An investigation into ball burnishing process of magnesium alloy on CNC lathe us

Suleymn Cinr Cgn,Ctlin Iulin Pruncu,Bert Bris Buldum

a Department of Mechanical Engineering,Mersin University,33343,Ciftlikkoy,Mersin,Turkey

b Department of Mechanical Engineering,Imperial College London,Exhibition Road,London SW7 2AZ,United Kingdom

cMechanical Engineering,School of Engineering,University of Birmingham,Birmingham B15 2TT,United Kingdom

Received 29 January 2020;received in revised form 5 May 2020;accepted 15 June 2020 Available online 7 August 2020

Abstract Ball burnishing routine permits through a simple,fast and economical manner to obtain free chip on the manufactured parts.It generates a superior surface finish by rotating a ball tool against a workpiece.The burnishing process is commonly developed in industry in order to improve the surface quality,which is a critical issue in the manufacturing sector.An experimental study were carried out to determine the best surface quality for magnesium alloy subjected to different medium.Burnishing of magnesium alloy was performed varying four different mediums and combining different burnishing parameters.To design the experiment were used the classical Taguchi method through which were developed the L16 orthogonal array.This strategy allowed to detect the driving parameters that generate the best surface roughness value by computing the signal-to-noise ratio.The driving parameters values for this study are 400N(force),0.05mm/min(feed rate),three number of passes and boron oil as medium.The results are paramount important for designing heavy parts used in transportation vehicles such as automobiles,airplanes,high-speed trains etc.

Keywords:Magnesium alloy;Surface roughness;Medium;Taguchi method;Ball burnishing.

1.Introduction

Magnesium alloy is considered the future material for transportation industry(i.e.automotive,aerospace)and/or defense industries because their high potential to be used for lightweight products which can enable reducing the CO2 emission.It has lower density(1.74g/cm3)when comparing to other materials(steel,titanium),a good balance between strength and weight ratio,superior durability and appropriate machining properties[1–5].Their superior properties make magnesium alloys attractive for designing of heavy parts used in transportation vehicles such as automobiles,airplanes,high-speed trains etc.Indeed,it is an appropriate choice cause the weight of magnesium is almost close to plastic materials one and its hardness properties is comparable to metal materials one[1–5].

In this period of global competition,the quality of the surface has to be improved in order to ensure the long life of machined part.The morphology of a machined surface is one of the most important criteria in terms of quality[6,7].Each parameter used during machining operation of the materials may influence the surface roughness[8].The literature endorse that the surface roughness is a key parameter to describe the quality of manufactured surface[9,10].The surface irregularities resulting from the cutting tool or other micro morphology generated during the manufacturing process are referred to as roughness.There,in order to generate a superior surface quality of manufactured parts is used an advanced burnishing process[11].

Fig.1.Ball burnishing operation[20].

Mechanical machining is considered a crucial step on production.It is particularly important for finishing of magnesium alloys used in a wide range of applications,from the automotive industry to the biomedical industry[12–15].One of the last step,during machining process,is the use of burnishing process which allows to improve the components quality(i.e.surface hardness,fatigue strength and/or wear resistance)by reducing the roughness of machined parts[16,17].This process is a method of mechanical surface treatment which is carried out quickly and practically with the help of an apparatus which is easy to manufacture and low cost[18].Because of these properties,burnishing method finds numerous application in industry[19].Ball burnishing is a cold working process used for removing the surface roughness.It requires an applied pressure to be enforced to a harder ball acting as workpiece surface.Fig.1.presents schematically the ball burnishing operation[20].

The main key parameters affecting the material surface quality during the ball burnishing process are:the workpiece materials,ball materials,ball size,lubricants,passes,feed rate,speed and passes.In literature are only few studies dedicated to investigate different materials submitted to specific forces,feed,passes and speed parameters in order to prove the benefits of ball burnishing operation[21–25].

The major studies dedicated to evaluate the development of material surface properties subjected to ball burnishing process were focused on aluminum,super alloys,titanium and nickel.It was noted that there is no research focused on magnesium and different composite materials[21–27].Therefore,it is believed very important to investigate the behavior of these later materials when are subjected to ball burnishing and the results helps to enrich the literature.Besides,it is important to stress the necessity of investigating the ball burnishing applications for different ambient conditions.In order to overcome this gap,in the present study we developed an experimental methodology that enabled to obtain fundamental knowledge in terms of surface quality produced by burnishing of magnesium AZ91D alloy using different medium.

Table 1AZ91D magnesium alloy chemical composition of(wt%).

2.Materials and methods

2.1.Material

AZ91D magnesium alloy bar samples procured from Yuanhong Alloy Materials Co.,Ltd,China)were used for experimentation.Table 1 presents the material chemical composition.In this survey,we have selected a magnesium alloy,because it fit well with automotive,aviation,aerospace,manufacturing and defense industries requirements.It has the lowest density among all structural metals,high specific resistance,superior castability and easy processing when subjected to high cutting speeds.

In the development protocol were used two specimens of AZ91D.They were bounded by the following dimension:a nominal 30mm diameter with a length of 250mm.Each specimen was sectioned in another 8 different samples that had 20mm length and contain 5mm groove between them.Each segment was subjected to a separate experiment.To avoid the oxidation problem the magnesium materials were covered with a stretch film.It allows to avoid the contact between air and magnesium,therefore reducing the potential of oxidation.

2.2.Burnishing conditions

A Liouy Hsing GNC-450L CNC lathe were used to simulate the ball burnishing process.Fig.2a presents the main experimental setup used in this survey whereas in Fig.2b were presented the configuration of the burnishing tool.The burnishing apparatus involved in these experiments protocol were made of AISI 52100 steel ball having a diameter of 8mm.Their polished surface has a hardness on the scale 2 Brinel of 105N×mm−2.The samples made of AZ91D alloy that were submitted to ball burnishing has a nominal hardness of 78.75N×mm−2.The selection of a tool with higher hardness in respect to workpiece enable superior crush strength linked to the applied thrust force.It may dictate the surface roughness peaks crush associated to feed rate of applied force and magnitude on to burnishing ball tool.In order to detect effect of burnishing force during the ball burnishing process we have analyzed the progress of force gauge.It was considered a continuum cleaning routine,to keep the ball free of defect,and to ensure no hard particles get at the contact interface between the tool and specimen,respectively.The hard particles generally have a detrimental effect,producing specific deep scratches that transform the later one in damage accumulation on the specimen burnished surface.To ensure the appropriate tolerances the workpiece were fixed into three jaw chuck and another tailstock center of the machine.

Fig.2.a)A picture with the experimental set up and b)The burnishing tool.

Fig.3.Schematic illustration of surface roughness profile.

2.3.Measurements

The measurement of surface quality was evaluated by using the surface roughness profile,because it provides details of material surface texture.It shows the micro-level changes that occur at the interface topographical structure.In this study,the surface roughness were computed using an arithmetic average of absolute values[24].Fig.3 presents the main entities of a rough surface that enable computing of arithmetic mean value(Ra).It is defined as:

Therefore,

The quality of surface roughness obtained in manufacturing sector allows predicting with accuracy the mechanical properties(i.e.fatigue strength,corrosion resistance)and the number of cycle during the working period of a manufactured part[28].Therefore,the surface roughness has a strategic applicability cause if there are any irregularities,on the surface,generates nucleation of cracks or corrosion that lead to catastrophic failure.Even there are always a strict requirements to produce superior surface quality in terms of surface roughness,the engineers prefer a balance between quality and material strength otherwise the machining process is cumbersome and too expensive to manufacture[28].

Here,in order to obtain a good reproducibility,the surface roughness values were measured in five distinct locations of the burnished workpiece(AZ91D magnesium alloy).A Mitutoyo portable roughness meter having the model Surftest SJ 201 were used during these surface roughness measurements.In addition,surface topography of the unburnished and burnished samples was examined through a non-contact optical stereo microscope(SteREO Discovery.V8;Zeiss,Jena,Germany).

2.4.Taguchi method and experimental design

The Taguchi method was developed to identify the optimal,reliable and optimized parameter from experimental combinations that finally permits to reduce the number of experiments.[29].The Taguchi method is a simple and very effective/systematic tool to optimize the design into an economical view and qualified systems[29,30].

The criterion applied during our measurement and analysis of results for the quality characteristics under the Taguchi test design method were the S/N ratio.In this method,the signal(S)represent the actual value that the system produces and is desired to measure.The noise(N)indicates the proportion of unwanted factors in the measured value.Three different factors are considered.Such as,we have used the following condition:‘Nominal is better’,‘Smaller is better’and‘Larger is better’[31].The S/N ratio were designed through evaluating the system performances sensitivity in respect to designed parameters and the system performances sensitivity against the noise factor.As target for computing the S/N ratio,here,were considered“Smaller is better”which allows to obtain superior results at higher performances.The mathematical equation describing‘Smaller is better’were introduced in(3).

Yirepresent the numerical value of surface roughness;n endorse the tests number developed and N represent the total number of data points used in Eq.(3).

The parameter design of the Taguchi method has been developed in order to optimize the burnishing process with the performance characteristics of the experiments applied,including the following steps[32]:

•Problem identification

•Performance characterization and determination of measurement systems

•Determination of the factors and levels affecting the performance characteristic

•Determination of the interaction between the factors

•Selecting the right orthogonal array

•Determination of performance statistics

•Realization of tests and recording of results

•Analysis of results

•Doing verification tests

Table 2 presents the L16orthogonal array developed from this experiment.It contains data of four different parameters which were simulated at four different levels.Further,Table 3 introduces the process parameters and their limits.During the experiment analysis were not considered an interaction process of the main factor,instead,we have used an individual computation for each experiment to obtain the S/N ratio.

Table 2Experimental design with Taguchi method.

Table 3The main burnishing parameters used and their limits.

Fig.4.Interaction graph of parameters for S/N ratio evaluated for the surface roughness value.

3.Results and discussion

3.1.Experimental analysis results

Table 4 presents details of results gathered for the values of surface roughness and the S/N ratios computed from the data tests results after manufacturing the AZ91D magnesium alloy with different burnishing parameters(i.e.Force,feed rate,number of passes and medium).The interaction of the main factors driving the burnishing process is shown in Fig.4.In this way,the factors that interfere between them and are crossing against each other within opposite direction are confirmatory.Otherwise,the lines that were built parallel to each other denotes a non-interaction process.

In Fig.5.was plotted the evolution of parameters effect generated with this experimental setup to improve the surface roughness.It was noted that when were applied polishing it enabled a higher polishing force,however,it leads to a reduced surface roughness value of the part manufactured.Once the pressure were increased at the surface interface,by contacting the tool against the workpiece,the sub-surface layer was modified through net shaping of cavitation[18,33].Only some minor modification the surface roughness value was generated by imposing a small feed rate(between 0.05 to 0.1mm/min).But,when the feed rate was over 0.1mm/min,there,major decreases in the surface roughness value were detected.The low feed rate enables large amount of plastic deformation which in turn cause a superior surface hardness.Nevertheless,during the slow feeding rate is possible to deform more the ball surface as compared with the same location of surface worked at high feed.The work hardening produced on the burnished surface is more accentuated at low feed rate and is reduces by increasing the feed rate[34].The ball burnishing apparatus creates feed marks between two consecutive indents larger than the contact surface between the tool and the workpiece at high feed rates.Thus,the surface quality is less developed.To avoid this,the feed rate must be less than the length of the contact area[33].It has been observed that the number of passes has no effect on the process.A superior surface roughness values was generated when were used 3 number of passes.It is stated that as the number of passes increases,the surface will harden and then cause the surface layer to exfoliate[35].By corroborating 3 passes with boron oil allows obtaining the best surface roughness value whereas the lowest surface roughness value is obtained from experiments in mineral oil environment.It was noted that the roughness values obtained in dry medium and grease oil environment are almost the same.It has been found that the use of machine oils in operations greatly increases the durability of the ball to be used and improves the quality of the surface to be obtained[36].Fig.6.presents the changes of surface roughness values according to parameters imposed.By analyzing the S/N ratio we can highlights that the optimum ball-burnishing process parameters which allows obtaining the best surface roughness is produced with the followings setups:400N force,0.05mm/min feed rate,3 number of passes and boron oil as environment.To simulate these parameters in order to detect the evolution of surface roughness were used the MINITAB 18 statistical software consideringeach parameter with their level.And the MINITAB 18 statistical software were used to evaluate the main effects plot with lead to achieve the S/N ratios.

Table 4Parameters and results after burnishing process.

Fig.5.Main effects graph associated to S/N ratios in respect to surface roughness value.

Fig.6.Change of surface roughness values according to parameters.

Fig.7.a)Microscopic image of the surface without ball burnishing process b)Microscopic image of the best surface obtained after ball burnishing process(Burnishing conditions:Burnishing force:400N,Feed rate:0.05mm/min,number of passes:3 and as medium:boron oil c)Microscopic image of the worst surface obtained after ball burnishing process(Burnishing conditions:Burnishing force:100N,Feed rate:0.2mm/min,number of passes:3 and medium:mineral oil).

In Fig.7.,were presented surface micrographs of workpiece submitted/unsubmitted to ball burnishing process.Untreated part was depicted in Fig.7a.The turning process path can be seen clearly.The deep grooves and peaks formed surface topography.In Fig.7b and c work-piece surface section which made ball burnishing process was depicted.Fig.7b shows an image acquired from the specimen which reveals the best surface roughness value.This image supports the numerical values obtained as a result of experiments.A microscope image represented the coarsest results obtained in the experiments was presented in Fig.7c.It was seen that the grooves and peaks in the image was no significantly rounded at the end of the process.

3.2.Analysis of variance(ANOVA)

In order to detect the optimum level for each control factor were used the analysis of means(ANOM)which permits to determine the S/N ratio.Here,were considered the best level for each factor as the one which have the highest S/N ratio value in respect to experimental data.The best surface roughness was detected through using the following combination of the control factor setting:A4,B1,C3 and D2.

By means of ANOVA we have generated the results from Table 5 that allows to identify the factors and process most effective and their statistical significance.Generally,the ANOVA have the role to evaluate the examined factors importance by selecting the output values and to verify their quality and as well as to indicate how some different levels may produce variability.Fig.8 presents the results from ANOVA simulation in terms of surface roughness in respect to S/N ratio.These results gathered with ANOVA claim that the surface roughness is controlled by the following parameters as:force(46.69%),feed rate(21.68%),number of passes(2.33%)and burnishing medium(10.25%).The strongest effect on the surface control is played by the force parameter,noted as 46.69%.Fig.8.present a graphical contribution in terms of percentage for each burnishing parameters.To further validate the most significant parameter role over the performance characteristic on the burnishing process were evaluated the F test.In literature was indicated a close relationship between the numerical value of F and its performance characteristics when are varied the burnishing(manufacturing)parameters[20,37,38].For calculations,the equations specified in 4–10 were used.

The random change in the factor is measured by the"error sum of squares"(SSE).Depending on the differences between the factors,the variation called factor effect is measured by the"sum of squares between factors"(SSA).The total variability in data measured by total squares(SST)is divided into these two components,as in regression:

Table 5Details of ANOVA results of surface roughness in respect to S/N ratio.

Fig.8.A pictorial view of contribution percentage based on burnishing parameters influence.

Fig.9.Schematic sketch of the fatigue-crack propagation in the burnished specimens[39].

Fig.10.Microstructures near the surface(a)before burnishing,(b)after dry burnishing[40].

Fig.11.Optical micrographs of(a)Homogenized alloy(b)The sample with depth of press:2mm,feed:450mm/min,passes:2(c)microstructure near the surface of that sample[41].

where k=number of factors,n=number of measurements,yij=j th measurement of factor i,=overall sample mean of all factors,i=sample mean of factor.

Factor mean squares(MSA)and error mean squares(MSE)are obtained by dividing the sum of the corresponding squares by their degrees of freedom.These degrees of freedom also depend on the size of the problem,that is,the number of factors and the number of measurements within the factors.The“Fcalc”value is calculated by dividing MSA by MSE.

The biggest F-value reveals a more significant effect of the parameters for the results determined.In our survey the force parameter has the biggest value,2.45 which is obviously the dominant factor to control the surface roughness.

3.3.Influence of burnishing process over microstructure

Ball burnishing is a work hardening process.The effect of ball burnishing on the microstructure is described as follows.The grain sizes are reduced after the ball burnishing process,and the microhardness properties of the material are improved due to the combined work hardening effect and grain refinement during the ball burnishing process.It was proved that the work hardening produced during the ball burnishing is beneficial to minimize and distribute homogenous the residual stresses on the material surfaces.It is generally accumulated on the surface after turning process.In addition,the ball burnishing process can prevents the formation of cracks by applying pressure to the surface and preventing high residual stresses on the surface.Fig 9 presents through a schematic sketch the mechanism crack propagation associated to ball burnishing routine.

Pu[40]proved by an optical microscopic analysis that the microstructures near to the surface suffer some changes because of burnishing.Fig.10 presents details of microstructure evolution before and after burnishing of magnesium alloy subjected to dry conditions.The occurrence of some deformation twins were noted on the microstructure before burnishing process(see details in Fig.10a).Once the burnishing process were applied the morphology of grain changes.Fig.10b shows that the grain size is significantly reduced by dry burnishing.

In other study carried out by Ramesh et al.[41]were investigated the ball burnishing process on different parameters of Mg–4% Zn-1% Ca alloy.The material microstructures generated through experimental tests were presented in Fig.11.Fig.11a shows the homogenized sample microstructure in which numerous grains contain twins.Fig.11b reveals clearly that after ball burnishing is created a superficial layer with different microstructure.BY applying a higher magnification(see Fig.11c)it is observed the formation of fine-grained structure which may contain harder particles.

The burnishing process helps to improve the material microstructure creating a novel substrate which is made as a gradient microstructure layer of nanocrystalline and ultrafine grain produced by deformation process.This explain why the surface roughness of the burnishing treated sample are significantly reduced and the surface microhardness is increased.There morphological changes leads to a better wear resistance[42].However,the improvement depends mainly on the setting parameters.

4.Conclusions

In this paper were have applied successfully the Taguchi method in order to decrease the number of experimentations.The results gathered from experimental and ANOVA simulations allows to fully optimize the main burnishing parameters that enable a superior surface roughness quality.The following conclusion were drawn:

•The use of signal-to-noise ratio is paramount important in order to identify the best optimum parameter combination that leads to generate the lowest surface roughness.The combination which produces a superior surface roughness value were detected by using the following parameters A4B1C3D2.

•The analysis of variance results indicated the force as driving parameter governing the surface roughness modification.It has a percentual contribution of 46.69%.The other parameters as importance are the feed rate,medium and the number of passes which contribute with 21.68%,10.25%and 2.33%,respectively.Therefore,by well setting the force and feed rate is possible to improve the burnishing process with up to∼75% These adjustments will have direct effects on surface quality and cost.

•A good balance between the force(a bigger force)and feed rate(low feed rate)enable superior surface roughness.The number of passes have only limited effects on the surface roughness quality.

•The better surface roughness values were generated by using boron oil as environment.It has been observed that the results of the experiments in the dry environment are almost the same as the results of the experiments in the grease oil environment.Mineral oil environment was found to have the lowers potential.

•As recommendation for industrial application,when is used burnishing process,in order to obtain superior surface roughness during processing of AZ91D magnesium alloy the following parameters are the best:400N force,0.05mm/min feed rate,3 number of passes and born oil as medium.

Declaration of Competing Interest

None.