handbook of hydraulic fluid technology

HRNDBOOK O F HYDRRULIC FLUID TCHNOLOGY

MECHANICAL ENGINEERINGA Series of Textbooks and Reference Books

Founding Editor

L. L. FaulknerColurnbus Division, Battelle Memorial Institute and Department o Mechanical Engineering f The Ohio State University Colurnbus, Ohio

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30, 31. 32. 33. 34. 35. 36.

Spring Designers Handbook, Harold Carlson Computer-Aided Graphics and Design, Daniel L. Ryan Lubrication Fundamentals, J. George Wills Solar Engineering for Domestic Buildings,William A. Himmelman Applied Engineering Mechanics: Statics and Dynamics, G. Boothroyd and C. Poli Centrifugal Pump Clinic, lgor J. Karassik Computer-Aided Kinetics for Machine Design, Daniel L. Ryan Plastics Products Design Handbook, Part A: Materials and Components; Part B: Processes and Design for Processes, edited by Edward Miller Turbomachinery: Basic Theory and Applications, Earl Logan, Jr. Vibrations of Shells and Plates, Werner Soedel Flat and Comgated Diaphragm Design Handbook, Mario Di Giovanni Practical Stress Analysis in Engineering Design, Alexander Blake An Introduction to the Design and Behavior of Bolted Joints, John H. Bickford Optimal Engineering Design: Ptinciples and Applications, James N. Siddall Spring Manufacturing Handbook, Harold Carlson lndustrial Noise Control: Fundamentals and Applications, edited by Lewis H. Bell Gears and Their Vibration: A Basic Approach to Understanding Gear Noise, J. Derek Smith Chains for Power Transmission and Material Handling: Design and Applications Handbook, American Chain Association Corrosion and Corrosion Protection Handbook, edited by Philip A. Schweitzer Gear Drive Systems: Design and Application, Peter Lynwander Controlling In-Plant Airborne Contaminants: Systems Design and Calculations, John D. Constance CAD/CAM Systems Planning and Implementation, Charles S.Knox Probabilistic Engineering Design: Principles and Applications, James N. Siddall Traction Drives: Selection and Application, Frederick W. Heilich Ill and Eugene E. Shube Finite Element Methods: An Introduction, Ronald L. Huston and Chris E. Passerello Mechanical Fastening of Plastics: An Engineering Handbook, Brayton Lincoln, Kenneth J. Gomes, and James F. Braden Lubrication in Practice: Second Edition, edited by W. S. Robertson Principles of Automated Drafting, Daniel L. Ryan Practical Seal Design, edited by Leonard J. Martini Engineering Documentation for CAD/CAM Applications, Charles S. Knox Design Dimensioning with Computer Graphics Applications, Jerome C. Lange Mechanism Analysis: Simplified Graphical and Analytical Techniques, Lyndon 0. Barton CAD/CAM Systems: Justification, Implementation, Productivity Measurement, Edward J. Preston, George W. Crawford, and Mark E. Coticchia Steam Plant Calculations Manual, V . Ganapathy Design Assurance for Engineers and Managers, John A. Burgess Heat Transfer Fluids and Systems for Process and Energy Applications, Jasbir Singh

37. Potential Flows: Computer Graphic Solutions, Robert H. Kirchhoff 38. Computer-Aided Graphics and Design: Second Edition, Daniel L. Ryan 39. Electronically Controlled Proportional Valves: Selection and Application, Michael J. Tonyan, edited by Tobi Goldoftas 40. Pressure Gauge Handbook, AMETEK, U.S. Gauge Division, edited by Philip W. Harland 41. fabric filtration for Combustion Sources.' fundamentals and Basic Technology, R. P. Donovan 42. Design of Mechanical Joints, Alexander Blake 43. CADLAM Dictionary, Edward J. Preston, George W. Crawford, and Mark E. Coticchia 44. Machinery Adhesives for Locking, Retaining, and Sealing, Girard S. Haviland 45. Couplings and Joints: Design, Selection, and Application, Jon R. Mancuso 46. Shaft Alignment Handbook, John Piotrowski 47. BASIC Programs for Steam Plant Engineers: Boilers, Combustion, Fluid Flow, and Heat Transfer,V. Ganapathy 48. Solving Mechanical Design Problems with Computer Graphics, Jerome C. Lange 49. Plastics Gearing: Selection and Application, Clifford E. Adams 50. Clutches and Brakes: Design and Selection, William C. Orthwein 51. Transducers in Mechanical and Electronic Design, Harry L. Trietley 52. Metallurgical Applications of Shock-Wave and High-Strain-Rate Phenomena, edited by Lawrence E. Murr, Karl P. Staudhammer, and Marc A. Meyers 53. Magnesium Products Design, Robert S. Busk 54. How to lntegrate CADLAM Systems: Management and Technology, William D. Engelke 55. Cam Design and Manufacture: Second Edition; with cam design software for the ISM PC and compatibles, disk included, Preben W. Jensen 56. Solid-state AC Motor Controls: Selection and Application, Sylvester Campbell 57. Fundamentals of Robotics, David D. Ardayfio 58. Belt Selection and Application for Engineers, edited by Wallace D. Erickson 59. Developing Three-DimensionalCAD Software with the ISM PC, C. Stan Wei 60. Organizing Data for ClM Applications, Charles S. Knox, with contributions by Thomas C. Boos, Ross S. Culverhouse, and Paul F. Muchnicki 61. Computer-Aided Simulation in Railway Dynamics, by Rao V. Dukkipati and Joseph R. Amyot 62. Fiber-Reinforced Composites: Materials, Manufacturing,and Design, P. K. Mallick 63. Photoelectric Sensors and Controls: Selection and Application, Scott M. Juds 64. Finite Element Analysis with Personal Computers, Edward R. Champion, Jr., and J. Michael Ensminger 65. Ultrasonics: Fundamentals, Technology,Applications: Second Edition, Revised and Expanded, Dale Ensminger 66. Applied Finite Element Modeling: Practical Problem Solving for Engineers, Jeffrey M. Steele 67. Measurement and lnstrumentation in Engineering: Principles and Basic Laboratory Experiments, Francis S. Tse and lvan E. Morse 68. Centrifugal Pump Clinic: Second Edition, Revised and Expanded, lgor J. Karassik 69. Practical Stress Analysis in Engineering Design: Second Edition, Revised and Expanded, Alexander Blake 70. An lntroduction to the Design and Behavior of Bolted Joints: Second Edition, Revised and Expanded, John H. Bickford 71. High Vacuum Technology: A Practical Guide, Marsbed H. Hablanian 72. Pressure Sensors: Selection and Application, Duane Tandeske 73. Zinc Handbook: Properties, Processing, and Use in Design, Frank Porter 74. Thermal fatigue of Metals, Andrzej Weronski and Tadeusz Hejwowski 75. Classical and Modem Mechanisms for Engineers and Inventors, Preben W. Jensen 76. Handbook of Electronic Package Design, edited by Michael Pecht 77. Shock-Wave and High-Strain-Rate Phenomena in Materials, edited by Marc A. Meyers, Lawrence E. Murr, and Karl P. Staudhammer 78. Industrial Refrigeration: principles, Design and Applications, P. C. Koelet 79. Applied Combustion, Eugene L. Keating 80. Engine Oils and Automotive Lubrication, edited by Wilfried J. Bark

8 1. Mechanism Analysis: Simplified and Graphical Techniques, Second Edition, Revised and Expanded, Lyndon 0. Barton 82. Fundamental Fluid Mechanics for the Practicing Engineer, James W. Murdock 83. Fiber-Reinforced Composites: Materials, Manufacturing, and Design, Second Edition, Revised and Expanded, P. K. Mallick 84. Numerical Methods for Engineering Applications, Edward R. Champion, Jr. 85. Turbomachinery: Basic Theory and Applications, Second Edition, Revised and Expanded, Earl Logan, Jr. 86. Vibrations of Shells and Plates: Second Edition, Revised and Expanded, Werner Soedel 87. Steam Plant Calculations Manual: Second Edition, Revised and Expanded, V. Ganapathy 88. lndustrial Noise Control: Fundamentals and Applications, Second Edition, Revised and Expanded, Lewis H. Bell and Douglas H. Bell 89. Finite Elements: Their Design and Performance, Richard H. MacNeal 90. Mechanical Properties of Polymers and Composites: Second Edition, Revised and Expanded, Lawrence E. Nielsen and Robert F. Landel 91. Mechanical Wear Prediction and Prevention, Raymond G. Bayer 92. Mechanical Power Transmission Components, edited by David W. South and Jon R. Mancuso 93. Handbook of Turbomachinery,edited by Earl Logan, Jr. 94. Engineering Documentation Control Practices and Procedures, Ray E. Monahan 95. Refractory Linings: ThermomechanicalDesign and Applications, Charles A. Schacht 96. Geometric Dimensioning and Tolerancing: Applications and Techniques for Use in Design, Manufacturing, and Inspection, James D. Meadows 97. An lntroduction to the Design and Behavior of Bolted Joints: Third Edition, Revised and Expanded, John H. Bickford 98. Shaft Alignment Handbook: Second Edition, Revised and Expanded, John Piotrowski 99. Computer-Aided Design of Polymer-Matrix Composite Structures, edited by S.V. Hoa 100. Friction Science and Technology, Peter J. Blau 101. lntroduction to Plastics and Composites: Mechanical Properties and Engineering Applications, Edward Miller 102. Practical Fracture Mechanics in Design, Alexander Blake 103. Pump Characteristics and Applications, Michael W. Volk 104. Optical Principles and Technology for Engineers, James E. Stewart 105. Optimizing the Shape of Mechanical Elements and Structures, A. A. Seireg and Jorge Rodriguez 106. Kinematics and Dynamics of Machinery, Vladimir Stejskal and Michael ValaSek 107. Shaft Seals for Dynamic Applications, Les Horve 108. Reliability-Based Mechanical Design, edited by Thomas A. Cruse 109. Mechanical Fastening, Joining, and Assembly, James A. Speck 110. Turbomachinery Fluid Dynamics and Heat Transfer, edited by Chunill Hah 111. High-Vacuum Technology: A Practical Guide, Second Edition, Revised and Expanded, Marsbed H. Hablanian 112. Geometric Dimensioning and Tolerancing: Workbook and Answerbook, James D. Meadows 113. Handbook of Materials Selection for Engineering Applications, edited by G. T. Murray 114. Handbook of Thermoplastic Piping System Design, Thomas Sixsmith and Reinhard Hanselka 115. Practical Guide to Finite Elements: A Solid Mechanics Approach, Steven M. Lepi 116. Applied Computational Fluid Dynamics, edited by Vijay K. Garg 117. Fluid Sealing Technology, Heinz K. Muller and Bernard S. Nau 118. Friction and Lubrication in Mechanical Design, A. A. Seireg 119. lnfluence Functions and Matrices, Yuri A. Melnikov 120. Mechanical Analysis of Electronic Packaging Systems, Stephen A. McKeown 121. Couplings and Joints: Design, Selection, and Application, Second Edition, Revised and Expanded, Jon R. Mancuso 122. Thermodynamics: Processes and Applications, Earl Logan, Jr. 123. Gear Noise and Vibration, J. Derek Smith 124. Practical Fluid Mechanics for Engineering Applications, John J. Bloomer

125. Handbook of Hydraulic Fluid Technology, edited by George E. Totten

Additional Volumes in PreparationHeat Exchanger Design Handbook, T. Kuppan

Mechanical Engineering Soflware

Spring Design with an ISM PC, AI Dietrich Mechanical Design Failure Analysis: With Failure Analysis System Software for the ISM PC, David G. Ullman

This page intentionally left blank

HRNDBOOK Of HYDRRULIC FLUID TKHNOLOGY

edited by

GORG. TOTEN

Union Carbide Corporation Tarrytown, New York

~-

MARCEL EKKER, D INC.I) E K K E R

N E WYORK BASEL

ISBN: 0-8247-6022-0This book is printed on acid-free paper.

Headquarters

Marcel Dekker, Inc. 270 Madison Avenue, New York, NY 10016 tel: 2 12-696-9000; fax: 2 12-685-4540

Hutgasse 4, Postfach 8 12, CH-4001 Basel, Switzerland tel: 41-61-261-8482; f a : 41-61-261-8896

Eastern Hemisphere Distribution Marcel Dekker AG

World Wide Web

http://www.dekker.com

The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special SalesProfessional Marketing at the headquarters address above.

Copyright 0 2000 by Marcel Dekker, Inc. All Rights Reserved.Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher. Current printing (last digit): 10 9 8 7 6 5 4 32 1

PRINTED IN THE UNITED STATES OF AMERICA

Preface

One of the most frustrating practices of my career has been the search for information on hydraulic fluids, which includes information on fluid chemistry; physical properties; maintenance practices; and fluid, system, and component design. Although some information on petroleum oil hydraulic fluids can be found, there is much less information on fire-resistant, biodegradable, and other types of fluids. Unfortunately, with few exceptions, fluid coverage in hydraulic texts is typically limited to a singlechapter overview intended to cover all fluids. Therefore, it is often necessary to perform a literature search or a time-consuming manual search of my files. Some time ago it occurred to me that others must be encountering the same problem. There seemed to be a vital need for an extensive reference text on hydraulic fluids that would provide information in sufficient depth and breadth to be of use to the fluid formulator, hydraulic system designer, plant maintenance engineer, and others who serve the industry. Currently, there are no books dedicated to hydraulic fluid chemistry. Most hydraulic fluid treatment is found in handbooks, which primarily focus on hydraulic system hardware, installation, and troubleshooting. Most of these books fit into one of two categories. One type of book deals with hydraulic equipment, with a single, simplified overview chapter covering all hydraulic fluids but with a focus on petroleum-derived fluids. The second type of book provides fluid coverage with minimal, if any, discussion of engineering properties of importance in a hydraulic system.iii

iv

Preface

The purpose of the Handbook o Hydraulic Fluid Technology is to provide a f comprehensive and rigorous overview of hydraulic fluid technology. The objective is not only to discuss fluid chemistry and physical properties in detail, but also to integrate both classic and current fundamental lubrication concepts with respect to various classes of hydraulic fluids. A further objective is to integrate fluid dynamics with respect to their operation in a hydraulic system in order to enable the reader to obtain a broader understanding of the total system. Hydraulic fluids are an important and vital component of the hydraulic system. The 23 chapters of this book are grouped into three main parts: hardware, fluid properties and testing, and fluids.

HARDWAREChapter 1 provides the reader with an overview of basic hydraulic concepts, a description of the components, and an introduction to hydraulic system operation. In Chapter 2, the rolling element bearings and their lubrication are discussed. An extremely important facet of any well-designed hydraulic system is fluid filtration. Chapter 3 not only provides a detailed discussion of fluid filtration and particle contamination and quantification, but also discusses fluid filterability. An understanding of the physical properties of a fluid is necessary to understand the performance of a hydraulic fluid as a fluid power medium. Chapter 4 features a thorough overview of the physical properties, and their evaluation and impact on hydraulic system operation, which includes: viscosity, viscosity-temperature and viscosity-pressure behavior, gas solubility, foaming, air entrainment, air release, and fluid compressibility and modulus.

FLUID PROPERTIES AND TESTINGViscosity is the most important physical property exhibited by a hydraulic fluid. Chapter 5 presents an in-depth discussion of hydraulic fluid viscosity and classification. The hydraulic fluid must not only perform as a power transmission medium but also lubricate the system. Chapter 6 provides a thorough review of the fundamental concepts involved in lubricating a hydraulic system. In many applications, fluid fire resistance is one of the primary selection criteria. An overview of historically important fire-resistance testing procedures is provided in Chapter 7, with a discussion of currently changing testing protocol required for industry, national, and insurance company approvals. Ecological compatibility properties exhibited by a hydraulic fluid is currently one of the most intensive research areas of hydraulic fluid technology. An overview of the current testing requirements and strategies is given in Chapter 8. One of the most inexpensive but least understood components of the hydraulic system is hydraulic seals. Chapter 9 provides a review of mechanical and elastomeric seal technology and seal compatibility testing. An often overlooked but vitally important area is adequate testing and evaluation of hydraulic fluid performance in a hydraulic system. There currently is no consensus on the best tests to perform and what they reveal. Chapter 10 reviews the state-of-the-art of bench and pump testing of hydraulic fluids. Vibrational analysis not only is an important plant maintenance tool but is also one of the most important diagnostic techniques for evaluating and

Preface

V

troubleshooting the operational characteristics of a hydraulic system. Chapter 11 provides an introductory overview of the use of vibrational analysis in fluid maintenance. No hydraulic system operates trouble-free forever. When problems occur, it is important to be able to identify both the problem and its cause. Chapter 12 provides a thorough discussion of hydraulic system failure analysis strategies.

FLUIDSAlthough water hydraulics do not constitute a major fluid power application, they are coming under increasing scrutiny as ecocompatible alternatives to conventional hydraulic fluids. Chapter 13 offers an overview of this increasingly important technology. The largest volume fluid power medium is petroleum oil. In Chapter 14, the reader is provided with a thorough overview of oil chemistry, properties, fluid maintenance, and change-out procedures. Chapter 15 reviews additive technology for petroleum oil hydraulic fluids. There are various types of synthetic hydraulic fluids. Chapter 16 describes the more important synthetic fluids, with a focus on aerospace applications. Chapters 17 to 20 describe fire-resistant hydraulic fluids. Emulsions, waterglycols, polyol esters, and phosphate esters are discussed individually and in depth in Chapters 17, 18, 19, and 20, respectively. This discussion includes fluid chemistry, physical properties, additive technology, maintenance, and hydraulic system conversion. Vegetable oils are well-known lubricants that have been examined repeatedly over the years. Currently, there is an intensive effort to increase the utilization of various types of vegetable oils as an ecologically sound alternative to mineral oil hydraulic fluids. Chapter 21 provides a review of vegetable oil chemistry, recovery, and properties. The applicability of these fluids as hydraulic fluid basestocks is examined in detail. Chapter 22 discusses electrorheological fluids, which are becoming increasingly interesting for use in specialized hydraulic applications. In Chapter 23, various standardized fluid maintenance procedures are discussed and a summary of equivalent international testing standards is provided. The preparation of a text of this scope was a tremendous task. I am deeply indebted to many colleagues for their assistance, without whom this text would not have been possible. Special thanks go to Dr. Stephen Lainer (University of Aachen), Prof. Atsushi Yamaguchi (Yokohama National University), Prof. Toshi Kazama (Muroran Institute of Technology), K. Mizuno (Kayaba Industrial Ltd.), and Jiirgen Reichel (formerly with DMT, Essen, Germany). Special thanks also goes to my wife, Alice, for her unending patience, and to Susan Meeker, who assisted in organizing and editing much of this material; to Glenn Webster, Roland J. Bishop, Jr., and Yinghua Sun, without whose help this text would never have been completed; and to Union Carbide Corporation for its support. George E. Totten

Contents

Preface Cont ribu tors

111

...

xi

HARDWARE 1. Basic Hydraulic Pump and Circuit Design Richard K. Tessmann, Hans M. Melie$ and Roland J. Bishop, Jr: 1

65

2 Bearing Selection and Lubrication .Faruk Pavlovich

3. Fluid Cleanliness and Filtration Richard K. Tessmann and I. T Hong4. Physical Properties and Their Determination George E. Totten, Glenn M . Webster; and F: D. Yeaple FLUID PROPERTIES AND TESTING

147 195

5 Fluid Viscosity and Viscosity Classification .Bernard G. Kinker6.

305

339

vii

Lubrication Fundamentals Lavern D. Wedeven and Kenneth C Ludema

viii

Contents

7. Fire-Resistance Testing Procedures of Hydraulic Fluids Glenn M. Webster and George E. Totten 8. Ecological Compatibility Gaty H. Kling, Dwayne E. Tharp, George E. Totten, and Glenn M. Webster 9. Seals and Seal Compatibility Ronald E. Zielinski 10. Bench and Pump Testing of Hydraulic Fluids Lin Xie, Roland J. Bishop, J K , and George E. Totten 11. Vibrational Analysis for Fluid Power Systems Hugh R. Martin 12. Failure Analysis Steven Lemberger and George E. Totten

393 427

469 523 577 601

FLUIDS13. Water Hydraulics Kari i Koskinen and Matti J. Vilenius ? 14. Mineral-Oil Hydraulic Fluids Paul McHugh, William D. Stofey, and George E. Totten 15. Lubricant Additives for Mineral-Oil -Based Hydraulic Fluids Stephen H. Roby, Elaine S. Yamaguchi, and P. R. Ryason 16. Polyalphaolefins and Other Synthetic Hydrocarbon Fluids Ronald L. Shubkin, Lois J. Gschwender; and Car1 E. Snyder; J K 17. Emulsions Yinghua Sun and George E. Totten 18. Water-Glycol Hydraulic Fluids George E. Totten and Yinghua Sun 19. Polyol Ester Fluids Robert A. Gere and Thoinas V Hazelton 20. Phosphate Ester Hydraulic Fluids W D. Phillips 21. Vegetable-Based Hydraulic Oils Lou A. ?: Honary 22. 675 711 795825

847 917 983 1025 1095 1153 1185

Electro-rheological Fluids Christian Wolfl-Jesse

23. Standards for Hydraulic Fluid Testing Paul Michael

Contents

ix

Appendixes 1. Temperature Conversion Table 2. SI Unit Conversions 3. Commonly Used Pressure Conversions; Fraction Notation 4. Volume and Weight Equivalents 5. Head and Pressure Equivalents 6. Flow Equivalents 7. Viscosity Conversion Charts 8. Number of U.S. Gallons in Round and Rectangular Tanks 9. Water Vapor Pressure Chart 10. Physical Properties of Ethylene, Diethylene, and Propylene Glycol 11. Common Wear Problems Related to Lubricants and Hydraulic Fluids Index

1209 1213 1215 1216 1217 1218 1219 1222 1225 1226

1242

1247

Basic Hydraulic Pump and Circuit DesignRICHARD K. TESSMANN FES, Inc., Stillwatec Oklahoma HANS M. MELIEF The Rexroth Corporation, Bethlehem, Pennsylvania ROLAND J. BISHOP, JR. Union Carbide Corporation, Tarrytown, New York

1 INTRODUCTIONHydraulics, according to Webster, is defined as operated, moved, or effected by means of water. In the 17th century, it was discovered that a fluid under pressure could be used to transmit power [ 11. Blaise Pascal (1623- 1662) observed that if a fluid in a closed container was subjected to a compressive force, the resulting pressure was transmitted throughout the system undiminished and equal in all directions Hydraulics is by far the simplest method to transmit energy to do work. It is considerably more precise in controlling energy and exhibits a broader adjustability range than either electrical or mechanical methods. To design and apply hydraulics efficiently, a clear understanding of energy, work, and power is necessary. In this chapter, fundamentals of hydraulic pump operation and circuit design will be provided. This will include the following: Hydraulic principles Hydraulic system components Hydraulic pumps and motors System design considerations1

HI.

2

Tessmann et al.

2 2.1

DISCUSSION Hydraulic Principles

Work is done when something is moved. Work is directly proportional to the amount of force applied over a given distance according to the following relation,

Work (ft-lbs) = Distance (ft)

X

Force (lbs)

(1.1)

Power is defined as the rate of doing work and has the units of foot-pounds per second. A more common unit of measure is horsepower (hp). Horsepower is defined as the amount of weight in pounds that a horse could lift 1 ft in 1 s (Fig. 1.1) [l]. By experiment, it was found that the average horse could lift 550 lbs. 1 ft in 1 s; consequently,1 hp =

550 ft-lbsS

Energy is the ability to do work. It may appear in various forms, such as mechanical, electrical, chemical, nuclear, acoustic, radiant, and thermal. In physics, the Law of Conservation of Mass and Energy states that neither mass nor energy can be created or destroyed, only converted from one to the other. In a hydraulic system, energy input is called a prime mover. Electric motors and internal combustion engines are examples of prime movers. Prime movers and hydraulic pumps do not create energy; they simply convert it to a form that can be utilized by a hydraulic system. The pump is the heart of the hydraulic system. When the system performs improperly, the pump is usually the first component to be investigated. Many times, the pump is described in terms of its pressure limitations. However, the hydraulic pump is a flow generator, moving a volume of fluid from a low-pressure region to a higher-pressure region in a specific amount of time depending on the rotation speed.

1 ft.Figure 1.1 Illustration of the horsepower concept.

Hydraulic Pump and Circuit Design

3

Therefore, the pump is properly described in terms of its displacement or the output flow rate expected from it. All pumps used in a hydraulic system are of the positive displacement type. This means that there is an intentional flow path from the inlet to the outlet. Therefore, the pump will move fluid from the inlet or suction port to the outlet port at any pressure. However, if that pressure is beyond the pressure capability of the pump, failure will occur. The pressure which exists at the outlet port of the pump is a function of the load on the system. Therefore, hydraulic system designers will always place a pressure-limiting component (i.e., a rdief valve) at the outlet port of a pump to prevent catastrophic failures from overpressurization. Most pumps in hydraulic systems fall into one of three categories: vane pumps, gear pumps, or piston pumps (Sec. 2.2). The action of the hydraulic pump consists of moving or transferring fluid from the reservoir, where it is maintained at a low pressure and, consequently, a low-energy state [2]. From the reservoir, the pump moves the fluid to the hydraulic system where the pressure is much higher, and the fluid is at a much higher-energy state because of the work that must be done by the hydraulic system. The amount of energy or work imparted to the hydraulic system through the pump is a function of the amount of volume moved and the pressure at the discharge port of the pump: Work = PressureX

Flow

(1.3)

From an engineering standpoint, it is common to relate energy to force times distance: Work ForceX

Distance

(1.4)

However, hydraulic pressure is force divided by area, and volume is area times distance: Force Pressure = Area Volume = AreaX

(1.5)

Distance

(1.6)

From these relationships, Eq. (1.7) shows that pressure times volume is equivalent to force times distance.P (lbs/in.)X

V ( h 3 ) = F (lbs)

X

D (in.)

(1.7)

The hydraulic pump is actually a three-connection component. One connection is at the discharge (outlet) port, the second is at the suction (inlet) port, and the third connection is to a motor or engine (Fig. 1.2) [l]. From this standpoint, the pump is a transformer. It takes the fluid in the reservoir, using the energy from the motor or engine, and transforms the fluid from a low-pressure level to a higherpressure level. In fact, the hydraulic fluid is actually a main component of the hydraulic system, and as we will see throughout this book, has a major influence in the operation of the system. Hydraulic pumps are commonly driven at speeds from 1200 rpm to 3600 rpm or higher and maximum pressures may vary from 6000 psi. Tables 1.1 and 1.2 [ l ] show typical pressure and speed limitations for various types of pumps and motors. In addition to pressure, there is also a temperature limitation imposed by the hydraulic fluid. This is caused by the decrease in viscosity as the fluid temperature

4

Tessmann et al.

Figure 1.2 Illustration of pump connections.

increases. Generally, pump manufacturers set upper and lower viscosity limits on the hydraulic fluids used in their pumps. The upper viscosity limit determines the minimum temperature for pump start-up to prevent cavitation. Cavitation occurs when there is insufficient fluid flow into the pump inlet. During cavitation, the fluid will release any dissolved gases or volatile liquids. The gaseous bubbles produced will then travel into the high-pressure region of the pump where they will collapse under high pressure and may cause severe damage to the pump. Also, overheating of the pump bearings could result because of insufficient cooling as a result of inadequate flow.

Table 1.1 Typical Pump Performance Parameters for One Manufacturer~~ ~

Hydraulic Pump Type External Gear Internal Gear Vane Vane Radial Piston Axial Piston (bent axis) Axial Piston (bent axis) Axial Piston (swash plate)

Maximum PressureFlow

(psi)

Maximum Speed (rpm1500- 5,000 900- 1,800 900- 3,000 750-2,000 1,000-3,400 950-3,200 500-4,100 500-4,300

Total Efficiency(%)

Fixed Fixed Fixed Variable Fixed Fixed Variable Variable

3,600 3,000 2,500 1,000-2,300 10,000 5,100-6,500 5,800 4,600-6,500

85-90 90 86 85 90 92 92 91


5

Table 1.2 Typical Motor Performance Parameters for One ManufacturerMaximum Pressure (psi) 3600 6 100-6500 6100 5800- 6500 6500 4600 - 5800 Maximum Speed (rpm) 500-3000 1-500 1-500 50-6000 50-8000 6-4900 Total Efficiency (%j

Hydraulic Motor Type Gear Radial Piston Radial Piston Axial Piston (bent axis) Axial Piston (bent axis) Axial Piston (swash plate)

Flow Fixed Fixed Variable Fixed Variable Variable

85 91 -92 92 92 92 91

The lower viscosity limit will establish the upper temperature limit of the fluid. If the upper temperature limit is exceeded, the viscosity will be insufficient to bear the high operating loads in the pump and, thus, lubrication failure will result in shortened pump life and/or catastrophic pump failure. Table 1.3 shows the effect of viscosity index on the pump operating temperatures for one major pump manufacturer [ 13. Viscosity index is a measure of the viscosity change or resistance to flow of a liquid as the temperature is changed. A higher viscosity index produces a smaller viscosity change with temperature than a fluid having a lower viscosity index (see Chapter 4).

2.1.1 *Torque and PressureThe input parameters to the hydraulic pump from the prime mover are speed and torque. The input torque to the pump is proportional to the pressure differential between the inlet port and the discharge port: (Pressure,,,,,,-

Pressure,,,,,)

X

Displacement

(1.8)

The torque required to drive a positive displacement pump at constant pressure is [3-71

T, = T, + T,, + Tf+ T,

where T, is the actual torque required, T, is the theoretical torque due to pressure differential and physical dimensions of the pump, T z ,is the torque resulting from viscous shearing of the fluid, Tf is the torque resulting from internal friction, and T, is the constant friction torque independent of both pressure and speed.

Table 1.3 A Pump Manufacturers Viscosity Index and Temperature GuidelinesTemperature (OF) Minimum Optimum Maximum Viscosity Index (50) Viscosity Index (95) Viscosity Index(150)

85- 130 155

18

5 80- 135 160

0 75- 140 175

6

Tessmann et al.

Substituting the operational and dimensional parameters (using appropriate units) into Eq. (1.9) produces the following expression:

T l = (Pi

- P2)Di, + C,,D,,pN

+ C,(Pi

-

P,)D,,

+ T,

(1.10)

where P i is the pressure at the discharge port, P , is the pressure at the inlet port, D , is the pump displacement, C,, the viscous shear coefficient, is the fluid viscosity, is N is the rotation speed of the pump shaft, Cjis the mechanical friction coefficient, and T, is the breakaway torque. Actual values of the parameters such as viscous shear coefficient, mechanical friction coefficient, and breakaway torque are determined experimentally. From Eq. (1. lO), the required torque to drive a pump is primarily a function of the pressure drop across the pump (P, - P 2 ) and the displacement (D,,) of the pump.2.1.2

Rotational Speed and Flow=

The output flow of a hydraulic pump is described byQCi

Qt

-

Q/ -

Qr

(1.11)

where Qclis the delivery or actual flow rate of the pump, Q, is the theoretical flow, Q, is the leakage flow, and Q,. is the losses resulting from cavitation and aeration (usually neglected). Substituting the operational (speed, flow, and pressure) and dimensional (pump displacement) parameters into Eq. ( 1.11) yields

.where Qclis the delivery or actual flow rate of the pump and C,is the leakage flow (slip) coefficient. From Eq. (1.12) it is apparent that pump delivery (QJ is primarily a function of the rotational speed (N) of the pump, and the losses are a function of the hydraulic load pressure ( P i - P 2 ) . The leakage flow or slip that takes place in a positive displacement pump is caused by the flow through the small clearance spaces between the various internal parts of the pump in relative motion. These small leakage paths are often referred to as capillary passages. Most of these passages are characterized by two flat parallel plates with leakage flow occurring through the clearance space between the flat plates (Fig. 1.3) [ 11. Therefore, the fundamental relationships for flow between flat plates are normally applied to the leakage flow in hydraulic pumps. However, to apply Eq. (1.12), it is necessary to obtain the value of the slip flow coefficient through experimental results [3,6].2.1.3

Horsepower (Mechanical and Hydraulic)

Of interest to the designers and users of hydraulic systems is the power (Hplnpu,) required to drive the pump at the pressure developed by the load and the output power ( H P ~ , , , ~ ~ ~ ~ )will be generated by the pump. The expression (using approwhich , priate units) which describes the mechanical input horsepower required to drive the pump is (1.13) where Hplnpllt the required input power (hp), T,, is the actual torque required (lbis


7

Where: U = viscosity p = pressure

Cylinder Block

Figure 1.3 Illustration of the leakage path between two flat plates in a piston pump.

ft), and N is the rotational speed of the pump shaft (rpm). The hydraulic output horsepower of a hydraulic pump is described by the expression

PQ Hpoutput = 1714

(1.14)

where Hpoutput the power output of the pump (hp), P is the pressure at the pump is discharge port (psi), and Q is the delivery or actual flow output (gpm). However, in the real world, Hpinput Hpoutput. is always true because no > This pump or motor is 100%efficient. As will be shown in the next subsection, this is mainly due to internal mechanical friction and fluid leakage within the pump.2.1.4

Pumping Efficiency

Hydraulic pumping efficiency or total efficiency (E,) is a combination of two kinds of efficiencies: volumetric (Et,) and mechanical (Em).Volumetric efficiency is given by

E,, =

Actual flow output Theoretical flow output

(1.15)

The second kind of efficiency is called the torque efficiency or the mechanical efficiency (E,,,).The mechanical efficiency is described by

E, =

Theoretical torque input Actual torque input

(1.16)

The overall or total efficiency (E,) is defined by

E, = E,,E,,,

(1.17)

8

Tessmann et al.

The total efficiency is also related to power consumption by

E, =

Power output Power input

(1.18)

Substituting Eqs. (1.13) and (1.14) into Eq. (1.18) produces the following expression for the overall efficiency of a hydraulic pump:

E, =

0.326T,, N

PQ

(1.19)

Theoretical pump delivery is also determined from the dimensions of the pump. However, if the dimensions of the pump are not known, they are determined by pump testing. The overall efficiency of a pump may also be measured by testing. However, it is difficult to measure mechanical efficiency. This is because internal friction plays a major role and there is no easy way of measuring this parameter within a pump. Rearrangement of Eq. (1.17) gives the mechanical efficiency as the overall efficiency divided by the volumetric efficiency: (1.20)

2.1.5

Hydraulic System Design

When designing a hydraulic system, the designer must first consider the load to be moved or controlled. Then, the size of the actuator is determined. The actuator is a component of a hydraulic system which causes work to be done, such as a hydraulic cylinder or motor. The actuator must be large enough to handle the load at a pressure within its design capability. Once the actuator size is determined, the speed at which the load must move will establish the flow rate of the system (gpm; gallons per minute): For hydraulic cylinders, AV gpm = 23 1 where A is the area (h3) V is the velocity (in./min) and For hydraulic motors, (1.22) where D is the displacement (in?/rev.) For example, calculate the hydraulic cylinder bore diameter ( D ) and flow ( Q ) required to lift a 10,000 lb load at a velocity of 120 in./min with a hydraulic load pressure not exceeding 3000 psi. Using the fundamental hydraulic expression, Force = PressureX

(1.21)

Area

(1.23)


9

Force is equal to 10,000 lbs and pressure is 3000 psi. The area of the head end of the cylinder is calculated by rearranging Eq. (1.23) and solving for the area: Area = Force 10,000 -Pressure 3,000 = 3.333 in.2 TTD2 -4 D = 2.06 in.~

(1.24)

Therefore, a double-acting single-rod cylinder with a cylinder bore of 2.25 or 2.5 in. may be used, depending on the requirements of the head side of the cylinder (Fig. 1.4) [l]. However, on the rod side, the area of the piston is reduced by the area of the rod. Because the effective area on the rod side of the cylinder is less than that on the head side, the cylinder would not be able to lift as large a load when retracting, because of pressure intensification at the rod end causing the system relief valve to open. For this reason, all single-rod cylinders exert greater force at the rod end when extending than retracting. On the other hand, a double-rod cylinder of equal rod diameters would exert an equal force in both directions (extending and retracting). Once the size of the cylinder is selected, the designer must consider the speed requirements of the system. The speed at which the load must be moved is dependent on the pump flow rate. Of course, the velocity of the cylinder rod and the speed of the load must be the same. By using the equationQ=--

VA 23 1

(1.25)

where Q is the flow rate into the cylinder (gpm), V is the velocity of the cylinder rod (in./min), and A is the cylinder area (in.). The flow rate ( Q ) needed to produce a specific velocity ( V ) can now be calculated by combining Eqs. (1.24) and (1.25); using a 2.5-in.-diameter cylinder as an example,Q = -

nD2V 924 - (3.14)(2.5)*( 20) 1 924 = 2.6 gpm

(1.26)

Once the pressure needed to support the load and flow to produce the specified load velocity is determined, the pump selection process and system design may begin. There are several factors in pump selection and hydraulic system design that have not been addressed. For example, the service life required by the system must be decided along with the contamination level that must not be exceeded in the system. The piping sizes and the inlet conditions to the pump must be considered. The fluid to be used in the hydraulic system is an obvious consideration. These and other factors will be addressed in later sections of this chapter.

10

Tessmann et al.

Figure 1.4

Illustration of a single-rod double-acting cylinder.

2 2 Hydraulic Pumps and Motors .There are three types of pumps used predominately in hydraulic systems: vane pumps, gear pumps, and piston pumps (Fig. 1.5) [l]. Although there are many design parameters that differ between a hydraulic pump and a hydraulic motor, the general description is fundamentally the same, but their uses are quite different. A pump is used to convert mechanical energy into hydraulic energy. The mechanical input is accomplished by using an electric motor or a gasoline or diesel engine. Hydraulic flow from the output of the pump is used to power a hydraulic circuit. On the other hand, a hydraulic motor is used to convert hydraulic energy back into mechanical energy. This is accomplished by connecting the output shaft of the hydraulic motor to a mechanical actuator, such as a gear box, pulley, or flywheel.

2.2.1

Vane Pumps

A typical design for the vane pump is shown in Fig. 1.6 [8]. The vane pump relies upon sliding vanes riding on a cam ring to increase and decrease the volume of the pumping chambers within the pump (Fig. 1.7). The sides of the vanes and rotor are sealed by side bushings. Although there are high-pressure vane pumps (>2500 psi), this type of pump is usually thought of as a low pressure pump ( = 1, at the most critical point of contact -the rolling element -raceway, a permanent deformation of 0.0001 of the rolling element diameter. Standards I S 0 76-1987 and DIN I S 0 281 provide a standard criterion for definition of the static load. A maximum value was adopted for the calculated contact stress at the most critical point of contact, the rolling element on the raceway. The value of the contact stress ( p ) varies with the type of bearing: 4600 N/mm for aligning ball bearings 4200 N/mm for all other ball bearings 4000 N/mm for all roller bearings These data refer to standard bearings whose assembly components are constructed from steel of the same hardness.

4.3.1 Determination of the Equivalent Static LoadThe static equivalent load is the calculated value of load; radial load for radial bearings and axial load for axial bearings, at the most loaded point between the

Bearing Selection and Lubrication

81

rolling element and raceway that causes the same effect as combined load. The expression for calculation of the static equivalent load (PO) is

load (kN), Xo is the radial load factor, and Yo is the axial load factor. Values of axial and radial load factors for various bearing assembly designs are provided by the bearing manufacturer. If the calculated equivalent PO < F,, then the radial force component is being neglected and the equivalent static load is assumed to be PO = F,.4.4

P = XOFr + YOFO O where PO is the static equivalent load (kN), F, is the radial load (kN), F, is the axial

Bearing Choice Based on Dynamic Load Rating

The load rating for dynamically loaded bearings is dependent on material fatigue strength and pitting. The calculation procedures are described in various national and I S 0 standards. The most commonly used equation is

L,,, = L = cP- x 10 rpmwhere LIois the rating life of 10 rpm, C is the dynamic load rating, P is the dynamic equivalent load, and p is the rating life power. Dynamic load rating is defined as load of constant direction and intensity at which 90% of the tested bearings achieve 106 cycles ( r ) without the appearance of fatigue. Data on dynamic load rating are provided by the bearing manufacturer. The value of the rating life power varies with the type of bearing: p = 3 for ball bearings and p = 3.33 for cylindrical rolling elements bearings. For a constant number of revolutions, the rating life in hours may be determined fromL h l O = L,l = ( L x 10)(n-) 60

where L,,10 is the rating life (required value), L is the rating life (tabular value), and n is the number of revolutions (min-). The following equations are derived from the definition of dynamic load rating:L, =

L500( 3 3.3 3)(60)n- I h 60

or the more general equations, Lh = C T ( 3 3 . 3 3 ) n - I,r;,l~500-1/? ( 3 3 . 3 3 ) l / ~ ~ - I / / ~ ~ p - I =

wherefL = L,11500-2 the dynamic factor andf, = (33.33)1fn-1f the number of is is revolutions factor; fL = 1, at a rating life of 500(h) and f,, = 1, at the number of revolutions of 33.33 min-. Using these substitutions, the following equation is obtained:fL

= f,,cp-

Values of fL and J1are available from the bearings manufacturer for different types of machine and different rotational frequencies.

82

Pavlovich

Determination of Dynamic Equivalent Load The influence of the outside axial and radial forces acting on a bearing can be determined experimentally or the dynamic equivalent load ( P ) which represents the effect of axial and radial forces on expected lifetime of the bearing. This represents the same rating life as when it is exposed to a combined load. Dynamic equivalent load is calculated fromP = XF,

+ YF,

where P is the dynamic equivalent load (kN), F, is the radial load (kN), F , is the axial load (kN), X is the radial load factor, and Y is the axial load factor. These load factors vary with rolling element bearing design and construction and are available from the bearings' manufacturer. For the axial bearings that are transmitting a radial force, the dynamic equivalent load is determined from P = F,, For needle bearings, the expression for determination the equivalent load is P = F where F is the radial outside load acting on the needle bearing.

Bearings Selection Based on the ModiJied Theory o Dynamic Load Rating f The relationship between the number of revolutions and the applied load was developed by Lundberg and Palmgren and has been the standard engineering practice for the prediction of rolling element bearing lifetime since 1947 [9]. However significant differences between calculated and determined lifetimes were observed. Some of the reasons for these differences include the effect of wear processes that accompany working surface coupling, the use of newer and better specification-grade steels, the effect contaminant particles and some wear products on initiation and development of bearing destruction, improved measures accuracy of surfaces finish, and development of methods for maintaining system cleanliness. Alternative equations has been proposed for determinating bearing life which include the influence of these significant parameters omitted in the Lundberg- Palmgren equation [S]:

L,," = a,a,a,L 'pmLh,rli

= ala2a3Lh

where L,,,, and Lh,,cr the bearing lifetimes determined by the modified theory, a , is are the probability choice factor, a, is the material factor, a3 is the working conditions factor, and L and L , are the rating lives. Contact surface fatigue obeys the Weibull's probabilities distribution. For rating lifetimes other than at 90% probability of failure, the appropriate factor is selected from Table 2.1 [ 5 ] . Instead of factors a, and u3, the complex factor of material and working conditions influences is introduced, obtained as a product a Z 3 ,and s, and a?? = a 2 3 , 1 s , [ S ] . The factor a23is calculated in several steps. The ratio of bearing load rating ( C ) and static equivalent load (P),f,, is used along with the factor that is dependent on bearing type to determine K , (Fig. 2.22) [5].


03

Table 21 .

The Probability Factor a,~ ~

Probability selection (%) Life rating Factor a ,

10 LO I 1

5 0.62

4

L 5

L4 0.53

0.44

3 L,

2L 2

0.33

0.2 1

1 Ll

To determine K2, the ratio of working and required viscosity must be determined. The required viscosity is determined from the half-sum of the diameters of the outside and inside raceways, or their mean values, and working number of revolutions (Fig. 2.23) [ 5 ] . Working viscosity is determined based on the working temperature obtained from Fig. 2.24 [ 5 ] . The value of K2 is determined from Fig. 2.25 [ 5 ] , using the ratio of working to required viscosity and the value offs = C,)/ P,, for fluids not containing additives or for fluids containing additives whose influence on wear is untested. For fluids containing additives, which are used on the proper way, K2 = 0. The value of the factor is determined from Fig. 2.26 [ 5 ] and is dependent on the ratio of working to required viscosities and the factor K , which is the sum of factors K , and K 2 . The lubricant purity factor s quantifies the influence of soiling on the rating life, depending on the degree of soiling V . Tables 2.2 and 2.3 [ 5 ] provide parameters suggested by international standard for the evaluation of the degree of lubricant soiling. The standard is based on experience about the influence of size and number of particles on initiation and development on the destruction process of the contact surfaces of rolling bearings:

4

3

1

0 0

2

4

1,.

6

-

a

10

12

Figure 2.22 Determination of the K I factor: (a) ball bearings; (b) cylindrical radial roller bearings and tapered roller bearings; (c) radial and axial convex-roller bearings; (d) cylindrical roller bearings without separator.

841 000

Pavlovich

500

- 100E El

I

200

.-

2

0

2010

>

U

2

3

10

20

50

100d,

200=

- [mm] 'd 2

500

1000

Figure 2.23 Required viscosity as a function of half-sum of diameters of the outside andinside raceways.

120 110

100

90

c

1

80

7060

EgY

3

50JO

E

8

U

6

'f5

30

3

20

104

6 810

20 3040

60

100

Worldng viscosity (mm*/sl

-

200 300

Figure 2.24 Working viscosity.


85

Figure 2.25 Determination of the K2 factor.

factor. Figure 2.26 Determination of the uz311

86

PavlovichApproximate Value for a Purity Factor V~ ~ ~ ~ ~ ~ ~ ~ ~~ ~~

Table 2.2

(D - 412 (mm)L 12.5

V 0.3 0.5 1 2 3 0.3 0.5 2 3 0.3 0.5 1 2 3 0.3 0.5 2 31 1

Point contact: required purity level according to IS0 4406 1118 1219 1411I 15/12 16/13 1219 13/10 15/12 16/13 18/14 13/10 1411 1 16/13 17/14 19/15 14111 15/12 17/14 18/15 20116

Approximate effectiveness of filtration according to I S 0 4572

Line contact: required purity level according to I S 0 4406 1219 13/10 15/12 16/13 17/14 13/10 1411I 16/13 17/14 19/15 14111 15/12 17/14 18/15 20116 1411 1 15/12 18/14 19/16 21/17

Approximate effectiveness of filtration according to I S 0 4572

>12.5 and 20 and 35

Table 2 3 .

Purity Class According to IS0 4406 (quotation) Number of particles per 100 mL

Over 5 pm More than~~~

Over 15 pm u p to 1,000,000 500,000 250,000 130,OOO ~,OOo 32,000 16,000 8,000 4,000 2,OOo 2,000 1,000 500 More than 64,000 32,000 16,000 8,000 4,000 2,000 1,000 500 250 130 64 32 32 u p to 130,00064,000

Code 12/17 19/16 18/15 17/14 16/13 15/12 14111 13/10 1219 1118 1117 1016 916

500,000 250,000 130,000 64,000 32,000 16,000 8,000 4,000 2,000 1,ooo 1,000 500 250

32,000 16,000 8,000 4,000 2,000 1,m 500 250 130 64 64


07

For the normal level of soiling V = 1, the purity factor takes the value s = 1. At realized increased and high purity, V = 0.5 and V = 0.3, respectively, depending on value of factor fs and ratio of working to required viscosity (Fig. 2.27, top), factor s takes values of s > 1. At increased soiling, V = 2, and high soiling, V = 3, depending on the factor fs-, purity factor takes values of s < 1 (Fig. 2.27, bottom). the The soiling factor V is dependent on bearing size, type of contact, and oil purity class [ 5 ] .When soiling occurs under high-load conditions, there is a stronger negative influence with decreasing contact surface area. As the size of rolling element bearings decreases, the sensitivity to soiling increases. Using ceramic rollers with or without a coating and steel raceway with a modification of surface, hard and soft coating or both, special heat treatment can improve bearings life, particularly in extreme condition such as in high speed, poor lubrication, high temperature, high vacuum, ceramically active environments, presence of water, and other situations in which acceleration of wear is indicated [ 10- 121. The effects of fire-resistant fluids on bearing wear must be treated differently than mineral-oil-based hydraulic fluids [ 131. For rolling element bearings lubrication with fire-resistant hydraulic fluids, the equation and Table 2.5 provided in Section 5.3 may be used. Bearing manufacturers often recommend hybrid and ceramic bearings for use with fire-resistant hydraulic fluids. The calculation procedures are iden-

m

UJ

m

mcn

K=O. 7~=0.6

2,s 3

4

5 6 7 8 910 12 14 16 Stressing index f,.

201

1

2

3 5 10 15 20 Cleanliness factor s

30

0.03

Cleanliness factor s depending on V, stress level, expressed by f,., and viscosity ratio K

Figure 2.27 Cleanliness factorviscosity ratio.

s depending on V , stress level, expressed by f s and the

aa

Pavlovich

tical as those used for steel bearings [5]. Under conditions of good lubrication, the absence of chemically active substances, and the absence of water and soiling, this approach is suitable because both steel and ceramic bearings have approximately equal working capacities. The working capacity of ceramic bearings with poor lubrication, the presence of soiling, high vacuum, and especially the presence of chemically active substances and water is several times higher than for steel.

5 COMMON WEAR MECHANISMSRolling element bearing wear varies with the loading history that will determine the mechanism of surface failure and wear. Proper analysis and identification of these failure mechanisms is critically important if future occurrences are to be avoided. Failure analysis procedures require consideration of bedding design for the bearing, type of bearing, material of construction, production procedures to increase contact surfaces resistance to wear, manufacturing precision, determination of the contact surface topography, type of lubrication, lubricant, and system maintenance procedures. With the exception of abrasive wear resulting from particular contamination, the most common failure mechanism of rolling element bearings is fatigue wear. Other types of wear, including adhesive, corrosive, and fretting, are less frequently encountered as the root cause of the bearing failure. However, they may occur as secondary wear processes.

5.1

Fatigue Wear

Fatigue wear is caused by cyclic action of contact stresses in surface layer. Figure 2.28 [14] shows the distribution of the contact stresses in the rough contact layer. As a result, complex stress states are present that are locally significantly and may exceed the yield stress after many repetitions, creating conditions for initiation of the fatigue crack. Figure 2.29 [ 151 illustrates the distribution of normal and shear stresses under conditions of rolling without sliding. The primary shear stresses T = 1/2(u, - 0,) achieve their maximum value at a depth of 0.78a, where a is the half-width of the Hertzian contact surface [15]. These shear stresses are the primary cause of initial crack creation, which, in the absence of other causes, appear in the zone of maximum stress. When both rolling and sliding of contact surfaces occur, stress distribution is determined by the magnitude of friction force as shown in Fig. 2.30 [15]. Under these conditions, maximum shear stress, with increase of shear force, moves toward the surface, causing the fatigue crack to form closer to the surface. Crack propagation is significantly influenced by physical, chemical, and metallurgical characteristics of the friction pair, as well as characteristics of the applied lubricant [ 161. Observation of the macro and micro records of the surface layer of bearings exposed to high loads shows that a dark bandlike zone forms at a depth of 0.1 -0.5 mm, preceding crack initiation. Upon enlargement, white spots are observed that are


89

1

I - 1.0.

-0.5a

I

0I

I

0.5.

I

1.0.

1

Contact in X axis h hcremeclts 04

(1/2 contact bngW

Figure 2.28 Contact stress in surface layer.

0

Q

2aDepth of surface layer

3a

Y

0,7& aout sliding.

3a

Y

Figure 2.29 Distribution of normal and shear stresses under the condition of rolling with-

90

Pavlovich

2a

3c

Y

I

9

a

2C

30

J

Figure 2.30 Distribution of stress under sliding and rolling conditions.

characterized by higher hardness and a cubic structure. Fatigue crack initiation occurs on the boundaries of these spots [16]. Figure 2.31 1171 illustrates typical changes of the material structure and also material properties, white bands, caused by repeated material stressing. This structural change is a decay of the martensite. Figure 2.32 [16] illustrates a microcrack in the initial failure phase. The mechanism and speed of fatigue wear contact-surface failure of rolling element bearings depend on various factors, including magnitude of normal and shear stresses, significant influence of the surface-layer structure, load history, presence of the hard particles during the contact-surface coupling, corrosive activity of environment, lubricant base fluid, and composition [ 16,18-251.

5.2

Abrasive Wear

The presence of low levels of abrasive particles in the contact zone accelerates fatigue crack initiations decreasing resistance to fatigue wear [25-271. In the presence of hard particles with intensive destruction processes occur at the contact surface resulting in rapid failure of the rolling element bearing. Although the mechanism of material removal from the contact surfaces is basically identical, two types of abrasive wear have been reported: wear caused by impact of the abrasive particles on the surface, and wear caused by coupling of the contact surfaces by the abrasive particles. The ratio of material surface hardness to abrasive particle hardness is directly related to the surfaces resistance to abrasive wear, as shown in Fig. 2.33 [ 7 ] . If particle hardness is lower than the hardness of the material working surface, the


91

Figure 2.31

Typical material changes of fatigue specimens- white bands.

Figure 2.32 Microcracks In the initial failure phasc (clectron micrograph, two-stage surface replica, magnification 17,SOOX).

Pavlovich

HBaFigure 2 3 .3rasive wear. Volume of wear particles depending on the hardness of the surface and ab-

quantity of the wear product increases slowly with increase of the hardness of the abrasive particle. This type of wear is designated light abrasive wear. When the hardness of the abrasive particle is approximately equal to the working surface hardness, greater quantities of wear products will result. If the hardness of the abrasive particle is greater than the working surfaces hardness, no significant increase in wear products will be obtained. However, a significant increase friction will result. This is called heavy wear. In rolling element bearings, depending on the lubrication regime and hardness and size of the abrasive particles, light and heavy wear frequently occurs in combination with other types of wear. Generally, it is possible to control this type of wear by controlling the size and quantity of abrasive particles present in the lubricant. The intensity of wear as a function of the concentration of the introduced abrasive is shown in Fig. 2.34 [20]. The solid line represent a new bearing with a fresh abrasive, and the dashed line represents the same relationship when an old lubricant is with the new bearing. These data show that the intensity of wear is most intensive during the first few hours and that wear increases with each addition of the new abrasive. In Fig. 2.34 wear intensity is significantly smaller with a used lubricant. In Fig. 2.35 [26] is shown the relative bearing life depending on the kind of indented particles [26].

5.3

Adhesive Wear

When the lubricant film thickness is less than the height of surface roughness, direct contact occurs between the surface asperities when the bearing surfaces are in relative motion. High localized pressures are present at the wear contact points. When the surface asperities come into contact, strong intermolecular bonds between the two surfaces may be formed, as illustrated in Fig. 2.36. This may result in the formation of microwelds. Because of tangential surface movement, existing bonds may be destroyed and new microwelds may be formed. This adhesive wear process is called

Bearing Selectionand Lubrication

93

Concentration of abrasive

hg//l

Intensity of wear as a function of concentration of the introduced abrasive; solid line represents new rolling bearing with fresh abrasive; dashed line represents new bearing with old lubricant.

Figure 2.34

0.01

U.............. .............. .............. ............................ :.............. . .... ... :::: ~:::x::~.... ..............

Relative bearing life depending on the kind of indented particles. (Reprinted with permission. SAE paper, Loesche T., Weingard M. Quantihing and Reducing the EfSect of Contanzination on Rolling Bearing Fatigue, c. 1995, Society of Automotive Engineers, Inc.)

Figure 2.35

94

Pavlovich

1 - metal 2 oxide layer 3 - monomolecular layer 4 - polymolecular layer

-

Figure 2.36 Modcl of contact surfaccs undcr the condition of mixed lubrication. (1) metal; (2) oxide layer; (3) monomolecular layer; (4) polymolecular layer.

scuffing and is characterized by significant changes in surface topography (Fig. 2.88b [17]) and structure (Fig. 2.37 [17]) and by the formation of wear debris. Rolling bearings subjected to high concentrated loads and significant raceway deformation with rolling elements characterized by a significant presence of inertial resistance to acceleration are frequently susceptible to adhesive wear. A change occurs in the perimeter velocity of the rolling element, and the angular velocity is significantly decreased with respect to the velocity of the forced rotation, as illustrated in Fig. 2.38 [17]. The reduction in surface velocity creates a situation of reduced elasto-hydrodynamic lubrication leading to surface asperity contact [ 171. The resulting stresses at these asperity contacts is significantly higher than the yield point of the metal which leads to the formation of frictional bonds. The formation and subsequent breaking of frictional bonds results in the destruction of contact surfaces and adhesive wear. The presence of wear debris from this process at the wear contact coupling zone may initiate subsequent abrasive wear, depending on the hardness of the wear debris and the initiation of cracks and fatigue wear as a secondary process.

5.4 Corrosive WearCorrosive wear refers to surface destruction of steel, nickel, aluminum, and other metal surfaces by chemical or electrochemical attack. Surface corrosion produces

Figure 2.37

Cross section of damaged cylinder roller.


95

Figure 2.38

The change in surface vclocity of the roller.

oxide coatings. passivating them to further attack. The type of surface passivation is dependent on the chemical characteristics of the corrosive environment and on the bearing material. For steel alloys, the oxide coating is typically paramagnetic iron oxide wFe,O,, ferromagnctic iron oxide y-Fe,O,, or magnetite FeiO,, which is typically formed in the presence of water, either in the form of humidity or may be present in the fluid. Magnetite may be transformed into rust (Fe,O,.HIO). The oxide coating thickness and tribological characteristics are dependent on the quantity of the fluid. The most frequently encountered forms of corrosive coating may be classified by its color and thickness; thin, relatively bright 500 pm. Intensive contact loadings in the presence of unresisting oxide surface coatings destroys the pasivating oxide layers, forcing the corrosion process deeper into the metal surface.

5.5

Fretting Corrosion

Fretting corro\ion, illustrated in Fig. 2.39 [ 281, is most often encountered for machine elements constructed from non-corrosion-resistant materials that undergo small but repetitive movements (1 Ibratioas) and are exposed to corrosive environments, Surface destruction caused by fretting corrosion forms \mall micron-sized craters filled

Figure 2.39

Raceway damaged with fretting corrosion

96

Pavlovich

Figure 2.40

Typical phases in a fretting corrosion process.

with wear debris. The surfaces are characteristically covered with red Fe,O, or dark Fe,O, oxides or Fe,O,.H,O if humidity is present. Figure 2.40 illustrates that fretting corrosion starts with hardening and cyclic plastic flow of the subsurface layer at the wear contact. This is caused by the vibration movement of the load. Surface asperities are plastically deformed where oxide layer destruction has occurred and intermolecular bonds develop. Mutual vibrational movement of the contact surfaces breaks these intermolecular bonds and, in the presence of the debris from fatigue material damage, creates conditions for contactsurface destruction. Initially, the wear debris consists of the metal oxide material. The second phase of the fretting corrosion wear is characterized by formation of the corrosive environment which consists of wear debris, absorbed oxygen, and moisture (H,O). A steady-state balance is established between the formation of wear products and their removal from the contact zone. Surface deformation activates the tribo-chemical reactions involved in the corrosion process. When the thin outer layer of surface oxides is destroyed, the work-hardened layer immediately below is subjected to the vibrational loads, leading to fatigue loads accelerated by the corrosion process. Upon removal, the metal oxide debris acts as a catalyst accelerating the adsorption of oxygen and moisture. A reactive electrolytic environment is formed between the wear contact surfaces. Corrosive destruction progresses into the subsurface layer, creating corrosive fatigue damage, thus accelerating the wear process. The investigation of the influence of the effect change on the angle of axle and the change in load showed that the angle of movement having a great effect on fretting wear is no more than 1 degree, but increases rapidly when the angle rate is greater (Fig. 2.41 [28]).

Bearing Selection and Lubrication180 r

97Innerring/

120 . 100 140

P

0

w.

0.5

'

1.0

1.5

1

Angle of rnoverntncAS

angle o i movemenr c;ianges (Load= 1470 N in this w e )

Figure 2.41 Influence of the angle of movement on the rate of increased breadth.

6

ENHANCEMENT OF ROLLER BEARING WEAR RESISTANCE

Roller bearings are typically subjected to various types of wear process throughout their lifetime, leading to a complex overall surface destruction process. The wear contact surface destruction process can be retarded by various measures, including wear surface material selection, process of producing these wear surfaces (heat treatment and surface roughness) of the bearings and raceways and resulting accuracy of the production process, protection of the bearing assembly against corrosive environments, lubricant selection, method of bearing lubrication, and system cleanliness.6.1

Influence of Bearing Design on Wear

Bearing wear can be affected by the operating conditions and design. Bearing operating conditions can be affected by various factors, including selecting proper bearing type, assuring necessary accuracy of shafts and beddings, providing adequate shaft and bedding stiffness, protecting against the operating environment, lubrication system, and lubricant selection. Aspects of bearing design that affect wear include material pairs for the wear contact surface, size of the wear contact, weight of rolling elements, maximum friction reduction, load distribution, increasing working capacity, decreasing size, high speeds, autonomy, and reduction of system contamination. This discussion will be limited to the effects of bearing design on wear. The best bearing designs should increase the wear contact surface area, provide homogeneous distribution of contact stresses, and organize spreading of deformations, as illustrated in Fig. 2.42 [29]. Usually, bearing design changes are made to modify the curvature of the rolling element and raceway contact with the inside and outside rings. Bearings with cylindrical and conical rolling elements exhibit line contact, and modification is performed to avoid peaks that appear at the end of

98

Pavlovich

0.60.4

0.2

0

Figure 2.42

(a) Direction of deformation in a Hertzian contact; (b) distribution of stress in Hertzian contact; (c) distribution of principal and resulting stress under contact point.


99

Resulted stressz0 .

.?.5

-/0

0.5

i.O

:,L .

1.0

(c>

Surface layer depthContinued

Figure 2.42

contact, as illustrated in Fig. 2.43 [29]. Macrostress distribution improvement may be achieved by rounding the ends of cylindrical and conical rolling elements or by using raceways with a convex profile as illustrated in Fig. 2.44 [29]. In convex-roller bearings, the roller profile radius is smaller than the radius of raceway profile. Theoretically, the contact is treated as a rigid body and forms a point contact. Because

Line profile

I

Convex profile

Figure 2.43 Distribution of presure; convex profile of raceway; straight-line profile of roller; straight-line profile of raceway; and convex profile of roller.

100

Pavlovich

Line profile

I

Convex profile

Figure 2.44 Distribution of pressure; straightline profile of raceway; straight-line profileof roller; convex profile of raceway; and convex profile of roller.

Very low load

F

< Fgr

Low load

F> Fgc

F General

I

I

I

I

I

I

Ellipse

Fragment of ellipse

Rectangle-~~~

Line contact

Figure 2.45 The forms of contact area.


101

Figure 2.46

Dependence of rolling element and the raceways curvature on wear intensity.

the profile curvature of the contact surfaces is negligible at small loads, F < F,,, the surface contact forms an ellipse, as illustrated in Fig. 2.45 [29]. Figure 2.46 [30] illustrates the effect of rolling element and raceway curvature as expressed by the curvature parameters CO, C,,,,: and

These studies, which were conducted by measuring the radioactivity increase within the inside raceway, suggest mutual dependence of the ratios of coupled surface profiles and wear intensity. Studies on the influence of different contact-surface profile curvatures in convex-roller bearings [30] suggest a positive influence of the small values of the curvature factor, C,,,,< l % , on wear reduction. At higher values of the curvature factor C, wear increases, which may be the result of contact-surface reduction and the corresponding increase in surface pressure within the contact zone. In addition, results of these studies suggest that wear debris in the contact zone can participate in the destruction of the roller-bearing surface. Reduction in surface wear and an increase in efficiency ratio can be achieved by modification of the contact surface in the main load direction. Modification can also be performed on the face ribs and on the rolling element separators to reduce friction parasitic losses. Figure 2.47 [31] illustrates the design of the rolling bearing with face contact over the conix and the convex front face. Investigations have shown that the friction losses in this case are less than losses of friction that occur during contact of flat front faces (Fig. 2.48) [31]. Decreasing parasitic friction losses in the bearing separator contact is most frequently accomplished using a material with a low friction coefficient (zinc-phosphate coatings, PTFE, MOVIC). The coil separator illustrated in Fig. 2.49 [32] is elasto-hydrodynamically lubricated if there is sufficient fluid flow through the wear contact. This type of separator has performed well for low-speed applications such as gyroscopes. One of the causes of vibration of the bearing is the shape error of bearing parts, such as the circularity of bearing track, cylindricity, eccentricity, paralellism, roughness of rotating surface, and the size variation of rolling elements per unit container. Roller bearings with a size accuracy of 175"C, steel alloys containing molybdenum, tungsten, and vanadium alloying elements are used. These steels are capable of maintaining a hardness of 58 HRC at working temperatures as high as 315-480C. Halmo and M50 steels are used for working temperatures up to 310C. For operation at extremely high temperatures (up to 427"C), T-1 and M-10 steels may be used. For working temperatures up to 492"C, AISI H-1 and H-2 steel alloys are used, and for temperatures up to 538"C, AISI WB-49 is used. Highly alloyed steels

lrticle

10

102

103

104

Lifetime. hoursFigure 2.54 Extension of lifetime through increased purity of bearing steel.

106

Pavlovich

that maintain high hardness at elevated temperatures typically exhibit lower resistance to contact surface destruction at working temperatures of ZOO0 HV and it is well connected to the base material. A special PVD device is used to deposit TiN after cleaning. The process of material deposition is conducted from the plasma state to the desired surface or surfaces. The working piece temperature during the coating deposition phase is ~500C. bright silver chrome-nitride Cr/CrN coating is used The for improving wear resistance. It exhibits a hardness up to 1500 HV. It is deposited by a PVD process at temperatures lower than 500C. The bright black Me-I-C coating is deposited to thicknesses of 1-4 p m and it is very well connected to the base material. The coating hardness is about 1000 HV and is used for improving sliding properties and wear resistance. This coating is limited to working temperatures of about 300C. The I-C coatings are carbon coatings of amorphous dark color and they are deposited to thicknesses of 1-3 p m with a hardness of 3000-6000 HV. The connection with the base material is good. It is deposited at temperatures of 150-200C. The coating improves the wear resistance and sliding properties of the base material. The most frequently applied CVD coatings are titanium nitride and titanium carbide. The TiN coating exhibits a shiny golden color and a hardness >2000 HV and is deposited to thickness of 2-4.5 pm. The coating produced by the CVD procedure exhibits a very good connection to the base material. The coating deposition temperature is higher than 1000C. The coating improves the base material wear resistance but worsens weldability. Titanium carbide (Tic) coatings are applied by ion implantation and exhibit a shiny silver color and a hardness of 3500 HV and it is well connected to the base material. It is deposited to a thickness of 3 pm. The deposited coating worsens the surface quality of the base material and requires additional machining to ensure the required quality. The deposition temperature is over 1000C. A film thickness of less than 0.5 p m improves the wear resistance of the base material. The appearance and surface structure remain identical to the initial ones. The maximum working temperature is determined by the base material. In combination with the PVD coating MoS, it is used for improving wear resistance and sliding characteristics of rolling elements of AISI 440C steel.


113

Ion-implanted CrN coatings are deposited to a thickness of less than 0.5 pm and are used to improve the wear resistance of the base material. The surface roughness is unchanged and the maximum working temperature is limited by the base material. The most frequently used coatings applied by dipping and spraying are zinc and ceramic coatings. Ceramic coatings (A1,O + Ti02) are metallic gray and are deposited to a thickness of about 250 pm. They exhibit a gray porous structure and are significantly harder, frequently over 1000 HV, than the base material. Their corrosive stability is excellent and they may be used at all normal working temperatures. The connection with the base material is good, but coated surfaces are additionally machined to achieve the desired surface quality. Ceramic coatings are used to isolate the influence of electric current. They are well suited for application to rolling bodies or the races. The physical and chemical properties of PVD layers have an intermediate position between metals, and ceramic PVD layers are electrically conductive and have a metallic luster, but they do not build up alloys with other metals and they are sluggiqh in reaction with other materials [12]. In the past few years. full-coated bearings can be found in a new approach to combining ceramic balls with coated races to improve the behavior of a tribological system [ 121. Because of their superior properties, four different kinds of coating have been selected for coating the races of the bearings: CrN, TiAlN, TiN C, and WC C. Figure 2.63 summarizes the properties of the different coatings [12].

+

+

6.3

Influence of Lubricant and Lubrication Systems

Elasto-hydrodynamic lubrication is dependent on many factors that include working surface geometry, elastic properties of the material, thermal conductivity, physical

Figure 2.63 Comparison properties of coatings for roller bearings.

114

Pavlovich

properties of the fluid, loads, speed, and the lubrication system. It has been shown that the use of bearings constructed from high-purity steel, high manufacturing accuracy, operating with a sufficient number of revolutions, with a fluid of adequate viscosity, with fluids free of contaminants, and under conditions of no contact deformation can exceed their life rating by up to 10 times. Under conditions of elastohydrodynamic lubrication, coupled surfaces are typically separated by an oil layer of 2-3 p,m. Therefore, to assure no asperity contact between the working surfaces is to assure both manufacturing accuracy and the high quality of the surface finish. To assure optimal bearing lifetimes, elimination of particles larger than the thickness of the oil layer is necessary. However, elasto-hydrodynamic lubrication rarely occurs under conditions of total absence of metal surface contact which occurs, for example, at the beginning of bearing rotation. Particles originating from wear debris and other sources, if their size is greater than the fluid film thickness, will increase the contact fatigue of the bearing surfaces. Elasto-hydrodynamic lubrication significantly reduces friction, therefore moving the point of the fatigue crack initiation deeper into the surface layer. Figures 2.64 and 2.65 [43] illustrate the pressure distribution in the fluid film and stress distribution with material depth, respectively. Changes of pressure and the physical form of the fluid film as a function of physical properties of coupled cylindrical surfaces and the lubricant are shown in Figure 2.66 [43]: Curve a refers to rigid cylinders with a constant lubricant viscosity of lubricant; curve b refers to rigid cylinders and viscosity dependent on pressure; curve c refers to elastic cylinders and constant viscosity; curve d illustrates elastic cylinders and viscosity dependent on pressure. Figure 2.67 [43] illustrates the variation of fluid pressure distribution and fluid film thickness as a function of material parameters. The pressure distribution and changes of the oil layer form as a function of rotational velocity is illustrated in Fig. 2.68 [42]. These data show that at constant load, the maximum pressure, minimum oil layer thickness, and oil layer gap are functions of the rotational velocity of the contact surface. Figure 2.69 [43] illustrates the pressure distribution of nonidealized rough surfaces. Studies of the influence of surface roughness on the rough layer carrying

/ ; - 7Figure 2.64

-

I------

Pressure distribution in fluid film.


115

E

3

c-!.6 -1.2I

-0.8I

-0.4I-I0

0I 0

0.4

0.61

inmm

r10

-30

-20

23

It

Figure 2.66 Change of pressure and physical form of the fluid film as a function of physical properties of coupled cylindrical surface.

116

Pavlovich

-2

I

0

I

I

I

-2

I

0X

I

-2

II

0X

I

Figure 2 6 Variation of fluid pressure distribution and fluid film thickness as a function .7of material parameters. W = 3 X lO-; U = l o p ; G = 5000 for (a); G = 2500 for (b).

capacity showed that increased longitudinal roughness significantly reduces the carrying capacity of the oil layer [42]. The oil layer not only provides lubrication, which is pressure and speed dependent, but it also cools the wear contact. Figure 2.70 [43] depicts the temperature field within the elasto-hydrodynamic contact. The lubrication properties of the roller bearing at the moment that motion begins is characterized by direct contact of working surfaces, localized interruptions of fluid film, insufficient conditions to provide hydrodynamic lubrication, and the tribological characteristics determined by the interaction of very thin fluid film and the material surface. Lubrication (FT)under these conditions is either mixed-film or boundary lubrication. Friction under these conditions is determined byFT

= ~ A , T ( I - 1 CPa

Ref.182

183184185 186

Max. 1 0 M P a

Max. 50 MPa

Apparent impact pressure Max. 15 MPa Erosive impact pressure >70 mPa

187

204

Totten et al.

Table 4.14 Mechanical Properties of Various MaterialsSteel JIS AISIs35c s55c HT80

Tensile strength U (MPa)420 615 706 853 196 145

Modulus of elasticity (GPa)204 204 204 204 122 71

Vickers Hardness (HI,)130 180 240 300 98 41

s 1oc

cuA1

Soitrce: Ref. 142.

impact force [142]. Therefore, the potential for cavitation is dependent on the material properties, especially tensile strength and the initial condition of the material surface [ 142,1431 and, in some cases, material structure [ 1441. Some illustrative examples are provided in Table 4.14 [142]. Okada et al. have determined that impact loads of 9.1 N (aluminum) 9.7 N (copper), and 13.7 N (mild steel) are required to form a 4-pm pit from a single implosion of a single bubble [ 1451. However, the bubble-collapse pressures will vary with the bubble size, shape, and location. Talks and Moreton evaluated the volumetric cavitation erosion rates of different hydraulic fluids toward different steels using an ultrasonic cavitation erosion test; see Table 4.15 [146]. Their results, summarized in Table 4.16, illustrate that the general order of cavitation of different fire-resistant hydraulic fluids toward the different steels was water > W/O emulsion > water-glycol > invert emulsion > mineral oil. Although cavitation erosion rates were dependent on fluid properties, the erosion damage was only a function of the material properties [ 1461. Tsujino et al. in another ultrasonic laboratory study with a different group of hydraulic fluids found the order of erosion rates shown in Fig. 4.101 to be [147] water > HWBF > water-glycol > phosphate ester > mineral oil; HWBF denotes a high-water-base fluid. These data are consistent with the results predicted by Hara et al. from cavitation number calculations shown in Fig. 4.86 [131]. Typically, cavitation bubbles exist in a large clouds. Figure 4.102 illustrates a cavity formed by cloud of bubbles from a HWBF. The actual eroded test specimen

Table 4.15 Volumetric Cavitation Erosion Rates for Different Hydraulic Fluids TowardDifferent Steels Volumetric Erosion Rates (mmh)Stee1

Distilled water1.47 0.275.54

WIO Emulsion 5.5 1.2 0.22

Water-glycol0.73 0.34 0.02

Invert emulsion0.21 0.05 0.02

Mineral oil0.27 0.02 0.0 1

Mild steel Stainless steel Bearing steel

Note: Data based on an ultrasonic open-beaker laboratory test. Source: Ref. 146.

Physical Properties and Their Determination

205

Table 4.16 Cavitation Rating of Different Fluids____________~ ~

Cavitation rating Very low Low Medium low Medium high High

Ce2,mg, 4 3 21

1 Straight mineral oil Tellus-68 Turbo-T68 Tonna-T220 Vitrea-150 Ondina-Vitrea-9

2 Mineral oil with additives

3 Biodegradable hydraulic oils BD 32

4 Synthetic-base oil Cassida-HF68 Cassida-GL 150 Cassida-HF46 Cassida-GL460 Cassida-HF32 Cassida-HF15

0

Naturelle-HE46 Naturelle-HFR32 Omala-320 Tellus-32 Tellus-T32

Cassida-GL46O Cassida-HF32

Source: Ref. 168.

is shown in Fig. 4.103 [148]. The varying distribution and shape of the eroded surfaces formed by different fluids was examined by scanning electron microscopy (SEM) and is illustrated in Fig. 4.104 [ 1481. Different studies have shown that every bubble in the cloud does not cause damage upon collapse. The damage frequency has been reported to vary between 1 in 16,000 [ 1491 and 1 in 30,000 [ 1501. Cavitation bubbles may undergo multiple collapse and reformation processes. Knapp and Hollander have reported that a single bubble may undergo as many as seven collapse and reformation processes [ 1511. The breakage and reformation process will cease as the viscosity of the surrounding fluid dissipates the energy generated from bubble breakage. Backe and Berger have reported that cavitation wear rate (mm'/h) can be estimated from material properties [ 1521:

where

E (N/mm2)= the elastic modulus of the material R,,, = the maximum yield strength at 0.2% elongation R, (N/mm2)= tensile strength W (N/mm') = the work to deform the test specimen for the tensile test (A&)) H = the Vickers hardness p,,.(g/cm3) the material density. =

2.5.8 Test MethodsThere have been a number of test procedures reported; they include spark-induced bubble formation and cavitation collapse [ 1531, hydraulic cylinders [ 154,1551, hydraulic circuits [ 131,1561, and other devices. Although these tests, particularly the use of hydraulic circuitry, continue to be of interest, there are primarily two types

Totten et al.

/

1

f

I

20% HWBF

/

t (min)Figure 4.101 Propensity for cavitation damage for various hydraulic fluid types. (FromRef. 147.)

of laboratory screening test to determine the cavitation potential of fluids: submerged jets and ultrasonic vibratory apparatus. Two common submerged-jet test configurations are in use. One is the configuration reported by Lichtarowicz and Kleinbreuer [ 157- 1631. An illustration of the system reported by Lichtarowicz is provided in Fig. 4.105 [159] and the system reported by Kleinbreuer is illustrated in Fig. 4.106 [162]. The second most common type of test cell that is used for laboratory studies of cavitation is the ultrasonic vibratory test [ 164- 1671. The ultrasonic vibrational cavitation test may be conducted in one of two possible configurations: the specimen or the fluid may be vibrated [ 1661; the test configuration of the ASTM G 32 test

Physical Properties and Their Determination

287

Figure 4.102 Illustration of a cavity formcd by a cloud of bubbles from a high-water-base fluid. (From Ref. 118.)

p

handbook of hydraulic fluid technology

Documents

systems design

machine design

electronic design

design of mechanical

mechanical design problems

applications handbook

knox design dimensioning

cam design software