USF Libraries
USF Digital Collections

Analysis of glass mold to enhance rate of heat transfer

MISSING IMAGE

Material Information

Title:
Analysis of glass mold to enhance rate of heat transfer
Physical Description:
Book
Language:
English
Creator:
Warude, Anand
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Molding
Cooling passages
Glass bottle
Dissertations, Academic -- Mechanical Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Narrow Neck Press and Blow (NNPB) process is used to produce light weight bottles. The gob of molten glass is delivered to the blank mold and a specially designed narrow diameter plunger is used to form the finish or mouth and the parison as it presses upwards. Invert and final blow takes place followed by take-out and annealing. Anchor Glass Container Corp. (AGC) uses NNPB technology in their glass making plants. The problem experienced by AGC in the process is that the heat dissipation through out the mold is not uniform and hence there is a non uniform temperature distribution in the finished bottle extracted from it. Specifically the shoulder region of the bottle stays at a higher temperature when compared with the other regions, becoming the limiting factor in determining the rate of bottle production.Excessive temperatures in any region leave the glass insufficiently rigid, allowing the bottle to sag or lean. An increased rate of production which demands faster and effective cooling of the bottle is desired and is the ultimate goal of this research effort. This problem can be effectively solved by increasing the amount of heat transferred from the mold to the cooling air, which can be done by increasing the surface area of the cooling passages. A mathematical model for calculating the amount of heat transferred to the cooling air is proposed in this thesis. The air properties at the exit of the mold and the amount of heat transferred by each cooling passage were obtained by using MATHCAD. A 2 dimensional numerical simulation for the final molding was carried out using ANSYS and the temperature distribution for the mold and glass were obtained.
Thesis:
Thesis (MSME)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Anand Warude.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 116 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681127
oclc - 62792166
usfldc doi - E14-SFE0000618
usfldc handle - e14.618
System ID:
SFS0025308:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681127
003 fts
005 20060215071259.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 060106s2004 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000618
035
(OCoLC)62792166
SFE0000618
040
FHM
c FHM
049
FHMM
090
TJ145 (Online)
1 100
Warude, Anand.
0 245
Analysis of glass mold to enhance rate of heat transfer
h [electronic resource] /
by Anand Warude.
260
[Tampa, Fla.] :
b University of South Florida,
2004.
502
Thesis (MSME)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 116 pages.
520
ABSTRACT: Narrow Neck Press and Blow (NNPB) process is used to produce light weight bottles. The gob of molten glass is delivered to the blank mold and a specially designed narrow diameter plunger is used to form the finish or mouth and the parison as it presses upwards. Invert and final blow takes place followed by take-out and annealing. Anchor Glass Container Corp. (AGC) uses NNPB technology in their glass making plants. The problem experienced by AGC in the process is that the heat dissipation through out the mold is not uniform and hence there is a non uniform temperature distribution in the finished bottle extracted from it. Specifically the shoulder region of the bottle stays at a higher temperature when compared with the other regions, becoming the limiting factor in determining the rate of bottle production.Excessive temperatures in any region leave the glass insufficiently rigid, allowing the bottle to sag or lean. An increased rate of production which demands faster and effective cooling of the bottle is desired and is the ultimate goal of this research effort. This problem can be effectively solved by increasing the amount of heat transferred from the mold to the cooling air, which can be done by increasing the surface area of the cooling passages. A mathematical model for calculating the amount of heat transferred to the cooling air is proposed in this thesis. The air properties at the exit of the mold and the amount of heat transferred by each cooling passage were obtained by using MATHCAD. A 2 dimensional numerical simulation for the final molding was carried out using ANSYS and the temperature distribution for the mold and glass were obtained.
590
Adviser: Crane, Roger.
653
Molding.
Cooling passages.
Glass bottle.
690
Dissertations, Academic
z USF
x Mechanical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.618



PAGE 1

Analysis of Glass Mold to E nhance Rate of Heat Transfer by Anand Warude A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Mechanical Engineering Department of Mechanical Engineering College of Engineering University of South Florida Major Professor: R. A. Crane, Ph.D. Thomas Eason, Ph.D. Glenn Besterfield, Ph.D. S. C. Kranc, Ph.D. Date of Approval: July 2, 2004 Keywords: Glass Bottle, Cooling passages, Molding Copyright 2004, Anand Warude

PAGE 2

DEDICATION To my family

PAGE 3

ACKNOWLEDGEMENTS I wish to express my gratitude to my ma jor professor, Dr. Roger A. Crane, for his continuous support, and guidance throughout my master’s research and thesis preparation. I would also like to express my appreci ation to my committee members, Dr. Glen Besterfiled, Dr. Thomas Eason and Dr Stanley Kranc for their comments and suggestions. Acknowledgement is made to Anchor Gl ass Container Corporation for supporting this research in part. This thesis would not have been possible without the love, encouragement and support of my parents.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v ABSTRACT ix CHAPTER 1 INRODUCTION 1 1.1 Basic Physics 1 1.2 Glass Forming Operations 2 1.3 History of Glass Molding 4 1.4 Objective 5 CHAPTER 2 LITERATURE REVIEW 7 2.1 Patents Relating to the Air Cooling of Molds 7 2.1.1 Apparatus for Uniform Cooling of Glass Mo lding Machines 7 2.1.2 Method for Blow Molding and Cooling Ho llow Glassware 8 2.1.3 Mold Cooling Arrangement for Use in Glassware Forming Machine 9 2.1.4 Forming Machine 11 2.1.5 Method of Cooling a Mold 12 2.1.6 Mold Cooling Arrangement for a Glassware Forming Machine 13 2.1.7 Mold Portion with Cooling M eans for Use in Molding Molten Glass 14 2.1.8 System and Method for the Cooling of Hot Molds 14 2.1.9 Mold Assembly for the Glass Articles 15 2.1.10 Cooling Arrangement for a Mo ld of a Glassware Forming Machine of the Individual Section Type 17 2.1.11 Method of Cooling a Mold 17 2.1.12 Mold Cooling Apparatus for a Glassware Forming Machine 18 2.1.13 Apparatus and Method for Cooling a Mold 19 2.2 Liquid Cooling of Molds 20 2.2.1 Fluid Cooling of Glass Molds 21 2.2.2 Mold with Exterior Heat Conducting Elements 22 2.2.3 Blow Mold Cooling 23 2.2.4 Cooling Molds Used in Forming Glassware Containers 24

PAGE 5

ii 2.2.5 Mold Cooling 25 2.2.6 Method and Apparatus for Mold Cooling 26 2.3 Liquid Metal Cooling 28 2.3.1 Molding Cooling System for th e Manufacture of Glass Articles or Similar Materials 28 2.3.2 Glass Forming Mold 29 2.4 Controlling Mold Temperature 30 2.4.1 Controlling the Temperature of a Glass Mold 30 2.4.2 Controlling the Temperature of a Mold 31 2.4.3 Mold Arrangement for Cyclic ally Operating Glassware Container Manufacturing Machine with Temperature Sensing Means 32 2.5 Dead Pan Cooling 33 2.5.1 Cooling Articles of Newly Molded Glassware 33 2.6 Neck Mold Assembly Cooling 35 2.6.1 Glassware Molding Machine with Unita ry Axis Molding 35 2.6.2 Glass Container Forming Machine Including Neck Ring Mold Cooling 36 2.7 Interior Cooling in the Finishing Mold 36 2.7.1 Apparatus for Forming Glass Articles with Treating Mean 36 CHAPTER 3 NARROW NECK PRESS AND BL OW PROCESS 38 3.1 Gob Delivery 38 3.2 Start of Press 39 3.3 Press Time 40 3.4 Plunger Down 40 3.5 Reheat 41 3.6 Reheat and Run 42 3.7 Final Blow and Vacuum Forming 43 3.8 Mold Open and Takeout 43 3.9 Dead Plate Time 44 CHAPTER 4 EXPERIMENTAL ANALY SIS OF AIR FLOW THROUGH COOLING HOLES OF THE MOLD 45 4.1 Experimental Setup and Procedure 46 4.2 Observations 48 4.3 Calculations 49 4.4 Results and Conclusions 52 CHAPTER 5 ONE DIMENSIONAL STEADY STATE ANALYS IS 53 5.1 Introduction 53 5.2 Steady State Analysis 54

PAGE 6

iii 5.2.1 Compressible Flow and Heat Transfer Modeling 55 5.2.2 Fin Analysis 58 5.2.3 Calculations for Fin Efficiency 59 5.2.4 Results of Fin Analysis 61 5.3 Parametric Analysis 62 5.4 Axial Distribution of Cooling th rough Constant Temperature Cooling Passage 64 5.5 Conclusions 65 CHAPTER 6 ANALYTICAL TRANSIENT ANALYSIS 66 6.1 Formulation of Semi-Infinite Solid Problem 66 6.2 Calculations 67 6.3 Plotting Transient Temperature Di stribution Profiles for Glass and Mold 69 6.4 Surface Heat Flux Calculations 71 6.5 Conclusions 72 CHAPTER 7 MATHEMATICAL MODELING OF THE PROBLEM 73 7.1 Governing Equations and Modeling 74 7.2 Finite Element Models 77 CHAPTER 8 RESULTS OF NUMERICAL SO LUTION 79 8.1 Problem Definition 80 8.2 Results of the Numerical Simulation 81 8.3 Temperature Distribution Data 85 8.4 Conclusions 85 CHAPTER 9 EXPERIMENTAL RESULTS 87 9.1 Measurement of Bottle Temperatures 87 9.2 Bottle Lean Data Analysis 89 9.3 Mold Temperature Measurements 95 9.4 Conclusions 98 CHAPTER 10 CONCLUSIONS AND RECOMMENDATIONS 98 10.1 Conclusions 98 10.2 Recommendations 98 REFERENCES 101

PAGE 7

iv LIST OF TABLES Table 4.1 Flow Meter Reading for Each Hole 49 Table 4.2 Flow Velocity through Each Hole 50 Table 5.1 Results of Fin Analysis 61 Table 5.2 Results of Increasing Ai r Coolant Passage Diameter in Finned and Un-Finned Passages 63 Table 5.3 Effects of Varia tions in Body Temperature on Existing and Prototype Molds 63 Table 5.4 Relative Cooling at Various Mo ld Sections 64 Table 6.1 Variation in the Temperat ure Distribution Depending on the Location and Time for the Glass 69 Table 6.2 Variation in the Temperat ure Distribution Depending on the Location and Time for the Mold 70 Table 6.3 Surface Heat Flux Variation with Time 71 Table 8.1 Comparison of Temperatur es between the Anchor Mold and USF Mold 85 Table 8.2 Comparison of Temperatures between Bottles Extracted from the Anchor Mold and USF Mold 85

PAGE 8

v LIST OF FIGURES Figure 1.1 Viscosity-Temperature Curve for Silicate Glass 2 Figure 2.1 Apparatus for Uniform Cooling of Glass Molding Machines 8 Figure 2.2 Method for Blow Molding and Cooling Hollow Glassware 9 Figure 2.3 Mold Cooling Arrangeme nt for Use in Glassware Forming Machine 10 Figure 2.4 Mold Cooling Arrange ment for Use in Forming Machine 11 Figure 2.5 Method of Cooling a Mold 12 Figure 2.6 Mold Cooling Arrangement for a Gla ssware Forming Machine 13 Figure 2.7 Mold Portion with Cooling Means for Us e in Molding Molten Glass 14 Figure 2.8 Systems for the Cooling of Hot Molds 15 Figure 2.9 Mold Assembly for the Glass Articles 16 Figure 2.10 Method of Cooling a Mold 18 Figure 2.11 Mold Cooling Apparatus for a Glassw are Forming Machine 19 Figure 2.12 Apparatus and Method for Cooling a Mold 20 Figure 2.13 Fluid Cooling of Glass Molds 21 Figure 2.14 Mold with Exterior Heat Conduc ting Elements 23 Figure 2.15 Blow Mold Cooling 24 Figure 2.16 Cooling Molds Used in Forming Gla ssware Containers 25 Figure 2.17 Arrangements for Mold Cooling 26

PAGE 9

vi Figure 2.18 Method and Apparatus for Mold Cooling 28 Figure 2.19 Molding Cooling System for the Manuf acture of Glass Articles 28 Figure 2.20 Glass Forming Mold 29 Figure 2.21 Controlling the Temperature of a Glass Mold 31 Figure 2.22 Controlling the Temperature of a Mold 32 Figure 2.23 Mold Arrangement for a Cyc lically Operating Gl assware Container Manufacturing Machine with Temperature Sensing Means 33 Figure 2.24 Cooling Articles of Newly Mold ed Glassware 34 Figure 2.25 Glassware Molding Machine with Unita ry Axis Molding 35 Figure 2.26 Apparatus for Forming Glass Articles with Treating Mean 37 Figure 3.1 Gob Delivery 39 Figure 3.2 Start of Press 39 Figure 3.3 Press Time 40 Figure 3.4 Plunger Down 41 Figure 3.5 Reheat 42 Figure 3.6 Reheat and Run 42 Figure 3.7 Final Blows and Vacuum Forming 43 Figure 3.8 Mold Open and Takeout 44 Figure 3.9 Dead Plate Time 44 Figure 4.1 Mold with Fins Cu t through it Exposing the Outer Row Holes 45 Figure 4.2 Experimental Setup to Analyze Air Fl ow through Cooling Holes in the Mold 46 Figure 4.3 Cross section of the Mold Showing I nner and Outer Row Holes 48

PAGE 10

vii Figure 4.4 Air Velocity (fps) Vs Hole Number in the Mold (Inlet Pressure of 20Psi) 51 Figure 4.5 Air Velocity (fps) Vs Hole Number in the Mold (Inlet Pressure of 20Psi) 51 Figure 5.1 One dimensional MATHCAD model 54 Figure 5.2 Schematic Representation of the Fin in side the Air Cooling Channel 59 Figure 5.3 Varying Heat Transfer Rate w ith Fin Thickness 62 Figure 5.4 Axial Distribution of Cooling through Constant Temper ature Passage 65 Figure 6.1 Semi Infinite Solid with Consta nt Wall Temperature 66 Figure 6.2 Transient Temperat ure Distribution in Semi Infinite Solid (Glass) 70 Figure 6.3 Transient Temperat ure Distribution in Semi Infinite Solid (Mold) 71 Figure 6.4 Plot of Surface Heat Flux with Time 72 Figure 7.1 Two Dimensional Model of the Mold 74 Figure 7.2 Two Dimensional Finite Element Model of the Mold 78 Figure 8.1 Mold Cross Section at the Shoulder 80 Figure 8.2 Temperature Distributio n of Glass at the Neck Section (Degree F) 81 Figure 8.3 Temperature Distribution of Mold at the Neck Section (Degree F) 81 Figure 8.4 Temperature Distribution of Glass at the Shoulder Cross Section (Degree F) 82 Figure 8.5 Temperature Distribution of Mold at the Shoulder Cross Section (Degree F) 82 Figure 8.6 Temperature Distributio n of Glass at the Neck Section (Degree F) 83 Figure 8.7 Temperature Distribution of Mold at the N eck Section (Degree F) 83 Figure 8.8 Temperature Distribution of Glass at the Shoulder Cross Section (Degree F) 84

PAGE 11

viii Figure 8.9 Temperature Distribution of Mold at the Shoulder Cross Section (Degree F) 84 Figure 9.1 Bottle Shoulder Temperatures Meas ured at Dead Plate 88 Figure 9.2 Bottle Neck Temperatures Measured at the Dead Plate 89 Figure 9.3 Measured Lean on Control Sample, 30-195o Air Circulation 90 Figure 9.4 Measured Lean on Control Sample, 60-175o Air Circulation 91 Figure 9.5 Measured Lean on Prototype Mold, 50-195o Air Circulation 92 Figure 9.6 Measured Lean on Prototype Mold, 70-195o Air Circulation 92 Figure 9.7 Measured Lean on Prototype Mold, 60-175o Air Circulation 93 Figure 9.8 Measured Lean on Prototype Mold, 60-175o Air Circulation 93 Figure 9.9 Measured Lean on Prototype Mold, 70-175o Air Circulation 94 Figure 9.10 Standard Mold Thermal Image during Tests 95 Figure 9.11 Prototype Mold Surface Image during Tests 96 Figure 9.12 Standard Mold Surface Image during Tests. 97 Figure 9.13 Prototype Mold Surface Image during Tests 97

PAGE 12

ix ANALYSIS OF GLASS MOLD TO EN HANCE RATE OF HEAT TRANSFER Anand Warude ABSTRACT Narrow Neck Press and Blow (NNPB) process is used to produce light weight bottles. The gob of molten glass is delivered to the blank mold and a specially designed narrow diameter plunger is used to form the fi nish or mouth and the parison as it presses upwards. Invert and final blow takes place followe d by take-out and annealing. Anchor Glass Container Corp. (AGC) uses NNPB technology in their glass making plants. The problem experienced by AGC in the process is that the he at dissipation through out the mold is not uniform and hence there is a non uniform temperature distribution in the finished bottle extracted from it. Specifically the shoulder region of the bottle stays at a higher temperature when compared with the other regions, becoming the limiting factor in determining the rate of bottle production. Excessive temperatures in any region leave the glass insufficiently rigid, allo wing the bottle to sag or lean. An increased rate of production which demands faster and effective c ooling of the bottle is desired and is the ultimate goal of this research effort. This problem can be effectively solv ed by increasing the amount of heat transferred from the mold to the cooling ai r, which can be done by increasing the surface area of the cooling passages. A mathemati cal model for calculating the amount of heat

PAGE 13

x transferred to the cooling air is proposed in this thesis. The air properties at the exit of the mold and the amount of heat transferred by each cooling pa ssage were obtained by using MATHCAD. A 2 dimensional numerical simula tion for the final molding was carried out using ANSYS and the temperature distributio n for the mold and glass were obtained. Results obtained from the above simulations ar e compared with the th ermal images of the bottle taken during the molding operation at AGC, Jacksonville.

PAGE 14

1 CHAPTER 1 INTRODUCTION 1.1 Basic Physics Molten glasses are Newtonian liquids fo r which rate of flow is directly proportional to applied stress. Quite small st resses or pressures can be used in glass making, so the apparatus used does not need to be massive. All the main methods of making glass rely on the very rapid variatio n of viscosity with temperature. Silicate glasses are very viscous compared with most familiar liquids and th eir rate of flow under applied stress increases rapidly as the temp erature rises. Glass manufacturing operations depend on adjusting the viscosity (by selecting the right temperatures) so that small stresses are sufficient for shaping the glass a nd subsequent cooling so that it becomes too viscous to flow under gravity once it has atta ined the desired form. The most important properties of glass thus are viscosity, densit y, thermal conductivity and specific heat. The glass forming operations take pl ace at high temperatures at which thermal radiation plays an important role in heat tr ansfer. At maximum melting temp erature the viscosity is about log = 2 and for stress release duri ng annealing it is around log = 14. Viscosity varies greatly with glass composition.

PAGE 15

2 1.2 Glass Forming Operations The whole cycle of glass forming can be conveniently divided into 3 stages, gathering the glass from the furnace, formi ng (pressing or blowing), and annealing which means slow cooling of glass to leave very l ittle internal stress at room temperature. Figure 1.1 Viscosity Temperature Curve for Silicate Glass Figure 1.1 shows the viscosity-temperature curve for silicate glass showing the ranges used at different stages of glass forming. Adjusting either temperature or the pressures used can vary rates of flow but the rates of heat transfer are not so easy to accelerate and heat transfer often controls the maximum rate of production. Conditions at the glass-mold interface are crucial. Temperatur e control is very important b ecause, if the glass is too hot it flows easily under gravity and cannot be contro lled and if the mold is too hot, even if the glass is at the right temperature, the tw o stick together. The glass flows to the shape

PAGE 16

3 of the mold when the mold and the glass condi tions are at the requir ed temperature. The glass and the mold may briefly stick to each ot her, however the glass is cooled somewhat and contracts whilst the mold is heated a nd expands. The stresses produced are sufficient to break the weak adhesion and separate th e glass and the mold. Therefore there is a limited range of temperature for both glass and mold within which good results can be obtained, which means that the heat transfer coefficient at the inner surface of the mold varies considerably with the time of contact. Molds are usually made in halves, split ve rtically, as they n eed to open and close and are mostly made from special grades of cas t iron or brass but other alloys are used for some components. Molds have to be desi gned and operated to keep the mold temperatures in the correct range (balancing en ergy loss with heat gain) and to give up as much energy while empty as they gain fr om the glass while th e container is being formed. The molding process will generally occur in two stages. In the first stage, a blob of glass is formed to produce a finished mout h and an interior cavit y is preformed; this crude bottle is called a parison. Contact of th e glass with walls of the parison mold chills the surface of the glass as th e parison is being shaped. Af ter removing the parison from this mold, heat continues to flow from the bulk of the glass toward the surface, reheating this surface to allow further processing. Very steep temperature gradients occur near the surfaces of the glass while in contact with the mold. When shaping of the parison is finished the mold halves are open just a fr action to create a small air gap between the glass and the mold which causes a great decrease in the rate of heat transfer from the glass, as a result the chilled surface layer regains temperature and its viscosity decreases so that the parison is more easily blown in the second mold. An appreciable proportion of

PAGE 17

4 heat removed from the glass is lost from th e inner surface of the mold by cooling it with compressed air while empty, but conduction through the wall and loss from the outer surface also plays an important role. In this study, we are examining methods of impr oving the energy removal process so as to permit shorter processing times and to speed production. 1.3 History of Glass Molding In 1859 the first vertically split iron mold with movable base plate and means for blowing compressed air to make bottles wa s designed by Mein of Glasgow. The most important change to the manual pr ocess was to make the mouth or finish first, so that the glass could be held in a neck ring while the rest of the opera tions were performed. In 1866 H.M.Ashley and Josiah Arnall proposed to use an inverted mold with a plug at the bottom to make the mouth of the container an d a sliding base plat e to press the glass down into the neckring to form the mouth. The base plate was retracte d to the top of the mold and compressed air was used to blow the top of the bottle. In 1887 Ashley came up with two crucial improvements, the provision of a separate neckring mold to hold glass during other manipulations and use of a separate parison mold which gave better distribution of wall thickness. A simple machine with one pair of mold operating on the above principl e claimed that the labor cost was reduced considerably. The machine needed to be fed by hand with gobs of glass. Achieving an almost constant weight of the glass and thus internal capacity Hollow glass articles such as bottles and jars, when molded by a forming machine individual section (“I.S.”) type using Narrow Neck Press and Blow (NNPB) process are

PAGE 18

5 molded in two steps. In the fi rst step, perform of the finished container, which is usually called a blank or a parison, is molded by an annular mold made up of a pair of mating blank mold sections. Upon the completion of the blank molding step, the blank molding step, the blank molding section separate and the parison is transferred to another mold station, often called blow mold station, where it is blown into its final shape by another mating pair of mold sections. During the mo lding operation a great deal of heat is transferred from the semi-molten glass to the mo lds, and it is necessary to cool the molds in a predetermined and controlled manner in order to ensure the consistent molding conditions which result in a glass container wi th uniform wall thickness and a sufficiently low heat content to enable it to stand afte r leaving the mold. At the conclusion of the blow mold process, the mating sections of the blow mold are separated, and the container is removed from the forming m achine for further processing. 1.4 Objective Anchor Glass Container Corp. uses the same technology discussed above in their glass making plant. The problem experienced by the Anchor Glass Company is that the heat dissipation throughout the mold is not uniform resulting in a non uniform temperature distribution in the bottle extracted from it. As a result the shoulder region of the bottle stays at a higher te mperature when compared with the other regions, resulting in a lean which is undesirable. Lean can be defined as the offset between a good bottle and a bottle which tilts at some cross-secti on. Also an increased rate of production which demands faster and effective co oling of the bo ttle is desired.

PAGE 19

6 This problem can be effectively solv ed by increasing the amount of heat transferred from the mold to the cooling ai r. The heat transfer rate is increased by increasing the surface area of the cooling pa ssages. A mathematical model for calculating the amount of heat transferred to the coolin g air is proposed in this thesis. The air properties at the exit of the mold and the amount of heat tran sferred by each cooling passage were obtained by using MATHCAD. Numerical simulation for the critical sections of the glass mold was carried out using ANSYS. The temp erature distribution trough out the mold and glass were obtained in this simulation. Resu lts obtained from the above simulations are compared with the th ermal images of the bottle taken during the tests conducted at Anchor Gla ss Corporation, Jacksonville.

PAGE 20

7 CHAPTER 2 LITERATURE REVIEW As a part of this study all recent pate nts related to glass mold cooling were reviewed to determine the particular design aspect that is being claimed. To a large extent, the patents selected were those ini tially identified by Anchor Glass. Where the patents, themselves, have referred to earlier patents deemed relevant are included in the overall survey. The patents were reviewed and categorized to classify the particular aspect of mold cooling by topic, summari zing the ideas in the paragraphs below. 2.1 Patents Relating to the Air Cooling of Molds 2.1.1 Apparatus for Uniform Coolin g of Glass-Molding Machines This invention, covering the air cooling of the neck ring and parison mold was filed in 1967 by A.E. Kurtz [1]. The figure to the right refers to that portion of the patent relating to cooling the parison. It is shown in th e upper portion of the figure as item 26. The power cylinder, item 32, is used to driv e a punch into the blank forming the blow hole. In this design air is introduced into the chamber, 22. The chamber is sealed from the cooling passages around the pa rison by a seal, 50. The seal is attached to a cylinder, 52.

PAGE 21

8 Figure 2.1 Apparatus for Uniform Cooli ng of Glass Molding Machines When the cylinder is rotated, the passages are opened allowing cooling air to pass uniformly around the outside circumference of the parison. The design is claimed to uniform and constant cooling of the mold. 2.1.2 Method for Blow Molding and Cooling Hollow Glassware This patent relating to cooling hollo w glassware was filed in 1979 by John K Martin and assigned to Vitro Tec Fedeicomiso [2]. This patent included a very brief explanation of the concept. Essentially the inverter has patented the idea of using a

PAGE 22

9 partial seal about the t op of the bottle so that blowing ai r simultaneously shapes the bottle and the continuous flow provides fo r continuous interior cooling. Figure 2.2 Method for Blow Moldi ng and Cooling Hollow Glassware 2.1.3 Mold Cooling Arrangement for Use in Glassware Forming Machine This patent, relating to a parison mold cooling arrangement for use in glassware forming machine was filed in 1986 [3] by Co nstantine W. Kulig and was assigned to Emhart Industries Inc. One feature of this invention is that air cooling passages come

PAGE 23

10 into registry when the mold is opened. The patent also covers a design for directing air flow to the neck ring mold Figure 2.3 Mold Cooling Arrangement fo r Use in Glassware Forming Machine (a) (b) (c)

PAGE 24

11 2.1.4 Forming Machine This patent, relating to cooling device fo r glass container forming machine, was filed in 1990 by Roger Erb and Ro bert Johnson [4]. It was a ssigned to Emhart Industries Inc. The invention comprises an arrangement to air cool the molds for a glass parison. Cooling flows to a plenum below the bolds and through axial cool ing passages in the mold when the air passages align themselv es with the plenum openings. The cooling passage extends through the axial dimension of the mold and thus provides cooling of separate neck finish ring. Figure 2.4 Mold Cooling Arrangement for Use in Forming Machine

PAGE 25

12 2.1.5 Method of Cooling a Mold This patent was filed in 1982 by Thomas V. Foster and assigned to Emhart Industries [5]. This invention is concerned wi th a method of cooling a mold in glassware forming machine where at least one porti on of the mold is supported on a movable support which opens and closes the mold. In th is invention an intermediate support is irremovably mounted on the mold support, which provides access to the spaces that insulate the mold from the mold support and pr ovides access to the space to blow cooling air, which impinges on mold portion through aper tures in intermediate support. Increased cooling can be obtained by blowing cooling ai r between fins on outer surface of the mold portion. A thermocouple embedded on mold potion also helps achieving better control of cooling airflow. Figure 2.5 Method of Cooling a Mold

PAGE 26

13 2.1.6 Mold Cooling Arrangement fo r a Glassware Forming Machine This patent, covering the design of an air cooled glassware forming machine (parison mold), was filed in 1984 by Thomas V. Foster and assigned to Emhart Industries [6]. The concept appears to be a variati on of certain of the other Emhart patents describing vertically orientated, radially a rrayed air passages arranged radially about the outer perimeter of the mold. The arrangem ent comprises of mold cavity having an upwardly facing opening. Cooling passages ex tending vertically in the mold body and opening on the upper surface, cooling air passes through these opening and cools the mold body. Air is delivered to cooling passage s by an annular chamber that is operative when mold closes so that, in this case, ai r flows downward. Provisions are made for water entrainment into the cooli ng air to enhance cooling. Figure 2.6 Mold Cooling Arrangement for a Glassware Forming Machine

PAGE 27

14 2.1.7 Mold Portion with Cooling Means for Us e in Molding Molten Glass Thomas V. Foster filed this patent in 1986 and again assi gned it to Emhart Industries [7]. This describes an invention to enhance cooling in spec ific hot spots inside the mold. A rod, of a material having higher thermal conductivity than that of the mold, is inserted into a hole projected toward th e mold cavity. This rod extends from region requiring higher heat extraction into a recess in an outer surface of the mold. An air passage enters this recess and air flow is such that it swirls around the rod, providing efficient cooling. Figure 2.7 Mold Portion with Cooling M eans for Use in Molding Molten Glass 2.1.8 System and Method for the Cooling of Hot Molds Rolando Cantu-Garcia filed this patent on 19 Jun 1986, assigning the invention to the company Vidriera Monterrey [8]. The in vention provides for enhanced heat transfer for the conventional vertical air coolant passages. This is achieved by means of a set of

PAGE 28

15 nozzles, located at the passage inlet, which impart a swirling component to the air flow pattern. This is an effort to implement one of several possible e nhanced heat transfer solutions to achieve more effective cooling within a given passage space. There are other more effective, means of enhancing heat tran sfer which would not fa ll within this patent. Figure 2.8 System for the Cooling of Hot Molds 2.1.9 Mold Assembly for the Glass Articles This patent, relating to mold assembly fo r glass articles was filed in 1997 by Dan Haynes and assigned to Owens-Brickway Glass Container Inc. [9]. The patent would clearly appear to be intended for air cooli ng, but the applicant has taken the precaution of explicitly claiming that the idea could be used with any fluid. What is unique is that the intended air flow path is directly radially inward against the center section of the mold

PAGE 29

16 and allowed to exit by splitting the flow path both upward and downward. The surface of the mold has finned ribs situated on the exteri or surface to further enhance heat transport. Distribution control of the cooling fluid is ac hieved by applying a split perforated screen to the interior of each section of the mold holder. Th e fabrication costs of molds are claimed to be reduced in comparison with t hose molds having coolant passages contained in molds themselves. This claim is due to the cooling system being on the mold holder itself and hence can be used with a variety of molds, reducing mold structure complexity. Figure 2.9 Mold Assembly for the Glass Articles

PAGE 30

17 2.1.10 Cooling Arrangement for a Mold of a Gla ssware Forming Machine of the Individual Section Type This patent, relating to the design of an inlet plenum to distribute uniform coolant flow rates to air cooled molds, was filed in 1985 by Stanley Peter Jones and assigned to Emhart Industries Inc. [10]. The arrangement includes an air blowing means, a ducting arrangement to conduct cooling air to a pl enum chamber and the chamber itself. The plenum chamber communicates with the entrances of the air cooling passages in the mold halves. It is claimed that this device opera tes with an air pressure of between 1100-200 mm of water at the entrance of cooling passages and that this value is substantially lower than the then current designs. 2.1.11 Method of Cooling a Mold This patent was filed in Canada and wa s accessed through the Canadian Intellectual Property Office. Unfortunatel y, this source does not list the inventor or the date that the patent was issued. This patent desc ribes a method of air cooling a blank mold used for making a parison. The primary element of the design is the baffle, i.e. the top portion of the mold contacting what would be the base of the inverted parison. When the baffle is in position on the mold; air is blown through the passages in the baffle into the passage in the mold to cool the mold. This method of cooling the mold blank can be applied whether the parison is formed by blowing or pressing.

PAGE 31

18 Figure 2.10 Method of Cooling a Mold 2.1.12 Mold Cooling Apparatus for a Glassware Forming Machine The above invention, relating to mold c ooling apparatus for a glassware forming machine was filed in 1994 by Richard T Kirk man and assigned to Wens Brockway Glass Container Inc. [11]. The invention relates to air cooling of single or multiple cavity molds and addresses both the plenum arrangement a nd the mold coolant passages. In this arrangement cooling air is di rected from the plenum, thr ough a diffuser plate and against the mold halves. The diffuser plate will cover th e greater part of the side of the mold so that the air flow pattern is primarily inward ra dially. Venting of the cooling air is through an exhaust port at the top of the mold.

PAGE 32

19 Figure 2.11 Mold Cooling Apparatus for a Glassware Forming Machine 2.1.13 Apparatus and Method for Cooling a Mold This patent, relating to the method fo r cooling a mold was filed in 1992 by Charles Trevor Lawrence and assigned to VHC Limited [12]. The invention provides an arrangement for air cooling glass forming mo lds which is adaptable to both blank and blow molds. The mold holder is designed wi th internal air passages to provide for a continuous passage of cooling air to the mold Air enters these passages through an opening on the side of the mold and, from he nce to a vertical cooling passage between top and bottom of the mold.

PAGE 33

20 Figure 2.12 Apparatus and Method for Cooling a Mold 2.2 Liquid Cooling of Molds The following patents describe various methods of introducing liquid cooling. The intention is to make use of the much larger convective coefficients generally associated with such fluids. Because of the low boiling temperature of water at low pressure, some consideration might be given to alternate coolants, but it appears that each of these designs is intended to use water. If excessive water temperatures are to be avoided, the designer must provide a system that will result in a large temperature drop between the mold and the coolant and that is the central feature of each of these patents.

PAGE 34

21 2.2.1 Fluid Cooling of Glass Molds This patent, describing the fluid cooli ng of glass molds, was filed by Millard Jones in 1977 and assigned to Owens-Illinois [ 13]. This particular patent has received extra consideration in that some have suggested that it may cover a wide range of liquid cooling designs. After a fairly careful re view, it is limited in scope and unlikely to encompass those ideas under consid eration in this design effort. Figure 2.13 Fluid Cooling of Glass Molds The basic cooling arrangement is very similar to that used in air cooled molds; several radially spaced, axial passages are provid ed, either in mold itself as in a mold holder. These are said to provide “a good heat transfer relationship” to a set of mold inserts. In order to provide a sufficient temperature drop between the cooled surfaces and the liquid coolant, these pa ssages are radially insulate d using 316 st ainless steel compacting powder for managing the heat tran sfer. They may exte nd through the mold

PAGE 35

22 from the top, substantially to the lower e nd of mold. The liquid coolant passes through the coaxially positioned metal tubes in passage, Proper selection of thickness of material, its degree of compression and its composition results in desired temperature at the molding surface of mold. The primary inventio n claimed herein is the use of compacted granules, i.e. 316 stainless stee l but compacted powders of othe r origin are also included. 2.2.2 Mold with Exterior Heat Conducting Elements This patent relating to the exterior heat conducting elements in a mold was filed in 1981 by Julius J Torok [14]. In his invention, heat flows from the forming surface to the outer surfaces of the mold. Th e outer surface of the mold is thermally connected to a set of fluid conduits via a set of thermally conducting elements. While the conducting elements are designed to maintain thermal c ontact, through proper sizing they are used to provide the necessary temperature drop betw een the mold surface and the circulating coolant. The inventor states that “The heat conducting elements are designed with respect to the heat transfer to be of specific shape, size and material such that when the mold is in service each segment of th e forming surface is at a predetermined temperature.” The patent also includes a de sign in which the mold includes two parts, a glass forming part and a heat dissipating part ; the two parts are said to be connected through a thermally conducting interface.

PAGE 36

23 Figure 2.14 Mold with Exterior Heat Conducting Elements 2.2.3 Blow Mold Cooling This patent, relating to blow mold c ooling, was filed in 1984 by Richard Alan Letellier and assigned Emhart Industries Inc. [15]. Th e invention describes a method to cool molds by spraying cooling a fluid (i.e. wate r) on to outside as well as the inside of each mold valve. Cooling is achieved by mean s of several cooling spray nozzles which spray water directly onto resp ective mold halves. The cooling is done by a control valve which operates such that cooling fluid is sprayed only when molds are open and stops when the molds close.

PAGE 37

24 Figure 2.15 Blow Mold Cooling 2.2.4 Cooling Molds Used in Forming Glassware Containers This patent, relating to water cooling of final molds in a glassware forming machine, was filed in 1982 by Stanley Jones and assigned to Emhart Indu stries Inc. [16]. Particular care was taken to limit the temper ature to which water is exposed to prevent vapor formation. This is accomplished by provi ding a thermal barrier between the mold and coolant passage. This barrier consists of a stand off bracket with limited contact area designed to produce a large temperature di fference. The thermal conductivity of the

PAGE 38

25 barrier can be varied by cutti ng holes or adding pads to the bracket. Th is is a very broad claim and may cover designs well beyond those originally envisioned by the inventor. Figure 2.16 Cooling Molds Used in Forming Glassware Containers 2.2.5 Mold Cooling This patent was filed by Stanley Peter Jones in 1986 and was assigned to Emhart Glass Machinery Inc. [17]. This patent desc ribes water cooling of a mold assembly. The cooling assembly is provided with a coo ling passage through which water flows. The

PAGE 39

26 cooling assembly and mold are assembled such that they enclose a thin chamber (which is heat barrier) between external surface of cooling assembly and mold. A mixture of gases preferably air and helium is provided to the enclosed chamber to control transfer of heat between mold and cooling assembly, cont rolled cooling of the mold can be achieved by varying the proportion of helium in cooling air. Figure 2.17 Arrangements for Mold Cooling 2.2.6 Method and Apparatus for Mold Cooling This patent was filed in 1988 by Guillermo Cavazos and M. De Cervantes [18]. In this patent he has adopted the concept of a reboiler to the cooling of a mold. In this case a phase change liquid (distilled water) is allowed to boil inside of a mold cavity.

PAGE 40

27 Each of these mold cavities is provided with a series of vertically arranged blind bores between the molding surface and the mold exterior. The bores are connected to a common manifold, acting as a va por separator, and which, in turn, is connected to a condenser. The bores are filled with the phase change liquid which is vaporized due to heat transfer. The vapor rising from the mold rises to the manifold/vapor separator. The liquid is allowed to flow by gravity back into the cooli ng chamber inside the mold. Figure 2.18 Method and Apparatus for Mold Cooling The vapor flows upward to a condenser, is condensed and flows back into the mold via the manifold. This is not a good arra ngement if the primary coolant is water. The problem is that water pressures will need to be quite high if water is to boil at the desired temperature.

PAGE 41

28 2.3 Liquid Metal Cooling Several patents have been issued to consid er the use of liquid metals to moderate the range of temperature variations within the mold. These are summarized below: 2.3.1 Molding Cooling System for th e Manufacture of Glass Articles or Similar Materials This patent, relating to molding cooli ng system for the ma nufacture of glass articles or similar materials, was filed in 1989 by Alfredo Martinez-Soto et.al. and assigned to Vitro Tee Fideicomiso [19]. The invention describes a cooling system for a parison glass mold with a plunger to pre-form the cavity. The mold used consists of first body having an internal cavity made of Fe, Cu or Ni base alloys and a second body to absorb heat and control heat extraction. Du ring the cycle, the second body will undergo an alternate freezing/melting process so as to moderate the magnitude of any temperature peaks within the combined system. Figure 2.19 Molding Cooling System fo r the Manufacture of Glass Articles

PAGE 42

29 Thermal load variations will affect amount of liquid and solid zone of the second body is preferred to have high thermal conduc tivity, high heat capacity, and high melting point. A third body complements the mold in or der to release heat and can be provided with grooves, holes, pipes for fluid c ooling in order to dissipate heat. 2.3.2 Glass Forming Mold This patent, relating to cooling glass forming mold, was filed in 1962 by Hanns Stinnes and Martin Strasse [20]. This particular patent is of particular interest in that it involves heat transport by radi ation. However, radiation is used in a substantially different manner than envisioned in the design under consideration. Figure 2.20 Glass Forming Mold

PAGE 43

30 The inventor has designed a cooling sy stem with which incorporates a liquid metal between the primary mold and the surr oundings. The benefit from the liquid metal is that it acts to ensure a uniform temp erature across the mold surface. Heat is unidirectionally conveyed from the forming su rface to the liquid metal and then to the second surface portion of the mold. The space be tween two surfaces is filled with a heat conducting substance (Na), so that it cannot react with atmosphere air from the second surface, heat is radiated to the surroundings via a set of radiat ing/convecting fins. 2.4 Controlling Mold Temperature The following patents refer to ideas for monitoring and controlling mold surface temperature. 2.4.1 Controlling the Temperature of a Glass Mold This patent was filed by Stanley P. Jones in 1984 and assigned to Emhart Industries [21]. This particular invention ha s focused upon controlling the temperature of a glass mold. This method involves the fo rmation of a passage in the mold wall, extending from external surface to a position adj acent to the cavity. An infra-red radiation transmitting device is inserted in this passage During the manufacture of molded articles the temperature of the mold is detected and this information is used to control the mold heating or cooling means. The transmitting devi ce is usually silica or glass rod, but an aluminum tube may be alternatively used.

PAGE 44

31 Figure 2.21 Controlling the Temp erature of a Glass Mold 2.4.2 Controlling the Temperature of a Mold This patent, relating to c ontrolling the temperature of a mold, was filed in 1984 by Stanley Peter Jones and assigned to Emhart I ndustries Inc. [22]. It appears to be very close in concept to the U.S. Patent Nu mber.4,519,827 granted to Stanley Jones the same year. In this invention an infrared, radiation transmitting device and an infrared radiation detection device are used to detect the temp erature of the mold during the manufacture of molded articles An infrared radiation transm itting device is inserted in a passage in the mold portion adjacent to the mold cavity. The infrared radiation detection device is connected by a fiber optic guide to the transmitting device. The device is connected to control means of the machine in which mold portion is mounted. This information is used to control the operati on of mold heating or cooling means.

PAGE 45

32 Figure 2.22 Controlling the Temperature of a Mold 2.4.3 Mold Arrangement for a Cyclical ly Operating Glassware Container Manufacturing Machine with Temperature Sensing Means This patent, relating to mold arrangem ent for a cyclically–operating glassware container manufacturing machine with temper ature sensing means, was filed in 1984 by Frank A. Fenton and assigned to Emhart Industr ies Inc. [23]. The obj ect of the invention is to provide a mold arrangement for cyc lic operating glassware manufacturing machine, in which a temperature detecting device is able to detect the temperature of the mold. The blank as well as the final mold carry a temperature detecting device which is operable to detect the temperature of at least one of the side mold portions. The temperature detecting device can be a thermocouple or an in frared radiation transmitting device as shown in the attached drawings.

PAGE 46

33 Figure 2.23 Mold Arrangements for a Cyc lically Operating Gl assware Container Manufacturing Machine with Temperature Sensing Means 2.5 Dead Pan Cooling A patent has been applied to the problem of cooling the bottom of freshly molded glassware. Because of the significantly grea ter thickness of the base of many such items, there will be a greater problem with excessive reheating in this regi on. One ideas to deal with this problem is presented below. 2.5.1 Cooling Articles of Newly Molded Glassware This patent, relating to cooling articles of newly molded glassware, was filed in 1984 by Hermann Heinrich Nabelung and assigned to Emhart Industries Inc. [24]. This invention is relating to c ooling of newly molded glassware with a dead plate arrangement. The object of the i nvention is to enable cooling of articles of glassware to take place on a dead plate without having to be held in position by the take out

PAGE 47

34 mechanism of the machine. The newly molded glassware is positioned on the horizontal dead plate with a central ope ning of the dead plate beneath the central region of the bottom of the article and a plurality of gr ooves around the periphery. Air is sucked from central opening and so that the glassware stays in place and is also blown on sidewalls of the articles through plurality of nozzles uniformly spaced around the article, thus effecting cooling of the bottom as well as sides of the glass article. Figure 2.24 Cooling Articles of Newly Molded Glassware

PAGE 48

35 2.6 Neck Mold Assembly Cooling The neck mold is a separate item and will require special consideration for cooling during the overall glass forming process. The neck ring contains somewhat thicker glass than the body of the molded it em. Moreover, since th is portion of the item is used to handle and transport the parison and finished item, it will undergo higher stresses. 2.6.1 Glassware Molding Machine with Unitary Axis Molding This patent was filed by Wilbur Orla nd Doud in 1987 and was assigned to Ball Corporation. Doud designed a glassware moldi ng machine with a plunger and neck mold assembly that works in conjunction with a car riage and a parison blank mold [25]. The assembly also acts to suspend the comp leted parison by the neck ring portion. Figure 2.25 Glassware Molding Machin e with Unitary Axis Molding

PAGE 49

36 The whole assembly is designed such that while one pair of forming molds transports the completed container for cooling, the other pair of form ing mold portions is opened to have a continuous operating cycle. 2.6.2 Glass Container Forming Machine In cluding Neck Ring Mold Cooling This patent, relating to Glass container forming machine including neck ring mold cooling, was filed in 1992 by Robert S. Johns on & Robert D. Hall and was assigned to American National Can Company [26]. It is an object of the invention to provide a simple low cost way of providing adequate cooling to a neck ri ng mold of a parison forming machine. Reduction in neck ring mo ld temperature allows increased speed of operation. An air supply is provided at the pari son forming station a nd directs a flow of cooling air at exterior surfaces of neck ri ng mold extensions, thereby cooling the mold halves even in the closed position. The mold halves also have a vertical cooling air passage for enhanced heat transfer through the mold. 2.7 Interior Cooling in the Finishing Mold 2.7.1 Apparatus for Forming Glass Ar ticles with Treating Mean This patent, relating to the apparatus fo r forming glass articl es with treating means, was filed in 1965 by John E. Cook and was assigned to Owens. The invention relates to a method and apparatus for contro lling the cooling air on a Westlake glass blowing machine. In this method, a gather of glass is placed on the upper end of a spindle. When the spindle rotates, end for end, about its axis, a small amount of air is introduced to the interior of the gather to form an initial cavity. Cooling air is also

PAGE 50

37 directed against the sidewall of parison fr om one direction in order to control its elongation. The amount of cooling air will dete rmine the rate of el ongation of parison. The spindle end which holds the upper end of pa rison is cooled by air directed against the side of the spindle. Just before the enclosi ng of parison within the paste mold, the spindle rotation is interrupted so that greater amount of glass is accumulated near the bottom of parison and increase in elongation rate uni nfluenced by centrifugal force. Now the cooling air directed against th e sidewall of parison is disc ontinued where as the cooling air directed against the spindle is conti nued so as to avoid uneven glass thickness distribution upon final expansion of the parison in the paste mold. Figure 2.26 Apparatus for Forming Gla ss Articles with Treating Mean

PAGE 51

38 CHAPTER 3 NARROW NECK PRESS AND BLOW PROCESS The narrow neck press and blow ( NNPB) process was introduc ed to gain better control over glass distributi on in the container. The improved control over glass distribution has enabled signi ficant reduction in glass weight of up to 33% without adversely affecting the mechanical performan ce of the container. A key component in the above process is the plunger, used to form the cavity in the parison during the forming stage. The function of the plunger is to evenly distribute the glass w ithin the blank mould cavity and to aid the removal of thermal energy from the inte rnal surface of the parison. The NNPB Forming Cycle can be split into the following steps 3.1 Gob Delivery With the plunger in loading position, the gob is delivered through the funnel of the blank and loads on top of the plunger. Gob shape and loading depth are the most important factors. Loading depth into the bl ank should be about inch below the top of the blank.

PAGE 52

39 Figure 3.1 Gob Delivery 3.2 Start of Press The pressing action starts as soon as possi ble after the glass loads into blank and the baffle is down. Once the glass enters the blanks, it cools rapidly hence, the pressing should start as soon as possible which a llows a lower pressure to be used. Figure 3.2 Start of Press

PAGE 53

40 3.3 Press Time During press, the plunger travels in an upward direction forcing the glass up against the baffle and then down into the neck ring to the parison. Correct weight is very critical there has to be just enough glass to fill the cavi ty between the plunger and the blank. The amount of pressing pressure app lied is approximately 8 to 12 lbs. excess pressure may cause defects such as split finishes, and blank tears. Figure 3.3 Press Time 3.4 Plunger Down In this step the plunger retracts to it s full down position where it compresses the receiver spring and is stoppe d by the loader spacer. At this point, there is enough clearance for the invert mechanism to transfer the parison to the mold side without the neckring making contact with the plunger. Paris on reheat will then begin on the inside of the parison. At this point the colder inner skin of the parison will be heated up by the hotter glass in the middle of the parison.

PAGE 54

41 Figure 3.4 Plunger Down 3.5 Reheat Total parison reheat is initiated when the baffle is up, the blank open, and the plunger is in the down position. This process wi ll continue until the final blow is applied. At this point, the parison is transferred to the mold as soon as possible. Correct invert speed and cushioning are important factors. To o fast or too slow i nvert speed can cause swung baffles. It can also cause the parison to be pinched by the mold. At this time, the effects of reheat will cause glass to soften and start to run at the bottom of the mold.

PAGE 55

42 Figure 3.5 Reheat 3.6 Reheat and Run Reheat will remain in process and continue until final blow is applied. As soon as the parison is transferred to the mold side, it starts running towards the bottom of the plate. The amount of run time is determin ed by the job being made. The longer the parison is allowed to run, the thicker the glass will be in the bottom area. Figure 3.6 Reheat and Run

PAGE 56

43 3.7 Final Blow and Vacuum Forming The amount of final blow and vacuum duration will be determined by the job running. Final blow is applied af ter the blowhead is on the mo ld and should be off before the blow head rises. Vacuum should be appl ied approximately 5 de g before Final blow “on”, and turned of 5 deg. after Final blow “off”. Vacuum is pulled through valves in the shoulders of the mold and ditches which are milled in the face of the mold. It is then ported through the bottom plate, Verti-flow mechanism, and to the vacuum pump. The air is then exhausted out to the atmosphere. Figure 3.7 Final Blow and Vacuum Forming 3.8 Mold Open and Takeout The mold opens after the blow head is completely up and the takeout closes around the finish before the mold opens. Th e takeout action is smooth as possible to lessen the change of defects.

PAGE 57

44 Figure 3.8 Mold Open and Takeout 3.9 Dead Plate Time The container is about inch above the dead plate while hanging in the takeout process which allows the air coming through the dead plate to cool the bottom and the sides of the bottle. The amount of time cont ainers are held above the dead plate is determined by the amount of cooling needed to do the job. Figure 3.9 Dead Plate Time

PAGE 58

45 CHAPTER 4 EXPERIMENTAL ANALYSIS OF AIR FLOW THROUGH COOLING HOLES OF THE MOLD The mold contains 44 cooling holes passi ng vertically throu gh upright mold. The analysis of individual vertical passage w ould normally be a relativ ely straight forward undertaking, but the problem was complicated by the presence of three circumferential, axially displaced grooves. Horizontal grooves, cut through the outer surface of the mold at an axial position about 1/3 from the top of the mould, penetrate into the outer row of holes leading to discontinuity in the flow of air through the mold Figure 4.1 Mold with Fins Cut thro ugh it exposing the Outer Row Holes Fi n Outer Row Hl Fins

PAGE 59

46 In order to establish the relative importance of these grooves on the overall heat transfer process a simple experiment was de veloped to compare th e flow of air through the inner and outer row of coo ling holes and to establish what error might be introduced into the analysis should they be ignored. 4.1 Experimental Setup and Procedure A simple experiment was conducted to analyze the flow of air through the cooling holes of the mold. The main objective of this experiment was to compare the air flow between the inner row of holes and the outer row of holes of the mo ld. The outer row of holes differ from the inner rows due to th e circumferential fins provided on the outer surface of the mold which cut in the mold making small openings in the outer row of holes allowing a venturi effect for the flow of air. Figure 4.2 shows the experimental setup. Figure 4.2 Experimental Setup to Analyze Air Flow through Cooling Holes in the Mold

PAGE 60

47 Procedure The mold and flow meter were placed on a stable table so that the calibration of the meter would not be disr upted during the experiment. The flow meter was placed on the table and was leveled such that the bubble on the spirit level was centered. This ensure s that the differential elevations within the meter are truly vertical so that accurate readings can be maintained. A small nozzle was attached to the outlet air hose of the air compressor to constrain the air flow through the mold holes. The flow meter was turned on. The needle pointer in the flow meter was adjusted to remove any backlash error. The air compressor was turned on and was set at value of 20Psi. The nozzle was inserted into the mold hol e and the lever provided to the nozzle was depressed to allow flow air flow into the cooling holes of the mold, the high pressure inlet of the flow meter was held at the other end of the mold; values read on the flow meter were noted. The same procedure was repeated for all the holes. The inlet pressure was then changed to 30Psi and a second set of readings was noted. This experiment was repeated on two nonconsecutative days with the order of readings being varied to ensure agains t the introduction of any data bias. The pressure was measured remotely from the test section. A long hose ran from the air compressor to the table on which the mold was placed, this accounted for pressure losses. To enable the flow of ai r into the cooling channels of the mold a

PAGE 61

48 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 nozzle was attached at the end of the hos e which further increased the pressure loss. The values of air flow velocities hence obtained appear to be on the lower side. The primary interest of this experi ment was however to compare the air flow between the cooling channels based on the relative velo cities and not absolute values. 4.2 Observations Flow meter readings for inle t pressure of 20Psi and 30 Psi are tabulated as shown in the Table 4.1. In the original glass mold the inner and outer row holes are arranged as shown in the Figure 4.3. From the figure it is evident that there are 12 inner row holes and 10 outer row holes. Figure 4.3 Cross-section of the Mold Showing Inner and Outer Row Holes

PAGE 62

49 Table 4.1 Flow Meter Reading for Each Hole 4.3 Calculations The values noted above are then used to calcu late the air flow velocity through the holes using a simple empirical formula described by equation (4.1) and are tabulated in the Table 4.2. 1000 93 FMR V Fps (4.1) Where FMR is the Flow Meter Reading in A 12 11 10 9 8 7 6 5 4 3 2 1 Inner Row Hole No. 25 26 26 22 21 21 20 25 24 25 24 21 Flow meter Reading in A (At 20Psi) 27 23 27 28 29 30 28 27 23 29 29 29 Flow meter Reading in A (At 30Psi) 10 9 8 7 6 5 4 3 2 1 Outer Row Hole No. 23 19 21 20 23 19 24 17 26 16 20 21 22 23 22 24 20 26 20 19 Flow meter Reading in A (At 30Psi) Flow meter Reading in A (At 20Psi)

PAGE 63

50 The Standard deviation for the air velocities is calculated to analyze the fluctuations in the air flow through the mold. Table 4.2 Flow Velocity through Each Hole The air velocities in fps calculated using equation (4.1) is plo tted against the hole number for both inner and outer rows at th e two pressure levels 20 Psi and 30 Psi. Outer Inner Outer Inner 17.399 13.795 12 17.193 17.327 12.936 13.889 Average Std Dev 11 10 9 8 7 6 5 4 3 2 1 Hole No. 0.550 13.477 14.996 13.795 13.477 13.477 13.153 14.105 14.408 14.105 14.408 13.477 Velocity at 20Psi (Fps) 14.705 14.105 14.105 14.408 13.795 14.105 14.105 14.408 14.705 13.477 0.614 17.646 16.895 17.399 17.399 16.895 16.895 17.149 17.149 17.646 16.895 16.637 16.895 17.149 16.637 16.375 17.149 16.895 18.130 18.367 17.889 18.367 Velocity at 30Psi (Fps)

PAGE 64

51 Variation of Air Flow Between Inner and Outer Row of Holes(Inlet Pressure 20 Psi) 0 5 10 15 20 123456789101112 Hole NumberFlow Velocity(fps) Inner Row Holes Outer Row Holes Figure 4.4 Air Velocity (fps) Vs Hole Number in the Mold at an Inlet Pressure of 20Psi Figure 4.5 Air Velocity (fps) Vs Hole Number in the Mold at an Inlet Pressure of 30Psi Variation of Air Flow Between Inner and Outer Row of Holes(Inlet Pressure 30Psi) 0 5 10 15 20 123456789101112 Hole NumberFlow Velocity(fps) Inner Row Holes Outer Row Holes

PAGE 65

52 4.4 Results and Conclusions From the standard deviation and average flow velocities for inner and outer row of holes for both the inlet pressure calcu lated above it is easily comprehend that the air flow through all the mold holes is almost constant. The graphs above show that the air flow velocity from the inner row and outer row of the holes is almost same. The fins cut on the outer surface of the mo ld do not affect the air flow through the outer row of holes. The effect of the fins can hence be neglec ted and the mold holes will be treated as individual vertical passages while calcu lating the amount of heat transfer and mass flow through each of them.

PAGE 66

53 CHAPTER 5 ONE DIMENSIONAL STEADY STATE ANALYSIS 5.1 Introduction Detailed analytical solutions to the 3 dimensional transient general heat conduction equations are restricted to simple geometries and boundary conditions. In this particular case the geometry and the boundary conditions preclude the use of analytical techniques, and recourse is made to finite differences methods or to one dimensional steady state approximations. The amount of heat transferred to the coo ling air from the mold is calculated in this one dimensional steady state numeri cal solution. Here we assume a similar performance in the steady state and the transient st ate solution and vari ations involved are hence neglected. This MathCAD code calcul ates the amount of he at transferred per cooling hole at a particular inlet pressure a nd temperature of air. The geometry of the mold is sufficiently complex; a code for cal culating the total amount of heat transferred from the mold to the cooling air was not assessable. The main objective of the problem is to increase the amount of heat transferred from the mold to the cooling air in a gi ven time interval. This can be achieved by increasing the area of heat transfer. The area available for heat transfer can be increased

PAGE 67

54 by drilling bigger diameter holes in the mold. There is a limitation to the increase in the size of the holes as they overlap each othe r and also drilling holes that are too close causes the web between the holes to be very thin not allowing sufficient area for heat transfer radially outward direction. In la ying out the geometry an increase in cooling channel diameter from to 9/32 inch is recommended. Introducing a thin metal fin in the holes increases the surfa ce area considerably. The MATHCAD code calculates the mass flow rate of ai r and the amount of heat transferred per hole taking into cons ideration the compressible air flow. 5.2 Steady State Analysis As described above we are trying to anal yze the air flow through the mold cooling holes. The problem is simplified and can be defined as shown in the Figure 5.1 Figure 5.1 One Dimensional Mathcad Model Cold Air in Constant Wall Temperature Fin Hot Air out

PAGE 68

55 5.2.1 Compressible Flow and Heat Transfer Modeling The MathCAD code developed solves for fr ictional and minor flow losses so as to adjust flow rate for a constant pressure drop, P This problem is subjected to boundary conditions of constant pressure drop and cons tant wall temperature. Assuming a pressure drop of 60 inches and a constant mold wa ll temperature of 800F, a single channel analysis is completed for the inner and outer row of cooling passages. The expression used for core pressure dr op so as to accommodate compressibility effects is that of Kays and London [28] 1 2 2 1 1 2 2 1 1 2 11 1 2 1 2 v v K v v A A f v v K P v g G P Pe m c F c c (5.1) Where, G= b b a aV V mass velocity based on free flow area 1vEntering specific volume 2vExiting specific volume mv Mean specific volume 1PInlet pressure P Core pressure drop cK Entrance loss coefficient eK Exit loss coefficient Ratio of free area flow to frontal area

PAGE 69

56 A Total heat transfer area cAFree flow area FfCore fanning section area For this particular case the values of the entrance and exit loss coefficients are estimated using the Kays and London’s recommendation, 28 0 & 4 0 e cK K The heat convected to the gas is equal to mcp(Tout – Tin), where m is the mass flow rate, cp, the fluid specific heat and Tout – Tin represents the coolant temperature change. Other than increasing the air flow rate, which would require higher capacity air blowers with increased operational costs, the on ly option open to us is to increase the air outlet temperature. An equation to describe th e air outlet temperature has been previously evaluated within the literature and the follo wing relationship is used to describe this arrangement [29]. Tout = TMold – (TMold-Tin) e-hA/mc p (5.2) Where the Mold temperature, TMold, is assumed uniform, the air enters at Tin, exits at Tout, the average convective coefficient is given as h the convective surface area as A the mass flow of air as m and the specific heat of air as cp The MathCAD code developed takes in ai r properties versus temperature in a matrix for cubic spline interpolation. The mold conditions and air inlet conditions are given as input.

PAGE 70

57 Density kgK kJa a ,5 4 3 2 (5.3) at SpecificHe C kgK kJ C C C C C Ca a ,5 4 3 2 1 (5.4) ductivity ThermalCon K mK watt K K K K K Ka a ,5 4 3 2 1 (5.5) ture AirTempera T K T T T T T Ta a ,5 4 3 2 1 (5.6) sNumber andtl Pr Pr Pr Pr Pr Pr Pr Pr ,5 4 3 2 1 (5.7)

PAGE 71

58ity is KinematicVacos ,5 4 3 2 1 (5.8) 5.2.2 Fin Analysis The size of the fins to be used was also an important criterion as it affected the heat transfer rate due to changes in the ai r flowing through the cooling passage and also making considerable impact on the losses due to friction. Hence, the efficiency of the fins was calculated, the fin with efficiency of approximately 90% or more having the most positive impact on the heat transfer rate wa s chosen. The available fin sizes from Small Parts in Miami varied in thickness as follows: 0.016in, 0.025in and 0.032 in. These sizes were then checked for efficiency and heat transfer enhancement. The fin thickness is selected to optimize the conflicting requirements of maintaining adequate air flow the cooling passage and effectively augmenting heat transfer. The fin is inserted in the outer row of c ooling holes. To calculate the efficiency of the fin the problem was simplified and compared to a fin representing an insulated tip as shown in the figure 5.2.

PAGE 72

59 Figure 5.2 Schematic Representation of the Fin inside the Air Cooling Channel The efficiency of an insulated tip as shown above is given as follows [29] mL mL tanh (5.9) Where, KA hP m (5.10) Where, P= wetted perimeter = 2Z and Surface area=Zt Z=the length of the fin t=thickness of the fin 5.2.3 Calculations for Fin Efficiency Let us assume the velocity of air flowing through the cooling passages asfps v 300 The diameter of the cooling passage, in d 32 9 Fin Cooling Channel

PAGE 73

60 Properties of air K m watt k 0338 0 ; 69 0 Pr ; 2 5sec 10 388 3 m Newtonw 38711 0 m kg ; 3sec 000023 0 m Newton Where K= Thermal conductivity w Kinematic viscosity Pr = Prandtl’s Number Density Thermal conductivity of fin, 1 1110 K m watt kf Hydraulic Diameter ) 2 ( d D (5.11) Thickness of the fin in t 016 0 D v Re (5.12) 410 512 1 Re The Nusslet’s Number can be calculated as follows [29] 586 77 Pr 027 033 0 8 0 wRE Nu (5.13) The heat transfer coefficient D k Nu h (5.14) R ft hr BTU Nu . 805 1052

PAGE 74

61 Therefore the efficiency of th e fin for the given thickness is, 2 2 2 2 tanh5 0 5 0d t k h d t k hf f (5.15) 899 0 5.2.4 Results of Fin Analysis Similar calculations are done for the othe r two fin sizes available. The heat transfer rate is calculated with the MathCAD code and the results of the calculations are tabulated in table 5.1. Table 5.1 Results of Fin Analysis Fin Thickness, t(In) Efficiency, (%) Heat Transfer Q(BTU/Hr) 0.016 89.99 1262 0.025 93.33 1155 0.032 94.70 1070 The graph in figure 5.3 clearly indicates th at as the thickness of the fin increases the heat transfer rate decreases, this is due to the fact that the increasing fin thickness hampers the air flow through the cooling channe ls of the mold hen ce reducing the amount of heat transferred from the mold to the cooling air.

PAGE 75

62 Figure 5.3 Varying Heat Transf er Rate with Fin Thickness From the above results it was concluded that the thickest fin even though was high in efficiency failed to achieve the desi red result, which is gain in heat transfer. Hence, the fin with a thickness if 0.0 16 in was chosen for further analysis. 5.3 Parametric Analysis The pressure drop in the cooling passage as per the date acquired from ACG is 60in of H20. The cold fluid properties are determ ined from the cubi c spline curve fits. The cold fluid temperature is not prescribed in this case and is hence solved here by trail and error using the code. In this program we have taken into consid eration the change in fluid properties with temperature, the efficien cy of the fin and analyze the problem with a fixed pressure drop and for constant wall temp erature. The use of hydraulic diameter is made for calculating the Nusselt’s number. The code is then run several times with different values of cooling passage diameters and fins. The output of the program gives the flow rate and the heat transfer Fin effectiveness Graph 1050.00 1100.00 1150.00 1200.00 1250.00 1300.00 0.010.0150.020.0250.030.035 Fin Thickness(in)Heat Transfer(BTU/hr )

PAGE 76

63 rate which are the most important parameters of this analysis. The results obtained from the runs are tabulated according to the numb er of fins used per cooling passage are tabulated in table 5.2 Table 5.2 Results of Increasing Air Coolant Passage Diameter in Finned and Un-Finned Passages While the results shown here were found for Mold temperatures of 800oF, results are not strongly affected by variat ions in temperature. The critical parts of this analysis were repeated at Mold temper atures ranging between 650 and 950oF with results shown in Table 5.3. Clearly variations in Mold temperature do not pr oduce significant changes in the relative improvements in heat transfer or air flow rates. Table 5.3 Effects of Variations in Body Temperature on Existing and Prototype Molds Inch Coolant Passage 9/32 Inch Coolant Passage Type of fin used Air Flow lbm/hr Total Heat Transfer BTU/hr Air Flow lbm/hr Total Heat Transfer BTU/hr No Fin 26.96 2187 36.06 2605 Strip Fin 18.07 2336 24.90 2958 Tri–Star Fin 15.05 2156 21.02 2862 Inch Coolant Passage, Unfinned 9/32 Inch Coolant Passage, Finned Change, % Mold Wall Temp. Air Flow lbm/hr Total Heat Transfer BTU/hr Air Flow lbm/hr Total Heat Transfer BTU/hr Total Heat Transfer Air Flow 650 oF 28.47 1763 26.57 2408 36.6 -6.7 800 oF 26.96 2187 24.90 2958 35.3 -7.6 960 oF 25.61 2589 23.43 3470 34.0 -8.6

PAGE 77

64 5.4 Axial Distribution of Cooling thro ugh Constant Temperature Cooling Passage The basic flaw with the current mold re lates to the cooling near the shoulder region of the bottle, hence it was necessary to compute the relative cooling at various mold sections. The change in the relative cool ing at the various mold section is monitored with the MATHCAD code .Table 5.5 shows the amount of heat removed from the mold by the cooling air as it passes through the cooling channel. Table 5.4 Relative Cooling at Various Mold Sections Fraction of Energy Transferred Below Elevation Frictional Axial Distance along Mold ” hole no Fin 9/32” hole No Fin ” hole Strip Fin 9/32” hole Strip Fin 0.1 0.683 0.704 0.609 0.62 0.2 0.729 0.745 0.682 0.686 0.3 0.771 0.784 0.745 0.744 0.4 0.811 0.82 0.8 0.796 0.5 0.848 0.855 0.847 0.841 0.6 0.883 0.887 0.887 0.881 0.7 0.915 0.918 0.922 0.917 0.8 0.945 0.947 0.952 0.948 0.9 0.974 0.974 0.978 0.976 1.0 1 1 1 1

PAGE 78

65 Figure 5.4 Axial Distribution of Cooli ng through Constant Temperature Passage 5.4 Conclusions As the number of fins increases the air flow rate decreases hence reducing the heat transfer rate hence, no consider ation is given to more elaborate fin arrangements. The best possible combination from the a bove results is the one where the cooling channel diameter is 9/32 in and the straight fin is used. The increase in the amount of heat transferred between the standard mold and the new design is 35.3%. The decrease in the cooling air mass flow rate between the standard mold and the new design is -7.6%. Axial Distribution of Cooling through Constant Temperature Passage 0.5 0.6 0.7 0.8 0.9 1 1.1 00.20.40.60.811.2 Fractional Axial Distance along the MoldFraction of Energy Transferre d Below Elevation 1/4" No Fin 9/32" No Fin 1/4" Straight Fin 9/32" Straight Fin

PAGE 79

66CHAPTER 6 ANALYTICAL TRANSIENT ANALYSIS The MATHCAD code used in the previous chapter was used to evaluate the cooling channel diameter and the fin dimensio ns for the maximum possible heat transfer rate. This particular chapter deals with the analytical solution to determine the depth of the transient heat penetrati on in the glass and the mold. 6.1 Formulation of Semi-I nfinite Solid Problem The problem can be simplified and re presented as follows in Figure 6.1 Figure 6.1 Semi Infinite Solid with Constant Wall Temperature T(x,0)= Ti T ( 0 t ) = Ts Mold G L A S S Constant Wall Tem p erature x T iMold =800F T iGlass =1700F

PAGE 80

67 The heat equation for transient conducti on in a semi-infinite solid is given by t T x T 2 12 2 (6.1) The initial conditions are given as follows, (6.2) In this problem we have a constant wall temperature; the surface heat flux for this particular condition is given as [29], 2 1) ( ) ( ) ( t T T k t qi s s (6.3) We also know that for th is particular case [29], ) 2 ( T T T t) T(x,s i st x erf (6.4) 6.2 Calculations Here we wish to find out how far the heat penetrates (x) into the glass and into the mold within the given 3 seconds time interval. The surface heat flux, 2 1) ( ) ( ) ( t T T k t qi s s (6.5) Where pC k (6.6) p i s sC k t T T t q 2 1) ( ) ( ) ( (6.7) iT x T ) 0 (

PAGE 81

68 Here we also assume that, The amount of heat lost by glass = th e amount of heat gained by the mold ) 1 )( ( ) ( i s mold p i s glass pT T t C k T T t C k (6.8) Where iM iG s iGT T T T is the temperature distributi on constant (6.9) Mold P Mold P Glass PC k C k C k ) ( ) ( ) ( (6.10) The following properties are known bot h for glass and for the mold. Glass properties Thermal conductivity, F in BTU kg. sec 10 3 25 Density, 30788 0 in lbg Specific heat, F lb BTU Cpg. 179 0 Mold properties Thermal conductivity, F in BTU kg. sec 10 3489 54 Density, 33081 0 in lbg Specific heat, F lb BTU Cpg. 119 0 Initial temperature of the glass, F TiG 1700 Initial temperature of the mold, F TiM 800

PAGE 82

69 Solving eqn. 6.10 we get =0.886 Solving eqn. 6.9 The interface temperature, sT = 814 F Solving eqn.6.4 for heat penetr ation into the glass we get, x=0.238 inches The distance penetrated by heat into the glass in a time inte rval of 3 seconds is x=0.238 inches. Solving eqn.6.4 for heat penetr ation into the glass we get, x=0.711 inches The distance penetrated by heat into the glass in a time inte rval of 3 seconds is x=0.711 inches. 6.3 Plotting Transient Temperature Dist ribution Profiles for Glass and Mold The calculations were done for the temper ature distribution in the glass and the mold depending upon the time and the location. Table 6.1 Variation in the Temperature Dist ribution depending on the Location and Time for the Glass Glass Temperature T(x,t) Degree F Distance in the solid x (in) Time, t=1 sec Time, t=2 sec Time, t=3 sec 0.06 1407 1242 1156 0.12 1164 1563 1475 0.18 1698 1674 1631 0.237 1700 1697 1684

PAGE 83

70 Temperature Distribution in Semi-infinite Solid (Glass) 1000 1100 1200 1300 1400 1500 1600 1700 1800 00.050.10.150.20.25 Distance in Solid,x(in)Temperature in Solid, T(x,t) Deg F Time t=1 sec Time t=2 sec Time t=3 sec Figure 6.2 Transient Temperature Distribut ion in Semi Infinite Solid (Glass) Table 6.2 Variation in the Temperature Dist ribution depending on the Location and Time for the Mold Mold Temperature T(x,t) Degree F Distance in the Solid x (in) Time, t=1 sec Time, t=2 sec Time, t=3 sec 0.711 700 701 702 0.534 700 703 708 0.356 704 716 726 0.178 725 752 762

PAGE 84

71 Temperature Distribution in Semi Infinite solid(Mold) 690 700 710 720 730 740 750 760 770 00.20.40.60.8 Distance in the solid,x(in)Temperature,T(x,t) Time t=1sec Time t=2sec Time t=3sec Figure 6.3 Transient Temperature Distribut ion in Semi Infinite Solid (Mold) 6.4 Surface Heat Flux Calculations The surface heat flux can be calculated using equation (6.5). The table below gives the values of the heat flux for di fferent times in the molding process. Table 6.3 Surface Heat Flux Variation with Time Time, t(sec) Surface Heat Dr op in Heat Flux with time 0.5 1450 -1.0 1025 425 1.5 836 189 2.0 724 112 2.5 648 76 3.0 591 57

PAGE 85

72 Surface Heat Flux Plot 300 500 700 900 1100 1300 1500 1700 00.511.522.53 Time, t(sec)Heat Flux,q"(BTU/hrin^2 ) Figure 6.4 Plot of Surface Heat Flux with Time 6.5 Conclusions The graphs indicate a uniform temperatur e profile in the glass and the mold The depth of heat penetration shows that the interior surface of the glass is cooling. The surface heat flux goes on decreasing with time; this implies that the cooling will not the hampered considerably, if the amount of time allowed for the glass to cool is reduced.

PAGE 86

73CHAPTER 7 MATHEMATICAL MODELI NG OF THE PROBLEM The thermal analysis of the mold consisted of the following steps Determining the problem domain nonlinear transient analysis. Creating the 2D model of the mold. Creating a finite element mesh. Applying boundary conditions. Setting initial conditions. Setting the thermal analysis parameters. Solving the problem. Post processing the results. The thermal analysis was solved by nu merical simulation using ANSYS. PLANE 55 element type was used in this analysis to mesh the model. These elements were used for calculations of temperature distri bution in regions, the glass and the mold. The problem was defined by the laws of conservation of energy. These laws were expressed in terms of partial differential equa tions, which were discretized with finite element based techniques.

PAGE 87

74 Assumptions made in the analysis are as follows The glass and the mold are in perfect thermal contact with each other. The glass and the mold are at individual uniform temperatures at the beginning of the analysis. 7.1 Governing Equations and Modeling The glass parison comes in contact with the mold for a brief period of 3 seconds, during this time the parison is blown into th e final product and the cooling takes place as air flows through the cooling channels. This s ituation is described by the equations stated below and numerical simulati on software, ANSYS is used to get the solution. The schematic drawing of the mold cross section wh ich is to be analyzed is shown in fig 7.1 Figure 7.1 Two Dimensional Model of the Mold

PAGE 88

75 The temperature field of the considered domain as a function of space and time can be obtained by solving the heat-c onduction equation. The equation in twodimensional Cartes ian coordinates ( x y ) is shown in equation (7.1). The temperature depends on space coordinates and time, i.e. ) , ( t y x T T (7.1) The density the specific heat capacity, c and the therma l conductivity, k, depend on temperature t T c y T k y x T k x (7.2) To complete the physical model the above equations are subject to the following boundary and initial conditions. Boundary Conditions: For 3 0 t At 1 3 1 4 1, 90 0 90 0 r r r r r r r r

PAGE 89

76 0 r T (7.3) This boundary condition applie s to all the coo ling channels and the fins that provide convection. At iR r ,1 where i changes for different cooling channel location for the inner row of holes Also atiR r ,1, where i changes for different cooling channel location for the outer row of holes ) ( ,2 1t T h a T k b R r a R r at (7.4) At, 90 ,3 jr r ) ( t T h a T k (7.5) For,

PAGE 90

77j jr x cos 03 0 r T (7.6) The initial conditions for the glass and the mold are as follows For0 t, At 90 0, 2 1 r r r F Ti o1700 (7.7) For 0 t At 90 ,3 2 jr r r F Ti o800 (7.8) 7.2 Finite Element Models Most of the heat removed from the glass is transported from the mould by cooling air in the axial channels. The heat flux is large in the radial direction compared to the flux in the circumferential and the axial directions. The model is axi-symmetric and the crosssection of the mould is projected on the radial section of the two-dimensional model. For simplicity and to reduce the time of the analysis only a quarter of the mold was analyzed

PAGE 91

78 taking into account the symmetry of the mold The two-dimensional model is presented in Figure 7.2 Figure 7.2 Two Dimensional Finite Element Model of the Mold The cooling channels are modeled by inserting convective boundary conditions on the periphery of the cooling holes and fins as shown in Figure 7.2. On the outside of the mould a convective boundary condition is used. ANSYS solves the above equations with Newton-Raphson method. In this case the full Newton-Raphson method is employed to calculate the solution. Mold Glass Fins

PAGE 92

79CHAPTER 8 RESULTS OF NUMERICAL SOLUTION As mentioned earlier the heat conductions and convection equations were solved numerically using ANSYS. The equations descri bed in the last chapters were solved numerically to predict the amount of heat transferred from the glass to the mold and from the mold to the cooling air flow ing through the cooling channels. As per the results obtained in the Mathcad code the mold was analyzed for the following cooling channel diameters: 8/32 in and 9/32 in with straight fin inserted in the outer row of cooling channels axially thor ough the length of the mold. The temperature distributions was observed in each of the cases for the mold and for the glass to analyze the maximum cooling that is attained by incr easing the cooling channel diameter and use of fins in the outer row cooling holes. The mold was analyzed at 2 critical sect ions axially, firstly at the shoulder region of the bottle and finally at th e neck of the bottle because theses are the areas where the lean occurs the most. The mold analysis was done for the standard mold used by anchor glass and then with the newly designed USF mo ld. Results obtained from the analysis are then compared on the basis of the temperatur e distributions obtaine d from the numerical simulation using ANSYS.

PAGE 93

808.1 Problem Definition As shown in the Figure 8.1 the following section was analyzed to simulate for the mold conditions around the shoulder region. Figure 8.1 Mold Cross Section at the Shoulder The glass parison at a temp erature of 1700 F (apprx.) co mes in contact with the mold at 800F just before the mold closes. Immediately after that the parison mold is blown to its final shape and cooling air flow through the cooling channels. This entire process last for almost 3 seconds, after which the bottle is taken out of the mold. Here we have made use of ANSYS to simulate this situation. Cooling Channels Convective BC

PAGE 94

81 8.2 Results of the Numerical Simulation The temperature distribution of the gla ss and the mold for the new design as obtained from ANSYS post processor are plotted in Figure 8.2 through 8.4 Figure 8.2 Temperature Dist ribution of Glass at the Neck Section (Degree F) Figure 8.3 Temperature Dist ribution of Mold at the Neck Section (Degree F)

PAGE 95

82 Figure 8.4 Temperature Distribution of Glass at the Shoulder Cross Section (Degree F) Figure 8.5 Temperature Distributi on of Mold at the Shoulder Cross Section (Degree F)

PAGE 96

83 The temperature distribution of the glass and the mold for the existing anchor mold as obtained from ANSYS post proce ssor are plotted in fig 8.6 through 8.9 Figure 8.6 Temperature Dist ribution of Glass at the Neck Section (Degree F) Figure 8.7 Temperature Dist ribution of Mold at the Neck Section (Degree F)

PAGE 97

84 Figure 8.8 Temperature Distributi on of Glass at the Shoulder Cross Section (Degree F) Figure 8.9 Temperature Distributi on of Mold at the Shoulder Cross Section (Degree F)

PAGE 98

85 8.3 Temperature Distribution Data The results above can be tabu late in to compare the average temperatures of the glass and the mold at the neck and the shoul der of the standard anchor mold and the newly designed USF mold. Table 8.1 Comparison of Temperatures between the Anchor Mold and USF Mold Temperature of Mold at Neck Sectio n Temperature of Mold at Shoulder Anchor Mold USF Mold Anchor Mold USF Mold Min Max Min Max Min Max Min Max 684.334 911.259 565.796835.252684.806893.080604.803 811.981 Table 8.2 Comparison of Temperatures be tween Bottles Extracted From the Anchor Mold and USF Mold 8.4 Conclusions From the tables above we can easily conc lude that the newly designed USF molds always run at a lower temperature than then existing Anchor molds. Temperature of Glass at Neck Sectio n Temperature of Glass at Shoulder Anchor Mold USF Mold Anchor Mold USF Mold Min Max Min Max Min Max Min Max 907.987 1177 813.9191082 888.9051065 831.593 1019

PAGE 99

86 The temperature of the glass coming out of the new USF molds is always less than the glass coming out of the existing Anchor mold. The temperature distribution profile on the glass as well as on the mold in the newly designed mold is almost similar to the existing Anchor molds, hence we can be sure that the lower temperature pr ofiles will lead us to a better shaped bottle. The difference in temperatures between the inner and the outer row of holes is small showing that the outer rows of holes have been effectively used for cooling. The radial pattern of temperature prof iles suggest that uni form cooling is obtained. The difference between the minimum and the maximum temperatures between the glass extracted from USF molds is always less then that extracted from the Anchor molds suggesting that the reheat th at occurs after the glass comes out of the mold will be less in the new molds as compared to the old ones thus achieving better shaped bottles.

PAGE 100

87CHAPTER 9 EXPERIMENTAL RESULTS As per the results of the MATHCAD code and the numerical simulation solution obtained with ANSYS the new molds were manufactured. The test on the new designs was conducted on 11 February, 2004. Three test molds were put on a bank for testing to avoid complications of cooling air supply. 9.1 Measurement of Bottle Temperatures The bottle temperature was measured at the dead plate and at the light box the former being closer to the mold. Temperature measurements were taken with an infrared thermal imaging camera. The camera is capable of reading thermal profiles over a portion of the bottle and recording thes e for individual pixe ls within image or averaging readings over a particular portion of the bottle. All th e readings taken were then averaged and a bar graph of the same is shown in Figure 9.1. Three different tests of the finned mold were conducted. In each test the finned mold passage (USF Mold) represents the centre line in the test group and was evaluated with cooling times set at 60-175 degrees which is oper ated by a cam that makes 360 degrees in 4 seconds. After the test USF 1 th e relative positions of the two molds were interchanged so as to achieve more uniform cooling between the molds,

PAGE 101

88 1160 1180 1200 1220 1240 1260 1280 1300 USF 1USF 2USF Std Std Mould 30-195 deg USF Mould 60-175 deg Std Mould 60-175 deg this test is labeled USF 2. The data labeled USF Std, a standard mold was placed in between the two test mold within a single blank. In each grouping c control is shown as th e left line. The control data was taken using a standard mold set with the standard cooling time of 30-195 degrees. On the right hand side of each group is another control samp le, a set of standard molds with a cooling time of 60-175 degrees, same as that used in the test mold. Figure 9.1 Bottle Shoulder Temperatures (Degree F) Measured at Dead Plate Figure 9.1 represents the read ings taken in the shoulder region of the bottle at the dead plate and Figure 9.2 represents the readings taken in the neck region of the bottle at the dead plate respectively. A similar set of temperatures were measured at the light box and results are shown in Figures 9.3 and 9. 4, for measurements taken at the shoulder

PAGE 102

89 1250 1260 1270 1280 1290 1300 1310 130013001300 Std Mould 30-195 deg USF Mould 60-175 deg Std Mould 60-175 deg region and the neck region, respectively The in itial arrangement used in the USF 1 set of runs resulted in uneven temperatures, the m easured shoulder temperatures tending to be low while the temperatures at the neck appeared high. USF 2 test data indicates that the finned Mold design, set at 60 to 175 degrees, consistently achieves between 30 and 75% of the additional cooling effect as when the standard Mold cooling time is increased from 60-175 degrees to 30-195 degrees. Figure 9.2 Bottle Neck Temperatures (Deg ree F) Measured at the Dead Plate 9.2 Bottle Lean Data Analysis Bottle lean was measured at both the shoul der and the neck on 25 to 40 bottles for each of the mold configurations. The shoulde r data was taken at a position 4.434 inches from the top of the bottle; the neck data was taken at 0.626 inches from the top. The

PAGE 103

90 overall height of the bottle is 9 1/16th inches. The data has been plotted so that the shoulder lean can be compared to the neck lean as a means of establishing where the problem had occurred. If the le an occurred at or near the ba se, we would expect that the shoulder and neck lean would be in proportion to the distan ce measured from the base. As measured from the base the shoulder and neck positions were calculated to be at 4.6285 and 8.4375 inches, respecitvely If the bottl e were to lean from the base, the neck would lean to a greater angle, the ratio being 1.823. If th e lean occurred higher in the bottle, we expect that the neck lean might be considerably greater than s houlder lean. Figure 9.3 Measured Lean (inc hes) on Control Sample, 30-195o Air Circulation In Figure 9 we see a plot of this lean da ta for the control sample, a standard mold with air blowing between 30 and 195 degrees. A heavy dashed line indicates a locus of Lean on Control Mold 30-195 degree0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 00.020.040.060.080.10.12 Lean at ShoulderLean at Neck

PAGE 104

91 points where the lean ratio is 2:1. For data above this line we assume that the lean occurred close to the neck, below the line towa rd the base. It may be observed that the data lie in two distinct groupings, but with the preponderance of data toward the upper grouping indicating that the primary cooling prob lem is in the neck region. After testing Figure 9.4 Measured Lean (inc hes) on Control Sample, 60-175o Air Circulation the Control molds with full air circulation, they were also tested with the reduced air circulation associated with the prototype molds. Measured lean for this test is shown in Figure 11. Patterns appear similar to those w ith full air flow, again showing two distinct regions of neck lean and base lean. Lean data for the prototype molds is shown in Figures 9.5-9.9. These show a progression of reduced air circulation with initial air flows of between 50 and 195o gradually being reduced to between 70 and 175o. Lean Tabulation of Control Mold 60-175 degrees0 0.05 0.1 0.15 0.2 0.25 00.020.040.060.080.10.120.14 Lean at ShoulderLean at Neck

PAGE 105

92 Lean Tabulation for Finned Mold 70-195 deg0 0.05 0.1 0.15 0.2 0.25 00.020.040.060.080.10.120.140.16 Lean at ShoulderLean at Neck Lean Tabulation For Finned Mold 50-195 degree 0 0.05 0.1 0.15 0.2 0.25 00.020.040.060.080.10.120.140.16 Lean at ShoulderLean at Neck Figure 9.5 Measured Lean (inc hes) on Prototype Mold, 50-195o Air Circulation Figure 9.6 Measured Lean (inc hes) on Prototype Mold, 70-195o Air Circulation

PAGE 106

93 Figure 9.7 Measured Lean (inc hes) on Prototype Mold, 60-175o Air Circulation Figure 9.8 Measured Lean (inc hes) on Prototype Mold, 60-175o Air Circulation Lean Tabulation For Finned Mold 60-175 degree0 0.05 0.1 0.15 0.2 0.25 00.020.040.060.080.10.120.140.16 Lean at ShoulderLean at Neck Mold Order: 1F2M3B Lean Tabulation for Finned Mold 60-175 deg0 0.05 0.1 0.15 0.2 0.25 00.020.040.060.080.10.120.140.16 Lean at ShoulderLean at Neck Mold Order: 3F2M1B

PAGE 107

94 Figure 9.9 Measured Lean (inc hes) on Prototype Mold, 70-175o Air Circulation The red line shown on the graph indicates this ratio of the distance between the neck to the base and the shoulde r to the base. Data falling on this line would be expected to represent bottles leaning from the base. Da ta to the lower right of the line represents bottles in which the neck leans disporportiona tely, an indication that the location of the slumping is above the base. Data lying to the upper left of the curve are thought to indicate cases in which the base has shifted so as to cause the shoulder to have a disporportuante lean. Lean Tabulation For Finned Mold 70-175 deg0 0.05 0.1 0.15 0.2 0.25 0.3 00.020.040.060.080.10.120.140.160.18 Lean at ShoulderLean at Neck

PAGE 108

959.3 Mold Temperature Measurements During the mold test infr ared measurements of the mo ld temperatures were also taken. These have been used to verify th e calculated temperature improvements and serve to indicate the degree of progress made in improved mold cooling. Typical comparisons are shown in Figures 17 and 18, indicating the operational mold internal temperatures for the standard mold and the prototype, respectiv ely. The measured average temperatures across the cavity surface are 674 and 488oF, respectively, even given that the prototype mold is being operated with minus 60o cooling. Figure 9.10 Standard Mold Thermal Image during Tests

PAGE 109

96 Figure 9.11 Prototype Mold Surface Image during Tests Figures 9.12 and 9.13 indicate temperatures on the outside surface of the standard and prototype molds. The prot otype does not include any horizontal slots so that it clearly stands out on the right. Some care should be exercised in viewing these images in that the hot air exiting from these slots may heat the mold in this area providing a false, high reading. What may be more informative is to look at the smooth regions away from these slits. The prototype mold exhibits a strong blue-purple tone as compared to the very red surface on a standard mold. Temperatures record ed around the lug are 622 and 574oF, respectively.

PAGE 110

97 Figure 9.12 Standard Mold Surface Image during Tests Figure 9.13 Prototype Mold Surface Image during Tests

PAGE 111

989.4 Conclusions Current data indicate that the prototype molds have been successful in providing substantial improvements in mold cooling rates for the side portions. In each of the bottle lean plots it is observed that an increased proportion of the lean has shifted toward the base of the bottl e. This resulted due to the fact that the base of the mold was left unaltered in this new design and hence, was unable to provide sufficient cooling w ith reduced cooling times.

PAGE 112

99CHAPTER 10 CONCLUSIONS AND RECOMMENDATIONS 10.1 Conclusions Increasing the surface area for convection in creases the overall energy transfer by over 35.3%. The increase in heat transf er rates increases the temper ature of air flowing through the cooling passages. Pressure loss in the air cooling passage due to a drop in the dens ity of air which is now at a higher temperature, restricting the air flow. The reduction in air flow is 7% hence the existing air blowers sh all be capable of handling the new cooling sy stem without any trouble. The temperature distribution obtained by ANSYS analysis and the actual experimental results generally indicates similar tends. Exact comparisons of result are hampered by uncertainties in the emissivity of the finished glass and molds. This leads to uncertainties when converting measured radiant fluxes into equivalent temperatures, hampering ef forts to make direct temperature comparisons.

PAGE 113

9910.2 Recommendations As a recommendation for future analysis a 3D model with air flow should be developed to get a better understanding of the actual pro cess. Finer mesh should be applied in the ANSYS finite element anal ysis. The present work was done on ANSYS Research version, having a limited number of nodes and elements’, hence developing a finer mesh for the analysis was not possible. For future work it is hence recommended to use the full version of ANSYS which would gr eatly eliminate the difficulties encountered in meshing the complex geometry with the current version, hence leading to more accurate results. The analysis in this thesis is a basic step in this complex finite element analysis. The temperature distribution data calculated are estimates due to software limitation. When the problem is analyzed in the full version of ANSYS, a more accurate data will be collected and deeper understanding of the temperature distribution will be possible. Furthermore, due to software limitations it was not possible to simulate the actual process as it takes place. In this analysis only the thermal part was solved, the flow was not take into consideration. A more accurate analysis can be done with the full version leading to simulations more clos e to the real life problem. The analysis is hampered by ill defined initial condition for the glass coming into the mold. We know the starting temperature of glass leaving the furn ace, but the cooling process which occurs as the glass gob is transported to the parison mold is not well defined. Additional cooling occurs as the gob enters the parison mold and begins to conform to its new shape. The plunger whic h performs the internal cavity and the top finish mold also provide a poorly defined te mperature transient. After the gob comes in

PAGE 114

100 contact with the plunger in the parison mo ld the temperature of glass is no longer uniform. The glass also undergoes some extra cooling while it is transferred from the parison mold to the final mold which is not accounted for, it is hence suggested that a detailed analysis should be done with all these factors in mind. The temperature of the glass can also be reduced by bringing in th e plunger at a lower temperature causing the glass parison temperature to drop. This aspe ct should be addressed in further analysis. In this case we have studied the straig ht fin for enhancement of cooling other configurations of cooling shoul d also be studied for further analysis. The analysis should go beyond the point where the bottle comes out of the mold. Once the bottle is out of the mold it has a large temperature gradient initiating a rehea ting phenomenon which weakens the bottle causing undesirable shape ch anges which are not accounted for in the current analysis. Hence, the bottle shoul d be analyzed for further temperature distributions after coming out of the mold. It is also not certain at this point if we are at the optimum temperature in between the two molding operations and the possibility of cooling of the glass as it transfers between the two molds should be analyzed.

PAGE 115

103REFERENCES 1. A.E. Kurtz “Apparatus for Uniform Cooling Of Glass-Molding Machines”, US Patent 3536469, 1967 2. John K Martin, “Method for Blow Mold ing and Cooling Hollow Glassware”, US Patent 4339258, 1979 3. Constantine W. Kulig, “Mold Cooli ng Arrangement for Use in Glassware Forming Machine”, EP 0242197, 1986 4. Roger Erb and Robert Johnson “M old Cooling Arrangement for Use in Forming Machine”, US Patent 4983203, 1990 5. Thomas V. Foster “Method of Cooling a Mold”, CA 1212234, 1982 6. Thomas V. Foster “Mold Cooling Arrangement for a Glassware Forming Machine”, US Patent 4502879, 1984 7. Thomas V. Foster, “Mold Portion wi th Cooling Means for Use in Molding Molten Glass”, US Patent 4657574, 1986 8. Rolando Cantu-Garcia “System and Method for the Cooling of Hot Molds”, US Patent 4668269,1986 9. Dan Haynes “Molds Mold Assembly for the Glass Articles”, EP 0819654, 1997 10. Stanley Peter Jones, “ Cooling Arrange ment for Mold of a Glassware Forming Machine of the Individual Section Type”, EP 0153801,1985 11. Richard T Kirkman “Mold Cooli ng Apparatus for a Glassware Forming Machine”, EP 0612699, 1994 12. Charles Trevor Lawrence “Apparatu s and Method for Cooling a Mold”, EP 0576745, 1992 13. Millard Jones “Fluid Cooling Of Glass Molds”, US Patent 4142884, 1977

PAGE 116

103 14. Julius J Torok “Mold with Exterior Heat Conducting Elements”, US Patent 4313751, 1981 15. Richard Alan Letellier “Blow Mold Cooling”, EP 0121346, 1984 16. Stanley Jones, “Cooling Molds Used in Forming Glassware Containers”, 1982 17. Stanley Peter Jones “Mold Cooling”, GB 2315746, 1986 18. Guillermo Cavazos and M. De Cervantes “Method and Apparatus for Mold Cooling”, US Patent 4824461,1988 19. Alfredo Martinez-Soto, “ Mold Cooli ng System for the Manufacture of Glass Articles or Similar Materials”, US Patent 4940480, 1989 20. Hanns Stinnes and Martin Strasse “Glass Forming Mold”, US Patent 3224860, 1962 21. Stanley P. Jones, “Controlling the Temp erature of a Glass Mold”, US Patent 4519827, 1984 22. Stanley P. Jones, “Controlling th e Temperature of a “Mold”, EP 0121335, 1984 23. Frank A. Fenton, “Mold Arrangement for a Cyclically Operating Glassware Container Manufacturing Machine with Temperature Sensing Means”, US Patent 4526604, 1984 24. Hermann Heinrich Nabelung “C ooling Articles of Newly Molded Glassware”, EP 0149890, 1984 25. Wilbur Orland Doud, “Glassware Mo lding Machine with Unitary Axis Molding”, US Patent 468050, 1987 26. Robert S. Johnson & Robert D. Ha ll, “Glass Container Forming Machine Including Neck Ring Mold Cooling”, US Patent 5358542, 1992 27. John E. Cook, “Apparatus for Forming Glass Articles With Treating Mean”, US Patent 3445219, 1965 28. Kays, W.M and London, A.L, “Compact Heat Exchangers. New York: McGraw Hill, 1997 29. Incropera, F.P., and Dewitt, D.P., “Int roduction to Heat Transfer”. New York: John Wiley, 1996