Study of Parametric Influence on Dry Sliding Wear of Al-SiCp MMC using Taguchi Technique

Study of Parametric Influence on Dry Sliding Wear of Al-SiCp MMC using Taguchi Technique

Available online at ScienceDirect Materials Today: Proceedings 2 (2015) 1825 – 1832 4th International Conference on Materials ...

473KB Sizes 0 Downloads 16 Views

Available online at

ScienceDirect Materials Today: Proceedings 2 (2015) 1825 – 1832

4th International Conference on Materials Processing and Characterization

Study of parametric influence on dry sliding wear of Al-SiCp MMC using Taguchi technique U.Prakasha,*,S.L.Ajith Prasadb, H.V.Ravindrab a

Department of Mechanical Engineering, Sri Chitra Thirunal College of Engineering, Thiruvananthapuram, Kerala-695 018, INDIA b Department of Mechanical Engineering, PES College of Engineering, Mandya, Karnataka-571 401, INDIA

Abstract Tribological properties are the primary factors controlling the performance of components subjected to relative motion. In re cent years, lightweight Metal Matrix Composites (MMC) have received wider acceptance as material for components subjected to tribological applications. In the present work an attempt has been made to study the influence of operating parameters like applied load, sliding speed and sliding distance on the dry sliding wear of A356 aluminium alloy reinforced with different percentages of 23 μm SiC particulates. Experiments were designed based on Taguchi technique. The composites showed better wear resistance under the parameters tested, compared to unreinforced alloy. Based on ANOVA, sliding distance and load were found to have highest influence on the sliding wear of the specimens. Optimum values of parameters which result in minimum mass loss was determined and confirmation test was conducted to verify the same. © The Authors. Ltd. All rights reserved. © 2014 2015 Elsevier Ltd. AllElsevier rights reserved. Selection andpeer-review peer-review under responsibility ofconference the conference committee members of International the 4th International on Selection and under responsibility of the committee members of the 4th conferenceconference on Materials ProcessingProcessing and Characterization. Materials and Characterization. Keywords:Metal Matrix Composites; Wear; Taguchi Technique; Orthogonal Array; ANOVA

1. Introduction Tribological characterization of materials is crucial for modern engineering applications. Common tribological components which are used in industrial applications, where wear is one of the significant design parameters, include sliding and rolling contact bearings, seals, gears, cams and tappets, piston rings, electrical brushes, and

* Corresponding author. Tel.: +91-9895516911; fax: +91-471-2492497. E-mail address:[email protected]

2214-7853 © 2015 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the conference committee members of the 4th International conference on Materials Processing and Characterization. doi:10.1016/j.matpr.2015.07.120


U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832

cutting and forming tools. Adhesive and abrasive wear are the most common failure modes in the components operating under moderate contact stresses. Composite materials are increasingly replacing traditional engineering materials because of their advantages over monolithic materials. The development of metal matrix composites has been one of the major innovations in the field of materials science in the recent past [1]. A metal matrix composite (MMC) is normally fabricated using a ductile metal (e.g., Al, Ti or Ni) as the base material, which is normally reinforced by a ceramic material (e.g., alumina, SiC or graphite). Composites, exhibiting higher toughness, specific strength and stiffness and good wear resistance can be obtained by combining the metallic properties such as good ductility and toughness of the matrix with properties such as high strength, hardness and elastic modulus of the ceramic reinforcement. Thus, in view of increased applications of composite materials in tribological components their characterization is of major significance.The wear resistance of composites has received much attention in recent times. Research works concerning the un-lubricated sliding wear behaviour of such materials have examined the influence of variables such as the contact pressure, sliding velocity, temperature, reinforcement particle volume fraction, and particle size. A number of mechanisms have been proposed to explain the sliding wear behavior of these composites [2].Dry sliding wear study of Al 2219 reinforced with SiC particulates showed better resistance to sliding wear compared with unreinforced alloy [3]. The wear resistance was found to increase with the volume percentage of reinforcement. The wear rate was found to decrease initially, with the sliding speed before showing an increasing trend. Three distinct regimes of wear have been observed in the sliding wear study of Al-4.5%Cu-13vol% SiCp composites, at different sliding speeds. Low rates of wear are observed in Regime I (below 3 m/s ); catastrophic failure (large amount of pin material adhering to the counterface) of the composites is observed in Regime II (between 3 and 8 m/s) and extensive melting of the composites is observed to take place in Regime III (above 8 m/s) [4].Influence of normal load has also been defined by three regimes in the sliding wear study of A356 Al-20% SiCp against SAE 52100 steel counterface [5]. At smaller loads, ultra-mild wear rate regime is observed, where, the wear resistance of the composites is marginally higher than the unreinforced A356 Al. At higher loads which produce stresses greater than the fracture strength of the reinforcement particulates, the wear rates of the composite and unreinforced A356 Al are almost identical. An extensive review work on the dry sliding wear characteristics of aluminium alloy based composites has been carried out [6], discussing the influence of reinforcement volume fraction, reinforcement size, sliding distance, applied load, sliding speed, hardness of the counter face and properties of the reinforcement phase on the wear behaviour. Influence of operating parameters on sliding wear rate and wear behaviour of aluminium based MMCs have also been reported [7, 8]. It is observed from the literature that the wear behaviour of composites is influenced by operating parameters like load, speed, sliding distance and temperature as well as the material parameters like particulate size and percentage composition of the reinforcement. The task of studying the parametric influence and identifying the optimized values becomes difficult by conventional testing methods. Taguchi’s parameter design offers a systematic method for optimization of various parameters with regard to performance, quality and cost. Taguchi’s method utilizes statistically designed orthogonal arrays (matrices) to combine the various parameters and their individual levels in conducting the experiment. Taguchi’s method uses a quadratic loss function to estimate the response characteristic. A unique feature of this method is the transformation of experimental data into special forms called S/N ratios. The S/N ratio is a concurrent statistic. A concurrent statistic is able to look at two characteristics of a distribution and roll these characteristics into a single number, or figure of merit. Whenever the experimental investigation involves more number of parameters and the range of individual parameters is also large, the study becomes exhaustive. Under these circumstances, Taguchi’s method of investigation minimizes the number of experiments while providing the reliable inference about the influence of parameters using statistical techniques such as Analysis of Variance (ANOVA). Analysis of variance is the decomposition of variance, which helps in getting a better feel of the relative effect of the different factors. This method also helps in determining the interactions, if any, between the individual parameters which is not possible in conventionally carried out experiments [9, 10]. The analysis of variance helps in deciding which factor dominate over the other and the percentage contribution of particular independent variable [11]. The main two factors i.e. speed (S) and time (T) are significant parameters affecting the wear behavior. But the factor speed (S) is less significant. The interaction between load and speed (LxS) is the significant interaction. [12] In view of the above facts, the present work is taken up with the objective of studying the influence of sliding speed, sliding distance and normal load on the dry sliding wear behaviour of A356 alloy reinforced with 10% and


U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832

20% by weight of α-silicon carbide particles of 23 μm size, using Taguchi technique. It is also intended to find the optimum level of operating parameters to minimize the material loss under test conditions. 2. Experimental details 2.1. Materials The composite specimens are fabricated using aluminum A356 alloy as the matrix material. The chemical composition of the alloy is given in Table 1. The reinforcement material used is α-silicon carbide particles (SiCp) of 23μm size. 2.2. Processing The composites are fabricated by liquid metal stir casting technique. The aluminum alloy A356 is melted in a graphite crucible using a resistance-heated furnace. The preheated SiC particles were added to the melt with controlled feed rate, stirring speed and melt temperature. The stirring speed was in the range of 700-750 rpm and the processing temperature was 720-740˚C. The SiC particles were preheated at 750˚C for 2 hrs to remove the volatile contaminants on the particle surface and to artificially oxidize the surface to obtain a layer of SiO 2 which could promote better wetting. Gravity casting is used for preparing the specimen using permanent moulds.The cast material was machined to produce cylindrical test specimens 10mm in diameter and 28mm length. Table 1. Composition details of matrix material Element

% Composition





% Composition 0.1





















3. Test Procedure Dry sliding wear tests of the composite specimens were conducted using pin- on- disc test apparatus conforming to ASTM G99 standards. EN-32 hardened steel disc with a hardness of 65HRC and Ra value of 2.5–3.5 μm was used as the counter surface. The details of the test parameters are shown in Table 2. The weight loss of the specimen at different operating conditions was measured using an electronic weighing machine with an accuracy of 0.1mg. Table 2. Process parameters with their values at three levels Level

Sliding Speed (SS), ms-1

Load (L), N

Sliding Distance (SD), m

Wt.% of SiCp (COMP)
















4. Design of experiments The test parameters selected for the experiments were based on Taguchi’s standard L27(313) orthogonal array. Orthogonal arrays allow rapid estimation of individual factor effects (or main effects), without the fear of distortion of results by the effect of other factors. In orthogonal arrays, for any pair of columns, all combination of factor levels


U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832

occur equal number of times. This is called balancing property and it implies orthogonality. The L27(313) array contains 27 rows and 13 columns. The rows represent number of trials and columns represent the number of parameters or factors that can be studied. Each parameter can be studied under three levels or values. In the present investigation the test parameters chosen are (i) sliding speed (ii) normal load (iii) sliding distance and (iv) weight percentage of reinforcement. Details of wear parameters with their levels are shown in Table 2. Since the mass loss has to be minimized, smaller the better S/N ratio was chosen for the study. The wear test results were analyzed using ANOVA. Analysis of means technique was used to find the optimum levels for each parameter so that the mass loss is minimized. 5. Results and discussions Typical test results, indicating the variation of mass loss (mg) and specific wear rate (mg m-1N-1) with respect to sliding distance are shown in Fig. 1(a) through (d). It can be observed that the mass loss of all the materials increases steadily with increase in sliding distance. Also it can be seen that at most of the test conditions, mass loss is marginally higher for matrix alloy specimens compared to the composite specimens. It can also be observed that at higher loads, the specific wear rate decreases marginally with increase in speed. During the initial periods of wear, there will be asperity contact between the two contacting surfaces, and the asperities of the weaker material get sheared resulting in higher wear rate. With increase in sliding distance, asperities get evened out, the contact area increases and the surface gets work hardened, resulting in reduced wear rate of the test surface. Experimental results also indicate larger difference in mass loss, between matrix alloy and reinforced composite at lower loads and at early stages, in case of higher loads. This can be attributed to the formation of a thin layer of oxide film along the wear track of the counter surface when it is rubbing against composite specimens.




(d) Fig. 1. Variation of wear rate with sliding distance

During the rubbing action between the surfaces, the exposed particles of SiC, causes the scratching of the counter surface, subsequently resulting in oxidation of the surface. The formation of thin film of oxide layer acts as a contaminant layer between the sliding surfaces, resulting in reduced wear. At higher loads and increased sliding distance, the oxide layer is removed, resulting in increased wear loss.

U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832


Fig. 2 (a) and (b) shows the typical variation of specific wear rate with respect to the sliding speed. A marginal decreasing trend in the specific wear rate can be observed with increase in sliding speed, which can be attributed to the reduced coefficient of friction between the mating surfaces at higher speeds. The wear loss observed within the operating range of sliding velocity appears to be mild in nature. Wear loss doesn’t show severity in the operating range, indicating that mostly adhesive wear is prevailing. Influence of load on the wear rate of different test materials is illustrated in Fig.3 (a) andFig. 3 (b). It can be observed that there is an initial increasing tendency in wear, with load, for matrix alloy and 10% SiC composites, whereas the composite with 20% SiC appears to be not greatly affected by increased load. Even the matrix alloy and the 10% SiC composites are observed to be little affected at higher loads. This could be attributed to the fact that the asperities at the surface undergo severe plastic deformation at higher loads, resulting in work hardening of the surface, which in turn results in reduced wear. The 20% SiC composite, which is much harder than the other test materials, is very little affected by the load within the operating range considered.


(b) Fig. 2. Variation of wear rate with sliding speed


(b) Fig. 3. Variation of wear rate with normal load

SEM photographs of typical worn surfaces under different magnifications are shown in Fig. 4. From SEM photographs it can be observed that delamination is the dominant wear mechanism under the load and speed conditions studied. The delamination wear results in the detachment of wear particles in the form of sheets or flakes. The worn surfaces also show the development of cracks which lead to delamination.


U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832

Fig. 4. SEM photographs of typical worn surfaces under different magnifications

6. Analysis of variance From the data obtained from the experimentation conducted as per Table 2, the analysis of variance (ANOVA) has been carried out to analyze the influence of various operating parameters on the wear loss. This analysis was carried out for a level of significance of 5% (i.e., the level of confidence 95%). Table 3 shows the results of ANOVA analysis.The results show that the load (17.5%) and sliding distance (37.5%) have the highest influence on wear of the composite. The interaction between sliding speed and load (20.67%), load and sliding distance (18.9%) is also significant. Optimum value of each factor to obtain minimum mass loss is found using the response table for S/N ratios. In the present experimentation, the requirement is to minimize the mass loss or maximization of S/N ratio. Hence, the level having maximum average S/N ratio value is optimum for the corresponding factor. Average S/N ratios (db) for the factor levels are tabulated in Table 4, Fig. 5 shows the main effects plot for S/N ratios. The mean in the main effects plot is the average of S/N ratios for the factor levels in the Table 4. From Table 4 considering the maximum average S/N ratio value, the optimum levels of the parameters chosen are(i) sliding speed of 2 ms-1(ii) normal load of 10N (iii) sliding distance of 500m and (iv) composite with 20 wt% SiCp. A confirmation sliding wear test was conducted applying the optimized test parameters and the mass loss under these conditions was found to be 0.4mg (Specific wear rate 0.8 mg m-1 N-1×104.), which was found to be the least among all the test results recorded.

U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832


Table 3. Results of ANOVA Source of Variance






Sum of Squares

F, 5%

% Contribution






17.5 37.5



































Table 4. Response table for S/N Ratios Factor

Levels 1



















Fig. 5. Main Effects Plot for S/N Ratios

7. Conclusions Dry sliding wear tests have been conducted on A356 matrix alloy and its composites with 10% and 20% SiC by weight at different operating parameters. Following conclusions can be drawn based on the test results. ¾ Addition of SiC reinforcement has been found to reduce the wear rate, compared to matrix alloy. ¾ Mass loss due to sliding wear increases with sliding distance, but the wear rate is found to decrease with sliding distance, because of flattening of asperities as well as work hardening of the surface. ¾ Within the operating range, sliding velocity is found to influence the wear rate only marginally. ¾ Mass loss due to sliding is found to increase initially with load, but the rate of wear if found to decrease at higher loads, mostly due to the work hardening effect on the sliding surface.


U. Prakash et al. / Materials Today: Proceedings 2 (2015) 1825 – 1832

¾ Results of ANOVA, indicate that sliding distance and load are the major contributing factors for the mass loss. ¾ S/N ratios provide the optimum values of the operating parameters for minimum wear, which has been validated by the confirmation test. Acknowledgment The authors wish to thank the staff members of PES College of Engineering, Mandya for the co-operation during the experimental work. The authors also wish to thank the Scientists and the supporting staff of National Institute for Interdisciplinary Science and Technology (NIIST), Thiruvananthapuram for providing the A356 MMC material and the co-operation extended during the work. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

J. Van der Geer, J.A.J. Hanraads, R.A. Lupton, J. Sci. Commun. 163 (2000) 51–59. W. Zhou, Z. M. Xu, Journal of Materials Processing Technology, 63(Part 1 – 3)(1997) 358 – 363. M. Ramachandra, K. Radhakrishna, International Symposium of Research Students on Materials Science and Engineering, (2004) 2022. S.Basavarajappa, G.Chandramohan, R.Subramanian, A.Chandrashekar, Materials Science-Poland, 242 (2006) 357-366. J.K.M. Kwok, S.C. Lim, Composites Science and Technology, 59 (1999) 55-63. S. Wilson, A. T. Alpas, Wear, 212(Part 1) (19997) 41–49. A. P. Sannino, H. J. Rack, Wear, 189(Part 1 – 2) (1995) 1–19. M. Narayan, M.K. Surappa, B.N.P. Bai, Wear, 181-183(Part 2) (1995) 563-570. C. Dengan, P. Shipway, J. Wood, Wear, 251 (2001) 1444-1451. M. S. Padke, “Quality Engineering using Robust Design”, Quality Engineering, Pearson Education Inc, 2008. P. Tapan Bagchi, “Taguchi Methods Explained”, Prentice Hall India, 1993. Ravindra Singh Rana, Rajesh Purohit, Anilkumar Sharma, Saraswathi Rana, Procedia Materials Science, 6 (2014) 503–511. Shovik Ghosh, Prasantha Sahoo, Goutam Sutradhar, Advanced Materials Manufacturing & Characterization, 4 2 (2014) 81-87.