Available online at www.sciencedirect.

com

ScienceDirect
Procedia Materials Science 6 (2014) 1303 – 1311

3rd International Conference on Materials Processing and Characterization (ICMPC 2014)

Metal Removal and Kerf Analysis in Abrasive jet drilling of Glass Sheets

D.V. Srikantha, Dr. M. SreenivasaRaob

a
Associate Professor, Department of Mechanical Engineering, Abhinav Hi-Tech College of Engineering, Hyderabad, A.P.India
b
Professor, Dept. of Mechanical Engineering, JNTUH, Kukatpally, Hyderabad, A.P. India.

Abstract

Drilling of glass with different SOD’s, Pressures and different Nozzle Diameters have been carried out by Abrasive Jet Drilling process (AJD)
in order to determine its mach inability under different controlling parameters of the AJM process. Abrasive jet machine (AJM) removes
material through the action of focused beam of abrasive laden gas. Micro abrasive particles are propelled by an inert gas of velocity. When
directed at a work piece, the resulting erosion can be used for cutting, etching, drilling, polishing and cleaning. In this paper optimization of
process parameters of Abrasive Jet Machining of glass by Taguchi methodology is presented. The Values obtained in Taguchi Analysis was
compared with the Analysis of Variance (ANOVA).Various levels of Experiments are conducted using L9 Orthogonal Array for both MRR and
KERF.
©
© 2014
2014 The Authors.
Elsevier Ltd. Published
This is an by Elsevier
open accessLtd.
article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of the Gokaraju Rangaraju Institute of Engineering and Technology (GRIET).
Selection and peer review under responsibility of the Gokaraju Rangaraju Institute of Engineering and Technology (GRIET)
Key Words: Abrasive Jet Machining, Erosion rate, Glass, AJD, SOD, Pressure.L9orthogonal Array.

1. INTRODUCTION

In Abrasive Jet Machining, fine abrasive particles (typically ~0.025mm) are accelerated in a gas stream (commonly air) towards
the work surface. As the particles impact the work surface, they cause small fractures, and the gas stream carries both the
abrasive particles and the fractured (wear) particles away. A high-velocity jet of dry air, nitrogen, or carbon dioxide Containing
abrasive particles is aimed at the work piece surface under controlled conditions. The jet velocity is in the range of 150-300 m/s
and pressure is from two to ten times atmospheric pressure. Abrasive Jet Machining is used for drilling, deburring, etching, and
cleaning of hard and brittle metals, alloys, and non-metallic materials (e.g., germanium, silicon, glass, ceramics, and mica). No
heat is required in the process of machining a piece with an abrasive jet. As a result, parts from an assembly do not experience
structural changes from overheating. There are no toxic wastes given off by abrasive water jets, and no oils are necessary in the
process of machining. Aluminium oxides, silicon carbides, Boron Carbides, Crushed glass, Sodium bicarbonate, Dolomite are
Various Abrasive Particles used for Machining in Abrasive Jet Machining. Re use of abrasives is not recommended since the
cutting ability of abrasive decrease after the usage and also the contamination of wear materials clogging the nozzle and the
cutting unit orifice. The Major Process Parameters that affects the MRR in AJM are 1. Gas Pressure 2. Velocity of Abrasive
Particles 3. Abrasive mass flow rate 4. Mixing ratio 5. Nozzle Tip Distance.

______

D.V.Srikanth Tel: +919703050004

E-mail Address: dvsk75@gmail.com.

2211-8128 © 2014 Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer review under responsibility of the Gokaraju Rangaraju Institute of Engineering and Technology (GRIET)
doi:10.1016/j.mspro.2014.07.109
1304 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311

Fig 1: Line Diagram of Abrasive Jet Machining Process

2. LITERATURE REVIEW

Till date there has been a through and detailed experimental study on the Machining of Glass. El-Domiaty et al (2009)
are conducted the experimentation on drilling of Glass with AJM. The abrasive grits (sand) were mixed with air stream ahead of
the nozzle and the abrasive flow rate was kept constant throughout the machining process. The Results was represented with
Graphs. Bhaskar Chandra Kandpal et al (2011) conducted Experimentation on machining of Glass and Ceramics with various
types of Abrasives by changing pressure, nozzle tip distance on different thickness of glass plates and ceramic plates. The effect
of process parameters of compared with theoretical results.Stephen Wan et al. (2010) present simple deterministic process
models for the prediction of the evolution of the cross-sectional profile of glass channels generated by erosive wear in micro air
abrasive jet machining using a round nozzle. Experiments were carried out on soda lime and borosilicate glass to verify the
process models. Predicted model results show fairly good agreement with experimental results. Lingyin et al (2001) investigated
the abrasive jet machining characteristics of a glass-infiltrated alumina used for fabrication of all-ceramic dental crowns were
investigated using a high-speed dental hand piece and diamond burs with different grit sizes. Apart from the Experimental works
detailed theoretical studies are also performed on Abrasive Jet Machining.Most of the studies argue over the hydro dynamic
characteristics of abrasive jets, hence ascertaining the influence of all operational variables on the process effectiveness including
abrasive type, size and concentration, impact speed and angle of impingement. Other papers found new problems concerning
carrier gas typologies, nozzle shape, sizeand wear, jet velocity and pressure, standoff distance (SOD) or nozzle tip distance
(NTD). These papers express the overall process performance in terms of material removal rate, geometrical tolerances and
surface finishing of work pieces, as well as in terms of nozzle wear rate. Finally, there are several significant and important
papers which focus on either leading process mechanisms in machining of both ductile and brittle materials, or on the
development of systematic experimental statistical approaches, Analysis and artificial neural networks to predict the relationship
between the settings of operational variables and the machining rate and accuracy in surface finishing. In recent years abrasive jet
machining has been gaining increasing acceptability for drilling applications.
3. METHODS:
3.1.Design of Experiments (DOE) Experimental design is a useful complement to multivariate data analysis because itgenerates
“structured” data tables, i.e. data tables that contain an important amount of structured variation. This underlyingStructure will
then be used as a basis for multivariate modelling, which will guarantee stable and robust models.TheDOETechnique helps to
study many factors simultaneously and most economically. By studying the effects of individualFactors on the results, the best
factor combination can be determined.
3.2. Optimization based on TAGUCHI approachis used to achieve more efficient cutting parameters. According to Roy.R.K
(2001), Jurkovic.Z (2009),Jurkovic. Z (2010) and Cukor.G (2011)Parameter design is the key step in the Taguchi approach to
achieve high quality without increasing cost. To solve this problem Taguchi approach uses a special design of orthogonal arrays
where the experimental results are transformed into the S/N ratio as the measure of the quality characteristic deviating from the
desired value.

3.3.Analysis of Variance (ANOVA)

Analysis of variance (ANOVA) and F-test (standard analysis) are used to analysis the experimental data as given follows

Notation:
Following Notation are used for calculation of ANOVA method
C.F. = Correction factor
T = Total of all result
n = Total no. of experiments
ST = Total sum of squares to total variation.
 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311 1305

Xi = Value of results of each experiments (i = 1 to 27)

SY = Sum of the squares of due to parameter Y (Y = P, S, A, T)
NY1, NY2, NY3 = Repeating number of each level (1, 2, 3) of parameter Y
XY1, XY2, XY3 = Values of result of each level (1, 2, 3) of parameter Y
FY= Degree of freedom (D.O.F.) of parameter of Y
fT = Total degree of freedom (D.O.F.)
fe = Degree of freedom (D.O.F.) of error terms
VY = Variance of parameter Y
Se = Sum of square of error terms
Ve = Variance of error terms
FY = F-ratio of parameter of Y
SY’ = Pure sum of square
CY= Percentage of contribution of parameter Y
Ce = Percentage of contribution of error terms
CF = T2/n
ST = ∑i=1 to 27 Xi2 – CF
SY = (XY12/NY1 + XY22/NY2 + XY32/NY3) – CF
fY = ( number of levels of parameter Y) – 1
fT = ( total number of results)-1
fe = fT - ∑fY
VY = SY/fY
Se = ST - ∑SY
Ve = Se/fe
FY = VY/Ve
SY’ = SY – (Ve*fz)
CY = SY’/ST * 100%
Ce = (1- ∑PY)*100%

4. EXPERIMENTAL WORK

4.1. Experimental set up:

Main machine structure of size 250x250x550 has been fabricated. Lifting mechanism parts have been machined & assembled as
shown in Fig. Mixing chamber & hopper have been machined out of M.S sheet and welded. The opening for the abrasive
Particles was arranged with air tight and leak proof system. The Nozzle was made of Tungsten carbide which can withstand for
many operations. Nozzles are designed and fabricated in different sizes( 1 mm,2mm,3mm) .Working chamber is surrounded by
transparent plastic sheets to view the progress of abrasive machining to make the machine environment friendly, exhauster with
dust collecting lags are fixed.

Fig 2 : Fabricated Experimental Setup of Abrasive jet Machining

1306 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311

The experimental work was carried on a test rig. The abrasive grits (Al2o3, Sic) were mixed with air stream ahead of the nozzle
and the abrasive flow rate was kept constant throughout the machining process. The jet nozzle was made of Tungsten Carbide to
carry high wear resistance and increase in Life of nozzle. Several nozzles were manufactured with different bore diameters of 1
mm, 2 mm and 3 mm. Drilling of glass sheets was conducted by setting the test rig on the parameters like Pressure,SOD,and
Nozzle diameter, Abrasive flow rate.

(a)Nozzles with various diameters(b) Impingement angle of Nozzle

Fig 3: Shows the Nozzles of different diameters and nozzle fixed to set up

a) CharacteristicsOfDifferentVariables

Table 1: Variables and characteristics in Abrasive jet machining

Medium Air , CO2 ,N2

Abrasive Sic, Al2O3 (of size 20μto50μ )
Flow rate of abrasive 3 to 20 gram/min
Velocity 150 to 300 m/min
Pressure 2 to 8 kg/cm2
Nozzle size 0.07 to 0.40 mm
Material of nozzle WC, Sapphire
Nozzle life 12 to 300 hr
Standoff distance 0.25 to 15 mm (8mmgenerally)
Workmaterial glass, ceramics, granites.
Metalsandalloysofhardmaterials like germanium,
silicon etc
part application Drilling, cutting, deburring,cleaning

Glass was used as a work piece material because of its homogeneous properties. The test specimens were cut into square and
rectangular shape for machining on AJM unit having thickness 3mm, 4mm. In machine the initial weights of glass specimens
were measured with the help of digital balance. After machining the final weights were measured with the help of digital balance
to calculate the material removal rate. First the abrasive that was Sic in powder form was fed in the hopper carefully. After that
compressor connections were checked. The glass specimen was properly clamped on cross slide with the help of various clamps.
As the Compressor was switched on, the hopper gate valve was opened so that abrasive grains were mixed with air jet coming
from the compressor and focused on the specimen with help of nozzle. Same experiments were with silicon carbide as abrasive
in AJM process. Different readings were taken based on the levels of Taguchi Analysis using different process parameters
(Pressure, Sod, and Nozzle diameter.). All results were analysed by Taguchi method and compared with ANOVA.
 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311 1307

5. EXPERIMENTATION, ANALYSIS & RESULTS

Experiments are conducted based on Taguchi’s method with three factors at three levels each. The values taken by a factor are
termed to be levels. The factors to be studied and their levels chosen are detailed in the Table 2.

Table 2: Factors of Taguchi and Levels

Machining Parameters Level 1 Level 2 Level 3

2
Pressure (kg/cm ) 6 7 8

Stand Of Distance (mm) 8 9 10

Nozzle Diameter (mm) 2 3 4

5.1. Signal-To-Noise Ratio

The term signal represents the desirable value (mean) and noise represents the desirable value (mean) and the “noise” represents
the undesirable value (standard deviation). So the S/N ratio represents the amount of variation presents in the quality
characteristic. Depending upon the objective of the quality characteristic there can be various types of S/N ratio. Here the
desirable objectives are higher values of MRR and lower values of KERF. So the Larger-the-better and Smaller the better type
S/N ratios are mentioned below.

§ n 1 ·
¨¦ ¸ (1)
¨ yi2 ¸
SN L  10 log i 1
¨ n ¸
¨ ¸
© ¹

§ n 2 ·
¨ ¦ yi ¸ (2)
SN L  10 log ¨ i 1 ¸
¨ n ¸
¨ ¸
© ¹

Where y is the observed data

nis the number of observations and
n is no of measurements in a Trail

Table 3: Experimental DesignMatrix and Results

EXPMachining Parameters
No of Abrasive Jet Machining
Pressure Sod N D MRRI MRR II SNRA1MEAN1KERFSNRA2

1 6 8 2 0.0432 0.0402-27.6141 0.04170 3.1 -9.8272

2 6 9 3 0.0441 0.0513 -26.5039 0.047703.5 -10.8814
3 6 10 4 0.0812 0.0800 -21.8740 0.08060 6.2 -15.8478
4 7 9 4 0.0637 0.0612 -24.09460.062454.1 -12.2557
5 7 10 2 0.0371 0.0298 -29.66750.033454.8 -13.6248
6 7 8 3 0.0765 0.0745 -22.4433 0.075503.6-11.1261
7 8 10 3 0.0986 0.1000 -20.06170.099305.3 -14.4855
8 8 8 4 0.0864 0.0964 -20.8201 0.09140 5.8 -15.2686
9 8 9 2 0.0542 0.0513 -25.5654 0.052754.4 -12.8691
1308 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311

Table 4: Response Table for Means(MRR)

Level Pressure Sod Nozzle Diameter
1 0.05667 0.06953 0.04263
2 0.05713 0.05430 0.07417
3 0.08115 0.07112 0.07815
Delta 0.02448 0.01682 0.03552
Rank 2 3 1

Main Effects Plot for MRR

Fitted Means
Pressure Sod
0.08

0.07

0.06

0.05

0.04
Mean

6 7 8 8 9 10
Nozzle Diamet er
0.08

0.07

0.06

0.05

0.04
2 3 4

Fig 4: Graphs indicates the Main Effect plot for MRR (Fitted Means)

Table 5: Response Table for Signal to Noise Ratios (Larger is better)

Level Pressure Sod Nozzle Diameter

1 -25.33 -23.63 -27.62
2 -25.40 -25.39 -23.00
3 -22.15 -23.87 -22.26
Delta 3.25 1.76 5.35
Rank 2 3 1

Main Effects Plot for SN ratios

Data Means

Pressure Sod
-22

-24
Mean of SN ratios

-26

6 7 8 8 9 10
Nozzle Diameter
-22

-24

-26

2 3 4

Signal-to-noise: Larger is better

Fig 5: Graphs indicates the Effect of plots on S/N Ratio Pressure, Sod, Nozzle diameter on MRR

Table 6: Response Table for Means (KERF)

Level Pressure Sod Nozzle Diameter
1 4.267 4.167 4.100
2 4.167 4.000 4.133
3 5.167 5.433 5.367
Delta 1.000 1.433 1.267
Rank 3 1 2
 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311 1309

Main Effects Plot for Means

Data Means
Pressure Sod
4.5

4.0
Mean of Means

3.5

6 7 8 8 9 10
Nozzle Diamet er
4.5

4.0

3.5

2 3 4

Fig 6: Graphs indicates the Main Effect plot for MRR (Data Means)

5.2. Taguchi Analysis: MRR I , MRR II,KERF versus Pressure, Sod, Nozzle DiameterPredicted values.

MRR KERF
S/N Ratio Mean StDev S/N Ratio Mean
-19.6920 0.10045 0.0030406 -10.5327 3.33333

Factor levels for predictions: Factor levels for predictions:

Pressure Sod ND Pressure Sod ND
8 10 4 6 9 3

The effect of Individual Parameter on entire process is not effectively rated by using Taguchi Method and the Predicted values of
Mean S/N Ratio, Standard deviation, Factor levels are identified. By using ANONA the contribution of Individual Parameter can
be well determined. The General Linear Model of ANOVA module was employed to investigate the effect of Parameters on
MRR.

5.3. General Linear Model:MRR, KERFversus Pressure, Sod, Nozzle Diameter

Factor Type Levels Values

Pressure fixed 3 6, 7, 8
Sod fixed 3 8, 9, 10
Nozzle Diameter fixed 3 2, 3, 4

Table 7: Analysis of Variance for MRR using Adjusted SS for Tests

Source DF Seq ss Adj ss Adj MS F P

Pressure 2 0.0013999 0.0013999 0.0007000 9.05 0.099

SOD 2 0.0004839 0.0004839 0.0002419 3.13 0.242

Nozzle 2 0.0027317 0.0027317 0.0013658 17.67 0.054

Diameter
Error 2 0.0001546 0.0001546 0.0000773

Total 8 0.0047701

S = 0.00879223 R-Sq = 96.76% R-Sq (adj) = 87.04%

1310 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311

Table 8 : Analysis of Variance for KERF using Adjusted SS for Tests

Source DF Seq ss Adj ss Adj MS F P

Pressure 2 1.8200 1.8200 0.9100 3.00 0.250

SOD 2 3.6867 3.6867 1.8433 6.08 0.141

Nozzle 2 3.1267 3.1267 1.5633 5.15 0.162

Diameter
Error 2 0.6067 0.6067 0.3033

Total 8 9.2400

S = 0.550757 R-Sq = 93.43% R-Sq(adj) = 73.74%

Table 7,8 shows Analysis of Variance for MRR and KERF. In Table (7)F Value (17.67) of the parameter indicates the Nozzle
Diameter is significantly contributing more towards cutting performance. F value (3.13) of parameter indicates the Contribution
of Stand-off distance is less. In Table (8) F Value (6.08) of the parameter indicates the SOD is significantly contributing more
towards KERF performance. F value (3.00) of parameter indicates the Contribution of Pressure is less. The R-Square values of
both MRR and KERF are nearer to 100 % .This shows the results obtained are optimal.

Fig7: Different glass Sheets drilled atdifferent Nozzlediameters, SOD’s and Pressures

CONCLUSIONS
This paper presents Ample results of experiments have been conducted by changing pressure, nozzle tip distance, SOD
on different thickness of glass plates. The effect of their process parameters on the material removal rate (MRR) is analysed by
using Taguchi Method and compared this by using Analysis of variance ANOVA. Optimal levels of Performance Found at
Larger is Better MRR was identified as Air Pressure (8 kg/cm2) SOD (10 mm) Nozzle diameter (4 mm).As per the Table (4)the
ranking of parameters are Nozzle diameter, Pressure and SOD. Where it was also observed from ANOVA F-test that the same
parameters of Taguchi are repeated in same order. Optimal levels of Performance Found at Smaller is Better KERF was
identified as Air Pressure (6 kg/cm2) SOD (9 mm) Nozzle diameter (3 mm).As per the Table (5)the ranking of parameters are
Nozzle diameter, Pressure and SOD. Where it was also observed from ANOVA F-test that the same parameters of Taguchi are
repeated in same order. Therefore the results obtained by TAGUCHI are nearly matching with ANOVA results.

ACKNOWLEDGMENT

The authors here by thank the authors of the below mentioned references for their valuable contribution which enabled us
make this comparison.
 D.V. Srikanth and M. Sreenivasa Rao / Procedia Materials Science 6 (2014) 1303 – 1311 1311

References
A.EI-Domiaty, H.M.Abd EI –Hafez, and M.A.Shaker (2009) “Drilling of glass sheets by abrasive jet machining”, World Academy of
Science, Engineering and Technology 56.
A. P. Verma and G. K. Lal Publication (1984) “An experimental study of abrasive jet machining”, International Journal of Machine Tool
Design and Research, Volume 24, Issue 1, pp 19-29.
Bhaskar Chandra Kandpal1* Naveen Kumar2 Rahul Kumar3 Rahul Sharma4 Sagar Deswal5 Machining Of Glass and Ceramic with Alumina
And Silicon Carbide in Abrasive Jet Machining, International Journal of Advanced Engineering Technology, IJAET/Vol.II/ Issue
IV/October-December, 2011/251-256.
Cukor, G., Jurković, Z., Sekulić, M.: Rotatable Central Composite Design of Experiments versus TaguchiMethodin the Optimization of
Turning,Metalurgija/Metallurgy, 50(2011)1, 17-20.
Jurković, Z., Brezočnik, M., Grizelj, B., Mandic, V.: Optimization of Extrusion Process by Genetic Algorithms and Conventional
Techniques, Tehnički vjesnik/Technical Gazette 16(2009)4, 27-33.
Jurković, Z., Cukor, G., Andrejcak, I.: Improving the Surface Roughness at Longitudinal Turning Using the Different Optimization
Methods, Tehnički vjesnik/Technical Gazette 17(2010)4, 397-402.
Ling Yina, Lewis K. Ivesa, Said Jahanmira, E. Dianne Rekowb & Elaine Rombergc Abrasive Machining Of Glass- Infiltrated Alumina With
Diamond Burs Machining, Science and Technology: An International Journal Volume 5, Issue 1, 2001 pages 43-61.
Marshall DB, Lawn BR, Evans AG., "Elastic/plastic indentation damage in ceramics: the lateral crack system", J Am Ceramic Society,
1982; 65(11).
M L Neema and P C Pandey. 'Erosion of Glass When Act Upon by an Abrasive Jet' Proceedings of International Conference on Wear of
Materials, 1977.
M. Roopa Rani & S. Seshan “AJM - Process Variables and Current Applications”, Publication- Journal of Metals Materials Process, 1995
Vol.7 No. 4 PP.279-290.
Roy, R.K.: Design of experiments using the Taguchi approach: 16 steps to product and process Improvement, Wiley-Interscience, New
York, 2001.
Ruff AW, Wiederhorn SM., "Erosion by solid particle impact"; Treat Mater Sci. Tech, 1979.
R. Balasubramaniam, J. Krishnan and N.Ramakrishnan (1999), “An experimental study on the abrasive jet deburring of cross drilled holes”,
Publication: Journal ofMaterials Processing Technology, Volume 91, Issues 1-3, and pp 178-182.
R. Balasubramaniam, J. Krishnan and N.Ramakrishnan (2002) “A study on the shape of the surface generated by abrasive jet machining”,
Publication: Journal of Materials Processing Technology, Volume 121, Issue 1, 14February 2002, pp 102-106.
R.H.M. Jafar*1, M. Papini2, 1, J.K. Spelt1, 2, Numerical simulation of surface roughness and erosion rate in abrasive jet micromachining of
Glass. 2013 – Elsevier.
Sheldon, G.L. and Finnie, I., Trans. ASME B (J. Eng. for Industry), 1966, 88 B, 387-392.
Stephen Wan, Weng Seng Fong, Qi Ying Leong-- Deterministic Process Modelling of Micro Air Abrasive Jet Machining of Glass Channel-
2010, Key Engineering Materials, 447-448, 272.
U.D.Gulhani, P.P.Patkar,S.P.Patel,A.A.Patel et al ,Analysis of AJM Parameters on MRR, Kerf width of Hard and Brittle Materials like
Ceramics,--IJDMT-April-2013.