Study and Analysis of the surface integrity of lead brass and unleaded brass

MASTER THESIS

Master's Programme in Mechanical Engineering, 60 Credits

LEAD FREE BRASS

Study and Analysis of the surface integrity of lead brass and unleaded brass

Amogh Vedantha Krishna, Vijeth Venkataram Reddy.

Thesis in Mechanical Engineering, 15 credits

Halmstad 2015-08-18

PREFACE

Lead, being a toxic substance is dangerous to society and environment.

This has raised concern on the limitation of lead in products which has direct influence on the society and environment. The primary motive of this project is to develop sustainable manufacturing technique for the production of lead free brass products.

Possible ways of exposure to lead may include ingestion of lead contaminated water, inhalation of lead containing particles of dust in air. Several measures have been considered to reduce the levels of lead in products in connection with the plumbing systems, as it is in direct contact with the drinking water. But lack of success in producing lead free products economically has triggered the expulsion of lead free concept. This raises the potential risk on the health of humans and the environment due to recycling problems. This research endeavors the possibility of sustainable production for brass components without lead.

The thesis deals with the surface study of the brass samples which includes

developing a reliable and valid method for the characterization of the surface

parameters. Based on the results, the guidelines for the production of the brass

components are generated in accordance with the research methodology. The

master's project in cooperation with Chalmers University-Lund University and

MMA in Markaryd, SECO Tools along with a Lund University consortia

conducted at Halmstad University under the guidance of Prof. BG Rosén.

Abstract

The addition of lead to the copper alloys increases its machinability and reduces the overall production cost, despite copper being expensive, which makes a challenging task to replace lead. But the alarming effects of lead on human health and the recycling problems has led to the increase in concern for reducing/eliminating the use of lead in brass and other copper alloys. Many materials are considered to replace lead in brass; silicon is one such alternative.

The turned brass sample are investigated using the state of the art equipments at Halmstad University. The results obtained are controlled readings of surface parameters and is categorized using surface imaging and mapping software, Mountains Map.

This thesis characterizes the lead and the lead free brass's surface integrity for a certain cutting data. The study deals with the evaluation of selection of appropriate surface integrity parameters and summarizes the appropriate combination of cutting data to maintain the surface of the eco brass/unleaded brass on par with the leaded brass surface. The 2D and 3D surface parameters illustrates the surface functionality and its effect on the material in contact.

The research results suggest a detailed methodology for the analysis of surface topography and a comparison exemplifying differences between the two materials under study. The research provides a perplexed results and forms the basis for further investigations of the samples machined at different cutting data.

Second set of test includes comparing the Leaded brass with the unleaded brass

samples machined at 0.06, 0.1, 0.15 and 0.2mm/rev respectively. The study

focuses on the correlation of cutting feed and the surface parameters. Comparing

the results of two tests, the unleaded brass machined @ feed rate 0.2mm/rev,

200m/min, 1.5mm depth of cut posses similar surface functionality as leaded

brass.

ACKNOWLEDGEMENT

The project endures social responsibility towards the study of environmental friendly brass product. We would like to thank the faculty of mechanical department, Halmstad University, for their support and help especially:

Project supervisor, Prof. Bengt-Göran Rosén, Halmstad University, for his supervision, guidance and moral support.

Mr. Stefan Rosén, Toponova AB, for his instructions and guidance for taking measurements using profilometer and interferometer.

Mr. Pär-Johan Lööf for his help to get the samples ready for the measurements.

CONTENTS

1. INTRODUCTION ... 1

1.1 BACKGROUND ... 1

1.1.1. AIM OF THE STUDY ... 1

1.1.2. PROBLEM IDENTIFICATION ... 2

1.2. LIMITATIONS ... 2

1.3. INDIVIDUAL RESPONSIBILITY ... 2

1.4. STUDY ENVIRONMENT ... 3

2. THEORY ... 4

2.1. LITERATURE REVIEW ... 4

2.1.1. INFLUENCE OF LEAD ON MACHINING OF BRASS ... 5

2.1.2. SURFACE MEASUREMENT INSTRUMENTS ... 5

2.1.3. SURFACE INTEGRITY ... 8

2.2. PRESENT SCENARIO ... 14

2.3. PREVAILING STUDY AND RESEARCH ... 14

3. METHOD ... 16

3.1. RESEARCH METHODOLOGY ... 16

3.2. DATA COLLECTION ... 18

3.3. ALTERNATE METHODS ... 18

3.4. CHOSEN METHODOLOGY ... 21

4. RESULTS AND DISCUSSIONS ... 22

4.1. STUDY 1 ... 22

4.1.1. PARAMETER SELECTION ... 22

4.1.2. EXPERIMENTAL RESULTS ... 32

4.1.3. OBSERVATIONS BASED ON SURFACE IMAGES ... 36

4.2 STUDY 2 ... 39

4.3 DISCUSSIONS ... 56

5. CONCLUSIONS AND FUTURE RESEARCH ... 59

5.1. RESEARCH CONCLUSION ... 59

5.2. FUTURE RESEARCH ... 60

6. CRITICAL REVEIW ... 61

REFERENCES LIST OF FIGURES LIST OF TABLES LIST OF GRAPHS

APPENDIX I: DATA COLLECTION

APPENDIX II: AVERAGE AND STANDARD DEVEIATION METHOD

APPENDIX III: SELECTING APPROPRITE PARAMETERS

1. INTRODUCTION

Lead is a toxic metal which exists in nature naturally. The WHO has recognized effects of lead has a major health concern and emphasized on the action by member states to protect affected sections of the society. Intake of lead above the recommended levels, which is 5 micrograms per liter in drinking water, is dangerous to human health commonly associated with acute poisoning, chronic poisoning and it interferes with multiple organ function and tissues including cardiac, brain and reproductive tissues. Exposure to lead is a potential concern to all humans, with infants and children more prone to health effects of lead which includes damage to brain and nervous system.

Another serious threat arising out of the leaded brass components is the contamination of the alloyed material which effects in recycling when the brass or bronze components is joined to other metallic components. This causes serious threat to the environment and raises concern on recycling of components with lead content and the component attached to it as well. The lead free components implies no recycling problems and significantly low environmental threat.

1.1. BACKGROUND

Allowance of 3% lead in the manufacture of household brass products has declined the research on lead free brass products in the recent past. This has led to the increase of lead content in the eco system which has dangerous consequences associated with it as explained in the earlier section. The lead particles are dispersed on surface of the brass, so the handling of the material causes direct exposure to lead. Further the use of these materials in plumbing systems causes the water to be vulnerable to lead poisoning due to corrosion. Using lead free brass products to avoid the effects of lead is as important as producing the alternative material economically. The surface integrity of the two materials under study are compared to notice the differences and to establish a standard methodology to identify the suitable cutting parameters to obtain optimum lead free brass surface. These results of the research forms the reference for further improvement in the manufacturing process parameters and finally to accomplish the sustainable manufacturing of lead free brass.

1.1.1. AIM OF THE STUDY

This research is a part of the project which aims to reduce/eliminate the use of lead in the manufacture of copper alloys by replacement with suitable alternative.

It also aims to eliminate the use and distribution of lead in the eco system and to

summarize the results of the project into guidelines for the sustainable

manufacturing of lead free copper alloys. This paper deals with the following

sections of the project:

i. The study includes characterization of the influence of final machined lead- and lead free brass alloys’ surface integrity.

ii. To perform laboratory studies and evaluation of performance of lead-free components frictional behavior.

iii. To differentiate the two materials based on the surface parameters and to identify the surface proximity to the leaded brass surface.

1.1.2. PROBLEM IDENTIFICATION

Lead is used in brass, normally 2%-4%, to improve its machinability. This improved machinability reduces the production cost and the cycle time for the products. But the hazards of lead on the society and environment has raised concerns on its usage. This has led to the need to replace lead and to eliminate/

reduce its usage to maximum. Lead has been replaced by silicon and the results suggest that the production cycle time and the manufacturing cost increases (≈30%- 40%).

This increase in production cost has led to the research for the manufacture of lead free products economically. The study requires machinability testing and surface testing as well, to ensure that the lead free component attains and exhibits better material properties and surface characteristics compared to its leaded counterpart.

The problem identified in this study includes filtering and characterization of the surface integrity parameters of the lead and lead free brass samples obtained from the state of the art equipments. This selection of appropriate parameters is important to define the quality and to quantify the difference between the two samples.

1.2. LIMITATIONS

The study involves the selection of appropriate parameters obtained from the brass sample. The surface integrity of these turned samples may differ with respect to the real time tool condition. Nevertheless the investigations are appropriated at ideal tool conditions. The research is conducted on the leaded brass alloy, CuZn

Pb

and the lead free brass alloy, CuZn

Si

P. These investigations can be contrived for different lead free copper alloys and is limited to the mentioned samples. The number of measurements taken from each sample is five and this consideration were mainly based on observations but need to be explored further.

1.3. INDIVIDUAL RESPONSIBILITY

The study is conducted by two master's students under the supervision of an academic supervisor from Halmstad University. The responsibilities are divided into two sections; individual and combined.

Responsibilities were divided based on individual's strength in the concerned

field:

i. Amogh and Reddy both had major role in conducting the tests in the state of the art equipments availed in the research investigations.

ii. Amogh and Reddy played major roles in the investigations of the samples and extracting the profiles using mountains map software.

iii. Amogh and Reddy were equally responsible for the research and both are accountable for writing the paper.

iv. Reddy performed a major role in compiling data on cutting data, tool material, coatings and the geometry of the cutting tool.

v. Reddy had a major role in the study of lead- and lead free brass alloys properties.

vi. Amogh performed major role in collecting information on surface integrity.

vii. Amogh also contributed in gathering information on the influence of the process parameters on the material's surface integrity.

Conclusions and results were cautiously deduced under the guidance of supervisor.

1.4. STUDY ENVIRONMENT

The advantages of using lead in the copper alloys outweighs the lead free counterpart. Since certain levels of lead is permitted in copper alloys and the failure to develop the sustainable manufacturing conditions for lead free components, gradual decline in the research of lead free materials has occurred.

This has raised concerns on the increase in percentage of lead in the eco system.

Thus an immediate requirement for the technology to produce lead free components economically has led to the development of this research study.

The need for the lead free brass's surface to replicate the properties of the leaded brass has influenced the study to focus on developing a significant approach to distinguish between the surface parameters. This project can be adopted as the base for all further research on sustainable manufacturing of lead free materials.

The testing of the materials and the surface study was conducted in Halmstad

University.

2. THEORY

The properties of the two materials under investigation are discussed in this section. The importance of surface integrity and its influence on the material properties together with the surface parameters availed in this study are briefed in this section.

2.1. LITERATURE REVIEW

Brass is an alloy of copper and zinc and used mainly in low friction application due to its good tribological properties.

F

1:

[15].

The two materials under study are free machining brass/ leaded brass and the Eco

brass/ unleaded brass. The material composition and the properties (Table 1) are

based on the document provided by the Wieland-Werke AG, specialist

manufactures of copper and copper alloys. Materials mentioned in the table 1

were used in the machinability testing conducted at Lund University.

TABLE 1:RESEARCH MATERIAL'S CHEMICAL COMPOSITION AND PROPERTIES [7].

Material Machining Brass/Leaded brass

Eco brass/ Lead free Brass

Designation (En) CuZn

Pb

/ CW614N CuZn

Si

P Chemical Composition

(Reference values in % by weight)

Cu- 57.5%

Pb- 3.3%

Zn- balance

Cu- 76%

Si- 3%

P- 0.05%

Zn- balance Pb- max. 0.09%

Physical properties (Reference values at room temperature) Thermal Conductivity

(W/m*k) 113 35

Density (g/cm

) 8.46 8.25

Modulus of elasticity (GPa) 96 100

Machining properties Machinability

(CuZn

Pb

=100%) 100% 80%

Corrosion Resistance

Machining Brass exhibits Stress corrosion cracking

and dezincification in warm, acidic waters.

Eco brass exhibits good corrosion resistance due to alloying additions.

The addition of silicon reduces the risk of stress

corrosion cracking and dezincification.

2.1.1. INFLUENCE OF LEAD ON MACHINING OF BRASS

Lead is soluble in molten brass but is rejected during solidification, precipitating particles usually 1 to 10 microns in diameter, which is dispersed uniformly to achieve higher machinability. During the machining phase, such as the turning operation, thin chips are produced and these are fragmented at very short lengths which makes it readily disposable. This reduces the tool forces which almost become independent of the cutting speed and this consequently leads to brass machined for longer periods without an interruption due to replace of tools or clearing the chips. The reduction in tool forces is mainly due to the action of lead at the tool/work interface. [13]

2.1.2. SURFACE MEASUREMENT INSTRUMENTS

The surface study of the samples includes measurement of the surface using stylus

profilometer to obtain 2D surface parameters, optical interferometer for areal

surface parameters and scanning electron microscope to produce high resolution

image of surface under study. A detailed description of these devices are

discussed in this section.

STYLUS PROFILOMETER [2]

A fine stylus is dragged smoothly and steadily across the surface under examination. The graphs displays the vertical displacement of the stylus as a function of the distance travelled along the surface. 2D parameters (Roughness, profile and waviness) of the surface are obtained by surface imaging and analysis of these graphs. The contact type profilometer availed in the study is Somicronic Surfascan Stylus Profilometer shown in the figure below.

FIGURE 2:SOMICRONIC SURFASCAN STYLUS PROFILOMETER

FIGURE 3: ROUGHNESS, WAVINESS AND TOTAL PROFILE OF A NOMINAL SURFACE (12)

P-profile: represents an overall profile of the surface.

R-profile: represents the roughness of the surface.

W-profile: represents the waviness of the surface.

Surface profilometry provide geometrical and metrological information

concerning the turned surfaces, but to investigate the difference in the surface

integrity at higher level of accuracy or at higher scale, 3D parameters were

obtained using the Optical Interferometer.

OPTICAL INTERFEROMETER [2]

Optical interferometer work on the principle of interference between two beams of light, reflected from the surface under examination and from a perfectly plane reference surface, creates fringes which are digitally recorded by an array of photodiodes linked to a microprocessor. A wide range of 3D parameters are obtained by extracting the area of the sample from these fringes. The optical interferometer shown in the figure below is MICROXAM Phase Shift Interferometer.

FIGURE 4:MICROXAM PHASE SHIFT INTERFEROMETER

The sampling area measured using the interferometer is analyzed and the readings are extracted from the Digital Surf Mountains surface imaging & metrology software. The investigations were carried out at 5X, 10X and 50X magnification levels.

FIGURE 5: EXAMPLE OF 3D SURFACE OF THE SAMPLE OBTAINED FROM MOUNTAINS MAP SOFTWARE

The visual difference between the surfaces are examined using images

from the mountains map software and the images generated by the scanning

electron microscope.

SCANNING ELECTRON MICROSCOPE [2]

The Scanning electron microscope scans a focused beam of electrons on the surface to generate high resolution surface images. These images are considered in the study to analyze and visually examine the magnified image at a resolution inconspicuous to the naked eye.

FIGURE 6:SCANNING ELECTRON MICROSCOPE

2.1.3. SURFACE INTEGRITY

Surface integrity [2] is the surface condition of a workpiece after being modified by a manufacturing process. Surface Integrity has two aspects: Surface topography characteristics made up of surface roughness, waviness, errors of form, and flaws and the Surface Layer Characteristics that can change through processing such as plastic deformation, residual stresses and cracks. This research deals with the surface topography characteristics of the lead and lead free brass samples.

IMPORTANCE OF SURFACE INTEGRITY

The surface integrity of a component has influence on the material's properties.

The outcomes of changes to the surface integrity of a component effects the

design, function and assembly consideration. Surface parameters under the study

includes 2D and 3D parameters. The impact of the surface integrity of a

component affects the mating component. Certain parameters will impact the

lubrication factor and the corrosive projections will affect the other material in

contact. [2]

TABLE 2:2D PARAMETERS OBTAINED FROM THE PROFILOMETER (ISO4287)

PROFILOMETER- 2D PARAMETERS AND ITS DESCRIPTION.

Pa - Arithmetical mean deviation of the primary profile Mean height P - prof il e Pt - Total height of the primary profile Max. height PSm - Mean width of the primary profile elements Hybrid

Mean width Ra (Gaussian filter, 0.8 mm, End-effects managed)

Arithmetical mean deviation of the roughness profile Mean height R - prof il e Rz (Gaussian filter, 0.8 mm, End-effects managed)

Maximum height of the roughness profile Max. height RSm (Gaussian filter, 0.8 mm, End-effects managed)

Mean width of the roughness profile elements

Hybrid Mean width Wa (Gaussian filter, 0.8 mm, End-effects managed)

Arithmetical mean deviation of the waviness profile Wt (Gaussian filter, 0.8 mm, End-effects managed) Mean Height W - pro file Total height of the waviness profile Max. Height WSm (Gaussian filter, 0.8 mm, End-effects managed)

Mean width of the waviness profile elements

Hybrid Mean width Areal surface parameters are categorized based on the type and region and are explained in the below section:

Mean Amplitude Parameters

Amplitude parameters characterize the surface of the material based on the vertical variations of the roughness profile from the mean line.

FIGURE 7:MEAN AMPLITUDE PARAMETERS:ARITHMETIC MEAN HEIGHT (LEFT), ROOT MEAN SQUARE SLOPE (RIGHT)[11]

Mean amplitude parameters represents an overall measure of surface texture. The

figure 7 represents the 3D height parameters, Arithmetic mean height (Sa) which

expresses the average of the absolute values of Z (x,y) in the measured area and

Root mean square slope (Sq)which helps in computing the standard deviation for

the amplitudes of the surface

However these parameters are insensitive in

differentiating peaks, valleys and spacing of various texture aspects. Hence the surface has to be predefined before using these amplitude parameters to explain the nature of the surface.

TABLE 3:3D PARAMETERS OBTAINED FROM OPTICAL INTERFEROMETER CATEGORIZED BASED ON TYPE AND DOMAIN (ISO25178)

3D

parameters Description Category 3D

parameters Description Category

Sa Arithmetic Mean

Height A m pl itu de Mean

Vmc (p = 10%, q =

80%)

Core Material Volume C ore

Sq Root Mean Square Height

Vvc (p = 10%,

q = 80%) Core Void Volume S5p (pruning =

5%) Five Point Peak Height

P ea k

Std (Reference

angle = 0°) Texture Direction

S pa ti al

Sha (pruning =

5%) Mean Hill Area Str (s = 0.2) Texture Aspect Ratio Shv (pruning =

5%) Mean Hill Volume Sal (s = 0.2) Auto-Correlation Length Smc (p = 10%) Inverse Areal Material

Ratio Sdq Root Mean Square

Slope

Hybr id

Smr (c = 1 µm under the highest peak)

Areal Material Ratio Sdr Developed Interfacial Area

Sp Maximum Peak Height Sku Kurtosis

Spc (pruning = 5%)

Arithmetic Mean Peak

Curvature Ssk Skewness

Spd (pruning =

5%) Density of Peaks S10z (pruning

= 5%) Ten Point Height Ampli tude Ma xim um Vmp (p =

10%) Peak Material Volume Sxp (p = 50%,

q = 97.5%) Extreme Peak Height

Vv (p = 10%) Void Volume Sz Maximum Height

S5v (pruning =

5%) Five Point Pit Height

V all ey

Sda (pruning =

5%) Mean Dale Area Sdv (pruning =

5%) Mean Dale Volume Sv Maximum Valley

Depth

Vvv (p = 80%) Pit Void Volume

Maximum Amplitude Parameters

The parameters which describes the highest peaks and deepest valleys fall under the maximum amplitude category. The figure 8 describes the Sz parameter which gives the maximum height of the surface that is the sum of the heights of the highest point and lowest point of the surface. This parameter group helps in identifying the unusual conditions such as sharp spike or burr on the surface that may be indicative of poor material or poor processing.

FIGURE 8:MAXIMUM AMPLITUDE PARAMETERS [11]

The parameters such as S10z, which represents the ten point height and Sxp which describes the extreme peak height are categorized into the maximum amplitude.

Peak, core and valley parameters

The parameters which describes the regions of the surface profile above the horizontal plane corresponding to the height of the material ratio of p=10%

represents the peak parameters. However it differs from the maximum amplitude parameters as it not only describe the highest peaks but also the volume and areas of the profile at peaks. The maximum peak parameter Sp is normally considered in application involving sliding contact between materials. The significant area under the peaks 50% - 97.5% are considered in accordance with the function of load support. It can be used to control the overall “peak-to valley” height of the surface by not accounting for the top of the surface which may likely be easily deformed/worn and the bottom which may be easily filled in during initial surface interactions.

The core parameters describes the region in between the horizontal plane

corresponding to the two material ratios p=10% and q=80% as shown in the figure

9. It depicts the material and void volumes at the core regions. The parameters

which illustrate the lowest regions below the horizontal plane corresponding to

the material ratio p = 80%. It explains the depth, areas and volumes of the surface

profile at valley regions.

FIGURE 9:VOLUME PARAMETERS DISTRIBUTED ALONG THE PEAK, CORE AND VALLEY REGIONS [11]

The figure 9 represents the material ratio curve of the sample material. It can be clearly witnessed the regions on the top represent the peak areas, the deepest regions represent the valley and the intermediate regions represent the core areas.

Spatial parameters

FIGURE 10:AUTO CORRELATION PEAK WITH THRESHOLD OF 0.2[18]

FIGURE 11:POLAR SPECTRUM REPRESENTING THE TEXTURE DIRECTION [18]

Spatial parameters are based on frequencies of features and include the texture direction of a surface, texture aspect ratio and auto correlation length parameters.

The texture aspect ratio and texture direction defines the spatial isotropy or directionality of the surface texture. Surface with Str > 0.5 are considered as strong isotropy and Str <0.3 indicates strong anisotropic. [18]

Auto correlation length is a quantitative measure as to the distance along the

surface by which one would find a texture that is statistically different from the

original location. It is the length of the fastest decay of ACF (auto correlation

factor) in any direction. The larger the Sal parameter of the rough surface, less the

number of contact points. Based on the Hertizan contact mechanics, it is shown that the real area of contact increases with the increase in number of contact points [18]. Adhesive friction force is proportional to the real area of contact, this suggests that the adhesive friction behavior of a surface is inversely proportional to its auto correlation length [23].

Hybrid parameters

FIGURE 12:SCHEMATIC DIAGRAM OF THE HYBRID PARAMETER (SDR)[2]

Hybrid parameters are the combination of spacing and height parameters. It includes Sdq, Root mean square slope is the general measurement of the slopes which comprise the surface and Sdr, Developed interfacial area ratio is the additional surface area contributed by the texture as compared to an ideal plane and these parameters may be used to differentiate surfaces with similar average roughness. Also the kurtosis and skewness parameters provide the degree of symmetry of the surface heights about the mean plane and also helps in identifying the erroneous peaks and valleys.

Kurtosis (Sku): Sku indicates the presence of inordinately high peaks/ deep

valleys (Sku>3.00) or lack thereof (Sku<3.00) making up the texture. Surfaces

described as gradually varying, free of extreme peaks or valley features, will tend

to have Sku <3.00. Sku is useful for indicating the presence of either peak or

valley defects which may occur on a surface. It not only detects the profile spikes

but provides a measure of the spikiness of the area.

2.2. PRESENT SCENARIO

The scattering of lead particles on the parent material makes it vulnerable to the environment in contact, be it either through air or water. The lead gets into the drinking water mainly through the plumbing systems and introducing the lead free components would certainly reduce the contamination of drinking water. Stricter regulation regarding the lead levels in products have been implemented. The permitted levels of lead in household products is found to be around 3%. These regulations has forced the industry to produce environment friendly products.

FIGURE 13: PRODUCTS CONTAINING LEAD- A RADIATOR VALVE OF LEADED BRASS FROM MMA (LEFT), LEADED BRONZE BUSHING FOR A CON-CRUSHER WITH A WEIGHT OF 100 KG FROM SANDVIK SRP (MIDDLE) AND A LEAD-FREE COPPER LID FOR NUCLEAR WASTE CAPSULE WITH A WEIGHT OF 300 K

This project is in collaboration with AB Markaryds Metallarmatur (MMA) which partly finances the project and serves as the major partner. MMA products include heating system components like the radiator valve shown in figure 13. Even though these components of MMA are not affected by the regulations, they have taken the initiative to set an example by producing lead free components and this altogether leads to societal benefits.

The use of diamond tools provide significantly higher surface quality but in practice, the carbide inserts are used for machining of brass in MMA. Therefore the research is limited to uncoated carbide inserts. Future research in this project includes the influence of coated carbide inserts on the surface quality.

2.3. PREVAILING STUDY AND RESEARCH

The surface study on lead free brass is an extension of the research project which

focuses on eliminating lead usage in copper and copper alloys cited in article by

Prof. J.-E. Ståhl [15]. The machinability of the leaded and unleaded brass samples

are determined at Lund University and the corrosive behaviour of the materials

are investigated at Chalmers University.

MACHINABILITY TESTING

An investigation was carried out in Lund University to compare and evaluate the machinability of four different types of brass alloys [5]. Lead free alternative, CW724R, was tested and their material property was compared to the other alloys.

The results show that CW724R has a considerably higher strength than the other alloys. However, the material properties for CW724R are poor, compared to the others, when machining and thus leads to increased costs during manufacturing.

The author mentioned that an increased knowledge about this alloy and the cutting process could potentially lead to decreased costs.

Testing for machinability of lead and eco brass was carried out in Lund University

[7]. The machinability testing included longitudinal turning of a lead-free brass

alloy performed at different cutting data, CuZn

Si

P, as compared to a

conventionally used, lead-alloyed alternative, CuZn

Pb

. Materials were supplied

as bars having an initial diameter of approximately 50 mm. The inserts used were

uncoated H10F cemented carbide DNGA150708F inserts with an edge radius of

approximately 10±5μm. The attained results show that CuZn21Si3P has a

considerably higher strength than the lead-alloyed material resulting in

significantly higher cutting forces. This increase in cutting forces was found to

result in more rapid deterioration of the cutting tool, thus raising concerns for

higher manufacturing cost.

3. METHOD

3.1. RESEARCH METHODOLOGY

This research is an empirical study involving quantitative investigations of the qualitative parameters to distinguish the two materials under study. The research was mainly conducted in a laboratory environment and a series of similar procedure are followed during investigation of the samples.

The research methodology includes:

i. Sampling: The profile parameters is by default, ISO 4288(1996), calculated on each sampling length with the default number of samplings is five.

With surface and areal parameters, the default is one sampling area per evaluation area and five number of samplings considered. Each sampling measured is probabilistically independent.

ii. Study and measurement of the surface integrity of the brass samples using state of the art equipment available at Halmstad University.

a. Profilometer

[2] - A random cut-off filter is considered to obtain the mean width of the roughness elements (RSm) value of the respective sampling. The value of RSm is found within 0.13mm and 0.4mm with a cut-off filter 0.8mm. Roughness sampling length for the measurement of the R-parameters of periodic profiles are shown in the table 4 (ISO 4288).

TABLE 4:ROUGHNESS SAMPLING LENGTH FOR THE R-PARAMETERS OF THE PERIODIC PROFILES

RSm mm

Roughness sampling length (lr)

mm

Roughness evaluation length (ln)

mm

0.013 < RSm ≤ 0.04 0.08 0.4

0.04 < RSm ≤ 0.04 0.25 1.25

0.13 < RSm ≤ 0.4 0.8 4

0.4 < RSm ≤ 1.3 2.5 12.5

1.3 < RSm ≤ 4 8 40

1

SOMICRONIC SURFASCAN Stylus Profilometer.

The selection of the stylus tip radius depends on the cut-off wavelength is shown in table 5 (ISO 3274).

TABLE 5: RELATIONSHIP BETWEEN THE CUT-OFF FILTER AND THE STYLUS TIP RADIUS

Cut-off wavelength λc (mm)

Stylus tip radius r

(μm)

0.08 2

0.25 2

0.8 2*

2.5 5

8 10

*} For surfaces with Ra > 0.5μm or Rz > 5μm, r

= 5μm can be used without significant difference in measurement result.

b. Optical Interferometer

[2] - A wide range of 3D parameters are obtained by extracting the area of the sample from these fringes.Gaussian filter of 0.8mm is applied to filter the profiles.

c. Scanning electron microscope

[2] - produces images of the sample surface by scanning it with a focused beam of electrons. The magnified images obtained demonstrates the sample's surface topography and composition.

iii. Surface imaging and Analysis using Mountains Map Software

[18].

a. The area of sample are extracted from the readings and fringes obtained from the stylus profilometer and the optical interferometer respectively.

b. Non measured points are filled.

c. Form is removed from the cylindrical surface of the sample. ISO 25178 (part 2) defines the form removal operator as the "operation which removes form from the primary surface". The removal of form allows the areal parameters to be evaluated on the modified points to minimise the influence of form on these parameters.

d. Surface texture characterization using Gaussian filter (0.8mm) presented in ISO 16610 (part 61).

iv. Selection of appropriate parameters

The study of surface metrology includes wide range of profile and areal parameters. The problem exists in selecting the parameters which are most suitable to distinguish the materials. The research is based on mathematical and statistical approach to classify and select the significant parameters.

1

MICROXAM - Phase Shift & EX,

Jeol JSM Scanning Electron Microscope,

3

Digital Surf, www.digitalsurf.com/

3.2. DATA COLLECTION

The data regarding the lead and lead free samples are based on the research article published, a research partner of this project. The experimental setup is mentioned in the results (section 4).

The state of the art equipments availed in this research are trail run before the actual measurement and the values measured from equipments are carefully examined. The measurements taken are five for each sampling and is same for all the investigations. The data collected are entered into the mountains map for surface imaging and analyzing. The parameter results obtained are extracted into Microsoft Excel for further statistical investigation.

Reference identification: The reference for each step is important to justify and extract reliable results. The manufacturing process data and material type is the reference for the study and measurement of the surface imaging and analysis. The surface parameter readings is the reference for the characterization of the samples.

The obtained results by characterizing is the reference to spot the difference between the two materials. All data collected are carefully formulated and stored for future reference.

3.3. ALTERNATE METHODS

The analysis of the parameters is performed by passing the quantitative data through an array of steps. Three methods were tried and tested to obtain the appropriate results. Each method follows a common string of steps which includes calculating the mean and standard deviation for each parameter. Mean (μ), Standard Deviation () are calculated for each parameter. The intervals I

, I

, are calculated with the coverage factor k = 2 and further the Relative Standard Deviation, Significance are calculated [18].

The intervals,

I

=μ+ k, I

= μ- k

For different k, the following probabilities are known, k=1: 68% is covered between μ- k and μ+ k;

k=2: 95% is covered between μ- k and μ+ k;

k=3: 99.7% is covered between μ- k and μ+ k;

The calculation of Mean, Standard Deviation, I

, I

, Relative Standard

Deviation, Significance are explained in Appendix II.

Method 1: The method consisted of setting two threshold values and the parameters selected with similar explanation are eliminated based on the value of Relative Standard Deviation.

FIGURE 14:FLOWCHART OF THE ALGORITHM FOR METHOD 1 OF SELECTING APPROPRIATE PARAMETERS

Method 2: In this method, the threshold is set and the parameter with similar meaning are filtered based on the Relative Standard Deviation.

FIGURE 15: FLOWCHART OF ALGORITHM FOR METHOD 2 FOR SELECTING APPROPRIATE PARAMETERS

Method 3:

FIGURE 16: FLOWCHART OF ALGORITHM OF THE METHOD 3 FOR SELECTING APPROPRIATE PARAMETERS

In this method, threshold is set on the significance of the parameter. The significance value is determined for the parameters to identify the significant parameters. Parameters with positive significance value are considered in this method to distinguish the materials. Setting threshold to eliminate the parameters that have significance value higher than 0.1. The parameters within this range was misinterpreted as significant. The threshold for significance may eliminate the parameter which are most significant in distinguishing the parameters and in some cases misconstrue parameters with high variability to be significant.

The variability of the surface parameter readings provide the spread of the data and the higher variability implies the inconsistency in the parameter values.

Illustration of normal distribution curve with disjunct interval

High variability

Medium variability

Low variability

3.4. CHOSEN METHODOLOGY Method 4:

FIGURE 17: FLOWCHART OF ALGORITHM OF THE CHOSEN METHODOLOGY FOR SELECTING APPROPRIATE PARAMETERS

The chosen methodology for study 1 includes comparing the Leaded and Unleaded brass sample machined @feed rate 0.2mm/rev, cutting speed 400m/min and 0.8mm depth of cut:

i. Selecting the surface texture parameters with disjunct intervals (positive significance values). The parameters with overlapping intervals are considered as insignificant for comparison.

ii. The surface parameters obtained are grouped into categories based on their type and domain.

iii. The correlation between these parameters are determined.

iv. Parameter with high correlation and low relative standard deviation (high variability) are considered in the study.

The selected surface texture parameters are employed in distinguishing the samples. The results are deduced from the mean difference in percentage of these parameters. The measured values and results of the average and standard deviation method are summarized in section 4.

The methodology for study 2 includes comparing the leaded brass with unleaded brass samples machined @ feed rate 0.06, 0.1, 0.15, 0.2mm/rev respectively and at cutting speed 200m/min with 1.5mm depth of cut

i. The surface parameters obtained are grouped into categories based on their type and domain.

ii. Parameter with high correlation and low relative standard deviation (high variability) are considered in the study.

The selected surface parameters are employed in comparing between the samples

and identifying the unleaded brass sample with surface functionality closer to the

leaded brass.

4. RESULTS AND DISCUSSIONS

This section of the report summarises the results considered to discriminate the surface topography between the Leaded and Unleaded Brass. Profilometer, Interferometer and Scanning Electron Microscopes were utilized to obtain the readings of the 2D, 3D surface texture parameters and surface images. All the parameters in the study are based on and defined by the international specification standard ISO 25178-2 (2012).The readings obtained from the surface measurement instruments generated many parameters which posed a major problem in rendering appropriate conclusions in discriminating the surfaces of the samples. It was necessary to develop a significant and accurate method to minimise the number of parameters. Thus correlation matrix is generated to enable the comparison between the parameters within the categorised groups.

Correlation gives a brief idea about the statistical dependence or connection between the parameters within the groups. Parameters with a high intrinsic variability can be effectively replaced by parameters with which they have a high correlation within its respective category. The selected surface parameters are used in the study to explain the surface topography of the samples.

4.1. STUDY 1

Machining Data

General turning operation was carried out on the samples of leaded brass CW614N and eco brass CW724R with initial diameter of 50mm and the uncoated H10F cemented carbide DNGA150708F inserts with an edge radius of approximately 10±5 μm was used as the cutting tool. The cutting data remained constant throughout the operation with cutting speed of 400m/min, depth of cut being 0.8mm and feed rate of 0.2mm/rev. The five different surface readings were taken into consideration from both the samples and were evaluated to obtain the results.

4.1.1. PARAMETER SELECTION

The average and standard deviation method identifies the parameters which show significant variations but at the same time it is very important to understand the pattern of the variations. If the variations are very large it shows that the readings taken are on the damaged surface and cannot be a reliable data to discriminate between the materials. Hence the parameters which show significant variations but do not exhibit a large variations can be utilized to differentiate between the leaded and unleaded brass.

The average and standard deviation method involves calculating the mean,

standard deviation and the intervals of each parameter. Significance (Si) for each

parameter are calculated. The parameters with disjunct intervals were selected for

the study as it indicates a significant difference between the materials. This

method remains same for selecting the parameters of both 2D and 3D

measurements. Leaded brass's parameters- Mean (

), Standard deviation (

), minimum and maximum Intervals-

and

Unleaded brass's parameters- Mean (

), standard deviation (

), minimum and maximum intervals-

and

I"max

and Significance value (Si). The readings (five iterations) of all the parameters are attached in the Appendix I.

TABLE 6:THE AVERAGE AND STANDARD DEVIATION METHOD FOR SELECTING 3D PARAMETERS –10X MAGNIFICATION

Name of the

studiable Description Unit

Average and Standard Deviation Method

Leaded Unleaded Leaded Unleaded

Mean

' SD s'

Mean

'' SD

s'' I'min I'max I"min I"max

Vmc (p = 10%, q = 80%)

Core Material

Volume µm³/µm²

1.91 0.059 1.39 0.071 1.79 2.02 1.24 1.53 Vvc (p = 10%,

q = 80%) Core Void Volume µm³/µm² 2.22 0.075 1.63 0.071 2.07 2.37 1.48 1.77 Str (s = 0.2)

Texture Aspect

Ratio no unit 0.09 0.005 0.12 0.006 0.08 0.10 0.11 0.13 Sal (s = 0.2)

Auto-Corelation

Length µm 28.15 1.676 38.26 1.991 24.80 31.51 34.27 42.24

Sz Maximum Height µm 8.89 0.345 7.06 0.209 8.20 9.58 6.64 7.48

Sa

Arithmetic Mean

Height µm 1.53 0.031 1.24 0.053 1.46 1.59 1.14 1.35

Sq

Root Mean

Square Height µm 1.85 0.026 1.52 0.047 1.80 1.91 1.43 1.61 Smc (p =

10%)

Inverse Areal

Material Ratio µm 2.39 0.066 1.81 0.076 2.26 2.52 1.66 1.97 Vv (p = 10%) Void Volume µm³/µm² 2.44 0.06 1.85 0.07 2.33 2.56 1.71 1.99

Name of the studiable Unit

Average and Standard Deviation Method

Category Condition: If I'min > I"max

OR I"min > I'max - Accept

Significance Si

Vmc (p = 10%, q = 80%) µm³/µm² Accept 0.157

Core

Vvc (p = 10%, q = 80%) µm³/µm² Accept 0.158

Str (s = 0.2) no unit Accept 0.085

Spatial

Sal (s = 0.2) µm Accept 0.083

Sz µm Accept 0.090 Max Amplitude

Sa µm Accept 0.081 Mean

Amplitude

Sq µm Accept 0.112

Smc (p = 10%) µm Accept 0.140

Peak

Vv (p = 10%) µm³/µm² Accept 0.155

The table 6 presents the average and standard deviation method employed on the

parameter readings and the parameters displayed are all the significant parameters

obtained, which have a disjunct intervals. The significance value is calculated for

all the parameters and the positive significance value indicate that the parameter

has disjunct intervals and negative value indicates that the parameters have overlapping intervals when the bell curve is plotted. In other words, the disjunct interval parameters show significant difference in the surface topography between the leaded and unleaded brass samples. Selecting the main parameters from each functional group would a brief idea of the overall topography difference between the two samples. The parameter correlation study is employed to provide substantial interpretation in the above raised concern.

TABLE 7:LEADED AND UNLEADED BRASS PARAMETERS READINGS –INTERFEROMETER 10X MAGNIFICATION

Core Spatial Maximum

Amplitude

Vmc (p = 10%,

q = 80%)

Vvc (p = 10%,

q = 80%) Str (s = 0.2) Sal (s = 0.2)

Unit µm³/µm² µm³/µm² <no unit> µm µm

Leaded Brass Readings

1.91 2.29 0.08 25.64 8.91

1.99 2.24 0.09 29.55 9.39

1.89 2.21 0.09 27.22 8.68

1.83 2.1 0.09 29.22 8.99

1.91 2.28 0.09 29.13 8.48

RSD 3.10% 3.40% 5.80% 6.00% 3.90%

Unleaded Brass Readings

1.27 1.6 0.12 36.47 6.93

1.44 1.75 0.12 37.3 7.41

1.41 1.63 0.12 36.86 7.1

1.43 1.58 0.13 39.55 6.93

1.39 1.57 0.13 41.11 6.93

RSD 5.10% 4.40% 5.10% 5.20% 3.00%

Mean Amplitude Peak

Sa Sq

Smc (p = 10%) Vv (p = 10%)

Unit µm µm µm µm³/µm²

Leaded Brass Readings

1.55 1.87 2.46 2.51

1.54 1.88 2.41 2.46

1.56 1.86 2.37 2.43

1.48 1.84 2.29 2.35

1.51 1.81 2.44 2.48

RSD 2.00% 1.40% 2.80% 2.40%

Unleaded Brass Readings

1.15 1.44 1.8 1.83

1.27 1.56 1.94 1.97

1.26 1.53 1.82 1.86

1.28 1.54 1.77 1.81

1.26 1.53 1.74 1.79

RSD 4.30% 3.10% 4.20% 3.80%

The table 7 represents the selected parameters which are significant and the five

surface readings of the leaded and unleaded brass sample. The Relative standard

deviation is calculated for each parameter and is expressed in percentage. The

Relative standard deviation (RSD) also called as Coefficient of variation is the

standardized measure of dispersion of a probability distribution. RSD is necessary

to check the consistency of the parameter, higher the value of RSD higher will be the variability of the parameter.

The correlation is applied to the parameters within the categorized groups. The correlation is enabled using the readings of the leaded and unleaded brass from table 7.

TABLE 8:CORRELATION MATRIX OF CORE PARAMETERS

% Vmc (p = 10%, q = 80%)

Vvc (p = 10%, q = 80%) Vmc (p = 10%, q =

80%) 100 98

Vvc (p = 10%, q = 80%) 98 100

The table 8 represents the correlation matrix of the core parameters. It is evident that the parameters Vmc and Vvc are highly correlated, hence the parameter which has the least variation can be selected for the discrimination of the surface between leaded and unleaded brass. From table 7, the Core void volume (Vvc) parameter has least average variability, considering the relative standard deviation of both the leaded and unleaded brass sample readings.

TABLE 9:CORRELATION MATRIX OF SPATIAL PARAMETERS

% Str (s = 0.2) Sal (s = 0.2) Str (s = 0.2) 100 100 Sal (s = 0.2) 100 100

Table 9 displays high correlation between the Texture aspect ratio (Str) and the Auto correlation length (Sal). The parameter, Str is suitable as it has lower variability.

TABLE 10:CORRELATION MATRIX OF MEAN AMPLITUDE

% Sa Sq

Sa 100 99

Sq 99 100

TABLE 11:CORRELATION MATRIX OF PEAK PARAMETERS

% Smc (p = 10%) Vv (p = 10%)

Smc (p = 10%) 100 100

Vv (p = 10%) 100 100

The table 10 and 11 represent the correlation matrix of the mean and peak

amplitude respectively. The amplitude parameters Sq – Sa and Smc - Vv exhibit

high correlation coefficient. The Root mean square height (Sq) and the Void

volume (Vv) have low variability and hence are selected for the comparison between the samples.

The maximum amplitude parameter, Maximum height (Sz) is the only one parameter within its functional group and hence it can be used directly for the differentiation of the surfaces. Also in cases where parameters have least correlation, then those parameters are considered as independent parameters and both are selected for the differentiating of the surface topography. The parameters, Vvc, Str, Sz, Sa and Vv are reliable as their intrinsic variability is less and also represent the respective functional groups which provide a detailed description of the surface topography.

Similarly the parameter correlation study is conducted to the 3D parameters obtained from the Interferometer with 5X magnification and 2D parameters from the Profilometer.

TABLE 12:AVERAGE AND STANDARD DEVIATION METHOD FOR SELECTING 3D PARAMETERS INTERFEROMETER READINGS-5XMAGNIFICATION

Name of the studiable

Description Unit

Average and Standard Deviation Method

Leaded Unleaded Leaded Unleaded

Mean

- ' SD - ' Mean

- '' SD - '' I'min I'max I"min I"max

Vmc (p = 10%, q = 80%)

Core Material

Volume mm³/mm² 0.0027 0.0002 0.0021 0.00004 0.0024 0.0030 0.0020 0.0022 Vvc (p =

10%, q = 80%)

Core Void

Volume mm³/mm² 0.0031 0.0001 0.0023 0.00004 0.0030 0.0033 0.0022 0.0024 Str (s =

0.2)

Texture

Aspect Ratio <no unit> 0.08 0.0017 0.10 0.0003 0.08 0.09 0.10 0.10 Sal (s =

0.2)

Auto- Corelation Length

mm 0.04 0.0007 0.04 0.0001 0.03 0.04 0.04 0.04

Sdr

Developed Interfacial Area

% 1.43 0.04 1.27 0.0273 1.35 1.51 1.22 1.33

Ssk Skewness <no unit> -0.21 0.03 -0.48 0.0342 -0.28 -0.14 -0.54 -0.41

Sa Arithmetic

Mean Height µm 2.04 0.04 1.80 0.0268 1.97 2.11 1.75 1.86

Sq

Root Mean Square Height

µm 2.49 0.03 2.13 0.0298 2.42 2.56 2.08 2.19

Smc (p = 10%)

Inverse Areal Material Ratio

µm 3.34 0.04 2.53 0.0399 3.26 3.42 2.45 2.61

Vv (p =

10%) Void Volume mm³/mm² 0.0034 0.00004 0.0026 0.00004 0.0033 0.0035 0.0025 0.0027

Name of the

studiable Description

Average and Standard Deviation Method

Category Condition: If I'min >

I"max OR I"min > I'max - Accept

Significance S

Vmc (p = 10%, q =

80%) Core Material Volume Accept 0.09

Vvc (p = 10%, q =

80%) Core Void Volume Accept 0.23

Str (s = 0.2) Texture Aspect Ratio Accept 0.130

Spatial Sal (s = 0.2) Auto-Corelation Length Accept 0.130

Sdr Developed Interfacial

Area Accept 0.017

Hybrid

Ssk Skewness Accept 0.38

Sa Arithmetic Mean Height Accept 0.06

Mean Amplitude

Sq Root Mean Square

Height Accept 0.10

Smc (p = 10%) Inverse Areal Material

Ratio Accept 0.22

Peak

Vv (p = 10%) Void Volume Accept 0.22

The table 12 represents the selected 3D parameters which show a significant variations between the materials.

TABLE 13:LEADED AND UNLEADED BRASS READINGS –INTERFEROMETER 5X MAGNIFICATION

Core Spatial Hybrid

Vvc Vmc Sal Str Ssk Sdr

mm³/mm² mm³/mm² mm <no unit> <no unit> %

Leaded Brass Readings

0.00321 0.00284 0.0365 0.08 -0.17 1.43 0.00316 0.00277 0.0361 0.08 -0.2 1.47 0.00307 0.00248 0.0352 0.08 -0.19 1.39 0.00306 0.00264 0.0359 0.08 -0.25 1.47 0.00319 0.00282 0.0373 0.08 -0.23 1.38

RSD 2.20% 5.55% 2.14% 0.00% -15.35% 2.99%

Unleaded Brass Readings

0.002344 0.002130 0.04 0.1 -0.46 1.26 0.002285 0.002106 0.04 0.1 -0.45 1.31 0.002239 0.002018 0.04 0.1 -0.51 1.24 0.002292 0.002118 0.04 0.1 -0.45 1.26 0.002249 0.002107 0.04 0.1 -0.52 1.28

RSD 1.82% 2.13% 0.00% 0.00% -7.16% 2.08%

Mean Amplitude Peak

Sq Sa Vv Smc

µm µm mm³/mm² µm

Leaded Brass Readings

2.5 2.06 0.00342 3.36 2.51 2.07 0.00341 3.36 2.43 1.99 0.00335 3.3 2.49 2.02 0.00335 3.3 2.52 2.06 0.00343 3.38

RSD 1.42% 1.66% 1.15% 1.12%

Unleaded Brass Readings

2.15 1.84 0.002636 2.6 2.14 1.8 0.00257 2.52 2.09 1.77 0.002542 2.5 2.17 1.81 0.002582 2.52 2.12 1.79 0.00254 2.51

RSD 1.43% 1.44% 1.52% 1.58%

The table 13 represents the Interferometer 5X magnification readings of the significant parameter obtained based on the average and standard deviation method and also the relative standard difference is calculated to understand the variations of each parameters.

TABLE 14:CORRELATION MATRIX OF CORE PARAMETERS –5X

% Vmc (p = 10%, q = 80%) Vvc (p = 10%, q = 80%)

Vmc (p = 10%, q = 80%) 100 98

Vvc (p = 10%, q = 80%) 98 100

TABLE 15:CORRELATION MATRIX OF SPATIAL TABLE 16:CORRELATION MATRIX OF

PARAMETERS –5X HYBRID PARAMETERS –5X

% Str (s = 0.2) Sal (s = 0.2) Str (s = 0.2) 100 100 Sal (s = 0.2) 100 100

TABLE 17:CORRELATION MATRIX OF MEAN AMPLITUDE PARAMETERS –5X

% Sa Sq

Sa 100 99

Sq 99 100

TABLE 18:CORRELATION MATRIX OF PEAK PARAMETERS -5X

% Smc (p = 10%) Vv (p = 10%)

Smc (p = 10%) 100 100

Vv (p = 10%) 100 100

% Sdr Ssk

Sdr 100 91

Ssk 91 100

The correlation matrix is produced using the readings of the leaded and unleaded brass samples from the table 13. The selection of a parameter among the highly correlated parameters depend upon checking the average of relative standard deviation of leaded and unleaded brass samples, if the average variability is less then that particular parameter is selected. In cases where the parameters are highly correlated and have the same variability in both the leaded and unleaded brass, the parameter which is more relevant for the study is taken into consideration. The parameters Vvc, Sal, Sdr, Sq and Vv are selected from each functional groups which are have least variability.

TABLE 19:AVERAGE AND STANDARD DEVIATION METHOD FOR SELECTING 2D PARAMETERS- PROFILOMETER

Name of the studiable Unit

Average and Standard Deviation Method

Leaded Unleaded Leaded Unleaded

Mean - 

SD -



Mean -



SD -



I'

I'

I"

I"

Pa - Arithmetical mean deviation

of the primary profile µm 2.06 0.026 1.87 0.026 2.01 2.11 1.82 1.92 Pt - Total height of the primary

profile µm 11.30 0.354 9.75 0.414 10.59 12.01 8.93 10.58

PSm - Mean width of the primary

profile elements mm 0.11 0.003 0.17 0.011 0.10 0.11 0.14 0.19 Ra - Arithmetical mean deviation

of the roughness profile µm 2.02 0.028 1.85 0.030 1.96 2.08 1.79 1.91 Rz - Maximum height of the

roughness profile µm 9.60 0.254 8.24 0.283 9.09 10.11 7.67 8.81 RSm - Mean width of the

roughness profile elements mm 0.10 0.007 0.17 0.010 0.08 0.11 0.15 0.19

Name of the studiable

Average and Standard Deviation Method

Category Condition: If I'min > I"max

OR I"min > I'max - Accept

Significance S

Pa - Arithmetical mean deviation of the

primary profile Accept 0.041 P -Pr o fi le

Pt - Total height of the primary profile Accept 0.001 PSm - Mean width of the primary profile

elements Accept 0.230

Ra - Arithmetical mean deviation of the

roughness profile Accept 0.027

R -Pr o fi le

Rz - Maximum height of the roughness

profile Accept 0.032

RSm - Mean width of the roughness

profile elements Accept 0.267

Table 19 displays the selected 2D parameters which shows a significant variation between the leaded and the unleaded brass.

TABLE 20:LEADED AND UNLEADED BRASS READINGS –2DPROFILOMETER

Maximum Amplitude Mean Amplitude Hybrid

Pt (

Rz

(

Pa (

Ra

(

PSm (mm) RSm

(mm)

Leaded Brass Readings

10.9 9.38 2.02 1.98 0.11 0.1

11.78 10.01 2.09 2.05 0.11 0.1

11.19 9.64 2.06 2.03 0.11 0.11

11.54 9.56 2.07 2.03 0.1 0.1

11.1 9.41 2.05 2 0.1 0.09

RSD 3.10% 2.60% 1.30% 1.40% 3.30% 6.90%

Unleaded Brass Readings

9.91 8.23 1.87 1.83 0.16 0.15

9.09 7.88 1.85 1.84 0.16 0.17

9.72 8.18 1.85 1.83 0.18 0.18

10.2 8.67 1.88 1.87 0.17 0.16

9.85 8.24 1.91 1.89 0.16 0.17

RSD 4.20% 3.40% 1.40% 1.60% 6.70% 6.00%

The table 20 exhibits the five selected parameter readings taken on the surface of leaded and unleaded brass samples using profilometer.

TABLE 21:CORRELATION MATRIX OF 2D PARAMETERS

Maximum

Amplitude Mean Amplitude Hybrid

% Pt Rz Pa Ra PSm RSm

Pt 100 98

Rz 98 100

Pa 100 100

Ra 100 100

PSm 100% 98

RSm 98 100%

The correlation matrix of 2D parameters are shown in the table and the same

methodology is employed in order to select the most relevant and appropriate

parameters which exhibit less variation. Parameters Rz, Pa and Psm are selected

from the roughness and primary profile for discrimination between the surfaces of

the samples.