Cutting Performance of Glass-Vinyl Ester Composite by Abrasive Water Jet

Mustafa Armaan1,2, A. Armaan Arici2

Address correspondence to Mustafa Armaan, Directorate of Construction & Technical

Received 18 Mar 2016 Accepted 11 Nov 2016

Abstract

Polymer matrix composite materials have been increasingly used in aerospace, defense,

automotive and marine industries. In these fields, non-traditional machining method of

abrasive water jet (AWJ) has been used significantly in order to form polymer matrix

components. In this study, glass fiber reinforced vinyl ester composite plates have been

investigated under various AWJ cutting parameters by using Taguchi experimental

design in detail. For Taguchi experimental design, experimental parameters of standoff

distance, abrasive mass flow rate, traverse speed, pressure, and the material thickness

were determined in three levels. Top kerf width and the surface roughness were

investigated in order to understand cutting performance. Finally, linear regression models

were conducted and all performance parameters were examined by using ANOVA

(Analysis of Variance) and main effects plots. According to overall test results, standoff

distance was determined as the most effective one. The optimal parameter levels were

obtained by the main effects plots, and finally, the predictive modeling were validated

by performing the optimal combination of parameter levels.

1
KEYWORDS: Abrasive; Waterjet; Cutting; Glass-Vinyl Ester; Composite; Roughness;

Kerf; Width; Experimental; Parameters; Taguchi; Design

1. INTRODUCTION

Abrasive water jet machining (AWJM) method is based on sending abrasive materials

accelerated by pressurized water onto surfaces of target materials in order to process and

cut engineering materials. Advantages of AWJ machining method compared to other

machining methods are; (i) heat-affected zone does not occur [1], thus the melting and

solidification are not seen in the cutting surface, (ii) cutting forces do not occur on the

cutting tools, thus assembly fixtures are not necessary for the connection of target

material [2], (iii) complex shapes can be easily formed with the aid of pressurized water

by adding abrasive material [3], (iv) various materials can be processed with the aid of

abrasive particles [4,5], (v) the overall process is completely environmentally friendly [6].

The cutting surface has a two-step mechanism, which includes cutting wear and

deformation wear zones. In the first stage indicated as cutting wear zone, the target

material is removed primarily by micro cutting action of particles which impacts with

shallow angles [7,8]. In the second stage indicated as deformation wear zone, the material

is removed by plastic deformation because of large impact angles [2,7].

It is possible to divide AWJ process parameters into four sections as (i) hydraulic, (ii)

mixing and acceleration, (iii) cutting and (iv) abrasives [9]. Most of the research studies

were focused on determining the effects of the parameters [1013]. The indicators are

generally kerf formation and surface roughness [1418].

2
The reinforcement elements in the structure of polymer matrix composites are found in

different forms [1921]. This caused to a negative effect and the machining of composite

materials becomes difficult [22,23]. Therefore, the use of non-traditional machining

methods of polymer composites are increased rapidly rather than the traditional

machining (turning, milling, etc.) methods [2,24]. AWJ machining method is generally

used in mining [13,2529] and metal [2933] industries. However, few numbers of scientific

studies are not enough for the polymer matrix composite materials. Also these materials

have been increasingly used in important and strategic industries, but generally traditional

machining methods are used in manufacturing process for these materials.

Aforementioned advantages of AWJ machining method, and developments of cutting

process were achieved.

Shanmugam and Masood [34] investigated AWJ cutting process of graphite/epoxy and

glass/epoxy composites in accordance with the Taguchi experimental design. According

to the study, kerf taper angle was decreased with increasing pressure. Also, it was

increased with increasing traverse speed and standoff distance. Wang [35] examined

processing parameters of the AWJ cutting process of polymer matrix reinforced with

Teflon fabric. According to this study, top kerf width was increased with standoff

distance and water pressure but it was decreased with traverse speed. Furthermore,

roughness was increased with water pressure and traverse speed. Ma and Deam [36]

studied the cutting of acrylic polymeric material and their results were correlated with the

predicted kerf (slot) profile. They concluded that the conicity of the kerf (formed by

3
cutting) was increased with according to the desired engineering tolerances in machined

to tapering. Thus, kerf profiles which were improved with different traverse rates, have

been found by regression analysis.

Azmir and Ahsan [37] investigated machining surfaces of epoxy matrix reinforced with

glass fibre composites by AWJ. The study was conducted in L18 Taguchi experimental

design with analysis of variance (ANOVA). The type of abrasive particles was the most

significant control factor for this study. In addition, abrasive mass flow rate, cutting

orientation and standoff distance parameters were insignificant control factors. In other

study, previous researchers [38] examined cutting process of glass/epoxy composite

cutting by AWJ. Type of abrasive material and pressure were the most important

parameters for kerf taper ratio and surface roughness, respectively. It is also told that

cutting orientation was not influenced for kerf taper ratio and surface roughness. Alberdi

et al. [39] used two different carbon fiber reinforced composite materials with two

different thicknesses for AWJ cutting process. According to the study, the percentage of

traverse feed rate relative to the separation speed and thickness were significant

parameters for surface roughness and taper angle, respectively. Sreekesh and Govindan
[10]
conducted review study for AWJ cutting processing and thereby comparative

parameter analysis table was created.

In this study, cutting performance of glass fiber reinforced vinyl ester composite plates

was investigated by AWJ. The effect of performance parameters on smooth cutting zone

surface roughness (initial and zone average surface roughness) and top kerf width was

4
determined. The optimal experimental conditions were determined by ANOVA analysis

and confirmation tests were performed.

2. MATERIALS AND METHODS

In this study, Metyx Composites trademark named E-glass fiber was used as the

reinforcement materials. Glass fiber orientation has been selected quadraxial (0 /-

45 /90 /+45 ) stitched knit type, tricot stitch type and the weight was 1200 g/m2.

Vinyl ester resin trademark named as Reichhold was used for matrix material. Hand lay-

up process was used to prepare glass fiber/vinyl ester laminates. Quadraxial stitched knit

fibers were cut into rectangular form of 600 mm400 mm. Composite plates with three

different thicknesses were prepared to examine the effect of the thickness. According to

the thickness of each plate, quadraxial stitched knit glass fibers for 9, 14, 19 pieces of

plies were stacked, and the final thicknesses of plates were measured as 18, 28, 38 mm.

The final glass fiber mass ratio of the plate was calculated as 0.63. AWJ cuttings were

performed on a Robjet RJP 50 HP model high pressure abrasive waterjet machine (Figure

1). The nozzle impact angle and the nozzle diameter were 90 and 1.02 mm,

respectively. Garnet was the abrasive material commercially named Supreme Garnet,

Industrial Minerals Co. and garnet mesh size was 80.

In this study, experimental design was planned to be five factors and three levels (3^5)

(Table 1). Taguchi orthogonal array design according to the experimental design was

performed. Based on the Taguchi's method, DoE L27 (3^5) was used, 27 runs (Table 2)

were carried out. The measured quality indicators were top kerf width and smooth cutting

5
zone (cutting wear zone) surface roughness (initial and zone average surface roughness).

Top kerf width measurements (Figure 2a) were carried out using Nikon MA100 optical

microscope device. Surface roughness (Ra) measurements (Figure 2b) were carried out

using Nanovea PS503D optical profilometer device. Surface roughness measurements

were performed at 3 mm intervals along the zone section. All measurements were

repeated at least three times and their averages were found. General linear models

(regressions) were conducted and ANOVA analysis were performed by Minitab 17

statistical software.

3. RESULTS AND DISCUSSION

Efficacy of the experimental parameters on AWJ cutting performance [top kerf width

(Wt) with initial (Rai) and zone average surface roughness (Raz) in the smooth cutting

zone] was performed by ANOVA analysis. 95% confidence level for all interval, two

sided type of confidence interval and type III for sum of squares tests were selected.

Analysis of variance regarding to top kerf width was shown in Table 3.

According to Table 3, standoff distance was found as the most effective experimental

parameter (73.35 %), and it was shown very high statistically significant (P-

Value<0.001). Traverse speed showed a high level of statistically significant (0.001<P-

Value<0.01) and it was found as the second highest experimental parameter (10.55%).

Material thickness was found as third highest experimental parameter (4.42%) and it was

shown statistically significant (0.01<P-Value<0.05). The pressure was indicated

(significant at the border) meaningfulness trend (0.05<P-Value<0.10) and thereby it was

6
considered insignificant (2.80%). Abrasive mass flow rate was also considered

insignificant (P-Value>0.10).

Analysis of variance regarding to initial surface roughness was shown in Table 4.

Standoff distance was shown very high statistically significant (P-Value<0.001).Thus, the

most effective experimental parameter was found (35.11%). Pressure and abrasive mass

flow rate showed a high level of statistically significant (0.001<P-Value<0.01) and they

were found almost equally significant effect (18.12% and 22.83%, respectively). Traverse

speed and material thickness were considered insignificant (P-Value>0.10).

Analysis of variance regarding to zone average surface roughness was shown in Table 5.

Standoff distance showed a high level of statistically significant (0.001<P-Value<0.01)

and the most effective experimental parameter was found (26.71%). Pressure, abrasive

mass flow rate and traverse speed showed statistical significant (0.01<P-Value<0.05) and

they were found almost equally significant effect (16.53%, 13.18% and 12.22%,

respectively). Material thickness was considered insignificant (P-Value>0.10).

The influence of each factors top kerf width, initial and zone average surface roughnesses

were illustrated in Figure 3. It was seen clearly that the parameter of the standoff distance

was the most effective one for all quality indicators.

ANOVA analysis for top kerf width was conducted and the experimental parameter

levels were shown in Figure 4. Standoff distance which was the most effective

7
experimental parameter increased from 5 mm to 10 mm, and top kerf width increased

significantly. Wang [35], Wang and Wong [40] explained this behavior with the divergence

of the waterjet when spreading out (at different angles) from the mixing tube. The

divergence of the water jet enlarges with increasing standoff distance which effects top

kerf width with same way. Ma and Deam [36] also reported that standoff distance and

cutting speed were the two cutting parameters changed the kerf profile. So, our result

about the effect of standoff distance on kerf width was compatible with these studies and

also compatible with Sreekesh and Govindan [10].

Traverse speed was found as the second highest experimental parameter and it was highly

effective. Top kerf width is inversely effected by traverse speed and this increased from

2.5 mm/s to 8 mm/s, therefore top kerf width was decreased. It is contemplated that the

number of particles hitting the material are decreased with increasing traverse speed. So

the decreasing cutting efficiency caused to different kerf width [15,35,40]. Also Gudimetla et

al. [41] reported this decreasing situation with particle fragmentation effect which is the

cutting efficiency of the jet and energy absorption by the material as the jet penetrates the

target material. Consequently, higher traverse speed generates a narrower slot [42].

Material thickness was the third highest effective experimental parameter. Material

thickness increased from 18 mm to 28 mm thus top kerf width increased. When material

thickness increased, resistance force against the jet stream was increased slightly

therefore top kerf width increased. After material thickness decreased from 28 mm to 38

mm thus top kerf width decreased. When material thickness more increased, dependent

8
on the other process parameters (above a threshold value) they were inversely effect top

kerf width.

Abrasive mass flow rate and pressure parameters were considered insignificant and

ineffective. Both of two increased from level 1 to level 2 but top kerf widths were in

different characters. Top kerf width was decreased slightly for pressure parameter while

it was increased slightly for the parameter of abrasive mass flow rate. Wang and Wong
[40]
and Gupta et al. [15] explained in their study, when pressure over a threshold value

connected with the other machining parameters, AWJ became less effective.

ANOVA analysis for initial surface roughness was conducted and the experimental

parameter levels were shown in Figure 5. Standoff distance was observed again that it

was the most effective parameter from 5 mm to 7 mm. Initial surface roughness increased

significantly at this stage. Herein the increase behaviour for surface roughness was

caused by jet divergence [43]. With the jet divergence effect, abrasive particles are spread

to the ends portions of the jet. Thus particle numbers are decreased by unit time per unit

area incoming on the target material. Also, with multiplication effect of water jet and

abrasive particles are shown in scattering until the critical deployment distance of the jet.

Scattering is increased and the effects of particles and jet are decreased, so surface

roughness is increased. But initial surface roughness decreased significantly with standoff

distance from 7 mm to 10 mm. When standoff distance increased more, particle numbers

in unit time per unit area are decreased. For the critical deployment distance values are

9
exceeded, water jet and multiplication scattering effects are decreased. Surface roughness

is improved by this situation.

Abrasive mass flow rate and pressure were found as the secondary highest experimental

parameters and they were highly effective. Initial surface roughness decreased

significantly with abrasive mass flow rate from 200 g/min to 250 g/min and it increased

with this parameter from 250 g/min to 350 g/min. Firstly, the surface roughness was

decreased significantly with increasing abrasive mass flow rate because the numbers of

cutting particles available per unit area were increased. But at the second stage inter

collision of particles among themselves were reduced the surface roughness [44]. Initial

surface roughness increased significantly with pressure from 260 MPa to 320 MPa.

Similar situation has been observed by Jani et al. [45]. Wang and Wong [40] explained this

increasing trend as the outer rim of the diverged jet will gain enough energy for cutting

the material which tends to increase the irregularity and roughness of the surface. When

the pressure increases, the kinetic energy of each particle is increased thereby material

removal ability is improved. Also, interparticle collisions may be seen randomly by the

acceleration. The water jet with the effects of abrasives spends more energy on target

material. Consequently, rougher surface is obtained [44].

Traverse speed and material thickness parameters were considered insignificant and

ineffective. Initial surface roughness increased with traverse speed from 2.5 mm/s to 4

mm/s and it decreased with this parameter from 4 mm/s to 8 mm/s. This trend was

observed in similar studies [29,43,45,46]. Initial surface roughness decreased with material

10
thickness from 18 mm to 28 mm and it increased with this parameter from 28 mm to 38

mm. This trend was observed in similar study [47].

ANOVA analysis for zone average surface roughness was conducted and experimental

parameter levels were shown in Figure 6. As seen here, zone average surface roughness

behaviour was almost identical with initial surface roughness behaviour (Figure 5).

Percent contribution of zone average surface roughness for standoff distance was

decreased. Because depth of cut along in material effect of decrease was reduced by jet

divergence and multiplication scattering effects. Pressure has been more effective than

abrasive mass flow rate. Because kinetic energy and sharpness of cutting particles were

decreased along the cutting depth. Also, traverse speed was in action for surface

roughness because of number of particles unit time per unit area with increasing traverse

speed. The studies for initial surface roughness of different researchers were assessed and

the same zone average surface roughness was concluded.

The confirmation tests were performed with optimal combinations of machining factors

as seen in Table 6. The predictive top kerf width (Wt), initial (Rai) and zone average (Raz)

surface roughnesses were determined using the following regression equations:

11
Wt 1601.7 (51.9 P _ 260 ) (20.1 P _ 290 ) (31.8 P _ 320 ) (31.1 A _ 200 )
(5.9 A _ 250 ) (25.1 A _ 350 ) (96.1 T _ 2.5 ) (19.7 T _ 4.0 ) (76.4 T _ 8.0 )
(214.1S_ 5 ) (32.2 S_ 7 ) (246.3 S_10 ) (44.7 M _18 ) (64.1 M _ 28 ) (19.4 M _ 38 )
R ai 10.87 (7.02 P _ 260 ) (2.47 P _ 290 ) (4.55 P _ 320 ) (7.02 A _ 200 )
(6.83 A _ 250 ) (0.19 A _ 350 ) (3.02 T _ 2.5 ) (3.56 T _ 4.0 ) (0.53 T _ 8.0 ) (1)
(6.32 S_ 5 ) (9.78 S_ 7 ) (3.46 S_10 ) (2.71 M _18 ) (2.11 M _ 28 ) (0.61 M _ 38 )
R az 9.37 (5.59 P _ 260 ) (2.27 P _ 290 ) (3.32 P _ 320 ) (4.80 A _ 200 )
(3.67 A _ 250 ) (1.14A _ 350 ) (3.53T _ 2.5 ) (4.63T _ 4.0 ) (1.10T _ 8.0 )
(3.51S_ 5 ) (7.15S_ 7 ) (3.63S_10 ) (2.94M _18 ) (3.21M _ 28 ) (0.26 M _ 38 )

Where P_260, P_290, P_320 are all constant for levels of pressure, A_200, A_250,

A_350 are all constant for levels of abrasive mass flow rate, T_2.5, T_4.0, T_8.0 are all

constant for traverse speed, S_5, S_7, S_10 are all constant for standoff distance, M_18,

M_28, M_38 are all constant for material thickness. Herein for all equations, values of

constants according to level of parameters are taken into account. By means of these

parameters, predictive confirmation values (prediction interval for regression) were

calculated by Minitab 17. Confirmation test values for response parameters of top kerf

width, initial surface roughness and zone average surface roughness were shown in Table

7. The zone behaviours for surface roughnesses were observed in similar study [4850]. It is

seen that the very few differences occurred between experimental and predictive results.

Consequently, the confirmation process was authenticated.

CONCLUSIONS

In this study, abrasive water jet cutting performance on glass-vinyl ester composite was

investigated and AWJ cutting process was performed successfully. Standoff distance was

emerged as the most effective experimental parameter for cutting performance. It was

seen that only one parameter the top kerf width affects highly and it was standoff distance

12
(73.35%). Divergence effect of the water jet was the mechanism for this higher

percentage. Besides, the top kerf width was slightly affected by traverse speed. The

decreasing number of the cutting particles hitting the material and particle fragmentation

effect were the reasons.

The effects of parameters for surface roughnesses of both initial and zone average were

nearly the same. Standoff distance was the most effective parameter, but not only the one

parameter. Pressure and abrasive mass flow rate were also effective parameters. Also the

other effective parameter was traverse speed for zone average surface roughness. The

variability of kinetic energy, amounted sharpness of the cutting particles along the cutting

depth were the reasons for both surface roughnesses. Taguchi experimental design,

ANOVA analysis and confirmation tests were performed successfully and thereby the

applicability of the process was ensured.

ACKNOWLEDGMENT

The authors would like to thank, Kasm BAYNAL at Kocaeli University and KOU-BAP

for financial support of this work (Project Number: 2014/078).

REFERENCES

[1] Lebar, A.; Junkar M.; Poredos, A.; Cvjeticanin M. Method for online quality

monitoring of AWJ cutting by infrared thermography. CIRP Journal of Manufacturing

Science and Technology 2010, 2, 170-175.

13
[2] Sheikh-Ahmad, J.Y. Machining of Polymer Composites; 1st ed.; Springer US: New

York, USA, 2009; 237-251 pp.

[3] Krajcarz, D. Comparison metal water jet cutting with laser and plasma cutting.

Procedia Engineering 2014, 69, 838-843.

[4] Ferenc, K. Cutting with water jet. Welding International 2007, 21, 730-735.

[5] Natarajan, Y.; Murugasen, P.K. Investigation of process parameters influence in AWJ

cutting of D2 steel. Materials and Manufacturing Processes 2016, Published online,

DOI: 10.1080/10426914.2016.1176183

[6] Dong, Y.; Liu, W.; Zhang, H.; Zhang, H. On-line recycling of abrasives in abrasive

water jet cleaning. Procedia CIRP 2014, 15, 278-282.

[7] Hashish, M. A. modeling study of metal cutting with abrasive waterjets. Journal of

Engineering Materials and Technology 1984, 106, 88-100.

[8] Arola, D.; Ramulu, M. Material removal in abrasive waterjet machining of metals

surface integrity and texture. Wear 1997, 210, 50-58.

[9] Kechagias, J.; Petropoulos, G.; Vaxevanidis, N. Application of Taguchi design for

quality characterization of abrasive water jet machining of TRIP sheet steels. The

International Journal of Advanced Manufacturing Technology 2012, 62, 635-643.

[10] Sreekesh, K.; Dr. Govindan, P. A review on abrasive water jet cutting. International

Journal of Recant advances in Mechanical Engineering 2014, 3, 153-158.

[11] Folkes, J. WaterjetAn innovative tool for manufacturing. Journal of Materials

Processing Technology 2009, 209, 6181-6189.

14
[12] Aydin, G.; Karakurt, I.; Hamzacebi, C. Artificial neural network and regression

models for performance prediction of abrasive waterjet in rock cutting. The International

Journal of Advanced Manufacturing Technology 2014, 75, 1321-1330.

[13] Karakurt, I.; Aydn, G.; Aydner, K. An experimental study on the depth of cut of

granite in abrasive waterjet cutting. Materials and Manufacturing Processes 2012, 27,

538-544.

[14] Begic-Hajdarevic, D.; Cekic, A.; Mehmedovic, M.; Djelmic, A. Experimental study

on surface roughness in abrasive water jet cutting. Procedia Engineering 2015, 100, 394-

399.

[15] Gupta, V.; Pandey, P.M.; Garg, P.L.; Khanna, R.; Batra, N.K. Minimization of kerf

taper angle and kerf width using Taguchis method in abrasive water jet machining of

marble. Procedia Materials Science 2014, 6, 140-149.

[16] Yuvaraj, N.; Kumar, M.P. Multiresponse optimization of abrasive water jet cutting

process parameters using TOPSIS approach. Materials and Manufacturing Processes

2015, 30, 882-889.

[17] Yuvaraj, N.; Kumar, M.P. Surface integrity studies on abrasive water jet cutting of

AISI D2 steel. Materials and Manufacturing Processes 2016, Published online, DOI:

10.1080/10426914.2016.1221093

[18] Selvam, R.; Karunamoorthy, L.; Arunkumar, N. Investigation on performance of

abrasive water jet in machining hybrid composites. Materials and Manufacturing

Processes 2016, Published online, DOI: 10.1080/10426914.2016.1198039

15
[19] Karger-Kocsis, J.; Mahmood, H.; Pegoretti, A. Recent advances in fiber/matrix

interphase engineering for polymer composites. Progress in Materials Science 2015, 73,

1-43.

[20] Tan, C.L.; Azmi, A. I.; Muhammad, N. Delamination and surface roughness

analyses in drilling hybrid carbon/glass composite. Materials and Manufacturing

Processes 2016, 31, 1366-1376.

[21] Chaudhary, V.; Gohil, P.P. Investigations on drilling of bidirectional cotton

polyester composite. Materials and Manufacturing Processes 2016, 31, 960-968.

[22] Dandekar, C. R.; Shin, Y. C. Modeling of machining of composite materials: A

review. International Journal of Machine Tools & Manufacture 2012, 57, 102-121.

[23] Debnath, K.; Sisodia, M.; Kumar, A.; Singh, I. Damage-Free hole making in fiber-

reinforced composites: an innovative tool design approach. Materials and Manufacturing

Processes 2016, 31, 1400-1408.

[24] Sheikh-Ahmad, J.Y. Hole quality and damage in drilling carbon/epoxy composites

by electrical discharge machining. Materials and Manufacturing Processes 2016, 31,

941-950.

[25] Chakravarthy, P. S.; Babu, N. R. A new approach for selection of optimal process

parameters in abrasive water jet cutting. Materials and Manufacturing Processes 1999,

14 (4), 581-600.

[26] Babu, M. N.; Muthukrishnan N. Investigation on surface roughness in abrasive

water-jet machining by the response surface method. Materials and Manufacturing

Processes 2014, 29, 1422-1428.

16
[27] Babu, M.N.; Muthukrishnan, N. Investigation of multiple process parameters in

abrasive water jet machining of tiles. Journal of the Chinese Institute of Engineers 2015,

38 (6), 692-700.

[28] Hu, D.; Tang, C.L.; Kang, Y.; Li, X. An investigation on cutting quality by adding

polymer in abrasive water jet. Particulate Science and Technology 2016, 00, 1-7.

[29] Boud, F.; Murray, J.W.; Loo, L.F.; Clare, A.T.; Kinnell, P.K. Soluble abrasives for

waterjet machining. Materials and Manufacturing Processes 2014, 29, 1346-1352.

[30] Uthayakumar, M.; Khan, M.A.; Kumaran, S.T.; Slota, A.; Zajac, J. Machinability of

nickel based superalloy by abrasive water jet machining. Materials and Manufacturing

Processes 2015, Published online, DOI: 10.1080/10426914.2015.1103859

[31] Hlav, L. M.; Hlavov, I.M.; Geryk, V.; Planr, . Investigation of the taper of

kerfs cut in steels by AWJ. The International Journal of Advanced Manufacturing

Technology 2015, 77, 1811-1818.

[32] Zhang, S.; Wu, Y.; Wang, S. An exploration of an abrasive water jet cutting front

profile. The International Journal of Advanced Manufacturing Technology 2015, 80,

1685-1688.

[33] Li, H.; Wang, J. An experimental study of abrasive waterjet machining of Ti-6Al-

4V. The International Journal of Advanced Manufacturing Technology 2015, 81, 361-

369.

[34] Shanmugam, D.K.; Masood, S.H. An investigation on kerf characteristics in abrasive

waterjet cutting of layered composites. Journal of Materials Processing Technology

2009, 209, 3887-3893.

17
[35] Wang, J. A machinability study of polymer matrix composites using abrasive

waterjet cutting technology. Journal of Materials Processing Technology 1999, 94, 30-

35.

[36] Ma, C.; Deam, R.T. A correlation for predicting the kerf profile from abrasive water

jet cutting. Experimental Thermal and Fluid Science 2006, 30, 337343.

[37] Azmir, M.A.; Ahsan, A.K. Investigation on glass/epoxy composite surfaces

machined by abrasive water jet machining. Journal of Materials Processing Technology

2008, 198, 122-128.

[38] Azmir, M.A.; Ahsan, A.K. A study of abrasive water jet machining process on

glass/epoxy composite laminate. Journal of Materials Processing Technology 2009, 209,

61686173.

[39] Alberdi, A.; Suarez, A.; Artaza, A.; Escobar-Palafox, G.A.; Ridgway, K. Composite

cutting with abrasive water jet. Procedia Engineering 2013, 63, 421-429.

[40] Wang, J.; Wong, W.C.K. A study of abrasive waterjet cutting of metallic coated

sheet steels. International Journal of Machine Tools & Manufacture 1999, 39, 855-870.

[41] Gudimetla, P.; Wang, J.; Wong, W. Kerf formation analysis in the abrasive waterjet

cutting of industrial ceramics. Journal of Materials Processing Technology 2002, 128,

123-129.

[42] Ay, M.; ayda, U.; Hasalk, A. Effect of traverse speed on abrasive waterjet

machining of age hardened Inconel 718 nickel- based superalloy. Materials and

Manufacturing Processes 2010, 25, 1160-1165.

18
[43] Santhanakumar, M.; Adalarasan, R.; Rajmohan, M. Experimental modelling and

analysis in abrasive waterjet cutting of ceramic tiles using grey-based response surface

methodology. Arabian Journal for Science and Engineering 2015, 40, 3299-3311.

[44] Aydn, G.; Karakurt, I.; Aydner, K. An investigation on surface roughness of

granite machined by abrasive waterjet. Bulletin of Materials Science 2011, 34 (4), 985-

992.

[45] Jani, S.P.; Kumar, A.S.; Khan, M.A.; Kumar, M.U. Machinablity of hybrid natural

fiber composite with and without filler as reinforcement. Materials and Manufacturing

Processes 2016, 31, 1393-1399.

[46] Lima, C.E.A.; Lebron, R.; Souza, A.J.; Ferreira, N.F.; Neis, P.D. Study of influence

of traverse speed and abrasive mass flowrate in abrasive water jet machining of

gemstones. The International Journal of Advanced Manufacturing Technology 2016, 83,

77-87.

[47] Cojbasic, Z.; Petkovic, D.; Shamshirband, S.; Tong, C.W.; Ch, S.; Jankovic, P.;

Ducic, N.; Baralic, J. Surface roughness prediction by extreme learning machine

constructed with abrasive water jet. Precision Engineering 2016, 43, 86-92.

[48] Hreha, P.; Radvansk, A.; Crach, J.; Lehock, D.; Monkov, K.; Krolczyk, G.;

Ruggiero, A.; Samardzi, I.; Kozak, D.; Hloch, S. Monitoring of focusing tube wear

during abrasive waterjet (AWJ) cutting of AISI 309. Metalurgija 2014, 53 (4), 533-536.

[49] Hloch, S.; Valek, J. Topographical anomaly on surfaces created by abrasive

waterjet. The International Journal of Advanced Manufacturing Technology 2012, 59,

593-604.

19
[50] Kinik, D.; Gnovsk, B.; Hloch, S.; Monka, P.; Monkov, K.; Hutyrov, Z. On-line

monitoring of technological process of material abrasive water jet cutting. Technical

Gazette 2015, 22 (2), 351-357.

20
Table 1 Experimental variable factors and their levels.

Experimental Parameters Design

Parameters Level 1 Level 2 Level 3 Units

Pressure 260 290 320 MPa

Abrasive mass flow rate 200 250 350 (g/min)

Traverse speed 8.0 2.5 4.0 (mm/s)

Standoff distance 5 7 10 mm

Material thickness 18 28 38 mm

21
Table 2 L27 Taguchi orthogonal array design and experimental results.

Experi Press Abrasi Traverse Standoff Material Top Intial Zone

ment ure ve speed distance thickness Kerf Surface Average

No. mass Width Roughn Surface

flow (Wt) ess Roughnes

rate (Rai) s (Raz)

Mpa (g/min) (mm/s) mm mm m m m

1 260 200 8.0 5 18 1400.5 4.60 4.30

2 260 200 8.0 5 28 1378.0 4.15 3.70

3 260 200 8.0 5 38 1217.5 3.28 3.89

4 260 250 2.5 7 18 1602.5 3.43 3.99

5 260 250 2.5 7 28 1933.5 3.77 3.55

6 260 250 2.5 7 38 1634.5 4.11 3.66

7 260 350 4.0 10 18 1849.0 3.72 3.92

8 260 350 4.0 10 28 1983.5 4.19 3.75

9 260 350 4.0 10 38 1883.5 3.36 3.24

10 290 200 2.5 10 18 1842.5 28.51 16.53

11 290 200 2.5 10 28 1949.0 4.35 3.85

12 290 200 2.5 10 38 1887.5 8.78 7.48

13 290 250 4.0 5 18 1300.0 3.21 3.64

14 290 250 4.0 5 28 1379.0 3.56 4.58

15 290 250 4.0 5 38 1382.5 4.46 19.05

16 290 350 8.0 7 18 1378.0 37.33 24.78

22
17 290 350 8.0 7 28 1565.5 12.87 8.16

18 290 350 8.0 7 38 1551.0 17.01 16.71

19 320 200 4.0 7 18 1409.5 33.63 46.18

20 320 200 4.0 7 28 1577.0 37.08 20.15

21 320 200 4.0 7 38 1474.5 36.62 21.48

22 320 250 8.0 10 18 1788.0 4.35 3.78

23 320 250 8.0 10 28 1787.5 4.52 4.48

24 320 250 8.0 10 38 1661.5 4.93 4.59

25 320 350 2.5 5 18 1443.0 3.46 3.68

26 320 350 2.5 5 28 1439.5 4.37 3.25

27 320 350 2.5 5 38 1548.5 9.83 6.61

23
Table 3 Analysis of variance for top kerf width.

Source DF Seq SS Adj SS Adj F- P- Contribution

MS Value Value

Pressure 2 36972 36972 18486 2.89 0.085 2.80%

Abrasive mass flow 2 14673 14673 7337 1.15 0.342 1.11%

rate

Traverse speed 2 139211 139211 69605 10.88 0.001 10.55%

Standoff distance 2 967779 967779 483890 75.64 0.000 73.35%

Material thickness 2 58376 58376 29188 4.56 0.027 4.42%

Error 16 102363 102363 6398 7.76%

Total 26 1319374 100.00%

DF: Degrees of freedom, Seq SS: Sequential sums of squares, Adj SS: Adjusted sums of

squares, Adj MS: Adjusted mean squares.

24
Table 4 Analysis of variance for initial surface roughness.

Source DF Seq Adj Adj F- P- Contribution

SS SS MS Value Value

Pressure 2 685.5 685.5 342.75 9.18 0.002 18.12%

Abrasive mass flow rate 2 863.8 863.8 431.91 11.57 0.001 22.83%

Traverse speed 2 198.7 198.7 99.33 2.66 0.101 5.25%

Standoff distance 2 1328.3 1328.3 664.15 17.79 0.000 35.11%

Material thickness 2 109.5 109.5 54.75 1.47 0.260 2.89%

Error 16 597.4 597.4 37.34 15.79%

Total 26 3783.2 100.00%

DF: Degrees of freedom, Seq SS: Sequential sums of squares, Adj SS: Adjusted sums of

squares, Adj MS: Adjusted mean squares.

25
Table 5 Analysis of variance for zone average surface roughness.

Source DF Seq Adj Adj F- P- Contribution

SS SS MS Value Value

Pressure 2 427.1 427.1 213.53 5.35 0.017 16.53%

Abrasive mass flow rate 2 340.3 340.3 170.17 4.26 0.033 13.18%

Traverse speed 2 315.7 315.7 157.84 3.95 0.040 12.22%

Standoff distance 2 689.9 689.9 344.93 8.64 0.003 26.71%

Material thickness 2 171.0 171.0 85.51 2.14 0.150 6.62%

Error 16 639.1 639.1 39.94 24.74%

Total 26 2583.1 100.00%

DF: Degrees of freedom, Seq SS: Sequential sums of squares, Adj SS: Adjusted sums of

squares, Adj MS: Adjusted mean squares.

26
Table 6 Optimum levels of experimental parameters for response parameter.

Response Optimum levels of experimental parameters

parameter Pressure Abrasive Traverse Standoff Material

mass flow rate speed distance thickness

Top kerf width (Wt) 320 200 8.0 5 18

Initial surface 260 250 2.5 5 28

roughness (Rai)

Zone average 260 250 2.5 5 28

surface roughness

(Raz)

27
Table 7 Confirmation test values for response parameter.

Response parameter Confirmation test values

Experimental Predictive % Error

Top kerf width (Wt) 1050.5 1037.89 1.20

Initial surface roughness (Rai) 3.01 2.823 6.21

Zone average surface roughness (Raz) 3.54 3.39 4.24

28
Figure 1 Cutting process with AWJ.

29
Figure 2 Measurements: (a) Top kerf width (Wt) measurements; Average surface

roughness (Ra) measurements.

30
Figure 3 The influence of experimental parameters.

31
Figure 4 Effects of level of experimental parameters on top kerf width.

32
Figure 5 Effects of level of experimental parameters on initial surface roughness.

33
Figure 6 Effects of level of experimental parameters on zone average surface roughness.

34