A Practical Approach to Scientific Molding 1569906866, 9781569906866

This hands-on guide helps today’s plastic molding technician to implement a scientific molding program. It shows how the

324 61 11MB

English Pages 190 [192] Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

A Practical Approach to Scientific Molding
 1569906866, 9781569906866

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Gary F. Schiller

A Practical Approach to

Scientific Molding

Schiller A Practical Approach to Scientific Molding

Gary F. Schiller

A Practical Approach to Scientific Molding

Hanser Publishers, Munich

Hanser Publications, Cincinnati

The Author: Gary F. Schiller, Erie, PA

Distributed in the Americas by: Hanser Publications 6915 Valley Avenue, Cincinnati, Ohio 45244-3029, USA Fax: (513) 527-8801 Phone: (513) 527-8977 www.hanserpublications.com Distributed in all other countries by: Carl Hanser Verlag Postfach 86 04 20, 81631 München, Germany Fax: +49 (89) 98 48 09 www.hanser-fachbuch.de The use of general descriptive names, trademarks, etc., in this publication, even if the former are not especially identified, is not to be taken as a sign that such names, as understood by the Trade Marks and Merchandise Marks Act, may accordingly be used freely by anyone. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal ­responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. The final determination of the suitability of any information for the use contemplated for a given application remains the sole responsibility of the user.

Cataloging-in-Publication Data is on file with the Library of Congress

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying or by any information storage and retrieval system, without permission in writing from the publisher. © Carl Hanser Verlag, Munich 2018 Editor: Dr. Mark Smith Production Management: Jörg Strohbach Coverconcept: Marc Müller-Bremer, www.rebranding.de, München Coverdesign: Stephan Rönigk Typesetting: Kösel Media GmbH, Krugzell, Germany Printed and bound by Druckerei Hubert & Co GmbH und Co KG BuchPartner, Göttingen Printed in Germany ISBN: 978-1-56990-686-6 E-Book ISBN: 978-1-56990-687-3

Preface

This book is designed to help today’s plastic molding technician deal with pro­ cessing issues found day to day in the injection molding environment. It not only describes the functions of the molding machine, but also the auxiliary equipment associated with the process to produce quality parts. The chapters in this book will help the user to have a more thorough and hands-on understanding of the molding machine and the material. It explains the process from the plastics point of view, and how the material is heated, flowed, packed, and cooled to produce the desired quality parts. This processing guide not only shows users how to find a solution to the problem but also lets them understand why they are making the change, and what effect it has on the plastic. It details solutions from a hot runner/cold runner standpoint. Each material has a different characteristic and will present problems in different ways, but through learning to read the part and analyzing the machine, the necessary insight will be provided to remedy most issues seen in everyday molding. The most important thing to remember when processing or making adjustments to any machine is to make just one adjustment, review the effects on the part, and if that change has no effect, return to the previous set point, before implementing another change. By making a lot of changes in the hope of solving the molding ­issue, it becomes unclear which change had the effect on the part. Look at the parts, watch the molding machine, and observe what each change is doing to the process and machine. Never neglect the details: ƒƒWalk around the machine and make sure the water is on to all lines going to the mold, or have any water lines been left off? Is the machine functioning properly (pressures, times, heating, with no unusual noises)? ƒƒMake sure the mold is functioning as intended and able to produce the quality parts desired. ƒƒObserve the material: make sure it is free of contaminants (dirt, foreign resin, or water) and is dried properly.

VI

Preface

ƒƒThen review the process and make sure there are no shortcomings (process is not pressure limited, transfer position is being achieved, not timing out and cycle time achieved). There are no magic solutions for eliminating all molding issues, but a solid understanding of these scientific molding principles will help eliminate the unnecessary waste and scrap generated from not knowing. There are three major components to the injection molding process: the injection unit, the clamping unit, and the mold. In the next chapters, we will discuss the different functions of each major com­ ponent and how they affect the process and conditions of the material. I would like to acknowledge and thank the following companies and people: ƒƒRJG Inc. Traverse City, MI, especially Gary Chastain, Pat Mosley, and Shane Vandekerkhof. ƒƒAIM Institute, Erie, PA, especially John Beaumont and Dave Hoffman. ƒƒTechnimark LLC, Asheboro, NC, especially Brad Wellington and Bruce Winslow. ƒƒMilacron LLC, Batavia, OH, especially Kent Royer. ƒƒI would also like to thank Gary Mitchell. Gary Schiller

About the Author

ƒƒ37 years in the plastics industry ƒƒCertified Master Molder I, II, & Train the Trainer; past RJG instructor with over 27 years of scientific molding experience ƒƒAIM Institute graduate and alumnus – Plastics Technology and Engineering ƒƒAIM Institute Advisory Board ƒƒPractical Rheology in Injection Molding – Penn State, Erie, PA ƒƒDesign of Experiments & Quality Engineering Methods – University of Colorado ƒƒTQM – Front Range Community College, Denver, Colorado ƒƒCertified Mechanical Inspector ASQ ƒƒCertified Quality Technician ASQ ƒƒProcessing expert with a wide array of plastics Core Competencies ƒƒStack molding ƒƒCube technology molding ƒƒTwo shot molding ƒƒInsert molding ƒƒHigh cavitation molding ƒƒEngineering and commodity resins

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V

About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII 1

Injection Unit: Screw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

1.1 Prepares the Melt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Flows the Melt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Pressurizes the Melt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.4 Sections of the Screw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.4.1 Feed Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.4.2 Transition or Compression Zone . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.4.3 Metering Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.5 L/D or Length/Diameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.6 Compression Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.7 Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.8 Injection Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.9 Injection High Limit Fill Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.10 Injection Pack Pressure/Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.11 Injection Hold Pressure/Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.12 Non-return Valve Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.13 Different Styles of Non-return Valves . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.14 Decompression/Pull Back/Suck Back . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.15 Screw Rotate Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.16 Mixing Head on a Reciprocating Screw . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.17 Barrier Screws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

X

Contents

2

Injection Unit: Barrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.1 Barrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2 Thermocouples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3 Heater Bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.4 Spacing of Heater Bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.5 Wattage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.6 Worn Barrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.7 Feed Throat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.8 Venting of the Barrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.9 Hopper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.10 Hopper Dryer Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.11 Filter Packs/Dispersion Disks/Screen Packs . . . . . . . . . . . . . . . . . . . . . 24

3

Clamping Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.1 Hydraulic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.2 Toggle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.3 Weakness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.4 Tie-Bar-Less . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.5 Single Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.6 Platen Wrap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.7 Mold Coverage Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.8 Cleanliness of the Platens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.9 Care of Bolt Holes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.10 Proper Bolt Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.11 Weight of Mold Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.12 Mold Height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.13 Calculating Clamp Tonnage for a Press . . . . . . . . . . . . . . . . . . . . . . . . . . 35

4

Ejectors/Controllers, Human Machine Interface (HMI) . . . . . . 37

4.1 Ejector Pattern and Spacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.2 Ejector Spacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.3 Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 4.3.1 Open Loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 4.3.2 Closed Loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 4.4 Key Pads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Contents

5

Machine Performance Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

5.1 Rear Barrel Zone Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5.2 Load Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 5.2.1 Purge Disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.3 Pressure Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.4 Dynamic Non-return Valve Test (FILL) . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.5 Static Non-return Valve Test (PACK/HOLD) . . . . . . . . . . . . . . . . . . . . . . 49 5.6 Injection Speed Linearity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

6

Process Development Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

6.1 Tonnage Calculation/Projected Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 6.2 On Machine Rheology Curve (Viscosity Curve) or Fill Time Study . . . 56 6.3 Construction of Viscosity Curve Graph . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6.4 Least Pressure Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 6.5 Plastic Flow Rate (Qp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 6.6 Shear Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 6.7 Gate Freeze, Gate Seal, or Gate Stabilization . . . . . . . . . . . . . . . . . . . . . . 64 6.8 Runner Weight Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 6.9 Range Finding for Gate Seal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 6.10 Manifold Imbalance and Balance of Fill Analysis . . . . . . . . . . . . . . . . . . 68 6.11 Cooling Optimization Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.12 Pressure Loss Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

7

Plastic Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

7.1 Molecular Structure of Common Materials . . . . . . . . . . . . . . . . . . . . . . . 73 7.2 Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 7.2.1 Amorphous Resin Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 7.2.2 Semi-Crystalline Resin Morphology . . . . . . . . . . . . . . . . . . . . . . 74 7.3 Glass Transition Temperature (Tg) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 7.4 Melt Transition Temperature (Tm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 7.5 Shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 7.5.1 Isotropic Shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 7.5.2 Anisotropic Shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 7.6 Melt Density versus Solid Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 7.7 Advantages/Disadvantages of Hot Runner versus Cold Runner . . . . . . 79 7.8 Induced Shear through a Hot or Cold Runner System . . . . . . . . . . . . . . 80

XI

XII

Contents

8

Plastic Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

8.1 Fountain Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 8.2 Flow of Plastic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 8.3 How to Calculate Flow Rate (Qp) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 8.4 Calculating Volume of Shot Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 8.5 Blocking a Cavity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 8.6 Flow through a Mold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 8.7 Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 8.8 Transfer/Cut-Off Position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 8.9 Viscosity Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 8.10 Intensifying Ratio (Ri) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 8.11 Pressure Limited Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 8.12 Safe Start-Up Shot Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 8.13 Runner Sizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

9

Plastic Pressure (Pack/Hold) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

9.1 Plastic Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 9.2 Dynamic versus Static . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 9.3 Viscosity Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 9.4 End of Cavity Pressure Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 9.5 Part Shrinkage versus Cavity Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . 94 9.6 Maximum Average Pressure at Parting Line before Flashing . . . . . . . . 95

10 Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 10.1 Plastic Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 10.2 Turbulent versus Laminar Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 10.3 Reynolds Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 10.4 Water Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 10.5 Area of Water Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 10.6 Series/Parallel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 10.7 Cooling Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 10.8 Ineffective Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 10.9 Cooling Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 10.10 Depth, Diameter, and Pitch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

Contents

10.11 Baffles/Bubblers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 10.12 How a Thermolator/Mold Heater Works . . . . . . . . . . . . . . . . . . . . . . . . . 108

11 Benchmarking the Injection Molding Process . . . . . . . . . . . . . . 111 12 Process Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 12.1 Black Specks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 12.2 Blush . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 12.3 Brittleness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 12.4 Burns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 12.5 Burns in Gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 12.6 Cloudy Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 12.7 Color Streaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 12.8 Deformation: Ejector Pin Marks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 12.9 Degraded Polymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 12.10 Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 12.11 Fish Hooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 12.12 Flash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 12.13 Flow Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 12.14 Hot Tip Drool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 12.15 Jetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 12.16 Long Gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 12.17 Nozzle Drool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 12.18 Parts Sticking in Mold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 12.19 Pulls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 12.20 Shorts/Non-Fills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 12.21 Sinks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 12.22 Splay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 12.23 Sprue Sticking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 12.24 Surface Imperfections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 12.25 Voids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 12.26 Warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 12.27 Weld Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

XIII

XIV

Contents

13 What is Important on a Set-Up Sheet? . . . . . . . . . . . . . . . . . . . . . . 145 14 Commonly Used Conversion Factors and Formulas . . . . . . . . . 149 14.1 Conversion Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 14.2 Common Formulas for Injection Molding . . . . . . . . . . . . . . . . . . . . . . . . 150

15 Machine Set-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 16 Things That Hurt the Bottom Line of a Company . . . . . . . . . . . . 157 17 Terms and Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 18 Reference List for Further Courses and Reading . . . . . . . . . . . . 173 18.1 Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 18.2 Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

1

Injection Unit: Screw

In this chapter we will discuss the components and functions of the injection unit, and how each play a role in the preparation of the plastic.

Figure 1.1 Injection unit

„„1.1 Prepares the Melt There is mechanical heating, caused by the friction or shear inside the barrel, from the plastic pellets being rubbed against the barrel wall and compressed inside the flights of the screw. There is electrical heating, from the heater bands on the barrel. They are used from a cold start to heat up the barrel and plastic. After a proper amount of soak time (30 minutes), start to rotate the screw. The barrel heater bands are to maintain the temperature in the barrel so the plastic does not hit any cold spots. Once the barrel is up to heat, start to extrude plastic through the barrel. About 80% of the heat comes from the shearing process and 20% from the electrical portion. In Figure 1.2 you can see the shaded sections representing the different sections of the screw.

2 1 Injection Unit: Screw

Metering

Transion

Feed

Figure 1.2 Reciprocating screw

In Figure 1.3 it is shown how the plastic in each section has a circular motion inside the flight. There is a melt pool on the back side so that as the screw rotates the melt pool pushes the unmelted pellets forward and up against the barrel wall. As the unmelted pellets rub against the barrel wall it creates friction, and that friction causes the pellet to melt and go into the melt pool. Metering

Transion



Feed



Figure 1.3 Melting of the plastic in different sections of the screw. Courtesy of AIM Institute

„„1.2 Flows the Melt There is a hydraulic unit and valves that provide the oil flow and pressure needed to inject the plastic. The injection velocity set point will give and maintain the speed of the ram coming forward and it must have ample pressure and flow to push the plastic. To ensure the injection high limit pressure set point is never reached (pressure limited) the valve is either restricted or opened depending on the feedback it receives from the linear transducer on what velocity or injection speed is desired. Also understand the influence the injection velocity has on the rheological properties of the material: plastics typically show non-Newtonian behavior, which means the faster the material is shot or the faster the flow rate of the plastic, the thinner the material becomes and the easier it will flow.

1.4 Sections of the Screw

„„1.3 Pressurizes the Melt The non-return valve (check ring) is what pressurizes the melt. It creates a seal on the inside of the barrel through the use of a sliding check ring, ball-check screw tip, and/or poppet check ring. There are also plunger-style screws that inject the plastic into the molds: there are no moving parts to this design. And as the plastic is pushed forward the non-return valve seals off, not allowing any plastic to return behind it. If it does there is either a worn non-return valve or possible wear in the barrel. This will be discussed later in the book.

„„1.4 Sections of the Screw There are many different screw styles available today, with a multitude of mate­ rials available. The reciprocating screw provides the function of conveying the ­material, and compressing and heating it to prepare it for the next shot.

Metering

Transion

Feed

Figure 1.4 Sections of the screw

Figure 1.5 Root diameter changes in each sections of the screw

1.4.1 Feed Zone The feed zone is used to convey the material from the feed throat and start the compaction process in the barrel. This section starts to compress the pellets within the flight and starts the friction process as the material rubs along the barrel wall as the screw rotates. Screws can have long or short feed sections depending on the material being run. Longer feed sections could be for shear sensitive materials or a material that melts easily with a low melt temperature.

3

4 1 Injection Unit: Screw

1.4.2 Transition or Compression Zone This is where the flight depth starts to get shallower. The material starts to receive greater compression and the friction or shearing of the material increases, ­contributing to the melting of the plastic. This is where most of the work is done on heating the material (see Figure 1.5).

1.4.3 Metering Zone This zone has the shallowest flight depth. By the time the material gets to this point it should be melted, and ready to be conveyed past the non-return valve to position itself in front of the screw building the next shot.

„„1.5 L/D or Length/Diameter Length (L) is measured from the front of the screw to the end of the flights. Dia­ meter (D) is measured from the highest point on the flight of the screw to the corresponding other side (see Figure 1.6). Keep in mind the value of L/D for the screw in the press: too short of an L/D results on non-melted pellets, while too long of an L/D and the result is too much residence time, which can burn or degrade the plastic.

Figure 1.6 Where to measure length and diameter of screw

„„1.6 Compression Ratio This refers to the depth of the feed section flight (Figure 1.8) divided by the depth of the metering section flight (Figure 1.7). If there is a 3 : 1 compression ratio screw this means that the depth of the feed section flight is three times the depth of the metering section flight. The measurement is taken from the root of the screw to the top of the flight.

1.6 Compression Ratio

Figure 1.7 Measuring flight depth in metering section

Figure 1.8 Measuring flight depth in feed section

Example: The depth of the feed section is 0.450″ and the depth of the metering section is 0.150″. It is expressed as 0.450″ divided by 0.150″ = 3 or 3 : 1 compression ratio. 

Compression ratios for materials: ƒƒLow compression screws range from 1.5 : 1 to 2.5 : 1, and are for shear sensitive materials ƒƒMedium compression screws range from 2.5 : 1 to 3 : 1, and are for general-purpose materials ƒƒHigh compression screws range from 3 : 1 to 5 : 1, and are for crystalline materials One way of determining whether the compression ratio is correct for the material is to check if the standard cycle creates black streaks or non-melts in the parts. If one of these two conditions exist, then the machine can have the wrong compression ratio screw for the application.

5

6 1 Injection Unit: Screw

„„1.7 Profile The profile of the screw refers to the number of flights in each section of the screw (see Figure 1.9). Some screws will have a profile of 10-5-5, which means that there are 5 flights in the metering section, 5 flights in the transition section, and 10 flights in the feed section. This would be representative of a general-purpose screw

5

4

5

4

10

13

Figure 1.9 Different profiles of a screw

The 13-4-4 profile would possibly be used for a shear sensitive material, with the long feed section not allowing the material to heat up or be under compression or shear for as long with the shorter compression or transition zone.

„„1.8 Injection Pressure Injection pressure is also known as boost pressure, fill pressure, or injection 1st stage pressure. Its purpose is to provide enough pressure (abundant pressure) so the process does not become pressure limited, which limits the filling velocity. If the fill pressure limits the velocity, the shear rates of the material can vary (when pressure limited you will not see the same velocity or flow rate in the plastic due to  pressure controlling the filling phase and not velocity, which can reduce the temperature from under-shearing of the plastic; differental temperatures can cause differental shrinkage). When establishing the injection pressure, start with abundant pressure and establish a 95% full part. After running the relative viscosity curve test (please see Chapter 6 for test and procedure) and selecting an optimum velocity, start coming down on the fill pressure until it affects the fill time. When the fill time starts to increase, that is when pressure is affecting the fill rate. The set point should be 150–250 psi above transfer pressure to account for any viscosity variation that might be seen during normal operation.

1.11 Injection Hold Pressure/Time

„„1.9 Injection High Limit Fill Time The injection high limit fill time is a safety feature added to the machine to protect the tool and process. This timer should be set just above the actual fill time of the press. It provides the protection that if the transfer position is not met, then the timer will engage and transfer the press. This timer provides the safety needed so that if one of the cavities blocks off in a multi-cavity tool, the high limit timer takes over and transfers the machine to the lower pack/hold pressure. For example, if the molding machine has a constant fill time of 0.74 seconds, set the timer 0.1 seconds above to 0.84 seconds. Another purpose is if there is a viscosity change and the fill time starts to increase, then this timer will alarm out and notify that something has changed in the process.

„„1.10 Injection Pack Pressure/Time Injection pack pressure is used to complete the filling process and imprint the plastic to the cavity surface. This pressure is used to pack all the material in that is needed to achieve gate seal and hold dimensional tolerances, in a two-step process (fill and pack). This pressure is usually lower than the fill pressure and depending on part geometry and wall thickness, this part of the process is time dependent and will require enough time to complete the process.

„„1.11 Injection Hold Pressure/Time Injection hold pressure is used to hold the material that was injected into the part; this phase requires just enough pressure so the screw does not move backwards. If the screw moves backwards then the plastic that was put into the cavity is now starting to push back out because of the internal cavity pressure and cause dimensional variation, and a reduction in pack around the gate area. Watch the cushion! (The cushion is the amount of plastic left in front of the screw at the end of fillpack-and-hold, and should never move backwards during any of these phases.) This part of the process is time dependent and will require enough time to complete the process.

7

8 1 Injection Unit: Screw

It is all about the pack rate of the material and part, and sometimes you do not want an actual gate seal because the flow front has stopped moving furthest from the gate and still packing at the gate causes differential pack rates or differential shrink rates, which can cause dimensional variation.

„„1.12 Non-return Valve Function The function of the non-return valve is to allow plastic in front of the screw while the screw is rotating back (see Figure 1.10, Figure 1.11, and Figure 1.13) but to seal off during injection. Even though various non-return valves are designed differently, they serve the same function. There is a clearance between the outside diameter of the non-return valve and the inside diameter of the barrel. this clearance is generally 0.003″ to 0.005″ per side but can vary depending on manu­ facturer; review the specifications for individual presses. This clearance allows for thermal expansion of the steel as it heats. When this gap increases or wear happens, leakage over the non-return valve will increase. When this happens, the non-return valve not only loses the ability to pressurize properly; there is a loss of the cushion of material left in front of the screw at the end of pack and hold. When the material is squeezed between the non-return valve and barrel wall, the material will shear heat. When material shear heats because of the leakage, there are two different viscosities of material, and therefore there will be two different shrink rates from two different cooling rates in the material.

Figure 1.10 Flow through a ball non-return valve (cut-away)

Figure 1.11 Flow through a sliding (3pc) non-return valve (cut-away)

1.13 Different Styles of Non-return Valves

Figure 1.12 Flow along a smear tip or plunger tip

Figure 1.13 Flow through a poppet tip (cut away)

The smear tip (Figure 1.12) is a low shear tip, where the plastic is extruded in front of the screw and there are no moving components; this type of screw tip is used for shear sensitive materials such as PVC.

„„1.13 Different Styles of Non-return Valves

Figure 1.14 Ball check non-return valve

Figure 1.15 Sliding check ring (3 Pc)

9

10 1 Injection Unit: Screw

„„1.14 Decompression/Pull Back/Suck Back Decompression is used to relieve the pressure off the front of the screw, which in turn relieves the pressure off the hot manifold or a cold runner. When the screw sucks back it will create a vacuum effect inside the barrel and draw the melt backwards, creating a void in the nozzle. This will allow the pressure in the manifold system to relax into that void and take pressure off of the tip area, which helps avoid gate or nozzle drool. There are two different decompressions phases: Pre-decompression is used in conjunction with a nozzle shut-off and to decompress the manifold before the nozzle shuts off completely and screw starts to rotate, again creating a void area for the pressurized plastic from the manifold to relax into. Post decompression is used if there is no nozzle shut-off and suck back of the melt is needed to keep the material from drooling, out of either the nozzle or gates. Post decompression can and should be used whether there is a nozzle shut off or not. This will create a slight gap in front of the screw. As the screw injects forward for the next cycle it gives room for the non-return valve to move back and seal properly. The minimum distance to suck back or decompress is the movement of the sliding ring or the ball. As a reference we add 0.250″ or ¼″ as a starting point, but ­sometimes more decompression is required depending on how much the melt is compressed within the system. This movement or decompression will also help with shutting off the non-return valve upon injection or forward movement because it creates a small gap and as the screw moves forward the ring and/or ball will have time to seat or shut off before the plastic starts moving.

„„1.15 Screw Rotate Delay A minimum of 0.5 second screw delay is needed before the machine starts to rotate. When the machine is done with the pack/hold phase of the cycle, the check ring seat and the check ring are pressed against each other tightly; if the screw rotates with no delay, the first movement is metal on metal, which can wear the check ring out prematurely. If a screw rotate delay time is added after pack/hold, there is a chance for relaxation before rotation starts, minimizing the wear to the seal area.

1.17 Barrier Screws

„„1.16 Mixing Head on a Reciprocating Screw The mixing screw allows the material to knead and mix better after the metering section. The mixing head is located between the non-return valve and the metering section. By the time the material travels through the metering section the pellets should be melted. This final phase allows the material to cross over the thin section cut into the mixing head (Figure 1.16) into the next channel, which will continue to move the heated material forward in front of the screw. This will also help with ­un-melted pellets getting into the shot or manifold system, by not allowing them to cross over into the exit channel. One thing to watch out for in the mixing screw is the possibility to generate more shear than normal. The purpose of the mixing screw is to mix and knead the material to get it to blend, but this can also generate extra shear in the process, which can change the color. Thus, darkening can result from burning of the pigment.

Figure 1.16 Typical mixing head on the end of a reciprocating screw. Photo courtesy of ­Technimark LLC

„„1.17 Barrier Screws The barrier screw is a screw style that helps prevent most non-melted materials from entering the melt. With the extra flight created on the screw it allows the melt to move forward while trapping the un-melted material and giving it more of a chance to melt completely. The second flight is lower than the main flight, only allowing the melted material to cross over (see Figure 1.17). In the case of Figure 1.17, the barrier screw manufacturer recommends a reverse heat profile but this is also material specific and in some cases a regular heat profile is needed.

11

12 1 Injection Unit: Screw

Consult the material and screw manufacturer to obtain the proper heat profile for the material being used.

Figure 1.17 Barrier screw configuration. Photo courtesy of Technimark LLC

2

Injection Unit: Barrel

„„2.1 Barrel The barrel is a steel chamber that houses a reciprocating screw or plunger screw that feeds material. It is designed to handle the high pressures associated with injection molding. There are holes drilled into the barrel for placement of the thermocouples (T/C’s), and a larger opening towards the back end of the barrel known as the feed throat.

Thermocouple holes

Feed throat

Figure 2.1 Drawing of barrel with thermocouple hole locations

Figure 2.2 Typical heater band and thermocouple placement. Photo courtesy of Technimark LLC

14 2 Injection Unit: Barrel

„„2.2 Thermocouples Thermocouples provide feedback to the HMI controller; they sense the temperature of the steel in that section of the barrel. Looking at the barrel on the injection unit, it can be noticed that the thermocouple is located between the two heater bands in that zone. This is to reduce the chance of over- or under-heating in a certain zone. Care must be taken at the thermocouple holes to avoid debris and contamination as this can cause a false reading. Avoid spilt material and make sure the T/C touches the bottom of the hole without interference.

Figure 2.3 Spade-type thermocouple. Photo Courtesy of Technimark LLC

Figure 2.4 90 degree bend. Photo Courtesy of Technimark LLC

2.2 Thermocouples

Thermocouples are two dissimilar materials or metals that have a predictive voltage at a specific temperature when combined. See Figure 2.11 for type and color coding. The two wires are fused or soldered together at the very end. This is the contact point (Figure 2.5). If the thermocouple wires are separated or broken away at the true reading location, the new touch area is where the reading will come from (Figure 2.6). No matter where the wires touch, that is where the voltage for reading is generated, and if it is in the wrong place, a false reading of temperature is received.

Figure 2.5 Proper T/C reading

Figure 2.6 Broken or pinched T/C wire

Figure 2.7 J type T/C plug

15

16 2 Injection Unit: Barrel

Figure 2.8 K type T/C plug

Figure 2.9 Color coding J type T/C

Reading the chart above (Figure 2.11), we can choose the USA J type thermo­couple, for example. Notice that the positive wire is white and the negative wire is red, and the coating or jacket is black (Figure 2.9).

2.3 Heater Bands

Figure 2.10 Color coding K type T/C

Figure 2.11 Thermocouple color code chart for wires and jacket

„„2.3 Heater Bands Heater bands are made up of a wire or filament and usually encased in a protective coating of mica. Heater bands provide the heat source to the barrel, to bring the temperature up to the set point and maintain it.

17

18 2 Injection Unit: Barrel

Figure 2.12 Nozzle heater band. Photo courtesy of Technimark LLC

Figure 2.13 Barrel heater bands. Photo courtesy of Technimark LLC

The heater band must be tightened completely to provide good contact. If a heater band is loose and does not receive a good contact point all the way around, or the heater band is too small, it can start to overheat in that area from the air gap, in which case the band will start to run away and turn cherry red, and have a shorter life expectancy. If the heater band is not working properly or not working at all, take a piece of plastic and touch it to the heater band (be careful: this is a safety concern and a heat-protection glove must be worn; also, this can generally only be accomplished on older machines where the heater bands are exposed, as newer machines normally will have the barrel encased in guarding). The heater band should start to melt the plastic. If the plastic does not start to melt and seems cold and sticking/ sliding across the heater band, then the band is faulty and needs replaced. When replacing, verify the new band is the same size, wattage, and voltage!

2.5 Wattage

„„2.4 Spacing of Heater Bands By observing the heater bands on an injection unit, notice a gap or space in between each heater band; this will allow the extra heat to dissipate. When the heater bands are pushed tightly together, meaning no gap in between, the barrel will overheat or cause the heaters not to cycle properly. This must be monitored closely when replacing a bad heater band.

Figure 2.14 Spacing of heater bands to allow dissipation of heat. Photo courtesy of ­Technimark LLC

„„2.5 Wattage To work out the wattage (power), what needs to be known are the amps (current) and the volts (voltage) in the power source. Most machines run off a 200 and something voltage system (208 V, 220 V, or 240 V); you must know what the voltage is to plug into the formula watts = amps × volts. Example: We want to know the wattage drawn. If there are 4.5 amps and 220 volts the equation is written as W = A × V or 4.5 × 220 = 990 watts. If there are 1.5 amps and a power source of 110 volts again W = A × V or 1.5 × 110 = 165 watts. Or if we wanted to know how many amps the controller should be drawing the formula is A = W/V. If there is a 990 watt heater band and 220 V, A = W/V or 990/220 = 4.5 amps. 

19

20 2 Injection Unit: Barrel

Figure 2.15 Watts–volts–amps triangle

„„2.6 Worn Barrel A worn barrel can be the result of constantly running the same shot size. The wear can come from fillers (glass, talc, calcium carbonate, also semi-crystalline materials). A worn barrel can cause cushion inconsistencies, poor recovery of the screw, contamination, or dark streaks from the material leaking back over the non-return valve and causing excessive shear. It can also leave a thicker skin in the barrel, making degradation causing dark streaks more likely. In describing a worn barrel, a common term used is “bell mouthed”, caused by a wash out of the steel from the constant friction in this area (Figure 2.16). It can also happen in the middle of the barrel where the transition or compression zone is; this is where most of the friction arises (Figure 2.16). On a new screw and barrel, the gap or distance between the non-return valve and the barrel wall should be around 0.003″ per side. Some manufacturers will have up to 0.005″ clearance (consult the screw and barrel manufacture for the proper clearance). One way to verify that the front of the barrel is not worn is to increase the shot size to a new section of the barrel (remember that whenever the shot size is moved up, the transfer ­position must be moved up, in order to still have a 95% full part).

2.8 Venting of the Barrel

Wash out condion

Figure 2.16 End of barrel washed out from constant use in that area

„„2.7 Feed Throat The feed throat is one of the most overlooked sections on the barrel. Not only is it used for preparing the material before it goes into the barrel, meaning that if the feed throat temperature is too low it can attract moisture (understand the relative humidity within the plant), but also if too hot, a hot ball or clump can be created in the feed throat. The recommended temperature of the feed throat is between 105 °F and 150 °F or as close to the drying temperature of the material as possible. The other way to look at it is this: hot enough not to get condensation and cold enough not to get a hot ball. Most importantly, never run chiller water on the feed throat because of the potential to accumulate moisture. Most molding shops run tower water on their feed throats, which can vary in temperature from day to night and summer to winter. Try to create a closed loop control system for the feed throat to maintain a constant temperature for optimization.

„„2.8 Venting of the Barrel One reason to run a warmer feed throat is the fact that it is also a vent for the barrel. As the material enters the barrel and starts the compaction and heating process, the air and volatiles must go somewhere, and that location is the feed throat. When a colder than normal feed throat is run, the volatiles and additives being released from the material will form a skin inside the barrel. These gases and volatiles will harden when they contact with the cold section around the feed throat. This skin will prevent the volatiles/air from escaping and now become trapped in the melt, causing bubbles or other defects in parts (Figure 2.17).

21

22 2 Injection Unit: Barrel

Volales forming seal

Figure 2.17 Additive forming skin when they get close to feed throat. From AIM Institute PTE Certificate Program

The additives have a shorter molecular chain, which means they will vaporize first and break away from the longer molecular chains. The easily melted additives and gases from the material want an escape route and the feed throat provides that feature. If the inside of the barrel has a film or seal in it, the gases have nowhere to escape to and will stay in the melt stream causing defects.

„„2.9 Hopper The purpose of a hopper is to provide a continuous flow of material to the feed throat without interruptions, dead spots, or hang-ups. Funnel flow hoppers have the tendency to flow down through the center of the hopper, which can cause the material to hang along the sides. This can be problematic in materials that need to be dried due to over- and under-drying of the materials. It can also result in poor color mixing, and poor part performance if blending materials. Hoppers with funnel flow have an angle at the bottom of less than 60°. This will promote the funnel flow effect.

Angle 30 degrees

Figure 2.18 Funnel flow designed hopper. From John Bozzelli AIM Institute PTE Certificate Program

2.10 Hopper Dryer Diagram

Mass flow hoppers have the material flow evenly down the hopper and allow for more even drying and distribution to reduce the problem of uneven mixture of lots of material, blended lots, or color mixtures. The whole object is to have a “first in, first out” process, and not have any material hang up in dead spots in the hopper.

Angle must be 60 degrees for mass flow

Figure 2.19 Mass flow designed hopper. From John Bozzelli AIM Institute PTE Certificate ­Program

„„2.10 Hopper Dryer Diagram

heater

Figure 2.20 Typical material dryer diagram

Hygroscopic material draws moisture inside the plastic pellet. Hydrophilic material draws moisture to the surface.

23

24 2 Injection Unit: Barrel

If the material is hygroscopic, meaning that it absorbs water, it must be dried before processing, and a dryer unit will perform that function. Things that are necessary for a dryer to function properly are as follows: 1. Heat source; must have dry air to be conveyed to the hopper. 2. Volume of air; must have a blower that pushes the air through the hopper at the proper flow rate (CFM) 3. Residence time; the material must stay in the dryer long enough to be dried properly 4. Moisture level/dew point; the dew point must be at −40 °F to supply the dry air, but the end result will be the moisture level in the material. And this number will be material-specific; most materials will range between 0.01% and 0.05% moisture content 5. The return air must be at 150 °F or lower, or otherwise drying efficiency decreases Dryers are the most overlooked piece of equipment on the molding press. There are different sized hoses that are taped together to try to convey the material, which can cause contamination if the tape gets sucked up into the hopper. The filters on the air return do not get cleaned on a regular schedule, which can inhibit air flow and drying capability. Also, if the material fines get into the desiccant bed, it can destroy the desiccant, and then desiccant beds will have to be ­replaced. If the heated air delivery line to the hopper or the return line has a kink or crushed section in it, it needs to be replaced.

„„2.11 Filter Packs/Dispersion Disks/Screen Packs This is secondary equipment that can be added to the nozzle end of the injection unit, and is used to filter out contaminants such as non-melted pellets, metal, etc (see Figure 2.21 and Figure 2.22). A pressure increase will occur, as it will take more pressure to push through smaller holes (Figure 2.21) or if the surface area of all the holes added up is smaller than the orifice of the nozzle opening. When ­installing this equipment, the material is filtering through very small orifices or channels which can be smaller than the gates. This helps protect the gates from clogging.

2.11 Filter Packs/Dispersion Disks/Screen Packs

This filter will go in between the nozzle p and the nozzle body, and can be easily removed and cleaned. There is an increase in transfer pressure, because it will take more pressure to push through the smaller holes.

Figure 2.21 Filter disk Linear edge melt filter goes into a nozzle body specifically designed for this applica on. The material enters into a flow channel and has to cross over into the next flow channel to con nue to move forward. The cross-over point is the filter, which can be purchased in different depths, depending on the applica on.

Figure 2.22 In-line filter that goes in nozzle body

Please see Plastic Process Equipment (www.ppe.com) catalog for styles.

25

3

Clamping Unit

The two illustrations below (Figure 3.1 and Figure 3.2) show how different clamp styles, hydraulic and toggle, support the molds, and what effect each can have on the process. The one thing to remember, whether using hydraulic or toggle, is that the front half platen will always be the weakest point and create issues while molding. The reason the front half is the weakest is the large hole where the locating ring and the nozzle must enter the platen.

„„3.1 Hydraulic A hydraulic clamp will support the mold in the center of the back half (Figure 3.1). Because of where the force is concentrated, the outsides can become weak and need to be reviewed with a larger mold that extends close to or past the outer edge of the tie bars. Hydraulic clamps are limited on stroke due to the stroke being dependent on how long is the shaft on the cylinder: the larger the mold the smaller the opening stroke. Clamp tonnage on a hydraulic press is not as influenced by mold temperature, but it will need mold touch position adjustment once operating temperature is reached. And clamp to whatever tonnage is programmed in.

Hydraulic clamp cylinder

Figure 3.1 Hydraulic clamp

Tie bars Nozzle opening in platen

28 3 Clamping Unit

„„3.2 Toggle A toggle press employs mechanical advantage by using a smaller cylinder for speed pushing on the center of the knuckle to lock over the clamp, but also supports ­differently to a hydraulic clamp (Figure 3.2). The toggle press supports the outer edges of the platen, leaving the center to become weak. Clamp tonnage on a toggle press is highly influenced by mold temperature. Once the mold is heated there will be thermal expansion, and on a toggle press, the tonnage is set to a certain position; once that position has changed, the machine will either under-clamp if mold temperature goes down, or not lock over if the temperature goes up.

Smaller hydraulic cylinder to provide speed and use a mechanical advantage

Tie bars Nozzle opening in platen

Figure 3.2 Toggle clamp

„„3.3 Weakness The weakest point on any machine will be the front half center regardless of whether a toggle or hydraulic clamp. The reason for this is there is a huge hole in the front half center where the barrel must go to get the nozzle to the sprue. This large hole creates a weakness. Anything outside the tie-bar area will be weaker.

„„3.4 Tie-Bar-Less A tie-bar-less machine is exactly as it sounds: there are no tie bars for alignment, and the back-half platens usually ride on linear rails to help with guidance. It has stiffer platens to help with the clamping of the mold and building tonnage without losing force. Tie-bar-less machines are only in the lower tonnage category. The plus

3.5 Single Point

side is larger molds can be installed in the press because there are no tie-bars to get in the way.

„„3.5 Single Point To single point a press, there must be four indicators. Take the indicators and apply one to each of the four tie bars. All indicators must be zeroed. Also needed is a frame to put the indicators on so they act independently from the machine. Then, clamp the mold up and take the reading off each indicator (watch how many times the indicator spins). They must all spin the same number of revolutions and the readings must be within 0.002″–0.003″ of each other. This way, all four corners of the mold are clamping up equally.

Figure 3.3 Placement of indicators on tie-bar ends for single point

The dial indicators must be set on an external frame independent from the press to get a true reading. If the indicators are attached to the press, the movement can influence the reading. The indicators must be zeroed before the press is to be locked over and understand the stretch of each tie bar. This test is performed when one corner of the mold is flashing more easily than the rest of the mold, or if there is some type of wear on the mold located in one of the four corners. Bluing or pressure film can be applied to see if there is a potential problem.

29

30 3 Clamping Unit

„„3.6 Platen Wrap Platen wrap is where the platens want to wrap around the mold because the mold is too small. Remember that the force is generated at the tie bars and as they squeeze together the corners will pull together more than the center. This creates a bow in the platen, and it will be different for toggle versus hydraulic (Figure 3.4 and Figure 3.5). Reducing tonnage can help with some flashing issues, by allowing the platens to straighten out and relax instead of wrapping. This can provide the extra support needed. See below how the molds are supported differently for each style press.

Figure 3.4 Platen wrap (toggle press)

Figure 3.5 Platen wrap (hydraulic press)

3.8 Cleanliness of the Platens

„„3.7 Mold Coverage Area The mold must cover at least two-thirds of the space inside the tie bars to minimize the platen wrap that is experienced on smaller molds (Figure 3.6; reference machine manufacturers’ recommendations). When the mold is pulled out of the press and the platen and the mold face that sits up against the platen are rusty, and this build-up flakes off, this is an indication that some platen wrap is happening. If the platens wrap or deflect, it allows moisture to get in behind the mold and rust starts to form. If there is little to no rust then the mold is sitting tight up against the platen and not allowing the moisture to collect in between the two pieces of steel. Making sure when cleaning the platen that an oil base spray is used, and when wiping it down, leaving a small film of oil to prevent rust from forming.

 Figure 3.6 Mold coverage area inside tie-bars

„„3.8 Cleanliness of the Platens The platens must be cleaned and maintained after every mold change. Wipe down with some form of oil base spray (to prevent rust) such as WD-40, and use a scrub pad and a stone if needed.

31

32 3 Clamping Unit

Figure 3.7 Platen cleanliness is critical

Remove any high spots that can hold the mold away from the platen and possibly cause damage from being out of alignment. Also check the mold for any rust build-up that must be removed before setting. These small steps will help prevent mold damage or premature wear.

„„3.9 Care of Bolt Holes If there is damage done to the bolt hole it MUST be cleaned out. Use a tap to re-tap the hole and remove any burrs that are present. Using the right size tap and thread pitch are extremely important. Remember that some manufacturers will have bolt holes in a standard thread and others with a metric thread pattern. Make sure to understand which is the correct thread pattern on the platens. Check for burrs on the hex head or socket head that could cause someone to cut their hand. The bolts should tighten up to hand tight to the clamp being installed. If not, there is something wrong with the bolt or bolt hole: find issue and fix.

3.10 Proper Bolt Location

„„3.10 Proper Bolt Location Wrong locaon

Bolts should be placed in these bolt hole locaons for best clamping force, as close to the mold as possible. Be careful that the water line is not in the way

Figure 3.8 Proper bolt location for maximum strength

Notice in this picture (Figure 3.8) that the bolt is placed on the back end of the clamp; this is when the clamp will lose clamping force on the front end. Remember that this a fulcrum point, and the bolt should be as close to the mold as possible to achieve the greatest pressure or force. It should be torqued down to a specification, not as tight as somebody can get them.

Figure 3.9 Bolt location and effects on strength

Make sure the clamp is as level as possible (Figure 3.10) to get the most clamping force or (surface contact area). Figure 3.11 (toe’d in) and Figure 3.12 (toe’d out) show improper ways to place a clamp in the mold slot; there is much less force due to the decrease in surface area to clamp against.

Figure 3.10 Level

33

34 3 Clamping Unit

Figure 3.11 Toe’d in

Figure 3.12 Toe’d out

„„3.11 Weight of Mold Calculations

Figure 3.13 Calculate mold weight in pounds and kilograms

3.13 Calculating Clamp Tonnage for a Press

The calculation above is for regular tool steel (H-13). This will change if you have aluminum, stainless steel, or beryllium copper. It all depends on the density of the material, which is measured in g/cm3 or lbs/in3. The formula will however always be the same: length × width × height × whatever the tool steel density. When using lbs/in3 you the get the weight in pounds (lbs), whereas using g/cm3 will give the weight in grams (g) then divide by 1000 to get kilograms. The harder the metal the poorer the thermoconductivity will be or the harder the metal poorer the heat transfer will be. Densities of Some Tool Materials Metal

Density in g/cm3

Density in lbs/in3

Aluminum

2.80

0.101

Beryllium copper (BeCu)

8.25

0.298

Stainless steel

7.88

0.285

Steel (H-13)

7.85

0.28

„„3.12 Mold Height The mold height is set differently on a hydraulic clamp to a toggle clamp. When setting a new mold, make sure that the mold touch position is taken down to zero so it will not be made during the closing of the back-half platen. If not, when the machine sees that position (mold touch), it will try to close at full pressure and could possibly cause damage to the mold or the press. On a toggle press, run the die height forward or backward until the press will not lock over any more. The main thing to know is what is the projected area of the parts and runner to set tonnage properly. Just because the press is a 300-, 400-, or 650-ton press does not mean that is where to set the tonnage. Excessive tonnage will not only damage a mold but will cause the press to wear prematurely.

„„3.13 Calculating Clamp Tonnage for a Press The following formulas are needed to calculate clamp tonnage for a press: Area of cylinder = D × D × 0.7854 Clamp force in pounds = area of cylinder × pressure available to clamp cylinder Clamp force in tons = clamp force in pounds ÷ 2000

35

36 3 Clamping Unit

Figure 3.14 Pascal’s equation

Hydraulic Machine Example: Clamp cylinder is 29″. Clamp pressure is 1900 psi. 1. What is the area of the clamp cylinder? 2. What is the clamp force in pounds? 3. How many tons is this machine? Clamp cylinder is 29″ so take 29 × 29 × 0.7854 = 660.52 in2. Clamp force in lbs: (area × psi) 660.52 × 1900 = 1,254,988 lbs. Clamp force tons: clamp force lbs ÷ 2000 = 627 tons. 

Toggle Machine Example: Let’s say you have a 500-ton press. Clamp pressure is 2000 psi. CP/T = psi for 1 ton. Machines can only go so low on pressure and be repeatable, usually 50% of max pressure. On a toggle press, it is pressure ÷ tons, so 2000 ÷ 500 = 4, or every 4 psi equals 1 ton. 

4

Ejectors/Controllers, Human Machine Interface (HMI)

„„4.1 Ejector Pattern and Spacing The ejection system on an injection molding press plays a significant role in the removal of parts from the mold. There are many different forms of part ejection. There is ejection from the press by which ejector bars run from the ejector plate through the platen and into the back of mold, tying into the ejector plate. Hydraulic cylinders are put on the mold to provide this ejector function by moving a stripper plate forward and back. There is also a mechanical function ejection system that operates with the movement of the mold.

„„4.2 Ejector Spacing Figure 4.1 and Figure 4.2 show the PIA ejector patterns (PIA is the Plastics Industry Association, formerly SPI—Society of Plastic Industry). Figure 4.3 and Figure 4.4 show two European patterns.

38 4 Ejectors/Controllers, Human Machine Interface (HMI)

Blue Paern is a 7” x 7”

Red paern is a 4”x 16” paern

Yellow Paern is a 6” x 28” paern

Figure 4.1 PIA ejector pattern spacing

Figure 4.2 Ejector pattern spacing and bolt pattern spacing (PIA pattern) 450 ton maxima. The dimensions are given in mm and inches. Used by permission of Milacron®

4.2 Ejector Spacing

Figure 4.3 This bolt pattern is a European pattern (in mm). Used by permission of Milacron®

Figure 4.4 European ejector pattern (Ferromatik 160t) (in mm). Used by permission of ­Milacron®

39

40 4 Ejectors/Controllers, Human Machine Interface (HMI)

„„4.3 Controllers Regarding controllers, there are two types: open loop and closed loop. The controllers integrate and control all functions of the press, including moving oil from the tank to a directional valve, moving a hydraulic cylinder, or have the screw move at a certain speed during injection. This controller will not only monitor position, pressure, speed, and temperature, but having a basic understanding of how both open loop and closed loop controllers work and how each will affect the plastic while processing will have a great effect on part quality.

4.3.1 Open Loop An open loop controller allows the use of valves but cannot provide feedback to the control unit. If an injection velocity is programmed into the machine, it will provide the oil to create flow, but will not provide feedback to maintain the velocity programmed into the controller; it will only try to achieve the set point and will overshoot or undershoot that programmed velocity.

4.3.2 Closed Loop A closed loop controller will provide constant feedback to maintain the programmed set point that is programmed into the controller. A thermocouple will provide information to the controller to maintain temperature, and this tells the heater to turn on or turn off to maintain the set point. A linear transducer will provide feedback to the controller to monitor velocity. The linear transducer will tell the controller how far the screw has traveled in a certain amount of time and ­adjust the oil flow accordingly to maintain the programmed set point. All newer presses are closed loop controllers and provide feedback. Some machines will give the option of controlling the open/closed loop control on the press. Our job is to understand the difference between the two and what benefit each has. If you are wanting to use the scientific molding principles, then using the closed loop feature will provide the best feedback.

4.4 Key Pads

„„4.4 Key Pads There are universal symbols on the touch pads on the machine; just familiarizing yourself with the symbols will allow processing on any machine possible. Figure 4.5 shows some common key pad function buttons.

1

2

3 1 Mold Open 2 Mold Close 3 Ejector Back

4

5 4 Ejector Forward 5 Nozzle Forward 6 Nozzle Back

6

7 7 Screw Forward 8 Screw Back 9 Screw Rotate

Figure 4.5 Key pad function buttons on an injection molding machine

8

9

41

5

Machine Performance Testing

„„5.1 Rear Barrel Zone Optimization The purpose of this test is to establish the optimum temperature for the rear barrel zone. This is an easy Excel chart that can show what is the optimized temperature for the fastest recovery. Use the recovery time to create the chart and label the X and Y axes correctly. In the example shown in Figure 5.1 below, notice that 480 °F and 6.6 seconds recovery time is the optimized temperature with the lowest recovery time. First make sure that everything is functioning correctly, from water on the feed throat to all the heater bands working correctly. Also review the barrel temperatures according to the material manufacturer’s recommendations. This test will take a while to perform due to the raising of the temperature and the need for a stabilization point. It is better to start at the low end of the temperature and work up. But it will be worth it in the long run due to a decrease in screw ­recovery time and decreased cycle time. Example: First create the Excel spreadsheet with two data columns: (1) rear barrel temperature and (2) recovery time, as in Table 5.1. Then plot the recovery time to create the chart, such as in Figure 5.1. Table 5.1 Table for Optimizing Rear Barrel Temperatures Test

Rear Barrel Temperature

#1

440 °F

Recovery Time 7.9 sec

#2

450 °F

7.5 sec

#3

460 °F

7.2 sec

#4

470 °F

6.8 sec

#5

480 °F

6.6 sec

#6

490 °F

6.8 sec

#7

500 °F

7.0 sec

44 5 Machine Performance Testing

Table 5.1 Table for Optimizing Rear Barrel Temperatures (continued) Test

Rear Barrel Temperature

Recovery Time

#8

510 °F

7.3 sec

#9

520 °F

7.5 sec

#10

530 °F

7.7 sec

Figure 5.1 Graph for optimizing rear barrel temperatures

„„5.2 Load Sensitivity The load sensitivity test will allow the examination of the press, to find out how it will respond under a load or how it will handle and compensate when the plastic is being injected into the mold. Think of it like cruise control: if you are going 55 mph on a flat surface (doing an air shot) and then you start to go up a hill (shooting into a mold), the car should still do 55 mph. This test compares fill times in the mold versus fill time in the air, and will also use transfer pressures of shooting into the mold versus shooting into the air. A purge disk will be required (see Figure 5.3). Procedure: 1. Set machine to run a standard 95%–98% filled process 2. This should be done with optimum velocity chosen from a rheology curve 3. Turn off pack/hold time and pressure (some machines may require putting in some minimum value to function properly)

5.2 Load Sensitivity

4. Make a shot into the mold and record the fill time and transfer pressure (hyd or ppsi) on cycle 5. Back injection unit off and install purge disk in locating ring (Figure 5.3), and bring injection unit forward as quickly as possible (remember the longer the material sits in the barrel the lower the viscosity becomes). Reset nozzle forward position or switch 6. Put press into semi-automatic and cycle press; the plastic will now be diverted and shoot through channel (remember, if machine tries to rotate to achieve ­recovery, there is no resistance to build a shot and could make a large purge). Record fill time and pressure (hyd or ppsi) 7. Insert times and pressures into formulas, and calculate to get results (see Figure 5.2) × 100 = ÷ (

)

= =

Acceptable range +/- 4%:

Figure 5.2 Calculation formula for load sensitivity. Used by permission of RJG Inc.

When using this formula, insert the proper value for the pressure. When documenting hydraulic pressure divide by 1000 in the second part of the formula, whereas if documenting plastic pressure then instead insert 10,000 into the formula.

5.2.1 Purge Disk The purge disk must be made from aluminum so that, when installed between the mold and the nozzle, no damage occurs when the nozzle moves forward to inject a shot. Cut a ½″ radius for the nozzle to sit against. If using a ¾″ radius nozzle tip then make a purge disk to accommodate it. In Figure 5.3, the purge channel is shown as 0.500″ deep. It can be any depth; it just important not to restrict the flow when doing an air shot. There must be a clean cut in the flow channel so the material will not stick. Install a magnet on the back side to help hold and align it to the mold.

45

46 5 Machine Performance Testing

For the diameter, a value of 3.990″ is given here. Most locating ring holes in the platen are 4.000″ so make the purge disk just under that so it does not stick in the locating ring hole. Some machines have a larger locating ring hole and that is where the magnet comes in handy.

Figure 5.3 Drawings for a purge disk

„„5.3 Pressure Response Pressure response is a test that is done to find out when the pressure stabilizes after machine shifts from transfers to the pack phase.

The amount of me the pressure takes to stabilize



Figure 5.4 Pressure response

The importance of this test is to understand when the pressure drops below the pack pressure setting. The melt front has a chance to hesitate, and if this happens then it starts to freeze, which could cause visual and dimensional defects along with variation in process.

5.3 Pressure Response

When the pressure dips down below set pack/hold pressure it can cause melt front to hesitate

Figure 5.5 Melt front hesitation caused by pressure drop

If pressure is dipping down below set point, then it is necessary to set a profiled pack phase where a higher pressure is added for a couple tenths of a second as the first pack profile. This will help stabilize the pressure that is coming down from transfer to pack pressure set point.

No hesitaon with profiled pack pressure

Figure 5.6 Pressure drop

Let’s say there is a transfer pressure of 1850 psi and a pack pressure of 800 psi. To resolve the dip in pressure or screw bounce, add a pressure of 1200–1400 psi for 0.2 seconds. This will help with the transition of pressure and keep the melt front moving forward. Experiment with the time and pressure to get the right combination; this will ­resolve the hesitation that was noticed previously.

47

48 5 Machine Performance Testing

„„5.4 Dynamic Non-return Valve Test (FILL) 1. Set machine to run 95%–98% fill process 2. Turn off pack/hold time and pressure 3. Increase the cooling time to compensate for the change in cycle time 4. Make 10 fill only shots, weigh the parts and runner(s), and record 5. Calculate the percentage change Remember, the reason to weigh the runner with the parts is to determine the % change in the fill only shot, and how the non-return valve is working and shutting off. The acceptable limit is less than 3%. If the % change is higher than 3% then add some extra decompression or suck back to the machine and run the test over again. Adding the extra suck back or decompression will allow more room for the non-­ return valve to seat shut-off properly. Heaviest shot − Lightest shot ×100 = %of change  Average shot weight Acceptable range =< 3% Example: If our heaviest shot is 110.45 gm, our lightest shot is 109.21 gm, and the average shot weight is 110.01 gm, the equation would be as follows: 110.45 − 109.21 = 1.24 then 1.24 divided by 110.01 = 0.01127 × 100 = 1.13% → Acceptable But if it had more variation shot to shot, then look at it like this: 111.25 − 106.24 = 5.01 then 5.01 divided by 110.01 = 0.0455 × 100= 4.55% → Unacceptable 

If there is too much in the variation seen shot to shot, then it is confirmed that our non-return valve is not working properly during the dynamic or fill phase. The example shown in Figure 5.7 below shows how an Excel spreadsheet is constructed.

5.5 Static Non-return Valve Test (PACK/HOLD)

Figure 5.7 Data information to set up a dynamic non-return valve test

„„5.5 Static Non-return Valve Test (PACK/HOLD) 1. Set machine to run 95%–98% fill process 2. Add an additional 10 seconds over gate seal to pack/ hold time 3. Adjust screw RPM’s accordingly 4. Enter the cushion end position for five consecutive shots The reason an additional 10 seconds is added over gate seal is to positively verify that the non-return valve is shut off and sealed during pack and hold. Largest cushion − Smallest cushion ×100 = %of change  Average Cushion Acceptable range =< 3% Let’s review the example below; see that the static non-return valve test shows only a 1.64% variation

49

50 5 Machine Performance Testing

Acceptable range =< 3%

Figure 5.8 Data information to set up a static non-return valve test

Formulas Total – Add shot 1 through shot 5 and get a total Average – Take the total and divide by 5 Highest – What is the highest value out of the five shots? Lowest – What is the lowest value out of the five shots? Difference – Take the highest value and subtract the lowest value Variation – Difference divided by Average

„„5.6 Injection Speed Linearity Note: all information can be taken from relative viscosity test. 1. Set machine to run a standard 95%–98% fill process 2. Turn off pack/hold time and pressure (some machines may require a minimum set point) 3. Increase cooling time to compensate for changes in cycle time 4. Raise injection forward timer high enough so as not to be reached during testing 5. Set injection velocity to the highest setting and adjust the transfer/cut-off position to achieve a 98% part visually 6. Linear stroke (LS) is calculated from (shot size + decompression) − transfer = LS; pull back might have to be increased to provide better accuracy 7. Record the injection fill velocity set point and fill time for each setting 8. Calculate the actual injection speed (linear stroke/actual injection time). Cal­ culate the % difference from the set injection velocity to the actual injection ­velocity

51

5.6 Injection Speed Linearity

Actual injection velocity − Set injection velocity ×100 = % Difference  Set injection velocity Acceptable range: >5% (up to 10% acceptable short term only)

Figure 5.9 Injection speed linearity test. Used by permission of RJG Inc.

The blue line represents set injection velocity and the red line represents actual injection velocity. As seen, the machine is not capable of achieving the set velocity that is asked for. Example: Linear stroke = (feed/shot size + decompression) − transfer/cut-off Actual injection speed = linear stroke/actual fill time % difference =

actual injection velocity − set injection velocity ×100  set injection velocity



52 5 Machine Performance Testing

Let’s review the chart above (Figure 5.9), using line # 10 for our example: Linear stroke = 1.8″ + 0.20″ − 0.50″ = 1.5″ Actual injection speed = 1.5/0.35 = 4.286 in/sec % difference: 4.286 − 5.7 = −1.414, then divide −1.414/5.7 = −0.248 × 100 = −24.8% difference 

An Excel workbook covering machine performance testing can be downloaded from http://files.hanser.de/fachbuch/9781569906866_Workbook.zip. 

6

Process Development Test

The purpose of these tests is to optimize the process as much as possible, and to get an understanding of how each aspect will affect the part. Through this optimization process you can establish high and low limits along with the centerline process.

„„6.1 Tonnage Calculation/Projected Area Why is the surface area of the part so important to injection molding? The reason is that this is where the clamp tonnage comes from. A certain force is needed to hold the mold closed against the forces of injection pressure. And without the proper calculation, damage to the mold can occur by flashing it, destroying the parting line, and rolling over the edges. Calculating Projected Area

Figure 6.1 Geometry diagrams to calculate surface

Circle or cylinder: A = D × D × 0.7854 or pR2 When looking at the diameter of a circle or cylinder, pick the widest part of the circle or cylinder.

54 6 Process Development Test

Example: In the example shown in Figure 6.2, there is a basic washer shape. It is a circle with a hole in it. First, calculate the largest area then subtract the smaller area from it to get projected area. Calculation for a circle or cylinder: Method 1: D × D × 0.7854, or 4.500 × 4.500 × 0.7854 = 15.90 in2 Method 2: pR2 (pi = 3.14) (R = ½ the diameter) R = 4.500/2 = 2.25, then 3.14 × (2.25 × 2.25) = 3.14 × 5.063 = 15.90 in2 There are two different calculations to formulate the area of a circle. Now, for the second part of the calculation, figure out the surface area of the smaller circle: D × D × 0.7854 = 1.025 × 1.025 × 0.7854 = 0.825 in2 pR2 = 3.14 × (0.513 × 0.513) = 3.14 × 0.263 = 0.826 in2 Method 1: Surface area = 15.90 − 0.825 = 15.075 in2 Method 2: Surface area = 15.90 − 0.826 = 15.074 in2 

The difference between the first answer and the second answer is just a small rounding error; when we calculate the in2 we have a 0.001 in2 difference. The diameter of the large circle is 4.500" and the diameter of the small circle is 1.025"

Figure 6.2 Round geometry

For a trapezoid shape (Figure 6.3), see the formula below: A = H × ((W1 + W2)/2) Example: Take H = 2.450″, W1 = 1.350″, and W2 = 0.950″ First add W1 + W2 = 1.350 + 0.950 = 2.300″, then divide by 2: 2.300/2 = 1.15″ Surface area = 2.450 × 1.15 = 2.818 in2 

6.1 Tonnage Calculation/Projected Area

Figure 6.3 Trapezoid shape geometry

For a square or rectangle (Figure 6.4), see the formula below: A = W × H, or L × W Example: Take W = 2.556″ and H = 1.895″ Surface area = W × H = 2.556 × 1.895 = 4.844 in2 

Figure 6.4 Square or rectangular geometry

For a triangle (Figure 6.5), see the formula below: A = (B × H)/2 Example: Take B = 1.559″ and H = 2.865″ Surface area = (B × H)/2 = (1.559 × 2.865)/2 = 4.467/2 = 2.233 in2 

55

56 6 Process Development Test

Figure 6.5 Triangle geometry

„„6.2 On Machine Rheology Curve (Viscosity Curve) or Fill Time Study The purpose of a viscosity curve is to plot the effects of speed and pressure on the viscosity of the plastic melt. The optimum injection velocity is where speed and pressure will have minimal ­effect on the viscosity of the melt. Fill as fast as quality will allow. Using the optimum velocity obtained from this study also helps to minimize the effect of material fluctuations or viscosity changes that would happen during normal production runs in the process. How is a viscosity curve completed? Run machine and mold at a free and easy process (make parts full but not over packed) before starting the graph. This will help ensure that the tool is warmed up and that the viscosity of the material in the barrel has had a chance to stabilize. Starting with the maximum injection speed, adjust shot and transfer so that parts are 95% to 98% full. Allow the machine to stabilize for two or three shots, then record fill times and transfer pressures. Change injection speed to the second setting (always start at the fastest velocity and work down to the slowest velocity).

6.3 Construction of Viscosity Curve Graph

„„6.3 Construction of Viscosity Curve Graph Construct Excel spreadsheet like the one in Figure 6.6 below using the formulas indicated. A14 through A24 are manually entered values for selected velocities. B14 through B24 are manually entered values for intensifying ratio of injection unit. C14 through C24, D14 through D24, and E14 through E24 are manually entered values for peak hydraulic pressure at transfer. F14 formula is as follows: =AVERAGE(C14:E14). This gives an average of three  ­ different transfer pressures. This formula would be applied for F15 =­AVERAGE(C15:E15) and so on, until F24 is reached. G14 through G24, H14 through H24, and I14 through I24 are manually entered values for fill time at transfer. J14 formula is as follows: =AVERAGE(G14:I14). This gives an average of three different fill times. This formula would be applied for J15 =AVERAGE(G15:I15) and so on, until J24 is reached. K14 formula for shear rate is =1/J14 and for K15 it would be =1/J15, and so on until K24. L14 formula for relative viscosity is =B14*F14*J14 and for L15 is =B15*F15*J15, and so on until L24 is reached.

Figure 6.6 Rheology Data

Then the graph is established from the shear rate column and the relative viscosity column.

57

58 6 Process Development Test

From this curve, the optimum velocity for the mold can be established. Pick the flattest part of the curve and establish a process window as viewed in Figure 6.7 below; there are the two black arrows to mark the process window and the red ­arrow shows the optimum velocity. Establish a viscosity curve or on the press rheology curve to find out how the material exhibits shear thinning (non-Newtonian) behavior, and set up the process window on the flattest part of the curve (Figure 6.7). This ensures that, if within that window a velocity change is made, meaning that the process is speeded up or slowed down, there will be very little effect on the viscosity of the material. The red arrow is just in the center of that process window. Never take the process window to the very end of the graph (because the process should never be run at maximum velocity, with no room to go up) and try to stay away from where the curve starts to rise (meaning that the slower velocity plays a larger role on how the material is shear thinning).

Figure 6.7 Rheology curve; black arrows mark the process window and the red arrow shows the optimized velocity

The graph might say that you can fill faster, but part quality will dictate injection ­velocity.

6.3 Construction of Viscosity Curve Graph

Figure 6.8 Rheology curve breakdown

Note that there are certain materials that do not like the fast injection velocities and must be run at slower velocities or closer to the crossover point (e. g., polycarbonate, PVC). Let’s look at the viscosity curve another way: it is all about managing variation within the process, and where on the curve is the smallest variation (Figure 6.9).

Figure 6.9 Rheology curve showing variation

59

60 6 Process Development Test

„„6.4 Least Pressure Curve The least pressure curve (Figure 6.10) is set up with the transfer pressure as the inputs and is designed to pick the least amount of pressure it takes to fill the cavity while still maintaining the proper shear rate. When running a viscosity test, the fastest plastic that is shot into the mold and the slowest plastic that is shot into the mold will usually be the highest pressures; this graph just tells where is the lowest pressure at transfer compared to the proper shear rate. The lower the pressure the easier the machine is working, minimizing wear. The boom of the curve is the opmum. This is the lowest pressure required to fill the mold.

Figure 6.10 Least pressure curve

„„6.5 Plastic Flow Rate (Qp) This is the rate or speed at which the plastic flows into the mold and can be represented as Qp or flow rate of plastic. It is normally measured in cm3/sec or in3/sec. As the plastic starts to flow into the cavity, it can become faster or slower depending on the part geometry. This is due to the area that the flow front is moving into. Keeping the flow channels as large as possible and having the same wall thicknesses will help prevent the changes in flow rate.

„„6.6 Shear Rates Divide Q as needed as the plastic flows through the delivery system. Qtotal is the total volume to be injected during fill or fill time. It refers to the total amount of plastic that is injected out of the nozzle (remember this is a certain amount of material at a specific point) and is the same as that observed in the sprue, as this would be handling the same amount of plastic as being injected out of the nozzle. As the material starts to go through different sections of the runner

6.6 Shear Rates

the amount of plastic will be divided down. In the example shown in Figure 6.11, when branched from the sprue to the main runner the total is divided in half, and when branched to the secondary runners the flow divides into fourths. And as the material goes into the final branch into the gates the Qtotal is divided into eighths.

Figure 6.11 Shear rates at different sections of runner

Let’s build a spreadsheet to establish what are shear rates are at different sections of the feed system. Exercise This is what is known for this example: Eight-cavity mold; 1 gate per part Material: ABS Total volume = 11.51 cm3 Injection time range = 0.1–2.1 sec, 0.5 sec increments Gate diameter = 0.1 cm

61

62 6 Process Development Test

Table 6.1 Example Data for Relationship of Fill Time, Qtotal, Qgate, and Shear Rate. Courtesy of the AIM Institute Fill Time (sec)

Qtotal (cm3/sec)

Qgate (cm3/sec)

Shear Rate (1/sec)

0.1

115.1

14.39

146,650

0.6

 19.18

 2.4

 24,459

1.1

 10.46

 1.31

 13,350

1.6

  7.19

 0.9

  9,172

2.1

  5.48

 0.68

  6,930

Let’s go through the math for 0.1 sec: Qtotal (cm3/sec): take 11.51 (total volume) and divide by 0.1 sec (fill time), which gives 115.1 cm3/sec; this is the volume coming through the sprue. Qgate (cm3/sec): now, take the 115.1 cm3/sec and divide by 8 (there are 8 gates), which comes to 14.39 cm3/sec of volume through each gate.

Figure 6.12 Shear rate through a round channel

Shear rate (1/s): the formula needed is

4Q pR 3

(Figure 6.12).

Q is the volume at gate Pi (p) = 3.14 R (radius) is ½ the diameter = 0.1/2 = 0.05 cm, so R3 = 0.053 = 0.00125 So the shear rate is

4 ×14.39 57.56 = 146,650 (1/s) = 3.14 ×0.000125 0.0003925

Now this work is done to understand what the shear rate of the plastic is going through the gate. To see what it tells us, look at the chart in Figure 6.15; see that the ABS molecular chain will start to fracture at 50,000 1/s (reciprocal seconds). What kind of additives are in the material and how easy they potentially degrade is also to be considered. The additive molecular chain will be shorter than the polymer chain, and it is known that the shorter the chain the easier it will potentially degrade.

6.6 Shear Rates

Now let’s look at the fill time of 0.1 sec, which gives the shear rate of 146,650 1/s. The acceptable range for ABS is up to 50,000 1/s, so the shear rate of material ­going through the gate is too high and is now fracturing the molecular chain, and potentially degrading as it flows. Look at a fill time of 0.6 sec: the shear rate is only 24,459 1/s, which is well within the acceptable range. Figure 6.13 shows the formula for shear rate in a square or rectangular gate.

Figure 6.13 Shear rate through a rectangular channel

Figure 6.14 Typical runner configuration for a two-cavity bar

This is what is known for this example: Two-cavity mold; 1 gate per part Material: ABS Total volume = 0.3942 in2 Injection time = 0.32 sec Gate width = 0.080″ Thickness = 0.040″ To calculate the shear rate 6Q wt

3

=

)

6×((0.3942 / 0.32 /2) 0.080×(0.040×0.040)

=

6×0.616 3.696 = = 28,875 1/s 0.080×0.0016 0.000128

63

64 6 Process Development Test

Figure 6.15 Published basic values from Mold Flow on upper limits

„„6.7 Gate Freeze, Gate Seal, or Gate ­Stabilization A gate seal or gate freeze test is performed to understand how much plastic is packed into a mold to achieve a seal or freeze. The chart is established from the shot weight: add a column for the individual part or piece weight (Figure 6.20 Anchor) and as the Excel spreadsheet is set up, just divide the shot weight by the number of cavities. Try to achieve a true gate seal (where the weight no longer increases) to get the most stable part, but there will be exceptions to this rule; does a true gate seal actually need to be achieved? (Remember: the packing rate of the plastic will be different at the gate vs the end of fill.) The plastic always has a greater tendency to freeze the further it gets away from the gate, and if the dimension is larger at the gate than at the end of fill, not wanting to achieve a true gate seal might be an ­option, because the dimensions will be larger where the plastic is still packing. A hot tip is designed to keep the plastic hot at the tip so the plastic can flow for the next shot. This is also true when trying to achieve gate seal. The hot plastic at the gate will be trying to come out once the pressure is let off. Plastic in the cavity is like a spring: all this pressure is used to push it in and imprint to cavity, and once that pressure is released it will try to come back out.

6.7 Gate Freeze, Gate Seal, or Gate ­Stabilization

A valve gate is designed to be a positive shut-off to the plastic. The pin is forced forward to help with the visual appearance of the gate, but also to help prevent plastic leaking back out. Normally a valve gate tool will run quicker on cycle time due to the plastic being positively stopped from coming back out. But whether it is a cold runner, a hot tip mold, or a valve gate mold, performing this test to see where the plastic stops packing into the cavity or where the weight no longer increases is still the goal. The reason hold pressure and time are added is that the pack pressure/time is used to get the plastic into the cavity and imprint the cavity with the pressurized plastic. Use hold just to maintain the plastic from back flowing out. This will help maintain dimensional control over the part. When running a cold runner mold, do not weigh the runner for this graph. Weighing the runner, though, is very important to find out when the part weights stop going up at which point the runner weight should continue to rise. This tells you that the runner is not influencing the gate seal time. If the runner weight stops going up when the part weight stops increasing, the runner is too small so it is ­influencing the gate seal.

Figure 6.16 Gate freeze, seal, or stabilization graph

ƒƒCycle time must stay consistent during all runs. ƒƒCooling time must change with each change to pack time. As pack time goes up, cooling time must go down. After all runs are complete, plot results on the graph. When the part/shot weight levels out, the gate is sealed. Determine whether gate seal is required. Optional: Record the amount of mold deflection at the parting line and carriage as read with a dial indicator. Goal: To determine the amount of mold & stationary platen deflection during the pack & hold phase of the cycle.

65

66 6 Process Development Test

Tools: Two dial indicators with magnetic bases. Instruction: Install one dial indicator at the back of the injection unit to measure the amount of movement of the injection unit (Figure 6.17). When the mold ­expands from being pressurized, it will push the injection unit back.

Figure 6.17 Placement of indicator on injection unit

Put one dial indicator at the parting lines to detect mold “blown” condition. Set the dial indicator to zero with the mold clamped up just before injection. Record the maximum amount of movement of the dial indicator during injection. The most movement should be seen during the pack phase when the cavity is pressurized, when the plastic is trying to overcome clamp force and push the two halves apart.

Figure 6.18 Placement of indicator on parting line of mold to measure movement

Let’s look at the gate seal curve another way: it is all about managing variation within the process, and where on the curve the smallest variation is seen.

6.8 Runner Weight Study

Figure 6.19 Shot weight variation with time

Figure 6.20 Gate stabilization with part weight added in

Construction of the graph is simple, by plotting weight over time, showing how much will the weight go up with more pack/hold time added. The other thing to understand is what pack time intervals to use when building this chart. On thin-walled parts (less than 0.040″) increase in 0.1-second increments, on medium-walled parts (0.125 to 0.150″) 0.5-second increments, and on thick-walled parts (above 0.150″) 1-second increments.

„„6.8 Runner Weight Study The runner weight study is designed to be run in conjunction with the gate freeze or gate seal test. This test establishes whether the runner is freezing off before the cavity is truly packed out. If the runner freezes off before the cavity is full (by weight), then under packing the parts is possible, which will not only affect the shrinkage of the plastic but also affect the plastic’s cooling rate.

67

68 6 Process Development Test

„„6.9 Range Finding for Gate Seal On thick-walled parts it might be best to range find before starting this test. By this, the following is meant: Go in 5-second increments, starting at 5 seconds then going to 10 seconds, and check whether the weight goes up. If not, the gate is sealed between 5 and 10 seconds. If the weight goes up, then go to 15 seconds. If the weight does not go up, the gate seal is between 10 seconds and 15 seconds, but if the weight still goes up, then go to 20 seconds, and so on. This is the easiest way to find out where the gate is sealed at and reduces the time it takes to find the ­answer.

„„6.10 Manifold Imbalance and Balance of Fill Analysis The manifold balance graph is established from the average imbalance column, and the balance of fill analysis graph is established from the analysis of fill from mean weight column (see Figure 6.21). See the formulas below for setting up the Excel spreadsheet. The formulas on the Excel spreadsheet are as follows: Average part weight in grams: =AVERAGE(G12:I12) Fill sequence: =RANK(B12, B12:B19) Average imbalance: =(MAX(B$12:B$59)-B12)/MAX(B$12:B$59) Analysis of fill from mean weight: =((K12-B12)/K12) Cavity with most imbalance: =MAX(E12:E15) Maximum imbalance: =MAX(D12:D15) Average imbalance: =AVERAGE(D12:D15) % range drop shot 1 to 3: =(G62-I62)/G62 Average: =AVERAGE(B12:B15) Max: =MAX(G12:G15) Min: =MIN(G12:G15) Range: =G60-G61 Average: =AVERAGE(G12:G15) Pop std: =STDEVP(G12:G15)

6.10 Manifold Imbalance and Balance of Fill Analysis

Figure 6.21 Manifold balance graph for 16-cavity mold

Figure 6.1 depicts the Excel spreadsheet with corresponding numbers and columns, to show where to use the formulas and how they play a role in setting up the spreadsheet. This will allow the graphing of the balance of a mold, along with the fill sequence, and finally how the balance compares to the average weight.

69

70 6 Process Development Test

Figure 6.22 Manifold balance for an eight-cavity mold

„„6.11 Cooling Optimization Study The cooling optimization study is performed on a mold to understand the effects temperature of the plastic plays on how the part ejects off the mold. The thought process behind this test is to take the cooling time down until the parts start to deform. The heat deflection temperature or heat distortion temperature (HDT) is a test performed in a lab and is found on most material data sheets (see Table 6.2). The purpose of the test is to apply a certain temperature and load and find out where the plastic starts to deform. There are normally two different pressures used: 0.46 MPa or 66 psi and 1.8 MPa or 264 psi. This test is performed on a test bar which is normally not relevant to the part design due to different part geometries, but the HDT number on the material data sheet will give the range in which to shoot for. Understand that this test should be performed during the prototype phase of the process, but can be performed on a production mold too. Also needed is a clear understanding that if the plastic comes out of the mold hotter then more shrinkage will occur. During the prototype stage the tool has usually not been cut to the final steel dimensions (steel safe), and having a complete understanding of the shrink rate of the material in the mold will be critical to hold a consistent dimensional tolerance.

6.12 Pressure Loss Study

Once a measurement of the part or parts has been taken, giving a clear understanding of the shrink rate of the material, then the steel can be cut to the final numbers. Table 6.2 Typical HDT Temperatures for Common Resins Material

Deflection Temperature in °C Deflection Temperature in °F

ABS

 98

208

ABS + 30% glass

148

298

POM

159

318

POM + 30% glass

195

383

PMMA

 96

204

PA 6

161

321

PA 6 + 30% glass

220

428

PA 6/6

150

302

PBT

120

248

Polycarbonate

141

285

Polyethylene

 86

186

Polyethylene terephthalate (PET)  70

158

Polypropylene

 93

200

Polypropylene + 30% glass

165

329

Polystyrene

 96

204

PPS

190

374

TPE

 60

140

Also note that adding any kind of fillers, whether it is glass fiber, glass spheres, carbon, or talc, to the material will only make the HDT number higher because the filler does not melt and will remain in a solid state.

„„6.12 Pressure Loss Study When talking pressure loss, let’s establish where the pressure loss is happening. Start with the plastic pressure in front of the screw: this plastic pressure is established by the hydraulic set point on the press and the intensifying ratio of that press. This will be the plastic pressure in front of the screw. Start with 20,000 ppsi in front of the screw, and then every time the plastic moves forward, through the end cap, nozzle body, nozzle tip, runner or hot manifold, and finally through the cavity, pressure will be lost.

71

72 6 Process Development Test

Figure 6.23 shows a typical example.

Through cavity

Through runner

Through nozzle

In front of screw

Figure 6.23 Pressure loss through each phase

Start with 20,000 ppsi in front of the screw, and see how the plastic pressure left is affected. ƒƒLoss through nozzle: 3200 ppsi, which means having 16,800 ppsi available. ƒƒLoss through runner: 3800 ppsi, so now 13,000 ppsi available. ƒƒLoss through cavity: 6000ppsi, so now 7,000 ppsi available. If there was any more pressure loss, there might not be enough pressure available to fill the part and the process will become pressure limited.

7

Plastic Temperature

„„7.1 Molecular Structure of Common ­Materials Figure 7.1 shows the repeating unit molecular structure of some common resins. The polymer chains are made up of this unit repeated many times.

Polyethylene

Nylon 6

Polypropylene

Styrene

PVC

Polycarbonate

Figure 7.1 Repeating units of typical resins from amorphous and semi-crystalline families

74 7 Plastic Temperature

„„7.2 Morphology The morphology of the plastic is the study of the structure or form of the plastic and how they differ. In the thermoplastic family, this concerns whether they have a crystalline structure or not, and if they have a crystalline structure, how large are the crystals. When dealing with the thermoplastic family, there are two groups of structures. There is amorphous resin (without form) and semi-crystalline resin (with form, crystalline structure).

7.2.1 Amorphous Resin Morphology Figure 7.2 shows what happens when an amorphous resin goes through a phase change. There is no crystalline form in either cold or heated states.

Amorphous resin in a cold state -

Amorphous resin in a heated state -

Amorphous resin in a cooled state -

Figure 7.2 Phase change for amorphous resins

7.2.2 Semi-Crystalline Resin Morphology Figure 7.3 shows what happens when a semi-crystalline resin goes through a phase change.

7.4 Melt Transition Temperature (Tm)

Semi-Crystalline in a cold state -

Semi-Crystalline in a heated state -

Semi-Crystalline in a cold state -

Figure 7.3 Phase change for semi-crystalline resin

Notice that the semi-crystalline resin has a crystalline structure and the amorphous material is void of structure in a cold state, but as the two materials, amorphous and semi-crystalline, are heated up the structure becomes the same. When the molecules are in a heated state, they relax and are free to move around. In this state they are now able to be injected into the mold.

„„7.3 Glass Transition Temperature (Tg) The glass transition temperature is where the molecules are no longer free to move and the plastic is returning back to a solid state (plastic cooling) or a crystallized state for semi-crystalline resin. Or the molecules are becoming free to move (plastic heating). See Figure 7.4 and Figure 7.5 for the phase change.

„„7.4 Melt Transition Temperature (Tm) The melt transition temperature is where the molecules are in the least tangled state and are free to move, so the plastic is now flowable; see Figure 7.4 and Figure 7.5.

75

76 7 Plastic Temperature

This is the softening of the material as it heats up.

Figure 7.4 Amorphous transition phase This is the point at which the crystalline structure melts. This happens quickly.

Figure 7.5 Semi-crystalline transition phase

The amorphous resin takes longer to soften, to get from Tg to Tm, and the semi-crystalline material has a shorter heat-up time because the amorphous regions are smaller, but will require more BTU’s (heat) in the final stage to melt the crystal.

„„7.5 Shrinkage As the plastic cools the molecules want to return to their natural state. The material expands while heated, and contracts as it cools, resulting in shrinkage. Shrinkage can be anisotropic or isotropic; the difference between the two is as follows.

7.5.1 Isotropic Shrinkage With isotropic shrinkage, the material will shrink the same amount in the direction of flow versus the transverse direction of flow. This is more inherent with amorphous resins because there is no crystalline structure in the molecular chain.

7.5 Shrinkage

Figure 7.6 Isotropic shrinkage

7.5.2 Anisotropic Shrinkage Anisotropic shrinkage means the plastic will shrink differently in the direction of flow versus the transverse direction of flow. It is most commonly associated with semi-crystalline materials because of the crystalline structure of the material. Remember the crystalline structure has to stretch out in order to flow and when it cools back down it acts like a spring, and will shrink more in the direction of flow because of the crystalline structure springing back into shape as it is cooled, with less in the transverse direction of flow.

Figure 7.7 Anisotropic shrinkage

This will all change when running a filled material. The glass fibers or filler structure do not melt, so as the material cools down the glass fibers or structure will not shrink. They also align with the direction of flow, so the shrinkage in the direction of flow will be less in the transverse direction of flow.

77

78 7 Plastic Temperature

„„7.6 Melt Density versus Solid Density Melt density and solid density are two completely different things. Solid density is the density of the plastic after it cools down, and the density will be greater (same molecules in a smaller amount of space). Like a spring, the molecules will contract when cooling down. Melt density will be much less. As the plastic gets heated the molecules start to loosen up and relax, which means they are spreading further apart. The more they spread apart the more space they take up, which in turn means that same molecules to take up a greater amount of space. Building the core and cavity larger than what the part is when it is cooled accounts for this difference between melt density and solid density. The core and cavity get filled with the melt density and cool down to the solid density. Semi-crystalline materials will have around a 20% decrease in density from solid to melt. If the solid density is 0.92 then the melt density after heating will be around 0.736. Amorphous material will have around a 10% decrease in density. If the solid density is 1.20 then the melt density will be around 1.08. The reason for the difference in melt density in the semi-crystalline versus the amorphous material is crystalline structure. The crystalline structure must relax when it is heated and when that relaxation happens it will take up more space. The next thing to consider is, if the machine is close to maximum shot size on the barrel and the material is changed, will there be enough shot size to fill the part out (each material has a different density)? That is why shrinkage plays a huge role in our final part dimensions. Let’s look now at Figure 7.8. It shows a pellet in a cool state or solid state, and a pellet heated or in a melt state. With the same number of molecules, the heated pellet is expanded due to the molecules relaxing. Melt and solid density must be taken into consideration when putting a mold into a machine with a shot size greater than 80% of maximum capacity. Running out of shot size can happen!

Figure 7.8 Solid density versus melt density

7.7 Advantages/Disadvantages of Hot Runner versus Cold Runner

„„7.7 Advantages/Disadvantages of Hot Runner versus Cold Runner Cold Runner Systems Advantages: ƒƒCheaper to produce and maintain ƒƒCan accommodate a wide variety of polymers, including commodity or engineered resins ƒƒColor changes can be made quickly (no manifold to clean) ƒƒFast cycle times if robot is used in removing runners Disadvantages: ƒƒCycle times are slower than hot runner systems due to thickness of runner ƒƒPlastic waste from runners (particularly if they cannot be reground and recycled) ƒƒShrink rate will be different with different resins Hot Runner Systems Advantages: ƒƒPotentially faster cycle times ƒƒEliminate runners and potential waste ƒƒNo need for robotics to remove runners ƒƒCan accommodate larger parts Disadvantages: ƒƒMore expensive molds to produce ƒƒColor cannot be easily changed ƒƒTrue gate seal takes longer ƒƒHigher maintenance costs and potential downtime ƒƒMay not be suited to certain thermally sensitive materials ƒƒLonger start-ups

79

80 7 Plastic Temperature

„„7.8 Induced Shear through a Hot or Cold Runner System The shear that is induced through a hot or cold runner system will determine how the parts fill during the injection phase. As the material flows into a cold runner the material freezes on the side walls, the plastic will flow down the center (fountain flow), and the layer just in from the frozen layer will have a lot of shear i­ nduced into it (like rubbing hands together). This in turn will heat up the outer portion of the flow.

Hoer material

Figure 7.9 Shear pattern through primary runner

Figure 7.10 Shear pattern through second and third branches of runner

As the flow front splits when it hits the secondary runner so do the hotter shear sections. At the tertiary runner the hotter melt would stay to the inside. This will allow the material on the inside of the center part to fill first (follow the red line) thus creating a shear imbalance. This will also happen in a hot runner manifold if there are less than three shots held within the manifold system. It all revolves around the flow length: the longer the flow length in a hot or cold runner system, the more shear is generated.

8

Plastic Flow

There must be two things available when talking about plastic flow: pressure and flow. Pressure enough that the process does not become pressure limited and sufficient flow available to achieve the velocity asked for.

„„8.1 Fountain Flow When plastic flows, it will always flow in a fountain flow state, meaning the plastic flows down through the center while the outside layer freezes or solidifies when it touches the steel.

Figure 8.1 Fountain flow of plastic

82 8 Plastic Flow

„„8.2 Flow of Plastic Plastic flow is expressed in volume per second. Looking at this a little deeper, it can be expressed in wt./second, in3/second, or cm3/sec. What this is saying is how much volume (weight) can be injected in 1 second. This is important to know when transferring a mold from one press to the other. Volume and flow can be calculated and transferred to the other press to take the guesswork out of it.

Figure 8.2 Plastic flow expressed in volume per second

„„8.3 How to Calculate Flow Rate (Qp) Flow rate is calculated using weight or volume and is expressed as wt./sec, in3/sec, or cm3/sec. The first thing to know is the area of the screw. Area of screw: Let’s say there is a 55 mm screw. Now to figure out the area in inches, take 55 mm and divide by 25.4: 55/25.4 = 2.165″ The next step is to calculate the area, which will be 2.165 × 2.165 × 0.7854 = 3.68 in2. The next step is to find out how much linear stroke is in the fill phase. The formula for that would be shot size plus (+) decompression minus (−) transfer position = LS. If there is a shot size of 3.25″ with the decompression set at 0.50″ and the transfer position set at 0.95″, then the equation is as follows: 3.25 + 0.50 – 0.95 = 2.80″

8.4 Calculating Volume of Shot Size

Take the 2.80″ and multiply by the area of the screw (3.68 in2) to give 2.8 × 3.68 = 10.304 in3. That is 10.304 inches cubed, because now we have taken a 2D object, the area of the screw, and multiplied it by a distance, which now turns it into a volume or 3D object.

2D area of screw

3D volume

Figure 8.3 Conversion of 2D shape and turning into 3D volume

If it is known that there is 10.304 in3 in fill volume and it takes 0.75 seconds to fill, then the flow rate for this example is 10.304/0.75 = 13.739 in3/second. This is the flow rate coming out of the nozzle. With an eight-cavity mold, take 13.739 and ­divide by 8, which equals 1.717 in3. What started as 13.739 in3/second flow rate at the nozzle is now 1.717 in3/second into each cavity.

„„8.4 Calculating Volume of Shot Size To calculate volume, two things are critical: the area of the screw and the stroke or shot size. Remember when calculating the area of a circle or cylinder, use D × D × 0.7854 or pR2. Example: We have a 35 mm screw. First, let’s convert millimeters to inches. Take 35 and divide by 25.4 to turn into inches: 35/25.4 = 1.378″. Then, D × D × 0.7854 = 1.378 × 1.378 × 0.7854 = 1.491 in2 (surface area of screw). And if our shot size is 3.25″ take 3.25 × 1.491 = 4.846 in3 (cubic inches). This is the volume of plastic in front of the screw that is able to be put into the cavity. 

83

84 8 Plastic Flow

„„8.5 Blocking a Cavity If there is a cavity that needs to be blocked, make sure that the cavity is safe to block off (no steel damage). The next step is to be sure why it is being blocked off. Is it dimensional or visual? Can the defect be processed out? And if the answer is no, block the cavity. On a cold runner, plug the gate and make sure that it is easily removable. If it is a hot runner mold, remember not to turn the cavity completely off; set the tip at 200 °F. This will keep some of the thermal expansion on the tip and prevent a leak in the manifold system. If the heater is bad then there is no choice and it must be turned off, but the heater must be replaced as soon as possible. When blocking a cavity, always remember that the flow rate of the plastic entering the mold and flowing into the cavity must be consistent (same fill time). You might have to slow down injection velocities to achieve the same fill time. This will help the molecular orientation and frozen skin thickness when running full cavitation, and maintain the same amount of shear seen in the plastic. Remember this principle when blocking cavitation.

Figure 8.4 Blocking a cavity

There are other principles that must be remembered. When blocking the cavity or cavities off, the projected area or total surface area of plastic pressure at the parting line is being reduced. Reducing the clamp tonnage is a must. Also necessary are reestablishing the shot size and changing the transfer position to maintain the 95%–98% full part at transfer. Therefore, understanding the part weight at transfer is important at start-up. Make sure the 5–10% cushion of shot size is being maintained.

8.7 Orientation

Example: Let’s say there is a four-cavity mold that fills in 1 second and it is necessary to block a cavity off. The fill time must remain as 1 second, which means to slow down the injection velocity to match the fill time of the four cavities and verify the 95% full part. Fill time is one of the critical variables that must be maintained. 

„„8.6 Flow through a Mold As the plastic flows through the cavity, the velocity at which it flows will change. If flowing into a thicker or larger section, the flow front will slow down because of the larger area, and the opposite occurs as it flows through a narrow section, where it will speed up. Although there is a programmed velocity in the press, the velocity will change as it flows through different geometries of the part. Be aware of this when processing visual defects.

„„8.7 Orientation Orientation only happens when the plastic is flowing. Within the sections of the part there will be highly orientated layers which are near the surface of the part and less orientated layers which are more towards the center of the part. The reasons the part gets highly orientated layers near the surface is from the shearing that is going on as the material flows through the cavity. There will be a frozen plastic layer against the wall of the cavity and a molten layer that is trying to slide by, which creates friction. The shearing decreases going towards the center of the part. Injection velocity will play a role in how thick the frozen layer is. With faster injection, the the frozen layer is smaller, whereas with slower injection, the frozen layer is thicker.

85

86 8 Plastic Flow

„„8.8 Transfer/Cut-Off Position There are four ways to transfer from the first stage to second stage on an injection molding press. ƒƒTime: When transferring by time, the process will not be consistent. The reason for this is telling the machine to go from shot size to transfer position in a specific time: if a viscosity change happens, the time to get to a certain point will either be shorter on longer depending on the viscosity of the material, and becomes very inconsistent. ƒƒPosition: This is one of the more consistent ways to transfer. When a viscosity curve was constructed, the injection velocity was optimized, and if there was a need to make a slight change to velocity it would have minimal effect on the viscosity of the material. And if a viscosity change happens, it will transfer at the same position, and it is just the transfer pressure that will either go up or down depending on how the viscosity shifted. ƒƒHydraulic Pressure: This is another inconsistent way to transfer, understanding that the machine will transfer at a certain pressure, but again if there is a viscosity change that pressure will be reached sooner or later depending on the viscosity. This increases the likelihood of an over- or under-packed part. ƒƒCavity Pressure: This is the most consistent way to mold, with fill and pack under velocity control, which means taking advantage of the rheological properties of the plastic, but also packing to a specific cavity pressure, which means the same part every time.

„„8.9 Viscosity Changes Viscosity changes will always happen in injection molding; our job is to try to control the viscosity change as much as possible. Viscosity changes happen from lot to lot of material or can happen within a single lot of material. When a viscosity shift happens, the material gets thicker or thinner compared to what was previously running. When the viscosity changes it means a couple things could have happened to the material. The chain length of the molecules or orientation could have changed; if the material gets stiffer or harder to push, it could mean the chain length is longer or less oriented. There is more strength or rigidity with longer chain length. On the other hand, if the chain length gets shorter or there is greater orientation, the viscosity will go down (easier to flow), but can result in a weaker or less rigid part with shorter chain lengths.

87

8.10 Intensifying Ratio (Ri)

If the viscosity goes up, the material becomes thicker and harder to push, and it will probably be noticed that the transfer pressure goes up. Remember that the material is thicker and harder to push so it will take more force to push it to the same point. If viscosity goes down, the material is thinner and easier to push so the transfer pressure will go down because it does not take as much pressure to push the material to the transfer position.

„„8.10 Intensifying Ratio (Ri) The injection end of the molding press has an intensifying ratio, denoted as Ri. What this means is the hydraulic oil is being intensified or amplified in front of the screw of the molding machine. It is calculated two different ways: area of the ­injection cylinder divided by the area of the screw, or maximum plastic pressure divided by maximum hydraulic pressure of the molding press (Figure 8.5).

Figure 8.5 Two ways to get information to calculate intensifying ratio

This intensifying ratio applies to all pressure functions of the injection end: ƒƒFill pressure ƒƒPack pressure ƒƒHold pressure ƒƒBack pressure So why is understanding the intensifying ratio important? Since it applies to all pressures on the injection end, this information is needed when transferring a mold from one press to another. Let’s see this in the example below. Example: A machine has a transfer pressure (when it goes from first stage pressure (fill) to second stage pressure (pack/hold)) of 1,900 psi hydraulic and the intensifying ratio is 10 : 1. 1,900 hydraulic pressure × 10 (which is the intensifying ratio) = 19,000 ppsi plastic pressure needed.



88 8 Plastic Flow

It is now moved to a press that has an intensifying ratio of 12.5 : 1. Take the 19,000 ppsi and divide by 12.5 to give the new specific/plastic pressure needed in the new press, giving 1,520 psi pressure hydraulic to achieve the same hydraulic pressure needed from the first press. Note that this also works the other way. Considering the same press that transfers at 1,900 hydraulic pressure with an intensifying ratio of 10 : 1, and therefore a specific/plastic pressure is 19,000 ppsi. Now move the mold to a press that has an intensifying ratio of 8 : 1. The new pressure needed is 2,375 psi hydraulic pressure. If our new press has a max pressure of 2,000 psi hydraulic pressure, then the machine is pressure limited, and runs out of the pressure needed. 

„„8.11 Pressure Limited Process Pressure limited is a term used when the machine is now filling on pressure and not velocity (injection speed), which does not allow the proper shear thinning of the material. When the material is not shear thinned properly it creates inconsistencies in flow and filling. See the example below showing the pressure curves of a 95%–98% fullest cavity process (using abundant pressure and flow to fill the mold 95% to 98%, then transferring to pack/hold phase) and a pressure limited process where the machine only uses enough pressure to fill and pack the mold.

Figure 8.6 Proper pressure versus pressure limited

8.12 Safe Start-Up Shot Size

Notice how the velocities drop off once the pressure limit is reached. Look at the Figure 8.7 now; this is a perfect example of a pressure limited process. Notice as the pressure is limited (red curve line) the velocity (green line curve) is not able to maintain the velocity the machine is asking for. Now the flow of the material is being controlled by pressure and not velocity. This means the material is not shear thinning properly due to the fact that pressure is controlling the flow instead of velocity.

Pressure limited

Velocies drop off

Figure 8.7 Actual pressure limited process (red line = pressure; green line = velocity)

„„8.12 Safe Start-Up Shot Size The purpose of this is to determine a safe amount of plastic or material to inject into the mold on the first shot without causing damage to the mold by over- or ­under-filling it, and have the parts eject off. Improper shot size can damage the mold by over-pressurizing the parting line; it can also break mold inserts or lock the mold up so that it cannot open. There are normally four ways to determine the initial shot size for a new mold: 1. Previous projects 2. Guess at it, test the guess, then retry 3. Experience 4. Calculate

89

90 8 Plastic Flow

Figure 8.8 Safe start-up calculation. Courtesy of AIM Institute Example: 95% of the time, it is possible to get a shot weight from the design group. With the formula above (Figure 8.8) let’s use the shot weight (volume cm3) based on the following data: Shot weight: 27.4 g Volume %: 80% full Solid density of PP: 0.92 g/cm3 (comes from material data sheet) Screw: 25 mm Conversion factor for cm3 to in3: 0.061023744 Now

27.4 ×0.8 21.6 = = 23.478 cm3 0.92 0.92

23.478 × 0.061023744 = 1.4377 in3 This is the volume needed to get an 80% safe start-up shot. Now let’s add our transfer volume of 0.85 in3: (1.4377 + 0.85) = 2.2877 in3 (remember volume is area of screw × stroke) 25 mm/25.4 = 0.984 in, so area of screw is 0.984 × 0.984 × 0.7854 = 0.76 in2. Now take 2.2877 and divide by 0.76. It gives us 3.01 in of stroke to get 2.2877 in3 of volume. 

8.13 Runner Sizing

„„8.13 Runner Sizing Below is one method for sizing the runner. One of the other methods is to keep the diameter the same all the way to the gate. At the end of the day we are trying to equalize the shear rate, minimize the pressure drop, and equalize velocity through each section of the runner for good part quality. Dgate feed = 1.5 × wall thickness at gate (0.060″ at gate is used for this example) Standard formulas: Dfeed = Dbranch × N⅓ Dfeed = 2 × √2 × area d-branch/p Note: N is the number of branches that the feed runner is feeding. Size the gate runner first and upsize if needed back to the sprue. Make the changes to keep the shear rate the same in each section. Dgate feed = 1.5 × 0.060 = 0.090″ Dfeed 1 = 0.090″ × 1.26 = 0.113″ Dfeed 2 = 0.113″ × 1.26 = 0.142″ (If we take N to the ⅓ power and N is the number of branches (2) for each section in the Dfeed 1 and in Dfeed 2, then we get 1.26.)

Figure 8.9 Sizing runner from gate back to sprue

91

9

Plastic Pressure (Pack/Hold)

„„9.1 Plastic Pressure The plastic pressure is what to measure on the process. It is calculated as the maximum hydraulic pressure multiplied by the intensifying ratio of the press (Ri). This will apply to the fill pressure, pack pressure, hold pressure, and back pressure. Even though the hydraulic pressure and the intensifying ratio will change from press to press, the plastic pressure (specific pressure) generated will remain the same because it is plastic pressure in front of the screw that we are trying to ­control.

„„9.2 Dynamic versus Static Dynamic pressure is the pressure that is experienced in the first stage, or boost portion of the cycle. This is where the machine is under velocity control and has abundant pressure and flow from the pump to make sure it maintains that velocity control, with shear thinning of the plastic. Static pressure comes under the pack and hold phase of the process. This is pressure control, where only pressure is controlling this phase of the process.

„„9.3 Viscosity Changes Viscosity changes happen from lot to lot of material or can happen within a lot of material. With a viscosity shift, the material gets thicker or thinner than what was previously running.

94 9 Plastic Pressure (Pack/Hold)

When the viscosity changes, the chain formation of the molecules is changing to more aligned or less aligned. The machine will see a viscosity change when going into pack phase because the machine is no longer under velocity control and no longer shear thinning the material. When the viscosity goes up, the material becomes thicker and harder to push.

„„9.4 End of Cavity Pressure Loss The end of cavity is where the largest pressure drop will be noticed. As plastic flows further away from the gate, its tendency to freeze increases, and every step of the way there is a pressure loss (so to achieve a fully packed out part, make sure to have ample pressure in front of the screw). It is known that there will be viscosity shifts in the material. If there is a viscosity shift that could use up the remaining pressure left in front of the screw, a 10% abundance of pressure must be maintained to accommodate viscosity changes. For example, if there is an injection pressure of 2000 psi hydraulic, then our transfer pressure cannot be above 1800 psi hydraulic or there is a chance of becoming pressure limited. If the viscosity goes up (meaning stiffer material), there is a good chance of running out of injection pressure before transfer.

„„9.5 Part Shrinkage versus Cavity Pressure With increased pressure applied to the pack or cavity, less shrinkage will be observed. When the molecules are packed tighter together they have less of a chance to move during the shrink phase. More shrinkage will occur in a semi-crystalline material because of the crystalline structure. Remember that the crystalline structure acts like a spring and the increased pressure inhibits the spring from returning to its original shape. But it can cause problems if the part is to be used in a heated environment; when the plastic is reheated the molecules will want to relax, which could cause warp or distorted parts.

9.6 Maximum Average Pressure at Parting Line before Flashing

Figure 9.1 Pressure vs shrink, from RJG’s Master Molder classSM

„„9.6 Maximum Average Pressure at Parting Line before Flashing Having a good understanding of why a tool flashes is important. Remember there is all this pressure over a specific area and eventually the force of injection pressure will overcome the force of the clamping unit and that is when flash appears. Figure 9.2 shows the relationship between clamping force, projected area, and maximum average pressure.

95

96 9 Plastic Pressure (Pack/Hold)

Figure 9.2 Relationship between clamp force, projected area, and maximum average pressure

Example: Consider a 300 ton press at full tonnage. Clamping force in pounds (lbs.): Clamping force = 300 × 2000 = 600,000 lbs. (2000 lbs. per ton) Projected area: Projected area is the area of the part measured at the parting line. If a part is rectangular and measures 3.5″ by 5.25″ then the projected area will be 18.375 in2. Maximum average pressure: Maximum average pressure is an average of the post gate cavity transducer pressure and the end of fill cavity transducer pressure. Let’s say there is 12,000 ppsi post gate and 3,000 end of fill: 12,000 + 3,000 = 15,000 ppsi Now take 15,000 and divide by 2 to get the average or 7,500 ppsi. (To get the average of anything, take all the set points and add them up and divide by the number of set points.) Now it is known that 7,500 ppsi is the maximum average pressure at parting line without flashing the tool. 

9.6 Maximum Average Pressure at Parting Line before Flashing

So what if there is no transducer in the tool? Take the clamp force in pounds and divide by projected area to get maximum average pressure at parting line before flashing. There are two ways to look at this: take clamp tonnage down until it flashes and now the mold has exceeded the maximum average pressure, or take pack pressure up until the mold flashes, again exceeding the maximum average pressure. When starting to get flash, the first two things that are tried are going up on clamp tonnage or down on pack pressure, therefore making the changes to the process that make sure the clamp force overcomes the injection force.

97

10

Cooling

„„10.1 Plastic Cooling The whole purpose of cooling is to run the fastest cycle possible with the most dimensionally stable part. Think of the cooling rate of the plastic and what effect it has on the physical properties of the plastic, fast cooling rate, slow cooling rate, and differential shrinkage. Let’s examine how and where the water lines are hooked up and their placement.

„„10.2 Turbulent versus Laminar Flow Turbulent flow is what is needed to have effective cooling. As the water churns inside the water channel, it allows the heat to be carried away. With laminar flow, the outer layers of the coolant near the wall will create an insulation barrier and ­inhibit the effective cooling needed.

Laminar Flow Figure 10.1 Laminar flow and turbulent flow

Turbulent Flow

100 10 Cooling

Another way to visualize this concept is to think of the faucet on a kitchen sink. If the faucet is turned on slowly, it creates a clear stream that is laminar flow (poor cooling) (Figure 10.2), and if the faucet is turned on full blast, the water stream is bubbly and foamy, which is turbulent flow (effective cooling) (Figure 10.3).

Figure 10.2 Laminar flow in a river

Figure 10.3 Turbulent flow in a river

10.3 Reynolds Number

„„10.3 Reynolds Number The Reynolds number tells us when turbulent flow is achieved. The value of ­Reynolds number strived for is 10,000, as this assures us that there is turbulent flow. The charts below give the gallons per minute (GPM; Table 10.1) or liters per minute (LPM; Table 10.2) needed for turbulent flow through a water line of specific size. Table 10.1 GPM Needed for Turbulent Flow (Reynolds Number = 10,000) Water channel

Temperature 180 °F

140 °F

100 °F

50 °F

1/4″

0.28

0.37

0.54

1.03

3/8″

0.41

0.56

0.81

1.55

1/2″

0.55

0.74

1.09

2.07

5/8″

0.69

0.93

1.36

2.58

Table 10.2 LPM Needed for Turbulent Flow (Reynolds Number = 10,000) Water channel

Temperature 82 °C

60 °C

38 °C

10 °C

 6 mm

0.99

1.32

1.95

3.7

 8 mm

1.32

1.77

2.6

4.93

10 mm

1.65

2.21

3.25

6.17

15 mm

2.47

3.32

4.87

9.25

Let’s look at the GPM chart (Table 10.1). Let’s go to the water channel column and choose 3/8″ water line and find the mold temperature of 100 °F. It takes 0.81 GPM to achieve turbulent flow. When creating the Excel spreadsheet (Figure 10.4), all the cells in yellow are manually entered, and all the cells in brown and blue are calculations created from the manually entered values. Table 10.1 and Table 10.2 give an easy view of how many GPM or LPM are needed per line to get a 10,000 Reynolds number. The other calculations in the spreadsheet below provide an easy way to calculate the GPM or LPM needed for the entire tool, so that volume of water coming into the system can be known. Below are the formulas to build the spreadsheet (Figure 10.4): #1 = (Viscosity)*(Diameter)*(10,000)/3,163 or (0.92 × 0.25″ × 10,000)/3,163 #2 = (Viscosity)*(Diameter)*(Diameter)/21,221 or (0.92 × 6mm × 10,000)/21,221

101

102 10 Cooling

#3 GPM × the number of water lines on the A half, B half, or center section #4 LPM × the number of water lines on the A half, B half, or center section

#1

#2

#3

#4

Figure 10.4 Excel spreadsheet for Reynolds number

„„10.4 Water Lines Try to get as much water on the mold as possible, but sometimes jumper water circuits are necessary. It is necessary to achieve a 4 °F or less delta difference ­between the water-in temperature and water-out temperature, and 2 °F delta difference on critical jobs.

10.6 Series/Parallel

„„10.5 Area of Water Line This is the most important thing to understand when watering up a mold and a great troubleshooting tool. The surface area of the supply line must be greater than all the surface area of all circuits added up. If not, a pressure loss will occur. And if the process was just making turbulent flow or 10,000 Reynolds number in a smaller diameter line, and continued into a larger diameter line, changing to laminar flow is possible. Example for troubleshooting water flow issue: Let’s say a 3/4″ water line feeds a water manifold that has eight ports that are 3/8″ diameter. Calculation: 3/4″ supply line means area = 0.750 × 0.750 × 0.7854 = 0.442 in2 3/8″ water lines each have area 0.375 × 0.375 × 0.7854 = 0.11 in2 Now multiplying this by 8 circuits gives 0.11 × 8 = 0.88 in2 Now notice that the surface area of the eight circuits is twice as large as the supply line, which means that twice as much water is needed to fill and flow and pressurize those circuits as the supply line, or whatever the water flow through the supply line will be reduced in half going through the eight circuits. 

„„10.6 Series/Parallel How should cooling channels be connected to the mold? Use the least amount of circuits that will provide turbulent flow. It is often said that jumpers on the mold are a bad thing, but this is not always true. The following should be taken into account: 1. How many jumpers? 2. What is the temperature delta between the IN and OUT? 3. Is there a large pressure drop?

103

104 10 Cooling

­Parallel channels:

Figure 10.5 Parallel channel cooling

ƒƒMaximum cooling is available ƒƒThe most cooling used ƒƒLowest pressure loss ƒƒFlow channel with restriction gets least cooling ­Series channels:

Figure 10.6 Series channel cooling

10.7 Cooling Rate

ƒƒUses the least amount of cooling ƒƒHas the largest pressure drop ƒƒLargest temperature delta ƒƒAll channels get the same flow Sometimes the use of jumpers will create fewer channels and will provide more water flow to the mold. It will depend on how much water flow is coming from the main line and how many circuits are on the tool.

Figure 10.7 Optimized water connections

„„10.7 Cooling Rate Cooling rate depends on the difference between the plastic temperature and the steel temperature. Remember that the heat put into the plastic must now be removed and happens very rapidly between melt transition temperature (Tm), or rubber phase, and glass transition temperature (Tg) or solid phase, because the molecules are still free to move. Once Tg is achieved, the molecules slow down or stop moving because they are freezing or re-crystallizing and becoming locked into place; then the cooling rate slows down.

105

106 10 Cooling

„„10.8 Ineffective Cooling Build-up in water lines can become a major problem for effective cooling and cycle time. Water lines and water channels inside the mold and on the press must be cleaned and maintained always. A 1 mm or 0.040″ build-up can reduce cooling ­capability by 40%. The build-up acts as an insulator and does not allow the heat to transfer properly to the turbulent water inside the channel to be carried away

„„10.9 Cooling Time Cooling time is the time added to the machine to allow the machine to recover the screw and build enough plastic in front of the screw for the next shot. By holding the plastic in the mold to allow the heat that was added to the plastic to get it to flow to be removed (BTU’s put into melt must be removed) and allow the part to eject off the mold without distortion. Adding this time ensures the plastic cools below the HDT or heat deflection temperature. This is the temperature at which the plastic is stable enough to eject off the mold; each material has its own HDT.

„„10.10 Depth, Diameter, and Pitch Here, we are referring to the following. Depth: The average distance from the center of the cooling channel to the surface of the part (1.5D–2.0D). Diameter: The diameter of a cooling channel, not the water line connection. Pitch: The average distance between the cooling channels (3 × D)

10.10 Depth, Diameter, and Pitch

Figure 10.8 Depth, Diameter, and Pitch

Make sure the water channels follow the shape of the part, or otherwise there will be hot spots within the core and cavity (notice in Figure 10.9 that the depth is not always the same). The hot spots will cause differential shrinkage, which in turn can cause warp or distortion.

Figure 10.9 Improper placement of water channels

107

108 10 Cooling

„„10.11 Baffles/Bubblers A baffle is installed into the mold to divert the water up a long core where conventional cooling methods are not as successful.

Figure 10.10 Baffle in mold

Bubblers are also installed to divert water into hard-to-reach areas. The water goes up the center tube and fountain flows down the outside. This way the water constantly carries away the heat. However, be careful: the bubbler can only be hooked up one way to work properly.

Figure 10.11 Bubbler in mold

„„10.12 How a Thermolator/Mold Heater Works Water is supplied by the water IN line. The water must go through a pressure ­sensor to make sure of proper flow (pressure is resistance to flow—no pressure, no flow). Once the pressure switch has seen enough pressure it allows the pump to turn on.

10.12 How a Thermolator/Mold Heater Works

The pump then sends water to the heater chamber. The thermal sensor will tell the solenoid valve what to do (if the mold is not up to temperature the solenoid valve shuts off and only allows the water to go through the mold and heat up). When the mold is up to temperature and needs to cool down the solenoid valve opens and releases the hot water and replaces it with cold water from the IN (closed loop). If there is a need for the thermolator to cool constantly, then it becomes an openloop system and the water going from the IN circulates through the mold and goes out of the OUT.

Figure 10.12 Typical thermolator diagram

109

11

Benchmarking the Injection Molding Process

Coefficient of Variation By performing this benchmarking procedure before the start of the next production run, the capability of the molding process to produce an acceptable product can be determined. This is part of the qualification procedure for injection molding tools. If there are machine problems, process limitations causing inconsistency, raw material variations, material handling differences, tool concerns, auxiliary equipment problems, or environmental concerns, these can be picked up and quantified using the benchmarking procedure below. This procedure is easy to perform and will put a quantifiable number on just how capable the production process will be. The idea is to use part weight to obtain a snapshot of just how robust the overall molding process is. To perform the benchmark, the following items are needed: ƒƒForty consecutive shots bagged (ensure these were pulled after the process has stabilized). ƒƒScale for measuring part weight. The scale must have resolution to three decimal places if the shot weight is less than 100 g, or resolution to two decimal places if the shot weighs more than 100 g. If the scale does not have this type of resolution, the data variation will not be sensitive enough to yield useful information. The following formula is used to calculate the coefficient of variation expressed as a percent: CV = (standard deviation/mean). This formula will allow the user to quantify if a molding process has an acceptable level of variation. The procedure for benchmarking is pretty simple. Collect 40 shots after the process has had a chance to stabilize. Determine the total shot weight by weighing (including runners if available). Document the shot weights, calculate the mean and standard deviation data, and compare the results to Table 11.1 below.

112 11 Benchmarking the Injection Molding Process

Table 11.1 % Chart for Machine Capability Coefficient of variation

Quality of molding

0.01 to 0.10%

Precision molding

0.10 to 0.32%

Fair to good quality

0.32 to 1%

Poor operating zone

>1%

Unacceptable

Newer injection molding machines should easily be able to produce a product in the precision molding range. Values above 0.32% need to be evaluated as to the source of the variation. To go into a manufacturing run with values greater than 0.32% will produce a product with significant variation and poor quality. If every molder used this procedure to evaluate the molding process prior to the start of a production run, improved consistency on molded parts, improved quality, and less downtime during the production run would result. A spreadsheet like that in Figure 11.1 is a convenient way to do this calculation.

#3

#4

#3

#4

#6

#2

#5

#7

Figure 11.1 Excel spreadsheet for coefficient of variation

Here is how to set up the Excel spreadsheet as shown in Figure 11.1. #1 The yellow cells are where to enter the individual shot weights for the 40 shots. #2 Calculate the mean of all the part weights. Formula for N15 is =(C25+H25)/2. #3 Columns D and I are used to subtract the shot weight from the mean part weight. Formula for D5 would be =C5-N15. Remember that N15 will be the mean part weight. #4 Columns E and J are the second part of the equation to figure out variance. Take the answer from column D and square it. The formula for cell E5 is =D5*D5.



#5 Now, to calculate variance, take the two totals in cells E25 and J25 and average them. Formula would be =(E25+J25)/2 for cell P15. #6 Calculate the standard deviation. It will be the square root of P15, and the formula is =SQRT(P15). #7 Calculate the coefficient of variation. Thus is the standard deviation divided by the mean part weight. The formula is =(L15/N15)*100. Then minimize columns D, E, I, and J and the calculations will work in the background (Figure 11.2).

Figure 11.2 Excel spreadsheet for coefficient of variation (simplified view)

The main thing to remember about this test is that it will detect variation within the process. It is our job to find out where the variation is being introduced and whether the machine is truly capable of running a stable process.

113

12

Process Troubleshooting

This chapter examines many of the most common problems encountered during injection molding. Possible causes are listed and solutions are suggested.

„„12.1 Black Specks

Figure 12.1 Black specks

ƒƒQuestions to ask before starting to troubleshoot: ƒƒWhat are the black specks? Is it carbon or dirt? ƒƒWhere are the black specks coming from? From the material, barrel, nozzle tip, or manifold system? Remember to check every possibility ƒƒIs it a hot runner or cold runner, and how does that affect the plastic flowing into the mold? ƒƒContamination in the material: ƒƒCheck surrounding area ƒƒIf there is a fan blowing above the press, check whether the filters need changed, or there is dust hanging on the rafters –– Watch for any kind of airborne source that will contaminate the material

116 12 Process Troubleshooting

ƒƒIf the machine was cleaned, consider whether contamination was blown into the material bin –– Again, airborne contamination ƒƒDust on the cover that is getting the material –– Remove and clean dust cover (remember not to blow contamination onto the next press) ƒƒRe-clean hopper and loading system in case of possible contamination ƒƒFind any dead spots (behind gaskets, stuck in screen, and in drain-out cap) ƒƒClean feed throat and magnet box area. Is there liquid color that needs to be cleaned out? ƒƒCheck all possible locations that can hold foreign material ƒƒHeater band malfunction ƒƒCheck for heater bands over-riding the set temperature ƒƒRelay or contact stuck (older style machines) ƒƒDegradation of the material in that zone. Where are the black specks? And what part of the barrel are they coming from? ƒƒNozzle has build-up of degraded material –– Clean degraded material from nozzle; material can be drawn into flow ­channel ƒƒIs there a sprue break causing degraded material to get into the flow channel? Eliminate or reduce stroke? ƒƒHeater band not working (e. g., burned out) creating more shear in the area of that heater band (80% of the plastic heating comes from shear or mechanical heating). If there is a skin of plastic left from the previous run, make sure the power level for that zone is not always at 100%, as this could mean one of the bands is dead and the other heater band is trying to make up for the heat loss. ƒƒMaterial change ƒƒHigher MFI or lower MFI (MFI = melt flow index)? ƒƒAre the heating bands running hotter or colder than for the previous job? Did the screw speed change, which will affect the heating and shear of the material? –– To eliminate previous skin of material, raise heats and run natural material, then lower heats to create new skin layer ƒƒIs the tact temperature the same from one material to the other? –– Tact temperature is the temperature at which the material starts to stick to the barrel to start the shearing process ƒƒDo the screw and barrel need to be cleaned? –– Eliminating two dissimilar materials, what are the MFI’s? or does new material have an additive or filler in it that could scrub the barrel?

12.2 Blush

ƒƒDo the tips need to be pulled to eliminate material from the previous job? –– Eliminate the skin layer from the previous job ƒƒClean nozzle and sprue bushing to remove degraded material

„„12.2 Blush

Figure 12.2 Blush

Blush is a hazy or dull surface imperfection opposite the gate or around the gate area. ƒƒGate blush is caused by the too fast injection velocity that can wash away the plastic on the opposite side of the gate ƒƒPlastic has a fountain flow, and if the skin layer does not have a chance to lay down or starts to skid across the surface this can cause blush ƒƒSlow down plastic as it enters the gate, reduce flow rate ƒƒIf running a cold runner tool, make sure to provide a cold slug well before plastic reaches gate ƒƒChange gate geometry ƒƒThis will change how plastic flows into the part ƒƒIncrease melt temperature, so plastic will flow more easily ƒƒIncrease mold temperature ƒƒThis will allow a smoother transition between the hot plastic and the cold mold ƒƒClean mold surface, to help avoid build-up of gases

117

118 12 Process Troubleshooting

„„12.3 Brittleness Brittleness is a loss of physical properties in the plastic, which means it breaks easily when subjected to stress, rather than deforming. ƒƒMoisture in material ƒƒVerify moisture content in original material. If material is wet, dry longer, but also verify temperature. Do not expose the material to excessive temperatures. –– Dry temperature and dry time of specific material. ƒƒMoisture in feed throat –– Feed throat should be set at 100 °F to 120 °F to avoid condensation ƒƒWater leak in mold –– Find and repair water leak ƒƒSweating of mold if not in climate-control setting or chiller water is below dew point in plant –– Raise chiller temperature –– Resolve issue with climate-control system ƒƒExcessive melt temperature ƒƒLower melt temperature to avoid degraded material ƒƒLower nozzle temperature ƒƒCheck residence time/temperature in barrel –– Should use 25% to 75% of barrel for shot capacity –– Profiled temperatures may be required: colder in rear and profile up to ­recommended melt temperature ƒƒDecrease screw speed –– Shear heating ƒƒMelt temperature too low ƒƒWeld lines or weak knit lines ƒƒContamination ƒƒInspect resin for contamination –– Mixed resins or incompatibility ƒƒPurge barrel for contamination ƒƒPull and clean barrel and screw (flights of screw can be contaminated with other material) ƒƒRegrind ƒƒHow many generations of regrind? –– Closed loop systems can have in excess of 10 generations

12.4 Burns

–– How many heat profiles has the plastic seen? –– Normal regrind percentage is 25% to 30% ƒƒImproper gate design ƒƒSize/location/shape ƒƒGate location affects directional strength and orientation of plastic in part ƒƒThin to thick or thick to thin? –– How is plastic flowing and packing?

„„12.4 Burns

Figure 12.3 Charred or burned resin

Burns are charred or discolored material usually (but not always) at the last point of fill. Trapped air in the cavity is compressed to the point that the air or gases ­ignite causing the burn mark. ƒƒReduce fill speed ƒƒThe over-shearing of the material from too high of an injection velocity can cause the breakdown of the additive chain length and cause degradation ƒƒVerify proper transfer position ƒƒTransfer at 95% full part; allow the gas a chance to escape ƒƒClean vents ƒƒOr add vents in proper location (see Figure 12.4) ƒƒClean ejector pins (ejector pins provide some venting) ƒƒChange gate location ƒƒProvides different flow pattern of the plastic as it enters the cavity

119

120 12 Process Troubleshooting

ƒƒReduce clamping force ƒƒAvoid not allowing the tool to breathe or over-clamping. Always think of the projected area of the part; just because it is a 300-ton press does not mean it needs to be to run at 300 tons ƒƒLower melt temperature ƒƒMaterial could be outgassing from elevated temperatures; the molecular structure of the additive has a shorter chain length, which means it will degrade sooner

Figure 12.4 Vent depth chart

Think of the gas in a volcano. The gas must go somewhere or an explosion can result. The same applies to a mold. If the gases inside a mold are compressed, they will eventually cause a small explosion inside the mold and that is where a burn mark comes from.

Figure 12.5 Volcano releasing gases

12.6 Cloudy Parts

„„12.5 Burns in Gates ƒƒBurrs and/or sharp corners at gate ƒƒPolish gate area, eliminating sharp corners or rough areas causing too much shear heating ƒƒGate diameter too small ƒƒGate size should be 50% to 80% of the nominal wall thickness ƒƒAnalyze the concentrate or additive to verify that it is not shear sensitive, creating the problem ƒƒInjection velocity is too fast ƒƒThink back to the shear rate of the material; if the material shoots in too fast through a small gate, it will fracture or break down the molecular chains of the material; understand the shear rate at the gate (see shear rate formula) ƒƒObstruction in the gate ƒƒThere could be a small piece of foreign material obstructing the flow path; when this happens the flow channel now decreases, and when the flow channel decreases the flow rate increases as a result of trying to push the same amount of material through a smaller opening

„„12.6 Cloudy Parts When any clear material is cloudy the main cause is contamination in the material. The other reason can be the cooling rate. ƒƒRemove contaminated material, whether it was introduced through regrind or left from the previous job. The mixing of two dissimilar materials causes the cloudiness of the materials ƒƒCheck regrind for contaminants ƒƒEmpty and clean hopper and loading system ƒƒCheck melt temperature ƒƒIncrease melt temperature –– Reducing stress in part ƒƒIncrease fill speeds ƒƒRemove screw and non-return valve; clean screw/barrel and non-return valve ƒƒIf material is hygroscopic try improving drying conditions ƒƒDecrease cooling rate for a semi-crystalline material to allow the crystalline

121

122 12 Process Troubleshooting

structure less chance to form. Remember the larger the crystalline structure, the less light that will pass through and bounce off creating a cloudy appearance ƒƒThe same material with just different cooling rates (see Figure 12.6)

Figure 12.6 Same material, two different cooling rates (left: fast cooling rate; right: slow cooling rate)

„„12.7 Color Streaks Color streaks happen when the material is contaminated or there is a mixing or degradation issue. ƒƒMaterial contaminated ƒƒPellets or liquid color from the previous job; review where the material is hanging up and eliminate ƒƒScrew/barrel needs to be cleaned ƒƒSkin layer from previous runs or colors will be in the barrel and on the screw. If a new skin layer does not cover the previous skin layer it will constantly bleed out ƒƒNozzle needs to be cleaned ƒƒDead spots may be present in a general purpose nozzle (Figure 12.7); it will need to be removed and cleaned to remove previous color or material

Dead spots

Figure 12.7 General purpose nozzle tip

12.8 Deformation: Ejector Pin Marks

ƒƒManifold/hot tips need to be cleaned ƒƒFrozen skin ƒƒDead spots ƒƒWhen trying to get the previous skin out of the manifold and tips, raise the temperature up 100 °F on the tips and 50 °F on manifold system and cycle machine, noticing the previous color start to be dragged out of the system. Run for 10–15 minutes with new color or natural material, and then shut ­system down and allow to cool. This will create a new skin when the system is heated back up ƒƒSlow screw speeds down ƒƒOver-shearing the color causes it to degrade and change color from the excess heat

„„12.8 Deformation: Ejector Pin Marks Ejector pin marks are caused by a stress or deformation of plastic in a concentrated area where ejection forces are higher than at other areas of the part.

Figure 12.8 Ejector pin mark

ƒƒDesign or tooling issue ƒƒNot enough ejector pins to equalize out force upon ejection ƒƒUndercuts causing part to stick ƒƒA vacuum is created causing unnecessary forces; is the core highly polished? ƒƒNot enough cooling; has the plastic stabilized enough to eject off the core? ƒƒPoor polishing of the detail feature around the ejector pin (less polish could be better for some resins because of the surface tension in the highly polished detail)

123

124 12 Process Troubleshooting

„„12.9 Degraded Polymer ƒƒMelt temperature too high, resulting in burning or degrading material. ƒƒWhen the material overheats it starts to turn color (to brown or black), with a breakdown of the molecular chains, turning long chains into shorter chains ƒƒScrew speed too high (over-shearing material) causing the material to turn brown or black, from generating too much friction and overheating the material ƒƒBreakdown of physical properties of the material (heat or moisture) ƒƒOver- or under-dried material ƒƒOver-dried: loss of physical properties, shorter molecular chain on the material; as the chains get shorter they flow more easily and can break down quicker causing the degraded material ƒƒUnder-dried: moisture in material, causing fracture or breakdown in the molecular chain, also degradation of the material and loss of physical properties ƒƒLong residence time in the barrel ƒƒReduce the time in barrel or take a melt temperature and verify the material is at the correct temperature

„„12.10 Design ƒƒSharp corners, causing shear or stress in the part ƒƒCausing the material in the flow to go faster on the inside corner than on the outside corner, resulting in the material overheating (reduce injection speed) ƒƒUndersized gates ƒƒCreating too much shear in the melt front causing gas or degraded material, or extreme pressures; also can cause premature gate freeze and not allow the cavity to pack out or imprint properly ƒƒDead spots in manifold system ƒƒAllow material to sit and overheat in the manifold system, eventually pulling degraded plastic into melt flow, or restricting the flow in general to create flow issues ƒƒPoorly designed runner system ƒƒSprue orifice too small so cannot handle the flow needed to fill the part ƒƒRunner channels too small, so again not enough flow to fill the part ƒƒRunner channels too large; plastic is compressible and if the plastic is not ­allowed to fill the cavity then it could get compressed inside the manifold ­system

12.11 Fish Hooks

ƒƒFaulty heater or not enough wattage on the heaters to heat material up properly to aid in the flow of the plastic ƒƒPoorly designed water channels ƒƒDepth diameter or pitch incorrect, creating hot or cold spots in the mold itself, or too much flow around the gate causing the gate to freeze off prematurely ƒƒThin to thick section, causing the flow of material to freeze off in the thin section and under-pack the thick section

„„12.11 Fish Hooks

Figure 12.9 Fish hook

ƒƒWith materials such as PET, PETG, polycarbonate, and acrylics, this issue is more noticeable because of the clarity ƒƒFish Hooks are usually caused by cold or unmelted material being dragged along the flow front of the material until it gets trapped ƒƒCheck for drool coming from the tip ƒƒIf running a valve gate, make sure there is no build-up on the pin –– Open the pin for a couple seconds before starting the injection of plastic, to re-melt the frozen plastic ƒƒIncrease or decrease injection velocity ƒƒFlow transition point for the plastic (hesitation in flow front)

125

126 12 Process Troubleshooting

„„12.12 Flash Flash is excessive plastic that was not contained within the parting line of the mold.

Figure 12.10 Flash

ƒƒThere are only two ways to get flash: 1. Damaged mold or tooling 2. More plastic pressure than clamping force ƒƒFirst establish when flash is happening in the injection molding process, first stage or second stage? ƒƒCommon locations: ƒƒEjector pins ƒƒShut-offs ƒƒParting lines ƒƒInserts ƒƒMelt temperature too high ƒƒMelt temperature above manufacturer recommended set point, causing plastic viscosity to go down, making the material thinner, which in turn allows the material to flow more easily ƒƒReduce residence time of material in barrel; the longer the material sits in the barrel the lower the viscosity becomes ƒƒReduce screw speed (minimize shear heating and viscosity changes) ƒƒMold temperature above manufacturer’s set point, causing plastic to go above its stability point and become too thin (instead of a rubber-like state, the plastic is in a water-like state) ƒƒExcessive plastic pressure (fill or pack) ƒƒMake sure part is filling to 95% full when transferring to hold (VP position) (over-pressurizing parting line) ƒƒPack pressure too high causing over-pressurizing of parting line ƒƒProjected area of part too large for available for tonnage

127

12.12 Flash

ƒƒSwitch to a larger tonnage press ƒƒMake sure clamp pressure is higher than plastic pressure at end of fill ƒƒUneven filling ƒƒVerify balance of fill to each cavity (no obstructions in gate) ƒƒVerify heaters and thermocouples are working properly ƒƒVerify tooling –– Vent depth to each cavity and proper location; verify tip height and gate geometry ƒƒRe-verify mold viscosity test –– Viscosity of material can affect filling pattern due to shear; verify at low/ medium/fast settings ƒƒCreate multi-stage injection profile –– Reducing velocity in area of flash will allow plastic to skin over at a lower pressure (fountain flow) ƒƒCreate a multi-stage packing profile –– Reduce pressure in flashing area, then once skin of plastic is frozen, increase plastic pressure ƒƒClamp pressure or clamp not adjusted properly (check tonnage to projected area) ƒƒVerify parting line for flash, damage, or obstruction ƒƒCheck platen parallelism ƒƒCheck single point of press (see Section 3.5 for single point) Increase clamp pressure only if mold is larger than 2/3 the space inside the tie bars. If the mold is less than 2/3 tie bar spacing, decreasing tonnage may help due to platen wrap (where the platen wraps around the mold due to high clamping force). Maximum Average Pressure at Parting Line without Flashing This is determined from the following formula: Maximum average pressure at parting line = (clamp force in pounds)/(part projected area) Example: We have a part that is 3″ × 5″, produced on a 50-ton press. The projected area of the part is 3 × 5 = 15 in2. The clamp force in pounds is 50 × 2000 (2000 lbs. per ton) = 100,000 lbs. Now take the 100,000 and divide by 15 (projected area of the part) = 6667 ppsi.



128 12 Process Troubleshooting

This means that the average pressure inside the mold has to be greater than 6667 ppsi to flash. To establish what the average pressure inside the mold is, take the postgate sensor and the end of cavity sensor and add those together and divide by 2, so if there is a reading of 10,000 at post gate and 2,000 at E.O.C. (end of cavity) that would give us 12,000. Then divide by 2 to get 6,000 ppsi, so it would not flash the mold, because 6,000 is lower than 6,667 ppsi. Now what happens if more projected area or more cavities are added? Let’s say we have a two-cavity mold and now our projected area is 30 in2. Just over 100,000/30 = 3,333 ppsi is needed to flash. If in keeping the same process and with an average pressure inside the mold of 6,000 ppsi, then the mold would flash because 6,667 ppsi is greater than 3,334 ppsi and flash would result. 

„„12.13 Flow Lines Flow lines are the result of the plastic melt flow slowing down as it is being injected into the mold. As the plastic flows from the gate it flows at a certain rate, and as it starts to fill the part, the flow front slows down slightly because now it is filling into a larger area. What is noticed is a hesitation of pressure and flow of plastic, and the ripples are caused by this action and the plastic freezing against the colder steel. When this is observed the velocity will need to be increased to overcome the freezing effect.

Figure 12.11 Flow lines or record grooves

12.14 Hot Tip Drool

ƒƒIncrease injection speed ƒƒFlow lines usually result from a hesitation in the melt front causing ripples or a fingerprint effect on the part ƒƒIncrease mold temperature to help with flow ƒƒThis allows the flow front to move a little quicker to get to the end of fill to start packing the part. When a flow front touches the cold steel it wants to start freezing, but the space in between the ripples starts to sink as it has not touched the surface yet. This is what we are trying to pack out before freeze happens, locking in the ripples due to two different cooling rates. ƒƒIncrease melt temperature to help with flow ƒƒAllows flow front to move a little quicker with a decrease in viscosity ƒƒIncrease pack pressure ƒƒWill help pack out ripples as long as the skin has not frozen ƒƒIncrease pack time ƒƒThis will also help pack out ripples, as long as the skin has not frozen

„„12.14 Hot Tip Drool The main cause of hot tip drool is an over-pressured manifold system. The plastic that was compressed into the manifold to fill and pack the previous shot has not had a chance to relax or depressurize and the pressure will push the plastic out of the gate. It will come in the form of a BB or a round ball of plastic that will need to be removed before molding the next shot or the cavity could block itself off. ƒƒNozzle temperature too high ƒƒReduce nozzle temperature to keep the hot plastic from entering back into the sprue ƒƒMoisture in material ƒƒMoisture can cause pressure to build up in the barrel and force the plastic back into the sprue ƒƒLack of decompression (suck back) ƒƒInternal pressure in the barrel from the back pressure can cause the plastic to migrate back into the sprue; increase suck back until drooling stops. BEWARE: too much suck back can cause bubbles or splay in the material ƒƒBack pressure ƒƒReview manufacturer’s recommendation for back pressure; use enough back pressure to achieve a homogeneous melt, but not too much to over-pressurize in front of the non-return valve

129

130 12 Process Troubleshooting

ƒƒShut-off nozzle ƒƒInvestigate if the machine has a nozzle shut-off and if it is in proper working condition ƒƒIf machine is equipped with a nozzle shut-off a pre-decompression might need to be applied

„„12.15 Jetting

Figure 12.12 Jetting

ƒƒFill rate too fast ƒƒMaterial does not have a chance to adhere to side wall (fountain flow), creating a worm-like flow where the plastic squirts through the cavity until it hits the back wall and then starts to adhere to the mold surface. Slow injection rate when going through gate, then speed up the flow rate once fountain flow has started ƒƒMelt temperature ƒƒMaterial too cold, so skin does not have a chance to lay down on the mold ­surface ƒƒMold temperature ƒƒMold temperature is too low so plastic does not lay down properly across the mold surface ƒƒDesign ƒƒCreate a fan gate (Figure 12.13) to allow some the resistance to the plastic flow and promote fountain flow ƒƒChange or lengthen the cold slug well before material goes into gate (Figure 12.14) ƒƒChange gate location to allow flow front to hit diverter or detail to interrupt the flow

12.16 Long Gates

Figure 12.13 Fan gate

By lengthening the cold slug well, the plas c will flow into the gate at a slower rate (flow front split) while it finishes the filling of the cold slug well, promo ng fountain flow

Figure 12.14 Adding longer cold slug well

„„12.16 Long Gates In a hot runner system, long gates are caused by the plastic solidifying in the gate and plastic being pulled out from around tip.

Figure 12.15 Proper tip height

Make sure the tip is level with the gate when heated. This will aid in the prevention of long gates

131

132 12 Process Troubleshooting

Figure 12.16 Tip too far back

If the tip is too far back the plastic has a chance to freeze and will then be pulled out, creating a long gate or a stump. In a cold runner system, the gate must be cut and trimmed to achieve a smooth feel and appearance. In a tunnel or sub gate, the plastic must be sheared upon ejection. If the gate is not sharp it can cause the plastic to smear leaving a long gate. The common (D) shape gate is good at achieving a flush gate as long as the D is sharp. Tunnel gates can be oval or round also, but the trick is to maintain sharpness with minimal land area.

Figure 12.17 Typical D gate for degating

ƒƒIncrease tip temperature ƒƒThis will keep the plastic hotter at the tip and allow the part to break properly when the mold opens up ƒƒIncrease speed on mold breakaway to break gate ƒƒSharpen gate geometry if tunnel/sub gated ƒƒIncrease mold temperature to create a more flexible plastic (careful with cycle time) ƒƒCheck gate geometry (gate could be too large or washed out) ƒƒCheck tip height; the tip could be too far back

12.18 Parts Sticking in Mold

„„12.17 Nozzle Drool ƒƒDecompression/suck back ƒƒAdd more stroke/speed to the post-decompression ƒƒBack pressure ƒƒReduce back pressure to minimize the pressure built up in front of screw ƒƒMoisture in material ƒƒMoisture can cause pressure to build up in the barrel and force the plastic back into the sprue ƒƒProper nozzle tip ƒƒMake sure to have the proper radius ƒƒThe proper orifice (opening) ƒƒThe proper style (GP, nylon)

„„12.18 Parts Sticking in Mold When a part sticks in the mold it can be in the front half or back half. Investigate where the part is sticking and resolve accordingly. ƒƒOver-packing material in mold ƒƒReduce pack pressure and time, reestablish gate freeze, and set timer accordingly ƒƒInsufficient draft ƒƒAdd more draft if possible (draft and reverse draft) ƒƒEjection temperature ƒƒMake sure plastic is below the HDT, the temperature at which the plastic can be ejected without being distorted ƒƒUndercuts ƒƒUndercuts can be too deep; look at part geometry and add more draft if possible ƒƒSprue sticking ƒƒPack time too long; gate has frozen off and the remaining pressure is concentrated in the sprue ƒƒEjection stroke/pin length ƒƒAdd more ejection stroke ƒƒEjector pins could be sticking in part; recess pins 0.005″ to create a small pad for ejector pin to push against

133

134 12 Process Troubleshooting

ƒƒDesign ƒƒCheck for reverse draft on mold ƒƒCheck for components not lining up properly

„„12.19 Pulls Pulls or deformation of the part can happen when there is tooling damage or the part does not eject out of the mold straight. ƒƒDamage to mold ƒƒVerify that the tool does not have damage or an undercut that could be causing the part to stick and deform; watch for EDM marks ƒƒOver-packing ƒƒWhen the part gets over-packed it can cause unequalized forces on the part causing the part to not come off the core parallel ƒƒDesign issue ƒƒWhere there is minimal or no draft on the geometry of the part, pulls can result if over-packed ƒƒMold temperature ƒƒIncrease the mold temperature on the ejection side of the mold; this will allow the material not to shrink as much ƒƒSurface finish ƒƒBe aware of surface tension on the parts; some materials like a rougher surface to eject the parts smoothly and other materials like a smooth or polished surface to eject the parts

„„12.20 Shorts/Non-Fills Shorts or non-fills mean just a lack of material that has entered the cavity to fill and pack the part out. ƒƒMold temperature ƒƒIncrease mold temperature to help the plastic flow further; this helps delay the freeze off of the material ƒƒMelt temperature ƒƒIncrease melt temperature to help lower the viscosity of the material (more so in amorphous resins than semi-crystalline resins) and have it flow further

12.21 Sinks

ƒƒInsufficient shot size ƒƒMake sure the screw is not bottoming out –– Add shot size as needed, reestablishing a 95% full part –– Whatever the shot size is taken up, increase the transfer position the same to maintain the 95% full part at transfer ƒƒPoor venting ƒƒMake sure to have proper venting, depth, land, and location; it could be a gas trap that has not charred or burned ƒƒGate diameter ƒƒIncrease gate diameter to increase flow rate into cavity ƒƒRestricted flow ƒƒGating from thin to thick causing the flow front to freeze off and not be able to fill or pack out the thick section? ƒƒPack pressure ƒƒMake sure to have sufficient pack pressure to overcome the pressure internal to the cavity and not stall or hesitate the flow front ƒƒIncrease injection velocity (fill rate) ƒƒCheck for blockage ƒƒFlow length

„„12.21 Sinks A sink (sink mark) is caused by the outside frozen skin, as a result of improper set-up, allowing the plastic to shrink when the packing/holding phase is complete. The frozen layer against the mold cannot withstand the forces of the plastic ­shrinking and has a chance to draw in causing a sink or depression on the molded surface.

Figure 12.18 Sink in thick section

135

136 12 Process Troubleshooting

ƒƒInsufficient shot size ƒƒMake sure the screw is not bottoming out; increase shot size as needed, and remember to increase transfer position the same to maintain a 95% full part at transfer ƒƒImproper design ƒƒUniform wall sections are the key to minimizing sink. Avoid thick sections around bosses, intersecting walls, gating thin to thick ƒƒImproper cooling (not enough flow channels) or not close enough to surface to draw heat out properly ƒƒInsufficient pack or hold pressure ƒƒIncrease pack pressure or hold pressure to overcome internal cavity pressure ƒƒInsufficient pack or hold time ƒƒRe-verify gate seal study ƒƒGate or runner design ƒƒReview gate and runner design; increase gate diameter if possible, and remember to re-verify gate geometry making sure the land area is still ok and not causing long gates ƒƒReduce mold temperature ƒƒThis will allow for a thicker wall being formed as the plastic flows through the cavity ƒƒSlow injection velocity ƒƒThis will increase the frozen skin layer of the plastic flowing through the ­cavity because of a decrease in shearing ƒƒReduce melt temperature ƒƒReducing the melt temperature will reduce the cooling rate of the plastic (cooling rate is determined by the delta or temperature difference between the plastic and the steel temperature)

12.22 Splay

„„12.22 Splay

Figure 12.19 Splay mark

Splay can be caused by a breakdown of the molecular chain of the plastic. The material can be degraded through processing, too fast an injection velocity, too much back pressure, too fast of a screw speed, too high temperatures, too much time in the barrel (residence time) or barrel temperature set above manufacturers recommended set point. If the material is not dried properly (temperature too low or too high) or sits in the hopper for too long a period at elevated temperatures, that can also lead to splay.

Figure 12.20 Molecular fracture

ƒƒMoisture in material ƒƒMoisture causes the material to degrade and breaks down the molecular structure of the material. Let’s understand where the moisture is coming from? Is the plastic not dried properly? Or is there condensation building up in the feed throat? ƒƒMelt temperature too high ƒƒHigh temperature causes the material to degrade and breaks down the molecular structure of the material

137

138 12 Process Troubleshooting

ƒƒShear heat ƒƒMoisture sources other than the material ƒƒMoisture causes the material to degrade and breaks down the molecular structure of the material ƒƒMoisture on mold face ƒƒMoisture in feed throat ƒƒThere are certain materials that can generate water if processed improperly; polycarbonate and nylon are classic examples

Figure 12.21 Polycarbonate unit molecular structure

Figure 12.21 shows the molecular structure for polycarbonate. Water reacts with the polycarbonate molecules at the C—O bonds, fracturing the chains. A similar situation occurs with nylon (Figure 12.22).

Figure 12.22 Nylon unit molecular structure

„„12.23 Sprue Sticking ƒƒNozzle orifice larger than sprue orifice ƒƒChange nozzle tip to reduce the orifice to smaller the sprue orifice on mold; the nozzle orifice should be 1/64 or 1/32 smaller

12.24 Surface Imperfections

ƒƒInsufficient taper on sprue ƒƒIncrease taper on sprue –– The draft on the sprue should be a minimum of 2° –– Increasing the taper will increase the size (volume) and there might be a need for more cooling time due to the increased mass of plastic ƒƒDamage to sprue ƒƒClean up damage to sprue with a tapered reamer, making sure the diameter is not increased too much, which would require adding cooling time and lengthening the cycle to solidify the plastic. If the sprue is enlarged, increase the nozzle orifice to increase flow and decrease pressure ƒƒOver-packing of material in sprue ƒƒPack time might be too long –– As the part freezes off and with the continuation of pack, the plastic has to go somewhere that is still molten –– Re-verify gate freeze study and enter proper time ƒƒNozzle temperature ƒƒIncrease nozzle temperature to allow material not to solidify while the nozzle sits against the mold ƒƒConsider adding an insulator in between the nozzle and mold (this could be cardboard or Kevlar) ƒƒRadius damaged ƒƒWhen backing the barrel off the mold, care must be taken when bringing the nozzle back in as it seats against the mold; there will always be some drool that will leak out due to the manifold relieving the pressure. ƒƒIn re-radius making, be sure to use the proper radius of ½″ or ¾″

„„12.24 Surface Imperfections Get a good understanding of what the surface imperfection is. Is it an imperfection in the plastic material that has made its way to the surface along the flow front, or is it something on the mold surface that has now transferred to the molded part? ƒƒMelt temperature too low ƒƒA low melt temperature will cause the melt viscosity to be higher (melt is thicker), in turn causing the melt front to be colder creating hesitation ƒƒMold temperature too low ƒƒA lower mold temperature can create a dull surface appearance; increase mold temperature to improve appearance

139

140 12 Process Troubleshooting

ƒƒInsufficient pack ƒƒUnder-packed parts will have different surface appearances due to the plastic shrinking away from the molding surface and also different cooling rates ƒƒInjection fill rate ƒƒIncrease injection fill rate to have material lay down and avoid the fountain flow marks or thumb print marks that would be visible by having too slow of an injection velocity ƒƒContamination on mold surface ƒƒWatch for moisture or grease on molding surface; make sure ejector pins, lifters, slides, and unscrewing cores are not allowing grease to migrate to the surface ƒƒInsufficient material volume ƒƒNot packing part out causes a dullness on the surface ƒƒInsufficient venting ƒƒCausing gas to be trapped in cavity and creating a film or haze on molding surface ƒƒMoisture in material ƒƒCausing splay marks ƒƒImproper steel finish ƒƒVerify that the mold has the proper finish to begin with, and that the polishing of the mold has not affected the finish in any way

„„12.25 Voids A void occurs when the plastic cools and the outside walls have solidified and will not move. As the plastic in the center continues to cool and shrink, it actually tears apart from itself, creating what looks like a bubble on the inside of the part, which appears mostly in thicker-wall sections. The easiest way to troubleshoot a void is to apply heat to the plastic. A void will shrink when heat is applied because the molecules can relax when heated and come back together.

12.26 Warpage

Figure 12.23 Vacuum created by shrink, or plastic ripping apart

ƒƒPack pressure/hold pressure too low ƒƒIncrease pack pressure/hold pressure to minimize the shrink of the plastic by packing more molecules into the part; less room for the molecules to move will minimize shrinkage ƒƒPack/hold time too low ƒƒIncrease pack/hold time to minimize the shrink of the plastic to promote a sealed gate so plastic cannot escape or discharge from the part ƒƒMold temperature ƒƒHeating up mold will reduce the cooling rate of the plastic and minimize the differential shrinkage and stresses noticed at the wall and internal to the part; this is especially true for thick-walled parts ƒƒGate size too small ƒƒIncrease gate size to achieve a better pack rate, and maintain flow into the cavity without freezing off ƒƒRunner size too small ƒƒIncrease runner size to help maintain the flow to the gate without freezing off ƒƒPart design ƒƒRedesign part to have a thinner wall section or gate into the thicker section of the part; you never want to gate thin to thick, as this will promote freeze off in the thin section and reduce the packing into the thick section

„„12.26 Warpage Warpage in the easiest terms is differential shrinkage. It can be the hardest problem to overcome. First, understand where the differential shrinkage is coming from. Is there orientational shrinkage, relating to how the molecules are aligned in the part? Or is it compressive shrinkage, caused by an over- or under-packed

141

142 12 Process Troubleshooting

condition on the part that is causing differential shrinkage? Or is it tensile shrinkage, which is due to different cooling rates within a part (thick or thin sections)?

Figure 12.24 Typical warpage

ƒƒMaterial temperature ƒƒThe material temperature must be below the HDT or heat distortion/deflection temperature in order to have the part stable enough to eject off the mold. The configuration and design of the mold can also cause one side of the mold to run warmer than the other even though the mold halves are set at the same temperature ƒƒDifferential mold temperatures ƒƒHaving one side of the tool warmer than the other will cause differential shrinkage because of the hot and cold sides will shrink at different rates. Parts will continue to shrink towards the warmer side when outside the mold ƒƒNon-uniform shrinkage ƒƒIs caused when different temperature remains in the part. With thick and thin sections, the hotter sections will continue to shrink causing warp ƒƒPack/hold times or pressures ƒƒPressures or time will cause differential shrinkage (too much or too little), because the material furthest away from the gate will receive less packing than that closest to the gate. Sometimes the pack and hold pressure must be profiled in order to equalize out the packing rates within the part ƒƒParts can shrink differently in the direction of flow compared to the transverse direction. This is anisotropic shrinkage, and happens in semi-crystalline resins in particular. (Isotropic shrinkage, which is shrinking the same in both directions, happens in amorphous resins.) This has to do with orientation of the molecules and the crystals in the molecular chain of the polymer; the crystal has to stretch out to flow, but will act like a spring when it cools down

12.27 Weld Lines

ƒƒShear ƒƒIncreasing flow rate will cause the end of fill melt temperature to be higher than at the gate; slowing down the velocity will equalize out the melt temperature throughout the part ƒƒEjection system design ƒƒUndue or differential forces on the part will also cause warpage because of the stresses created from the forces upon ejection ƒƒGate location ƒƒMove gate location to change location or direction of orientation ƒƒResin ƒƒUse a resin with less shrink; the lower the shrink rate the lower the warp due to differential shrinkage ƒƒCooling time ƒƒIncrease cooling time to help minimize the shrinking differential seen in the plastic

„„12.27 Weld Lines Weld lines are where two melt fronts come back together after going around some geometry on the part. As the melt fronts touch they hesitate for a split second in the center. When this hesitation occurs, the material starts to cool down and the hotter material will go to the outside, also having a thinner skin towards the outside; hotter material will create a better packing situation, but also note that the orientation of the plastic molecular chains will not overlap and a weak spot or line will be created. ƒƒMelt temperature ƒƒIncrease melt temperature to allow better bonding of the two plastic flow fronts, but be careful not to raise the melt temperature above the degradation point of the material ƒƒMold temperature ƒƒIncrease mold temperature to allow better bonding of the two plastic flow fronts ƒƒInsufficient pressure at end of fill ƒƒIf the end of cavity pressure too low, increase pack pressure to get better ­bonding of the two melt fronts; the two hottest areas will be towards the wall section and these areas will receive the greatest pack pressure

143

144 12 Process Troubleshooting

ƒƒAir trap ƒƒAdd vents at end of fill to allow the two melt fronts to knit properly. Make sure that the vent is in the proper location. A poorly placed vent serves no purpose ƒƒFill rate too slow ƒƒFlow fronts do not have a chance to reach the end of fill, and also the material wants to freeze the further away from the gate. Make sure that shear thinning of the material is happening properly, by increasing the flow rate and increasing the shearing thinning of the material will increase the temperature at the flow front, increasing the strength of the weld line ƒƒFlow distance from gate ƒƒThe flow distance from the gate might be too long (L/T, length divided by thickness, should be 250 : 1 or less). When L/T is above 250 : 1 it is considered thin-wall molding. Decrease the flow length or add another gate, which will automatically create a weld line ƒƒGate location ƒƒPlan the gate location so that the two melt fronts that form the weld line continue on the flow path ƒƒPart thickness ƒƒBy increasing the wall section in certain areas of the part, the weld line can be moved by changing the flow pattern (flow leaders)

Figure 12.25 Weld line forming

13

What is Important on a Set-Up Sheet?

Let’s use the four plastic variables: plastic temperature, plastic flow rate, plastic pressure, and plastic cooling. 1. Plastic Temperature Melt temperature – Understanding what the temperature of the plastic is, and not the barrel temperature. Recovery time – How fast is the screw turning to get the screw back before the cooling timer ends, but also how much is the material being sheared and heated up? Back pressure – How consistent is packing the material in front of the screw and how much plastic pressure does it take to push screw back? Volume of shot – in3 or cm3 to understand the shear rate. Charge delay – Add some screw delay to relieve the pressure on the non-return valve before the screw starts to rotate; this will help minimize damage to the non-return valve, and will also minimize residence time if there is a quick recovery of screw. 2. Plastic Flow Rate Fill time – Matching the fill time ensures that the flow rate of plastic is the same, even with changes to the machines or barrel sizes. Fill only part weight – Once a fill only part weight is established, then use this information every time there is a start-up; this will help with consistency. Mold shot (ppsi) – Mold shot in plastic pressure is the transfer or switch over pressure it takes to fill a 95%–98% part; it must be converted to plastic pressure. Air shot (ppsi) – This must be done with a purge disk and in a semi-automatic or automatic cycle; remember that the machine will reduce to pressure in manual when purging.

146 13 What is Important on a Set-Up Sheet?

3. Plastic Pressure Transfer pressure – The pressure at which the machine switches from velocity control to pressure control. Pack pressure (ppsi) – How much pressure it takes to pack to part out. Pack time – The time at which no more plastic is entering the cavity at that pressure. Hold time – How much time does it take the gate to seal or freeze to have a dimensionally stable part? Sometimes the gate will not be completely sealed. Hold pressure (ppsi) – How much pressure does it take to pack out the part to have a dimensionally stable part? Gate seal time – This is the time it takes to achieve true gate seal. Final part weight – After the process has been optimized and the dimensions are good, then a final part weight should be recorded. This will also help with troubleshooting. If the part is lighter then it is under-packed, whereas if the part is too heavy then it is over-packed. 4. Plastic Cooling Cooling temperature in – This is the temperature of the water going into the tool. Cooling temperature out – This is the temperature of the water as it exits the tool. The difference in temperature compared to the water going in should be 4 °F or  less, or 2 °F or less on critical jobs. This confirms that enough heat is being ­removed out of the tool. Cooling timer – The amount of time it takes to stabilize the part for ejection. Coolant flow – Measured in gallons per minute (GPM) or liters per minute (LPM) per line. Cycle time – Achieving the quoted cycle time. Clamp open/closed time – This is total time for the clamp to open, eject, and close. Clamp force (tons) – This is developed from the projected area of the part and the material tonnage factor from the material supplier. The last things needed are as follows: Shot size – Documented in inches or mm. Transfer position – Where does the machine switch from 1st stage pressure to 2nd stage or from fill to pack/hold? Nozzle orifice – What size is the opening on the nozzle tip? It must remain the same for all runs. If not, transfer pressure will change.

13 What is Important on a Set-Up Sheet?

Mold number Material type Number of cavities Date Color Name

147

14

Commonly Used Conversion Factors and Formulas

„„14.1 Conversion Factors

150 14 Commonly Used Conversion Factors and Formulas

„„14.2 Common Formulas for Injection ­Molding Volumetric shot = (screw dia. × screw dia. × 0.7854) × screw travel Expected fill time =

Actual speed =

(shot size + suck back) − transfer fill rate or fill speed

(shot size + suck back) − transfer actual fill time

Calculating clamp force =

area of clamp cylinder × max hydraulic pressure 2000 lbs. (1ton)

Max pressure at parting line before flashing = area of clamp cylinder × max hydraulic pressure

( )

projected area of part (s) in2

How many tons does cavity require = projected area of part(s) × material tons per in2 area of screw ´ linear distance traveled Volumetric flow rate (Qp) = fill time Volumetric shot size = area of screw × linear distance traveled Injection speed linearity =

expected fill time − actual fill time ×100 expected fill time

14.2 Common Formulas for Injection ­Molding

Triangle Formulas The triangle formulas provide an easy way to calculate of each of the quantities within them from the other two. Example: If there is a need to solve for distance, hold a thumb over distance in Figure 14.1, and the formula is: Distance = time × speed Or to solve for time, hold a thumb over time in Figure 14.1: Time = distance/speed

Figure 14.1 Triangle formula for distance, time, and speed

Figure 14.2 Triangle formula for force, area, and pressure

151

152 14 Commonly Used Conversion Factors and Formulas

Figure 14.3 Triangle formula for mass, density, and volume

Figure 14.4 Triangle formula for clamp force, projected area, and maximum average pressure

15

Machine Set-Up

Step 1: Turn barrel heat on (material mid-range available) Step 2: Turn die heat on (warm mold) (material mid-range available) Step 3: Adjust clamp movement 1. Adjust die height 2. Adjust open stroke a) Stroke (adjust for optimization and safety) b) Speed (adjust for optimization and safety) c) Pressure 3. Adjust closing stroke a) Speed (adjust for optimization and safety) b) Pressure (set low) c) Clamp force I) Measure projected area of part and runner II) Multiply by 2–5 tons per in2 (material specific) III) Can lower from there until it flashes—then increase until flash stops d) Set mold protect I) Check with business card Step 4: Adjust ejectors 1. Speed (adjust for optimization and safety) 2. Distance 3. Pressure 4. Number of strokes (try to keep to one stroke)

154 15 Machine Set-Up

Step 5: Adjust metering Note: Set injection pressure—boost at the press’s maximum (do not pressure limit process; also be careful not to flash tool; use safe start-up formula for shot size). 1. Shot size (see formula) a) Zero out all pack/hold pressures (some machines do not like zero; must enter value) b) Set injection speed at the middle of the available range (something safe) c) Start transfer position high enough so as not to flash mold d) Establish a shot size that provides a 5–10% cushion. Increase shot size by increasing stroke (remember to maintain 95% full part (visually)) I) Then add that time to the screw rotate delay (back pressure and screw speed will fill part if activated right after fill) e) RPM (set fast enough to get screw back before cooling ends) 2. Back pressure (500–1000 plastic pressure) 3. Suck back (reduce drool at gate) (0.25 in or 6 mm; need to understand how far the non-return valve moves) a) Speed (adjust accordingly) Step 6: Perform mold viscosity test 1. Find the maximum injection velocity for mold (if the machine does 100 in/sec, is that needed?) 2. Set press at maximum injection pressure—boost (should be there from the last test) 3. Re-establish a 95% full part at maximum speed/velocity by adjusting transfer position (up or down) a) Use auto cycle b) Then go down on injection velocities in 10 separate increments 4. Chart should start with the maximum injection velocity a) Have at least 10 data sets in the chart b) Example: if maximum velocity is 5.0 in/sec, then it would be 5.0 divided by 10 or 0.5 in/sec increments c) Last few sets should be set low (0.5, 0.2 in/sec) or fill time above 6.0 sec. This develops the shape of the curve d) Get the best shape of the curve (see viscosity curve procedure) 5. Three shots for each cavity at each injection speed (allow two shots between speed changes)

15 Machine Set-Up

6. Graph the results a) Pick the optimum injection speed at the flattest part of the curve (see Chapter 6), but not at the curves end (this will tell if a velocity change is made, it will have minimal effect on the viscosity of the material) Step 7: Dynamic non-return valve 1. After viscosity test, grab 10 shots and weigh with runner (95% full) 2. Fill out chart Step 8: Injection speed linearity 1. Use data from viscosity test and graph Step 9: Load sensitivity 1. Use fill time and pressure from optimum velocity 2. Back barrel off, install purge plate 3. Bring nozzle up and make contact with purge plate, and re-zero injection unit 4. Put machine in semi-auto or automatic, run one shot 5. Collect fill time and transfer pressure for air shot Step 10: Manifold balance 1. Establish a 95% full part at selected velocity/injection speed from mold viscosity test a) All tips should be at the same temperature 2. All parts should be visually short (95% full or less) 3. Run mold and take three shots from each cavity 4. Artificially balance the mold by increasing or decreasing the tip temperature of each cavity until all parts fill the same (new mold) a) Note: Do not go more than 25 °F from start settings on new mold 5. Then weigh three shots off each cavity and complete the chart Step 11: Gate freeze Note: Before starting, make sure pack/hold pressure is 50% of transfer pressure (just a good place to start!) 1. Establish a 95% full part at zero pack and hold time 2. Set pack and hold pressure at half the peak boost pressure 3. Change injection hold time per the chart a) Note: Raise hold time and lower screw delay time and cooling time 4. Weigh full shot without runner

155

156 15 Machine Set-Up

Step 12: Static non-return valve 1. Grab five cushion positions after gate freeze has been optimized 2. Graph chart Step 13: Cooling time study 1. Take cooling time down until parts distort or are no longer functional 2. Use pyrometer to measure part temperature (HDT) Step 14: Process 1. Set process to the gate freeze hold time (optimum) 2. Set process to the injection speed from the mold viscosity test (optimum) 3. Range find for pack and hold pressure a) Take pressures down until sinks/shorts are visible—this is the low limit. Then come up with the pressure a little (50–100 psi or 500–1000 ppsi) b) Take pressures up until flashing is visible, being careful not to damage tool. Then go down a little (50–100 psi)—this is the high limit c) This is the natural window range—middle is the centerline 4. Set screw delay to 0.5 sec minimum 5. Range find for cooling time a) Take the cooling time down until sinks or distorted part—this is the low limit b) Add a second or two for centerline and another for the maximum c) Make sure screw recovery time does not limit the cooling time. Screw should recover before cooling is complete (1–2 seconds) Step 15: Residence time (ideally between 1 and 3 minutes) ƒƒIf below 1 minute residence time, profile barrel temperature low at the nozzle end and high at the hopper end ƒƒIf residence time between 1 and 3 minutes, leave all the same—straight profile ƒƒIf residence time above 3 minutes, profile barrel temperature high at the nozzle end and low at the hopper end Final safety set points: ƒƒSet the transfer position/shot size so to maintain 5% to 10% cushion ƒƒSet high limit injection time 0.1 sec above fill time ƒƒInjection pressure 10% above actual transfer pressure ƒƒOverall cycle timer 10% above actual ƒƒTake melt temperature Used by permission of Gary Mitchell

16

Things That Hurt the Bottom Line of a Company

Long cycles: Cycle times are a concern and play a huge role in the profits of the company. A 10% increase in cycle time will decrease profits by 50%. The same goes for a quicker cycle time: a 10% decrease in cycle time will give a 50% increase in profits. Blocked cavities: These are a huge profit killer because of the fact that it will take longer to produce the same amount of cavities. It will also change the flow pattern inside the mold, and can cause dimensional variation. Downtime: Downtime is considered as any time the mold is down and not making a quality product. This includes mold changeovers, maintenance issues, material issues, or bad procedures. The cost of rejects: Once a bad product is made, you have to find it, which takes time, grind it, again a cost, and remake it. This all takes time and money. It takes four good parts to make up for the one bad part produced. Floor operators not working in the optimum conditions: Is their work cell flow efficient? Do they have all boxes and packaging needed to complete their task? Can the mold/tool be fixed to eliminate some of the secondary work being performed to make a quality product? The floor operator has a tough enough job running the press for 8–12 hours. Anything to minimize fatigue and make their job easier will only result in a superior quality product.

17

Terms and Definitions

Terms and definitions from RJG Master Molder series Accumulator: An auxiliary oil source for fast delivery of plastic melt. Used on injection molding machines. Where oil is stored in a container under pressure and used on a molding machine to boost injection rate. Additive: Material or compound added in small amounts to resins or to improve a plastic’s performance during processing, or to enhance a plastic’s performance for end use. Air shot: Injecting plastic into the air from the nozzle of an injection molding ­machine. Amorphous: Without structure. Amorphous polymers: A family of polymers characterized by the randomness of entangled polymer chains. Will have lower shrinkage than semi-crystalline materials due to having no crystalline structure. Anisotropy: The tendency of a material to react differently to stresses applied in different directions, especially with respect to flow orientation. Annealing: Applying heat at a specific temperature, then gradually cooling to relax stress with no shape distortion in an injection molded part. Antistatic agent: An additive placed in the resin at a low percentage or substance applied to the surface of the plastic part for the purpose of eliminating or lessening static electricity. Can also be used as a lubricating agent when leeched to the ­surface. Back pressure: The hydraulic pressure developed in the back side of an injection cylinder of an injection molding machine which must be overcome by the plastic pressure in front of the screw to start moving the screw back, or otherwise the screw stalls. Barrel: Cylinder or tube portion of the injection molding machine. The cylinder forms the chamber within which the plastic resin is converted from a solid form into a viscous melt. The barrel also contains the reciprocating screw or plunger.

160 17 Terms and Definitions

Barrel capacity: The maximum weight of material a machine can produce from one forward motion of the ram, screw, or plunger. Blind hole: The hole that is formed or molded or drilled into a part but not entirely through the part. Blowing agent: Additive for resins to be foamed. When heated to a specific temperature, it decomposes to yield a large volume of gas that creates cells in foamed plastics. Bluing: A transfer paste that is applied to one side of the tool to verify shut-offs or contact with the other side. When the tool is closed and clamp tonnage applied, the transfer paste or bluing will transfer to other side. Thickness is important: too thick will give a false reading. Boss: A small projection from a part’s surface designed to add strength, facilitate alignment with another part during assembly, or permit attachment to another part. May result in sink if not designed properly. Bubble: Gas or air trapped within a molded plastic product. The bubble rides in the melt stream and gets encapsulated within the part. Differs from a blister, which appears just under the surface. Also differs from a void, which is created by a vacuum from the shrink of the part during cooling. The plastic rips away from itself. Burned: Showing evidence of thermal decomposition with some discoloration. The gases that are not allowed to escape through venting get compressed and ignite (dieseling). Cartridge heater: Cylindrical-bodied electrical heater for providing heat for molds, injection nozzles, hot runner molding systems, or hot stamping dies. Cavity: A detailed impression in a plastics mold which forms the outer surfaces of the molded parts. Usually on the front half of the mold. Check ring: The sliding ring of the non-return valve on the front of the screw which, together with the seat, allows the flow of melted plastic forward in front of the screw during rotation and prevents flow back over the flights during injection. A high wear item that often leaks during injection. Chiller: A self-contained system comprised of a refrigeration unit and a coolant reservoir with a pump. Chillers maintain a colder temperature than tower water to increase the cooling rate and minimize cooling time. Clamping area: The largest molding area an injection molding machine can hold closed under full pressure. Clamping force: The force generated by the clamp on the mold to hold it closed against the forces of the injection pressures. Clamping plate: A portion of the mold used to fasten the mold to the machine. Clamping pressure: Pressure which is applied to an injection mold to hold it closed.

17 Terms and Definitions

Clamping system: Part of the injection molding machine that provides the capability to open and close the mold (front half and back half platen), to hold the mold closed during injection, and to eject the part. Clamping tonnage: Rated clamping capacity of an injection molding machine. Color concentrate: The plastic resin which contains a high loading of pigment. Concentrates come in pellet or liquid form. Colorant: Dyes and pigments used to color plastics. Compounding: The process of mixing the polymer with all the materials necessary for the finished resin to be shipped to the processor. Compression ratio: The depth of the feed flight section versus the depth of the metering section of a reciprocating screw. It is an important indication of the degree to which the plastic will be compressed as it is conveyed from the feed to the metering section of the screw. Compressive stress: The compressive load per unit area of original cross-section carried by the specimen during the compression test. Conversion process: The process of converting thermoplastic pellets into parts. Cooling fixture: A fixture used to distort the part to maintain the shape or dimensional accuracy of a molded part after it is removed from the mold. Copolymer: A compound resulting from the chemical reaction of two chemically different monomers with each other. Core pin: A pin which is used to make a hole in a molded part. Crystalline polymers: A family of polymers characterized by areas of order in which the molecular chains line up and lay tightly together in an otherwise amorphous mass. Crystallinity: A state of molecular structure in some resins denoting uniformity and compactness of the molecular chains. This characteristic is attributable to the existence of solid crystals with definite geometric form. Crystallization temperature: The temperature at which a crystalline resin begins to recrystallize upon cooling. Cure time: In the molding of plastics, it is the time it takes for the material to get below the heat distortion temperature so the part can be ejected without distorting. Cushion: The amount of plastic left in front of the screw at the end of fill-pack-andhold Cycle: One full sequence in a molding operation, from the clamp closed, inject, pack, cool, open, and eject. Daylight opening: The distance that a mold opens to safely eject the parts.

161

162 17 Terms and Definitions

Degradation: A change in the chemical structure of a plastic that is reflected in the appearance or physical properties. Many times caused by excessive time at an elevated temperature or improper processing of material. Desiccant: An adsorbent material installed in the desiccant beds of a drying unit to remove the moisture from the air being supplied to the drier. Dimensional stability: The ability of a plastic part to retain the precise shape in which it was molded. Draft: The tapered design of a mold wall which facilitates removal of molded parts. Any angle that is not “0” positive or negative. Every 1 degree of draft equals 0.017453″. Drooling: Leakage of material from a nozzle, sprue, or gate in injection molding. Dynamic: During flow, which is always before the cavity is filled. Dynamic Pressure: The pressure being built up while flow into the mold during fill is occurring. Effective viscosity: The processing viscosity of a material in response to all process variables as well as material characteristics. Ejection: Removal of the molded part from the mold by mechanical, hydraulic, or compressed air force. Ejection ram: Small hydraulic piston(s) which is attached to and operates ejector plates. Ejector plate: The metal plate in the mold used to operate ejector pins; designed to apply a uniform pressure through a guided system in the process of ejection. Elongation: The fractional increase in the length of a material stressed in tension. Fan gate: An opening between the runner and the cavity which has the shape of a fan. This shape helps reduce stress in the gate area by spreading the flow over a wider area. Also used to reduce jetting. Fatigue strength: The maximum cyclic stress a material can withstand for a given number of cycles before failure occurs; the residual strength after being subjected to fatigue. Feed section of the screw: The first section or zone of a screw, which is fed from the hopper and feed throat, which picks up pellets and carries them forward. Fiber orientation: Fiber alignment, where the majority of fibers are in the same direction, resulting in a higher strength in that direction. Fiberglass reinforcement: The major material used to reinforce plastics. Available as a mat, roving, fabric, etc. It is incorporated into both thermosets and thermoplastics. The glass increases mechanical strength, impact resistance, stiffness, and dimensional stability of the matrix.

17 Terms and Definitions

Fill: Filling a cavity or cavities on first stage/boost. Does not include pack or hold. Fill time: Time to fill a cavity or cavities to a 98% fill. Does not include packing or hold. Filler: A material which is added to plastics to make it less costly. Fillers can be glass, talc, or any inert substance which can alter various properties of the plastic. Flame retardant: Reactive compounds and additive compounds to render a polymer fire retardant. Reactive compounds become an integral part of the polymer structure, while additive chemicals are physically dispersed in the polymer. Flash: The excess plastic around the area of the mold parting line on a molded part. Where there is damage to the mold or the plastic pressure has overcome the clamping force. Flexural modulus: The ratio of applied stress to strain in outer fibers of the plastic specimen during flexure. Flexural strength: The resistance of a plastic material to cracking or breaking during bending. Flight: The outer surface of the helical ridge of metal on an extruder or injection molding screw. Flight depth: The distance from the edge of a flight to the core of the screw. The flight depth of an injection molding screw is greater at the narrower feed section than at the wider metering section. Flow: The movement of the plastic; a measurement of its fluidity. Flow line: The area of a molded part where two melt fronts come together during molding. Flow marks: Distinctive surface marks caused when two flow fronts meet and weld together during molding. Force: Defined as pressure × area. Gate: In injection molding, the channel through which the molten resin flows from the runner into the cavity. Generally, it is small and is where solidification first occurs. Gate mark: A blemish on the molded part left by the mold gate. Glass transition temperature: The approximate midpoint of the temperature range at which a non-crystalline (amorphous) polymer changes from brittle (glassy) to rubbery. Guide pin: A pin which guides mold halves into alignment on closing. Also called a leader pin. Guide pin bushing: The bushing into which the guide pin mates upon closing of the mold.

163

164 17 Terms and Definitions

Heat deflection temperature (HDT): The temperature at which the plastic starts to deform. Heater band: Electrical heater used as the primary source of heat on barrels and nozzles of injection molding machines and extruders. Homopolymer: A polymer that is formed by the polymerization of a single monomer. Hopper: In injection molding, the container holding a supply of molding material to be fed to the screw. Hopper blender: Mixes multiple materials such as virgin resin, regrind, blowing agents, fillers, and colorants. Materials to be blended are metered in ratio to a mixing chamber and then discharged into the feed throat of the processing machine. Hopper dryer: A combination feeding and drying device for injection molding of thermoplastics. Dry hot air flows upward through the hopper to drive moisture out of the pellets. Hot sprue bushing: Mold component that contains a heating element to keep the resin melts hot within the bushing. The bushing is inserted into the mold to provide a hot channel between the molding machine nozzle and the mold cavity. Humidity: The moisture in the air. Hydraulic clamp: Used in a variety of molding and forming machines. The hydraulic clamp consists basically of a high-speed, variable hydraulic pump, valving, a fast-acting cylinder, and a high-pressure cylinder. Cylinders can be single or combination units. The clamp closes the mold halves to form the part. Hydrolysis: The splitting of a molecule with the addition of water in the presence of heat and pressure, as in processing. Decomposition of a substance by reaction with water. Hydrophilic: Capable of absorbing water. Hydrophobic: Capable of repelling water. Hygroscopic: Capable of absorbing and retaining environmental moisture. Impact resistance: A material’s ability to absorb a great amount of energy before breaking. Impact strength: Ability to withstand shock. Impact test: Measures the energy necessary to fracture a standard notched bar by an impulse load. Injection pressure: Pressure applied to the injection ram to force the plastic from the barrel and into the mold (measured in psi). Insert molding: The process by which components, such as terminals, pins, studs, and fasteners, may be molded into a part.

17 Terms and Definitions

Isotropic: The ability to react the same regardless of the direction of measurement. Isotropic materials will react consistently even if stress is applied in different directions. Stress-strength ratio is uniform throughout the material. Jetting: Instability caused by improper gate design and too high flow rate. Fountain flow does not have a chance to form and it looks like work tracks on the plastic part. Knockout bar: A bar which holds and actuates ejector pin(s) in the mold. Used in the ejection of molded piece from the mold. Knockout pin: A pin that ejects a molded piece from the mold. LDR: Let down ratio used in coloring plastics with color concentrates, etc. (e. g., 20 : 1 is 20 lbs. of natural resin to 1 lb. of colorant). Length/diameter (L/D) ratio: Describes the relationship between the length of the screw and its diameter. An L/D of 20 : 1 is suggested to ensure enough residence time for thorough mixing without adding excessive mechanical energy. Locating ring: A ring attached to the front half and possibly the back half of a mold which aligns a mold with the center hole in the stationary/moving platen of an injection molding machine. Locking force: The force exerted in the clamping system (hydraulic, toggle, mechanical) of a molding machine to hold the mold closed. Lot: A quantity of resin produced at one time. Manifold: The configuration of a single channel flow with division into various flow channels to feed more than one area, as in a hot manifold system for polymers or a water manifold system for in-mold cooling. Mechanical property: Properties of plastics which are classified as mechanical include modulus, strength, impact resistance, hardness, and elongation. Melt: Plastic which is in a molten state or above the melting point. Melt air shot: A sample of the melt when taken on cycle under representative molding conditions which will indicate the actual temperature of the melt. Melt index: Extrusion rate of a thermoplastic material through an orifice of spe­ cified diameter and length under specified conditions of time, temperature, and pressure. Melt strength: The strength of plastic in the molten state. Melting point (Tm): The temperature at which the crystalline regions break apart and begin to flow. Metering section of the screw: Shallow end of the screw which does final plasticizing of the melt in injection molding.

165

166 17 Terms and Definitions

Mold base: An assembly of precision steel plates that holds or retains the cavities or cores in a mold. Provides a means for the melt to be injected into the cavities and provides a means to eject the solidified parts from the mold. Mold height: The overall thickness of the mold as it is located between the platens of the molding machine. Mold release agent: Provides a layer between two surfaces to prevent adhesion of one to the other. Molded-in stress: Orientation stress within molecules and compressive and tensile stress between molecules as a result of the molding process. Molding cycle: The time required to complete the molding of a part(s). In injection molding, the cycle begins when the mold closes and ends with the opening of the mold and ejection of the molded part. Molecular weight: The sum of the atomic weights of all atoms in a molecule. Molecule: The smallest unit of a substance which can exist by itself and retain all the properties of the original substance: Molecules are composed of one or more atoms. Morphology: Refers to the structure of the polymer material. Multiple stage screw: Extruder and injection molding machine screws which ­contain changes in the flight helix to perform specific functions, such as feeding, mixing, and metering. Non-return valve: The valve that permits material to flow in one direction and closes to prevent back flow. Located in front of the injection screw. Nozzle: Provides a leak proof connection between the injection mold and the molding machine, through which the melt flows. Orange peel: An uneven surface on a plastic part somewhat resembling that of an orange peel. Orientation: The molecular alignment in a plastic product. Caused by flow or when the plastic is stretched while hot. Packing: The final filling of the mold cavity to build up the proper static pressure distribution in the cavities to achieve proper surface finish, dimensions, and physical properties without over-packing, which creates flash and parts sticking. Parting line: Where the core, cavity, and plastic meet. Pinpoint gate: Gate that is less than 1 mm in diameter. This small gate minimizes the mark left on the molded part during the break of plastic. Plastic deformation: When an object does not return to its original shape or size after pressure, stress, or load is removed.

17 Terms and Definitions

Plastic: A material that contains as an essential ingredient one or more organic polymeric substances of large molecular weight, is solid in its finished state, and, at some stage in its manufacture or processing into finished articles, can be shaped by flow. Platen: Steel plates on a molding machine to which the mold is attached. Generally, three platens are used: stationary, movable, and tailstock. Polymer: A chemical compound formed by many small molecular units (monomers) linked together to form a large, chain-like molecule. Polymerization: A chemical reaction in which the molecules of monomers are linked together to form polymers. Projected area: Area of a molded part which is projected onto a plane at right angles to the direction of the mold. Properties: The characteristics of a material that indicate how well it will perform in a variety of applications. Properties are used to compare and select thermo­ plastic materials. Property loss: A reduction in how well the material will perform caused by the shortening of the polymer chain molecules reducing the molecular weight. Pull-in cylinder: The hydraulic cylinder(s) on an injection molding machine which holds the nozzle to the sprue bushing by pulling the injection unit carriage forward. It also retracts the nozzle for purging and shutdown. Purging: In injection molding, the emptying or removal of contaminated material from the machine by forcing it out with the new color or material. Purging compound: A chemical compound used to flush contaminated material from the injection molding press. Pyrometer: A device used to read temperature; it consists of a readout device and a sensor. Ram: Rod, plunger, or screw which forces the resin through the barrel and into the mold of an injection molding machine. Ram travel: The distance the ram travels to force the resin through the heating cylinder (barrel) and into the mold of a plunger injection molding machine. Reciprocating screw: A combination melting, softening, and injection unit in an injection molding machine. Regrind: A by-product of leftover runners or rejected parts. It is then blended back in with virgin resin or sold as scrap. Reinforced plastics: Molded, formed, filament wound, or shaped plastic parts consisting of resins to which reinforcing fibers, mats, fabrics, etc. have been added before the forming operation. Strength properties are improved.

167

168 17 Terms and Definitions

Reinforcement: A material or compound used to reinforce, strengthen, or give ­dimensional stability to another material. Relative humidity: The percent of moisture in the air at a given temperature. Resin: Most resins are of high molecular weight and consist of long chain or network molecular structure. Usually, resins are more soluble in their lower molecular weight forms. Resin content: The amount of resin in a laminate expressed as a percentage of ­either total weight or total volume. Resin-rich area: Localized area filled with resin and lacking reinforcing material. Resin-starved area: The localized area containing excess reinforcement and in­ sufficient resin. Reverse temperature profile: The barrel temperature settings, starting higher in the rear and going lower toward the front of the barrel. Rheology: The study of flow. Screw: A helically flighted shaft which rotates within a barrel to mechanically process and advance a material being prepared for extrusion or injection molding. Screw delay time: The amount of time the screw is idle after pack and before screw rotate. Screw flight: The helical metal thread of a screw in an extruder or injection molding machine. Screw speed: A measurement of revolutions per minute (rpm) of an extruder or screw on an injection molding machine. Screw tip: The device on the front of the screw that pushes the melt into the mold. It contains the shut-off valve (non-return valve), which prevents the melt from sliding backward along the flights of the screw. Shear heating: Heat produced within the plastic material as it slides against itself and the barrel wall of the injection molding machine. Shear strength: The maximum shear stress which a material is capable of sustaining. Shot: The material injected during a given cycle. Shot capacity: The maximum amount of material the machine can produce in one stroke. Shot size: The amount of plastic injected into a mold that will fill the cavity or cavities, runner, and sprue. It should be between 20–65% of the barrel capacity. Shrinkage: A volume reduction in polymers that occurs during cooling due to a reduction in space between the molecules. Usually from an under-packed con­ dition.

17 Terms and Definitions

Sink mark: Depression in a molded part caused by shrinking or collapsing of the resin during cooling. Slip agent: Provides surface lubrication during and immediately following processing of plastics. Compounded into the plastic, the additive acts as an internal lubricant which gradually migrates to the surface. Solvent resistance: The ability of a plastic to resist swelling and dissolving in a solvent. Specific gravity: The density (mass per unit volume) of any material divided by that of water at a standard temperature. It provides a more accurate means of comparing material costs because plastic parts are sold by volume, not weight. Splay marks: Marks or tracks caused by gas or liquid present or trapped in a material migrating to the surface of the mold, which then slide over the surface in the direction of flow or toward a vent, leaving tracks. A cosmetic blemish on a plastic part. Sprue: In an injection mold, the main feed channel that connects the mold-filling orifice with runners leading to each cavity gate. The plastic piece formed in the sprue bushing. Sprue bushing: The channel or feed opening for the passage of plastic from the nozzle of the injection unit to the runners of the mold. Sprue gate: The passageway through which the resin melt flows from the nozzle directly to the mold cavity. Sprue puller: A pin designed to remove a sprue from a sprue bushing. Stack mold: A two-level mold with two sets of cores and cavities stacked one behind the other for molding higher cavitation tools, and minimizing the increase in clamp tonnage. Usually, add 15% to clamp tonnage due to opposing forces. Static: When not moving, such as during packing when no flow front exists. Static pressure: The pressure built up when no flow front exists such as during packing and holding. Static pressure losses are large. Stationary platen: In a horizontal/vertical injection molding machine, the platen immediately adjacent to the machine nozzle. This platen does not move during mold cycling. Stiffness: Load bearing capability without deflection. Strain: Elastic deformation due to stress. Measured as the change in length per unit of length in a given direction, and expressed in percentage or inches per inch, etc. Strength: The maximum load a material withstands before breaking or yielding. Stress: The unit force or component of force at a point in a body acting on a plane through the point. Expressed in psi.

169

170 17 Terms and Definitions

Stress crack: External or, more commonly, internal cracks in a plastic caused by tensile stresses less than that of its short-time mechanical strength. Can be caused by external or internal forces. Stripper plate: A plate in a mold which removes a molded piece from the core pins or plunger. Submarine gate/Tunnel gate: A submarine gate breaks the molded item from the runner system upon ejection from the mold. The gate is located where the opening from runner to the mold is below parting line or mold surface (in conventional edge gating, the opening is machined into the mold surface). Suck back: The technique used to partially clear the resin from the injection nozzle after the injection cycle by pulling the screw rearward, thus drawing the resin back into the injector. Used to prevent gate or nozzle drool. Tensile strength: The maximum tensile stress sustained by the specimen before failure in a tension test. Usually expressed in psi. The cross-sectional area is that of the original specimen at the point of rupture, not reduced by the break. Thermal expansion (CTE): The tendency of a plastic or steel to expand in the heat and contract in the cold. Thermal properties: Thermal properties that are important when selecting and processing a material are heat deflection temperature (HDT), thermal conductivity, the coefficient of thermal expansion (CTE), dynamic mechanical analysis curves, differential thermal analysis (DTA), and thermogravimetric analysis (TGA). Thermocouple: A device which uses a circuit of two wires of dissimilar metals or alloys, the two junctions of which are at different temperatures. A net electro­motive force (emf) occurs as a result of this temperature difference. The minute electro­ motive force, or current, is sufficient to drive a galvanometer or amplifier. Thermoplastic: Capable of being repeatedly softened by heating and hardened by cooling. Thermoset: A plastic which changes into a substantially infusible and insoluble material when cured by application of heat or chemical means. Thermoset processing is irreversible. Tg: Abbreviation for glass transition temperature. Tie bar: In molding machines, the bars that tie the stationary platen and the ­hydraulic clamping mechanism together. During mold clamp-up, the tie bars resist the strain created by the hydraulic cylinder clamping and moveable against the stationary mold. Tm: Abbreviation for melting point. Toggle: A mechanism that exerts pressure developed by applying force on a knee joint. It is used to close and exert pressure on a mold in a press.

17 Terms and Definitions

Tool: In injection molding, the term sometimes used to describe the mold. Transition section of the screw: The section of a plasticating screw between the feed and metering sections in which the plastic resin is in both a solid state and molten state. Transducer force: A force measuring device. It has the characteristics of providing an output, usually electrical, which serves as the measurement of load, force, compression, pressure, etc. when placed along the sensitive axis of the force cell. Two-shot molding: The technique of molding parts in two colors or two materials in a single mold or set of molds. This process is accomplished by injecting the thermoplastic into a closed mold, transferring half of the mold to mate with another mold half of different cavity shape, and injecting the second color or material around the first part. Unit mold or die: Mold designed for quick changing of interchangeable cavity parts. UV stabilizer: A chemical compound additive to a thermoplastic resin which ­selectively absorbs UV rays. Variation: The difference in things that are supposed to be the same. Vent: Shallow channel or path in a mold which allows air, gas, or volatiles to exit as the melt enters the cavity. Vented barrel: Port in a barrel through which volatiles or moisture can be r­ emoved. Vented screw: Two-stage screw with a vent in the second stage to remove volatiles or water from the plastic. Viscosity: The measure of the resistance of a fluid to flow (either through a specific orifice or in a rotational viscometer). The absolute unit of viscosity measurement is the poise (or centipoise). Kinematic viscosity is expressed in stokes. Voids: Pockets of unfilled space or vacuum in a molded part generally caused by shrinkage during cooling of the part. Warpage: Distortion caused by non-uniform shrinkage of internal stresses. Water absorption: The ratio of the weight of water absorbed by a material to the weight of the dry materials. Weld lines: The marks visible on a finished part made by the meeting of two flow fronts of the resin during molding.

171

18

Reference List for Further Courses and Reading

„„18.1 Courses RJG Associates RJG Associates, 3111 Park Dr. Traverse City, Michigan, 49686 Visit website for full list: www.rjginc.com/training ƒƒRJG Systematic Molding ƒƒRJG Master Molder I ƒƒRJG Master Molder II AIM Institute AIM Institute, 6100 West Ridge Rd., Erie, Pennsylvania, 16506 Visit website for full list: aim.institute ƒƒPlastics Technology and Engineering Certificate ƒƒMold Start-up and debug ƒƒUnderstanding Shrink and Warp

„„18.2 Reading ƒƒJohn P. Beaumont, Runner and Gating Design Handbook, 2nd ed., (2007) Hanser Publishers, Munich ƒƒJohn P. Beaumont, Robert Nagel, and Robert Sherman, Successful Injection Molding, (2002) Hanser Publishers, Munich ƒƒSuhas Kulkarni, Robust Process Development and Scientific Molding, 2nd ed., (2017) Hanser Publishers, Munich ƒƒTim A. Osswald, Lih-Sheng Turng, and Paul Gramann, Injection Molding Handbook, 2nd ed., (2007) Hanser Publishers, Munich

Index

A

F

amorphous resin  74, 76 anisotropic shrinkage  77

feed throat  13, 21, 22 feed zone  3 fish hook  125 flash  126 flow lines  128, 129 fountain flow  81 four plastic variables  145 frozen layer  85

B black specks  115 blocking a cavity  84 blush  117 bolt location, proper  33 brittleness  118 burns  119, 121 C closed loop  40 cloudy parts  121 coefficient of variation  111 color streaks  122 compression ratio  4, 5 cooling optimization study  70 cooling rate  105 D

G gate seal or gate freeze  64 glass transition temperature  75 H heat deflection temperature  70 high limit fill time  7 hopper  22–24 hydraulic clamp  28, 29, 35 I

decompression  10 degraded polymer  124 dynamic non-return valve test  48 dynamic pressure  93

injection speed linearity  50 injection unit –– pressure limited  2, 6 intensifying ratio  87, 88, 93 isotropic shrinkage  76

E

J

ejector pin marks  123

jetting  130

176

Index

L

R

laminar flow  99, 100, 103 L/D  4 least pressure curve  60 load sensitivity  44, 45 long gates  131, 136

Reynolds number  101–103 runner weight study  67 S

non-return valve  3, 4, 8–11 nozzle drool  133

semi-crystalline resin  74, 75 shear  80 shear rate  57, 60, 62, 63 shear thinning  58 shorts or non-fills  134 shot size  82–84, 86, 89 shrinkage  76, 77 single point  29 sink  129, 135, 136 splay  137 static non-return valve  49 static pressure  93

O

T

open loop  40 orientation  85

thermocouples  13–15 tie-bar-less  28 toggle press  35, 36 transfer position  82, 84, 86, 87 turbulent flow  99, 100

M manifold balance  68 melt density  78 melt transition temperature  75 mold weight calculation  34 N

P plastic cooling  145 plastic flow rate  145 plastic pressure  145 plastic temperature  145 platen wrap  30 pressure limited  88 pressure loss  71, 72 pressure response  46 profile of the screw  6 projected area, calculation  53 pulls  134 purge disk  44–46

V viscosity  93 viscosity changes  86 viscosity curve  56, 58, 59 void  140 W warpage  141 wattage  18, 19 weld lines  118, 143 worn barrel  20