Performance Enhancement of Abrasive Waterjet Cutting
Performance Enhancement of Abrasive Waterjet Cutting
Proefschrift
ter verkrijging van de graad van doctor
aan de Technische Universiteit Delft,
op gezag van de Rector Magnificus prof. dr. ir. J.T. Fokkema,
voorzitter van het College voor Promoties,
in het openbaar te verdedigen
op maandag 19 mei 2008 om 10.00 uur
door
Vu Ngoc PI
Master of Engineering
Hanoi University of Technology, Vietnam
Geboren te Thai Binh, Vietnam
Dit proefschrift is goedgekeurd door de promotoren:
Prof. Dr.-Ing. habil B. Karpuschewski toegevoegd promotor: dr. ir. A.M. Hoogstrate
Samenstelling promotie commissie:
Rector Magnificus voorzitter Prof. Dr.-Ing. habil B. Karpuschewski Otto-von-Guericke-Universität Magdeburg, promotor Dr. ir. A.M. Hoogstrate TNO Science and Industry, toegevoegd promotor Prof. dr. ir. J.R. Duflou Katholieke Universiteit Leuven Prof. Dr.-Ing. H. Louis Leibniz Universität Hannover Prof. dr. ir. A.J. Huis in ‘t Veld Universiteit Twente Prof. dr. U. Staufer Technische Universiteit Delft Prof. dr. M.A. Guitierrez De La Merced Technische Universiteit Delft, reservelid
ISBN: 978-90-9023096-2
Printed by PrintPartners Ipskamp, Rotterdam, The Netherlands.
Copyright © 2008 by Vu Ngoc Pi
All rights reserved. No part of this publication may be reproduced, utilized or stored in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without permission from the copyright holder.
iv
Dedicated to my wife Hoang Thi Tham
Acknowledgement
First of all, I would like to express my deep and sincere gratitude to Prof. Dr.-Ing. habil. Bernhard Karpuschewski, my promoter, for his guidance, consideration, and critical review of the present thesis.
I would like to give sincere appreciation to Dr. ir. André Hoogstrate, my adjunct supervisor, for his useful discussion, for his detailed and constructive comments, and for his important support throughout this work.
Then I would like to express my special thanks to Dr. Mohamed Hashish (Flow International Cooperation), Dr. Jey Zeng (OMAX Corporation), Dr. Eric Chalmers (AccuStream Inc.), Dr. Greg Mort (KMT Waterjet Systems Inc.), Dr. Andreas Höfner (GMA Garnet (Europe) GmbH), and Prof. Deng Jianxin (Shandong University), for their discussions, documents, and encouragement.
Furthermore I would like to thank Paolo Golfiotti, my Italian MSc student, for his helping in the abrasive recycling experiments.
Not to forget I would like to record my gratitude to the staff members of the department of Precision and Microsystems Engineering, especially Associate Prof. Marcel Tichem, Dr. Sebastiaan Berendse, Harry Jansen and Marli Guffens, for their supports of my works.
My work is a cooperation between the Delft University of Technology and the Vietnamese Government. The work is supported by Training Scientific and Technical Cadres in Institutions Overseas with the State Budget (Project 322) and the Management Centre for International Cooperation (CICAT). I would like to give my appreciation to all members of CICAT and 322, especially Dr. Paul Althuis, Veronique van der Varst, Willemijn van der Toorn, and Ngoc Lien, for their helps and encouragement.
I would like to thank all my colleagues and friends in and outside TU Delft for their encouraging and helping. Very special thanks to Tolga Susuzlu for his help in my experimental work and to Dr. Thieu Quang Tuan and Jeroen Derkx for their review of my thesis. Also, thanks to Nguyen Thanh Hoan for his help by taking pictures of my experimental setup.
Also I would like to give my appreciation to Prof. Nguyen Dang Binh, Associate Prof. Phan Quang The, Associate Prof. Nguyen Dang Hoe and Associate Prof. Vu Quy Dac from the Thai Nguyen University of Technology, for their support and encouragement.
i Last but not least, I would like to thank my mother, my mother-in-law, my sister and brothers, for their love and their encouragement. Thanks to my nephew Vu Quang Dien for his help by designing the cover of my book. I would like to thank my wife, Hong Tham, for her love, patience, enormous support, review of my thesis, and finally for taking care of our children. I also would like to thank my daughters, Thu Trang and My Hanh, for their love and back up.
Delft, May 2008,
Vu Ngoc Pi
ii Summary
Abrasive Waterjet (AWJ) Machining is a recent non-traditional machining process. Major part of this technology is a very high-pressure beam of water and abrasives, which is used for machining. The typical water pressure ranges from 300 to 380 MPa and the typical beam diameter varies between 0.6 and 1.2 mm. This technology is widely used in industry for cutting difficult-to-machine- materials, milling slots, polishing hard materials, cleaning contaminated surfaces, etc. AWJ machining has many advantages, e.g. it can cut net-shape parts, no heat is generated during the cutting process, it is particularly environmentally friendly as it is clean and it does not create dust.
Although AWJ machining has many advantages, a big disadvantage of this technology is its relatively high cutting cost. Consequently, the reduction of the machining cost and the increase of the profit rate are big challenges in AWJ technology.
To reduce the total cutting cost as well as to increase the profit rate, this research focuses on performance enhancement of AWJ cutting with two possible solutions including optimization in the cutting process and abrasive recycling.
The first solution to enhance the AWJ cutting performance is the optimization of the AWJ cutting process. As a precondition, it is necessary to have a cutting process model for optimization. In order to use that model for this purpose, several important requirements are given. The most important requirement for such a model is that it can describe the “optimum relation” between the optimum abrasive mass flow rate and the maximum depth of cut.
To develop a cutting process model which can be used for the AWJ optimization, many available models have been analyzed. Since the most important requirement for a process model (see above) can be obtained from Hoogstrate’s model, an extension of this model is carried out. The extension model consists of three sub-models including pure waterjet model, abrasive waterjet model and abrasive-work material interaction model. The pure waterjet model enables to determine the energy transfer from pressurized water to the pure waterjet. The abrasive waterjet model is used to calculate the energy transfer from pure waterjet to the abrasive particles. The abrasive– work material interaction model is used to identify the relation between the work material characteristics, the abrasive characteristics and the cutting efficiency in the process of removing work material chips by using the kinetic energy of abrasive particles.
The extension cutting process model is more accurate than the original one and it is capable to
iii optimize AWJ systems. The influence of many process parameters such as the water pressure, the abrasive mass flow rate, the nozzle diameter, the abrasive particle diameter etc. have been taken into account. By modeling the work material coefficient, the extension model can be used for various work materials. Also, by giving a model for the abrasive material coefficient, the model can be applied for several most common abrasive types.
Up to now, there has not been a model for the prediction of AWJ nozzle wear. Therefore, modeling the nozzle wear rate has been carried out and a model for the wear rate of nozzles made from composite carbide has been proposed. The model can be used in the optimization problems as well as in the calculation of the AWJ cutting regime.
Based on the extension cutting process model, two types of optimization applications have been carried out. They are related to technical problems and economical problems. The optimization problems have been solved in order to determine the optimum exchange nozzle diameter and the optimum abrasive mass flow rate for getting different objectives including the maximum depth of cut (for technical problems), the minimum total cutting cost and the maximum profit rate (for the economical problems). From the results of these considerations, regression models for determining the optimum nozzle exchange diameter and the optimum abrasive mass flow rate for various objectives have been proposed.
In AWJ machining, there are many cutting process parameters. Therefore, the ways to select other process parameters optimumly have also been investigated. The procedure for the determination of an optimum cutting regime then is given.
The other solution to enhance the cutting performance is abrasive recycling. In the present study, GMA garnet, the most popular abrasives for blast cleaning and waterjet cutting, has been chosen for the investigation. The recycling of GMA abrasives has been investigated on both technical side and economical side. On the technical side, the reusability and the cutting performance of the recycled and recharged abrasives have been analysed. The influence of the recycled and recharged abrasives on the cutting quality was studied. Also, the optimum particle size of recycled and recharged abrasives for the maximum cutting performance has been detected. On the economical side, first, the prediction of the cost of recycled and recharged abrasives was done. Then, the economic comparisons for selecting abrasives have been carried out. In addition, the economics of cutting with recycled and recharged abrasives have been studied. Several suggestions for an abrasive recycling process which promises a more effective use of the grains have been proposed.
By optimization in the cutting process and by abrasive recycling, the cutting performance can be increased, the total cutting cost can be reduced, and the profit rate can be enlarged considerably. Consequently, the performance of AWJ cutting can be enhanced significantly.
iv Samenvatting
Abrasief waterstraal snijden (AWS) is een recent, niet conventioneel verspaningsproces. Het is een technologie waarbij een waterstraal onder hoge druk, gemengd met abrasief, gebruikt wordt voor de verspaning van diverse materialen. De waterdruk ligt tussen de 300 en 380 MPa; de waterstraaldiameter ligt tussen de 0.6 en 1.2 mm. Waterstraaltechnologie wordt veel gebruikt in de industrie voor het snijden van moeilijk bewerkbare materialen, het boren van gaten, het polijsten van harde materialen, het reinigen van vervuilde oppervlakken etc. AWS bewerken heeft vele voordelen waaronder: het maken van “near-net-shape” onderdelen, geen warmte ontwikkeling tijdens het verspaningsproces en het is bijzonder milieuvriendelijk omdat het schoon is en er geen fijnstof of gevaarlijke stoffen vrij komen.
Naast de vele voordelen die AWS snijden biedt zijn de hoge kosten een belangrijk nadeel. Daarom zijn de reductie van de bewerkingskosten en het verhogen van de winstmarge belangrijke uitdagingen in de AWS technologie.
Om zowel de totale bewerkingskosten te reduceren alsook de winstmarge te verhogen, concentreert dit onderzoek zich op de prestatieverbetering van AWS snijden. Twee mogelijke oplossingen worden bekeken: optimalisatie van het bewerkingsproces en hergebruik van abrasief.
De eerste oplossing om de prestatie van AWS bewerken te verbeteren is de optimalisatie van het AWS proces. Voorwaarde hiervoor is de beschikbaarheid van een procesmodel van de verspaning voor de optimalisatie. Om een model te kunnen gebruiken voor dit doel moet het aan enkele belangrijke voorwaarden voldoen. De belangrijkst daarvan is dat het model de relatie tussen de abrasief massa stroom en de maximale snedediepte beschrijft.
Voor de ontwikkeling van een procesmodel dat gebruikt kan worden voor de AWS optimalisatie zijn vele beschikbare modellen geanalyseerd. Omdat aan de belangrijkste voorwaarde voor een proces model (zie boven) kan worden voldaan door het model van Hoogstrate, wordt een uitbreiding van dit model uitgevoerd.
Het uitgebreide model bestaat uit 3 deelmodellen: het pure waterstraalmodel, het abrasieve waterstraalmodel en het abrasief-werkstukmateriaal interactie model. Het pure waterstraal model maakt het mogelijk de energie overdracht te bepalen van het samengeperste water naar de pure waterstraal. Het abrasieve waterstraalmodel wordt gebruikt om de energie overdracht te berekenen van de pure waterstraal naar de abrasieve deeltjes. Het abrasief-werkstukmateriaal interactie model
v wordt gebruikt om de relatie te bepalen tussen de werkstuk materiaal eigenschappen, het abrasief en de verspaningsefficiëntie tijdens het verwijderen van spanen van het werkstuk door gebruik te maken van de kinetische energie van de abrasieve deeltjes.
Het uitgebreide procesmodel is nauwkeuriger dan het originele model en kan gebruikt worden voor het optimaliseren van AWS systemen. De invloed van diverse procesparameters zoals de waterdruk, de abrasief massastroom, de orifice diameter, de deeltjesgrootte etc. zijn in het model meegenomen. De introductie van een werkstukmateriaal-coëfficiënt maakt het model bruikbaar voor diverse werkstuk materialen. Tevens kan het model gebruikt worden voor verschillende abrasief materialen door de introductie van een abrasiefmateriaal-coëfficiënt.
Tot dusver was er geen model beschikbaar voor de voorspelling van de slijtage van de AWS focusbuis. Daarom is de slijtagesnelheid van de focusbuis gemodelleerd en een model voor de slijtagesnelheid van gesinterde wolfraamcarbide focusbuizen voorgesteld. Dit model kan zowel worden gebruikt voor de optimalisatie van het AWS proces.
Twee types van optimalisaties zijn uitgevoerd, gebaseerd op het uitgebreide procesmodel. Deze zijn gerelateerd aan technische en economische optimalisatie. De optimalisatie functie is zodanig opgesteld, dat de optimale focusbuis wissel diameter en de optimale abrasief massastroom konden worden bepaald. Dit is gedaan voor verschillende doelstellingen waaronder de maximale snedediepte (de technische doelstelling) en de minimale bewerkingskosten en maximale winstmarge (de economische doelstellingen). Gebaseerd op de resultaten van deze overwegingen zijn regressie modellen voorgesteld voor het bepalen van de optimale focusbuis wissel-diameter en de optimale abrasief massastroom voor de verschillende doelstellingen.
Er zijn vele procesparameters in AWS bewerken. Daarom zijn de diverse methodes om de optimale procesparameters te bepalen ook onderzocht. De procedure voor het bepalen van een optimaal verspaningsregiem wordt vervolgens gegeven.
De andere oplossing om de verspaningsprestatie te verbeteren is het hergebruik van het abrasief. In dit onderzoek is gebruik gemaakt van het meest populaire abrasief voor waterstraal snijden en reinigen: GMA garnet. Zowel de technische als de economische kant van het hergebruik van GMA garnet zijn onderzocht. Op het technische vlak zijn de herbuikbaarheid en de verspaningsprestatie van het hergebruikte abrasief geanalyseerd. Hierbij is zowel het batch-gewijze hergebruik van abrasief, alsook het gradueel opmengen van gebruikt en nieuw abrasief geanalyseerd. De invloed van beide recycle-methodes op de verspaningskwaliteit is onderzocht. Tevens is de optimale deeltjesgrootte voor hergebruik bij beide methodes, gerelateerd aan de maximale verspaningsprestatie bepaald. Op het economische vlak is allereerst een voorspelling gedaan van de kosten van beide recycle-methodes; vervolgens is een economische vergelijking voor de selectie van abrasieven gedaan. Daarbij zijn ook de kosten bestudeerd van het bewerken met hergebruikt
vi en met toegevoegd abrasief. Er zijn verschillende suggesties gedaan voor een hergebruik proces dat een effectiever gebruik van het abrasief materiaal belooft.
Door optimalisatie van het verspaningsproces en hergebruik van het abrasief kan de verspaningsprestatie worden verhoogd, de totale verspaningskosten worden gereduceerd en de winstmarge aanmerkelijk worden vergroot. Daardoor kan de prestatie van AWS bewerken significant worden verbeterd.
vii
viii
Contents
Acknowledgement ...... i Summary ...... iii Samenvatting...... v Nomenclature...... xiii 1 Introduction ...... 1 1.1 Historical review...... 1 1.2 Introduction to AWJ Technology...... 2 1.2.1 Introduction to an AWJ cutting system ...... 2 1.2.2 Parameters of an AWJ machining process ...... 5 1.2.3 Advantages and disadvantages of AWJ Technology ...... 6 1.3 Challenges in AWJ Technology ...... 6 2 State of the art in optimization of AWJ machining...... 9 2.1 State of the art in AWJ technical optimization...... 10 2.1.1 Optimum combination of focusing tube and orifice diameter ...... 10 2.1.2 Optimum focusing tube length ...... 11 2.1.3 Optimum abrasive mass flow rate...... 12 2.1.4 Optimum abrasive particle size...... 15 2.1.5 Optimum standoff distance ...... 16 2.2 State of the art in AWJ cost calculation and cost optimization...... 16 2.2.1 State of the art in AWJ cost calculation ...... 16 2.2.2 State of the art in AWJ cost optimization...... 18 2.3 State of the art in AWJ abrasive recycling ...... 21 2.4 Conclusions...... 25 3 Project definition ...... 27 3.1 Aim of the investigations...... 27 3.2 Outline of the thesis ...... 28 4 Used experimental and measuring equipment ...... 29 4.1 AWJ machining setup ...... 29 4.2 Abrasive particles ...... 30
ix 4.2.1 Abrasive properties...... 30 4.2.2 Abrasive size distribution and abrasive particle diameter ...... 33 4.3 Work materials...... 35 4.4 Experimental setup for measuring the water flow rate ...... 36 4.5 Experimental setup for measuring the reaction force ...... 37 4.6 Experimental setup for determining the maximum depth of cut...... 38 4.7 Experimental setup for collecting abrasives ...... 38 4.8 Experimental setup for determining surface roughness ...... 40 4.9 Other measuring equipment...... 41 4.9.1 Microscope...... 41 4.9.2 Surface roughness measurement device ...... 41 5 Frame work of modeling and AWJ optimization approach...... 43 5.1 Frame work of modeling ...... 43 5.2 AWJ optimization approach ...... 45 5.2.1 Introduction to optimization ...... 45 5.2.2 Statement of an AWJ optimization problem...... 51 5.2.3 Solutions for AWJ optimization problems...... 51 6 Modeling the cutting process for AWJ optimization ...... 53 6.1 Requirements for an AWJ cutting process model ...... 53 6.2 State of the art in AWJ cutting process modeling...... 54 6.2.1 Studies of Hashish...... 54 6.2.2 Studies of Zeng and Kim...... 56 6.2.3 Other studies ...... 57 6.3 Introduction to Hoogstrate’s model...... 58 6.3.1 Model description ...... 58 6.3.2 Discussion...... 61 6.4 Extension of Hoogstrate’s model ...... 62 6.4.1 Pure waterjet modeling...... 62 6.4.2 Abrasive waterjet modeling...... 71 6.4.3 Abrasive - work material interaction modeling ...... 80 6.5 Modeling the AWJ cutting process ...... 91 6.6 Conclusions...... 94 7 Optimization in AWJ cutting process ...... 95 7.1 Cost and profit analysis...... 95 7.1.1 Cost analysis...... 95 7.1.2 Profit analysis...... 99 7.2 Optimization for determining optimum nozzle lifetime...... 100
x 7.2.1 Nozzle lifetime and nozzle wear in AWJ machining...... 101 7.2.2 Relation between the nozzle lifetime and the feed speed ...... 106 7.2.3 Optimization for determining optimum nozzle lifetime for minimum cutting cost..... 109 7.2.4 Optimization for finding optimum nozzle lifetime for maximum profit rate ...... 115 7.2.5 Benefits of cutting with optimum nozzle lifetime...... 117 7.2.6 Conclusions...... 120 7.3 Optimization for determining the optimum abrasive mass flow rate ...... 121 7.3.1 Optimization for determining the optimum abrasive mass flow rate for maximum cutting performance...... 121 7.3.2 Optimization for determining the optimum abrasive mass flow rate for minimum cutting cost ...... 124 7.3.3 Optimization for determining optimum abrasive mass flow rate for maximum profit rate 126 7.3.4 Benefits of cutting with the optimum abrasive mass flow rate ...... 130 7.3.5 Conclusions...... 132 7.4 Selection of process parameters for the optimum cutting regime...... 133 7.4.1 Optimum selection of the number of jet formers, the orifice diameter, and the nozzle diameter 134 7.4.2 Optimum selection of abrasive type and size ...... 136 7.4.3 Procedure for determination of the optimum AWJ cutting regime ...... 137 7.5 Conclusions...... 138 8 Recycling and recharging of abrasives ...... 139 8.1 Reusability of abrasives ...... 139 8.1.1 Experimental setup...... 139 8.1.2 Results and discussions...... 140 8.2 Cutting performance and cutting quality of recycled abrasives...... 141 8.2.1 Experimental setup...... 141 8.2.2 Results and discussions...... 142 8.3 Cutting performance and cutting quality of recharged abrasives ...... 147 8.3.1 Experimental setup...... 147 8.3.2 Results and discussions...... 148 8.3.3 Multi-recharging of abrasive...... 150 8.4 Conclusions...... 151 9 Economics of abrasive recycling ...... 153 9.1 Cost calculation for recycled and recharged abrasives...... 153 9.1.1 Cost analysis ...... 153 9.1.2 Results and discussions...... 154
xi 9.2 Economic comparisons for selecting abrasives...... 157 9.3 Economics of cutting with recycled and recharged abrasives ...... 159 9.3.1 Economics of cutting with recycled abrasives ...... 159 9.3.2 Economics of cutting with recharged abrasives...... 166 9.3.3 Comparisons among cutting with new, recycled and recharged abrasives ...... 170 9.4 Suggestions for abrasive recycling process...... 170 9.5 Conclusions...... 171 10 Conclusions and recommendations for further research ...... 173 10.1 Conclusions...... 173 10.2 Recommendations...... 176 References...... 177 Appendix: Recycling system ...... 183 A.1 WARD 1 ...... 183 A.2 WARD 2 ...... 184 About the author...... 187
xii Nomenclature
Symbols
Symbol Unit Definition
A m2 cross section area C € cost c - coefficient d m diameter E MPa elasticity
3 ec J/m specific cutting energy F N force
Grecy kg/h recycling capacity g h m depth k - coefficient l m length
Nm - machinability number n - number m kg/s mass flow rate
P w power Pr € profit p Pa pressure Q - quality number r % reusability R - abrasive load ratio
Re - Reynolds number T s time v m/s velocity x - number
xiii η - momentum transfer efficiency coefficient κ - power transfer efficiency δ m/s wear ξ - cutting efficiency coefficient ρ Kg/m3 density
Subscripts
Subscript Definition a related to average abr related to abrasive actual related to actual awj related to abrasive waterjet c related to cutting com compressible d related to discharge de related to depreciation e related to electrical en related to energy f focusing tube / nozzle h related to hour inc related to incompressible int related to interest l related to length la related to labor m related to mass ma related to maintenance max maximum min minimum msh related to manned shifts mt related to machine tool inc incompressible
xiv op related to optimum ori orifice ov related to overhead p related to particle q related to quality rech related to abrasive recharging recy related to abrasive recycling ro related to occupied room rpl related to replacement sal related to sale sqm related to squared meter sh related to shift th theoretical use related to time of use ut related to utilization w related to water wa related to wages wj related to waterjet wor related to working y related to year
xv
xvi 1 Introduction
Abrasive Waterjet (AWJ) Machining is a recent non-conventional machining process. In this technology, a very high-pressure beam of water and abrasives is used for machining. This technology is widely used in industry as it has many advantages.
In this chapter an introduction to Abrasive Waterjet (AWJ) Technology is provided. A review of the AWJ history is first carried out to draw a picture of the progress in this technology. Brief descriptions of the schema and the main components of an AWJ system are also given. Advantages and drawbacks of the AWJ technology are then evaluated. Challenges of the technology are discussed in the end.
1.1 Historical review
AWJ machining has been developed from Waterjet machining. The earliest use of the water beam in coal mining was in the former Soviet Union and New Zealand [Summ95]. This mining technique was also used for removing blasted rocks from working areas into collection tunnels.
From 1853 to 1886, pressurized water was used for excavating soft gold rocks. The pressurized water for coal mining was also used in Prussia in the early 1900s and then in Russia in the 1930s [Summ95].
In 1936, Peter Tupitsyn, who was working for the Donetsk Coal Basin in Ukraine, proposed the idea of using a waterjet beam to cut boreholes in the coal bed [Chri03].
In the 1950s, Dr. Norman Franz, a forestry engineer, was the first who studied the use of a waterjet beam as a cutting tool for wood processing [Flow08]. However, the first patent of a waterjet cutting system was granted for the staff of McCartney Manufacturing Company, a division of the Ingersoll-Rand Corp. [Tikh92]. In 1971, the first commercial waterjet machine was introduced into the market by this company [Tikh92].
In 1979, Dr. Mohamed Hashish, who has worked for Flow International Cooperation, invented the abrasive waterjet cutting method by adding abrasives into the pure waterjet [Flow08]. Soon after this, in 1980, abrasive waterjet was first used to cut glass, steel, and concrete [Flow08]. The invention of AWJ led to a huge expansion of applications of cutting with high-pressure water. Since then, AWJ has been widely used in various industries such as cutting of a wide variety of sheet
1 materials, cleaning of contaminated surfaces, polishing of hard-to-machine materials, etc.
1.2 Introduction to AWJ Technology
1.2.1 Introduction to an AWJ cutting system
There are two types of waterjets: pure (or plain) waterjet and abrasive waterjet. In pure waterjet cutting, only a pressurized stream of water is used to cut through materials. This type of cutting is used to cut soft materials such as cardboard, leather, textiles, fibre plastics, food or thin plates of aluminium. In AWJ cutting, an abrasive waterjet entrainment system mixes abrasives with the waterjet in a mixing chamber following an orifice (Figure 1.1). The abrasive particles are accelerated by the water stream and then leave the focusing tube (or the nozzle) with the stream. AWJ cutting is used for cutting harder materials such as stainless steel, glass, ceramics, titanium alloys, composite materials, and so forth.
Intensifier Attenuator High-pressure water
Mixing chamber
Orifice
Abrasive particles
Electric Hydraulic Directional Focusing motor pump control Inlet tube valve water Presure Water generation preparation Jet system former system
Figure 1.1: AWJ entrainment system schema
A typical AWJ entrainment system (as shown in Figure 1.1) consists of four main parts: the water preparation system, the pressure generation system, the jet former, and the abrasive supply system. A brief description of these parts is given below:
• The water preparation system:
2 The water preparation system is used for supplying purified water for the pressure generation system. Generally, particles larger than 1 μm have to be filtered out to prevent unacceptable wear of the critical parts of the pressure generation system [Hoog00].
• The pressure generation system:
This system is equipped with a pump to ensure a continuous and stable flow of high pressure. Three types of pumps, namely intensifier, crankshaft and direct pumps can be distinguished.
Figure 1.2: Direct pump (Courtesy of Flow International Cooperation)
Direct pumps are used for applications with low pressure such as cleaning, or washing a desk or a work place etc. In a direct pump, the movement of three plungers is transmitted directly from the electric motor (see Figure 1.2).
Figure 1.3: Double-acting intensifier
Intensifier pumps (Figure 1.3) are used for applications with water pressure up to 600 MPa. In an intensifier pump, a double-acting cylinder in which the movement of the piston is driven by a hydraulic system is used. Two small diameter cylinders at each end of the hydraulic cylinder help to
3 pressurize the water alternately as the hydraulic piston moves back and forth. By connecting two intensifier pumps in series, the output water pressure can be up to 800 MPa [Susu04].
The third type is the crankshaft pump, which can provide the pressure up to 345 MPa [Chri03]. An example of this pump is shown in Figure 1.4. It is known that the efficiency of crankshaft pumps is higher than that of intensifier pumps because crankshaft pumps do not require a power-robbing hydraulic system.
Figure 1.4: Crankshaft pump (Courtesy of OMAX Corp. Kent, WA)
• The jet former:
The jet former is used to transfer part of the hydraulic water energy into kinetic energy of water, and then into kinetic energy of abrasive particles. Figure 1.5 shows a typical jet former for AWJ cutting [Hoog04]. To form the abrasive waterjet, first, the high pressure water is forced through an orifice to create a high speed waterjet. Then the high speed waterjet passes through a mixing chamber, which is installed downstream of the orifice. Because of the Venturi effect, a vacuum is created in the mixing chamber. As a result, the abrasive particles and some air are sucked into the mixing chamber through a feed line. After entering the mixing chamber, the particles are accelerated by the high-speed waterjet (velocity about 600 to 900 m/s) and then passing through a focusing tube (or nozzle).
As mentioned above, the orifice, the mixing chamber and the focusing tube are the main parts of a jet former. Orifices can be made of sapphire, ruby or diamond with a diameter ranging from 0.08 to 0.8 mm [Hoog00]. The lifetime of a diamond orifice is about 1000 to 2000 hours while it is only 40 to 70 hours for sapphire [Koel02]. However, sapphire orifices are most commonly used because they are much cheaper than diamond orifices (the price of a diamond orifice can be $435 while it is
4 only $14.5 for a sapphire one [Bart08]).
Most of AWJ nozzles are made from composite carbide materials. They are available on the market under specific product names such as ROCTEC 100 and ROCTEC 500 from Kennametal Inc. ROCTEC composite carbide is a very dense, sintered, tungsten carbide based hardmetal. The common inner diameter of the focusing tube is from 0.5 to 1.5 mm, and the common length is from 70 to 100 mm.
• The abrasive supply system:
The abrasive supply system is used for accurate supply of abrasives with a pre-required mass flow rate. In practice, there are many types of abrasives which are used in AWJ machining. They can be garnet (for example Barton garnet (a trade mark of Barton Mines Company) and GMA garnet (a trade mark of GMA garnet Pty Ltd) – two most common garnets), olivine, aluminum oxide, silica- sand etc. Generally, in AWJ machining, the abrasive mass flow rate is about 0.08 to 0.5 kg/min (15 to 30 kg/h [Trum97]) and the abrasive size varies between 0.1 and 0.3 mm.
high pressure water orifice abrasive supply
mixing chamber focusing tube
Figure 1.5: A typical jet former for AWJ cutting [Hoog04]
1.2.2 Parameters of an AWJ machining process
There are many parameters involved in an AWJ machining process. In general, these parameters can be divided into two groups: process parameters and target parameters [Momb98]:
• Process parameters:
The process parameters include parameters relating to the forming of the AWJ beam. These parameters can be sorted into four following sub-groups [Momb98]:
-Hydraulic parameters including water pressure and orifice diameter.
5 -Mixing parameters including focusing tube (or nozzle) diameter and focusing tube length.
-Abrasive parameters including abrasive material, abrasive particle size, abrasive shape, and abrasive mass flow rate.
-Cutting parameters including standoff distance, impact angle, traverse rate and number of passes.
• Target parameters:
The target parameters consist of parameters related to the target of the machining. These parameters are the work material, the depth of cut and the cutting quality.
1.2.3 Advantages and disadvantages of AWJ Technology
AWJ cutting has various advantages over other non-conventional techniques such as laser and Electrical Discharge Machining (EDM). The advantages can be presented as follows:
-AWJ can machine a wide range of materials including titanium, stainless steel, aerospace alloys, glass, plastics, ceramics, and so on.
-AWJ can cut net-shape parts and near net-shape parts.
-No heat is generated in the cutting process. Therefore, there is no heat-affected area and thus no structural changes in work materials occur.
-AWJ cutting is particularly environmentally friendly as it does not generate any cutting dust or chemical air pollutants.
-The abrasives after cutting can be reused which allows for possible reduction of the AWJ cutting cost.
-Only one nozzle can be used to machine various types of work materials and workpiece shapes.
-AWJ machining can be easily automated and therefore can be run with unmanned shifts.
Although AWJ cutting is a truly useful machining process and can be used for various applications, the technology still has two major disadvantages:
-The total cutting cost is relatively high;
-The cutting quality is not always satisfying and unstable.
1.3 Challenges in AWJ Technology
As mentioned above, although AWJ cutting has many advantages, its high cutting cost is the most
6 prominent disadvantage. In AWJ cutting, the total cutting cost depends on many cost components such as machine tool cost, abrasive cost, nozzle wear cost, wages including overhead cost and so on. High AWJ cutting cost, for example, in Europe, the cutting cost per hour is about 150…200 (€/h), makes the AWJ business less competitive. As a result, the reduction of the total cutting cost and cutting time as well as the increase of the profit rate (or profit per hour) in AWJ machining are big challenges for this technology.
7
8 2 State of the art in optimization of AWJ machining
As addressed in Chapter 1, one of the biggest disadvantages of AWJ cutting is its high cost. The AWJ cutting cost per hour, for example, can be 150…200 (€/h) in Europe. Therefore, finding solutions to reduce the total cutting cost to increase the profitability for AWJ users is an important task of AWJ technology.
In the AWJ cutting cost, the abrasive cost (including disposal cost) is usually the largest component (Figure 2.1). This can amount to 20% up to 70% of the total cutting cost, depending on parameters such as the abrasive mass flow rate, the number of cutting heads, the abrasive price, the AWJ system’s cost and so on. However, the abrasives after cutting can be reused, which can reduce the abrasive cost and the disposal cost.
Machine tool cost (23.94%)
Abrasive cost (53.98%) Wages including overhead cost (16.89%)
Orifice cost (0.92%)
Nozzle cost (3.05%) Water cost (1.31%)
Figure 2.1: A typical AWJ cost breakdown [Hoog06]
In practice, the AWJ optimization and abrasive recycling are two main ways to increase the profitability for AWJ users. Especially, optimization can reduce both the cutting time (or increase the cutting performance) and the cutting cost and can increase the profit rate. Therefore, AWJ optimization and abrasive recycling have been the objectives of many studies.
The optimization problems in AWJ machining can be divided into two categories including AWJ technical optimization and AWJ economical optimization. The technical optimization, based on the physical relationships between process parameters, aims to determine optimum values of the
9 process parameters in order to fulfill the maximum cutting performance or the minimum cutting time. The economical optimization, based on the economical relations as well as the physical relations between the process parameters, aims at the optimum values of the process parameters for getting the minimum total cutting cost per product (or per unit length of cutting) or the maximum profit rate.
Up to now, there have been many studies on both AWJ optimization and abrasive recycling. To have a clear picture on this, a literature review is carried out. The review is split into three parts: AWJ technical optimization in Section 2.1, AWJ cost calculation and cost optimization in Section 2.2, and abrasive recycling in Section 2.3.
2.1 State of the art in AWJ technical optimization
In the AWJ cutting process, there are various factors affecting the material removal process or the cutting performance. These factors include the jet-parameters (the water pressure, the orifice diameter, the focusing tube diameter, the focusing tube length, the abrasive mass flow rate, the abrasive size, the abrasive shape and type) and the cutting parameters (e.g. the standoff distance, the workmaterial, the feed speed).
2.1.1 Optimum combination of focusing tube and orifice diameter
90 35 /s)
3 80 (mm) max 70 30
60
50 25 40 p =240 MPa; d =0.25 mm p =240 MPa; v=1.67 mm/s w ori w v =1.67 mm/s; l =50 mm d =0.25; d =1.2 mm 30 f f ori f
R=0.3; AlMgSi0.5 Maximum depth of cut h R=0.3; AlMiSi0.5 Material removal rate Q (mm 20 20 0.5 1 1.5 2 2.5 3 0 20 40 60 80 100 Focusing tube diameter d (mm) Focusing tube length l (mm) f f Figure 2.2: Focusing tube diameter versus Figure 2.3: Focusing tube length versus material removal rate [Blic90] maximum depth of cut [Blic90]
H. Blickwedel [Blic90] investigated the relationship between the focusing tube diameter and the volume removal rate. The author notes that the final abrasive particle velocity depends on the density of the abrasive-water-air mixture: a denser mixture creates a higher particle velocity. Also,
10 as the focusing tube diameter increases, the density of the mixture decreases and therefore the particle velocity decreases. However, a small focusing tube diameter leads to more interactions between particles and nozzle wall, and particles with each other and thus reduces the particle velocity. Therefore, an optimum value of the focusing tube diameter exists for the material removal rate (see Figure 2.2).
H. Blickwedel [Blic90] proposed an optimum ratio between the focusing tube diameter and the orifice diameter as follows:
d f = 34… (2.1) d ori U. Himmelreich and W. Riess [Himm91] confirmed that the above ratio is a good value for AWJ formation. E.J. Chalmers [Chal91] observed that the maximum depth of cut will occur for the ratio of nozzle to orifice diameter of 3. Zeng and Munoz [Zeng94] also reported that the highest cutting performance is achieved when using the following optimum combination of focusing tube/orifice: 3.3 (0.023”/0.007”), 3.2 (0.032”/0.01”), and 3.14 (0.044”/0.014”).
2.1.2 Optimum focusing tube length
Figure 2.3 shows the relation between the focusing tube length and the maximum depth of cut [Blic90]. The depth of cut, at first, increases linearly with the increase of the nozzle length. This is because a certain acceleration distance is necessary to accelerate the injected abrasive particles [Momb98]. Beyond this critical acceleration distance, the friction due to the spreading water jet increases. This leads to a reduction of the particle velocity and therefore a decrease of the depth of cut [Momb98]. The optimum acceleration distance, as noted by M. Heβling [Heβl88], depends strongly on the abrasive material density. Figure 2.4 shows the relation between the focusing tube length and the maximum depth of cut for different abrasive materials [Heβl88]. It is observed that round steel cast abrasive material is most influenced by the nozzle length while broken abrasive material and quartz sand are only lightly affected (Figure 2.4).
H. Blickwedel [Blic90] suggested the optimum focusing tube length lf,op based on his experimental results:
l fop, = 25… 50 (2.2) d f
M. Hashish [Hash91] indicated that the depth of cut and the kerf width both depend on the length of the focusing tube. The depth of cut and the kerf width reduce as the focusing length increases up to a length of about 50 to 70 times of the focusing tube diameter. Also, it is noted that no change in the depth of cut and the kerf width occurs when the focusing tube length increases further beyond 50 to 70 times of the tube diameter [Hash91].
11 Generally, the wear of the focusing tube is affected by the tube length. M. Hashish [Hash94] addressed that the nozzle exit bore wear rate reduces as the nozzle length increases. This conclusion was also confirmed later by K.A. Schwetz et al [Schw95] and M. Nanduri et al. [Nand00]. Figure 2.5 illustrates the relation between the tube length and the exit bore diameter wear rate [Nand00]. It follows that when the tube length is smaller than a certain value (in this case around 75 mm), a decrease of the tube length will lead to a significant increase of the exit bore wear rate. Beyond this value, the exit bore wear rate is almost unaffected by the tube length.
In practice, the length of the focusing tubes is determined for both a high cutting performance and a long nozzle lifetime. The nozzle lengths are standardized in some common sizes of 76 mm (3”), 89 mm (3.5”) and 101.6 mm (4”). It is known that the most commonly used nozzle length is 76 mm, offering the best cost-to-wear-life ratio [Chal06].
80 30 3 p =310 MPa; m =3.8 g/s Steel cast, angular (7400 kg/m ) w a 70 3 d =0.38; d =1.14 mm Steel cast, round (7400 kg/m ) 25 ori f0
(mm) Nozzle material: WC/C Quartz sand, round (2650 kg/m3) o
max 60 Abrasive: aluminum oxide #80 20 50
40 15
30 10 20 p =200 MPa; v =20 mm/s w f d =0.6; d =600 µm ori p m =30 g/s 5 10 a Maximum depth of cut h Exit diameter increase rate (%) 0 0 0 50 100 150 20 40 60 80 100 120 Focusing tube length l (mm) f Focusing tube length l (mm) f Figure 2.4: Focusing tube length versus Figure 2.5: Nozzle length versus nozzle exit bore maximum cutting depth [Heβl88] increase rate [Nand00]
2.1.3 Optimum abrasive mass flow rate
Typical relations between the abrasive mass flow rate and the maximum depth of cut are shown in Figure 2.6. It follows that the depth of cut, at first, increases as the abrasive mass flow rate increases. However, when the abrasive mass flow rate exceeds a certain value, the depth of cut will drop (Figure 2.6). This relation can be explained by the following equation [Hash89]:
v v =⋅η wj (2.3) awj 1/+ mmaw
In which, vwj is the velocity of water leaving the orifice, vawj is the velocity of abrasive particles
12 leaving the nozzle, η is momentum transfer efficiency, m a is the abrasive mass flow rate, and m w is the water mass flow rate.
Previous studies ([Mill91], [Clau98] and [Susu06]) indicate that the momentum transfer efficiency η decreases as the abrasive mass flow rate increases. Hence, it can be deduced from Equation (2.3) that an increase of the abrasive mass flow rate can lead to a decrease of the abrasive particle velocity vawj . Moreover, an increase of the abrasive mass flow rate also results in an increase of the impact frequency of particles. Therefore, a critical value of the abrasive mass flow rate exists at which the benefit of the impact frequency balances the loss in particle velocity [Zeng94]. This critical value is the optimum abrasive mass flow rate for the maximum depth of cut.
34
32
(mm) 30
max 28
26 d =0.8 mm 24 f d =1.2 mm f d =1.6 mm 22 f 20 p =240 MPa; v =1.67 mm/s 18 w f d =0.25 mm; l =50 mm ori f
Maximal depth of cut h 16 AlMgSi0.5 14 0 5 10 15 20 25 Abrasive mass flow rate (g/s)
Figure 2.6: Abrasive mass flow rate versus maximum depth of cut [Owei89]
The optimum abrasive mass flow rate for the maximum cutting performance (or for the maximum depth of cut) depends on many parameters. These are the water pressure [Chal91], [Guo94a], [Guo94b], orifice diameter [Chal91], [Guo94b], the focusing tube diameter [Chal91], [Guo94a], [Hoog05] and the focusing tube length [Guo94a].
Table 2.1: Optimum abrasive to water mass flow rate [Chal91]
Nozzle/orifice combination mm aw/ for hmax mm aw/ for 0.85⋅ hmax 0.76 mm / 0.25 mm 0.3 0.17 1.14 mm / 0.38 mm 0.19 0.12 1.65 mm / 0.53 mm 0.19 0.1
Figure 2.7 shows the effects of jet-parameters on the optimum abrasive mass flow rate according to experimental data of Guo [Guo94a]. It follows that the optimum abrasive mass flow rate increases with the increase of the water pressure (Figure 2.7a), of the water mass flow rate (Figure 2.7b), and of the focusing tube diameter (Figure 2.7c). The relation between the optimum abrasive
13 mass flow rate and the focusing tube length is shown in Figure 2.7d.
To determine the optimum abrasive mass flow rate, E.J. Chalmers [Chal91] found that cutting with the ratio of nozzle to orifice of 3:1 results in the maximum depth of cut at a specific value of
mm aw/ for a given size of the nozzle. In addition, to avoid excessive use of abrasives, Chalmers
[Chal91] assumed the optimum depth of cut is defined as occurring at 0.85hmax. The optimum abrasive to water flow rate is shown in Table 1 [Chal91].
9 9
8.5 8.5
8 8
7.5 7.5
7 7
6.5 6.5
6 d =0.25; d =0.95 mm 6 ori f v =1.67 mm/s; d =355 µm p =240 MPa; v =1.67 mm/s f p w f 5.5 d =0.95 mm; d =355 µm minersiv 5.5 f p minersiv 5 5 100 150 200 250 300 350 15 20 25 30 Optimal abrasive mass flow rate (g/s) Water pressure p (MPa) Water mass flow rate (g/s) w Optimal abrasive mass flow rate (g/s)
a) b) 12 10 p =240 MPa; v=1.67 mm/s w 9.5 d =0.25 mm; d =0.95 mm 11 ori f d =355 µm; minersiv 9 p
10 8.5
8 9 7.5 8 7
7 6.5 p =240 MPa; v=1.67 mm/s 6 w 6 d =0.25 mm; d =0.95 mm ori f d =355 µm; minersiv 5.5 p 5 5 0.5 0.7 0.9 1.1 1.3 1.5 20 30 40 50 60 70 80
Optimal abrasive mass flow rate (g/s) Focusing tube diameter d (mm) Focusing tube length l (mm) f f Optimal abrasive mass flow rate (g/s)
c) d) Figure 2.7: Effect of factors on the optimum abrasive mass flow rate [Guo94a]
14 In spite of recent efforts, the optimum abrasive mass flow rates are predicted for specific combinations of the focusing tube and the orifice diameter only. No model has been developed for determination of the optimum abrasive mass flow rate for more general combinations.
2.1.4 Optimum abrasive particle size
The effect of abrasive particle sizes on the depth of cut was investigated by J. Ohlsen (Figure 2.8a) [Ohls97]. This effect can be classified into two cases: brittle behaving materials (e.g. glass in Figure 2.8a) and ductile behaving materials (e.g. AlMgSi0.5 in Figure 2.8a). In the former case, the workmaterial seems less sensitive to the impact frequency [Momb98]. Therefore, the maximum depth of cut increases with the increase of the particle diameter. In the latter case, for small particles (smaller than 100 μm), a larger particle diameter causes a higher depth of cut. This is
3 because a larger particle means a higher kinetic energy, i.e. Edp ∝ p [Momb98]. In contrast, for larger particles (larger than 100 μm), an increase in size of abrasive particles can lead to a reduction of the maximum depth of cut (Figure 2.8a). Momber et al. [Momb98] noted that this phenomenon was also observed by Nakamura et al. [Naka89], Guo et al. [Guo92] and Momber et al. [Momb96].
H. Oweinah [Owei89] investigated the effect of the abrasive particle diameter on the depth of cut for various abrasive mass flow rates (Figure 2.8b). It is concluded that large particles have a significant influence on the depth of cut when the abrasive mass flow rate varies, while smaller particles are not sensitive against the changes of the abrasive mass flow rate (Figure 2.8b).
45 80
p =400 MPa; v =0.83 mm/s AlMgSi0.5 70 w f 40 d =0.25; d =1.08 mm Glass ori f AlMgSi1; corundum 60 (mm) (mm)
35 max
max 50
30 40
30 25 p =300 MPa (AlMgSi0.5) d =0.25 mm w p p =100 MPa (glass) w 20 d =0.029 mm d =0.25; d =0.9 mm p ori f 20 v=1.67 mm/s; m =5 g/s a 10 garnet Maximal depth of cut h
Maximal depth of cut h 15 0 0 100 200 300 400 500 600 0 5 10 15 20 Abrasive particle diameter d (g/s) p Abrasive mass flow rate (g/s) a) [Ohls97] b) [Owei89]
Figure 2.8: Abrasive particle diameter versus maximum depth of cut
15
2.1.5 Optimum standoff distance
The effect of the standoff distance on the maximum depth of cut was first investigated by R.E. Barton [Bart82]. It was found that the depth of cut decreases almost linearly with the increase of the standoff distance. Figure 2.9 shows the relationship between the standoff distance and the maximum depth of cut [Blic90]. R.A Tikhomirov et al. [Tikh92] reported the same result for the relation between the standoff distance and the maximum feed speed. The authors noted that at a small increase of the standoff distance, the maximum feed speed first remained constant and then decreased according to an almost linear relation [Tikh92]. The effect of the standoff distance on the depth of cut was also confirmed by Blickwedel [Blic90], Kovacevic [Kova92] and Guo et al. [Guo94b]. In addition, Guo et al. suggested that the optimum standoff distance is about 2 mm [Guo94b].
45 d =0.25; d =1.2 mm ori f 40 l =50 mm; v =1.67 mm/s f f m =8 g/s (mm) 35 a max
30
25
20
15 p =300 MPa 10 w p =200 MPa
Maximal depth of cut h w 5 0 10 20 30 40 50 Standoff distance (mm)
Figure 2.9: Standoff distance versus maximum depth of cut [Blic90]
2.2 State of the art in AWJ cost calculation and cost optimization
2.2.1 State of the art in AWJ cost calculation
• Study of J. Zeng and T. J. Kim:
To calculate the cutting cost per length, J. Zeng and T. J. Kim [Zeng93] first introduced a model for prediction of the feed speed:
1.15 ⎛⎞Np⋅⋅1.25 m 0.687 ⋅ m 0.343 v = mw w a (2.4) f ⎜⎟0.618 ⎝⎠Cqhd⋅⋅⋅f
16 The cutting cost per length was then determined by the following equation:
C h C l = (2.5) v f
Where, Ch is the total hourly cost ($/h), which is calculated as follows:
CChmhlhthphdh=++++ CCC C (2.6)
In which Cmth is the machine hourly cost, Clh is the labor hourly cost, Cth is the material hourly cost which considers the abrasive cost, water cost, focusing tube cost and orifice cost, Cph is the power hourly cost, and Cdh is the cost of maintenance and disposal.
In this study, many cost components were taken into account. Additionally, the effects of many jet- parameters on the cutting cost were also investigated through a model for prediction of the feed speed. After all, the effect of the number of jet formers as well as the effect of the nozzle wear on the cutting cost was still not well-understood.
• Study of D.A. summers et al.:
To compare the AWJ cutting cost per part in both cases with and without abrasive recycling, D.A. Summers et al. [Summ01] carried out a study in which the influence of many cost parameters on the total cutting cost, e.g. the abrasive cost, the disposal cost, the power cost, the water cost and the nozzle wear cost were investigated. Also, the optimum cutting performance was predicted by a tabulated method. The authors concluded that by cutting with the recycled abrasives using particles larger than 100 µm, the cutting cost can be reduced significantly. Nevertheless, the effects of several cost components such as the machine cost, the labor cost and the maintenance cost were not considered. Besides, although the nozzle wear cost was taken into account empirically, there is still no model for calculation of the nozzle wear.
• Study of M. Hashish:
M. Hashish [Hash04] compared the cutting cost in two cases: with water pressure of 400 MPa and 600 MPa. The effects of the water pressure on different cost elements such as abrasives, pump and machine maintenance, water, power and the nozzle wear were studied.
Hashish noted that the feed speeds when cutting at 600 MPa should be at least equal to those at 400 MPa while using 33% less abrasives and water. The author also found that cutting at a pressure of 600 MPa can save the total cutting cost 10 % to 25 % over that at 400 MPa [Hash04].
Although the study compared the cutting cost when cutting with high and low pressure, the effect of water pressure on the AWJ system’s utilization and the nozzle wear were not investigated. In practice, cutting with high pressure can increase not only the pump maintenance cost but also the downtime due to the pump’s maintenance. As a result, the total available cutting time when cutting with high pressure will be reduced and thus the cutting cost will increase accordingly. In addition,
17 due to the high pressure, the nozzle wear also increases (in the study the nozzle wear was constant) and leads to an increase of both the nozzle wear cost and the downtime because of replacement of the nozzle. These effects of high pressure on the utilization and on the nozzle wear cost should therefore be taken into account.
2.2.2 State of the art in AWJ cost optimization
• Study of P. J. Singh and J. Munoz:
P.J. Singh and J. Munoz [Sing93] noted that the AWJ cost optimization problem is very complicated to solve because there are a lot of parameters affecting the total cutting cost. However, local sub- optimization can be used as a solution for the problem. For cost analysis, the authors divided the cost elements into three main components: the operating costs, the labor costs Cl, and the capital investment costs Ce [Sing93]. The operating costs consist of the abrasive cost Ca, the power cost
Cp, the water cost Cw, the focusing tube cost Cf, orifice cost Cori, and maintenance cost Cmai. The total cutting cost per centimeter is then determined as follows [Sing93]:
10 ⋅C h C lc, = (2.7) 60 ⋅⋅vkfcf
Where, vf is the feed speed (mm/s) calculated by the model by Zeng and Kim (Equation 2.4); kcf is a contour factor which considers the necessary slow down of the system during turns; Ch is the total cutting cost per hour which is calculated by the following equation [Sing93]:
CCCCCCChapwforimaile=++++ + ++ CC (2.8)
A sub-optimization problem is performed by considering the orifice diameter as an independent variable. Other parameters are then chosen based on this variable [Sing93]. From the results of the optimization problem, the authors concluded that use of smaller orifices is more cutting efficient, i.e. the cutting length per unit of power is higher. However, larger orifices are more cost efficient as the feed speed can be increased so that the labor cost and the capital cost are reduced [Sing93]. In addition, the authors found that cutting with multiple-heads reduces the total cutting cost since the combination of the higher efficiency of smaller orifices with higher throughput of multiple-heads [Sing93]. It is noted that careful cost analysis and cost optimization can save 10 to 30% of the total AWJ cutting cost [Sing93].
As the above sub-optimization study was carried out by considering only one variable, the applications are therefore limited. The effect of process parameters as well as the effect of cost elements on the total cutting cost should be taken into the sub-optimization problem.
• Study of J. Zeng and J. Munoz:
18 J. Zeng and J. Munoz [Zeng94] presented a study on optimum selection of the abrasive mass flow rate in order to fulfill the minimum cutting cost. In this study, the total cutting cost per length
Cl ($/m) is calculated according to the approach of Zeng and Kim [Zeng93]: