings range from 50 to 3000 W

An investigation into the machining characteristics of titanium 145

and a controlled static load is applied to the tool. Abrasive slurry, which is a mixture of abrasive material; for example, silicon carbide, boron carbide or aluminium oxide suspended in water or some suitable carrier medium is continuously pumped across the gap between the tool and work (∼25–60 µm). The vibration of the tool causes the abrasive particles held in the slurry to impact the work surface leading to material removal by microchipping (Moreland, 1984). Figure 1 shows the basic elements of an USM set up using a magnetostrictive transducer.

Figure 1 Ultrasonic machining set up

Variations of this basic configuration include:

1 Rotary Ultrasonic Machining (RUM). In this process, tool vibrates and rotates simultaneously thereby improving the Material Removal Rate (MRR) and reducing the geometric inaccuracies; for example, oversize and out of roundness, etc. (Gilmore, 1991).

2 USM combined with Electric Discharge Machining (EDM) (Farago, 1980).

3 Ultrasonic assisted cutting/machining. Ultrasonic assisted turning is the most common process in this category.

4 Other non-machining applications such as cleaning, welding, coating and forming, etc.

The aim of this research work carried out, which is being presented in this paper, is to identify potential significant factors contributing to machining performance of USM of titanium (ASTM Grade I) in terms of three machining performance indices-MRR, Tool Wear Rate (TWR) and surface roughness of the machined surface. The problems experienced in machining titanium with conventional and other unconventional


machining methods have also been reviewed. Finally, the most significant levels for each of the factors undertaken in this study have also been established using statistical testing.

2 Literature review on USM of titanium

Titanium and its alloys are used extensively in aerospace because of their excellent combination of high specific strength, which is maintained at elevated temperature, their fracture resistant characteristics and their exceptional resistance to corrosion. They are being used increasingly in other industrial and commercial applications such as petroleum refinery, nuclear reactors, surgical implants and marine applications. They are also being used exclusively for aerospace industry, mainly in airframe construction, where maximum ease of formability is desired. Other applications include compressor blades, rocket cases and offshore pressure vessels Ezugwu and Wang (1997).

The machining of titanium and its alloys is generally cumbersome owing to several inherent properties of the material. Titanium is very chemically reactive and therefore, has a tendency to weld to the cutting tool during machining thus, leading to premature tool failure (Ezugwu and Wang, 1997). Its low thermal conductivity increases the temperature at the tool-work interface thus, affecting the tool life adversely. Additionally, its high strength maintained at elevated temperature further impairs the machinability. Owing to all these problems, it is very difficult to machine titanium and its alloys by conventional machining processes and moreover, by conventionally used tool materials. In the last few decades, there have been great advancements in the developments of cutting tools, including coated carbides, ceramics and cubic boron nitride and polycrystalline diamond. However, none of these newer developments had successful application in the machining of titanium and its alloys with conventional processes due to their peculiar characteristics. Straight tungsten carbide cutting tools have proven their superiority in almost all machining processes of titanium alloys, but still there is a great scope for improvement in machining efficiency.

Non-traditional machining methods such as Electric Discharge Machining (EDM) and Laser Beam Machining (LBM) has been applied to the machining of titanium and its alloys during recent times but even these established processes have their limitations; particularly in machining of small and deep holes in titanium and its alloys. With EDM process, one problem is that the debris in machining gap cannot be eliminated easily, and the machining status is unstable during the process (Lin and Yan, 2000). Another reason is that titanium has a low heat conduction efficiency and high tenacity. LBM can be applied for machining of titanium, but even this process has its own problems in forming pear shaped holes and tapering of holes, holes with straight profile are difficult to obtain. USM can be suitable for machining titanium and its alloys due to the following characteristics Kazantsev et al. (1973).

1 Titanium and its alloys have low thermal conductivity and in USM there is a thinner zone affected by machining, generous quantity of cutting fluid is used resulting in better heat dissipation, efficient slurry flow can be maintained, depth of cut can be maintained due to rigidity of tool fixed in tool holder, and chemically active medium can be used to transfer heat efficiently and reduce cutting forces between the tool and workpiece.

2 USM is superior to hybrid processes in terms of simplicity, economic and provides a better control.


The application of USM for titanium and its alloys has been reported by a few researchers. Sharman et al. (2001) have discussed the application of ultrasonic assisted turning to titanium aluminide. As compared to conventional turning, the cutting forces are reported to be of very small magnitude (approx. 12%) in this process, thereby, improving both the tool life and surface finish. Aspinwall and Kasuga (2001) have reported the use of USM for production of 3 mm holes in γ -titanium aluminide. In machining with conventional methods, titanium aluminide; an alloy of titanium, encounters problems of surface integrity and microcracking. When machined with USM, satisfactory results have been achieved with polycrystalline diamond tooling. Grit size has been identified as the greatest factor affecting MRR followed by static load, tool type: solid/hollow and power level. In contrast to brittle materials, the combination of fine grit size, low power level and solid tool type gives maximum TWR. The surface finish obtained is superior to ceramics.

Wansheng et al. (2002) have investigated the effect of ultrasonic vibration introduction in EDM process to machine microholes in titanium alloy; concluding an increase in MRR as well as the process stability along with reduction in arcing phenomenon. When applied to machining of titanium alloy, the combined EDM-USM process has been found to demonstrate better performance in terms of improved MRR, discharging efficiency and reduced thickness of the recast layer (Wansheng et al., 2002). Lin and Yan (2000) have also shown a similar result; the introduction of ultrasonic vibration into deep hole EDM of titanium alloy can improve the machining quality and efficiency distinctly. Singh and Khamba (2006) have investigated the machining characteristics of titanium and its alloy (Ti-6Al-4V) using USM process. They have reported optimum MRR and TWR with boron carbide as abrasive material with grit size 220 and Stainless steel as tool material. The surface finish has been reported to be better than that obtained while machining brittle materials such as ceramics.

3 Design of experiments and experimentation

From the literature survey undertaken regarding the application of USM for titanium and its alloys, it is evident that use of USM as a machining process for titanium has not been explored to a great extent. Moreover, the power rating of USM as a factor has not been reported by any investigator up to the best knowledge of the author. Hence, power rating of the ultrasonic machine was selected as a process parameter in this study.

In all, three factors were selected for experimentation – tool material; with two levels: High carbon steel and titanium alloy, grit size of abrasive and power rating of the ultrasonic machine. High carbon steel (1095) and titanium alloy (ASTM Grade-V) were selected as tool materials because they possess a wide spectrum of mechanical and physical properties, which could be of significance in USM of titanium. The chemical composition and few important mechanical properties of both tool materials are given in Tables 1 and 2, respectively. Both tools were prepared as solid cylindrical type with diameter 8 mm. The response factors to be studied were fixed as: MRR of titanium workpiece, TWR of both the tools used and surface roughness of the machined surface. To study the influence of power rating of the ultrasonic machine and grit size of the abrasive material, a pilot experimentation was performed using both tools (of HCS, titanium alloy) and different levels of power rating (from 100 to 500 W) with equal intervals of 100 W and different grit sizes (100–600). Experiments were performed with


Sonic Mill-AP 500 W set up manufactured by Sonic Mill, Albuquerque. Aluminium oxide was used as an abrasive material for preparing the slurry with a concentration of 25%. Three grit sizes of 220, 320 and 500 were selected for final experimentation as they represent a wide range of average particle size (16–64 microns) and also exhibit strong influence on the variation in machining performance as observed in pilot experimentation. Power levels beyond 400 W could not be selected as the machining status was highly unstable after crossing this value. Moreover, the machining performance was very low for power rating less than 100 W. Hence, four power levels were finalised for the experimentation: 100, 200, 300 and 400 W.

Table 1 Chemical composition and important properties of Tool material (HCS)

Chemical composition (by weight %) of HCS (1095 series)

C Mn Residual Fe

1.01 0.35 0.3 balance

Note: Density = 7.89 g/cm3; Hardness = 56 HRC.

Table 2 Chemical composition and important properties of Tool material (Titanium alloy)

Chemical composition (by weight %) of titanium alloy (ASTM Grade V)

O N C H Fe Al V Ti

0.20 0.05 0.08 0.015 0.40 6.02 4.27 balance

Note: Yield strength = 900 MPa; Ultimate strength = 1010 MPa; Hardness = 42 HRC; Density = 4.45 g/cm3; Mod. of elasticity = 114 GPa.

Titanium (ASTM grade I) was used as the work material in the experimentation. The chemical composition and other important properties of titanium are given in Table 3. A summary of the work material, process parameters undertaken in the study and the parameters that were kept constant is given in Table 4. MRR was calculated as loss in weight per unit time (mg/min). TWR was also computed in the same manner. Surface roughness was assessed with perthometer manufactured by Mahr (M4Pi) with a cut off value of 0.25 mm and tracing length of 1.5 mm. Three observations were taken for each hole and were averaged to get the value of roughness (Ra).

There are several approaches to design the experiments such as full factorial designs, fractional factorials, Latin square designs and taguchi’s robust design of experiments technique. The identification of all the significant two-way interactions between the different factors cannot be realised effectively by using techniques such as fractional factorials and taguchi’s robust design methodology (Astakhov, 2004). Moreover, if any crucial factor is missed at the time of experimentation or any particular combination of the selected factors is not put into trial, sometimes the results obtained might just be inadequate from the point of view of optimisation. Full factorial designs usually result in more experimentation requirements and hence more time and resources consumption in performing it but at the same time, have been reported to provide best accuracy in design and subsequent analysis of experiments (Hicks and Turner, 1999). Hence, this approach was used in designing and analysing the experimental runs. Each experiment was replicated twice to take the inherent variability of the process into consideration.


Table 3 Chemical composition and important properties of Work material (Titanium)

Chemical composition (by weight %) of titanium (ASTM Grade I)

O N C H Fe residual Ti

0.18 0.03 0.08 0.01 0.2 0.4 99.1

Note: Yield strength = 220 MPa; Ultimate strength = 340 MPa; Hardness = 115 HV; Density = 4.51 g/cm3; Mod. of elasticity = 103 GPa.

Table 4 Process parameters and their values at different levels

Symbol Process parameter

Level 1 Level 2 Level 3 Level 4

A Work material Titanium (Gr 1) B Tool material High carbon

steel Titanium (Gr 5)

C Grit size 220 320 500 D Power rating 100 200 300 400 Constant parameters Frequency of vibration

21 KHz

Static load 1.63 Kg Amplitude of vibration

25.3–25.6 µm

Depth of cut 1 mm Thickness of workpiece

10 mm

Slurry concentration

25%

Tool geometry Straight cylindrical with diameter 8 mm Abrasive type Alumina Slurry temperature 28°C (ambient room temperature) Slurry flow rate 26.4 × 103 mm3/min Slurry media Water

The 24 experimental runs were completely randomised to minimise the effect of noise factors and error. Table 5 depicts the design matrix for the experimental runs (trial number has been indicated in brackets). The results obtained for MRR, TWR and surface roughness has been indicated in Table 6. All the values indicated are averages of three samples for each run as each experiment was replicated twice.

Table 5 Design matrix for experimentation

Grit size Tool material High carbon steel Titanium alloy

Power rating (W) Power rating (W) 100 200 300 400 100 200 300 400

220 (20) (6) (11) (22) (4) (19) (24) (1) 320 (16) (14) (17) (2) (5) (7) (21) (13) 500 (3) (10) (8) (18) (9) (12) (23) (15)


Table 6 Experimental results

Run Tool material

Grit size Power rating (W)

MRR (mg/min)

TWR (mg/min)

Roughness (Ra) µm

1 TI 220 400 1.7 1.67 1.25

2 HCS 320 400 2.41 4.83 1.12

3 HCS 500 100 0.25 1.00 0.75

4 TI 220 100 0.71 0.67 1.12

5 TI 320 100 0.4 0.43 0.78

6 HCS 220 200 1.62 3.66 0.77

7 TI 320 200 0.63 0.31 0.96

8 HCS 500 300 0.52 1.61 0.87

9 TI 500 100 0.17 0.18 0.81

10 HCS 500 200 0.50 1.16 0.72

11 HCS 220 300 1.00 2.50 0.99

12 TI 500 200 0.13 0.16 0.83

13 TI 320 400 0.65 0.69 0.81

14 HCS 320 200 0.42 1.30 0.93

15 TI 500 400 0.45 0.48 0.80

16 HCS 320 100 0.55 1.52 0.97

17 HCS 320 300 0.70 1.64 0.61

18 HCS 500 400 1.68 3.37 0.69

19 TI 220 200 1.04 0.86 0.98

20 HCS 220 100 1.94 4.27 0.92

21 TI 320 300 0.37 0.37 0.77

22 HCS 220 400 3.50 7.00 1.50

23 TI 500 300 0.87 0.73 0.35

24 TI 220 300 0.5 0.45 1.20

4 Analysis of data

The results obtained were analysed with Minitab 14 software. Analysis of Variance (ANOVA) was performed on the experimental data depicted in Table 7. Main effects for all the factors undertaken in the study were assessed for data means of the response variables-MRR, TWR and surface roughness. The two-way interactions among the process parameters were also tested for their significance. Finally, the Newman-Keuls test was performed on each process parameter to identify the level of each factor contributing most to machining performance.


Table 7 ANOVA using balanced designs (MRR)

Source DF Seq. SS Adj. SS Adj. MS F P

Tool 1 2.33813 2.33813 2.33813 38.14 0.001* Grit size 2 3.85638 3.85638 1.92819 31.45 0.001* Power rating 3 4.91845 4.91845 1.63948 26.74 0.001*

Tool × Grit size 2 0.52014 0.52014 0.26007 4.24 0.071**

Tool × Power rating 3 2.00403 2.00403 0.66801 10.90 0.008*

Grit size × Power rating 6 1.18672 1.18672 0.19779 3.23 0.090**

Error 6 0.36787 0.36787 0.06131 Total 23 15.19173

*Significant at 5% level.

**Significant at 10% level.

Note: S = 0.247612; R2 = 97.58%; R2(adj.) = 90.72%.

4.1 Main effects due to parameters

The main effects are assessed by level average response analysis of the raw data. This analysis was done by averaging the raw data at each level of each parameter and plotting the values in graphical form. The level average responses from the raw data help in the analysis of the trend of the performance characteristic with respect to the variation in the factor under study. Figures 2 and 3 shows the main effects due to all the three parameters.

Figure 2 Main effects of parameters for MRR, TWR


Figure 3 Main effects of process parameters for surface roughness

4.2 Interactions among the process parameters

Two-way interactions among the process parameters were assessed for their effects on the performance characteristics. Three interactions that were tested are: Tool-power rating, Grit size-power rating and Tool-grit size. Figure 4 shows the interactions plots among the three process parameters for their effects on MRR. The interactions plot for TWR and surface roughness have been shown in Figures 5 and 6.

Figure 4 Interactions effects for MRR


Figure 5 Interactions effects for surface roughness

Figure 6 Interactions plot for TWR

4.3 Analysis of variance

ANOVA technique helps in obtaining the information about the effect of each process parameter on the performance characteristic of the interest. The total variation in the result is sum of variation due to various controlled factors and their interactions and


variation due to experimental error. ANOVA on the raw data has been performed to identify the significant factors and the levels for each factor that contribute most to the variation in the performance characteristics. The pooled ANOVA results for MRR, TWR and surface roughness have been given in Tables 7–9, respectively. The identification of the levels for each process parameter that contribute most to the variation in the three performance indices has been done by conducting the Newman-Keuls test.

The mathematical model for this experimentation and design would be

( )ijkm i j ij k ik jk ijk m ijkY µ T G TG P TP GP TGP ε= + + + + + + + +

Table 8 ANOVA using balanced designs (TWR)


Tool 1 30.1773 30.1773 30.1773 264.42 0.000*

Grit size 2 10.7844 10.7844 5.3922 47.25 0.000*

Power rating 3 13.6077 13.6077 4.5359 39.74 0.000*

Tool × Grit size 2 4.5691 4.5691 2.2846 20.02 0.002*

Tool × Power rating 3 7.2212 7.2212 2.4071 21.09 0.001*

Grit size × Power rating 6 2.4991 2.4991 0.4165 3.65 0.070**

Error 6 0.6848 0.6848 0.1141

Total 23 69.5436


**Significant at 10% level.

Note: S = 0.337829; R2 = 99.02%; R2(adj) = 96.23%.

Table 9 ANOVA using balanced designs (SR)


Tool 1 0.00122 0.00122 0.00122 0.03 0.864

Grit size 2 0.54192 0.54192 0.27096 7.11 0.026*

Power rating 3 0.16520 0.16520 0.05507 1.45 0.320

Tool × Grit size 2 0.03385 0.03385 0.01692 0.44 0.661

Tool × Power rating

3 0.05655 0.05655 0.01885 0.49 0.699

Grit size × Power rating

6 0.22775 0.22775 0.03796 1.00 0.502

Error 6 0.22857 0.22857 0.03810

Total 23 1.25506


Note: S = 0.195181; R2 = 81.79%; R2(adj) = 30.19%.

With Yijkm as the measured response variable (MRR, TWR, etc.), µ as a common effect in all observations (the true mean of the population), Ti as the tool type effect (where i = 1, 2) with level 1 the HCS tool, Gj as the grit size effect (j = 1, 2, 3) with level 1 the 220 grit size, Pk as the power rating effect (k = 1, 2, 3, 4) with level 1 the 100 W


power rating, TGij as the effect of the two way interaction between tool and grit size and so on. εm(ijk) is the effect produced by random experimental error.

5 Results and discussion

After analysing the results obtained with ANOVA, the effect of each parameter on MRR, TWR and surface roughness was studied. The trends of variation for various performance characteristics have been observed and are discussed here.

5.1 Material removal rate

MRR for titanium (ASTM Grade I) workpiece was calculated as the loss of weight per unit time at all combinations of tool-grit size-power rating. The values against different experimental conditions have been tabulated in Table 3 and have also been plotted as mean effects of each process parameter in Figures 2 and 3.

It can be observed that high carbon steel as tool material results in better machining performance in terms of MRR of titanium workpiece. This can be attributed to higher hardness of the high carbon steel than titanium alloy. In USM, the indentation of abrasive grains in work and tool is inversely proportional to the hardness ratio of tool and work materials. Hence, use of a harder tool results in more indentation in the workpiece as compared to tool, increasing the MRR (Komariah and Reddy, 1993).

The MRR obtained has been found to increase with the increase in coarseness of the abrasive grains. This is again, in accordance with the findings of most investigators (Lee et al., 1997; Smith, 1973; Treadwell et al., 2002; Zhang et al., 1999). As shown in Figure 2, MRR drops rapidly with change in the grit size from 220 to 320 and drops further from 320 to 500, but at a diminishing rate. Coarser grains result in more energy input to workpiece per impact thereby, removing larger chunks of material. After crossing the grit size of 320, the mean particle size of the abrasive grains comes closer to the mean gap between the work and tool as well as the amplitude of vibration, which promotes efficient machining of workpiece. The effect of lesser energy input to the workpiece is thus, lesser-pronounced, leading to lesser variation in MRR.

As far as the effect of power rating of USM on MRR is concerned, any increase in power rating is expected to improve MRR as the momentum with which the abrasive particle striking with the workpiece increases manifold with a small increment in power rating. As titanium is comparatively a tough material with good strain hardening capacity, the relative tool-work hardening plays a large role in the indifferent behaviour of MRR with change in power rating. At all grit sizes of alumina, the MRR obtained has been found to increase sharply while increasing the power rating from 300 to 400 W. The increase in MRR from 100 to 200 W has been found to be marginal, whereas MRR drops marginally with increasing power level from 200 to 300 W.

The significance of the process parameter for their effects on MRR has been assessed by performing ANOVA. The results have been summarised in Table 7. All the three parameters have been found to be highly significant at 95% confidence level. Two-factor interaction of tool-power rating has also been found significant. The remaining two interactions grit size-power rating and tool-grit size are significant at 90% confidence level.


5.2 Tool wear rate

TWR for both the tools at different grit size-power level combinations has been plotted in Figure 2. TWR has been calculated as loss of weight of the tool after each experiment per unit time. It can be observed that TWR for high carbon steel tool is more than tool made of titanium alloy by 3–5 times at each power rating. This again, can be explained on the basis of hardness ratio of both the materials involved. Titanium tool experiences lesser tool wear due to its work hardening capability. As a result of the repeated impacts of abrasive grains on the tool surface, titanium being lesser harder and strain hardenable, undergoes significant amount of plastic deformation before fracture. Hence, the TWR of titanium alloy is remarkably less. High carbon steel, on other hand being harder and brittle is worn out at a rapid rate by brittle fracture of the surface.

Regarding grit size of the slurry used, TWR pattern is almost same as that followed by MRR. Use of coarse abrasive grains results in stronger impacts on the tool surface and hence, the rate of fracture increases. This can also be visualised from the interactions plot of various parameters for TWR (Figure 6) that for a given tool material, TWR is maximum at that particular grit size-power level combination which corresponds to maximum MRR. In other words, TWR is maximum at the points of maximum MRR. This phenomenon has also been put forward by other investigators (Adithan, 1981; Jadoun et al., 2006; Smith, 1973; Thoe and Aspinwall, 1998). In USM process, the parameter settings that result in maximum MRR also involve maximum TWR. This is the inherent characteristic of the process.

The behaviour of TWR with respect to power rating is in line with MRR. So, it can be explained on the similar grounds. Power level of 400 W leads to a rapid increase in TWR for both tools. ANOVA results conform the significance of all the three parameters-tool type, grit size and power rating at 5% level. All the process parameters are equally significant as far as their effect on TWR is concerned. Two-way interactions of tool-grit size and tool-power rating are significant at the chosen level of significance and interaction of grit size-power rating is marginally significant (Table 8).

5.3 Surface roughness

The centre average value of surface roughness for the holes machined in the titanium workpiece was measured by using digital perthometer. For each hole, three readings were taken and averaged to get the final value for each hole. As each experimental run was replicated twice, so the procedure was repeated for three samples for each run.

From the main effects plot (Figure 3) for SR, it can be observed that the factor tool has almost negligible effect on surface roughness. In case of TI31 tool, surface roughness decreases with increase in MRR (with grit size 220 and 500), which is contrary to the normal observation. However, it behaves differently with 320 grit size of alumina (Figure 5).

Power levels of 300 and 400 W have significant effect on roughness whereas grit size has a strong effect at all three levels. Surface roughness of the machined surface has been found to be least at power level of 300 W. This can be attributed to the least MRR observed at this particular power rating. The rate at which the fracture propagates through the work material affects the height of microcavities generated in the surface. Any increase in MRR of the work material corresponds to the creation of larger size


microcavities in the surface, which is associated with more roughness. The sharp increase in roughness at power level of 400 W and grit size of 220 can also be explained on this basis.

From the ANOVA results for surface roughness as response variable (Table 9), grit size is the only factor that appears to be significant. No two-way interaction is of significance, whereas the interaction plots for SR depict strong interactions between all the factors particularly, tool and grit size. However, the conclusions given by ANOVA summary are final in this regard.

6 Identification of most significant levels for each parameter

After testing the process parameters and their interactions for significance for their effect on the variation in the response variables, it is equally important to establish the most significant levels for each factor that can contribute most to the machining performance of USM. Newman-Keuls test was applied to the data means for each level of each parameter. The data means for different levels of each factor are summarised in Table 10.

Table 10 Data means for different levels of parameters

Factor Factor level Mean for MRR Mean for TWR Mean for SR

HCS (level 1) 1.27 2.75 0.90 Tool

Ti alloy (level 2) 0.62 0.60 0.88

220 (level 1) 1.50 2.66 1.09

320 (level 2) 0.74 1.35 0.87

Grit size

500 (level 3) 0.56 1.10 0.72

100 (level 1) 0.67 1.35 0.90

200 (level 2) 0.74 1.25 0.86

300 (level 3) 0.64 1.21 0.80

Power rating

400 (level 4) 1.72 3.04 1.03

6.1 Test on power rating for data means (MRR)

Newman-Keuls test has been performed on data means for various levels of power rating factor (for MRR as a response). The steps in conducting the test are detailed here.

1 The means associated with power rating are:

a Level 1 2 3 4

b Mean 0.67 0.74 0.64 1.72

2 Mean error sum of squares for MRR is 0.06 (Table 4) with 6 DF. Hence the standard error is calculated as square root of 0.06/6, which comes out to be 0.1.


3 From Statistical table for ranges, with n = 6 and a = 0.05 (level of significance), the values of three ranges (2, 3, 4) obtained are:

a 2 3 4P

b 3.46 4.34 4.90R

4 The Least Significant Ranges (LSR) are calculated by multiplying standard error 0.1 with each of the values of R. Hence LSR obtained are

a 2 3 4P

b LSR 0.346 0.434 0.49

5 By starting with the difference between the largest and the smallest value for the means and comparing it with largest value of LSR and then moving towards the difference between largest and second smallest for data means and comparing it with second largest LSR and so on, we get

4 3 4 3

4 1 4 1

4 2 4 2

1.72 0.64 1.08 0.49 hence

1.72 0.67 1.05 0.434 hence

1.72 0.74 0.98 0.346 hence

Y Y

Y Y

Y Y

µ µµ µµ µ

− = − = > >

− = − = > >

− = − = > >

2 3 2 3

2 1 2 1

1 3 1 3

0.10 0.434

0.07 0.346

0.03 0.346

Y Y

Y Y

Y Y

µ µµ µµ µ

− = < ≈

− = < ≈

− = < ≈

As it is evident from the comparisons given above, level 4 of power rating (400 W) has its mean value (µ) significantly higher than the mean values at other levels (levels 1–3). So it can be concluded that for power level factor, level 4 contributes the most to the variation in MRR. The other levels of this factor have their means almost equal so their contribution is very less.

All the factors were subjected to the Newman-Keuls test for all levels and the results obtained have been summarised in Table 11. On the basis of these results, the levels for each factor that contribute most to the variation in MRR, TWR and SR have also been identified and listed in Table 12. It can be concluded that level 1 (HCS) of the tool is most significant for MRR and TWR whereas level 4 of the power rating factor that is 400 W. For grit size factor, level 1 (220 mesh size) have been found to be most significant for all performance characteristics. For surface roughness, as tool and power rating factors are not having a significant effect (ANOVA Table 9) hence no level of these factors can be concluded as most significant. The macromodel developed for optimised conditions of MRR, TWR and surface roughness has been given in Table 13. It can be seen that the conditions that correspond to optimum surface roughness are only dependent on grit size of the abrasive slurry.


Table 11 Newman-Keuls test results for all factors

Factor Results for MRR Results for TWR Results for SR

Tool µ1 > µ

2 µ

1 > µ

2 µ

1 ≈ µ

2

Grit size µ1 > µ

2, µ

3

µ2 ≈ µ

3

µ1 > µ

2, µ

3

µ2 ≈ µ

3

µ1 > µ

2, µ

3

µ2 ≈ µ

3

Power rating µ4 > µ

1, µ

2, µ

3

µ1 ≈ µ

2 ≈ µ

3

µ4 > µ

1, µ

2, µ

3

µ1 ≈ µ

2 ≈ µ

3

µ1 ≈ µ

2 ≈ µ

3 ≈ µ

4

Table 12 Most significant levels for all factors

Factor For MRR For TWR For SR

Tool Level 1 (HCS) Level 1 (HCS)

Grit size Level 1 (220) Level 1 (220) Level 1 (220)

Power Rating Level 4 (400 W) Level 4 (400 W)

Table 13 Macromodel for MRR, TWR and SR

For MRR

Tool material : high carbon steel (1095)

Grit size : 220

Power rating : 400 W

For TWR

Tool material : titanium alloy (ASTM Gr. V)

Grit size: 500

Power rating: 300 W

For surface roughness

Tool material: HCS/Ti Alloy

Grit size: 320/500

Power rating: 100–400 W

7 Conclusions

This work presents the identification of process parameters for USM process that put a significant effect on the machining performance of the process for titanium as work material and subsequently, the most significant levels for each parameter undertaken in the study. ANOVA was performed on the response variables data produced from the experimentation which was designed by using full factorials approach. The significance of all the three parameters studied was conformed by ANOVA results. For MRR and TWR, all the three parameters were found to be significant for their main effects. Two-way interactions among tool and power rating, tool and grit size and grit size-power rating were also significant for MRR and TWR. For surface roughness, grit size of the abrasive slurry was the only factor found significant. None of the two-factor interactions were found to have any significance for surface roughness.


High carbon steel as the tool material was found to contribute most to the variation in MRR and TWR. For Grit size factor, level 1 (mesh size 220) was identified as the most significant for its effect on the variation in MRR, TWR and surface roughness. For power rating factor, level 4 (400 W) has been most significant for MRR and TWR. The other important conclusions that can be drawn from this study are

1 The MRR titanium (ASTM Grade I) has been found to depend on the tool material, grit size of the slurry used and power rating. Best results for MRR have been obtained with high carbon steel, grit size 220 of the alumina slurry and power rating of 400 W.

2 The TWR for titanium alloy (ASTM Grade V) is very less as compared to high carbon steel as tool material.

3 TWR has been found to be maximum at the points of maximum MRR. Optimum results for TWR were obtained with titanium alloy as tool material, grit size 500 and power rating of 200 W.

4 Surface roughness of the machined surface has been found to depend on grit size of the slurry used. Tool material and power rating have negligible effect on surface roughness. Optimum values for surface roughness were obtained with grit size 500 for alumina for both tool materials.

5 The surface finish obtained in USM of titanium is better than many other non-traditional machining processes.

Acknowledgements

The authors would like to thank Mr. K. Ramesh (General Manager, Mishra Dhatu Nigam Limited, Hyderabad), Mr. S.S. Arora (Manager, Punjab Abrasives Limited, Mohali) for providing the necessary materials for our research work. The authors are thankful to Mr. Trilok Singh and Mr. Sukhdev Chand (Lab Superintendents, T.I.E.T. Patiala) for providing laboratory facilities. The authors are also thankful to Mr. Charlie White (SONIC-MILL, Albuquerque, NM) and Dr. Rupinder Singh Khalsa (Assistant Professor, Guru Nanak Dev Engineering College, Ludhiana) for providing technical advice and support.

References

Adithan, M. (1981) ‘Tool wear characteristics in ultrasonic machining’, Tribology International, Vol. 14, No. 6, pp.351–356.

Aspinwall, D.K. and Kasuga, V. (2001) ‘The use of ultrasonic machining for the production of holes in Y-TIAL’, Proceedings of the 13th International Symposium for Electro Machining ISEM XIII, Vol. II, Spain, pp.925–937.

Astakhov, V.P. (2004) ‘An application of random balance method in conjunction with Plackett-Burman screening tests in metal cutting’, Journal of Testing and Evaluation, Vol. 32, No. 1, pp.32–39.

Ezugwu, E.O. and Wang, Z.M. (1997) ‘Titanium alloys and their machinability – a review’, Journal of Materials Processing Technology, Vol. 68, pp.262–274.

Farago, F.T. (1980) Abrasive Methods Engineering, Industrial Press, Vol. 2, pp.480–481.


Gilmore, R. (1989) ‘Ultrasonic machining and orbital abrasion techniques’, SME Technical Paper (series) AIR, NM89-419, pp.1–20.

Gilmore, R. (1990) ‘Ultrasonic machining of ceramics’, SME Paper, MS90-346, p.12.

Gilmore, R. (1991) ‘Ultrasonic machining-a case study’, Journal of Material Processing Technology, Vol. 28, pp.139–143.

Haslehurst, M. (1981) ‘Manufacturing technology’, English Universities Press Ltd., pp.270–271.

Hicks, R. and Turner, V.K. (1999) Fundamental Concepts in the Design of Experiments, 5th edition, Oxford University Press, pp.134–139.

Jadoun, R.S., et al. (2006) ‘Manufacturing process optimization for tool wear rate in ultrasonic drilling of engineering ceramics using the taguchi method’, International Journal of Machining and Machinability of the Materials, Vol. 1, pp.94–114.

Kazantsev, A.F. (1973) ‘Ultrasonic cutting’, in L.D. Rozenberg (Ed). Physical Principles of Ultrasonic Technology, Vol. 1, New York: Plenum Press, pp.3–37.

Kennedy, D.C. and Grieve, R.J. (1975) ‘Ultrasonic machining-a review’, The Product Engineering, Vol. 54, No. 9, pp.481–486.

Komariah, M. and Reddy, P.N. (1993) ‘Relative performance of tool materials in ultrasonic machining’, Wear, Vol. 161, Nos. 1–2, pp.1–10.

Kremer, D. (1991) ‘New developments on ultrasonic machining’, SME Paper, MR91-522, p.13.

Lee, T.C., et al. (1997) ‘Mechanism of ultrasonic machining of ceramic composites’, Journal of Materials Processing Technology, Vol. 71, pp.195–201.

Lin, Y.C. and Yan, B.H. (2000) ‘Machining characteristics of titanium alloy using a combination process of EDM with USM’, Journal of Materials Processing Technology, Vol. 104, pp.171–177.

Moreland, M.A. (1984) ‘Ultrasonic impact grinding: What it is: What it will do, 1984’, Proceedings of 22nd Abrasive Engineering Conference: Abrasives and Hi-Technology, pp.11–117.

Moreland, M.A. (1988) ‘Versatile performance of ultrasonic machining’, Ceramic Bulletin, Vol. 67, No. 6, pp.1045–1047.

Sharman, A.R. et al. (2001) ‘Ultrasonic assisted turning of gamma titanium aluminide’, Proceedings of 13th International Symposium for Electromachining, 9–11 May 2001, Spain, pp.939–951.

Singh, R. and Khamba, J.S. (2006) ‘Ultrasonic machining of titanium and its alloys-a review’, Journal of Materials Processing Technology, Vol. 173, pp.127–131.

Smith, T.J. (1973) ‘Parameter influences in ultrasonic machining’, Ultrasonics, Vol. 11, No. 5, pp.196–198.

Thoe, T.B. and Aspinwall, D.K. (1998) ‘Review on ultrasonic machining’, International Journal of Machine Tools and Manufacture, Vol. 38, No. 4, pp.242–246.

Treadwell, C., et al. (2002) ‘Material removal rate analysis in rotary ultrasonic machining: designed experiments’, Journal of Materials Processing Technology, Vol. 129, pp.334–339.

Wansheng, Z., et al. (2002) ‘Ultrasonic and electric discharge machining to deep and small hole on titanium alloy’, Journal of Materials Processing Technology, Vol. 120, pp.101–104.

Zhang, Q.H., et al. (1999) ‘Material removal rate analysis in ultrasonic machining of engineering ceramics’, Journal of Materials Processing Technology, Vol. 88, pp.180–184.

jatinder kumar j.s. khamba s.k. mohapatra*libvolume6.xyz/.../ultrasonicmachiningusmnotes1.pdf ·...

Documents