Document NGQYwXeEk6okQRGvKz6ByLMY8

Report 68-04-0020 BRAKE EMISSIONS: EMISSION MEASUREMENTS FROM BRAKE AND CLUTCH LININGS FROM SELECTED MOBILE SOURCES M. G. Jacko R. T. DuCharme Bendix Research Laboratories Bendix Center ' Southfield, Mich. 48076 March 1973 Final Report for Period May 1971 - March 1973 Prepared for ENVIRONMENTAL PROTECTION AGENCY Office of Air and Water Programs Ann Arbor, Mich. 48105 HWBUI0002287 BRAKE EMISSIONS: EMISSION MEASUREMENTS FROM BRAKE AND CLUTCH LININGS FROM SELECTED MOBILE SOURCES M. G. Jacko R. T. DuCharme HWBUI0002288 FOREWORD The work described herein, "Emissions Measurements from Powered Brake and Clutch Linings From Selected Mobile Sources," was performed for the Office of Air and Water Programs, Environmental Protection Agency, Ann Arbor, Michigan. This work was carried out under EPA Con tract 68-04-0020 from May 10, 1971 to March 31, 1973. The EPA Project Technical Officer was Dr. Joseph H. Somers. The program was contracted to Bendix Research Laboratories, Southfield, Michigan. The work was performed in the Materials and Pro cesses Department, Dr. William M. Spurgeon, Manager, of the Mechanical Sciences and Controls Laboratory, Mr. L. B. Taplin, Director. The program was under the supervision of the Principal Investigator, Dr. Michael G. Jacko. Sections 4, 5, and 6 of this report were prepared by Mr. R. T. DuCharme. The data base of Section 10 was prepared by Mr. A. R. Spencer. All other portions were prepared by Dr. Michael G. Jacko. Other contributors to the program were as follows: Mr. R. D. Stapleton (sample collection, sample processing, calculations, and numer ous other tasks); Dr. R. M. Rusnak and Mr. D. G. Jones (microscopy analy ses) ; Mr. L. Smith (vehicle test driving); Mr. H. M. Danbert (vehicle preparation and test driving); Mr. P. R. Stewart (fabrication of collec tors); Mr. T. N. Vinson (technical writer); Mr. J. Gulvezan, Mr. A. Tomala, and Mr. W. R. Kee of Machine Systems Department (rotating seals designs), and Mr. C. Cowan and Mr. C. Morton of Bendix Automotive Controls Systems Group (information on vehicle testing and friction materials). Computer programming and processing of computer data were performed by Miss J. Lindsay. Mr. C. L. Gray and Mr. W. H. Houtman of EPA are acknowledged for their suggestions incorporated into the brake emissions collectors. Special thanks go to Dr. W. M. Spurgeon, Manager of the Materials and Processes Department, for his encouragement and suggestions through out the program. ii HWBUI0002289 ABSTRACT In order to define the extent of gas and particulate emissions from automotive brakes (and clutches), a combination separation and storage collection system was devised. Unique emissions collectors for both disc and drum brakes and for a clutch were conceived, designed, and built as the main embodiment of this instrumentation. The hardware was installed on a vehicle which was then driven through various test cycles to deter mine the extent and type of brake emissions generated at low and high operating temperatures. Typical driving conditions, such as the Detroit Traffic Test, as well as more abusive driving conditions such as fade tests, were included. Typical original equipment and aftermarket friction materials for both disc and drum brakes were used in the tests. Brake relines were made to simulate typical consumer-type practices. The brake emissions generated were removed from the various collectors and mass balances were performed. The particulates were processed and analyzed by a com bination of optical and electron microscopy to ascertain the asbestos content and the asbestos particle size distribution in the wear debris. Comparisons of emissions from new and used friction materials, disc and drum brakes, and original equipment and aftermarket materials were made. Finally, an estimate was made of the total emission of asbestos by all of the vehicles in the country: on the average, more than 99.7 percent of the asbestos is converted; the contribution to the atmosphere is 5060 pounds, or 3.2 percent of the total asbestos emissions. iii HWBUI0002290 TABLE OF CONTENTS SECTION 1 - INTRODUCTION AND BRIEF SUMMARY SECTION 2 - OBJECTIVES AND BACKGROUND 2.1 Objectives 2.1.1 2.1.2 Objectives of this Program Bendix Technical Approach 2.1.2.1 2.1.2.2 2.1.2.3 2.1.2.4 2.1.2.5 2.1.2.6 Select Suitable Vehicle and Friction Materials Conceive, Design, and Build Brake and Clutch Emissions Collectors Test a Typical Collector and Instrument the. Vehicle Select and Run Suitable Driving Test Schedules Collect and Analyze Emissions Estimate Asbestos Emissions for Cars and Trucks 2.2 Prior Work 2.2.1 2.2.2 Published Data Unpublished Data 2.3 Background 2.3.1 Energy Absorption by Friction Materials 2.3.1.1 Brakes 2.3.1.2 Clutches 2.3.2 Compositions of Friction Materials 2.3.2.1 2.3.2.2 2.3.2.3 2.3.2.4 Generalized Classifications Asbestos Resin Binders Property Modifiers 2.3.2.4.1 2.3.2.4.2 Non-Abrasive Modifiers Abrasive Modifiers 2.3.3 Friction Material Reactions 2.3.3.1 2.3.3.2 Absorption and Conversion of Energy Physical and Chemical Changes in Linings During Use 2.3.4 Brake Lining Wear Mechanisms and Generation of Emissions 2.3.4.1 Types of Wear Mechanisms Page 1-1 2-1 2-1 2-1 2-1 2-1 2-1 2-1 ^"2 2-2 2-2 2-2 2-2 2-3 2-3 2-3 2-3 2-3 2-4 2-4 2-4 2-5 2-5 2-8 2-8 2-8 2-8 2-9 2-13 2-13 v HWBUI0002291 2.3.4.1.1 2.3.4.1.2 2.3.4.1.3 2.3.4.1.4 2.3.4.1.5 Thermal Wear Abrasive Wear Adhesive Wear Fatigue Wear Macroshear Wear 2.3.4.2 PredominantWear Mechanisms 2.3.4.3 Wear Equation 2.3.5 Emissions of Particulate and GaseousEmissions 2.3.5.1 2.3.5.2 2.3.5.3 Drum Brakes Disc Brakes Clutches 2.3.6 Distribution and Changes in Brake andClutch Emissions 2.4 References SECTION 3 - MATERIALS SELECTIONS 3.1 Background 3.2 Materials Selections forProgram 3.2.1 3.2.2 Friction Material Selections Contents of the Friction Materials SECTION 4 - VEHICLE AND INSTRUMENTATION 4.1 Vehicle Description 4.2 Brake Test Instrumentation SECTION 5 - EMISSIONS COLLECTION SYSTEMS 5.1 Operation Requirements 5.1.1 5.1.2 Operation and Design Criteria Emission Expected 5.2 Overall Design Concept 5.2.1 5.2.2 5.2.3 Air Flow of a Typical Collection System Collection Systems Used Electrical System 5.3 Particulate Filters 5.3.1 5.3.2 Filter Holders Filter Flow Characteristics 5.4 Gas Collection Systems 5.4.1 5.4.2 Design and Fabrication of Traps Operation Page 2-13 2-16 2-16 2-19 2-19 2-21 2-21 2-23 .2-23 2-23 2-24 2-24 2-25 3-1 3-1 3-1 3-1 3-3 4-1 4-1 4-2 5-1 3-1 5-1 5-2 5-2 5-2 5-5 5-5 5-5 5-5 5-9 5-9 5-9 5-14 vi HWBUI0002292 5.5 Disc Brake Collector 5.5.1 5.5.2 Shroud Design and Construction Rotating Seal 5.5.2.1 5.5.2.2 Initial Design and Problems Final Design 5.5.3 Thermal Control Aids 5.6 Drum Brake Collector 5.6.1 5.6.2 Initial Designs and Rotating-Seal Problems Final Design 5.7 Clutch Collector SECTION 6 - INERTIA DYNAMOMETER TESTING OF THE DISC-BRAKE COLLECTOR SYSTEM 6.1 Objectives 6.1.1 6.1.2 6.1.3 Demonstration of Collection System Operation Thermal Response Rotating Seal Durability 6.2 Dynamometer Test Results 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 Sealing Practices for Disc Brake Collector Thermal Response Results Rotating Seal Operation Emission Collections System Integrity SECTION 7 - VEHICLE TEST SCHEDULES 7.1 Rationale 7.1.1 7.1.2 Brakes Clutches 7.2 Typical Test Cycles 7.2.1 7.2.2 7.2.3 7.2.4 7.2.5 7.2.6 Burnish Baseline Detroit Traffic Test 10-Stop Fade and Recovery 15-Stop Fade and Recovery Reburnish 7.3 Selected Vehicle Driving Schedule 7.3.1 7.3.2 7.3.3 Burnish After-Burnish (A.B.) Baseline Detroit Traffic Test Page 5-14 5-14 5-14 5-14 5-22 5-22 5-27 5-27 5-31 5-31 6-1 6-1 6-1 6-1 6-1 6-1 6-1 6-2 6-2 6-7 6-10 7-1 7-1 7-1 7-1 7-1 7-1 7-2 7-2 7-4 7-4 7-4 7-5 7-5 7-5 7-5 vii HWBUI0002293 7.3.4 7.3.5 7.3.6 7.3.7 10-Stop Fade and Recovery After-Fade (A.F.) Baseline 15-Stop Fade and Recovery Final Baseline SECTION 8 - SAMPLING AND ANALYTICAL PROCEDURES 8.1 Rationale Based on Collector Designs 8.1.1 8.1.2 Particulate Emissions Collection Gaseous Emissions Collection 8.2 Particulate Emissions Removal FromCollectors 8.2.1 8.2.2 8.2.3 8.2.4 8.2.5 Collection of Particulate Emissions Disc Brake Samples Drum Brake Samples Clutch Samples Improved Friction Material Weighing Procedure 8.3 Analytical Problem Definition 8.3.1 8.3.2 8.3.3 Composition of Wear Debris Analytical Methods for Asbestos and Their Rationale Microscopy Methods for Asbestos 8.4 Bendix Analytical Method for Analysis of Asbestos in Brake Emissions 8.4.1 8.4.2 8.4.3 8.4.4 8.4.5 Criteria and Flow Chart Analysis of Brake Emissions Representative Sampling and Low Temperature Ashing (LTA) Sample Distribution for Microscopy Selected Microscopy Methods 8.4.5.1 8.4.5.2 8.4.5.3 Initial Studies and Rationale Optical Microscopy Analysis for Asbestos Transmission Electron Microscopy Analysis for Asbestos 8.4.6 Calculations From Microscopy Results 8.4.6.1 8.4.6.2 8.4.6.3 Rationale and Microscopy Count Sheet Sample Calculation Computerization of Calculations and Particle Size Distribution 8.4.7 Possible Sources of Error Page 7-5 7-7 7-7 7-7 8-1 8-1 8-1 8-1 8-1 8-1 8-1 8-3 8-3 8-3 8-3 8-3 8-4 8-6 8-7 8-7 8-7 8-7 8-12 8-15 8-15 8-16 8-16 8-18 8-18 8-19 8-26 8-26 HWBUI0002294 8.5 Analysis of Gaseous Emissions 8.5.1 8.5.2 8.5.3 8.5.4 Method of Analysis Gas Handling System Calibration of Gas Chromatograph Problems and Interferences 8.6 References SECTION 9 - VEHICLE TEST RESULTS 9.1 Vehicle Operations 9.1.1 9.1.2 9.1.3 Test Scheduling and Problems Encountered Mileage Accumulations Reporting of Vehicle Test Data 9.2 Performance of Emissions Collection Systems 9.2.1 9.2.2 Rotating Seal Life Operating Temperatures 9.2.2.1 Normal Brake Stops 9.2.2.2 Heavy-Duty Fade Stops 9.2.3 Collection Efficiencies 9.2.3.1 Recovery of Brake. Particulate Emission 9.2.3.2 Mass Balance 9.2.3.2.1 Amount ofEmissions 9.2.3.2.2 Distribution ofEmissions 9.3 Percent Asbestos Contents 9.3.1 9.3.2 9.3.3 9.3.4 Program Results Battelle Analyses Johns-Manville Analyses Comparison of AsbestosAnalytical Results 9.4 Asbestos Emissions Factors 9.5 Asbestos Emissions Trends 9.5.1 9.5.2 9.5.3 9.5.4 9.5.5 New versus Burnished Materials Disc versus Drum Brake Materials Heavy (Abusive) versus Moderate Duty Effects of Friction Materials Airborne versus Sump versus SurfacesSamples 9.6 9.7 Changes in Lining Wear Rate After Use at Higher Temperatures Results of Gaseous Emissions Analyses 9.7.1 9.7.2 9.7.3 Summary of Analytical Methods Attempted Gaseous Emissions Detected Shroud Extract Analysis Page 8-28 8-28 8-30 8-33 8-33 8-36 9-1 9-1 9-1 9-5 9-5 9-5 9-5 9-8 9-8 9-12 9-15 9-15 9-19 9-19 9-19 9-24 9-24 9-28 9-28 9-28 9-30 9-37 9-.37 9-37 9-39 9-41 9-43 9-46 9-46 9-^6 9-^8 9-53 HWBUI0002295 SECTION 10 - ESTIMATED EMISSIONS FOR CARS AND TRUCKS 10.1 Approach 10.2 Supplementary Data Base 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 Number of Vehicles in Use Motor Vehicle Usage Estimates of Brake Usage and Abuse Asbestos in Friction Materials Amount of Friction Material Actually Worn 10.3 Interpretation of Test Results 10.3.1 Factors Influencing Rate of Emissions . 10.3.2 Weighting Factors 10.3.2.1 Distribution Weighting Factors 10.3.2.1.1 10.3.2.1.2 10.3.2.1.3 Location of Wear Debris Comparison of Normal versus Shrouded Brakes Summary 10.3.2.2 Severity Weighting Factors 10.3.2.2.1 10.3.2.2.2 10.3.2.2.3 10.3.2.2.4 New versus Burnished Friction Materials Severe or Abusive Braking Renewal of Friction Surface Normal Brake Operation 10.3.2.3 Summary 10.3.3 Asbestos Emissions Per Vehicle 10.3.3.1 Total Asbestos Emissions Estimate 10.3.3.2 Asbestos Emissions and Their Fate 10.4 Estimated Asbestos Emissions 10.4.1 Passenger Vehicles 10.4.1.1 Total Asbestos Emissions 10.4.1.2 Distribution of Asbestos Emissions 10.4.2 Trucks and Buses 10.4.2.1 10.4.2.2 10.4.2.3 10.4.2.4 10.4.2.5 Estimated Severity Factors Estimated Emissions Factors Calculations of Total Asbestos Emissions for Trucks and Buses Total Truck Asbestos Emissions Distribution of Truck Asbestos Emissions Page 10-1 10-1 10-1 10-1 10-4 10-4 10-9 10-15 10-16 10-16 10-18 10-18 10-18 10-19 10-22 10-23 10-23 10-23 10-24 10-24 10-24 10-24 10-24 10-25 10-25 10-25 10-25 10-25 10-28 10-28 10-28 10-29 10-31 10-32 x HWBUI0002296 10.5 10.6 10.4.2.5.1 10.4.2.5.2 Estimated Distribution Weighting Factors Truck Asbestos Emis sions and Their Fate 10.4.3 10.4.4 Estimated Asbestos Emissions Alternate Estimate for Total Asbestos Emissions Summary of Asbestos Emissions References Page 10-32 10-32 10-32 10-34 10-37 10-38 SECTION 11 - SUMMARY 11-1 APPENDIX A - WEAR DEBRIS CALCULATIONS A-l APPENDIX B - NUCLEP0RE FILTER FLOW CHARACTERISTICS FOR BRAKE LINING WEAR DEBRIS B-l APPENDIX C - COLLECTION OF BRAKE AND CLUTCH EMISSIONS C-l APPENDIX D - ANALYSIS OF BRAKE AND CLUTCH EMISSIONS D-l APPENDIX E - ELECTRON MICROSCOPY ANALYSIS OF BRAKE EMISSION SAMPLES E-l APPENDIX F - COMPUTER PROGRAM FOR COMPUTATION OF ASBESTOS CONTENT USING ANALYTICAL MICROSCOPY TECHNIQUES F-l APPENDIX G - VEHICLE TEST DATA REPORTS G-l APPENDIX H - ASBESTOS EMISSIONS ANALYSIS REPORTS FROM BATTELLE COLUMBUS LABORATORIES AND JOHNSMANVILLE RESEARCH AND ENGINEERING CENTER H-l ,IS II || . " i Hi xi HWBUI0002297 LIST OF ILLUSTRATIONS Figure No. 2-1 2-2 2-3 2-4 2-5 2-6 2-7 2-8 2-9 2-10 2-11 2-12 2-13 2-14 2-15 2-16 2-17 2-18 4-1 5-1 5-2 5-3 5-4 5-5 5-6 5-7 Title Schematic Structure of Chrysotile Asbestos DTA and TGA Thermograms for Chrysotile Asbestos (Bendix Data) X-Ray Diffraction Patterns of Heated Chrysotile Asbestos (Bendix Data) Photomicrograph of Brake Lining Cross Section after 8550 Miles of Test-Track Service Electron Micrograph of Friction Material Wear Debris (Made during 1966 studies.) - 10,000X Close-Up View of Bendix Drag Dynamometer Showing Wear Debris Collector Photomicrograph of Cross Section of Friction Material Test Specimen after More than 10 Successive Stops from 100 mph on Bendix1 Full-Brake Dynamometer Changes Occurring in the Friction Material as the Temperature Rises Changes Occurring in the Rotor as the Temperature Rises Mechanism of Explosive Thermal Wear Mechanism of Two-Body Abrasive Wear Mechanism of Three-Body Abrasive Wear Scanning Electron Photomicrograph Showing Abrasive Wear Tracks on Friction Material Surface and Adherent Wear Debris Particles Mechanism of Adhesive Wear Mechanism of Fatigue Wear Mechanism of Macroshear Wear Predominant Wear Mechanisms of Friction Materials Variation of Friction Material Wear with Temperature Vehicle Instrumentation for Friction Material Testing Schematic Diagram for Air Flow of Typical Particulate and Gas Emission Collection System Air Pump Installation in Engine Compartment Vehicle Instrumentation (Front Seat Area) Vehicle Instrumentation (Rear Seat Area) Electrical Schematic for Particulate Trap and Transfer Line Heaters Condensable-Gas Trap Activated-Charcoal Gas Trap - Sectional View Page 2-6 2-6 2-7 2-10 2-10 2-11 2-11 2-12 2-12 2-15 2-17 2-17 2-18 2-18 2-20 2-20 2-22 2-22 4-3 5-3 5-4 5-6 5-7 5-8 5-10 5-11 xiii HWBUI0002298 Figure No. 5-8 5-9 5-10 5-11 5-12 5-13(a) 5-13(b) 5-14 5-15 5-16 5-17 5-18 5-19 5-20 5-21 5-22 5-23 5-24 5-25 5-26 5-27 5-28 5-29 5-30 5-31 5-32 5- 33 6- 1 6-2 6-3 6-4 Title Activated-Charcoal Gas Trap - Top View Activated-Charcoal Gas Trap Exploded View of Activated-Charcoal Gas Trap Gas Collection System Sectional Viey of Disc Brake Shroud Major Portions of Disc Brake Emission Collector Assembly Major Portions of Disc Brake Emission Collector Assembly Disc Brake Emission Collector Assembly - Exploded View Front Disc Brake Emission Collector Installed on Vehicle Back View of Front Disc Brake Emissions Collector Showing Instrumentation Initial Design (Mkl) for Rotating Seal of Disc Brake Collector Rotating Seal for Disc Brake Emission Collector Details of Disc Brake Collector Rotating Seal Front Disc Brake Emissions Collector Showing Thermal Control Aids Right Front Wheel of Test Vehicle First Seal Design (Mkl) for Drum Brake) Mkll Rotating Seal for Drum Brake Emission Collector Details of Mkll Rotating Seal Rear Drum Brake Emission Collector Rear View'of Rear Drum Brake Emissions Collector Showing Instrumentation Completely Worn Seal from MKII Rear Drum Brake Collector Sectional View of Drum Brake Shroud (MKIII) Outboard View of Rear Drum Brake Emissions Collector Interior Assembled Rear Drum Brake Emissions Collector Inboard View of Assembled Rear Drum Brake Emissions Collector Emission Collection System for Clutch General View of Covered Clutch Disc Brake and Emissions Collector Assembly Installed on Dynamometer Disc Brake Comparative Thermal Response Time (Static Air) Comparative Thermal Response Tests on a Disc Brake Seal Leakage versus Flow for Disc Brake Emissions Collector Showing Differences between Dry and Lubricated Seals Page 5-11 5-12 5-13 5-15 5-16 5-17 5-18 5-19 5-20 5-21 5-22 5-23 5-24 5-25 5-26 5-27 5-28 5-29 5-30 5-32 5-32 5-33 5-34 5-35 5-36 5-37 5-38 6-3 6-4 6-4 6-5 xiv HWBUI0002299 Figure No, 6-5 6-6 6-7 6-8 8-1 8-2 8-3 8-4 8-5 8-6 8-7 8-8 8-9 8-10 8-11 8-12 8-13 8-14 8-15 8-16 8-17 8-18 8- 19 9- 1 9-2 9-3 9-4 10-1 10-2 Title Seal Leakage Rate versus Speed Showing Differences between New and Used Seal Seals from Disc Brake Emissions Collector Showing Normal and Abnormal Wear Rotating Seal Leakage Rate versus Wheel Speed for Disc Brake Emissions Collector Disc Brake Emission Collector Temperature Thermal Analysis of Wear Debris Flow Chart for Particulates Analyses Representative Sampling for Analysis Wear Debris Collected on 8y Nuclepore Filter Low Temperature Ashing SEM Photomicrographs of LTA Wear Debris Samples Uniform Sample Distribution onto Filter Phase Contrast Optical Microscopy at 400X Phase-Contrast Photomicrograph of Wear Debris - 400X Transmission Electron Microscope Asbestos Fiber in Brake Emission Microscopy Count Sheet Fiber Volume Calculations Computer Printout of Asbestos Analytical Results Variation of Maximum Deviation of Asbestos Analytical Results with Asbestos Percent Found Gas Handling System Gas Handling System and Gas Chromatograph Detector Response for Carbon Dioxide Typical Calibration Chromatogram for Low Molecular Weight Gas Brake Temperature Data - Vehicle Test 1, 10-Stop Fade Brake Temperature Data - Vehicle Test 1, 15-Stop Fade Decrease in Clutch Airborne Asbestos Emissions with Continued Normal-Duty Service Variation of Asbestos Emissions During Moderate Duty with Asbestos Content of Disc Pads Frequency Distribution of Braking Decelerations Variation of Wear Debris Retention with Mileage (for Rear Drum Brakes) Page 6-5 6-6 6-8 6-9 8-5 8-9 8-10 8-10 8-11 8-13 8-14 8-17 8-17 8-18 8-20 8-21 8-22 8-23 8-30 8-31 8-32 8-34 8-35 9-13 9-14 9-39 9-42 10-11 10-22 xv HWBUI0002300 Table No. 2-1 2-2 3-1 3-2 3-3 4-1 6-1 7-1 7-2 8-1 8-2 8-3 8-4 8-5 8-6 8-7 9-1 9-2 9-3 9-4 9-5 9-6 9-7 9-8 9-9 9-10 9-11 9-12 9-13 LIST OF TABLES Title Composition of Typical Automotive Friction Materials (Volume Percent) Wear Mechanisms in Friction Materials Vehicle Configurations and Lining Selections Surface Roughness Measurements on Turned Discs and Drums for Vehicle Test 3 Estimated Asbestos Content Test Vehicle Description Summary of Dynamometer Test Results Detroit Traffic Test Route (Modified) Vehicle Test Schedule Comparison of Disc Brake versus Drum Brake Operating Conditions Samples Produced by Brake Emissions Collection Estimated Chemical Composition Characteristics of Wear Debris Transmission Electron Microscopy Methods Sample Calculation Possible Sources of Error Extent of Error in Asbestos Emissions Analytical Results Actual versus Estimated Test Schedules for Vehicle Test 1 Actual versus Estimated Test Schedules for Vehicle Test 2 Actual versus Estimated Test Schedules for Vehicle Test 3 Summary of Vehicle Mileage Accumulations Running Log of the Rotating Seals Mileages Obtained in the Vehicle Test Program Rotating Seal Life (Miles) Brake Temperatures During Burnish Tests (F) (Vehicle Test 1) Brake Temperatures During Detroit Traffic Circuits (Vehicle Test 2 - A.B. Baseline) Weights of Brake Particulate Wear Debris Collected (Vehicle Test 1) (gm) Weights of Particulate Wear Debris Collected (Vehicle Test 2) (gm) Weights of Brake Particulate Wear Debris Collected (Vehicle Test 3) (gm) Brake Friction Material Initial Weights and Weight Losses (gm) Material Percent Recovery as Particulate Emissions (Vehicle Test 1) Page 2-4 2-14 3-2 3-3 3-4 4-1 6-10 7-3 7-6 8-2 8-2 8-6 8-8 8-24 8-27 8-29 9-2 9-3 9-4 9-6 9-7 9-9 9-10 9-11 9-16 9-17 9-18 9-20 9-21 xvii HWBUI0002301 Table No. 9-14 9-15 9-16 9-17 9-18 9-19 9-20 9-21 9-22 9-23 9-24 9-25 9-26 9-27 9-28 9-29 9-30 9-31 9-32 9-33 9-34 9-35 9-36 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 Title Material Percent Recovery as Particulate Emissions (Vehicle Test 2) Material Percent Recovery as Particulate Emissions (Vehicle Test 3) Summary of Asbestos Analytical Restuls (Vehicle Test 1) (Weight Percent) Summary of Asbestos Analytical Results (Vehicle Test 2) (Weight Percent) Summary of Asbestos Analytical Results (Vehicle Test 3) (Weight Percent) Comparison of Analytical Results for Asbestos Weight of Asbestos Generated During Braking (Vehicle Test 1) (mg) Weight of Asbestos Generated During Braking (Vehicle Test 2) (mg) Weight of Asbestos Generated During Braking (Vehicle Test 3) (mg) Weight of Asbestos Generated During Braking (Vehicle Test 1) (yg/mile) Weight of Asbestos Generated During Braking (Vehicle Test 2) (yg/mile) Weight of Asbestos Generated During Braking (Vehicle Test 3) (yg/mile) Summary of Clutch Asbestos Emissions Comparison of Asbestos Emission from New and Burnished Materials Asbestos Emissions from Heavy-Duty Tests Comparison of Asbestos Emissions Generated by Different Friction Materials Comparison of Asbestos Emissions for Various Sample Locations (yg/mile) Summary Comparison of Surfaces, Sump, and Airborne Samples Change in Lining Wear Rate after Operation at Increased Temperatures (Vehicle Test 2) Data Analysis of Gaseous Emissions (Vehicle Test 1) Data from Analysis of Gaseous Emissions (Vehicle Test 2) Elution Order from Poropak Q Elution Order from W-98 U. S. Braked Motor Vehicle Population U. S. Braked Trailer Population U. S. Motor Vehicles with Dry Clutches Yearly Additions to U. S. Braked Vehicle Population Motor Vehicle Usage by Vehicle Type and Roadway Type and Usage of Public Roads Motor Vehicle Usage by Purpose of Trip Motor Vehicle Usage by Length of Trip Page 9-22 9-23 9-25 9-26 9-27 9-29 9-31 9-32 9-33 9-34 9-35 9-36 9-38 9-38 9-40 9-42 9-44 9-45 9-47 9-49 9-50 9-51 9-52 10-2 10-2 10-3 10-3 10-5 10-5 10-6 10-6 xviii HWBUI0002302 Table No. 10-9 10-10 10-11 10-12 10-13 10-14 10-15 10-16 10-17 10-18 10-19 10-20 10-21 10-22 10-23 10-24 10-25 10-26 10-27 10-28 10-29 Title Results of Cross-Country Brake Test Trip Where Vehicles Are Used Some U. S. Driving Characteristics Brake Applications Per Mile Asbestos Content of Automotive Friction Materials Brake Friction Material Weights for Test Vehicle Number of Newly Surfaced Rotors Used Each Year Compilation of Annual Motor Vehicle Brake Friction Material Usage Compilation of Annual Clutch Friction Material Usage Relative Brake Relining Frequency Total Asbestos Emissions Calculated from the Test Vehicle (Average for the Three Vehicle Tests) Distribution-Weighted Asbestos Emissions from the Test Vehicle Estimate for Total Asbestos Emissions Expected for a Light Truck Estimate of Total Asbestos Emissions Expected for a Medium Truck Estimate of Total Asbestos Emissions Expected for a Heavy Truck Distribution-Weighted Asbestos Emissions from a Light Truck Summary of All Brake and Clutch Emissions (lbs per year) Calculation of Weighted Average Unconverted Asbestos Percent for Sump Sample Calculation of Weighted Average Asbestos Percent for Surfaces Sample Calculation of Weighted Average Asbestos Percent for Airborne Samples Data and Calculation of Weighted Asbestos Percent for All Vehicles. Page 10-8 10-8 10-10 10-10 10-11 10-12 10-12 10-13 10-14 10-14 10-26 10-27 10-29 10-30 10-30 10-31 10-31 10-34 10-34 10-36 10-36 xix HWBUI0002303 SECTION 1 INTRODUCTION AND BRIEF SUMMARY The chrysotile form of asbestos is a major constituent in automo tive brake linings and disc pads, and in the friction materials used in clutches. These components are expendable. In typical passenger car service, drum linings wear out in about 40,000 miles, disc pads in about 30,000 miles, and clutch facings in about 70,000 miles. There are, of course, wide variations in the service lives of these components, which depend on the nature of the friction materials used, on driver habits, and on driving conditions. The expendable members of the common automotive friction couples contain not only asbestos, but also various organic and inorganic fric tion and property modifiers. These constituents are bound together by phenolic resins. Because of their high content of organic substances, such friction materials are classified as "organic." Asbestos-free friction materials with a high content of metallic substances are classi fied as "semi-metallic." Over 99 percent of the automotive friction ma terials used today are organics. The mechanisms by which organic friction materials wear can be classified under five headings: thermal, abrasive, adhesive, macroshear, and fatigue. Thermal, abrasive, and adhesive wear are considered to be the most important mechanisms, thermal wear predominating above 450F, abrasive and adhesive wear predominating below 450F. (the temperature being measured by a thermocouple in the rotor, about 0.050 inch from the sliding interface). The organic constituents pyrolyze or oxidize, and are emitted to the air as carbonaceous solid particles or gaseous reaction products. The asbestos fibers are pulverized into small parti cles which are trapped in the brake (or clutch) housing, or fall to the road, or are emitted to the air. Disc brakes do not trap as much as drum brakes. Most of the asbestos is heated to temperatures high enough to cause chemical conversion, and is therefore trapped or emitted as olivine or forsterite particles. When the large number of existing cars and trucks is considered, each with eight or more pieces of friction material in the brake system, plus more in the clutch mechanism, it becomes apparent that an air pollu tion problem may exist. Considered in another light, there are perhaps one billion pieces of friction material in vehicles in the United States, gradually being pyrolyzed and ground to dust. The brake systems in the vehicles utilizing these friction materials can thus be considered as chemical reactors, each emitting organic and inorganic compounds, includ ing asbestos and its decomposition products, to the atmosphere. The com positions of the gaseous and solid emissions have not been well known. 1-1 HWBUI0002304 and the particle sizes and shapes of the solid emissions have not been well known. Because of the possible damage which these emissions can produce in human respiratory systems, it is desirable to identify and quantify them, with particular regard to asbestos. That is the purpose of the program described in this report. The amount of asbestos emitted is a function of many variables: the number of vehicles on the road, the number of miles driven per year by each, vehicle size, the number of vehicles using manual clutches, conditions of driving (city, expressway, mountain; driver habits, etc.), brake system design, and brake type (drum or disc), rotor condition (surface roughness, scoring, cracking, pitting, etc.), rotor design (e.g., solid or ventilated discs, finned drums, etc.), rotor metallurgy (cast iron or aluminum), volume percent of asbestos in the friction material, and inherent wear resistance of the friction material. In this program a passenger car of medium weight (test weight: 4800 pounds), produced in high volume, was used. It was equipped with a manual transmission and clutch, ventilated cast iron disc brakes on the front, and standard cast iron drum brakes on the rear. A passenger car was selected for testing because of the following factors: (1) vehicle braking cannot be simulated realistically on small-scale friction test-machines with the present state of the art; (2) full-scale brake and clutch dynamometers provide fairly good simulation, and can be used for checking out the emissions collection system; separate tests for brakes and clutches would be required, however, thus increasing program costs, whereas on the vehicle, disc and drum brakes and clutches can be tested simultaneously; (3) air flow in the dynamometers differs from that in vehicles, and emissions from drum brakes will therefore be different in the two cases. 1-2 HWBUI0002305 Three series of tests were run, using friction materials as tabulated below: Vehicle Test Simulation Friction Materials Remarks 1 Original OEM1 Disc Front Class A organic disc pads Equipment OEM Drum Rear 2 Partial AM2 Disc Front No metallic chips; Reline Used OEM Drum Discs and drums not Rears turned 3 Full AM Disc Front Disc pads and secondaries Reline AM Drum Rear contain brass chips Discs and drums turned This selection has several features. It provides a replication of one test, thus giving an estimate of repeatability. It includes a primary/ secondary combination that serves both OEM and aftermarkets. It includes one combination (two materials) of widely used OEM materials. It includes one combination (two materials) from one of the larger aftermarket sup pliers. It includes a set of used linings (about 3,000 miles) to give data on the emissions at various stages of lining life. It includes two different disc pad materials from two of the larger aftermarket suppliers. Five different manufacturers were represented in the friction material selections. The collection of particulate and gaseous emissions from any vehicle friction couple has never before, to the authors' knowledge,been reported in the literature. To define the extent of these emissions, a combination separation and storage collection system was devised. Unique emissions collectors for both disc and drum brakes, and for a clutch, were conceived, designed, and built as the main embodiment of this instrumentation. The collectors separated the wear debris into three different fractions: a sump sample which included the wear debris on the lining surfaces, in the rivet holes, and on the brake drum; a surfaces sample which included the wear debris on the brake and collector shroud surfaces; and an airborne sample collected on membrane filters. The vehicle was equipped with standard brake test instrumentation. A front disc brake collector and a rear drum brake collector were installed on the right side of the vehicle. The clutch was sealed by closing the J0EM - Original Equipment Manufacturers 2AM - aftermarket 1-3 HWBUI0002306 holes in its casing. The left wheel brakes were left in their normal configuration and were used to monitor the operation of the shrouded brakes. Wear debris was also taken from the left brakes; the amounts of debris collected and its composition were used to demonstrate that the brake shrouding did not significantly affect the operation of the brakes within the emissions collectors. Although there are many acceptable original equipment manufacturers' brake and friction material tests for measuring a specific collection or combination of collections, no one test adequately matches driving con ditions which would be representative of all vehicles. Consequently, it became necessary to devise a rational and meaningful vehicle test. A total of seven schedules were chosen and each was followed by a measure ments procedure (take emissions samples, measure wear, inspect systems, and replace worn parts). The first three schedules - Burnish, After Burnish (A.B.) Baseline, and Detroit Traffic - Represented low-temperature testing. The final four schedules - 10-Stop Fade, After Fade (A.F.) Base line, 15-Stop Fade, and Final Baseline - represented high-temperature and abrusive testing. The mileage accumulations for each schedule of each vehicle test verified the possible test mileage reproducibility. A critical component of the front disc brake collector was a unique rotating seal. For the 8100 miles of testing, only eight rotating seals were required -- an ex ceptional improvement over the anticipated need at the start of the program. Prior to vehicle test 1, a Burnish schedule was performed and the temperatures for the right front disc brake were found to be near 400F. The temperature difference, AT, between the right and left front brakes ranged from 200 to 250F. This was unacceptable. The addition of several thermal control aids reduced the AT to a range of 115 to 160F. Most import ant was the reduction of the right front brake operating temperature from about 400F to about 300F. The brake shrouding did not present any pro blems on hot days. During the Detroit Traffic Test, the AT between the front brakes remained between 100 and 140F while that for the rear drum brake remained between 10 and 30F. The range of asbestos content in the brake emissions for vehicle tests 1 and 2 went from a high of 1.65 percent to a low of 0.03 percent for the 90 analyses; only three were above 1.00 percent. The overall average for vehicle test 1 was 0.40 percent and for vehicle test 2 was 0.25 percent. For vehicle test 3, the range of asbestos content was from a high of 0.51 percent to a low of 0.003 percent. The overall aver age was 0.07 percent. In this work, all values for heavy-duty (abusive) braking were less than one percent, which is significantly less than the 15 percent asbestos content reported by Lynch of the U.S. Public Health Service. To provide a systematic, independent check on the asbestos analyses, the EPA Project Officer arranged a second contract for Battelle Columbus Laboratories to analyze twenty-four samples generated by Bendix Research 1-4 HWBUI0002307 Laboratories. Johns-Manville also provided three analyses. The JohnsManville and Battelle data agreed very well. In 19 of the 24 analyses, the Bendix results were slightly higher than the corresponding Battelle results. The Bendix average was 0.25 percent, while the Battelle aver age was 0.17 percent, or 68 percent of the Bendix average. The Bendix data were expected to be higher for the following reasons: all fibers found were assumed to be cylinders of projected diameter and length, and a fiber with an elliptical cross section was actually smaller in volume than reported; and all fibers were assumed to be pure asbestos, (partially degraded asbestos or olivine fibers were mechanically reduced to nonfibrous material by the other analytical techniques, so only the stronger asbestos fibrils that remained were counted). Analysis of the gaseous emissions was made difficult by the large quantities of water and carbon dioxide which by-passed the gas condition ing trap and entered through the rotating seals. Most samples showed traces of to C3 hydrocarbons only as lower concentrations of higher molecular weight gases could not be detected. When the gases were con centrated from the fade tests, C4 to Cg compounds such as butanes, ben zene, toluene, phenol, and cresols were detected in the gas traps. The corresponding shroud extracts showed 0^2 to C30 materials, including cardanol. To compare the generation of asbestos emissions properly, it was found necessary to calculate the emissions factors for each sample in micrograms/mile (yg/mile). These values were obtained by multiplying the weight of generated brake debris by the appropriate asbestos percent, followed by dividing by the miles per test schedule. Thus, either a high asbestos content or a large sample gave a high factor. The following observations were made: Asbestos emissions were higher for new friction surfaces and decreased with use. Initially the drum brake produced more asbestos emissions than the disc brake. The difference decreased as the materials continued to be used. Heavy (abusive) duty did not necessarily give a higher percent asbestos, however, the large amount of debris produced gave a significant rise in asbestos emissions. The order of decreasing asbestos emissions from brakes was found to be: Fade or heavy-duty stops Decreased Burnish asbestos Moderate braking emissions i.1 1-5 HWBUI0002308 o For the disc pads only, there was an increase in asbestos emissions with increased asbestos content in the friction material (there was no such trend for the drum brake materials). Both the front disc pads and the drum linings of vehicle test 3 had wear comparable to that of the other two vehicle tests, yet the asbestos emissions were significantly lower. Both pads and linings contained brass chips which may have been in part responsible for the more complete conversion of the asbestos in the brake emissions, possibly because of better heat transfer to the asbestos. For both disc and drum brakes, the surfaces sample was the largest of the three (^92 percent), the sump sample was next (^7 percent), and the airborne sample was the smallest (M. percent). Two series of weighting factors were developed. The distribution weighting factors were calculated from the material recoveries in the shrouded and unshrouded brakes as indicated earlier. Estimates were made for both disc and drum brakes from the test vehicle for the amounts of the different samples distributed from the brakes. These estimates were made for the Burnish schedule, for the end of a complete vehicle test, and for the end of all three vehicle tests. For the drum brake, estimates were made up to 40,000 miles. The severity weighting factors were cal culated from the percentages of different braking modes: for new versus burnished friction materials, for severe (or abusive) versus moderate braking, and for the renewal of friction surfaces (A.F. Baseline). The total asbestos emissions estimate for the test vehicle was calculated from the asbestos emissions factors and the severity weighting factors. The average asbestos emissions factor for the disc brakes, drum brakes, and clutch were used along with the Burnish (new surfaces), A. B. Baseline (surface preparation), Detroit Traffic (moderate braking), 10Stop Fade (severe braking) and A. F. Baseline (surface renewal) weighting factors. The value obtained for the test vehicle was 28.51 yg/mile. The fate of the asbestos emissions was calculated with the use of the distribution weighting factors as follows: Road Drop-out 81.9% Airborne: 3.7% Brake Retention: 14.4% The annual total asbestos emissions estimate for 96,400,000 pass enger vehicles travelling an average of 9,978 miles per year was found to be 60,400 pounds, distributed as follows: Road Drop-out: 49,470 lbs Airborne: 2,230 lbs Brake Retention: 8,700 lbs 60,400 lbs 1-6 HWBUI0002309 The annual total asbestos emissions estimates for light trucks, medium trucks (and buses), and heavy trucks were found to be: Light trucks 32,300 lbs Medium trucks (and buses) 16,300 lbs Heavy trucks 22,900 lbs 81,500 lbs The distribution of the total asbestos emissions estimate for trucks was calculated to be as follows: Road Drop-Out: 87.9% Airborne: 2.9% Brake Retention: 9.2% The estimated asbestos emissions for all vehicles in pounds/year is summarized as follows: Total Asbestos Emissions (lbs) Distribution (lbs) Drop-Out Airborne Retention Passenger Cars Light Trucks Medium Trucks (and buses) Heavy Trucks Miscellaneous (motorcycles, trailers, etc.) 60,400 32,300 16,300 32,900 16,300 49,470 28,420 14,330 28,920 2,230 940 470 950 14,330 470 8,700 2,940 1,500 3,030 1,500 158,200 135,470 5,060 17,670 Distribution by percent: 85.6% 3.2% 11.2% These estimates are considered to be maximum values. 1-7 HWBUI0002310 SECTION 2 OBJECTIVES AND BACKGROUND 2.1 OBJECTIVES 2.1.1 Objectives of this Program The objectives of this program, as stated in the Request for Proposal and in the Contract, were as follows: "In order to define the extent of non-engine asbestos emissions, the Office of Air Programs of the Environmental Protection Agency is seeking the assessment of these emis sions on a single vehicle basis. The efforts are to include emissions measurements from several different types of brakes as well as a clutch assembly. The end objective of the work will be the documentation of these emissions to assess their overall contribution to the atmosphere." 2.1.2 Bendix Technical Approach 2.1.2.1 Select Suitable Vehicle and Friction Materials For the results to be representative of the average consumer-type vehicle usage, it was necessary to select a vehicle inter mediate between the compacts and luxury vehicles. Moreoever, it was desirable to select a vehicle with original-equipment friction materials, produced by more than one supplier, and equipped with a clutch. Aftermarket friction materials, produced by several manufacturers, were also selected. 2.1.2.2 Conceive, Design, and Build Brake and Clutch Emissions Collectors Since only 100-percent collection could give a mass balance, it was necessary to conceive, design, and build brake and clutch emissions collectors which could collect all the emissions and differen tiate among wear debris remaining on the friction materials, debris re maining on the local surfaces of the foundation brakes, and debris entering the atmosphere. Moreoever, provision was necessary for the collection and storage of gaseous decomposition products. 2.1.2.3 Test a Typical Collector and Instrument the Vehicle To ensure that the concepts and hardware were technically sound, it was necessary to install the most critical collector on an inertial dynamometer to verify its operation. Once proven, that collector and the others, along with brake test instrumentation, could be installed on the vehicle. 2-1 HWBUI0002311 2.1.2.4 Select and Run Suitable Driving Test Schedules Because of the wide variety of driving which takes place throughout the various regions of the country, it was necessary to select a vehicle test program in which brake emissions could be collected for new brake linings and rotor combinations. The program included typical consumer-type driving on flat terrain, and heavier-duty braking as en countered in hilly areas. 2.1.2.5 Collect and Analyze Emissions As the main goals were to determine the asbestos content of the emissions and their particle-size distributions, it was necessary to select an analytical method which would not disturb the particle size distribution and yet be sensitive to very small amounts of asbestos. Further, it was necessary to collect and separate very low concentra tions of gaseous emissions from extremely large quantities of air and other molecules present in large concentrations. 2.1.2.6 Estimate Asbestos Emissions for Cars and Trucks Because of funding and time limitations, only a few selected tests could be run. Based on the measurements taken and a knowledge of the overall vehicle population, it was possible to prepare estimates for total emissions from cars and trucks operating in the United States. 2.2 PRIOR WORK 2.2.1 Published Data The only known published data on asbestos emissions were those reported by Jermiah Lynch of the U.S. Public Health Service, Department of Health, Education, and Welfare.(2-1)* The paper discusses tests which were performed with brake test machines, using one-inch-square samples, and tests performed on an inertial dynamometer where a complete brake and clutch were used. Emis sions samples were collected on membrane filters and analyzed by electron microscopy. In all but a few tests, the drum brake linings showed less than 1 percent free fiber in the decomposition product. In some tests reported, where the lining temperatures were extremely high and accompanied by a rapid drop in coefficient of friction (fade), a significant mass of free fiber was released - approximately 10 to 15 percent. The clutch test gave no asbestos fiber. Numbers in parentheses designate References at end of section. 2-2 HWBUI0002312 2.2.2 Unpublished Data At the time of writing of this report, there were three known works which had been completed or were in progress. The amount of brake wear debris and its composition were determined on a full-brake inertial dynamometer using a duo-servo drum brake in 1968-69.(2-2) The determination of the amount of asbestos in airborne brake debris sampled downwind of a disc brake installed on an inertial dynamometer was in progress. (2-3) The determination of asbestos content in airborne brake debris sampled downwind of vehicle-mounted disc and drum brakes was also in progress. (2-4) 2.3 BACKGROUND 2.3.1 Energy Absorption by Friction Materials 2.3.1.1 Brakes During a stop, the automobile brake converts the kinetic energy of the moving vehicle into heat, absorbs the heat, and gradually dissipates it to the atmosphere. The sliding friction couple in the brake consists of a cast-iron rotor (drum or disc) and a stator on which is mounted a friction material or lining consisting of asbestos fiber, property modifiers, and an organic resin binder. Lining is considered expendable. Over a long period of time and many miles of service, much of it is converted to dust and gases. Smaller amounts of the rotor are also converted to wear debris. Particulate debris is partially retained and partially emitted from the vehicle in normal service. 2.3.1.2 Clutches The "manual" automobile clutch transfers the kinetic energy of the engines rotating crankshaft smoothly to the transmission and wheels. Any slippage results in generation of heat, which is ab sorbed and eventually dissipated to the atmosphere by the clutch. Thus the clutch is basically a dry-running static friction couple (which slides momentarily during gear shifts). The mating materials are usually cast-iron or steel and a friction material or lining which is frequently organic-bound. Organic-bound friction materials consist of asbestos fiber reinforcement, property modifiers, and an organic resin binder. The composition is qualitatively similar to that of brake linings. Clutch linings are also expendable, gradually wearing away in use. The wear debris can escape from manual clutches. Wear debris from friction materials in automatic transmissions is retained almost completely in the transmission fluid. 2-3 HWBUI0002313 2.3.2 Compositions of Friction Materials 2.3.2.1 Generalized Classifications Table 2-1 gives typical compositions of drum lining, disc pad,and clutch facing friction materials. The foundation, or major constituent, of practically all organic friction materials is asbestos fiber. Asbestos is chosen because of its thermal stability, its relatively high friction level, and its reinforcing properties. Since asbestos alone does not offer all of the desired friction properties, other materials called friction modifiers are added. Modifiers are varied in type and content to provide desired levels of effectiveness, wear, fade, recovery, and noise. A resin binder is also added to hold the other materials together with adequate strength. 2.3.2.2 Asbestos The asbestos usually used in friction materials is chrysotile from Quebec or Vermont. Chrysotile, the principal mineral of the serpentine group, has the approximate composition formula Mg3(Si205)(0H)4 or 3Mg0.2Si02-2H20 Table 2-1 - Composition of Typical Automotive Friction Materials (Volume Percent) Resin Binder Asbestos Reinforcement Friction Modifiers Organic Friction Dust Elastomer Inorganic Carbons/Graphites Zinc Chips Oxides Copper/Brass Misc. Inorganics Sulfur Drum Linings Primary Secondary 20-30 Major 25-35 Major 10-15 5-15 5-15 0-5 0-5 5-10 Disc Pads Class A Standard Class B Heavy-Duty 18-22 15-18 Major Major 20-25 10-15 1-2 1-2 0-5 0-15 0-10 2-15 2-7 5-10 5-15 Woven Clutch Facings Standard Heavy-Duty 20-30 Major 15-25 Major 0-5 0-5 5-10 2-5 0-5 5-10 5-10 2-5 2-4 HWBUI0002314 Structurally, it is a pseudohexagonal network of SiC>4 tetrahedra forming a sheet in which all the tetrahedra point one way. A layer of brucite, Mg(0H)2, is joined to the Si04 network in such a way that, on one side, two out of every three hydroxyls are replaced by oxygens at the apices of the tetrahedra (Figure 2-1). The macromolecule consists of parallel sheets of brucite-silica layers structured in cylindrical lattices in volving closed concentric cylinders, spirals, and sometimes helical arrangements.(2-6) Bendix Research Laboratories' scientists have been study ing asbestos for several years. Figure 2-2 presents the results of dif ferential thermal analysis (DTA) and thermogravimetric analysis (TGA) studies for a common short-fiber grade of chrysotile asbestos.(2-7) The transition at approximately 700F (370C), in both the DTA and the TGA thermograms represents the dehydration of the brucite (magnesium hydrox ide) , the 1.4 percent weight loss corresponding to an approximate con centration of 4.5 percent brucite. The transition at 1250F (680C) in both thermograms represents the major dehydroxylation reaction in chrys otile asbestos. At 1580F (860C), a sharp exotherm, not associated with any weight change, represents the conversion of the dehydroxylated chrys otile to forsterite (2Mg0*Si02) or olivine [2 (Mgl. 0O-xFex^' si^2 > where x is approximately 0.02 to 0.16], TGA results for the asbestos were identical in air and in nitrogen. Figure 2-3 shows the X-ray diffraction patterns for asbes tos heated to different temperatures.(2-8) The crystal structure changes from chrysotile to forsterite (olivine) between 600 and 800C. Asbestos fibers are present in greater concentration (by volume percent) than any other ingredient in friction materials. As a consequence, both the relative concentration and the type of asbestos used (with respect to fiber length and "openness" or absorptiveness) significantly affect the amount of resin binder required and the process ing and performance characteristics. 2.3.2.3 Resin Binders Synthetic resins, such as phenolics and modified phenolics, are commonly used as friction material binders. They are prepared as the condensation product between the appropriate phenol (sometimes modi fied) and formaldehyde in the presence of an acidic catalyst to yield the novolak. When mixed with an appropriate curing agent, they poly merize at elevated temperatures to an insoluble, infusible mass. Other resin systems in wide use are based on elastomers, drying oils, or combinations. 2.3.2.4 Property Modifiers Friction modifiers can in general, be divided into two classes: non-abrasive and abrasive.(2-9) Organic materials are generally less abrasive than inorganic as described below. 2-5 HWBUI0002315  OXYGEN O MAGNESIUM SILICON SHEETS ARE CONTINUOUS IN DIRECTION OF CURVATURE AND NORMAL TO PLANE OF SECTION; FIBER AXIS IS ALSO NORMAL TO PLANE OF SECTION. Figure 2-1 - Schematic Structure of Chrysotile Asbestos 200 400 600 800 1000 1200 1400 1600 1800 F TEMPERATURE Figure 2-2 DTA and TGA Thermograms for Chrysotile Asbestos (Bendix Data) 2-6 HWBUI0002316 i (i 1 S'* r| m 8uu^ c Au -j A. i * Tfl It * rt FORSTERITE 1000 c 11 \-nra * y X D tofu CO <T) 1200 c CoNpI CL. 104 96 88 80 72 64 56 48 40 32 29 ANGLE (DEGREES) 24 16 8 Figure 2-3 - X-Ray Diffraction Pattern of Heated Chrysotile Asbestos (Bendix Data) 2-7 HWBUI0002317 2.3.2.4.1 Non-Abrasive Modifiers Nonabrasive friction modifiers can be classified further as low friction and high friction. The most common and best known of the high-friction materials is known as friction dust. This is a cured resi nous material. The most frequently used variety is derived from cured or polymerized cashew-nut-shell liquid, chemically a phenolic compound. When heated with hardening agents, such as hexamethylenetetramine or formaldehyde, it becomes sufficiently hard or polymerized to be granu lated. Many other cured resinous or polymeric materials, some with fillers, are also used. Certain friction dusts are combinations of these materials and cashew resin. Ground rubber is normally used in particle sizes similar to, or slightly coarser than, those of the cashew friction dusts for noise, wear, and abrasion control. Carbon black, graphite, petroleum coke flour, or other carbonaceous materials may also be added as friction modifiers to lower the friction coefficient or to reduce noise. These materials are normally used in the form of fine powders or particles, although graphite is some times used in coarse particles or pellets. The amount of friction modi fier added is dependent upon the properties desired in the final composite. 2.3.2.4.2 Abrasive Modifiers Abrasive modifiers, such as alumina and the silicas, are usually used in relatively small amounts and only in very fine particle sizes (generally 100 mesh or finer). Particle size is limited by the fact that large particles of such hard materials would groove and wear the mating surfaces. Minerals are generally added to improve wear resis tance at minimum cost. Those most commonly used are ground limestone (whiting) and barytes (barium sulphate), though various types of clay, finely divided silicas, and other inexpensive or abundant inorganic powders may also perform this function. Such materials are inorganic in nature and tend to detract from noise properties and mating surface compatibility. Metals or metal oxides may also be added to perform spe cific functions. Brass chips are frequently found in heavy-duty friction materials where, as scavengers, they break up undesirable surface films. Zinc and aluminum are also used. Zinc chips, in relatively small amounts, can contribute significantly to recovery of normal performance following fade. 2.3.3 Friction Material Reactions 2.3.3.1 Absorption and Conversion of Energy The automobile brake converts the kinetic energy of the moving vehicle into heat, absorbs the heat, and eventually dissipates it to the atmosphere. The clutch absorbs the friction energy generated when the transmission and drive-shaft are engaged or disengaged on manual 2-8 HWBUI0002318 transmission vehicles. Phenolic-bound brake lining compositions contain ing asbestos serve as one of the best classes of low-cost expendable members of the friction couple for all-around performance. At low energy absorption rates, the temperature of the friction material, approximately 40 mils from the interface, may rise to 300F; the interface itself is hotter. Most of the heat is removed from the interface by convection and conduction.(2"10) At high energy absorption rates, heat is generated faster than it can be dissipated and the temperature at the sliding inter face reaches a point where chemical reactions occur rapidly. Surface temperatures up to 1600F have been observed or calculated; the presence of olivine in the wear debris confirms that such temperatures have been reached. 2.3.3.2 Physical and Chemical Changes in Linings During Use Figure 2-4 illustrates the heterogenous nature of a typical friction material composed of the following ingredients: phenolic binder (30 volume percent), asbestos fiber (55 volume percent), and cashew, friction modifier (15 volume percent). As a result of the energy conversion at the stator-rotor interface, the composition of the brake lining surface is altered. At low service temperatures, the phenolic resin and organic friction modi fier convert, on the surface, to compositions with slightly higher carbon/ oxygen ratios. At somewhat higher use temperatures, the resin and or ganic friction modifier convert to carbon more readily, and this in turn oxidizes to carbon dioxide. At the same time, the asbestos wears away more rapidly because of the reduced binder strength and volume at the surface. Figure 2-5 is a photomicrograph of typical wear debris collected from Bendix Research Laboratories' dynamometer, operated at 600F. Asbestos fibers are prevalent in several of the wear debris particles. Figure 2-6 is a photograph of the improved wear debris collector designed and built for friction and wear research.(2-11) At still higher temperatures, the asbestos converts from its fibrous form to an olivine powder with little or no reinforcing value. Figure 2-7 shows the frictional-heat-affected layer of lining which had experienced more than 10 successive stops from 100 mph without sufficient cooling between stops.(2-12) (a large piece of material is about to separate from this sliding interface - an example of macroshear wear, described later.) When heat is generated at the sliding interface between stator and rotor much faster than it can be dissipated, the temperature rises and may reach a point where the lining or pad components decompose, oxidize, or melt. The stator material then wears rapidly and the fric tion coefficient, p, may decrease. This is fade. Fade has been attri buted by some to evolution of gas at the sliding interface; on heating, the gas expands and exerts a force on the lining or pad, tending to push it away from the rotor.(2-13) Fade can generally be ascribed to the formation of a gaseous, a liquid, or a low-friction solid phase at the interface, or to some combination of these phenomena. Figure 2-8 2-9 HWBUI0002319 Figure 2-4 - Photomicrograph of Brake Lining Cross Section After 8550 Miles of Test-Track Service Figure 2-5 - Electron Micrograph of Friction Material Wear Debris (Made during 1966 studies) - 10,000 X 10 HWBUI0002320 Figure 2-6 - Close-up View of Bendix Drag Dynamometer Showing Wear Debris Collector Sliding Surface Heat Affected Zone Unchanged Material Figure 2-7 - Photomicrograph of Cross Section of Friction Material Test Specimen After More Than 10 Successive Stops from 100 mph on Bendix Full-Brake Dynamometer 2-11 HWBUI0002321 PHYSICAL OR CHEMICAL CHANGE E-66-28-d <U a) Au 4J d +J d cd H u a) 60 04 hd &a) Uu H d a) a .d o u o co CO cd 60 M s 4oJ ,ad O o\ CN cu d 60 rH TEMPERATURE PHYSICAL OR CHEMICAL CHANGE U U) O) N AV 55 AV o< zg O Oui | o < S 5 <Ss> mi 2ls 2<y cc Q Ui V *mw to IO XO 2g 00 to ?< C/5 00 O D Hcuc Wcc MS2 Uz 3wr<f co UO j r: cQ2<o io 9 Xo Da' , XO O E< sio 5 a- o ZO UMJ ON UJ g Z N oo z o z 9 x . ic=o- jy- uj On c/5 m r; -- ui uj go aX. Xa. < 2-12 a) 0) .d .d u d CO H cd 60 r-1 CO d cd a) H H co U U rH U 0) P=S d 4J o cd a) ou od d 4J CO o cd a) H H 60 4J a) dO d rH e< g rC a) o P>H H I. 00 <N a> u d 60 pHLt HWBUI0002322 summarizes the physical and chemical changes which take place in the friction material as the temperature rises.(2-14) Figure 2-9 summarizes the physical and chemical changes which take place in the cast-iron rotor as the temperature rises. 2.3.4 Brake Lining Wear Mechanisms and Generation of Emissions 2.3.4.1 Types of Wear Mechanisms Emissions are generated by wear. To the consumer, wear is an economic consideration. To the environmentalist, these emissions are substances to be characterized and, if harmful, to be eliminated. From the basic friction material engineering standpoint, wear resistance is inversely related to friction level and other desir able performance characteristics. Linings with faster wear rate and more frequent surface renewal generally have greater frictional sta bility. -This circumstance is used to advantage in the development of primary-type linings for duo-servo brakes, where, because of design, the stability of frictional properties is of greater significance to brake performance than is wear resistance; use of primary linings with too much wear resistance results in poor fade resistance, less fric tional stability, and generally poor overall performance. Except in the case of primary linings, however, one of the formulator's objectives must be to achieve the highest possible level of wear resistance. Contrary to the opinion of some, maximum life does not require a friction material of maximum physical hardness. What is desirable is a minimum wear rate in normal low-temperature use, a moderately increased rate at elevated temperatures, and a return to the original low-temperature wear rates after being subjected to ele vated temperatures. Total emissions generation may be defined as the sum total of the losses associated with five basic types of wear.(2-15) abrasive wear, thermal wear, adhesive wear, fatigue wear, and macroshear wear As shown in Table 2-2, there are several possible mechanisms for each type of wear. 2.3.4.1.1 Thermal Wear Thermal wear encompasses a group of physical and chemi cal reactions in the course of which interatomic bonds are broken by ther mal energy. These reactions include pyrolysis (thermal decomposition), 2-13 HWBUI0002323 Table 2-2 - Wear Mechanisms in Friction Materials THERMAL WEAR PYROLYSIS OXIDATION THERMOPARTICULATION MELTING EVAPORATION SUBLIMATION EXPLOSION ADHESIVE WEAR ORGANIC MATERIAL ADHESION BINDER FRICTION DUST RUBBER INORGANIC MATERIAL ADHESION METAL PARTICLES ICE ABRASIVE WEAR TWO-BODY ASPERITIES SCORING CRACK EDGES PIT EDGES THREE-BODY WEAR DEBRIS ROAD DUST FATIGUE WEAR THERMAL CYCLING SHOCK MECHANICAL REPEATED BRAKING DRUMS OUT OF ROUND DISCS OUT OF PARALLEL MACRO-SHEAR WEAR ACROSS THE PAD AT CORNERS AND EDGES THROUGH HEAT-AFFECTED LAYERS oxidation, thermoparticulation, explosion, melting, evaporation, and sublimation. Their rates increase exponentially with temperature, and in extreme cases they result in fade. Pyrolysis probably occurs predominantly at the centers of linings and pads and to a lesser extent at the corners and edges. Oxidation, on the other hand, probably predominates at the corners and edges and is less severe at the center. Explosive reactions occur under highly abusive braking conditions, where the rate of heat input is so high that solids are converted to gases well beneath the surface; because these gases are greater in volume than the solids they displace, they create a pressure that ruptures the lining in an explosive manner. Four such reactions are represented in Figure 2-10. It might be thought that rapid thermal wear occurs only at high soak temperatures.* In fact, however, thermal wear probably k Soak temperatures are temperatures measured during dynamometer or vehicle testing, using a thermocouple located in the friction material or in the rotor, generally at a distance of 20 to 50 mils from the sliding interface. 2-14 HWBUI0002324 0 CaC03(s)-^Ca0(s) + C02(g) * 0 CASHEW DUST (s) ^C02 (g) + HzO (g) + C (s) 0 ASBESTOS->H20 (g) + OLIVINE (s) + ASBESTOS FIBRILS (s) 0 PHENOLIC RESIN ~>C02 (g) + H20 (g) + C (s) *THE BRASS CHIP CONDUCTS HEAT RAPIDLY INTO THE FRICTION-MATERIAL SURFACE. (S): SOLID (G): GAS Figure 2-10 -- Mechanism of Explosive Thermal Wear occurs at appreciable rates at low soak temperatures as well. It has been shown that the materials in a sliding friction couple contact each other at the tops of asperities on their surfaces. There is consider able evidence to indicate that the temperature increase at these asperi ties for low sliding speeds is represented by the equation^> 2-17) T - Tq = (gyvw/4aJ)[l/(k1 + k2)] where T is the "flash" temperature Tq is the soak temperature g is the acceleration due to gravity p is the coefficient of friction v is the sliding velocity w is the load on the asperity 2-15 HWBUI0002325 a is the radius of the asperity junction J is the mechanical equivalent of heat is the thermal conductivity of the disc is the thermal conductivity of the slider Modifications of this equation have also been developed for high sliding speeds. Experiments indicate that the flash temperature may be as high as 1400F (760C) if the soak temperature is 650F (343C). This flash temperature is high enough to cause rapid localized pyrolysis of the organic compounds used in friction materials, as well as conversion of fibrous asbestos to powdery olivine and transformation of pearlite to martensite in the cast-iron mating surface. 2.3.4.1.2 Abrasive Wear Abrasive wear is of two types: two-body wear (Fig ure 2-11) and three-body wear (Figure 2-12). Plowing by asperities on the rotor surface and plowing by edges in rough-machined, scored, cracked, or corroded (pitted) rotors are examples of two-body wear. Grinding by wear debris and by foreign particles such as sand, clay, mud, or salt in the sliding interface is three-body wear. Abrasive wear would be expected to increase with tem perature. It increases also with the braking load and with the concen tration, size, and hardness of foreign particles. It may be accompanied by negative fade. Figure 2-13 is a Bendix scanning electron photomicro graph of a dynamometer-tested friction material showing the abrasive wear tracks on the surface. Several adherent wear debris particles are similar in size to the wear track widths. 2.3.4.1.3 Adhesive Wear Adhesive wear, illustrated in Figure 2-14, involves the adhesion of organic and/or inorganic materials to the rotor and the sub sequent tearing or separation of these materials from the lining or pad. Organic constituents of friction materials include the binder, friction modifiers and elastomeric compounds. Inorganic consti tuents include zinc powder, brass chips, and asbestos. Little is known about the tendency of these materials to adhere to the cast iron used in today's drums and discs, though experience with hot elastomeric com pounds suggests that elastomers are very likely to stick. Experiments at Bendix Research Laboratories indicate that brass chips will alloy with certain nonferrous rotor alloys.(2-18) when this occurs, metal is transferred from the rotor to the lining, scoring the surface of the rotor and in turn subjecting the lining to a high rate of two-body abra sive wear. 2-16 HWBUI0002326 DIRECTION OF RELATIVE MOTION ROTOR WEAR DEBRIS Figure 2-11 - Mechanism of Two-Body Abrasive Wear -----------------------------------DIRECTION OF RELATIVE MOTION Figure 2-12 - Mechanism of Three-Body Abrasive Wear 2-17 HWBUI0002327 Figure 2-13 - Scanning Electron Photomicrograph Showing Abrasive Wear Tracks on Friction Material Surface and Adherent Wear Debris Particles ROTOR Figure 2-14 - Mechanism of Adhesive Wear 2-18 HWBUI0002328 Adhesive wear probably increases with temperature, braking load, and the concentration of adherent components in the friction material. It is also caused by the presence in the friction material of metal particles capable of solution in or reaction with the rotor alloy, increasing with the size and concentration of such particles. Because this type of wear can cause stick-slip, it may be associated with brake noise. 2.3.4.1.4 Fatigue Wear Fatigue wear (Figure 2-15) is most commonly^observed in metal bearings. However, it may also be experienced by organic fric tion materials, occurring in two forms, thermal and mechanical. Thermal fatigue is caused by repeated heating and cooling. Such repeated temperature changes impose cyclic stresses on the surface material, which heats and cools more rapidly than the bulk material and thus expands and contracts to a greater extent. Because of their low thermal conductivities, the thermal gradients and resultant thermal stresses in organic friction materials may be higher than those in metals. Thermal-shock cracking may be thought of as a special case of thermal fatigue, occurring as a result of a single abusive loading. Mechanical fatigue is caused by repeated mechanical stressing. Because of the frequency of brake use, friction materials are subjected repeatedly to both compression and shear forces. Addi tional stresses are also applied repeatedly if drums are out-of-round, if discs are out-of-parallel, or if rotor thickness is not uniform. Figure 2-13 shows the initiation of fatigue wear around the large cashew particle. This material swells when heated and thus the surface wears. After cooling, the particle contracts, cracks, and pulls away from the edges. When the particle is completely loose, it can fall out. 2.,3.4.1.5 Macroshear Wear Three types of material removal fall in the category of macroshear wear, which is illustrated in Figure 2-16. A fracture, in shear, may occur across the entire brake pad under extremely heavy loading, especially if the friction material has previously been weakened by heat. This is a sudden rather than a gradual type of wear. Fracture at the edges and corners of a piece of friction material is another type of macroshear wear. Oxidation of organic com ponents weakens the friction material at these locations. 2-19 HWBUI0002329 Figure 2-15 - Mechanism of Fatigue Wear DIRECTION OF RELATIVE MOTION GENERALIZED? FAILURE THROUGH HEAT-AFFECTED ZONE <V FRICTION MATERIAL -ROTOR LOCALIZED FAILURE AT EDGE WEAKENED BY OXIDATION Figure 2-16 - Mechanism of Macroshear Wear 2-20 HWBUI0002330 Fracture at the edges and corners of a piece of fric tion material is another type of macroshear wear. Oxidation of organic components weakens the friction material at these locations. A third type of macroshear wear can occur when a fric tion material such as that shown in Figure 2-15 is subjected to heavy loading following thermal abuse. The heat-affected layer in the material pictured is about 100 mils deep, and from this layer organic material has been removed by pyrolysis. If sufficiently weakened, such a layer can be sheared away by a single, heavy brake application. Whatever the mechanism, macroshear wear is promoted by nonuniform loading and by oxidation and thermal degradation of the binder. It is most likely to occur at elevated temperatures and under severe braking conditions. Its end products are typically large frag ments of friction material that leave the surface rough. These frag ments subject the remaining friction material to plowing or abrasive wear, and while they are present in the interface, friction behaviour will be erratic. 2.3.4.2 Predominant Wear Mechanisms The predominant wear or emissions-generation mechanisms may be summarized simply (Figure 2-17):(2-19) Below 450F: Abrasive and adhesive wear. Above 450F: Thermal wear. Figure 2-18 illustrates the wear loss of three different kinds of fric tion materials as a function of temperature, illustrating these predomi nant wear mechanisms. 2.3.4.3 Wear Equation Recent work at Bendix Research Laboratories has indicated that each friction material wears according to a universal wear equa tion. (2-20) AaWtt = ik p a ,,Vb t c or log (AW) = log k + a log P + b log V + c log t / 2-21 HWBUI0002331 I ISOTHERMAL TEMPERATURE Figure 2-17 - Predominant Wear Mechanisms of Friction Materials Figure 2-18 - Variation of Friction Material Wear with Temperature i 2-22 HWBUI0002332 where AW is the weight lost k is the wear factor p is the load V is the sliding velocity t is the time of sliding, and a,b,c are constants characteristic of a given friction couple. When further developed, this equation will be useful for predicting wear (and emissions) under various driving conditions. 2.3.5 Emissions of Particulate and Gaseous Emissions* 2.3.5.1 Drum Brakes Gaseous decomposition products in drum brakes are emitted from two areas: The space between the rotating drum and the stationary backing plate. The small holes in the backing plate itself (e.g., clearance holes, brake adjusting slot, etc.) Particulates are emitted from the same two areas, although only small particulates would come through the holes in the backing plate. In drum brakes, some particles are not emitted but remain in the rotating drum. It is expected that some of these particles would eventually be emitted through the space between the rotating drum and stationary back ing plate. Also, some large particles would be "ground up" and emitted as a number of smaller particles if they are trapped between the friction material and drum on subsequent braking or if they are trapped between the rotating drum and stationary backing plate. If the brake shoes are of riveted construction, then some particles can be trapped in the rivet holes and thus retained in the brake.. If convective cooling of drum brakes is improved in the future, as is desirable for safer operation and longer driving life, it is likely that fewer particles will remain trapped. The particles emitted, however, will probably be larger, as they will be subjected to grinding action for shorter periods of time. 2.3.5.2 Disc Brakes As the disc brake is essentially an open braking system, it is expected that all particulate and gaseous decomposition products 2-23 HWBUI0002333 will be emitted. However, a small amount of particulates collect in the rivet holes of riveted disc brake pads and on some areas of the caliper. 2.3.5.3 Clutches In the dry clutches commonly in use on "stick-shift" cars, there are very few openings for gases and particles to escape. The two main areas are: The area around the clutch fork, which generally has a rubber boot on it for protection of the clutch from road spray, sand, etc. A small opening at the lower front of the clutch to drain any oil, water, etc., which may get into the clutch. Because of the construction, many particulates remain in the clutch housing. 2.3.6 Distribution and Changes in Brake and Clutch Emissions Entrapped gases and pyrolysis gases released during braking or clutch slipping will be hot enough to mix rapidly with the surrounding air and become part of the air wake of the car. The higher molecular weight vapors will condense and eventually settle to the ground. The low molecular weight gases will disperse as part of the local atmosphere. Some of the condensible vapors may, of course, settle on small, air-borne dust particles and could be carried into the respiratory system as part of the breathing air. The particle emissions will disperse in different ways depend ing on their size, shape, and density. Submicrometer-size particles, re gardless of shape, will ordinarily remain air-borne until they are scrubbed out by rainfall or some other separating action. In effect they become a long-time part of the atmosphere. Spherical particles in the 1.0 pm to 10 ym diameter range will settle, under gravity and in still air, at velocities from 12 to 1000 cm/hr, according to Stokes' law. Irregularly shaped particles of similar size and in turbulent air will settle at much slower velocities and could remain air-borne for hours or days. These particles cause the greatest aggravation in the human respiratory system because they collect in the tracheo-bronchial regions and, to some extent, in the alveoli. Particles from 20 ym to 100 ym in diameter will settle as part of the dust trail left by the car or will collect with any larger particles in mechanical traps in the car. Eventually these larger-size particles will fall or be shaken out of the car and become part of the road dust. 2-24 HWBUI0002334 2.4 REFERENCES (2-1) J. Lynch, "Brake Lining Decomposition Products," J. Air Poll. Control Assoc., 18, No. 12, 824 (1968). (2-2) S. Spiel, Johns-Mansville Research and Engineering, private communication, 1971 (2-3) Private Communication, 1972. (2-4) J. Moran, Environmental Protection Agency, private communica tion, 1972. (2-5) B. W. Klein, "Semi-Metallic Outer Pads for Disc Brakes, "Bendix Technical Journal", 2^, (No. 3), 109-113 (Autumn 1969). (2-6) A. A. Hodgson, "Fibrous Silicates, "Lecture Series No. 4, Royal Institute of Chemistry, London, England, 1965. (2-7) M. G. Jacko, W. M. Spurgeon, R. M. Rusnak, and S. B. Catalano, "Thermal Stability and Fade Characteristics of Friction Materials," SAE Transactions, TT_, 1474 (1969). (2-8) M. G. Jacko, "Thermal Analysis of Friction Materials," presented at the National Meeting of The American Chemical Society, San Francisco, Calif., April 1968; also S. Porter (ed.). Analytical Colorimetry, Plenum Press, N.Y., 1968. (2-9) F. W. Aldrich and M. G. Jacko, "Organic Friction Materials," Bendix Technical Journal, (No. 1), 42-54 (Spring 1969.) (2-10) R. M. Rusnak, H. W. Schwartz, and W. P. Coleman, " A Com parison by Thermal Analysis of Rotor Alloys for Automobile Disc Brakes," SAE Paper 700137, January 1970. (2-11) A. R. Spencer, W. M. Spurgeon, and J. L. Winge, "Four Tests for Consistency of Automotive Brake Linings," SAE Paper 660412, presented at the Mid-year Meeting, Detroit, Michigan, June 1966. (2-12) M. G. Jacko and R. M. Rusnak, "Physical Properties of Disc Pads and Rotors Dynamometers. Tested at 100 mph," Bendix Progress Report, May 1969. (2-13) J. M. Herring, "Mechanisms of Brake Fade in Organic Brake Linings, "SAE Paper 670146, presented at the SAE Annual Meeting, Detroit, Michigan, January 1967. (2-14) W. M. Spurgeon, "Increasing the Service Life of Friction Materials," Bendix Progress Report, June 1969. (2-15) W. M. Spurgeon and A. R. Spencer, "Reliability and Durability of Automative Friction Materials," Bendix Technical Journal, 2_, (No. 3), 57-66 (Autumn 1969). 2-25 HWBUI0002335 (2-16) F. P. Bowden and D. Tabor, The Friction and Lubrication of Solids, Clarendon Press, Oxford, England, p. 53, 1950. (2-17) E. Rabinowicz, Friction and Wear of Materials, John Wiley and Sons, New York, p. 89, 1965. (2-18) S. K. Rhee, J. L. Turak, and W. M. Spurgeon, "An Inertial Dynamometer Evaluation of Three Alloys for Automotive Brake Drums," SAE Paper 700138, presented at the SAE Automotive Engineering Congress, Detroit, Michigan, January 1970. (2-19) S. K. Rhee, "Influence of Rotor Metallurgy on the Wear of Friction Materials in Automotive Brakes," SAE Paper 710247, presented at the SAE Automotive Engineering Congress, January 1971. (2-20) S. K. Rhee, "Wear Equation for Polymers Sliding Against Metal Surfaces," Journal of Wear, 16 p. 431 (1970). 2-26 HWBUI0002336 SECTION 3 MATERIALS SELECTIONS 3.1 BACKGROUND Friction materials and rotor materials vary in composition accord ing to their source. There are original equipment manufacturers (OEM) and aftermarket (AM) friction materials in addition to foreign friction materials; there are wide variations according to use and origin. Friction materials are the expendable item of the friction pair. Con sequently, friction materials are usually replaced two to five times during a vehicle's life-time, while rotor materials are seldom replaced. However, rotor materials are occasionally turned, that is, the rubbing surface is machined so that fresh metal is exposed as a new rubbing surface. The asbestos content of any particular friction material is not known, as this is information proprietary to the manufacturer. With the use of modern analytical tools, it is possible to estimate the asbestos content to the nearest + 2 weight percent. Thermogravimetric analysis yields the organic/inorganic ratio as well as the inorganic residue remaining above 800C. X-ray diffraction indicates the inorganics, other than asbestos, and their relative amounts. 3.2 MATERIALS SELECTIONS FOR PROGRAM 3.2.1 Friction Material Selections A completely balanced program would contain friction materials comparable to most of those in normal customer usage. If a large number of tests can be run, it is easy to fulfill that requirement. If only a small number of tests can be run, the selections are less representative and much more critical. The objective of the program was to obtain brake emissions from disc and drum friction materials, OEM and AM friction materials, friction materials from several different manufacturers, both new and used friction materials, and repeats of at least one axle set of friction materials. This was a considerable task since only a limited number of tests were funded. Three vehicle tests were planned in such a way as to accomplish these goals. It is believed that the friction materials used are representative of a high proportion of those on the road today. A custom-size vehicle was chosen as a median between the compacts and luxury-size vehicles prevalent in the country. The vehicle was equipped with the front disc pads and the rear drum primary and secondary lining combinations, listed in Table 3-1. Each of the front axle and rear axle friction materials was produced by different manu- 3-1 HWBUI0002337 3-2 Table 3-1 - V e h ic le C o n fig u ra tio n s and L in in g S e le c tio n s TP CO XJ 3 cu CO d) o U CU CO ccl > *H H 4-1 X) d) P <O rl co 3 co cd 3 60 H Ci uo C-66*28-d 3 3 iH Ci xJ Tl 33 33 3 Td CU H 3 3 3 rl Ci Cl O 3 33 3 UO xd CU 4-1 4-J 3 Tl 3 rl 3 3 3 3 3 4-J 4-J 3 3 Ci OO O X) 3 C 3 3 3 3 Ci 3 4-J 33 3 43 4-J rl 60 3 3 3 rH 3 Ci Cl 3 Td 3 rH 3 3 3 Ci 3 33 > 5 33 3 5 4-J H CU rl & 0) rH 3 3 33 ts O H O 4J a e to 3 3 3 3 3 3 o 3 H Cl O rl Ci 525 3 P p P u P P CuO 3 Cl 33 4-J o o 3 -rl ct3 O 4-1 *4-1 rl Cd 3 CO Cl 3 rl O <0 > Q, rl Cl Po o <u 4-> rH !> o HP at o Cd 42 p-l h 3 O rl CO > rl P CO 3 O 3 rl rl 4-J Cl 3 3 Cl 4-J O 3 CU >1 o o 3o o rl X 4J H a Td rl 3 Cl 3 PM P CU H O CJ 3 3 4-1 4-> 3 3 3 3 1 3 O 4-J 3 3 P 3 Jo 3 p CO 3 60 o Cl H H 4-> 3o *H P i--d)\ cu co co o o rH | a co H P 4-J 3 O Ci Pm CO i-H CO <T rH vd CO SJ Pm Pm 1 l 00 CM m in Cl 3 Cl Td o 33 3 eO CU H CJ p Cl 3 EM co g 3 Ci P o 3 rl P 4-1 3 Ou Pm 3 o 4J rH G CU H 3 6 Cl rl CU O 60 *rl CO rl 3 3 P OW 4- 3 P +J 03 33 g P CU 3 4-> 3 *4H <C W Cl 4-J 3 3 rQ 3 f" H3 rH CM 3 O rl 3 > rl P 33 o 3 rl 4J Cl 3 3 Cl 4-1 O 3 CU s Cl <) 3U o H X 4J rl O XJ 3 Cl 3 Pm p o cj Cl 3 43 3 43 O3 H EM 3 u3 > rl 3 P Cl rl 3H O -P 3 33 3O 43 4-1 Td 3 rH 3 Cl P4 o rl 5 Pm Ci o o p CJ o 4J 3 P o3 Ct 3P >G Q X JjJ 34. 3 a rl 4J 3 3 H P H o rH u 4J T! 3 Tti 3 o H Cl 3 XJ R 3 fl) o P 3 Ci P <4-1 e 3 C P Pm Pm 1 o v> 1 O 3 rl P 4-J 3 u Ci Pm M CJ PW EM CO 1l Pm Pm Pm Pm 11 ON O <t in CM CM 11 pp Cl to 3 M Cl Td Ci 3 G 3 BO 3 rl CJ CM O 3 CM CO F! 3 C P 4-J 3 p 4-1 Cl 3 3 Bg Cl Cu 4-J <W CO HWBUI0002338 facturers. The vehicle was received new with less than 5 miles on the odometer. Prior to vehicle test 1, all rotors were adjusted to their OEM condition and unused disc pads and linings were installed to replace those with partial use. Vehicle test 1 was made with the OEM test con figuration. Vehicle test 2 was a simulation of a typical consumer prac tice: the replacement of the front axle friction materials only (Table 3-2). The discs and drums were not turned. The rear lining combination was considered "a repeat." The rear lining combination was also consider ed as an extended test of "used" friction materials to ascertain the nature of the brake emissions with extended use against their original rotor. The front disc pads were purchased as aftermarket friction materials sold in the Detroit area. The manufacturer was among the larger AM manufacturers. Table 3-2 - Surface Roughness Measurements on Turned Discs and Drums for Vehicle Test 3 Specification As Machined Front Brakes Right (Shrouded) Left (Normal) 15-80 microinches 28-30 27-32 Rear Brakes Right Left (Shrouded) (Normal) 60-120 microinches 90-95 75-85 Vehicle test 3 was a simulation of a second consumer prac tice: the replacement of friction materials on both axles in addition to turned discs and drums (Table 3-1). The final conditions of the rotors and drums are listed in Table 3-2. The front and rear axle fric tion materials were produced by manufacturers different from those for the friction materials used in any other vehicle tests, and both manufacturers were among the larger AM manufacturers. Briefly summarizing the three vehicle tests, five different friction material manufacturers are represented. Both OEM and AM disc pads and drum linings were selected. The OEM manufacturers represented supply a large part of the OEM market and the AM manufacturers repre sented supply a large part of the aftermarket. All AM friction materials were purchased over-the-counter at various locations in the Detroit area. 3.2.2 Contents of the Friction Materials The asbestos content of the friction materials used was determined by the analytical methods mentioned earlier. The percent of asbestos estimated for each material is listed in Table 3-3. This infor mation was obtained in order to ascertain, in later stages of the program, whether there is any correlation between the asbestos content of a fric tion material and the asbestos content of its wear debris. 3-3 HWBUI0002339 T able 3-3 - E stim a te d A sbestos C ontent CO O 03 o) rO 3 **0M & . 4J o xj 0 u cd E u (1) W03 rl rH o *00) X5 H 03 &0 m VO vO vO <t vo mo co t- in in cm <r vO CO m vo m in co in vo m oo CO VO O rH vo VO oo Ov vo m vO CO vo VO O co vO r-v r-* Ov I CM co CM m vO vO CO vO r*- vo vO 03 O H 60 4J cd *h a -H h mm P -H 0) XJ U 1M a> a) CO -U o O i--l M o 1 Cm H --v Cm Pm W pq O P a vW-' CO rH CO rH vO CO g 03 03 <U 4-1 i-H i--i cd rH ov VO X) Ov 0) 03 co P p to Jo rt O *M x> 03 <d QH hE oo >M Q> o PM C/3 4OJ 3 *M U rH Pm Q O cd rl P Pm EO Cm O CO 1 co oo CM Pgh rH CO m OP vfr g XJ i--l 0 CO VO 2P O 03 QU O cd PUm Pa4) WO Ctf W CM 1 C1O Cm Pm Pm Pm Cm 1 1 Pm av O 1 <r m O CM CM vO 1 1 cp pq pq ppqq << U cd U X5 CO EH O o U a) H CM co uo tad) Pm p4 O H H g W U o gPM a Q003) MPi HM P o O' W *4 U (1) H003) ,SQg CM co gO go P o O o Ov u cd 03 rl 03 >> rH cd cd # O *H H a cd mm MM H n3 a Co *H cd Uu 4J i a) X E *rl E >o cd u u MM 60 O X) PS J <u 4-1 a) cd . E 4J rl 4-1 CO pq w K K K 3-4 HWBUI0002340 SECTION 4 VEHICLE AND INSTRUMENTATION 4.1 VEHICLE DESCRIPTION In this program, a passenger car of medium weight (4850 lbs test weight), produced in high volume was selected. It was equipped with a three-speed manual transmission, ventilated cast-iron disc brakes on the front, and cast-iron drum brakes on the rear. This vehicle was selected in order that the emissions from clutches and from disc and drum brakes could be tested simultaneously on a vehicle size representing a large percentage of the present vehicle popula tion. A detailed vehicle description is given in Table 4-1. Table 4-1 - Test Vehicle Description General Description 1971 two-door hardtop V-8 engine Three-speed standard-shift transmission Power brakes (disc front/drum rear) Power steering Air conditioning Vehicle Weight Curb weight: Test weight: Approximately 4000 lbs 4850 lbs (with driver and passenger) Tires H 78-15 Brakes Front Disc: Rear Drum: Inner Pad Area: Outer Pad Area: 2 8.5 in 2 11.2 in 11 inch diameter by 2-1/2 inch Primary Lining Area: 2 21.0 in 2 Secondary Lining Area: 27.3 in 4-1 HWBUI0002341 4.2 BRAKE TEST INSTRUMENTATION The vehicle was instrumented with standard brake test instrumentation (Figure 4-1): Brake pedal application counter Front and rear brake hydraulic pressure gages U-tube decelerometer Brake lining temperature pyrometers for each wheel (thermocouples were placed in one lining of each wheel in accordance with the SAE specification J-843 Twelve-position thermocouple switch and pyrometer to monitor gas trap and particle trap collector temperatures 4-2 HWBUI0002342 -:k: 4 j 11 Figure 4-1 - Vehicle Instrumentation for Friction Material Testing 4-3 HWBUI0002343 SECTION 5 EMISSIONS COLLECTION SYSTEMS 5.1 OPERATION REQUIREMENTS 5.1.1 Operation and Design Criteria Only 100 percent collection can assure a reliable measure of the total emissions from the clutch, drum and disc brakes. The com posite nature of the emissions, including gases, volatile liquids, and solid particles, makes it impossible to obtain a representative sample on anything less than a 100 percent collection. The first requirement for the emissions collector, then, was that it collect all the particle and vapor emissions from the clutch and brakes. It also had to store these emissions during a complete test cycle without changing them. Another important requirement for the emissions collector was that it should prevent intrusion of contaminants, or at least of contaminants that could not be positively distinguished from brake emis sions. Oxygen and nitrogen had to be excluded from this restriction since operating in an inert gas atmosphere would seriously change the character of the emissions. The collection systemshould not, of course, significantly change the ordinary operating conditions of the clutch or brakes. This means it should not materially change the heat flow and that the air flow rate through clutch and brake chambers had to be similar to that under ordinary operating conditions. The flow pattern did not need to be iden tical since this would not affect the generation of emissions at low tem peratures, but only the way in which they were carried out of the brake and clutch chambers. The front disc collector had to be capable of operating at brake temperatures to 650E (340C) with the sealing surface operating up to 400F (205C). The rear drum collector had to be capable of operat ing at brake temperatures to 500F (260C), the sealing surface operating to approximately 350F (177C). The clutch collector had to be capable of operating at temperatures to 250F (121C). Finally, the collection system had to permit easy separation and analysis of the different types of emission - large and small parti cles down to 0.2 pm size, condensable vapors, and volatile gases. If expensive preliminary separations had to be carried out in the laboratory, an already difficult and time-consuming analysis would become even more extended and costly. 5-1 HWBUI0002344 The major requirements for the emissions sample collection system may be summed up, therefore, as follows: Collect all particulate and gaseous emissions, Store collected emissions unchanged, Prevent intrusion of contaminants, o Maintain ordinary braking and clutching conditions, o Be operable at high temperatures, and Permit easy separation and analysis of emissions. 5.1.2 Emission Expected Appendix A shows a series of wear debris calculations which included the chemical composition and the amounts of wear debris expected. This information was necessary to insure proper sizing of the emission collection hardware. 5.2 OVERALL DESIGN CONCEPT 5.2.1 Air Flow of a Typical Collection System The basic design concept of each brake emissions collection system is shown in Figure 5-1. Air from inside the vehicle was pulled into a bed containing Drierite and calcium oxide pellets which remove water and carbon dioxide from the air. After passing through the bed, the air passed through a filter to remove all particles larger than 0.2 ym (8 x 10"6 inches). The filtered air was then routed to the sealed brake or clutch assembly through clean Teflon tubing. The air stream, along with wear particles and gases, then passed out of the brake or clutch through Teflon tubing surrounded by heating coils to keep the gases at 200F (93C) in order to avoid condensation. This stream passed through two filters to remove the airborne particles and preserve them for analysis. The stream then passed through three cold traps in order to condense and trap the gases evolved during braking and preserve them for analysis. Finally, the air stream passed through a flowmeter, the vacuum manifold, and an air pump run by the vehicle engine, finally being expelled to the atmosphere. Figure 5-2 shows the engine-driven air-pump installation under the hood. The pump was run by a belt attached to an added pulley attached to the drive shaft of the alternator. Thus the air pump had two modes of operation: During engine idle, the pump ran at a low. fixed speed. During vehicle operation, the pump ran at a speed proportional to the engine speed. .. 1 5-2 HWBUI0002345 1-999-28-d INLET AIR FILTER ui * CD <cc m Ol CO C<O d) 4-1 --cdi 3a H 4-1 u <d Pk i--i cad P;>< H 'b 4k <1) O 4-1 CO IS O CO pk u <*rk u o 4k cd c uo 00 -H cd co Qrl -CgO aw rk 4-1 CO cd cd 0aj O rC O C3 co cd I LOI ua) b3O H Pk 5-3 HWBUI0002346 F ig u re 5~2 A ir Pump I n s t a lla t io n in E n g in e C om partm ent HWBUI0002347 5.2.2 Collection Systems Used Collection systems such as those described above were attched to the following: right front disc brake, right rear drum brake, and the clutch. The left front and left rear brakes were not modified in any way but were monitored to insure that the wear on the right side was not changed significantly by the collectors. Figure 5-3 shows, mounted in the front seat of the test vehicle, the front-brake inlet air filter and gas collection system, the clutch gas collection system, and the twelve-position thermocouple switch and pyrometer used to monitor gas and filter trap temperatures. Fig ure 5-4 shows, as mounted in the rear seat of the test vehicle, the rear-brake air filter and gas collection system, the clutch air filter, and the electrical system. 5.2.3 Electrical System The electrical system mounted in the rear seat consisted of two batteries, one inverter, and two variable transformers, and was used to supply AC voltage to the heating coils. Figure 5-5 shows the schematic for the electrical system. The inverter was capable of changing 12-volt DC power to 60-Hz, 120-volt power with an output of 400 watts continuously. Two extra batteries were installed to provide sufficient electrical capacity for 10 hours of testing. The inverter was separated by relays from the batteries in order that peak voltages occurring during vehicle start up did not damage the inverter. This system consisted of power and latching relays and a pushbutton switch. After the vehicle was started, the pushbutton switch was used to energize the latching relay which in turn energized the power relay, providing DC power to the inverter. When the vehicle was shut down for any reason, the power to the latching relay was removed and thus the power relay was also de-energized. The AC output was regu lated to the heater through use of two variable transformers thus allowing temperature adjustment. The batteries were recharged each night in order to provide sufficient power for the following day. 5.3 PARTICULATE FILTERS 5.3.1 Filter Holders Two Gelman 47 mm filter holders joined in series were used for the collection of airborne particulates. These holders were sealed with 0-rings, fabricated from stainless steel for protection against corrosion, and loaded with standard 47 mm diameter filters. Nuclepore filters were used. These filters were made from a smooth film of poly carbonate plastic with a number of circular pores through them. The pores did not restrict the flow of gases significantly as particle col lection proceeded. The filter holders were heated by tape heaters wound 5-5 HWBUI0002348 Figure 5-3 - Vehicle Instrumentation (Front Seat Area) 5-6 HWBUI0002349 :i Figure 5-4 - Vehicle Instrumentation (Rear Seat Area) I;':':' 5-7 HWBUI0002350 5-8 HWBUI0002351 zo DCO Xo X cDo CO Q_ C-66-28-d F ig u re 5-5 - E le c tr ic a l Schem atic fo r P a rtic u la te Trap and T ra n s fe r L in e H eaters around the outside and controlled by a variable transformer within the test vehicle. The same configuration was used in all three collectors (front brake, rear brake, and clutch). 5.3.2 Filter Flow Characteristics At the beginning of the program, no data were available for the flow characteristics for any type of filter when loaded with brake wear debris. Consequently a series of simple tests was performed; the results are included as Appendix B, The following results were noteworthy: The pressure drop change across an 8.0y Nuclepore filter loaded with up to 1.0 gram of wear debris was not high enough to present problems in the program. Wear debris was easily removed from Nuclepore membranes such that no significant error was introduced when debris from several membranes was combined. 5.4 GAS COLLECTION SYSTEMS 5.4.1 Design and Fabrication of Traps To withstand handling and road shocks, it was deemed neces sary to design and build rugged gas traps for both condensable and non condensable gases. Figures 5-6 shows the schematic for the condensable-gas trap. The body of the trap^and the tubing were made of stainless steel. The top was threaded so that it seated against an 0-ring to provide a vacuum-tight seal. Figure 5-7 shows the schematic for the noncondensable-gas .traps. The trap was designed so that it could be filled with granular activated charcoal. The charcoal was held 3/8 inch above the bottom of the.trap by a stainless steel screen. This allowed a small volume for the gases to permeate toward the outside walls before they were absorbed Mfi; on the charcoal surfaces or passed through the charcoal bed to the exit port. Figure 5-8 shows the schematic of the activated-charcoal trap top and the plug used to fill the trap once the top had been threaded on. An 0-ring against a 45-degree bevel provided a vacuum seal. Figure 5-9 shows the complete gas trap. Figure 5-10 shows an exploded view of the trap. 5-9 HWBUI0002352 Figure 5-6 - Condensable-Gas Trap 5-10 HWBUI0002353 P-82-99-3 P-82-99-3 Figure 5-7 - Activated-Charcoal Gas Trap - Sectional View 1/8" PLUG Figure 5-8 - Activated-Charcoal Gas Trap - Top View 5-11 HWBUI0002354 5-12 HWBUI0002355 Exploded 5-13 HWEU10002356 5.4.2 Operation The gas traps were connected in series such that the gases flowed through the condensable-gas trap first and then through two noncondensable-gas traps containing activated charcoal. Valves were in stalled at the inlet and exit of each trap. The traps were kept in a custom-made metal box lined with styrofoam, 1-1/2 inches thick, as shown in Figure 5-11. The three gas traps were held at approximately -70C (-95F) with solid carbon dioxide. For the disc brake collector, which had higher gas flow rates, a 2-foot length of 1/4-inch OD tubing was added before the condensable gas trap. This was found necessary to prechill the gases, or otherwise no condensation would occur in the first trap. 5.5 DISC BRAKE COLLECTOR 5.5.1 Shroud Design and Construction The brake emissions collector for the right front disc brake consisted of a sheet-metal duct, or shroud, completely covering the caliper and disc, and an emission storage unit consisting of particle filters and gas traps. The shroud is shown in Figure 5-12. It served as a collector and duct to.carry the gaseous emissions into the delivery tube to the gaseous-emission storage unit. Figure 5-13(a) and 5-13(b) show the 2 views of major sheetmetal portions of the disc brake collector (parts were stainless steel to avoid corrosion) and Figure 5-14 shows the entire collector system with the disc brake. Front and rear views of the collector system, as mounted on the vehicle, are shown in Figures 5-15 and 5-16. 5.5.2 Rotating Seal 5.5.2.1 Initial Design and Problems The original design of the rubbing seal for the disc brake shroud was a face seal of Teflon (Figure 5-17). Problems arose with this design because of difficulty in realigning the seal after dismantling the shroud for maintenance. When the collector was opened for inspection and then resealed, the out-board shield was not relocated in the exact same position each time. The out-board face was usually slightly cocked in a different manner each time. The out-board face held the rotating seal. A 1-degree shift of the face cause a 0.060-inch deflection at the groove where the Teflon was sealed. A deflection greater than 0.020 inch was found too great for the seal's Teflon 0-ring to make up and consequently nonreproducible sealing and leaking occurred. 5-14 HWBUI0002357 5-15 HWBUI0002358 F ig u re 5 -1 1 - Gas C o lle c tio n System BRAKE PAD TIRE THERMOCOUPLE (FOR BRAKE TEMPERATURE) BEARING DUST CAP ROTATING SEAL P-82-99-3 TO EMISSION STORAGE UNIT Figure 5-12 - Sectional View - Disc Brake Shroud 5-16 HWBUI0002359 I i ! 5-17 HWBUI0002360 I 5-18 '-1 3 (b ) - M a jo r P o rtio n s o f D is c B rake E m issio n C o lle c to r Assem bly HWBUI0002361 I i Figure 5-14 - Disc Brake Emission Collector Assembly - Exploded View 5-19 HWBUI0002362 I i i \ 5-20 BRAKE HYDRAULIC LINE F ig u re 5-15 - F ro n t D isc Brake E m ission C o lle c to r In s ta lle d on V e h ic le HWBUI0002363 u o 4-1 a aj i--i r*H o U CO a o H CO CO H e w 0) cd PQ ua CO o H H p 4-1 cd 4J 4-1 dC o <U J-l g u 4-1 4-1 O CO a aIS) M H bo >c H IS ao cd rc pq CO I vi--OI mI aj u d 60 H :I 5-21 HWBUI0002364 SLIDING SEAL P-S 2 -9 9 -3 Figure 5-17 - Initial Design (MKI) for Rotating Seal of Disc Brake Collector 5.5.2.2 Final Design A new sealing system was designed, using a shaft-type spring-loaded seal, of commercial manufacture, made of graphite-filled Teflon. Figure 5-18 shows the drawing of the sealing system and Fig ure 5-19 shows the details at the sealing surface. 5.5.3 Thermal Control Aids During preliminary testing, it was apparent that the shrouded disc brake was running measureably hotter than the unshrouded disc brake. Therefore, various cooling aids were devised to reduce this temperature differential. Three cooling aids were used: (1) Black oxide coating of the shroud to improve heat radiation. (2) Water sprinkling system connected to the windshield washer pump for cooling. (3) Open or mag-type wheel for improved convective cooling of the shroud. These additions are shown in Figures 5-20 and 5-21. 5-22 HWBUI0002365 5-23 HWBUI0002366 F ig u re 5-18 - R o ta tin g S eal fo r D isc Brake E m ission C o lle c to r i HUB DISC 4340 PRESS FIT RING - RUBBING SURFACE GRAPHITE FILLED TEFLON SEAL DISC BRAKE SHROUD Figure 5-19 - Details of Disc Brake Collector Rotating Seal P-82-99-3 5-24 HWBUI0002367 i 5-25 HWBUI0002368 SHROUD F ig u re 5 -2 0 - F ro n t D is c B rake E m issio n s C o lle c to r Showing Therm al C o n tro l A ids 5-26 HWBUI0002369 F ig u re 5-21 - E ig h t F ro n t Wheel o f Test V e h ic le 5.6 DRUM BRAKE COLLECTOR 5.6.1 Initial Designs and Rotating-Seal Problems The original concept for the sample collector on the right rear collector was planned such that the drum and backing plate formed the shroud along with a rotating seal. This approach was used in order to minimize changes to the cooling efficiency of the drum. The first design concept (MKI) was a face seal of Teflon kept in contact with the polished edge of the backing plate by a silicone rubber O-ring (see Figure 5-22). This seal had a short service life, be cause of alignment problems and the high rubbing speeds. The second design concept again attempted, to use the back ing plate and drum as the major components of the shroud. However, this design (MKII) incorporated a shaft-type spring-loaded seal, of commercial manufacture, made of graphite-filled Teflon. The seal manufacturer in dicated that this seal could sustain the rubbing speeds and temperatures involved (2000 ft/min and 400F). This seal design is shown in Fig ure 5-23. Figure 5-24 shows the modified backing plate and drum made from this design. Figures 5-25 and 5-26 show front and rear views of the system as mounted on the vehicle. The seal in this system failed Figure 5-22 - First Seal Design (MKI) for Drum Brake 5-27 HWBUI0002370 P-82-99-3 .M 5-28 HWBUI0002371 F ig u re 5 -2 3 - M K II R o ta tin g S e a l f o r Drum B rake. E m is s io n C o lle c to r (a) Modified Rear Brake Drum Backing Plate Showing Spring-Loaded Teflon Seal (b) Modified Rear Brake Drum Showing Hardened Steel Ring for Rotating Seal Figure 5-24 - Details of MKII Rotating Seal 5,-29 HWBUI0002372 5-30 HWBUI0002373 after a short time as shown in Figure 5-27. On investigating the failure, three major problems were found: (1) High interface speeds (1450 ft/min at 40 mph). (2) Increase in seal interface pressure because of lack of concentricity and expansion of the drum. (3) Heat input to the seal from the braking action and dis tortion of the drum from the braking action. 5.6.2 Final Design In order to correct the problems described above, a third seal design (MKIII) was developed. This design removed the rubbing seal from the drum to the axle housing area, decreased the diameter (and thus the rubbing speed) of the seal and improved concentricity by using ac curate alignment of components and by modifying the rear axle hub to align the drum concentric to the axle bearings within 0.001 inch. A tapered lead was used on the axle hub to facilitate mounting of the drum. This design is shown in Figure 5-28. An outboard view of this sytem without the drum is shown in Figure 5-29. Outboard and inboard views of the assembled system are shown in Figures 5-30 and 5-31. This is the system which was used in all testing. 5.7 CLUTCH COLLECTOR The clutch collector system consisted mainly of a filtered air inlet and an outlet to the sample storage system with all other external holes sealed. A schematic of the system is shown in Figure 5-32 and a photograph of the system installed is shown in Figure 5-33. 5-31 HWBUI0002374 .1 \ I Figure 5--26 - Rear View of Rear Drum Brake- Emissions Collector Showing Instrumentation i i Figure 5-27 - Completely Worn Seal from MKII Rear Drum Brake Collector 5-32 HWBUI0002375 DRUM Figure 5-28 - Sectional View of Drum Brake Shroud (MKIII) 5-33 HWBUI0002376 P -8 2 -9 9 -3 5-34 F ig u re 5 -2 9 - O utboard V iew o f R ear Drum B rake E m issio n s C o lle c to r HWBUI0002377 HWBUI0002378 5-36 AXLESE F ig u re 5 -3 1 - In b o a rd V iew o f A ssem bled R ear Drum B rake E m issio n s C o lle c to r ! HWBUI0002379 FIREWALL F ig u re 5-32 - E m ission C o lle c tio n System fo r C lu tc h DUCl Ul > VCM OSo ui g DC O*e* < CC Ul <z z o oIDo Stl5l [: ' 5-37 HWBUI0002380 5-38 F ig u re 5-33 - G eneral View o f Covered C lu tc h HWBUI0002381 i-. . SECTION 6 INERTIA DYNAMOMETER TESTING OF THE DISC-BRAKE COLLECTOR SYSTEM 6.1 OBJECTIVES There were three major objectives for the inertia dynamometer tests. First, it was necessary to demonstrate that a brake can be completely enclosed and not change in its mechanical operation. Second, it was necessary to demonstrate that the thermal response would not be seriously altered compared to the normal brake operation. Finally, the most critical unknown was the rotating seal operation. 6.1.1 Demonstration of Collection System Operation The disc brake on the vehicle was designed to operate as a relatively open brake with a substantial air flow over its component parts. Shrouding of the caliper and rotor was a radical change from its original designed operating mode. The shrouded brake for emissions collection was a unique piece of hardware and thus it required testing on the inertia dynamometer, where modifications and adjustments could be made rapidly and rationally. 6.1.2 Thermal Response It was known that shrouding would change the brake cooling. However, the effects of pulling air through the brake shroud even while the vehicle was stationary would negate some of the shielding effect of the shrouding. Thus it was necessary to determine the thermal response of the shrouded brake compared with that of its original design and to ascertain the effects of any thermal differences. 6.1.3 Rotating Seal Durability At the beginning of the program, it was believed that the rotating seal life would be the most critical problem. Several questions arose. Would the initial design be suitable? Was a good back-up design available? Would any rotating seal survive under the conditions of opera tion? Were the claims of the seal manufacturers accurate and reliable? Thus the most critical objective of the inertia dynamometer test was to demonstrate the operation of a suitable rotating seal and to demonstrate its durability. 6.2 DYNAMOMETER TEST RESULTS 6.2.1 Sealing Practices for Disc Brake Collector Figure 5-14 shows the major portions of the front disc brake collector. The silicone rubber gaskets form the major portion of the j, | I ; j k: [ l. ' |;; j ! i j j j ! 6-1 i- HWBUI0002382 seals between the joints of the metal shroud. Once assembled on the wheel assembly, there were many leaks around the small screws holding the shroud together.Other leaks appeared at the welds on the outboard shield and the fittings on the modified dust shield. The leaks around the welds were eliminated by returning the parts to the shop for additional welding. The other leaks were eliminated by the liberal use of cured-in-place silicone rubber (see Figures 5-15 and 5-16). 6.2.2 Thermal Response Results The first testing performed was to determine the thermal response of the shrouded disc brake as compared with that of the unshrouded disc brake. The inertia dynamometer used was set up to simulate a load of 1200 lbs on the brake, which is approximately 25 percent of the vehicle test weight (Figure 6-1). Each stop was from 30 mph to 0 mph at 300 psi line pressure. Stops were made at the rate of one per minute. The tests were run in still air. Figure 6-2 is a plot of temperature versus time for a thermocouple located in the outer pad and a thermocouple mounted on the backing plate. The solid curves are for the regular or unshrouded system, while the dashed curves are for the shrouded system, with no air being pumped through the system. Figure 6-2 shows that the shrouded brake heats faster than the unshrouded brake which was to be expected. Figure 6-3 is a plot of the data from a series of tests on the shrouded disc brake system for thermal response of the outer brake pad under various air flow conditions. The plot shows temperature versus total energy input to the brake. The lowest curve on the plot is for the un shrouded brake and the highest is for the shrouded brake with no air flow as previously shown in Figure 6-2. The two middle curves are for air flows of 64 and 110 standard cubic feet/hour. The results show that increasing the air flow above about 60 SCFH (i.e., doubling the flow) had little effect on the cooling of the brake. Thus 60 SCFH was chosen as the flow to be set in the test vehicle. 6.2.3 Rotating Seal Operation The initial rotating seal design described in Section 5,5*2.1 did not provide a suitable seal. It was therefore replaced with the de sign described in Section 5.5.2.2. The following tests were performed with the second design. Testing was done to determine rotating seal effectiveness. The method used to determine the seal leakage rate was to apply air pres sure to the closed shroud at 3.5 psi and measure the time required for the primary pressure regulator reading to drop from 1500 psi to 500 psi while the tank valve was closed. The total flow through the secondary regulator, set at 3.5 psi, was about 137 cc during this time. Figure 6-4 shows a plot of the leakage times versus wheel speed for a new seal, both lubri cated and unlubricated. As can be seen, the leakage time increases as speed increases for both lubricated and unlubricated seals. Since the 6-2 HWBUI0002383 Figure 6-1 - Disc Brake and Emissions Collector Assembly Installed on Dynamometer lubricated seal had lower leakage rates, the vehicle tests were all run with lubricated seals. This leakage test method was used throughout vehi cle testing and seals were changed whenever the leakage rate fell below 20 seconds on a cold system. One seal was then tested for a total of seven hours at 50 mph and the results of the leakage tests versus speed are shown in Fig ure 6-5. The seal was removed after testing and inspected. Figure 6-6(a) shows that most of the seal experienced little wear during the test. However, one section, approximately one inch long, showed abnormally high wear to a point of imminent failure (Figure 6-6(b)). This was apparently caused by spring distortion in this area causing higher pressure. Sub sequent seals were inspected for this defect prior to assembly and this problem was avoided in all subsequent tests. 6-3 i I' I HWBUI0002384 TEMPERATURE P -8 2 -9 9 -3 Figure 6-2 - Disc Brake Comparative Thermal Response Time (Static Air) Figure 6-3 - Comparative Thermal Response Tests on a Disc Brake 6-4 HWBUI0002385 Figure 6-4 - Seal Leakage Versus Flow for Disc Brake Emissions Collector Showing Differences Between Dry and Lubricated Seals Figure 6-5 - Seal Leakage Rate Versus Speed Showing Differences Between New and Used Seal 6-5 HWBUI0002386 (a) Normal Worn Seal (b) Abnormal Wear to the Point of Imminent Seal Failure Figure 6-6 - Seals From Disc Brake Emissions Collector Showing Normal and Abnormal Wear 6-6 HWBUI0002387 A new rotating seal was installed at the beginning of the tests described in the next section. The initial seal integrity was equivalent to that of the seal used during the durability test runs. Figure 6-7 shows the data for this seal. The seal integrity was found to improve as the seal wore in. Moreover, the leakage rate further decreased as the wheel speed increased. Finally, the data for this seal agreed with all the trends shown by the earlier seals. 6.2.4 Emission Collections A series of dynamometer tests were made to check out the operation of the entire disc brake emissions collection system. Identify ing the tests by a year-month-day date code, a total of three tests were run: Test 11222: Baseline Test 20105: Baseline repeat Test 20108: Baseline plus spray Test 20108 was similar to the other two tests except that a mixture of 5 percent sodium chloride and 5 percent barium sulphate in water was periodically sprayed at the emissions collector. The salt and barytes (50 percent through 200 mesh and 50 percent through 325 mesh) were used to simulate a fall-winter slushy-wet road condition. Barium sulfate was used in place of sand so that the barium could be used as a tracer along with the chlorine in subsequent X-ray fluorescence analyses if needed. Each of the tests was run for two hours on the inertia dy namometer. Pre-run and post-run time added to the total. During the two-hour run, 40 brake stops were made from 30 mph at 8ft/sec2 decelera tion rate (considered a normal stop). The wheel was run at 30 mph between stops. The temperature-versus-time plots for both the backing plate and the outer brake pads are shown in Figure 6-8. The reproducibility of Tests 11222 and 20105 was good. As expected, Test 20108 ran cooler be cause of the cooling spray. Table 6-1 summarizes the dynamometer test results for flow rates, temperature control, and particulates collected. A slight decrease in carrier gas flow rate was seen as the filter picked up wear debris; this agreed with previous laboratory test data of pressure drop versus flow rates for Nuclepore membranes. The temperatures of the transfer line and filter assembly were held at 200 + 20F, the target temperature range, for the two base-line tests. The transfer-line temperature for Test 20108 fell below this for a brief period because of cooling effect of the spray. However, this was not considered a problem. The particulates data are also given in Table 6-1. The wear debris on the filter was representative of normal vehicular brake wear debris found in drum brakes and trapped from previous Bendix sample 6-7 HWBUI0002388 DECAY TIME (SEC) P -8 2 -7 9 -3 Figure 6-7 - Rotating Seal Leakage Rate Versus Wheel Speed for Disc Brake Emissions Collector 6-8 HWBUI0002389 TEMPERATURE (F) 300 RUN NO. 11222 -- 20105 -- -- 20108 ................. 0 30 60 90 120 TIME (MIN.) (a) BACKING PLATE TEMPERATURE: TEMPERATURE (F) C? 01 01 CN) 00 a. Figure 6-8 -- Disc Brake Emission Collector Temperature 6-9 HWBUI0002390 dynamometer tests. The shroud samples contained mainly grease and some Teflon-graphite wear particles from the rotating seal. The amount of grease required was not known at the start of these tests. A liberal amount was used initially. Due to the heat of the rotating seal and to the air flow through the rotating seal (slight vacuum in the shroud), the grease flowed into the shroud and coated a portion of the inside surface near the rotating seal. This grease film attracted and held the wear debris. Assuming that the brake lining wear debris fraction* of the shroud sample is a constant (^10 weight percent), the total particulates produced for each of the three tests agreed very well. The mean was 17.4 +0.3 mg. The air-borne fractions collected on the filters showed a wider variation with a mean of 12.8 + 0.'5 mg. 6.2.5 System Integrity The third test in the preceding section was carried out to ascertain system integrity under simulated harsh road conditions. This test was run in same manner as the first two tests except that a mixture of 5-percent sodium chloride and 5-percent barium sulphate in water was sprayed at the emissions collector system. A total of two liters of this mixture, used to simulate fall-winter slush conditions, was sprayed on the system during the test. No traces of salt or of barium sulphate were observable on the inside of the collection system after testing. The temperature data for the heated transfer line and particulate filters and the particulate collection results for these tests are shown in Table 6-1. These results show that no salt entered the shroud system, and prove the integrity of the system used in vehicle testing. Table 6-1 -Summary of Dynamometer Test Results RUN NUMBER 11222 20105 20108 CARRIER AIR FLOW (SCFH) Initial Final TEMPERATURES (F) Transfer Line Filters 36 31 190-210 180-205 36 25 180-190 190-220 36 26 170-220 190-215 PARTICULATES COLLECTED (mg) 8 y Filter 0.2 y Filter Shroud* Total** 10.6 9.6 10.2 1.7 48.4 17.1 3.6 41.8 17.4 2.7 46.6 17.6 REMARKS Baseline Baseline repeat Salt spray used; none detected inside collector *Includes grease and Teflon-graphite wear particles from seal (probably >90% of amount shown) **Includes 10% of shroud figure * No assumption as to the actual content of wear debris in the samples taken from the brakes during the vehicle tests was necessary. Much less grease was used and practically none of it entered the shroud during these tests. 6-10 HWBUI0002391 SECTION 7 VEHICLE TEST SCHEDULES 7.1 RATIONALE 7.1.1 Brakes ----------- For a given vehicle brake system, friction material wear is primarily a function of the duty cycle. For light duty cycles, where pad and lining temperatures remain under 350F, low wear results and is primarily due to the abrasive and adhesive wear mechanisms. Heavier duty cycles at higher temperatures give some combination of thermal, abrasive and adhesive wear. Under severe heavy duty where the use borders on abusiveness brake fade may occur. The inability of the brakes to hold the friction level results in part from the formation of gases (and wear debris) at the interface. The amounts and types of gaseous and particulate emissions, as a function of brake duty, are not known. Moreover, as brake linings age, changes such as "hardening" occur, possibly with changes in the wear properties. Also, until disturbed, portions of the particulate emissions are known to be retained inside certain areas of the foundation brakes. Consequently, it is necessary to select a vehicle test program which will provide some combination of normal and heavy duty service. 7.1.2 Clutches For a given vehicle, the friction material wear in the clutch is primarily a function of the duty performed. However, in most clutch applications, the energy absorption and temperature rise are usually less than in normal brake operation. Thus the wear and emission rates in clut ches were expected to be very low. 7.2 TYPICAL TEST CYCLES Several acceptable vehicle test cycles have been established and used by friction materials and brake system manufacturers. These are varied according to life and performance requirements, and for various stages of friction material conditions. The following test cycles appeared to be suitable for this program. 7.2.1 Burnish A portion of the SAE Recommended Practice: Brake System Road Test Code - Passenger Car (SAE J843a) is called the Burnish. This jv ! i |^; I; ' j 7-1 HWBUI0002392 test cycle is always used when the brakes are relatively new and thus is a part of the break-in procedure for new brakes. The test requirements Stop speed: Stop deceleration: Stop interval: Cooling speed: Stops required: Optional: 40-0 mph 12 fpsps As required to achieve 250F initial brake temperature in a maximum of one mile. 40 mph (moderate acceleration to cooling speed). .200 Inspect and/or adjust brakes after burnish cycle. 7.2.2 Baseline Since there was no cycle in SAE J843b comparable to a re peated, low duty-type baseline test, it was found necessary to adapt a portion of some other longer test cycle. This test would be used to serve as a reference for wear rate and brake emissions after other low and heavy duty test cycles. The baseline test selected was three days (12 circuits) of the Detroit Traffic Test schedule described below. 7.2.3 Detroit Traffic Test The Detroit Traffic Test is used primarily to evaluate friction, life, and noise of commercial brake products. The normal oper ating range for passenger car brakes in the test is 180 to 280F. This condition constitutes what a brake engineer calls a low-temperature wear test, but is actually somewhat more severe than for most driving. Test ing is intentionally accelerated to reduce test costs, consequently the test tends to be more severe than average driving. Table 7-1 describes the detroit Traffic Test route as modi fied for this program. The basic downtown circuit remains the same; only the starting and finishing routes are slightly altered. Each circuit begins with elevated brake temperatures. The vehicle merges and flows with the traffic during the prevailing traffic conditions. The following variations occur: Stop speeds: Snubs*: Stop deceleration: Stop interval: Variable: 45-0 mph maximum. Variable: 45 mph maximum. Variable: 20 fpsps maximum. As required to maintain flow in traffic. Decelerations to lower speeds, but not stops. 7-2 HWBUI0002393 Table 7-1 - Detroit Traffic Test Route (Modified) A. Drive from Bendix Research Laboratories (BRL) to 12-Mile Road and Woodward Avenue via the following route: Left on Civic Center Drive to Greenfield Avenue Left on Greenfield Avenue to 12-mile Road Right on 12-Mile Road to Woodward Avenue Turn right on Woodward Avenue and park on Service Drive to check instrumentation. B. Detroit Traffic Route: Starting on Woodward Avenue (just South of 12-Mile Road), drive south to Adams Avenue (starting point of downtown loop). Right on Adams Avenue (West) to Park Avenue Left on Park Avenue to Bagley Avenue Right on Bagley Avenue to Third Avenue Left on Third Avenue across Michigan Avenue to Fort Street Left on Fort Street to Griswold Street Right on Griswold Street to Jefferson Avenue West on Jefferson Avenue to Washington Boulevard Right on Washington Boulevard to Fort Street Right on Fort Street to Cadillac Square Right on Cadillac Square to Randolph Street Left on Randolph Street to East Fort Street Right on East Fort to Brush Street Left on Brush Street to Macomb Street Left on Macomb Street to Randolph Street Cross Randolph Street to Broadway Avenue Right on Broadway to Witherell Street Right on Witherell Street to Adams Avenue Left on Adams Avenue to Woodward Avenue This completes one downtown loop. Complete a second downtown loop by continuing west to Park Avenue and following around and returning to the corner of Adams and Woodward Avenues After the Second downtown loop, turn right and follow Woodward Avenue (north) to 12-mile Road Turn left, making U-turn around center island and return south on Woodward Avenue, crossing 12-Mile to park on Service Drive again to recheck instrumentation This completes one circuit of the Detroit Traffic Test. C. Return to BRL from 12-Mile Road and Woodward Avenue by following route: After checking out instrumentation, drive west on 12-Mile Road to Greenfield Avenue Left on Greenfield Avenue to Civic Center Drive Right on Civic Center Drive to BRL D. Estimated time scheduling Time 8:00 - 8:15 8:15 - 8:30 8:30 - 10:15 10:15 - 12:00 12:00 - 12:45 12:45 - 2:30 2:30 - 4:15 4:15 - 4:30 4:30 - 4:45 AT (min) 15 15 105 105 105 105 15 15 Vehicle preparation BRL to 12-Mile and Woodward First circuit Second circuit Lunch Third circuit Fourth circuit Return to BRL Prepare vehicle for overnight layover E. Estimated Miles BRL to route and return Each circuit 26.5 x 4 12 106 118 miles/8-hour shift. Estimated total miles: 12 x 118 = 1416 miles. j i | !; 7-3 i..' i. HWBUI0002394 Cooling speed: Normal traffic speeds. Stops required: As required in flow of traffic. The duration of the Detroit Traffic Test is not fixed for normal original equipment manufacturers' testing. For the purposes of this program, a 12-day test duration was chosen. Each day of testing was expected to provide approximately 118 test miles which, in turn, was expected to yield a total of approximately 1416 test miles. 7.2.4 10-Stop Fade and Recovery A portion of SAE J843a is called the fade sequence. This test cycle is used as the first high-temperature operation of a friction material. The test is used to simulate a vehicle making repeated heavyduty stops. The fade sequence is usually followed by a recovery sequence. The test requirements are: Fade: Initial brake temperature: 150 F before first stop. Stops required: 10* Stop speed: 60-0 mph. Stop deceleration: 15 fpsps (in normal driving gear). Stop interval: 0.4 mile. Cooling speed: 60 mph (accelerate at maximum speed). Recovery: Cooling speed: 40 mph Cooling interval: Stops required: 1.0 mile. 12. Stop speed: 30-0 mph. Stop deceleration: 10 fpsps. Lining temperatures are recorded b ore each fade stop and before each recovery stop. 7.2.5 15-Stop Fade and Recovery The 15-Stop Fade and Recovery is comparable to the 10-Stop Fade and Recovery described earlier, except that an additional 5 stops are made when the brakes are already very hot. 7.2.6 Reburnish The Reburnish portion of SAE J843a is essentially similar to the Burnish portion except that only 35 stops are required. 7-4 ! i HWBUI0002395 7.3 SELECTED VEHICLE DRIVING SCHEDULE Although there are many acceptable original equipment manufacturers' friction material tests to measure a specific condition or combination of conditions, there is no one test which adequately matches normal driving conditions which would be representative of all vehicles. Consequently, it became necessary to devise a rational and meaningful vehicle test driving schedule. Table 7-2 lists the selected vehicle driving schedule for each ve hicle test configuration used in the program. A total of seven test schedules was chosen and each was followed by a measurements procedure (take emissions samples, measure wear, inspect systems, and replace worn parts). The first three tests -- Burnish, After-Burnish Baseline, and Detroit Traffic -- represent the low-temperature tests. The final four tests -- 10-Stop Fade, After-Fade Baseline, 15-Stop Fade, and Final Baseline -- represent the high-temperature tests. 7.3.1 Burnish The initial test is the Burnish. The original equipment and aftermarket linings mate and conform to the initial condition of the rotor whether it is in its original configuration, as-used condition after vehicle test 1, or its turned condition prior to vehicle test 3. It was anticipated that the emissions would be different than those for wornin linings. 7.3.2 After-Burnish (A.B.) Baseline The second test is After-Burnish (A.B.) Baseline. This three days of Detroit Traffic Test was run at this time so that the other Baselines run at the later times could be compared to the A.B. Baseline for the amount and type of emissions. 7.3.3 Detroit Traffic Test The third and longest test employed was the 12-day Detroit Traffic Test. This test represents the low-temperature-wear or normalduty-wear tests. The results of this test could also be compared to the A.B. Baseline Test. The end of the Detroit Traffic Test was the end of the lowtemperature-wear testing. 7.3.4 10-Stop Fade and Recovery The 10-Stop Fade and Recovery was the fourth test sequence and the start of the heavy-duty or high-temperature testing. During this test, the front brakes experienced temperatures above 500F, while the rear brakes experienced temperatures above 350F. The combination of increased temperature and increased duty was expected to change the amount and type of wear debris generated, as compared to the previous low-duty tests. 7-5 I HWBUI0002396 Table 7-2 - Vehicle Test Schedule Test Burnish Measurements* Estimated Time (Days) Total Time (Days) 2.0 1.5 3.5 Estimated Lining Loss (Mils) 5 Purpose Lining Preparation After Burnish (A.B.) Baseline Measurements 3.0 " 1.5 4.5 5 Base Line Data Detroit Traffic (Modified) Measurements 12.0 1.5 13.5 10 Low Temperature Wear 10-Stop Fade and Recovery Reburnish Measurements 1.5 1.0 1.5 4.0 10 High Temperature Wear After Fade (A.F.) Baseline Measurements 3.0 _ 4.5 5 Base Line Data 1.5 15-Stop Fade and Recovery Reburnish Measurements 1.0 1.0 1.5 3.5 10 Abusive Wear Final Baseline Measurements 3.0 - 4.5 5 Base Line 1.5 Data 38.0 * Take emissions samples, measure wear, inspect systems, and replace worn parts. P-82-99-3 7-6 HWBUI0002397 sequence. The Fade and Recovery were then followed by the Reburnish 7.3.5 After-Fade (A.F.) Baseline This fifth test sequence was identical to the second and seventh test sequences (and similar to the third), but only the conditions of the friction materials had changed. This test was expected to deter mine the permanent changes in the amount and type of wear debris generated for friction materials taken through high-temperature and heavy-duty stops. 7.3.6 15-Stop Fade and Recovery The sixth test sequence was the 15-Stop Fade (and Recovery) in which the friction materials were taken to even higher temperatures than for the 10-Stop Fade. The five additional stops performed after the brakes were already hot produced temperatures above 575F. This is abusive braking, and rarely occurs. The emissions generated were expected to be different again, as compared with the low-duty tests and the 10Stop Fade. sequence. The Fade and Recovery were then followed by the Reburnish 7.3.7 Final Baseline The results of this seventh and last test sequence can be compared with the results of the identical second and fifth test sequences. This test concluded the determination of the effects that high-temperature and heavy-duty tests have on brake emissions. 7-7 HWBUI0002398 SECTION 8 SAMPLING AND ANALYTICAL PROCEDURES 8.1 RATIONALE BASED ON COLLECTOR DESIGNS 8.1.1 Particulate Emissions Collection During the braking process, particulate emissions are gen erated by the rubbing action of the friction materials and the rotors. The OEM brake configurations (left wheels of test vehicle) release emis sions which reside in either of two areas depending on the brake type (Table 8-1). The disc brake releases much of the emissions to the atmos phere while a small amount remains in the brake. The drum brake releases a lesser amount to the atmosphere and retains sizeable portions on the drum rubbing surface and on the brake parts. The brake emissions collec tors perform two functions not encountered in normal brake operation: collect all the debris generated and fractionate the debris. Table 8-2 summarizes the particulate emissions samples produced. 8.1.2 Gaseous Emissions Collection During the braking process, gaseous emissions are also generated by the heat associated with the rubbing action of the friction materials and the rotors. The OEM brake configurations release gaseous emissions which again reside in either of two areas. Most of the gaseous emissions are released into the atmosphere; some of the high molecular weight materials distill from the hot surfaces and condense on the cooler brake parts. The emission transfer lines used in conjunction with the collectors therefore were heated so that almost all gaseous emissions could be transferred to the gas collection traps. 8.2 PARTICULATE EMISSIONS REMOVAL FROM COLLECTORS 8.2.1 Collection of Particulate Emissions A detailed step-by-step sequence was prepared for the col lection of brake and clutch emissions. This information is included as Appendix C. The weights of each sample collected are reported in Section 9 under Vehicle Test Results. 8.2.2 Disc Brake Samples In all cases, three samples were obtained from the right front disc brake collector: sump sample (including debris in rivet holes, on lining surfaces, and inside caliper piston), wear debris which remains in the brake; surfaces sample (including all other debris remaining in the brake collector); and the airborne sample (debris collected on the 8-1 HWBUI0002399 Table 8-1 - Comparison of Disc Brake Versus Drum Brake Operating Conditions Parameter Type of system Air flow Emissions Third-body abrasive wear Disc Open High Relatively few trapped Relatively none Drum Closed Low Much trapped in surfaces and in sump Some Table 8-2 - Samples Produced by Brake Emissions Collection Sample Sump Surfaces Airborne Disc Brake Trapped debris* - includes that in rivet holes, on pads and in caliper piston Accumulated on shroud normally road dropout Airborne samples collected on filters Drum Brake Trapped debris* - includes that in rivet holes, on linings, and on drum surface* +Provides third-body wear Accumulated on brake surfaces* Airborne samples collected on filters *Normally not completely released during brake operation. (This may be a "controllable" material which could be disposed of properly). P-82-99-3 8-2 HWBUI0002400 8.Op and 0.2y Nuclepore filter train). The contract did not require Bendix Research Laboratories to sample the left front disc brake; however, some samples were taken so that it would be possible to assess the quantity of material remaining in an OEM configuration disc brake. These data were prepared for Section 9. 8.2.3 Drum Brake Samples In all cases, three samples were obtained from the right rear drum brake collector: sump sample (including debris in rivet holes, on lining surfaces, and on drum rubbing surface), wear debris which remains in the brake; surfaces sample (including all other debris remaining in the brake collector and on the brake surfaces); and the airborne sample (debris collected on 8.Op and 0.2y Nuclepore filter train). The contract required that selected left rear drum brake samples be taken in order to assess the quantity, of material remaining in an OEM configuration drum brake. These data were prepared for Section 9. 8.2.4 Clutch Samples Since the airborne samples taken from the clutch for vehicle test i were extremely small compared to the corresponding airborne samples from the brakes-, it was decided to accumulate the entire sump and surfaces samples for all three vehicle tests. These data were prepared for Sec tion 9. 8.2.5 Improved Friction Material Weighing Procedure The original equipment drum brake linings and disc brake pads were riveted to their corresponding1 brake shoes. The weight of the metallic shoe placed the combined weights in excess of 400 gm; the weight could then be recorded to the nearest 0.1 gm. The anticipated weight differences of approximately 0.5 gm would have had less than desired accuracy. By removing the rivets, the weights of the friction materials alone could be determined. Since all friction material weights were less than 200 gms in this condition, weighing on an analytical balance was possible. After the friction materials were dried in a vacuum oven at 90C, the weights were recorded to the nearest 0.01 gm, giving the desired accuracy. The linings were then refastened to the shoes by flat head nuts and bolts for vehicle installation and testing. 8.3 ' ANALYTICAL PROBLEM DEFINITION 8.3.1 Composition of Wear Debris The brake converts the kinetic energy of the moving vehicle into heat, absorbs the heat, and eventually dissipates it to the atmos phere* As a result of the energy conversion at the stator-rotor inter face, the compositions and microstructures of the brake lining and rotor surfaces are altered. 8-3 i K ' HWBUI0002401 At low service temperatures, the phenolic resin and organic friction modifiers in the friction material undergo ploughing and mechani cal damage before they convert on the surface to compositions with slightly higher carbon/oxygen ratios. At somewhat higher use temperatures, the resin and organic friction modifiers convert to carbon more readily, and this in turn oxidizes to carbon dioxide. At the same time, the asbestos, converted asbestos, and other converted materials wear away more rapidly because of the reduced binder strength and volume at the surface. At still higher temperatures, the asbestos converts from its fibrous form to an olivine powder having little or no reinforcing value. When heat is generated at the sliding interface much faster than it can be dissipated, the temperature rises and reaches a point where the lining or pad compon ents decompose, oxidize, or melt. The region near the sliding surface of the lining containing these reacted components is called the frictional heat affected layer (FHAL). At low service temperatures, the surface asperities of the rotor are mechanically worked and the surface becomes essentially free of graphite. Also, the cementite plates in the surface layer are broken up and distributed as fine particles in a ferrite matrix. When the sur face is subjected to very high temperatures, martensite forms at the rubbing surface, with an accompanying increase in surface hardness. Consequently, the thermal and mechanical forces which act at the friction couple interface produce a complex chemically and physi cally altered microstructure which cannot be accurately described for the resultant wear debris. Figure 8-1 shows the thermogravimetric analysis of typical rear drum brake wear debris. Table 8-3 gives an estimate of the chemical composition of wear debris. 8.3.2 Analytical Methods for Asbestos and Their Rationale As pointed out earlier, asbestos is a complex inorganic material with the approximate composition formula: Mg3 (Si05) (OH)^ or 3MgO 2Si02 2H20 The basic unit has a fibril form. Asbestos is readily identified when alone or in simple mixtures at high concentrations by the following ana lytical methods: X-Ray diffraction Thermal methods Microscopy Infrared analysis 8-4 HWBUI0002402 % J.HDI3M TEMPERATURE 8-5 HWBUI0002403 Table 8-3 - Estimated Chemical Composition Characteristics ` of Wear Debris Description Percent Low Molecular Weight Degradation Products Polymeric Components Carbonaceous Material 2-10 \ 12-20 > 1-5 ) 20-30 Inorganics: Mineral Olivine Oxides (from Lining) ^e2^3 ^rom Rtor) Asbestos VL0-30 \ ^20-40 f 0-10 > 5-15 \ <1 / 70-80 However, in complex mixtures, or at very low concentrations, the analysis for asbestos is very difficult. In brake wear debris, the problem is compounded because the reaction products of asbestos, forsterite and olivine, have similar elemental ratios and similar X-ray diffraction pat terns. The only sensitive method which can be used is microscopy. In extremely low concentrations, asbestos is identifiable in the electron microscope by its physical shape (tubular fibril), which is distinguish able from that of other fibers and particles. 8.3.3 Microscopy Methods for Asbestos At the beginning of this program, there were one optical and three electron microscopy methods for the determination of asbestos content. The optical microscopy method was established by the U.S. Public Health Service with the Asbestos Textile Indus try (8-1.) * to monitor ambient air quality for large asbestos fibers in areas of high concentration. In this method, an asbestos content with more than 90 percent fibers col lected on a membrane filter was examined at 400 or 430 diameters magni fication with phase contrast optics. The phase contrast rendered the background filter transparent and enhanced the viewing of the fibers. Only fibers greater than 0.5y in diameter and 5.0y in length were detected and reported. No fiber size degradation took place in this analysis. Two electron microscopy methods (developed by Battelle Columbus Laboratories(^-2) and Mt. Sinai School of Medicine(^-3)) existed * References appear at the end of this section. 8-6 HWBUI0002404 at the start of the program while the third was just being introduced (University of California, School of Public Health(8-4)). These methods are compared in Table 8-4. The Battelle and Mt. Sinai methods are char acterized by asbestos fiber size degradation to produce more fibrils. This size degradation was not desirable for this program, as one of the objectives was to ascertain the size distribution of the asbestos fibers in the brake emissions. On the other hand, the University of California method was desirable because of particle size preservation. However, the magnification range was found to be low, making it difficult to see the smaller asbestos fibers and fibrils found in brake wear debris. 8.4 BENDIX ANALYTICAL METHOD FOR ANALYSIS OF ASBESTOS IN BRAKE EMISSIONS 8.4.1 Criteria and Flow Chart Three important criteria had to be met by the analytical method used: the very small portion of brake debris used had to be re presentative of the much larger sample collected; the analytical steps could not degrade the particle size distribution of the asbestos fibers obtained; and the results obtained must be indicative of the actual as bestos content of the sample. These criteria were met by the analytical scheme outlined in Figure 8-2. 8.4.2 Analysis of Brake Emissions A detailed step-by-step sequence was prepared for the ana lysis of brake emissions. This information is included as Appendix U, The asbestos content of each sample analyzed is reported in Section 9 under Vehicle Test Results. 8.4.3 Representative Sampling and Low Temperature Ashing (LTA) After desiccating the samples to remove adsorbed water and weighing for material balance purposes, two representative sampling pro cedures were used. In the first sampling, 10 fractions, each approximately 2 to 5 mg, were used to accumulate approximately 20 to 50 mg of wear de bris (Figure 8-3). If the total sample weight was less than 100 mg, the entire sample was used. The large amount of carbonaceous material was responsible for the fluffy appearance of the wear debris. Figure 8-4 shows a scanning electron photomicrograph of a group of several fibers greater than 50 microns. Their ends were not visible and their exact length could only be obtained by LTA or by destroying the wear debris particle which, in turn, might have reduced fiber length. The LTA process is described in Figure 8-5. 8-7 HWBUI0002405 E-66-Z8-d 0.05 X 1.5 0 .5 0 X 40 N ot Known A sbestos W orkers U nion U OF CALIFORNIA 3,000X No Unchanged No 0.01% ' T able 8 -4 - T ra n sm issio n E le c tro n M icroscopy Methods P p H sbOi <pH co p~i OX t--1 03 oc PH #* U CM cd H i--1 CO nd QbOJ o *>'i P d cd o ON oO o i--1 XI X! oCO o CM oO *pH o CO CO bO iH d H H P de0)u P H P aO CM O !--1 Xi o1 nC3O t--1 r> rH V__' P O NO ao rH a P CO p QJ a; P O P CO pCoO QJ P CO < w P X! P W H o o o H <} o PQ <3- I--1 <D P qj i--i Cu t-H CO Bo O CO O oj o QJ O >-* O nd O (D bO o d <? cd rd o ON o o oP XI XI CO o o CM oO CO b0 /-*N d CO H rH d H 0) P PP o H pH iH O PO O rH r* O CM < P W 4-1 *iH P 'd 0) do do H QJ CO P i--1 P cd 3. do *H P Pd cd P do P o do 0) H BH T3 *H H P 4- 6 ou BP cd P P 0) O d d CO 03 P cd p 0J E U bo CO P P H H H H 0) H p e QJ d X P 4-1 Q Q d d Q H cd CO a aH QJ E *H d QJ 0) o H QJ P a abO N cd *rH N H od d N i-t H CO CO u CO cd p CO P QJ oB p CO d u 8 [ 11 1 1 1 i !; i ; ! j , i ' HWBUI0002406 P-82-99-3 SUMP SAMPLE (SAME AS FOR SURFACES SAMPLE) SURFACES SAMPLE (WEIGHED) i1 WEIGH REPRESENTATIVE FRACTION AIRBORNE SAMPLES 0.2 JU FILTER 8.0jU FILTER COMBINE (SAME AS FOR SURFACES SAMPLE) f LOW TEMPERATURE ASH (LTA) t WEIGH REPRESENTATIVE FRACTION t DISTRIBUTE ON 0.2/x NUCLEPORE FILTER OPTICAL MICROSCOPY: (GLASS SLIDE MOUNT) i COUNT (400X PHASE CONTRAST) DIVIDE ELECTRON MICROSCOPY: (CARBON SANDWICH) J COUNT (22,000X) RESULTS Figure 8- - Flow Chart for Particulates Analyses 8-9 HWBUI0002407 i: i Figure 8-3 - Representative Sampling for Analysis Figure 8-4 - Wear Debris Collected on 8p Nuclepore Filter 8-10 HWBUI0002408 < S-ZZl-S8-d [ F ig u re 8-5 - Low Tem perature A shing C/2 O o _i LU m oO LU CD < oz 2 < O 3) > DO X DC .o o < O C3 O O D O o h- < tl go in:U~ LU DC < x O < > > oo 1-0 0 DQ UJ o O Q ^5 m<> DC - 00 a LU Ic--n O DC Q. X o DC LU r-- DC O o. o < U_ X < < X DC g ^ > LU I-- o LU CD LU (3 > X o 8-11 I. V. HWBUI0002409 A series of tests was run to determine the organic content removal by the oxygen plasma on several types of samples. The removal rate varied according to sample type and state of subdivision: Sample Removal Rate mg/hr Carbon Rod Small Piece (Large Area) Large Piece (Small Area) Phenolic Resin Filter Paper Wear Debris 31 21 >15 40 4 Carbon rod was the standard recommended by the instrument manufacturer who claimed a removal rate of 90 mg/hr. This rate was never obtained. For the carbon rod, the phenolic resin, and the filter paper, the removal rate was dependent on surface area exposed to the plasma. The wear de bris, in spite of a large surface area, exhibited very low removal rates. Consequently, long periods were required for removal. The LTA process removed 15 to 30 percent of the sample weight, which, in turn, effectively increased the concentration of the asbestos content for counting purposes, as performed later. Figure 8-6 shows LTA-treated wear debris as distributed on a 0.8y Nuclepore filter. The large groups of asbestos fibers shown in Figures 8-4 and 8-6 are not typical; in almost all cases single fibers and fibrils exist and the larger groups are shown for illustrative purposes only. The second representative sample step is described in Sequence 7B of Appendix D. The appropriate LTA sample was taken in the following manner: approximately 10 fractions of approximately 0.05 mg each (material held on tip of fine spatula) were placed in a tiny platinum boat on a five-place analytical balance until 0.55 + 0.05 mg,weighed to the nearest 0.01 mg, was obtained. This material was then suspended in a liquid and distributed onto a 0.2y Nuclepore filter as outlined in Sequence 7B of Appendix D and shown in Figure 8-7. The manner in which the value of 0.55 + 0.05 mg was obtained is described in the following section. . 8.4.4 Sample Distribution for Microscopy The amount of material to be examined by microscopy was arrived at by trial-and-error. A 47 mm filter loading of 10.0 mg was too great, as the smaller particles were not seen or separated from larger particles. The sample shown in Figure 8-6 was prepared with 7.0 mg load ing and several areas of severe crowding or multi-layers of particles were seen. A further series of samples was made with filter loadings i. 8-12 I HWBUI0002410 (a) LTA Wear Debris on 0.8y Nuclepore Filter - 500X (b) Asbestos Particle Located in LTA Wear Debris - 5000X Figure 8-6 - SEM Photomicrographs of LTA Wear Debris Samples 8-13 HWBUI0002411 J, \:r: d\ I'; [ : t--.: Figure 8-7 - Uniform Sample Distribution onto Filter down to 1.0 mg. At the 1.0 mg level, the particles were sufficiently spread out to permit viewing of each particle by electron microscopy. However, to achieve consistently stable films, a further study was carried out (Appendix E) and it was found that good electron microscopy results could only be obtained by implementing the following changes in the ana lytical procedure: Use of a smaller grid opening (400 mesh screens) to give good support and higher thermal conductivity. o Lower loading on filter (0.50 to 0.60 mg) for more uniform distribution. 8-14 HWBUI0002412  A,second layer of carbon to effectively "sandwich" the wear debris for both strength of sample and thermal conductivity improvements. In early tests, wear debris was collected on both Millipore and Nuclepore filters. The smaller particles entered into the fibrous network of the Millipore filters as shown by scanning electron microscopy. On the other hand, the Nuclepore membranes readily exhibited the collected wear debris on a relatively flat surface. To avoid future qualitative and quantitative problems with fiber identification, Nuclepore filters were therefore selected for all further work in this program. 8.4.5 Selected Microscopy Methods 8.4.5.1 Initial Studies and Rationale A survey of the available analyses for asbestos particles in friction material wear debris or in ambient air samples was made. All methods found were satisfactory for total asbestos. All methods found were unsatisfactory for asbestos particle size distribution, as they all involved a particle size reduction step for transmission elec tron microscopy observation. Fiber bundles similar to those shown in Figure 8-6 did not permit sufficient electron beam penetration and good images could not be obtained for qualitative and quantitative observation when fibril ends were not present. In the very early tests, wear debris from a sample dynamometer was collected on the filter train of the 8.Op and 0.2p Nucle pore filters. Both filters were examined by a combination of optical and transmission electron microscopy. The following observations were made: When the total airborne sample was small, the particulate distribution between the filters was almost even. As the total sample (sum of debris on both filters) became larger, the larger amount of debris remained on the 8.Op filter. In effect,'it became a finer filter and less material passed to the 0.2p filter. The size distribution on the 8.Op filter varied over a larger range; the material on the 0.2p filter was finer and more uniform in size. Fibers were found on both filters; fibers up to 2.Op diameter and 50p long were located on the 8.Op filter, and up to 0.9p diameter and 3.6p long on the 0.2p filter. The electron beam could not penetrate the thicker fibers. Although many nonfibrous particles were checked for crystal linity by electron diffraction, very little crystallinity was found. 8-15 HWBUI0002413 Additional particulate analyses were carried out for the airborne samples collected from the inertia dynamometer tests described in Section 6. The filters were examined "as collected", without any sample redistribution onto a single filter. The following observations were made: o Asbestos fibers longer than 20 microns were obtained. o Very fine fibrils were extremely difficult to detect in "as-collected" airborne wear debris. Based on the above studies, it was deemed neces sary to employ two magnifications. To detect fibers greater than 5y, a magnification in the range 400 to 1000 diameters, was required. To detect the finer fibers and the ultimate fibrils, a magnification in the range. 10,000 to 40,000 diameters was required. The following optical plus transmission electron microscopy magnifications were selected: Transmission Electron at 22,000X Optical (Phase Contrast) at 400X Fibers Detected (y) Minimum dx1 Maximum dx1 0.03 x 0.10 0.50 x 35 0.50 x 1.50 10 x 100 8.4.5.2 Optical Microscopy Analysis for Asbestos The U.S. Public Health Service procedure outlined in Section 8.3.3 was modified for this program. The procedure(8-1) called for Millipore filters and the fluid used for phase contrast had a refractive index of 1.460. In the Bendix procedure with Nuclepore Filters, a fluid matching one of the refractive indices of Nuclepore filters at 1.585 was substituted. The procedure used is described in Sequence 8H of Appendix D. Figure 8-8 shows the microscopist behind the Reichert Zetopan as equipped with Heine phase contrast optics at 400 diameters. A typical photomicrograph (Figure 8-9) shows fiber among the remaining wear debris. 8.4.5.3 Transmission Electron Microscopy Analysis for Asbestos The University of California School of Public Health procedure, also described in Section 8.3.3, was modified for this program. A magnification of 22,000 diameters was used instead of the prescribed 3000 diameters because of the difficulty in seeing fibers below O.ly in diameter. I 8-16 HWBUI0002414 8-17 HWBUI0002415 The procedure used is described in Sequence 8B of Appendix D. Figure 8-10 shows the Hitachi Type HU-125E transmission electron microscope used in this program. A typical photomicrograph (Figure 8-11) shows asbestos fiber among the remaining wear debris. 8.4.6 Calculations From Microscopy Results 8.4.6.1 Rationale and Microscopy Count Sheet The ultimate asbestos fibril has been found to be circular in cross section(8-5). When the wear debris was spread thinly over the flat Nuclepore filter membranes, each fiber remained flat when viewed in the microscope, so that its projected diameter and length I ' Figure 8-10 - Transmission Electron Microscope 8-18 ! i t. HWBUI0002416 could be measured. The1 volume of each asbestos fiber was readily deter mined from the following formula: ird V= x where V = volume of material d = projected diameter measured = length measured When two or more fibrils make up the larger fiber bundles, a larger-thanactual projected diameter results. Kven fibers with elliptical cross section were assumed to lay such that their largest diameter was parallel to the filter. In all cases, the volume calculated from the projected diameter and length was thus on the high side. For each fibril or fiber found, it was necessary to determine its dimensions and calculate the volume of asbestos found. Because of the large number of fibers, this procedure became tedious. To simplify fiber sizing, a Microscopy Count Sheet was developed .(Figure 8-12). The crosshatched areas are the boxes which did not follow the definition of a fiber: the length was required to be three or more times greater than the diameter. Each fiber found by the appropriate microscopy method was classified into the proper box containing the projected diameter and length of the fiber in question. A typical microscopy count for a sample generated in this program is superimposed on Figure 8-12. 8.4.6.2 Sample Calculation A typical calculation is carried out in the following manner. For each box, an average volume is calculated from,the formula: where V = average volume for box ~2 2 d - average of diameter for box = average length for box 8-19 HWBUI0002417 These values are given in Figure 8-13. Then it is necessary to multiply the number of fibers given in each box in Figure 8-12 by the volume given for the corresponding box as shown in Figure 8-13. The typical microscopy count shown in Figure 8-12 is used to calculate asbestos content in the following manner. The fibers of each group are totaled as shown in Table 8-5. The total number of fjibers per group (n) is then multiplied by the average volume per group (V) to obtain the product (n x V). All products are then totalled to arrive at the total volume of asbestos detected. Since only a fraction of the total area is ac tually scanned, the volume observed must be normalized to the entire area containing wear debris: Total Fiber Volume (by EM or OM) . . , , s ______________ Total Area of Filter = Volume o axne cc x 0 Fieids Counted) x (Area of a Field) v ^ __________ 990 mm2 nX X ZF x (Area of Field) I j i. i 8-20 HWBUI0002418 MICROSCOPY COUNT SHEET Vehicle Test: Schedule: Collector: Sample: Date: / A-8- SAstZLftfe &&n-r ScW/e 7- Z./- 7Z. fteAf/e Total Sample Weight: /SS'. / LTA Weights (I/F): 38.8 / 3/. 2- Count Weight: O .(oZ. Comments: mg mg mg EM Field Size: 3.0 x 3,On No. of Fields: ZOO 33-fiBezs 2 2 ,000X 400X Date Counted: OM Field Size: No. of Fields: 7-7S-1Z- /M-MW Mt W-m* /` 8i8 x 8&U -- MV- #*'**'**r'*,r. > /`f' JZ 00 Microscopist: Q Figure 8-12 - Microscopy Count Sheet 8-21 HWBUI0002419 FIBER VOLUME CALCULATION x io ,ZCC V= a 786 X10 ,ZX dz xJL xooo'zz xoov Date Counted: OM Field Size: No. of Fields: ____________ 88^ x 88u Microscopist: Figure 8-13 - Fiber Volume Calculations 8-22 HWBUI0002420 * * * * ASRFSTOS ANALYTICAL RESULTS * * * * F ig u re 8-14 - Computer P rin to u t o f A sbestos A n a ly tic a l R e su lts LLI Ufrt OC 00 -4- cn O CO o fll <NJ co <E <E l/> (S> 5! X u.- C n*i m oo K Uj a: lli 1113W>_| C _J , i-- n. KO <_j Z UJ UJ < lli I _f (/): i U mo. O in in o o in o in in r-- O' c -o in vO h- -o a- .n f-- r- vj- m m 0000m00*-i0rsj00c00-*0'i'c0m00000in0 0 0in00>t'0>0n.000000 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO O'OinuMAininin^^'f-t^ f\j <\l <\| Csl i -< o o o o o o- O' a (7> c I I I I I I I I I I I I l i > l l l I I I I I 1 i t < l l i 1 I0 o 0 o o LU UJ LLI UJ LLI UJ UJ UJ UJ UJ UJ UJ in hi in in iii in III IIfII III III III II, III III UJ UJ UJ UJ UJ ii in in in ii. in iii UJ UJ UJ UJ UJ 1 t o - "<0 0 1O r- tn *o -O CD o -J H H >0 H *4 M m >t w o r OfflO-t 4- >o -- in -c nj m -t -O 4- -j o f- in >o CO o -t -I H . -h <\j m 4 o f** >0 ^o H4' m M** >c (<1 J- w o f** in sC COO't-*^ o ^ -i m m 1 HOMnO't locoo't-wo %r *o --t nj <n st HOI* -- in <\i >c cn v4O- 0^0 o o 00000000000000000000000000000000000000000.0000 8-23 HWBUI0002421 Table 8-5 - Sample Calculation ELECTRON MICROSCOPY Fiber Fiber Number x Volume = nxV (n) (V x 10-15 cc) (x 10 cc) 1 0.20 0.20 3 0.50 1.50 5 1.31 6.55 12 3.50 42.0 1 11.3 11.3 1 10.2 10.2 9 27.1 243.0 1 88.0 88.0 I n x V = 402.7 Total Volume.EM Z (n x V) x (Area of filter)* 3 (No. of fields) x (Area of field) 402.7 x 10~15 990 x IQ6 y2 200 X . , 2 3x3p = 202 x 10~9 cc OPTICAL MICROSCOPY Fiber Fiber Number x Volume = nxV (n) (V x 10"12 cc) (x 10-12 c 9 0.982 9 3.19 2 13.5 5 12.8 1 29.4 1 33.2 4 76.4 cr> CM 8.92 28.7 27.0 64.0 33.2 305.0 Z n x V = 496.22 Total Volume,OM Z (n x V) x (Area of filter) (No. of fields) x (Area of field) 496.2 x 10~12 990 x IQ6 pi2 200 X 2 88 x 88 y = 316 x 10 9 cc Total Volume (Total Volume)^ + (Total Volume) = (202 + 316) 10 9 cc = 518 x 10~9 cc Total Weight 518 x 10,-9' x 2.50 ^ x 1000 ^ cc gm 1.33 x 10 ^ mg Asbestos Percent = 0.00133 0.62 x 312 388 x -nn _ n 100 `173/ P-8 2-99-3 8-24 HWBUI0002422 For the electron microscopy analyses at 22,000 diameters, the area of a field was 3 x 3 For the optical microscopy analyses at 400 diameters, the area of a field was 88 x 88 y^. jn electron microscopy analysis, 402.7 x 10-^ cc was found for 200 fields: Total Fiber Volume (by EM) 402.7 x 990 mm x 1in0"15 cc (200) x (3 x 3 y ) = 202 x 10 ^ cc Total Fiber Volume (by EM) Specific Gravity 2.56 = Fiber Weight (by EM) 202 x 10 cc X 2.56 -2^ = 0.00052 mg Fiber Weight (by OM) 0.00081 mg Combined total for all asbestos found 0.00133 mg. This amount of asbestos calculated is for debris which had been slightly concentrated in the LTA process. Thus the asbestos percent, as related to the original material, is obtained by the following expression: Asbestos Percent Total Asbestos Found x LTA Ratio y x 100 Count Weight For the example given in Figure 8-12, the percent asbestos for the sump sample of the right rear brake collected for the A.B. Baseline Schedule of vehicle test 1 is found: Asbestos Percent 0.00133 0.62 31.2 38.8 x 100 = 0.173 percent d:|' r |: I: f 8-25 HWBUI0002423 8.4.6.3 Computerization of Calculations and Particle Size Distribution The calculation of asbestos content was lengthy and thus susceptible to computational errors. Moreover, over 150 such calculations were anticipated for the program. A computer program was therefore developed in conjunction with the Count Sheet given in Fig ure 8-12. The uppermost horizontal row and the extreme left vertical column of markers are used to identify the box with its corresponding volume. The circled number is the number of fibers found which corres pond to that volume. Unlike the manual calculation, grouping of boxes in threes is not required and for this reason the computer calculation is slightly different and slightly more accurate than the manual calculation. The computer program is given in Appendix F, Figure 8-14 shows the computer printout for the sample calculation given in the previous section. The computer printout gives the asbestos percent = 0.179 percent, which compares with the manual calculation of asbestos percent = 0.173 percent. 8.4.7 Possible Sources of Error The possible sources of error for the sampling and analysis of particulate wear debris are listed in Table 8-6. The possible sources of error associated with sample pro cessing were minimized through the use of statistical methods. For sample selection, two stages of representative sampling took place: weighing prior to sample low temperature ashing, and weighing for distribution onto the filter. Uniformity of sample distribution was checked by low power optical microscope observation immediately after distribution. This was further checked by the microscopy results as shown in Figure 8-12. Just under the number of fields reported for the electron microscopy re sults, the notation used shows 15 fibers found in 100 fields examined on 2 grids from one side of the filter, while 18 fibers were found for 100 fields examined on 2 grids from the other half on the filter. Also, for the optical microscopy results, 17 fibers were found in 100 fields on a sample taken from one side of the filter, while the other side yielded 13 fibers. These distributions were typical. As indicated in Appendix E, a series of tests was run to insure good samples for electron microscopy. Membrane shrinkage did not occur and membrane curling was minimized. No observations in the elec tron microscope were taken near membrane edges. The sample distribution was finalized after more than ten different sample weights were examined and then several different samples at the same sample weight gave repro ducible films in terms of stability and counting. 8-26 HWBUI0002424 Table 8-6 - Possible Sources of Error SAMPLE PROCESSING Low Temperature Ashing Sample Selection Extent of Ashing Weighing Sample Distribution Sample Selection (From Vial) Sample Weighing Characteristics of Sample and Variations in Size Distribution Loss through Filter Filter/Funnel Interface Effects Sample Handling Air-Borne Dust Contamination Sfatic Electricity Air Currents MICROSCOPIC METHODS Sample Preparation Sample Identification Sample Redistribution during Cutting Membrane Stretching, Shrinkage, or Curling Sample Counting Depth of Field Width x Length Estimates Large Fiber Contributions Fiber Identifications (Asbestos Versus Olivine) Magnification Variations Statistics O o<J>> cvi <9 0. i 8-27 HWBUI0002425 Perhaps the largest sources of error occur in the sample counting procedure, for the following reasons: Width and length estimates of fibers and their insertion into the proper box in Figure 8-12 (a borderline diameter such as 0.05 y which is between lines 12 and 13 gives an average volume difference of approximately 3). Fiber identification difficulty, especially in optical microscopy (asbestos or olivine). Number of fibers, found and their statistics. These errors were minimized by counting a large number of fibers and/or fields, whichever came first. The target for fiber count ing was 100 fibers in 100 fields in both electron and optical microscopy. This target was rarely obtained because of the very low content of as bestos present in the sample. A maximum number of fields to be counted for low fiber concentrations was also established (200 EM fields and 300 OM fields), to avoid unnecessary delays for very low concentrations. The largest effect found on the asbestos analysis was the occurrence of a large fiber. Its volume contribution significantly in creased the contribution, in either microscopy method,by 100-200 percent. This could give up to a 50 percent change in the final calculation. If additional fields were counted or a replicate analysis made, a comparable fiber was usually not found. The large fiber was thus considered an "outlier" and was not used in the calculations. In some instances, more than one large fiber was found in a sample. Replicate analyses did not yield similar fibers and the asbestos emissions result was significantly different as shown in Table 8-7 for the following sample: Vehicle Test 1, Detroit Traffic Schedule, right front brake, sump sample (Example 1). The precision in the asbestos analytical results can be estimated by considering the replication data given in Table 8-7. The percent deviations range from 0 to 79 percent; the average is +27 percent. On the other hand. Figure 8-15 shows the variation of the maximum devia tion (given in Table 8-7) with the asbestos percent found. The data show that the maximum deviation for results with asbestos contents above approximately 0.10 percent have a maximum envelope at approximately 20 percent, with averages at 12-15 percent. The maximum deviation increases for asbestos content below 0.10 percent, where errors for the small amounts of asbestos emissions become less significant. 8.5 ANALYSIS OF GASEOUS EMISSIONS 8.5.1 Method of Analysis The gas collection system (described in Section 5.4) was designed to collect and separate condensable and noncondensable gases from the carrier air. Because of the high vapor pressure of the noncon densable gases in the activated-charcoal traps, storage took place at 8-28 HWBUI0002426 Table 8-7 - Extent of Error in Asbestos Emissions Analytical Results EXAMPLE NUMBER SAMPLE ASBESTOS PERCENT AVERAGE MAXIMUM DEVIATION PERCENT DEVIATION Vehicle Test 1 0.247* 1 Detroit Traffic Right Front Brake 0.631 0.186 0.355 0.276 Sump Sample +79% Vehicle Test 1 0.137* 2 Detroit Traffic Right Front Brake 0.085 0.091 0.107 0.022 Surfaces 0.116 +21% Vehicle Test 1 0.052* 3 Detroit Traffic 0.025 0.056 0.031 Right Front Brake 0.085 Airborne 0.068 +55% Vehicle Test 1 0.432* 4 A.B. Baseline 0.554 0.493 0.061 Right Rear Brake Airborne +21% Vehicle Test 3 0.013* 5 Detroit Traffic 0.034 0.023 0.011 Right Front Brake Surfaces +48% Vehicle Test 3 0.057* 6 Detroit Traffic 0.045 0.051 0.006 Right Rear Brake Surfaces +12% Vehicle Test 3 0.087* 7 Final Baseline 0.061 0.074 0.013 Right Front Brake Surfaces +18% Vehicle Test 3 0.512* 8 Final Baseline 0.386 0.470 0.084 Right Rear Brake 0.511 Sump +18% Vehicle Test 3 0.237* 0.228 0.010 9 Final Baseline Right Rear Brake Surfaces + 5% Cumulative for Vehicle Tests 10 1, 2, and 3 Clutch Surfaces 0.047 0.047 0.047 0.000 0 P-82-99-3 *Result reported in Tables 9-16, 9-17, and 9-18. 8-29 I HWBUI0002427 M AXIM U M D E V IA TIO N (%) 0.10 0.20 0.30 0.40 ASBESTOS PERCENT FOUND 0.50 0.60 Figure 8-15 - Variation of Maximum Deviation of Asbestos Analytical Results with Asbestos Percent Found -70C with solid carbon dioxide packed around the traps. The analytical procedures are given in Appendix D. The primary method of analysis was gas chromatography for both the noncondensable and condensable gases. 8.5.2 Gas Handling System The gas handling system used for the gas analysis is shown in Figure 8-16. The system included various sections designed to perform the following tasks: Gas inlet and fraction traps - used to separate air from the gases of interest. Toepler pump and gas buret - used to transfer the gases and measure their volume. Gas chromatograph transfer U-tube - used with gas buret and Toepler pump to inject a sample into the gas chromatograph (Figure 8-17). 8-30 HWBUI0002428 Figure 8-16 - Gas Handling System 8-31 HWBUI0002429 8-32 -F ig u r e 8 -1 7 Gas H a n d lin g S yste m and Gas C h ro m a to g ra p h ii l-i- HWBUI0002430 I o Calibration gas storage volumes - used to store pure gases or known mixtures which were used to calibrate the gas chromatograph. The noncondensable gases were to be handled first. The complete step-by-step procedure is described as Sequence 10A in Appen dix D. The condensable gases were then handled and their procedure is described as Sequence 10B in Appendix D. 8.5.3 Calibration of Gas Chromatograph Prior to any analyses of samples generated by the vehicle, it was necessary to check out the gas handling system and calibrate the gas chromatograph. The gas handling system was set up in good working order. A calibration gas from one of the gas storage volumes was expanded into the high-vacuum manifold. By means of the Toepler pump, all of the sample could be collected and measured in the gas buret. The sample was then introduced into the G.C. transfer U-tube. The helium flow of the gas chromatograph was diverted so that the volume of gas in the gas chro matograph transfer tube could then be introduced directly into the gas chromatograph. The volume of gas introduced was calculated and the gas chromatograph detector response was determined. For a series of different volume sizes added to the gas chromatograph, a calibration curve could be made. Figure 8-18 is the typical detector response calibration curve obtained for carbon dioxide. Similar curves were prepared for other gases expected. The retention data for the column used were obtained in the following manner. A gas mixture containing air (nitrogen and oxygen), carbon monoxide, methane, carbon dioxide, water, and propane was prepared exterior to the system. The mixture was stored in a separate volume. Small samples could be introduced through the gas inlet of the gas handl ing system. The sample was collected in the gas buret and transferred to the gas chromatograph. Several different temperature programs for the gas chromatograph column were evaluated in order to obtain a reproducible program which would elute all gases predictably and well separated. Figure 8-19 shows the typical retention time calibration for low molecular weight gases which may be present in the noncondensable gas traps. The components were separated on a column 6 feet long and 0.125 inch O.D., packed with Porapak Q. The column was initially.cooled to less than -70C with dry ice. This condition was needed to separate N2, O2, and CO. The column temperature was then raised rapidly to 30C to elute CH4 and CO2. The column was then heated to 180C at 10C/min to elute water, C3's, and C4*s. 8.5.4 Problems and Interferences In Section 6.2.3, the integrity of the rotating seal was discussed along with its leak rate as measured under both static and dynamic conditions. It was impossible to eliminate the leak completely. Under static conditions, the flow through the seal was Vl37 cc/min. This 8-33 HWBUI0002431 1200 1000 DETEC TOR TEMPERATURE: 180C DETEC TOR CURRENT: 180 mA <fi 800 X CZf/-3 D Oo 600 cc LU CO 400 200 0 0.20 0.40 0.60 0.80 1.00 NUMBER OF MOLES (X 10`5) Figure 8-18 - Detector Response for Carbon Dioxide 1.20 r> CoT>> cOoi a. improved to about 50 cc/min under the most optimum conditions. However, in all cases there was a gas flow through the rotating seal into the emissions collection system. The rotating seal was downstream of the gas drier and car bon dioxide removal bed (described in Section 5.2.1). Thus the air entering the rotating seal contained the normal atmospheric components including carbon dioxide and water in addition to oxygen, nitrogen, and argon. In heavy traffic, measurable amounts of carbon monoxide and nitro gen oxides were also expected to be present. The large amounts of carbon dioxide and water obtained proved to be the major problem when the gases were collected at low gaseous emission concentrations such as in the 8-34 HWBUI0002432 DETECTOR OUTPUT P-82- 99-3 0 DRY ICE HEAT FAN TO ON 32C -70C -0C 30C 6 12 30C TO 180C @ 10/ mm * Figure 8-19 - Typical Calibration Chromatogram for Low Molecular Weight Gas 18 8-35 Il: HWBUI0002433 Detroit Traffic Test. During the later stages of the program, attempts were made to circumvent the carbon dioxide and water problems by selec tively collecting gases only during periods when gas emissions were ex pected to be adequate, such as the fade tests. 8.6 REFERENCES (8-1) S. G. Bayer, T. A. Brown, and R. D. Zumwalde, "Equipment and Procedures for Mounting Millipore Filters and Counting Asbestos Fibers by Phase Contrast Microscopy," H.E.W. Public Health Service, February 1969. (8-2) R. E. Heffelfinger, C. W. Melton, and W. M. Henry, "Develop ment of a Rapid Survey Method of Sampling and Analysis for Asbestos in Ambient Air," Interim Report to National Center for Air Pollution Control, July 1970. (8-3) W. J. Nicholson, A. N. Rohl, and E. F. Ferrand, "Air Pollu tion in New York City," Presented to Second International Air Pollution Conference, Washington, D.C., December 1970. (8-4) J. Murchio, Private Communication, 1971. (8-5) K. Yada, "Study of Chrysotile Asbestos by a High Resolution Microscope," Acta Chrystallegraphica, Vol. 23, 704 (1967). SECTION 9 VEHICLE TEST RESULTS 9.1 VEHICLE OPERATIONS 9.1.1 Test Scheduling and Problems Encountered Early in the program, it was estimated that approximately 38 days would be required to complete each vehicle test. This estimate was based on the optimistic assumption that the vehicle measurements procedure (take emissions samples, measure wear, inspect system, and replace worn parts) would require 1.5 days. Tables 9-1, 9-2, and 9-3 are the actual versus estimated test schedules for vehicle tests 1, 2, and 3, respectively. Vehicle test 1 took 56 days because of the problems listed. Moreover, the mea surements sequence actually required approximately two days. In addition, approximately one-half day was lost per schedule because of rain (the summer of 1972 was the wettest summer in recent history). The sum of the additional one-half day for measurements plus the one-half day for rain loss gave an average of an additional one day per schedule, or an additional seven days per vehicle test. Thus, the addition of seven days to the estimate of 38 days gave a value of 45 days, achieved in vehicle tests 2 and 3. Near the end of vehicle test 3, there was a very good chance of finishing in approximately 42-43 days. However, melting snow and freezing rain made the driving conditions hazardous at times. Thus two days were lost, and the program was completed just prior to the even more wintry weather which began in late November in the Detroit area. Some of the problems encountered in the program and their solutions can be summarized as follows: During the 10-stop fade schedule of vehicle test 1, the hardened steel ring which rotates aginst the Teflon seal slipped from its anchored position after shearing the locking pins. The leak thus created was detected immediately after the fade stops and prior to the reburnish. The ring was reanchored and no further problem was encountered during the 15-stop fade or any further test schedule. During the Detroit traffic schedule of vehicle test 1, water was found in the right rear brake collector. This was pulled in through the rotating seal. This problem was not encountered again, as the vehicle was not operated during heavy rain. 9-1 HWBUI0002435 Table 9-1 - Actual Versus Estimated Test Schedules for Vehicle Test 1 TEST SCHEDULE ESTIMATED TIME (DAYS) ACTUAL TIME (DAYS) START FINISH TOTAL REMARKS OR PROBLEMS Burnish Measurements" 2.0 1.5 JJ 5/1 5/10 8 Heaters shorted. Invertor failed. Loose collector screws. Slow silicone cure. A. B. Baseline Measurements 3.0 1.5 4 3) 5/11 5/18 6 Detroit Traffic Measurements 12.0 1.5 U J 5/19 6/13 20 Burned-out heater. Water in rear collector (due to rain). Replaced rear seal. Burned air line. Replaced filters. Rain - test stopped. Slow silicone cure. 10-Stop Face Reburnish Measurements 1.5 1.0 1.5 4.0 6/14 6/21 6 Rain. Slow silicone cure. Burned-out T.C. Measurements - fronts only. Rotor ring slippage. A. F. Baseline Measurements 3.0 1.5 /, c; 6/22 6/29 6 Rain. 15-Stop Fade Reburnish Measurements 1.0 1.0 1.5 6/30 3.5 7/5 5 Final Baseline Measurements 3.0 1.5 /, ^ 7/6 7/12 5 38.0 56 Take emissions samples, measure wear, inspect systems, replace worn parts. 9-2 i HWBUI0002436 T able 9-2 - A c tu a l V ersus E stim a te d T est Schedules fo r V e h ic le T est Take em issions sam ples, m easure w ear, in s p e c t system s, re p la c e w orn p a rts . CO 2 W XI PQ O CM P4 P-i P3 O CO p <lj sw P <J CO HO <>5s H P PQ 4aOHJ O mcd 'dd T--| cd a cd e rH A d +0J) H C4 "C--J_ H 0) > wa < m CM rH r^ CO rH CM 00 nd d a cd i--1 cu a) n aH CJ cd \ P$ H Oh CM X) 00 Oh U H rH 2 u rH 4rHH CO a) C3 H Od d Xcd u CdO pq 4oJ CO <D a d d 4-1 U 4-1 i--1 cd Q* H 4J do CO V-l iH O a Hci u O P -H O CHi 4aJ) m |m I <3- CO rH Oh co --l CM r^ OW W >H H <J <5 P ^ '-- M HW cn s WH &H w H !=> ca Q WW H EC U W m CO o CM in om CO i-l in CO rH om CM iH rH * CO 4-1 CJ a) QJ U XI d CO CO aH u cd d SH d PQ a) CO d 4-1 a rH a) a) CO cd du PQ cj CO cd PQ d 2 <J CJ H CO 4-4 4-1 4H a cd d JH H ud 4- d H CO Ou cd d 4-1 2 d P hO CM O iH CM CO r^. 00 CO CO Ch o m in m <r CO <r mom i--( rH rH o in CO rH .oOm rH rH rH o in CO rH O 00 CO CO 4-4 d CJ -a X a) cd CO S FH H. d CJ H p. U d o d CO 4-1 X cd CO d d 1 P3 2 o rH d CO C! 4-1 H CJ rH d dg CO d cd u pq 3 CO cd d 2 c d -d X cd CO H CJ cu u od 4-1 X CO d 1& m rH CO 4-> a d d d CO cd d d CO d 4-1 H d r--1 d d CO d cd u PQ d CO rH cd cd d CJ H i 9-3 HWBUI0002437 T able 9-3 - A c tu a l V ersus E stim a te d T e st S chedules fo r V e h ic le T est Take em issions sam ples, m easure w ear, in s p e c t system s, re p la c e w orn p a rts . P H cd U CO bO T3 P c0 H O NH prl cd CD CD UO PH P <H3 co O >} H P W P MCO SMP I HO H P <: <H2 CO P W CO H <2 ^P w H CO w CM On 0I--0I as m CO om CM iH 00 CO CM CM -- Oh O CM Oh CM CM Oh Oh in m co om co t--\ om CM rH vO CM O rH CM O i--1 o -d" mom rH rH iH LO CM i--I CM o rH m om CO iH co m co oom rH iH rH Oh 1--I ?--I m 'd- om CO rH m <- 00 CO W ,_q E-i P CO P WW g-i P Cj> CO co 4J cs Q) s 0) p CO CO H cd 0) J-l S n P cu CO p 4-1 rH a tH a) <D G CO cd ua) PQ P co cd P a) S <2 H CO m 4-> 4H a cud 0) G H 0) u 4-1 p H CO uO cd a) 4J S 0) P a) rd ,P cd CO P *H P p oP 4- P CO 0) 1p o rH CO 4-4 P a) G CD P CO cd 0) <D CO P 4J H P rH 0) <D g CO a) cd H PP CO cd p 0) s <2 CO 4-1 <D P T3 X a) cd CO G P H a) P J-4 PU P o P CO 4-1 P CO 0) cd a) m1 p 2 rH 0) CO P +J H P H a) 0) G CO 0) cd pP CO rH cd cd CD Ps H P 9-4 HWBUI0002438  During the final baseline schedule of vehicle test 2, pressure was lost for the rear brakes and brake fluid was found in the collector. The combined sump and surfaces sample, for the eleven completed circuits, was salvaged by filtering and re peatedly washing the particulates to remove the brake fluid. The linings were "destroyed" as the fluid could not be removed completely. This problem was not encountered again. 9.1.2 Mileage Accumulations Table 9-4 lists the mileage accumulations for each schedule. The A.B. Baseline, A.F. Baseline, and Final Baseline were run according to the same schedule, repeated at different times within the overall vehicle test program. The average baseline mileages varied from 343 to 348 miles and the maximum variation between vehicle tests was +5.7 percent. This range for the baseline schedules could also be compared to one-fourth of the Detroit traffic test for which the average was 352 miles (1408 v 4) and for which the overall variation was +3.8 percent. The overall vari ation for the total mileage accumulation of all three vehicle tests was very low at +2.4 percent. These mileage variations were primarily due to the varying times taken to reach the Detroit traffic circuit. Normally only three round trips were required to obtain the twelve circuits. Rain, seal failure, thermocouple malfunctions, and other minor difficulties which required a premature return to base added to the minimum mileage possible. 9.1.3 Reporting of Vehicle Test Data The data for each vehicle test are accumulated in a separate volume which includes a section for each test sequence: Burnish; A.B. (After-Burnish) Baseline; Detroit Traffic; 10-Stop Fade; A.F. (AfterFade) Baseline; 15-Stop Fade; and Final Baseline. Each section includes the following data sheets: Vehicle Log Sheet; supplementary vehicle data sheets such as Detroit Traffic data sheets; Wear Sheet - Front Axle; Wear Sheet - Rear Axle; Collection of Brake Emissions (Right Side); Col lection of Brake Emissions (Left/Control Side); and Test Summary Sheet. The complete list and examples of each type of data sheet are included in Appendix G. 9.2 PERFORMANCE OF EMISSIONS COLLECTION SYSTEMS 9.2.1 Rotating Seal Life The success of the vehicle tests depended heavily on the proper operation of the rotating seals. Table 9-5 is a running log of rotating seal mileages (life) obtained during the vehicle test program. The original intent was to change the rotating seals at the start of each test and prior to the 10-Stop Fade. Both premature seal failure and the unexpected fact that disastrous rotating seal wear did not occur during the fade led to abandonment of that plan. 9-5 HWBUI0002439 Table 9-4 - Summary of Vehicle Mileage Accumulations Schedule Test Miles Burnish 1 2 3 A.B. Baseline 1 2 3 Detroit Traffic 1 2 3 10-Stop Fade 1 2 3 A.F. Baseline 1 2 3 15-Stop Fade 1 2 3 Final Baseline 1 2 3 Totals 1 2 3 174 127 100 328 348 368 1439 1430 1355 47 53 58 356 339 347 47 79 52 352 326 352 2756 2702 2632 8090 Average Maximum Miles Deviation Percent Deviation 134 34 348 20 5.7 1408 53 3.8 53 6 347 9 2.6 59 20 343 17 4.9 2697 65 2.4 9-6 HWBUI0002440 T able 9-5 - R unning Log o f th e R o ta tin g S eals M ileages O btained in th e V e h ic le T est Program In clu d e s 393 M ile s from Test 1 CM * 9-7 HWBUI0002441 Table 9-6 summarizes the rotating seal life in miles ob tained for each seal during the three vehicle tests. These data were used to obtain a maximum average life for each rotating seal. collector. follows: Seven rotating seals were used on the front disc brake They were replaced at different mileage accumulations as Two seals failed after 12 and 329 miles. Three seals survived 1965, 2059, and 2664 miles and were replaced as worn, but not failed. Two seals were used for 815 and 404 miles and were replaced although they were not completely worn or failed after vehicles tests 1 and 3, respectively. collector. follows: Eight rotating seals were used on the rear drum brake They were replaced at different mileage accumulations as Three seals failed after 62, 475, and 1483 miles. Four seals survived 1233, 1515, 1315, and 1248 miles and were replaced as worn, but not failed. One seal survived 751 miles and was removed in good condition after the third vehicle test. 9.2.2 Operating Temperatures 9.2.2.1 Normal Brake Stops Throughout the vehicle tests, it was necessary to monitor the temperatures at all four wheels. The driver was instructed to take specified actions when greater than desired temperature differences occurred. For example, during the Detroit Traffic Circuits, the maximum normal operating temperature was 330F before proceeding. These cool-down periods were seldom necessary. The effectiveness of the thermal control aids attached to the right front brake described in Section 5.5.3 is shown in Table 9-7. The AT = 200 to 250F shown for the initial burnish was reduced to a AT = 115 to 160F. Most important was the reduction of the right front brake operating temperature from about 400F to about 300F. Since approximately 90 percent of the test miles were driven on the Detroit Traffic circuit, it was necessary to monitor the temperatures obtained and maintain test reproducibility. Although the temperatures did vary slightly, the normal operating range for the right front brake was 260 to 300F while the normal operating range for the right rear brake was 170 to 200F. The brake shrouding did not present any difficulties on hot days. Table 9-8 illustrates the brake tempera tures measured during four consecutive Detroit Traffic Circuits on a hot 9-8 HWBUI0002442 VEHICLE TEST 1 2 3 Table 9-6 -- Rotating Seal Life (Miles) SEAL INSTALLATION 1 2 3 4 5 6 5 (cont'd) 6 (cont'd) 7 8 Maximum Expected Life (miles) FRONT DISC BRAHE FI 2059 F2 815 F3 1965 F4 392 F5 12 F6 (393)* F6 2664 F7 404 2059 1965 Average: 2664 2229 REAR DRUM BRAKE R1 1233 R2 1515 R3 475 R4 1483 R5 (744)* R5 1315 R6 62 R7 1248 R8 751 1233 1515 1483 Av1e3r5a9e: 1315 1248 * Continued usage into Vehicle Test No. 3. HWBUI0002443 T able 9-7 - Brake Tem peratures D u rin g B u rn is h T ests (F ) (V e h ic le T e st 1) Eh < O *rl 4-1 cuod bO UCO soW O to i--I oLO om COo o LO o o CO LO lo r-- m co X) 4- XaJ) P3 bHO O^ PP CO Eh < a O H 4-J cd U0 -u bO 44 *H <D 4-4 PC O O S Wo Xd a) 4-1 Xl p^ bO O H M P X! CO mom in o st M CM M o oo LO CM O OJ CM CM O vo ooO o st vO St CO CO lo in LO o O O st H 40 CM CM CM CM 1--I O m LO O CO T-- 1 i--4 CO O O oo LO LO LO CO T-- 1 1-- 1 O O LO O O O 00 T--1 vO o 00 O CM CO CM CM CM CO CQO M < P o P H EE! P CO O 1--1 a p P 1--3 P <q PQ S P PW <P J--1 H Eh MO 525 P M CO CO CO aPp ooo 4-1 4J 4-1 CO CO CO ooO CM st VD UUU 0) a) <D 4-1 4-1 4-J 4H 4H 4H <<< K K CPO <M o p EH P O u Pp CO 1--1 3 sw pp PQ Eh 2P o Eh p h4 wP p K -JC * CO CO CO CO CO CO P p PPppo o o o o o 4-1 4-1 4-1 4-1 4-J 4-1 CO CO CO CO CO CO Oo o o O O o VO i--! St vO 00 1--1 1--1 M u 14 }4 u 0) CD at <D CD 0) 4-1 4-1 4-J 4-J 4-J 4-J 4-4 44 44 44 44 44 < < < < < < i--1 cd Q) CO i--cacdo)I B3 u XJ >4 uo cd OuJ XJ V-l cd ua) P S 4-1 O Q0)) P <D & U P 4-1 or--I c H LCHd *4 co XCJO Jo rl cd cd cd XJ o Q) --i X) XI o 3 uo XJ 4-1 u CD S CD P o CO Pu O 44 4-1 --1 O CO cd B 4-> 4-1 u CO <D cd a> p 4-1 Eh H CO *** ** K 9-10 HWBUI0002444 Table 9-8 - Brake Tem peratures D uring D e tro it T r a ffic C irc u its (V e h ic le Test 2 - A .B . B a se lin e ) o co H ooooo o- O o o O rH CM CM CO ooOoO 00 CO CM rH CO O o CO oco o CO o CM 1 o rH u ol M-l S a) u hJ o ooooo CM in ooOOO m i--i <X> rH Mi--OI sO rH sO rH ooOOO SO p- p SO rH rH rH rH rH oOooO so 00 00 p m rH rH rH rH rH O 00 rH 1 O rH a) U Td 4odo 3o ,G CO o rH rH o oOoOO I 1m r-. 00 00 os i--i t--l -- rH I--1 ooOoO oos OS 00 Os rH CM rH rH rH oOOoo sO i--1 rH c > prH CM CM CM rH CM 1 O rH O MT H o OoOO oOOoO oO oOO rH in t--1 rH 1--1 r--1 CM t--1 Mf 1--t rH i--1 rH CO CM m rH i--1 rH rH rH O CO sr i--1 i--l rH rH rH 1 O rH j-i ol 4-t B d) G hJ O O i--1 o OoO oO m SO SO m OS OoOoo os 1-- P- sO SO ooOOo rH CO 00 P- so CM 1 i--i rH rH rH rH rH !---1 rH rH rH CM 1--1 rH rH rH O m rH /-\ Td <U 4-J Td ,G P H ^ Pd ,G CO s./ 00 oco oOOoo OOOOO OooOo co op- r-* O O 00 CO CO rH rs OO rH rH CM rH rH 1 co co oCM CM CO CO CM CO CO CM CO CO CO CO CO CO so GQ) P, H gh rO <U O in oo oo os os OSOCMCM-vt 00 00 CO CO ovot- omo O^ iC--T1\ iC-T1i i(-T1i o* >s O (D 03 CM HaH q <S-Ho P *-> mo cm in so so vo aa CO o<or CO CO o o co os <tc)u0 03 a) G u OG O G pj O rH <J U H cd M-t o M-t o td u E-l Td Td Uu cd Td Td id cd 5uuu& Td H cd cd TJ o jd 5 o o H Td Td o 13 o o 13 Td o o Td G ;s 12 Td G cd G td nd Td cd GGG a> cd cd cd <U rH 00 rH rH rH co co rH xz a a o CO cd CM rH Td Td CM rH <2 < rH Td Td Uu cd Td Td Td cd 5 3-1 u u 5 Td rH cd td Td o jd S & o o EH Td Td o !3 o o U Td o o Td G & & Td G td G cd Td Td cd GGG 0) cd cd td <D rH 00 rH -H rH CO CO rH xz a a a cd cd CM H Td Td CM rH < < rH Td Td Up cd Td Td Td cd &U uu & Td H cd cd Td o XZ & o EH Td Td o o & o o 12 Td o o Td G Td c cd G cd Td Td cd GGG <u cd cd cd a rH oo r--1 H H co co rH XZ a Ic3 g to CM H Td Td CM rH < <2 i--H Td Td up cd Td Td Td cd 3-i U p & Td H cd cd Td o XZ 2 15 o H Td Td o o o O Td o o Td G & Td G cd G cd tj Td cd GGG a) cd cd cd <D i--i 00 rH H H co w Xi a s o cd cd CM H Td Td CM rH <2 < rH II I I 9-n : HWBUI0002445 July day with ambient temperatures up to 91F. The At between the front brakes remained between 110 and 140F while that for the rear drum brakes remained between 10 and 30F. The temperature trends discussed here were typical for all three vehicle tests. 9.2.2.2 Heavy-Duty Fade Stops During the Detroit Traffic Test, a series of stops from 40 mph, in rapid succession, every two blocks caused all four brakes to rise in temperature faster than* normal. The shrouded brakes would then cool slower than the normal brakes during subsequent running. Conse quently, it was anticipated that during the fade tests it would be necessary to program the right front disc brake according to the known temperature versus stop number for the same friction materials previously run on an inertia dynamometer and on a normal test vehicle. The dynamometer and normal vehicle test baseline results are given in Figure 9-1. Two different dynamometer tests are included. The data for the shrouded right front disc brake and the normal left front disc brake are superimposed on the baseline results. After the 7th stop, the time increment was increased for the vehicle test to pre vent the right front disc brake from overheating the lining and creating an abnormal situation. The left brake showed a slight temperature reduc tion during this increased interval. The right front brake was thus held to the illustrated time-versus-stop curve by increasing the interval as necessary. A temperature of 525F was recorded as the temperature before the 10th stop. Temperature excursions after the last two stops were higher than 540F. The recovery stops demonstrated the rate at which the brakes returned to their normal operating temperatures. The data for the rear drum brakes are also shown in Fig ure 9-1. The shrouded rear drum brake followed the baseline curve closely. The left rear drum brake showed a slight temperature reduction with the increase in interval. During the recovery stops, both rear brakes followed the baseline cooling curve closer than the corresponding disc brakes followed their baseline cooling curve. Figure 9-2 shows the corresponding data for the 15-Stop Fade schedule. Baseline data for two inertia dynamometer tests are also plotted on the graph. During the vehicle test, the shrouded right front disc brake began to heat more rapidly than expected (based on the 10-stop fade experience) and the increased interval for cooling between stops was first necessary after the 5th stop. Additional increased intervals were then necessary after each succeeding stop. Consequently, the left front disc brake never did increase above 300F. A temperature of 575F was recorded as the temperature before the 15th stop. Temperature excur sions after the last 6 stops were higher than 600F. 9-12 ii n': HWBUI0002446 FRONT BRAKES TEMPERATURE (F) TEMPERATURE (F) REAR BRAKES FADE STOP NUMBER RECOVERY Figure 9-1 - Brake Temperature Data - Vehicle Test 1, 10-Stop Fade 9-13 HWBUI0002447 FRONT BRAKES: TEMPERATURE (F) TEMPERATURE (F) REAR BRAKES: Figure 9-2 - Brake Temperature Data - Vehicle Test 1, 15-Stop Fade 9-14 i'-' HWBUI0002448 The long time period at which the brakes were above 500F produced an additional effect within the shrouded brake. The heat gener ated at the rotor/pad interfaces was retained within the shroud and caused the brake fluid within the right front caliper to heat. The resultant boiling of fluid produced a "loss of brake pedal" for the 14th and 15th stops; the brakes bottomed out, and pumping the brakes did not restore adequate line pressure to maintain the planned deceleration. The recovery data illustrate the rate at which the shrouded brakes recovered from the high-temperature testing. The fluid boil (soft pedal) persisted through the first eight recovery stops until the indicated lining temperature of the shrouded brake fell below 400F. This behavior demonstrates the effectiveness of air cooling in keeping brake temperature low on normal unshrouded brakes. The data for the drum brakes are also illustrated in Figure 9-2. The increased interval early in the test sequence held the temperatures lower than for the baseline data. The temperature trends discussed for the fade schedules were typical for all three vehicle tests. 9.2.3 Collection Efficiencies 9.2.3.1 Recovery of Brake Particulate Emission The weights of brake particulate wear debris collected during vehicle tests 1, 2, and 3 are given in Tables 9-9, 9-10, and 9-11, respectively. For the right front disc brake, the surface sample has the largest weight of the three samples taken. The sump and airborne samples are comparable in weight. For the left front brake, the sump sample was comparable in weight to the right front brake. Moreover the total weight of the samples taken from the left front brake of vehicle tests 1 and 2 were comparable when added for each schedule to arrive at a total weight for the entire vehicle test. On the other hand, the cumulative sample taken for all of the vehicle test 3 was significantly less. For the right rear drum brake, the surfaces sample has the largest weight of the three samples taken. The sump and the airborne samples are comparable in weight. In some instances, the results for the sump value were reported negative. This occurred because of the 80 per cent return of material to the sump from each preceding schedule. No negative values occurred during the normal driving up to the end of the Detroit Traffic Test schedule. During the more severe fade tests, some of the 80 percent return was redistributed to the surfaces samples. The overall wear debris from each wheel did, however, have a positive weight gain. 9-15 HWBUI0002449 Table 9-9 - Weights of Brake Particulate Wear Debris Collected (Vehicle Test 1) (gm) Schedule Collector Burnish Sump Surfaces Airborne A.B. Baseline Sump Surfaces Airborne Detroit Traffic Sump Surfaces Airborne 10 Stop Fade Sump Surfaces Airborne A.F. Baseline Sump Surfaces Airborne 15 Stop Fade Sump Surfaces Airborne Final Baseline Sump Surfaces Airborne Totals Sump Surfaces Airborne Entire Test Front Brakes Right (Shrouded) Left (Normal) 0.142 0.412 0.061 0.116* 0.029 - 0.083 0.644 0.130 0.128 0.055 - 0.201 2.010 0.237 0.458 0.109 " 0.580 1.212 0.013 0.244 0.076 - 0.191 1.764 0.259 0.153 0.030 - 0.616 5.943 0.120 0.150 0.027 - 0.154 2.011 0.375 0.138 0.083 - 0.967 13.996 1.195 1.387 0.409 - 17.158 1.796 Rear Brakes Right (Shrouded) Left (Normal) 0.144 1.170 0.046 0.220 0.394 0.185 0.590 0.049 0.208 0.205 - 0.436 4.052 0.285 1.654 0.878 -- -0.335 0.371 0.017 1.120 0.275 -- 0.060 0.652 0.030 -0.516 1.543 - -0.046 0.100 0.004 -0.330 0.169 -- 0.097 0.440 0.003 0.017 0.448 - 0.541 7.375 0.434 2.373 3.912 - 8.350 6.285 *Estimated from cumulative sample (0.145 gm) taken during this schedule only. 9-16 HWBUI0002450 Table 9-10 - Weights of Brake Particulate Wear Debris Collected (Vehicle Test 2) (gm) Schedule Collector Burnish Sump Surfaces Airborne A.B. Baseline Sump Surfaces Airborne Detroit Traffic Sump Surfaces Airborne 10 Stop Fade Sump Surfaces Airborne A.F. Baseline Sump Surfaces Airborne 15 Stop Fade Sump Surfaces Airborne Final Baseline Sump Surfaces Airborne Totals Sump Surfaces Airborne Entire Test Front Brakes Right (Shrouded) 0.109 0.432 0.039 Left (Normal) 0.122 0.032 - 0.049 0.481 0.015 - - 0.042 2.828 0.106 0.343 0.126 - 0.170 1.437 0.023 0.128 0.055 - 0.098 1.114 0.107 0.084 0.036 - 0.098 2.205 0.099 0.084 0.049 " 0.051 1.193 0.138 0.111 0.141 - 0.555 9.690 0.527 0.984 0.439 - 10.772 1.423 Left Brakes Right (Shrouded) 0.105 0.166 0.004 Left (Normal) -0.132 0.157 -- 0.081 0.596 0.007 0.109 0.166 -- 1.247 4.159 0.056 0.955 0.769 -0.217 1.322 0.010 -0.597 0.290 -- -0.047 0.749 0.004 -0.255 0.328 - -0.047 0.181 0.008 -0.255 0.360 -- 0.260* 1.070* 0.011 0.295 0.172 - 0.941 8.293 0.104 0.637 2.242 9.288 2.879 Value may be in error because of brake fluid in collector during Final Baseline schedule. 9-17 I HWBUI0002451 Table 9-11 - Weights of Brake Particulate Wear Debris Collected (Vehicle Test 3) (gm) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne A. B. Baseline Sump Surfaces Airborne Detroit Traffic Sump Surfaces Airborne 10 Stop Fade Sump Surfaces Airborne A.F. Baseline Sump Surfaces Airborne 15 Stop Fade Sump Surfaces Airborne Final Baseline Sump Surfaces Airborne Totals Sump Surfaces Airborne Entire Test FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.121 0.408 0.045 Cumulative Sample 0.319 2.056 0.033 Cumulative Sample 0.266 1.097 0.138 0.215 1.234 0.052 0.208 3.773 0.495 0.379 1.154 0.027 0.438 0.515 0.188 0.151 0.248 0.015 0.033 1.389 0.328 0.118 0.295 0.050 0.391 3.075 0.402 0.088 0.184 0.018 0.105 2.860 0.337 0.246 2.288 1 r 0.017 1.562 13.117 1.933 0.182 0.169 - 1.516 7.459 0.212 0.974 1.908 - 16.612 0.351 9.187 2.882 J i 9-18 i HWBUI0002452 9.2.3.2 Mass Balance 9.2.3.2.1 Amount of Emissions The Initial weights of the brake friction material and the weight losses per wheel are given in Table 9-12. The weights of the particulates collected (Tables 9-9, 9-10, and 9-11) and the data in Table 9-12 were used to calculate the mass balances given in Tables 9-13, 9-14, and 9-15 (numbers not in parentheses). The right front brake data ranged from approximately 42 percent recovery in the initial tests to approximately 70 percent in the later tests. The low initial recovery may have been due to excess gassing of the green linings resulting in a lower recovery as particu lates. This condition changed as the linings became burnished to a greater extent and the frictional heat-affected layer contained more inorganic content. The recovery for the 10-stop fade of vehicle tests 1 and 3 were low as the debris was "tarry" and the surfaces fraction could be collected only with difficulty. The recovery for the 15-stop fade of vehicle tests 1 and 3 was somewhat low, possibly for the same reason. The left front brake showed the same basic trends as the right front brake, increasing throughout the vehicle tests where normal driving was carried out. During the two fade tests and after fade and Final Baseline, approximately 2 to 3 times more wear debris remained in the unshrouded brake. This may have been due to the high concentration of wear debris produced, which in turn was scraped into the rivet holes during the braking. The material recovery totals for the rear brakes show wide variations. The problem, in part, was due to the carryover from test-to-test of the sump debris, 80 percent of which was returned to the brake after each test schedule. The data can be explained partially by examination of the distribution of particulate emissions obtained at each sample location. 9.2.3.2.2 Distribution of Emissions The data given in parentheses in Tables 9-13, 9-14, and 9-15, represent the percent of total particulates collected - the distribution of the particulate emissions. The bottom portion of Tables 9-13, 9-14, and 9-15, shows the ranges for the distribution into sump, surface, and airborne samples. The data for the rear brakes fell into fairly narrow ranges for the first three schedules only. From the 10-Stop Fade to the Final Baseline, the data show both negative recoveries and greater than 100 percent recoveries. For both the 10- and 15-Stop Fade schedules of vehicle test 1, the negative recovery figures were not reflected by corresponding increases in the surface or airborne figures. However, the reverse situation prevailed in vehicle test 2 for the 10-Stop Fade and the ensuing A.F. Baseline. 9-19 HWBUI0002453 9-20 Table 9-12 - Brake F ric tio n M a te ria l I n i t i a l W eights and W eight Losses (gm) sassoq qtqSrafj D u ty M oderate Heavy to A b u s iv e C um ulative As Prepared S c h e d u le B u rn is h A. B. B aseline D e tro it T ra ffic T otals 10-S top Fade A. F. B aseline 15-S top Fade F in a l B aseline T otals T o ta l Wear Inner O u ter Prim ary Secondary T otals V ehicle Test 1 Front Rear R ight L e ft R ight L e ft i 1.48 0.96 3.33 1.35 1.86 1.29 0.92 1.21 5.64 4.17 5.80 4.96 V ehicle Front R ight L e ft Test 2 ! |------------------------------------------------------ Rear V ehicle Front R ight I L e ft R ight L e ft i Test 3 Rear R ight ' L e ft 1.45 0.94 0.66 0.79 1.32 0.62 3.11 3.06 1.15 1.11 1.21 1.17 2.81 1.39 1.89 0.80 4.11 3.53 5.31 4.16 5.02 5.05 1.83 4.67 oco <r mi--i m L1--O1 <J\ CN| I--1 vO cr-o1 Ci--N1j r- co CmO m ri---i vO mOJ CN 8.98 6.42 10.05 7.52 7.06 6.83 8.53 5.72 0.78 0.45 0.67 2.69 1.58 1.01 0.54 4.44 0.82 1.22 1.90 3.34 1.14 1.02 2.39 2.87 1.28 0.28 0.95 2.11 1.17 0.61 1.34 11.87 0.50 0.29 0.25 3.62 0.49 0.35 0.55 7.79 0.72 1.20 0.80 3.73 1.22 0.49 1.13 1.89 0.68 1.57 0.83 4.14 2.44 1.19 1.45 24.66 3.64 4.44 11.07 4.03 2.87 18.48 5.49 33.64 10.06 12.30 11.96 17.78 i--------------- --------------- i i--------------- 95.8 93.8 106.4 117.4 116.5 105.0 108.1 106.4 189.6 189.3 9.61 106.1 105.6 10.39 104.9 181.3 8.99 102.4 180.2 27.63 12.21 11.13 14.02 115.3 152.3 1 113.0 160.3 1 113.2 153.5 ! 113.7 167.7 213.2 210.3 297.7 295.7 211.4 211.7 286.2 282.6 1 267.6 273.3 266.7 280.4 ! sqqSrnM XEiqxux HWBUI0002454 Table 9-13 - Material Percent Recovery as Particulate Emissions (Vehicle Test 1) SCHEDULE Burnish A. B. Baseline Detroit Traffic 10 Stop Fade A. F. Baseline 15 Stop Fade Final Baseline Ranges COLLECTOR Sump Surfaces Airborne Total Sump Surface Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces. Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) (24)* (67) (9) 10 28 4 42 (81) (19) 9 2 11 (ID (24) (15) 5 34 7 45 (71) (29) 10 4 - 14 (10) (81) (9) 4 35 4 43 (79) (21) 11 3 - 14 ,(31~) (63+) (6~) 10 20+ 2 32+ (76) (24) 31 10 - 41 (9) (79) (12) 6 52 8 66 (81) (19) 13 3 - 16 (9) (89) (2) 5 50 1 56 (88) (17) 30 6 - 36 (6) (79) (15) 4 54 10 68 (61) (39) 11 7 - 18 ( 6-31) (63-89) ( 2-15) (61-83) (17-39) REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) (10) (87) (3) (22) (72) (6) 4 36 1 41 20 65 5 90 (36) (64) (51) (49) 16 29 - 45 17 16 - 33 (9) (85) (6) 7 70 5 82 (65) (35) 33 18 51 -75 166 83 41 412 207 (8) (88) (4) 6 64 3 73 -22 65 - 43 -16 -133 35 68 120 -65 20 1 90 40 1111 41 ( 8- 22) (72- 88) ( 3- 6) (36-65) (35-65) ^Numbers in parenthesis are percents of total particulates collected: 10/42 = 24%. +Not all of the sample was recovered. 9-21 HWBUI0002455 Table 9-14 - Material Percent Recovery as Particulate Emissions (Vehicle Test 2) SCHEDULE Burnish A.B. Baseline Detroit Traffic 10 Stop Fade A.F. Baseline 15 Stop Fade Final Baseline Ranges COLLECTOR Sump Surfaces Airborne Total Sump Surface Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) (20) * (73) (7) 8 30 3 41 (81) (10) 13 3 - 16 (7) (71) (22) 4 41 13 58 Not Sampled (2) (93) ( 5) 1 65 3 69 (72) (28) 10 4 14 (10) (88) (2) 6 33 1 60 (72) (28) 8 3 11 (2) (89) (9) 1 39 4 44 (83) (17) 15 3 18 (5) (90) (5) 3 61 3 67 (63) (37) 17 10 - 27 (A) (86) (10) 3 63 7 73 16 21 37 (2- 20) (71- 93) (2- 22) (63-83) (17-37) REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) (38) (60) (2) 16 25 1 42 -17 20 3 (12) (86) (2) 7 49 1 57 (39) (61) 9 14 23 (22) (77) (1) 23 78 1 102 (56) (44) 23 18 - 41 -22 - 110 131 54 1- 111 -56 181 (44) 28 267 (56) 35 1" 87 63 -13 52 2 41 -46 65 - 19 Brake Fluid in Linings (63) (37) 36 21 57 (12-22) (60-86) (1-2) (39-63) (37-61) *Numbers in brackets are percents of total particulates collected: 8/41 = 20% I 9-22 i i HWBUI0002456 Table 9-15 - Material Percent Recovery as Particulate Emissions (Vehicle Test 3) SCHEDULE Burnish A. B. Baseline Detroit Traffic 10 Stop Fade A.F. Baseline 15 Stop Fade Final Baseline Ranges COLLECTOR Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) (21)* (72) (7) 9 31 3 43 Cumul ative Sam pie (19) (72) (9) 10 39 5 54 (5) (85) (11) 4 75 10 89 (38) (46) (16) 10 12 4 26 (2) (79) (19) 2 66 16 84 (20) (70) (10) 10 34 5 49 r (4) (86) (10) 3 69 8 80 (52) 1.5 (48) 1.4 ~3~ ( 2-38) (46-86) ( 7-19) RIGHT (SHROUDED) REAR BRAKES LEFT (NORMAL) (13) (83) ' (4) 10 66 3 79 Cumu] ative Sair pie (9) (87) (4) 7 65 3 75 (25) (74) (1) 21 63 1 85 (36) (61) (3) 12 20 1 33 (25) (64) (11) 19 48 8 75 (32) (64) (4) 7 14 1 22 H2O in Linings r (34) (66) 9 17 26 ( 9-36) (61-87) ( 1-11) *Numbers in brackets are percents of total particulates collected. 'i. ;- 9-23 t i HWBUI0002457 The ranges for the sump, surfaces, and airborne samples given at the bottom of all three tables agreed rather well for all brakes irregardless of the brake friction materials used or the rotor conditions prior to the start of the test. 9.3 PERCENT ASBESTOS CONTENTS 9.3.1 Program Results The microscopy results were converted by the computer into the asbestos emissions results summarized in Tables 9-16, 9-17, and 9-18. For vehicle test 1, the range of asbestos contents in brake emissions varies from a high of 1.65 percent down to a low of 0.05 per cent. Of the 47 analyses reported, only two were above 1.00 percent. The overall average was 0.38 percent. For vehicle test 2, the range of asbestos contents in brake emissions varies from a high of 1.42 percent down to a low of 0.03 per cent. Of the 43 analyses reported, only one was above 1.09 percent, and only three were in the range 0.50 to 0.96 percent; all others were less than 0.50 percent. The overall average was 0.25 percent. For vehicle test 3, the range of asbestos content in brake emissions varies from a high of 0.51 percent down to a low of 0.003 per cent. Of the analyses reported, three were in the range 0.20 to 0.51 per cent; all others were less than 0.20 percent. The overall average was 0.07 percent. For all three vehicle tests, the overall average of asbestos content in the brake emissions was 0.23 percent. The following trends were noteworthy: The right front disc brake yielded higher asbestos contents than the right rear drum brake (for two of the three vehicle tests): Vehicle Test R.F. Disc (%) R.R. Drum (%) 1 0.45 0.30 2 0.26 0.24 3 0.07 0.07 In general, the airborne samples contained a lower asbestos weight percent than the corresponding sump and surfaces samples In general, the asbestos weight percent in brake emissions was higher when the materials were new; the asbestos content de creases with brake use. For all vehicle tests, the asbestos weight percent increased only slightly for the severely burnished brakes during the fade tests. 9-24 HWBUI0002458 Table 9-16 - Summary of Asbestos Analytical Results (Vehicle Test 1) (Weight Percent) Schedule Burnish Collector Front Brakes Right (Shrouded) Left (Normal) Sump Surfaces Airborne 0.813 0.398 0.144 (0.452)* Not ** Analyzed Rear Brakes Right (Shrouded) Left (Normal) 0.865 0.349 (0.607) No t Analyzed A. B. Baseline Sump Surfaces Airborne 1.650 0.472 0.290 (0.804) 0.179 0.415 0.493 (0.362) 1 Detroit Traffic Sump Surfaces Airborne 0.247 0.137 0.052 (0.145) 0.129 0.373 0.092 (0.198) 0.991 0.265 10-Stop Fade Sump Surfaces Airborne 0.463 0.162 0.154 (0.260) 0.280 0.071 0.132 (0.161) 0.275 0.416 A.F. Baseline Sump Surfaces Airborne 0.656 0.507 0.134 (0.432) 0.419 0.242 0.079 (0.247) Not Analyzed 15-Stop Fade Sump Surfaces Airborne 0.602 1.548 0.175 (0.775) 0.071 0.257 0.222 (0.183 0.439 0.472 Final Baseline Sump Surfaces Airborne 0.435 0.290 0.167 (0.297) 0.585 Not 0.324 Analyzed 0.145 1 ' (0.351) Average 0.452 Average 0.301 *Average value for'wheel where collection took place. **Not Analyzed - indicates that analysis was not required. 9-25 Ii. - . h--' HWBUI0002459 Table 9-17 - Summary of Asbestos Analytical Results (Vehicle Test 2) ' (Weight Percent) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne A. B. Baseline Sump Surfaces Airborne Detroit Traffic Sump Surfaces Airborne 10-Stop Fade Sump Surfaces Airborne A.F. Baseline Sump Surfaces Airborne 15-Stop Fade Sump Surfaces Airborne Final Baseline Sump Surfaces Airborne Average FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.402 0.968 0.186 (0.515)* No t Anal yzed 0.236 0.311 0.211 (0.253) 0.304 0.548 0.234 (0.362) 0.244 0.261 0.107 (0.260) 0.443 0.173 0.165 (0.260) 0.119 0.193 0.095 (0.136) 0.117 0.071 0.032 (0.073) < 0.258 Average REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.484 1.416 0.087 (0.636) Nc>t Anal .yzed 0.268 0.098 0.344 (0.237) 0.249 0.113 0.115 (0.159) 0.199 0.619 0.194 (0.185) 0.138 0.237 0.180 (0.185) 0.065 0.136 0.048 (0.083) 0.058 0.067 (0.063) 0.243 1r *Average value for wheel where collective took place. 9-26 HWBUI0002460 Table 9-18 - Summary of Asbestos Analytical Results (Vehicle Test 3) (Weight Percent) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne A. B. Baseline Sump Surfaces Airborne Detroit Traffic Sump Surfaces Airborne 10-Stop Fade Sump Surfaces Airborne A.F. Baseline Sump Surfaces Airborne 15-Stop Fade Sump . Surfaces Airborne Final Baseline Sump Surfaces Airborne Average FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) REAR BRAKES RIGHT (SHROUDED) LEFT (NOEMAL) 0.218 0.101 0.099 (0.109)* Cumul ative Sam pie 0.063 0.046 0.060 (0.056) Cumul ative Sam pie 0.106 0.024 0.125 (0.085) 0.063 0.056 0.044 (0.054) 0.018 0.013 0.025 (0.019) 0.046 0.057 - 0.037 (0.047) 0.045 0.041 0.055 (0.047) 0.102 0.055 0.014 (0.057) 0.158 0.073 0.047 (0.093) 0.034 0.022 0.003 (0.019) 0.109 0.047 0.035 (0.064) 0.089 0.087 0.037 (0.071) 1r Not Analyzed 0.065 0.025 0.014 (0.035) 0.512 0.237 0.035 (0.261) r 0.026 0.031 0.070 Average 0.074 *Average value for wheel where collection took place. 9-27 HWBUI0002461  The data for vehicle test 2 showed slightly lower emissions than for vehicle test 1 (Test 2 started with smoother used rotor surfaces.) 9.3.2 Battelle Analyses The contractual requirements for the asbestos particle size distribution in addition to the asbestos content placed undue burden on the accuracy of the asbestos content calculation. To provide a systematic independent check on the asbestos emissions analyses, the EPA Project Officer arranged a second contract for Battelle Columbus Laboratories to analyze 24 samples generated by the program. The analytical method used by Battelle was developed for EPA under a separate contract. The method is outlined in Appendix H. The Battelle results are given in Appendix H. Mr. Carl Melton was the Battelle Technical Contact. 9.3.3 Johns-Manville Analyses During the course of the program, Johns-Manville volunteered to analyze three samples from the program using an analytical procedure which they had developed for the determination of low asbestos contents in brake wear debris. Drs. Sidney Spiel and James Leineweber conducted these tests. The results are given in Appendix H. 9.3.4 Comparison of Asbestos Analytical Results Table 9-19 gives the comparative data for the asbestos analytical results obtained by the program, Battelle, and Johns-Manville. The three Johns-Manville results agreed with the Battelle data as expected. In 19 of the 24 analyses, the program results were higher for asbestos percent than the corresponding data from Battelle. The five Battelle results that were larger than the corresponding program results were larger by the following factors: 7.5, 2.6, 2.1, 2.1, and 1.2. There were three program results larger than the corresponding Battelle results by a factor of less than 1.5. Of the remaining 16 comparative sets of data, two program results were greater by a factor in the range 1.5 to 2.5, eight program results were greater by a factor in the range 2.5 to 10, and six program results were greater by a factor in the range 10 to 100. The program average was 0.25 percent, while the Battelle average was 0.17 percent, or 68 percent of the Program average. The program data were expected to be higher for the follow ing reasons: All fibers found were assumed to be cylinders of projected diameter and length as indicated by microscopy. A fiber with an elliptical cross section will actually be smaller in volume than that reported. All fibers found were assumed to be 100 percent asbestos. Par tially degraded asbestos fibers are similar to asbestos and were counted. This is especially true for the larger fibers. 9-28 HWBUI0002462 B a tte lle T a b le 9-19 - Com parison o f A n a ly tic a l R e s u lts fo r A sbestos r-. 1 1 o o 42 o o 4.J p 1--1 cj e cd 1--1 Mi o\ 00 1 CM o o PM ''' 11 1 111 11 111 11 1 11I 11 I 1 0) m vo rH r- QJ 4J CM o o ooo CM o CM 1 o o 00 rH CO 1 CO 4J a) cd o o o o o o o 4*5 pq cd pq jsj pi "3 M p CM VO oo 1 oo oo 11 1 1il Ml cd <D B cd CT\ CO CM o CM VO Ml CM r"- 00 CO 00 rH 00 --1 CO o CM i rH -d- -d- 1 o u o o o o o o 1--1 Ph rH in o r*- in vo rH --1 -d- r-. ooo o o o ooo o o o 11I 1 1 1 Ob CO m vO r-- ON -d" r--1 rH < CO 00 CM rH rH o o CM ooo o o o 0) I--1 1--i vo in VO m 1(U t--1 r--1 o CM iH 00 o CM o CO rH vo CO 4-> CM o o CM | CO CM p--1 vo o --1 o CM a) M 4*5 cd o o o rH o o o o o o o o cd pq Mi PQ am CO s o H J 1o 1 1f 1 1 1 ii 1 1 1 1 Q o 4-1 a O Ml cd r-. r*'. CM CO CM VO <r Kf 00 in Ml -d- CO in vO m o CO o CO rH CM o 00 CM iH o i CM iH CO 1 CM o o CM u ooo o o o o o o o o o PM Ml o CO 0) .4- CJ 0) pj o Mi 0) cd o PM M-t 43 e u Mi o P pi H o CO C/3 < CO 0) 0o) cd c Moi PM <4-4 4-1 E n M4 PJ P H CO CO < CO 0) a) pi a cd Moi PM <4-1 4J E Mi M4 P P H CO CO < CO a) <D Pi u cd Mol PM IH rQ B Ml u p p H CO CO < CO ID a) a CJ cd Moi PM VH 40 e Mi Mi p p H CO CO c CO (D 00 cd Mi ID > < CD 4-> o p T3 H o MH 0) u M-t 42 4- u 0) Mi CO H PM o 4-> CO <D 1 *0 o cd iH Pn I 3 sax aioxqaA & 4-1 O CO H *H H O IH Pi Ml *H M 4-J Cd p a) Mi pq p p Z }S31 aioxqaA 4.1 o H o <H Ml *H 4-> cd <D Mi P2 H c nsax aXOTqaA GvO\ O II u P0uoH0 --<DI ID CmN rH-. ufJ cd O O PQ . oj .. QJ 00 O 00 Cd H ucd ua) ucd 0) > P*S >< < a) gcud t-aH) 00 4-> O 4- u cd PH pq i 9-29 i; HWBUI0002463  Other analytical techniques rely on reducing all asbestos to its fibril nature where dimensions and structure are well characterized. At the same time, all olivine and partially degraded asbestos are mechanically reduced to a nonfibrous nature so that only the stronger asbestos fibrils that remain are counted. 9.4 ASBESTOS EMISSIONS FACTORS To compare the generation of asbestos emissions properly on a mass-weighted basis, it was found necessary to calculate the asbestos emissions factors in weight and in weight per mile. These weights of asbestos emissions were obtained by multiplying the weight of brake wear debris generated (Tables 9-9, 9-10, or 9-11) by the appropriate asbestos percent determined for that weight of debris produced (Tables 9-16, 9-17, or 9-18). These data are expressed in mg in Tables 9-20, 9-21, and 9-22. Example: Vehicle Test 2 Burnish Right Front Brake Sump From Table 9-10: Weight of sample collected: 0.109 gm From Table 9-17: Asbestos weight percent: 0.402 Data given in Table 9-21: 0.10 gm x 1000 2S- x 0.00402 = 0.438 mg gm Using the mileages reported in Table 9-4, the asbestos emission in yg per mile are given in Tables 9-23, 9-24, and 9-25. Example: Sample same as that used in above example. From Table 9-21: Asbestos weight produced for entire test: 0.438 mg From Table 9-4: Mileage for Burnish Sequence of vehicle test 2: 127 miles Data given in Table 9-24: 0.438 mg x 1000 mg z 127 miles = 3.45 yg/mile. | 9-30 HWBUI0002464 Table 9-20 -- Weight of Asbestos Generated During Braking (Vehicle Test 1) (mg) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne Total A. B. Baseline Sump Surfaces Airborne Total Detroit Traffic Sump Surfaces Airborne Total 10-Stop Fade Sump Surfaces Airborne Total A. F. Baseline Sump Surfaces Airborne Total' 15-Stop Fade Sump Surfaces Airborne Total Final Baseline Sump Surfaces Airborne Total FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 1.160 1.640 0.088 2.888 Not Analyzed 0.137 0.340 0.377 0.854 0.497 2.754 0.123 3.374 2.685 1.963 0.080 4.728 1.253 8.943 0.347 10.543 3.708 91.998 0.210 95.916 0.670 8.748 0.626 10.044 ' REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) N/A 10.121 0.161 10.282 Not Analyzed 0.331 2.449 0.242 3.022 0.562 15.114 0.262 15.938 -- 1.039 0.022 1.061 ' ' 16. 350 2. 315 18. 665 3. 180 1. 143 4. 323 0.251 1.578 0.024 1.853 -- 0.257 0.009 0.266 Not Analyzed 0. 798 V 0. 798 0.567 1.426 0.004 1.997 Not Analyzed k 9-31 HWBUI0002465 Table 9-21 - Weight of Asbestos Generated During Braking (Vehicle Test 2) (mg) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne Total A.B. Baseline Sump Surfaces Airborne Total Detroit Traffic Sump Surfaces Airborne Total 10-Stop Fade Sump Surfaces Airborne Total A.F. Baseline Sump Surfaces Airborne Total 15-Stop Fade Sump Surfaces Airborne Total Final Baseline Sump Surfaces Airborne Total FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.438 4.190 0.026 4.654 Not Analyzed 0.116 1.495 0.032 1.643 0.127 15.620 2.510 18.237 0.415 3.755 0.024 4.194 0.160 1.932 0.177 2.269 0.117 4.250 0.094 4.461 0.051 0.851 0.044 0.946 REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.508 2.350 0.035 2.893 Not Analyzed 0.217 0.584 0.028 0.829 3.108 4.705 0.064 7.877 -0.430 8.190 0.019 7.779 -0.672 1.775 0.007 1.110 -0.031 0.246 0.004 0.229 nU 0o4/. ~/7 0.074 0.921 T 9-32 HWBUI0002466 Table 9-22 - Weight of Asbestos Generated During Braking (Vehicle Test 3) (mg) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne Total A. B. Baseline Sump Surfaces Airborne Total Detroit Traffic Sump Surfaces Airborne Total 10-Stop Fade Sump Surfaces Airborne Total A.F. Baseline Sump Surfaces Airborne Total 15-Stop Fade Sump Surfaces Airborne T otal Final Baseline Sump Surfaces Airborne Total FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 0.264 0.412 0.045 0.721 Cumulative Sample 0.201 0.946 0.020 1.167 Cumulative Sample 0.282 0.263 0.173 0.718 0.135 0.691 0.023 0.849 0.037 0.490 0.124 0.651 0.174 0.658 0.010 0.842 0.197 0.211 0.103 0.511 0.154 0.136 0.002 0.292 0.052 1.014 0.154 1.220 0.040 0.065 0.002 0.107 0.426 1.445 0.141 2.012 0.093 2.488 0.125 2.706 T Not Analyzed 0.057 0.046 0.003 0.106 1.260 5.422 0.006 6.688 1rg .0 253 0.591 .0 844 9-33 HWBUI0002467 Table 9-23 - Weight of Asbestos Generated During Braking (Vehicle Test 1) (yg/mile) Schedule Collector Front Brakes Right (Shrouded) Left (Normal) Burnish Sump Surfaces Airborne Total 6.77 9.42 0.51 16.70 Not Analyzed A. B. Baseline Sump Surfaces Airborne Total 0.04 1.04 1.15 2.23 Detroit Traffic Sump Surfaces Airborne Total 0.35 1.91 0.09 2.35 10-Stop Fade Sump Surfaces Airborne Total 57.13 41.77 1.70 100.60 A.E. Baseline Sump Surfaces Airborne Total 3.52 25.12 0.97 29.61 15-Stop Fade Sump Surfaces Airborne Total 78.89 1957.40 4.47 2040.76 Final Baseline Sump Surfaces Airborne Total 1.90 24.85 1.78 28.53 11 Rear Brakes Right (Shrouded) Left (Normal) N/A 58.11 0.92 59.09 Not Analyzed 1.01 7.47 0.74 9.22 0.39 10.50 0.18 11.07 13.52 1.61 - 14.13 - 22.11 0.47 22.58 67.85 24.35 92.20 0.71 4.43 0.07 5.21 __ 5.47 0.19 5.66 1.61 4.05 0.01 5.67 Not Analyzed - .1 r 9-34 HWBUI0002468 Table 9-24 - Weight of Asbestos Generated During Braking (Vehicle Test 2) (pg/mile) SCHEDULE COLLECTOR Burnish Sump Surfaces Airborne Total A.B. Baseline Sump Surfaces Airborne Total Detroit Traffic Sump Surfaces Airborne Total 10-Stop Fade Sump Surfaces Airborne Total A.F. Baseline Sump Surfaces Airborne Total- 15-Stop Fade Sump Surfaces Airborne Total Final Baseline Sump Surfaces Airborne Total FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 3.45 32.70 0.20 36.35 Not Analyzed 0.33 4.29 0.09 4.71 0.09 10.92 1.74 12.75 7.83 70.85 0.45 79.13 0.47 5.70 0.52 6.69 1.48 53.85 1.19 56.52 0.16 2.61 0.14 2.91 T REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 4.63 18.52 0.28 23.43 Not Analyzed 0.62 1.68 0.08 2.38 2.17 3.29 0.04 5.50 - 8.22 154.50 0.36 146.64 -1.98 5.24 0.02 3.28 -0.39 3.11 0.05 2.77 9 fin 0.23 2.83 T 9-35 HWBUI0002469 Table 9-25 - Weight of Asbestos Generated During Braking (Vehicle Test 3) (pg/mile) SCHEDULE COLLECTOR Burnish A. B. Baseline Detroit Traffic 10-stop Fade A.F. Baseline 15-Stop Fade Final Baseline Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total Sump Surfaces Airborne Total FRONT BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 2.64 4.12 0.45 7.21 Cumulative Sample 0.77 0.72 0.47 1.96 0.03 0.36 0.09 0.48 3.40 3.64 1.78 8.82 0.15 2.94 0.44 3.51 7.22 24.49 3.20 34.91 0.27 7.25 0.36 7.88 T Not Analyzed REAR BRAKES RIGHT (SHROUDED) LEFT (NORMAL) 2.01 9.46 0.20 11.67 Cumulative Sample 0.04 1.88 0.06 1.98 0.13 0.49 0.01 0.63 2.66 2.34 0.03 5.03 0.12 0.19 0.01 0.32 0.97 0.78 0.05 1.80 3.67 15.81 0.02 19.50 T 0.10 0.22 - 9-36 HWBUI0002470 The data for the clutch emissions are given in Table 9-26. During vehicle test 1 and the early stages of vehicle test 2, it was decided to combine the very small airborne samples collected as indicated in Table 9-19. The treatment of the data to arrive at the asbestos emis sions in yg/mile was similar to that for the brake emissions given earlier. 9.5 ASBESTOS EMISSIONS TRENDS 9.5.1 New versus Burnished Materials The comparison of the Burnish, A.B. Baseline, and Detroit Traffic Test overall asbestos emissions data is given in Table 9-27. Both the right front disc brake and the right rear drum brake show that the asbestos emissions generated during the Burnish schedule are higher for the new friction materials. Asbestos emissions decrease for slightly used materials during the A.B. Baseline schedule. The asbestos emissions increase slightly with increased use as shown with the Detroit Traffic test. The rear drum brake data showed two effects brought out by the vehicle test scheduling and material selections: The data above indicate that the Detroit Traffic asbestos emissions are higher than for the A.B. Baseline. This effect is probably due to the extent of conditioning of the lining prior to each test. The Burnish duty is slightly heavier than the A.B. Baseline duty and this may be responsible for the dif ferences between the A.B. Baseline and the Detroit Traffic results, even though both latter tests are identical in duty. Vehicle test 2 used the same linings as vehicle test 1. The Burnish, A.B. Baseline, and Detroit Traffic schedules showed lower asbestos emissions than vehicle test 1. The data demon strate that asbestos emissions decrease with increased lining life. The amount of clutch airborne wear debris collected for vehicle test 2 was less than for vehicle test 1. The progression through vehicle test 3 would have continued, except for the fact that several severe clutch applications and clutch slippages were made to increase the amount of debris to simulate generation of emissions under more se vere conditions. In spite of this practice, the asbestos percent of the debris and the airborne asbestos emissions content for the clutch samples decreased with use as shown in Figure 9-3. 9.5.2 Disc Versus Drum Brake Materials trends: The data given in Table 9-27 also indicate the following For all moderate-duty schedules, the drum brakes produced more total emissions than the disc brake. 9-37 HWBUI0002471 Table 9-26 - Summary of Clutch Asbestos Emissions SAMPLE VEHICLE TEST SCHEDULE AIRBORNE SURFACES SUMP Burnish A.B. Baseline Detroit Traffic 1 10-Stop Fade A.F. Baseline 15-Stop Fade Final Baseline Burnish Cumulative 2 (A.B.Baseline To Final Baseline) Cumulative 3 (Burnish to Final Baseline) Cumulative for all 3 Vehicle Tests Cumulative for all 3 Vehicle Tests WEAR DEBRIS WEIGHT (gm) 0.0005 0.0042 0.0097 0.0001 0.0009 0.0003 0.0040 0.0004 0.0058 ASBESTOS WEIGHT PERCENT H/A N/A 0.291 0.493 N/A 0.136 ASBESTOS (Mg) 0.028 0.026 0.008 ASBESTOS (ug/mile) 0.019 0.032 (0.023) 0.003 0.0126 1.9519 14.8472 0.035 0.004 0.099 0.047 1.930 6.990 0.001 0.239 0.864 Table 9-27 - Comparison of Asbestos Emissions from New and Burnished Materials Location Vehicle Test Burnish (ug/mile) A,B. Baseline (ug/mile) Detroit Traffic (Mg/mile) RIGHT FRONT DISC BRAKE RIGHT REAR DRUM BRAKE 1 2 3 Totals i 2 3 Totals 16.70 36.35 7.21 60.26 59.09 24.43 11.67 94.19 2.23 4.71 1.96 8.90 9.22 2.38 1.98 13.58 2.35 12.75 0.48 15.58 11.07 5.50 0.63 17.20 Burnish A.B. Baseline 7.5 7.7 3.7 Average: 6.3 6.4 10.3 5.9 Average: 7.5 Burnish A.B.B. + D.T. 2 7.3 5.3 9.2 Average: 7.3 5.8 7.3 12.2 Average: 8.8 i. 9-38 ! i I HWBUI0002472 ii P -8 2 -9 9 -3 VEHICLE TESTS <---------- 3 -------- Figure 9-3 - Decrease in Clutch Airborne Asbestos Emissions with Continued Normal-Duty Service The ratio of drum brake/disc brake emissions decreases with use: for the Burnish schedule, the ratio was 1.55; for the A.B. Baseline schedule, the ratio was 1.53; and for the Detroit Traffic schedule, the ratio decreased to 1.10. 9.5.3 Heavy (Abusive) versus Moderate Duty The asbestos emission results for the 10-Stop Fade and 15-Stop Fade - considered heavy-duty or abusive tests - and the cor responding A.F. Baseline and Final Baseline are given in Table 9-28. i 9-39 HWBUI0002473 9-40 T a b le 9.-28 - A sbestos E m is s io n fro m H e a vy-D u ty T e s ts a rl 0) 44 44 cd cd En M H tx o 4J 4-1 rl. CO O 1M o 44 I--1 a) Q <r CM CM 00 CM CO 0) CM VO CO 00 cd M 0) > 5 CM CM o o d CM vO CO 50 CM cud d <> d e H r-H /""S d a) (0 rH cd H pq a i--l 50 cd d a rl PH CO rH 00 CM m OV 00 CO 00 CM r-. On CM CO r- CO o o vO 00 m o in CM Ov 00 rH CM d cd PH d CX rl oa +j CO 00 1o IT) i--1 vO CM I-"- m OV CM o VO sT CM m CO CO o i--1 CM CM vO f"-. o CO VO r^. 00 CM m CM rH O rH <D c H rH 0) 0) rH (0 td 00 d pH < rH a\ rH rH vO vO to 00 CV vO CO o\ CM CO i--i 00 CM rH CM CM CO 00 m CO o 00 0) "d cd Ph 0) rH PX-H Oa 44 CO 00 1o OW rH O CO CM in VO CO in O ov CO 00 O c*. 00 rH 00 CO in m vO o CM CM vO LO CM r^. i--1 0) r-1 a 44 rl CO 0) d E-4 > 1E5 f4 44 cd iJ iD hj CO rH CM CO cd 44 O H 0) CM CO cd 44 o H ,34- 44 <u pi O M 50 O to rt H u H U PS pH O 44 a rC *4 e 50 crt 3 P rl cu U fP=S ccj Q pq !' 1 ! i i i ! I1 j i ; j t i | HWBUI0002474 The following trends are noteworthy: \ The right front brake emissions were higher than the correspond ing rear brake results as only the front brake experienced the true fade temperatures because of the brake designed balance. The 15-Stop Fade total emissions for the front brake were higher than for the 10-Stop Fade total, again because of the higher temperatures experienced, as compared to the 10-Stop Fade. The A.F. Baseline and Final Baseline total emissions were higher than the corresponding A.B. Baseline total given in Table 9-25 by a factor of approximately 4.4 for the front brake. The right rear drum brake show higher emissions for the 10-Stop Fade than for the 15-Stop Fade due to a shift in duty to the rear brakes as the front brakes fade during their first heavyduty experience. The above total-wheel asbestos emission trends indicate that there is a rise in total asbestos emissions produced with high temperature use. This trend was only partially reflected in a slight asbestos percent increase indicated earlier in Tables 9-16, 9-17, and 9-18. The greatest portion of the asbestos emissions increase was due mainly to the largerthan-normal amount of wear debris produced during and after the heavy duty and high temperatures experienced by the friction materials, as indi cated in Tables 9-9, 9-10, and 9-11, and considering the mileages for each schedule. The following observation was also noteworth in the compari son of moderate and heavy-duty tests: The ratio of asbestos emissions for the 10-Stop Fade to Detroit Traffic Tests (far right column of Table 9-27) was higher than the ratio of asbestos emissions for the Burnish to A.B. Baseline plus Detroit Traffic test average (far right column of Table 9-26). The order of decreasing asbestos emissions from brakes can be summarized: Fade (or heavy-duty stops) Burnish A.B. Baseline + Detroit Traffic (moderate driving) Decreased Asbestos Emissions 9.5.4 Effects of Friction Materials The data for the asbestos emissions were rearranged for each brake as shown on Table 9-29. This is the best way to compare the results due to the different friction materials. The following trends were noted: i; |i ij 9-41 HWBUI0002475 Table 9-29 - Comparison of Asbestos Emission Generated by Different Friction Materials Duty Schedule Asbestos Emissions (pg/mile) Vehicle Test 1 Vehicle Test 2 Vehicle Test 3 Front Rear Front Rear Front Rear Moderate Burnish A. B. Burnish Detroit Traffic 16.70 2.23 2.35 59.09 9.22 11.07 36.35 4.71 12.75 23.43 2.38 5.50 7.21 1.96 0.48 11.67 1.98 0.63 Totals 21.28 79.38 53.81 31.31 8.65 13.98 Heavy to Abusive 10-Stop Fade A. F. B seline 15-Stop Fade Final Baseline 100.60 29.61 2040.76 28.53 22.58 5.21 5.66 5.67 79.13 6.69 56.52 2.91 146.64 3.28 2.77 2.83 8.82 3.51 34.91 7.88 5.03 0.32 1.80 19.50 Totals 2199.50 39.12 145.25 155.52 55.12 26.75 ASBESTOS CONTENT (WT %) Figure 9-4 - Variation of Asbestos Emissions During Moderate Duty With Asbestos Content of Disc Pads 9-42 HWBUI0002476  For the disc pads, there was an increase in asbestos emissions for increases in pad asbestos content: Vehicle Test Estimated Asbestos Content (Weight percent) Asbestos Emissions (yg/mile) 3 60 8.65 1 65 21.28 2 70 53.81 These data are plotted in Figure 9-4 and show an almost linear relationship, predicting a zero asbestos emission content when the pad asbestos content falls below 55 weight percent. There was no such trend for the drum brake data. Both the disc pads and the drum linings of vehicle test 3 showed the lowest asbestos emissions of all three tests. Both the disc pads and the secondary linings contained brass chips which may have been responsible for the almost complete conversion of the asbestos in the brake emissions of the respective wheels. 9.5.5 Airborne versus Sump versus Surfaces Samples The comparisons of the asbestos emissions found in the sump, surfaces and airborne sample locations can be determined from the data given in Table 9-30. The data can be summarized and compared (Table 9-31) : Front Disc Brake Trends For the moderate-duty tests, the airborne asbestos emissions were 6.12 percent of the total emissions. The surface samples contributed 76.80 percent while the sump samples were 17.02 percent. For the heavy-duty tests, and as a result of the severely burnished rubbing surfaces, the airborne asbestos emissions were only 0.71 percent and the sump samples 6.71 percent as the surfaces samples increased to 92.58 percent. Rear Drum Brake Trends For the moderate tests, the airborne asbestos emissions were 2.00 percent of the total emissions. The surfaces samples contributed 88.28 percent while the sump samples were 9.72 percent. 9-43 HWBUI0002477 Table 9-30 - Comparison of Asbestos Emissions for Various Sample Locations (yg/mile) Duty Moderate Heavy or Abusive Vehicle Test 1 2 3 1 2 3 Schedule Burnish A. B. Baseline Detroit Traffic Burnish A. B. Baseline Detroit Traffic Burnish A. B. Baseline Detroit Averages: 10-Stop Fade A. F. Baseline 15-Stop Fade Final Baseline 10-Stop Fade A. F. Baseline 15-Stop Fade Final Baseline 10-Stop Fade A. F. Baseline 15-Stop Fade Final Baseline Averages: Front Disc Brake Sump Surfaces Airborne 6.77 0.04 0.35 3.45 0.33 0.09 2.64 0.77 0.03 1.61 57.13 3.52 78.69 1.90 7.83 0.47 1.48 0.16 3.40 0.15 7.22 0.27 13.52 9.42 1.04 1.91 32.70 4.29 10.92 4.12 0.72 0.36 7.28 41.77 25.12 1957.40 24.85 70.85 5. 70 53.85 2.61 3.64 2.94 24.49 7.25 185.04 0.51 1.15 0.09 0.20 0.09 1.74 0.45 0.47 0.48 0.58 1.70 0.97 4.47 1.78 0.45 0.52 1.19 0.14 1.78 0.44 3.20 0.36 1.42 Rear Drum Brake Sump Surfaces Airborne N/A 1.01 0.39 4.63 0.62 2.17 2.01 0.04 0.13 1.38 -- 0.71 -- 1.61 -- -- -- 1.30 2.66 0.12 0.97 3.67 1.58 58.11 7.47 10.50 18.52 1.68 3.29 9.46 1.88 0.49 12.38 22.11 4.43 5.47 4.05 154.50 5.24 3.11 1.30 2.34 0.19 0.78 15.81 18.28 0.92 0.74 0.18 0.28 0.08 0.04 0.20 0.06 0.01 0.28 0.47 0.07 0.19 0.01 0.36 0.02 0.05 0.23 0.03 0.01 0.05 0.02 0.13 9-44 HWBUI0002478 Table 9-31 - Summary Comparison of Surfaces, Sump, and Airborne Samples BRAKE SYSTEM SUMP (ug/mile) SURFACES (ug/mile) AIRBORNE (ug/mile) Disc Brake Moderate Duty Heavy Duty 1.61 (17.02)* 13.52 (6.71) 7.28 (76.86) 185.04 (92.58) 0.58 (6.12) 1.42 (0.71) Drum Brake Moderate Duty Heavy Duty 1.38 (9.72) 1.58 (7.90) 12.38 (88.28) 18.28 (91.45) 0.28 (2.00) 0.13 (0.65) Average ' (wt) 4.52 . (7.33) 55.75 (91.68) 0.60 (0.99) Data in parentheses are the percents of the total samples for the condition considered. 9-45 I j I HWBUI0002479  For the heavy-duty tests, and as a result of redistribution of the simp sample, the airborne asbestos emissions were only 0.65 percent and the sump samples 7.90 percent as the surfaces samples increased to 91.45 percent. 9.6 CHANGES IN LINING WEAR RATE AFTER USE AT HIGHER TEMPERATURES One of the goals of the project was to ascertain the changes in lining wear rate and the corresponding asbestos contents after precon ditioning at elevated temperatures. The lining wear changes are given in Table 9-32. The A.B. Baseline schedule (and the other Baselines) consisted of a three-day Detroit Traffic driving test. The Detroit Traffice schedule consisted of twelve days of the same road pattern. Thus one-fourth of the Detroit Traffic schedule was equivalent to the A.B. Baseline schedule. The data for the front brakes showed good consistency for each of these schedules. The data for the rear brakes showed wider variations. The A.F. Baseline and Final Baseline schedules were run after the 10- and 15-Stop Fade schedules followed by a Reburnish schedule. The right front brake indicated >550F during the fades and had a wear factor approximately 250 percent greater than the one-fourth Detroit Traffic schedule. The left front brake indicated only 300-370F during the 10-stop fade (due to increased intervals) and had a wear factor approximately 120 percent greater than the A.B. Baseline and the onefourth Detroit Traffic schedules. 9.7 RESULTS OF GASEOUS EMISSIONS ANALYSES 9.7.1 Summary of Analytical Methods Attempted Analysis of the gaseous emissions was made difficult by the large quantities of water and carbon dioxide present in the samples. Calculations were made which indicated that the flow through the rotating seals was approximately 0.5 to 0.7 percent of the total. Even when gaseous emissions collections were made during their peak formation such as the fade tests, sufficient quantities of untreated air entered through the rotating seals (downstream of the Drierite and calcium oxide) to present serious analytical interferences. Several different analytical schemes were used and proved only partially successful. Initially, the entire contents of the char coal gas traps were condensed in the gas handling system described in Section 8.5.2. Traps cooled with acetone-dry ice and LN2 were employed. The noncondensable gases proved to be essentially 100 percent air; no other gases could be detected except for a small amount of C02. When all this gas was pumped away and the LN2 removed, the trap was found to contain 100 percent C02. In one instance, the CO2 vapor pressure built up so fast that a stop-cock barrel was shot out of the system. In attempts to trap heavier gases and reduce the danger of a high pressure 9-46 i ! HWBUI0002480 T able 9-32 - Change in L in in g Wear Rate A fte r O p e ra tio n a t In cre a se d Tem peratures Brake f lu id in f r ic t io n m a te ria ls - no re s u lt G H <J Fh S w2 a o }-J 4J x~v cd co do a) s <u CO CM st r--1 rs st a> t--i CM i--> 1--1 2; ' aCO o O O w x in rs CO cd 03 o --i CM i--1 SH ' PQ 55 P. <o w ^ w^ V cd co oo O O st m o 0) s CM CO VO st E- W 12 p s'-' 33 Q 03 <H CM co CM CJ JD PL, i-- o pi pi 33 CO a /-n e fh Cd 03 o S Hw OOO CM O st CM CM CM cx o P -U) ^-N CT\ O CO av CO CO VO CM G* H -5 03 s P2 P ^ 03 CM CM CM Pn s w2 P-i w. CO O O O w x eh rs st rs r--1 CM --I S H^ PQ H tx IS O O M co CO o o Pi cd co 60 rs CO St LO fl-l 03 HW 12 H ' CM CM to co 35 Q 0) O 3) 3, MO Pi Pi PC CO e a/-s xh OOo \--1 CM CO Cd 03 o CO CO CO SH ' VO co uO CO CM CO in CO CM m VO rs rs in rs CM O CT\ co tH rs rs 00 iH O Oo CO st tn i--l CM i--i ooo st CM CM co co CO in o O 00 vO co CM CM co CO CO --i vO VO CO vf> CO CM CM CM CM CO CM i--i in rs CM o co * co X o CM co OOO VO CO st CM 1-- CM in o o rs CPi rs co CM co in o O CM 00 co co CM st rs m rs o CO m vO rs rs vO cr\ st co CO co O CM CM vO rs in o St <J\ CM CM ooo rs i-- m i--l CM i--I oOo vO rs o CO co CO in o O rs co CM co CO co O r o m CO VO o CM CM m CM o vD C-i i--I in CO m CM co rs rs rs CM o 00 o m ooo vo CO o co CO CO ooo m m CM m in m moo CM o vo vO m t--1 cd <--i T3 (X 03 (X FH <3 CM O o CO CO r CO CO co iH i--1 i--1 03 CP O CM co O 00 CO CM VO st in 03 03 i--1 CO 00 co *H CM -3- vO s CO co CO 03 i--1 ah H 03 O j3 aj S 03 H > 60 i--1 CM CO > <3 Ww i-J S P3 i--l Q H-l W W 33 PQ C/3 CJ <3 <3 W CM m rs i--i rs o sf co m i--1 tH i--i CM m 00 co st cr> o CM rH lH rH X St CM st O CM iH tH co co rH CM o G vO rs in m m rs C CM co CO St sf st m 00 X X St st <y\ vO m st CM m co 00 rs co m in sf co co CO VO CT\ rs m co St co co co -X X CO vo in CM CO CO CM in co 60 CM co > <3 . 60 CM co > < # 60 CM > < Ho i--i M o P-, Pi In St H W < Ph * '__y QH W IS H hJ >W CO <1 <3 PQ W 2; i-J <3 w Is LO M <3 Fh PQ CO JJ 3OH U OH 9-47 HWBUI0002481 build-up, the CO2 was allowed to evaporate into the vacuum after the LN2 trap was replaced with acetone-dry ice. The final fraction proved to be mainly C02 and water with traces of hydrocarbons. An attempt to eliminate CO2 and H2O was made by passing the gases from the charcoal traps through Drierite and calcium oxide traps placed between the gas handling and the gas collection systems. This proved partially successful. The traces of gases obtained gave only small peaks in gas chromatography with the thermal conductivity detector (T.C.D.). This problem was overcome by employing the flame ionization detector (F.I.D.) for the hydrocarbon analyses. However, all attempts proved fruitless in detecting anything other than minor amounts of C^'s to C3's. Because of all of these difficulties, it was decided to minimize the effort devoted to future gas analyses. In a last attempt to detect gaseous emissions products the gas collection system was connected directly to the gas injector of the gas chromatograph and the sample was analyzed simultaneously with both T.C.D. and F.I.D. This method proved successful for the fade test products. Although the T.C.D. detected only the major inorganices, the F.I.D. gave several peaks discussed in the following section. This method indicated that the initial tests with T.C.D. alone did not have adequate sensitivity for the organic products in the C4-C7 range where the entire test was sampled. The major drawback of this revised method of direct injection is that it is impossible to analyze the total volume of products obtained. 9.7.2 Gaseous Emissions Detected Table 9-33 lists the data from the analysis of the gaseous emissions from vehicle test 1 while Table 9-34 lists the data from vehicle tests 2 and 3. For vehicle test 1, the activated-charcoal gas traps from the Burnish and Detroit Traffic tests showed traces of Cj to C4 hydro carbons in large quantities of air, carbon dioxide, and water (Table 9-33). No other materials could be found. For the 10-Stop Fade test, the initial gases released from the traps again showed traces of C^ to C4 hydrocarbons when analyzed. The traps were then connected directly to the gas chromatrograph (G.C.) as described in the previous section. The charcoal gas trap samples from the right front brake, 10-Stop and 15-Stop Fade schedule of all vehicle tests were analyzed with a Porapak column. At times, the large amount of air and CO2 gave slightly earlier than normal elution times as compared . to the calibration blend. The same sample was also run on a W-98 sili cone gum rubber column. Table 9-35 lists the elution order of different compounds from the Porapak column. The table lists the compounds used for calibration, those detected, and possible compounds whose elution occurs between the calibration compounds. Table 9-36 lists the elution 9-48 HWBUI0002482 T a b le 9-33 - D ata From A n a ly s is o f Gaseous E m issio n s (V e h ic le T e s t 1) C lu tch C harcoal Traps Not sam pled Not sam pled #t VO M CM CO u- o rH * NO |I| O MU o td <u U rt S3 O 00 o Is 3 SB 3 rl M (O g to 4-j cj fVtO CM a <r cm o S3 O CM O U S3 3 S3 O 0 1-13 3 -H U g 3 4-1 CM O o CM o in fOt * 1-4 d *H ft CM O S3 O CM CJ CJ S3 3 S3 -3 3 l-i 3 3 -H is g 3 4J <u ftrH CO S3 cd CO ucd dH <u <u Td CO ucd do cd O PQ u M U<U cd S e uId a u i--i cd cd ft *\ GJ o CO CM O S3 ft a u ft cd cd u O NO CJ SB OJ S3 H u rod u~ o cd 3 -H M g 3 4-J Td M U<1) u d) <D 4o-1 1"l QU u 0) 4J ft d +J cd cd cd cd s IS CO s <t S3 cj o rH 3 > SB O 00 <t 3 l-i SoB~i a | H CJ ft o CM Td *K S3 (3 CO s cd ** ct 00 S3 " S H 30 <5 CJ rH S SB > <3" S3 CM -3-0 60 o SB 1 ft CJ CJ -rH S3 Td cu 4-1 rH o ft gg cd CO O CM S3 \ u H <3 #\ CM o CJ Td GJ ft4J i--1 O S 3 cd CO t3 cu 4-J rH o ft gg cd CO <D rH S3 P4 cd cd 0) CO (3 H cu0 <u Td CO PQ do cd o Oo CO H o U 0) 4J cd a 4J GO u i--i cd o CO aa CM O S3 O CM CJ ft uo ft cd td u S3 H CJ S3 roH ~ o<u ^ cd u 3 -H M g n) 4J u a) 4-1 cd g * <f CM O 33 O CM CJ a S3 0) S3 - O O l-i 3 3 -H S g 3 SJ Td 0) 4-1 fOt T--1 ft CO I--1 S3 3 o) d) ' ft i--i Td CO f3 cd 00 ft a a (U d) CO o C a) 3 ft rH <u o o 4-J N 3 3 CO CJ 3 3 S3 i-H O ft 3 S3 ft u<u o Td 4-J d) O I--1 g ft g cd CO g Vo u ^1 O -dft CJ Td <u ft4-J r-H O ge cd CO Td (U ft4-J tH O gg cd CO cu rH S3 d CO *H (U d o in d PQ +j a H *H O MH M ft 4J Cd <u u QH Poh 4- c1o a) T3 O3 rH ft G) a H 1--1 pH 0) CO cd < PQ fot 4J co oj 1 Td ifnt f3t d) (3 H rH i--1 cd <D (3 CO ftH cd PQ *M.W. = M o le c u la r W eight 9-49 HWBUI0002483 V e h ic le T est 2 Table 9-34 - Data From Analysis of Gaseous Emissions (Vehicle Tests 2 and 3) Schedule A.B. Baseline Front Disc Brake Charcoal Traps Condensable Gas Trap CO2, air h2o, trace CH4 Nc)t ana! _yzed Rear Drum Brake Charcoal Traps Condensable Gas Traps C02, air h2o, trace CH4 N ot ana lyzed Detroit Traffic 10-Stop Fade 15-Stop Fade Burnish Benzene, tolueme, up to cardanol Some high M.W.* peaks Some high M.W. peaks C02, Air, h2o, trace CH4 C02, air h2o, trace Cj to Cg Some high M.W. peaks CO2, air, H20 trace CH4 Detroit Traffic 10-Stop Fade 15-Stop Fade CO2, Air, h2o, trace CH4 plus high M.W. peaks Some high M.W. peaks Some high M.W. peaks C02, H20, air c2h6 + c2h6 + traces C4 to C8 Not analyzed 1 r *M.W. = Molecular Weight V e h ic le Test 3 9-50 HWBUI0002484 Table 9-35 - Elution Order From Porapak Q Air Carbon monoxide Methane Carbon dioxide Ethane Water Propane Methanol Formaldehyde Acetaldehyde Butane Ethanol Propylene oxide Propionaldehyde Acetone Isopropanol Acetic acide Methyl acetate Propanol Pentane Isobutraldehyde Butraldehyde 2-Butanone Methyl-ethyl ketone Ethyl acetate Isobutanol Hexane Butanol Benzene C7+ Calibrant X X X X X X X X X X X X X X Detected X X X X X X X X X X X X Possible X X X X i r: 9-51 HWBUI0002485 Table 9-36 - Elution Order From W-98 Pentane Benzene Toluene m+p-Xylene o-Xylene Benzaldehyde Phenol 1,2,3TM-Benzene Indene o-Cresol m-Cresol 2,6DM-Phenol 1,2,3,5TM-Benz ene 2,4DM-Phenol n-Dodecane n-Hexadecane Calibrant X X X X X X X X X X X X X X X Detected X X X X X X X Possible X X X X 9-52 HWBUI0002486 order of different compounds from the W-98 column. The table lists the compounds used for calibration, those detected, and other possible com pounds "expected. 9.7.3 Shroud Extract Analysis During normal fade tests, a dark brown cloud of high mole cular weight organics with a pungent odor appeared after the seventh to fifteenth fade stops for each front brake. During the 10-Stop Fade of vehicle test 1, a cloud was observed for the left front brake only as the brake shroud prevented any such release for the right front brake. The inside surface of the shroud was covered with a "tarry" layer which presented the same pungent odor after all fade tests. The G.C. chromatogram of the shroud extracts from the 10Stop and 15-Stop Fade tests of all vehicle tests were chromatographed on an OV-101 column. The results indicated that the shroud extract contained organics with molecular weights higher than n-dodecame [mole cular weight 170 and boiling point 214C (420F)]. Cardanol is present in all samples. 9-53 HWBUI0002487 SECTION 10 ESTIMATED EMISSIONS FOR CARS AND TRUCKS 10.1 APPROACH The purpose of this section is to develop an estimate of total brake and clutch emissions from the U.S. population of motor vehicles in use on the roads and highways. Inputs include all of the test re sults described earlier in this report, together with a supplementary data base on U.S. motor vehicle population and usage and on friction material usage and consumption obtained from other sources. Because of the uncertainties in all of the data available, emis sions can only be considered as a broad range, rather than as a single value, with the expectation that the true average emissions for both individual vehicles and the motor vehicle population in general will be within the range of values calculated below. 10.2 SUPPLEMENTARY DATA BASE The data base reported in the accompanying tables was obtained from a wide variety of trade publications and from industry and govern ment contacts. Credit is given to the sources of major contributions. Although they vary considerably in quality, the data are, in general, adequate for the purpose of estimating total annual emissions, because emissions themselves can vary so widely from vehicle to vehicle, from operator to operator, and from locality to locality. 10.2.1 Number of Vehicles in Use Table 10-1 lists the total U.S. motor vehicle population as estimated by both Department of Transportation (DOT) (National Highway Traffic Safety and Federal Highway Administration)(10-1)* and by the R. L. Polk Company/10-2) The Polk census is said to be more precise because it eliminates multiple registrations of the same vehicle. Polk considers that DOT estimates may be as much as 13 percent too high. Both Polk and DOT omit military vehicles, but include all other motor vehicles that are state-licensed including police cars, school buses and the like. Independently, the General Services Administration reports that 190,000 Department of Defense vehicles are in use world-wide. Table 10-2 lists the U.S. braked trailer population and Table 10-3 lists the U.S. motor vehicles with dry clutches. if Numbers in parentheses refer to References at end of section. 10-1 HWBUI0002488 Table 10-1 - U. S. Braked Motor Vehicle Population Private Passenger Cars (R.L..Polk Estimate 7-1-71: 83,137,324) D.O.T. Estimate 7-1-72: 96,397,000 Trucks: (R.L. Polk Estimate 7-1-71: 18,462,287) D.O.T. Estimate 7-1-72: (Includes 897,456 Truck Tractors) 20,800,000 Light M Medium Heavy (Under 6000 lbs GVW* 1 6 -10,000 /10-14,000 114-16,000 j16_19>500 119.5-26,000 (26-33,000 (Over 33,000 12,109,000 4,952,000 125,000 125,000 645,000 1,290,000 395,000 1,165,000 Buses: D.O.T. Estimate 7-1-72 403,000 Miscellaneous Motor Vehicles: motor cycles. etc. D.O.T. Estimate 7-1-72 3,787,000 Total Motor Vehicles in Use** D.O.T. Estimate 7-1-72 121,400,000 *Gross vehicle weight; estimates from 1970-71 sales data extended by this study. **This total includes approximately 1,5 million public vehicles (450,000 cars, 920,000 trucks, and 185,000 buses), but does not include 190,000 Department of Defense vehicles in worldwide use. Table 10-2 - U. S. Braked Trailer Population Class Recreational Travel Camping Boat Miscellaneous (Animal, Utility, etc.) Commercial Utility Trucking Full Semi Number In Use 1,800,000 1,200,000 2,500,000 1,700,000 1,000,000 278,000 1,552,000 Percent With Brakes Number With Brakes 95 1,710,000 20 240,000 35 875,000 80 1,360,000 60 100 100 600,000 278,000 1,552,000 6,615,000 10-2 HWBUI0002489 Table 10-3 - U. S. Motor Vehicles With Dry Clutches Vehicle Type Passenger Cars Domestic Imported Trucks and Buses Light Medium Heavy Number Of Vehicles In Use 87,100,000 9,300,000 17,100,000 2,600,000 1,600,000 Percent Containing Clutches With Asbestos 15 90 75 95 100 Number Of Dry Clutches 13,064,000 8,370,000 12,900,000 2,470,000 1,600,000 38,404,000 Source: estimate made from industry canvas. Table 10-4 - Yearly Additions to U. S. Braked Vehicle Population New Vehicles Placed in Service (Calender Year 1972): Passenger Cars: Net U.S. Production Net Imports Total 9,312,502 1,593,605 Trucks: Total (estimated) Estimate Total: 10,906,107 2,600,000 13,506,107 Source: Wall Street Journal 10-3 HWBUI0002490 The DOT estimate of 117-million total vehicles, including 20.8-million trucks, is probably sufficiently accurate for estimating brake emissions for 1972. It should be noted that the net growth in vehicles each year is given approximately by the sum of U.S. cars, trucks and other vehicles sold plus imports sold (all adjusted for exports) (Table 10-2) less annual retirement of about six- to sevenmillion cars and one-million trucks.(10~3, 10-4, 10-5) The scrappage rate tends to rise and fall with the sales rate and may well have been closer to eight-million cars than to seven-million for 1972. Figures are not yet available. Assuming a total retirement of nine-million vehicles, the net growth in vehicles for 1972 was approximately 4.5 million: 10.9 million cars sold + 2.6 million trucks sold 13.5 million total - 9.0 million scrappage 4.5 million net growth The net growth in car and truck registrations from 1961 to 1971 averaged 3.12-million vehicles per year and will probably tend to . follow the same patterns over the near future. 10.2.2 Motor Vehicle Usage Table 10-5 shows the approximate total number of miles traveled by all U.S. motor vehicles to be 1.249-trillion miles for 1972, according to a recent DOT estimate(10-1). This total is said to be grow ing at a 4.5- to 5.2-percent annual rate. To accomplish this mileage, about 100 billion gallons of motor fuel were consumed for an overall average of a little over 12 miles per gallon. The detailed breakdown of mileage by vehicle type and roadway type (Table 10-6) was adjusted for 1972 total mileage from 1970 DOT data. Roadway mileage and usage are further identified in Table 10-6 by type of surface and by locale, whether urban or rural. More insight into vehicle usage is given in Tables 10-7 and 10-8 where motor vehicle usage is classified by the purpose of the trip and length of the trip. All of these factors cast light,to some extent, on the degree of vehicle brake usage and emissions. 10.2.3 Estimates of Brake Usage and Abuse Brake usage varies widely from gentle to abusive. In the following paragraphs the term abusive braking will be defined and its extent will be explored. i i: |- |: j 10-4 HWBUI0002491 Table 10-5 - Motor Vehicle Usage by Vehicle Type and Roadway Type of Vehicle Passenger Cars Motorcycles and Motorbikes Buses: School Commercial Trucks-Total Single Unit Trucks Truck Combinations Total-all vehicles Billions of Vehicle Miles Traveled By Road Type and Location* Local Rural Main Rural 109* _* 337 - Urban 534 - Total Mileage 980 10.9 Average Annual Mileage Per Vehicle (1970) 9,978 3,605 --35 119 91 6. - 245 13,306 7,274 32,591 11,450 33 88 77 198 9,807 1.9 150 31 464 14 635 47 1,249 41,903 10,076 *Blanks indicate insufficient data. Other data are adjusted for 1972 total analysis from 1970 D.O.T. breakdown on best-estimate basis. **Weighted total average Table 10-6 - Type and Usage of Public Roads Road Type Miles of Road 1. By Classification Interstate (42,500 authc rized) Rural 30,000 Urban 5,000 Primary-Highways Rural Urban 705,000 -- -- Secondary Highways Rural Urban 2,990,000 -- -- Municipal Streets 2. By Location Rural Urban 3,169,000 561,000 3. Bv Quality Surfaced Non-surfaced 784,000 2,946,000 Grand Totals 3,730,000 Billions of Annual Miles of Vehicle Travel (1972 estimate) 117 113 222 153 275 .92 277 -- -- -- -- 1,249 Source: World Almanac, 1973 and 1972 Automobile Facts and Figures; adjusted l i. ; i; : 10-5 i HWBUI0002492 Table 10-7 - Motor Vehicle Usage by Purpose of Trip Purpose of Trip Earning a living (driving to work, etc.) Family business Educational Social (includes vacations) Proportion of Total Use 42.1% 19.6 5.0 33.3 Average Trip Length 10.2 miles 5.5 4.7 13.1 Table 10--8 - Motor Vehicle Usage by Length of Trip Length of Trip (one way) Percentage of Trips Under 5 miles 54.1 5 -10 19.6 10-15 13.8 15-20 20-30 4.3 4.0 30-40 1.6 40-50 0.8 50-100 Over 100 1.0 0.8 100.0 Average trip length = 8.9 miles Proportion of Total Use Miles, Percent 11.1 13.8 18.7 9.1 11.8 6.6 4.3 7.6 17.0 100.0% DOT Preliminary Survey 1969-70 | i 10-6 HWBUI0002493 Abusive braking, with the organic-bonded friction materials used almost universally on motor vehicles today, implies heating the bulk friction material to temperatures above about 500 to 600F. At these temperatures, the phenolic resin binders gradually pyrolyze, weaken, and become brittle, allowing wear to increase rapidly. Because degrada tion is a time-temperature phenomenon, total brake-use history is important. following: Overheating can be caused by one or a combination of the Dragging brakes caused by system defect or operator error. Repeated use of brakes to retard the vehicle on a long downgrade. Repeated high-speed stops. Maiadjustment. Severe vehicle overloading. Partial brake system failure, overworking one or more brakes. The degree of overheating is further influenced by ambient air tempera ture, wind velocity and relative direction, air flow around the brake (design) and use of engine braking to assist the foundation brakes. In predominantly flat terrain and on most interstate-quality highways with gentle grades, the need to use brakes to overcome gravity is minimal. Here, brake overheating is rare and usually comes from abusive use as in the fade tests carried out under this program. In very hilly or mountainous regions, overheating can occur quite easily as was demonstrated by a team of professional test drivers on cross country runs in 1967. Zechel, et.al.(10-6), described a study where a group of professional test drivers drove a total of 44,000 miles with six 1966 U.S. passenger cars in order to perform an "Automotive Brake Evaluation Under Customer Usage Conditions." Results of this trip (see Table 10-9) show that abusive temperatures were reached only in mountainous regions (and in the hot and hilly Los Angeles traffic for one disc brake car). Intentionally aggressive braking behavior in Detroit, St. Louis, New York, and other areas of heavy traffic failed to produce excessively high brake temperatures. The proportion of mountain driving mileage on these trips was as follows: Eastern leg, 7.8% of the total; Western leg, 7.4%, of which 2.4% was in Los Angeles traffic. Repeated runs were intentionally made over some of these mountain roads for increased severity of the test. Because we have shown that total emissions tend to increase with abusive braking, its prevalence will be explored further. Table 10-10 gives the results of an attempt to estimate the proportion of U.S. driving that takes place in predominantly hilly or mountaninous areas where long steep grades are common. Basic data on 10-7 HWBUI0002494 Table 10-9 Results of Cross-Country Brake Test Trip Local Eastern Trip: Highway and rural N.Y. City traffic Peters and Potts Mountain Western Trip: Los Angeles traffic Other city traffic Highway and rural in hilly areas Rocky Mountains Maximum Brake Temperatures (F) Drum Brakes (4 cars) Average Peak Temperature Maximum Disc Brakes (2 Cars) (not used on Eastern Trip) Car "E" Car "F" 250 343 471 421 300 526 350 390 580 465 360 410 650 -- -- 320 300 300 430 -- -- 630 410 375 748 Source: SAE Report SP 338 (1967) *Report covers 44,180 miles of coast-to-coast driving by professional drivers using 6 different makes of passenger cars. Test were deliberately aggressive to generate maximum temperatures to check brake fluid boil. Car "F" ran hotter on the front wheels by more than 300F compared to the rear wheels. Table 10-10 - Where Vehicles Are Used Plains States - 30, with 79 million vehicles Partly mountainous States20, with 38 million vehicles Estimated Usage: 70% flat terrain 30% steep grades Totals Percent of all driving Type of Braking Expected Estimated Total Annual Usage Predominantly Flat Terrain with Infrequent, Gentle Grades Steep Hills to Mountainous with Frequent, Long, Steep Grades 777 billion miles Minor Amount 328 144 billion miles 1,105 88.5% Normal 144 11.5% Severe to abusive brake use may be needed. Source: Estimate made for this study. 10-8 i j- i i HWBUI0002495 miles driven in each state are readily available from DOT. The percentage of driving that takes place in mountainous areas, where abusive brake use may be necessary, is estimated to be 11.5% of all driving. The esti mate is subjective, to a large degree, and may be high, but no better data have been located. Table 10-11 lists some additional U.S. driving characteristics recently gathered by the American Petroleum Institute. Table 10-12 lists typical numbers of brake applications per mile as measured under this and other brake programs. It is an illustrative rather than an exhaustive study of the subject. Figure 10-1 illustrates the frequency of high-g stops, as determined at the Highway Safety Research Institute of the University of Michigan (10-7). This shows that 50 percent of all stops are less than 0.15 g (4.8 fps^), 99 percent are less than 0.36g (11.6 fps^) and 99.9 percent are less than 0.5g (16 fps^). Carpenter^lO-8) confirms one point on the University of Michigan curve and lends credibility to the data. 10.2.4 Asbestos in Friction Materials Almost all of the commercial friction materials used for braking passenger cars and trucks today are based on the unique strength and thermal stability of natural asbestos (chrysotile) fibers. This has been true for over fifty years. Table 10-13 shows that the asbestos content of friction materials averages about 60 percent. Table 10-14 lists the weights of the brake lining materials used in this program. They are typical of the weights of friction material used on passenger cars today and will be considered average for the cal culations which follow. Compilations of data on brake rotors and friction materials usage appear in Tables 10-15, and 10-16, and 10-17 respectively. Based on the data in Table 10-16, the total amount of asbestos contained in all of the automotive brake friction materials sold each year is calculated to be about 103 million pounds which corresponds to ''<118 million pounds used prior to grinding. The two most knowledgeable sources(l0"9,10-10) report that 90 to 120 million pounds per year are used. A report from IIT Research Institute suggests that 67 million pounds are used annually in brake linings (10-11). A Bureau of Mines Report t,asecj on 1968 data credits only 50 million pounds of asbestos to automotive use, a number the authors find to be much too small. Ignoring the two latter values, the three estimates of asbestos contained in linings range from 90 to 120 million pounds. Using the calculated value of 118 million pounds and assuming a 15-percent grinding and drilling loss, the maximum amount of asbestos incorporated in brake friction materials is taken to be 103-million pounds per year. A compilation of data on clutch friction materials usage is given in Table 10-17. Based on these data, the total amount of asbestos contained in all automotive clutch friction materials sold each year is calculated to be about 4.5--million pounds. 10-9 HWBUI0002496 Table 10-11 - Some U. S. Driving Characteristics City Average Daily Mileage Average Speed Daily Driving Time Chicago 35 22.9 88 minutes Houston 36 25.2 85 minutes Los Angeles 32 24.7 70 minutes Minneapolis-St. Paul - 26.3 - New York 38 25.4 85 minutes Detroit Traffic test (this study) 110 17 7.0 hours Source: American Petroleum Institute (except for last item) Table 10-12 - Brake Applications Per Mile Source Type of Traffic This study University of Michigan Highway Safety Research Dept. Zechel et al, (SAE SP-338) Detroit traffic Ann Arbor and Michigan traffic Cross Country Eastern trip (4 cars) Western trip (4 cars) Source: as noted No. of Pedal Applications Per Mile Total No. of Miles 4.4 1.43 8,091 1.43 1.9 0.98 6,255 8,400 24,400 No. of Stops Per mile 2.6 - Average Decelera tion fps 4 to 6 - - 10-10 I HWBUI0002497 Table 10-13 - Asbestos Content of Automotive Friction Materials Vehicle and Brake Type Passenger Cars Drum Linings Disc Pads Truck Segments Clutch Facings "Average composition" for use in the calculations in this report. Average Percent Asbestos Content by Weight* 68 + 5 60 + 5 55 + 5 50 5 60 *Sources Estimates made for this study. Figure 10-1 - Frequency Distribution of Braking Decelerations 10-11 HWBUI0002498 Table 10-14 - Brake Friction Material Weights for Test Vehicle Vehicle Test Friction Material Description Inner Pad 1 Outer Pad Primary Lining Secondary Lining Inner Pad Outer Pad 2 Primary Lining Secondary Lining Inner Pad 3 Outer Pad Primary Lining Secondary Lining Average Per Wheel Average Per Axle Average Per Car Weight of Friction Material (gms) Front Disc Pads Rear Drum Linings Right Left Right Left 95.8 93.8 - - 117.4 116.5 - - - - 108.1 106.4 - - 189.6 189.3 106.4 106.1 - - 105.0 105.6 - - - - 104.9 102.4 - - 181.3 180.2 115.3 113.0 - - 152.3 160.3 - - - - 113.2 113.7 - - 153.5 167.7 231. 3 311.3 462. 6 622.6 1085.2 Table 10-15 - Number of Newly Surfaced Rotors Used Each Year Rotors replaced Rotors turned or ground New rotors on new vehicles (1972) Total rotors with new surfaces per year Percent of vehicles with new surfaces: ____ 68^570,000-----4 x 123,323,000 ^00 _ ' 4,700,000 9,810,000 54,060,000 68,570,000* *Source; Stanford Research Institute Data. 10-12 i HWBUI0002499 T able 10-16 - C o m p ila tio n o f A nnual M otor V e h ic le Brake F r ic tio n M a te ria l Usage 4-> o o o o o o o O P o o o o o o oO 60 o o o o o o o o H 0> A 0k 0k 0k 0k o o o o o o oo E2 CO vO o o o o o o vO 0P VO n- t-H 1"1 I-1 0k ^ 0 rH rH o CO ^ 0k 0k 0k 0k <d W <T> . co CM co O CM o rH U CM rH CO rH r- O EH ~N CO U i-H rC U '-S 60 Q) O 00 CO O O O H PH W i--i t-H rH m in rH a) *h rH &d !=> 4-> H t3 d a) d> CO d 4J Cd c PI n O 0) >9 <g p a co CO co 4-1 +J 4-> <u a) CU CO CO CO CO co CO +J 4-> 4-> (1) <u a> cu cu CO C/3 CO rH rH a) Oao a) cu <u rl H H rH i--} rH p ,d p <3 <3x X X (U cu cu <C > > > oooo oo o o oo oo oooo oo 0k 0k 0k 0k 0k o o o o o rH VO o o co r-* 00 vO co n*. VO a 0k a\ CM CM CO 1 1 i i 1 i i i >> CO CO co CO 0 rH rH 00 60 00 60 o a gc _c a a 3 d rl CO 3O 6O H H H H 4-1 rl rH p addd Ph a d O co Jh (0 H H H H o t>0 Q 00 i-5 J j i-5 4- 4J C P U C CO d -H M H g g g g CO o o c cd a P P P p a) |H Cd H *H a) *h Pi n Pi u pH PH PH pi ij Q Q O a 0) rH u Oa H 00 ,d d <u CD > CC a: Pi cC cd PH u 4J rX .c o 00 p H H i-5 H CO p PQ O g 3 Pi H O T3 3 a) J-t SH rX >o cd 3 CU M EC H * o CO 4J 3 <U o 0) 0k d to cd Pi rH 0) rH rH CU -H a cd co u H H a 0* CU 4- P 4J 4-J CO d H to cd H3 -U Pi cd cd P a cu cd 4J 4-J p CO 4- rl CO 4- rH CO cd d rH M t-l 0) rd 4J O cd Pi S cd CU d co o cu H pc5 4J a H u uo pH MH e d * O 4J U CO TO -a d cu cd 4J cd > rH 3H o cd CJ O a) o M p o CO 10-13 j I HWBUI0002500 Table 10-17 - Compilation of Annual Clutch Friction Material Usage Vehicle Passenger Cars: Domestic Imports Totals Trucks: Light Medium Heavy Totals New 900,000 1,350,000 2,250,000 Annual Clutch Facing Use Replacement New Rebuilt Total Units Installed 1,760,000 280,000 1,710,000 2,580,000 4,290,000 Weight Per Unit (lbs) 0.88 0.60 1,500,000* 290,000* 190,000* 1,980,000 935,000 1,170,000 3,100,000* 600,000* 385,000* 4,085,000 1.26 1.56 2.26 Total Clutch Weight (lbs) 1,500,000 1,550,000 4,000,000 950,000 680,000 8,740,000 *Estimated for ratio each truck size with clutch (Table 10-3) Source: Estimated from market studies and industry sources for this study Table 10-18 - Relative Brake Relining Frequency Locale Northeast South Large cities (over 1 million) Medium cities (100,000-500,000) Small cities (under 25,000) Rural areas Average - U.S. No. of Wheels Relined Per 100 Vehicles Each Year 74 '49 82 93 54 44 60* *Approximately equivalent to relining all four wheels on 15% of all vehicles each year. 10-14 ii HWBUI0002501 Frequency of brake relining is another measure of lining use. Table 10-18(^-13), Multiplying the relining rate of 15 percent per year times the vehicle population of 123 million and adding the number of new vehicles registered last year, 13.5 million, gives 32,100,000 ve hicle sets as an estimate of annual friction materials usage. This is a somewhat smaller number than that derived in Table 10-16, probably be cause insufficient emphasis was given to medium and heavy trucks in these data. One word of caution is in order at this point: the amount of fric tion material worn away each year is significantly less than that which is installed for a variety of reasons to be discussed in the next sub-section. 10.2.5 Amount of Friction Material Actually Worn For several reasons, the amount of brake lining (and asbestos) worn away during a year is significantly less than the total amount installed on vehicles. Brake linings are seldom completely worn away (due to the geometry of brake systems, perhaps 10 percent of the lining is still left on a brake when lining wear-through occurs). Linings are usually replaced in sets rather than piecemeal. As a consequence of this and other actions, about 25 percent of all brake linings is discarded at relining time. Furthermore, millions of vehicles are retired from service each year, many with brake linings that are far from worn out. Perhaps the average retired vehicle has about 1/3 of its usable lining remaining, or 1/3 x 90 percent + 10 percent = 40 percent of the installed weight of the lining. Combining these figures, the amount of brake friction material worn away each year is calculated as follows: Installed weight each year Less Discarded on relining - 25% Less discarded on vehicles retired (7 million cars x 2.38 lb x 40% = 6.7 + (0.75 million light trucks x 5 lbs x 40% = 1.5)+ (0.10 million medium trucks x 15 lbs x 40% = 0.6)+ (0.15 million heavy trucks'x 41 lbs x 40% = 2.4)+ (miscellaneous = 0.5) Amount worn away, annually Total Brake Friction Material 171.4 million lbs 42.7 million lbs 11.7 million lbs 117.0 million lbs 1 l yi 10-15 HWBUI0002502 Employing the same reasoning for clutches, the amount of clutch material worn away is calculated as follows: Total Clutch Friction Material Installed weight each year 8.76 million lbs Less discarded on relining - 15% 1*31 million lbs Less discarded on vehicles retired 0.89 million lbs (1.5 million cars x 0.75 lbs x 40% = 0.45)+ (0.50 million light trucks x 1.26 lbs x 40% = 0.45)+ (0.10 million medium trucks x 1.56 lbs x 40% = 0.06)+ (0.15 million heavy trucks x 2.26 lbs x 40% = 0.13) Amount worn away, annually 6.56 million lbs Summarizing, the combined total of brake and clutch fric tion material worn away annually is 117.0 + 6.6 = 123.6 million pounds. Assuming that the average asbestos content is 60 weight percent, the amount of asbestos worn away as friction material wear debris is ^74 million pounds. 10.3 INTERPRETATION OF TEST RESULTS 10.3.1 Factors Influencing Rate of Emissions Emissions measured during this program were generated by an intentionally severe schedule, one that is used by the industry to accelerate brake testing. Consequently, emissions were undoubtedly gen erated at a rate somewhat above average for the population as a whole. Emissions per mile for a given brake can vary, for ex ample, with the following: Composition of the friction material Composition, metallurgical structure and hardness of the cast iron rotor (drum or disc). Surface roughness of the rotor. Previous use (history) of the friction material - primarily thermal history, both recent and overall. Third-body contamination of the sliding interface by road dust, wear debris, rain water, salt and the like. Vehicle usage - miles driven: urban, surburban, rural, expressway. Vehicle weight, including passengers and cargo. Vehicle speed at beginning (and end) of stop. 10-16 HWBUI0002503  Deceleration (average or typical, as well as instantaneous). Frequency of braking due to traffic, terrain, and driver habits. Interface temperature, sliding speed, and unit load on the friction material. Roadway surface and grade. Air temperature, wind velocity, and wind direction relative to direction of vehicle motion. Brake design, including brake balance front to rear, and swept area in relation to vehicle weight and horsepower. Vehicle design including brake cooling adequacy. Operator habits and behavior - riding the brake, gentle stops versus rapid deceleration versus abusive deceleration, etc. Brake adjustment and maintenance - balance from side to side and front to rear. Some of the above factors are not independently variable, but are mentioned for emphasis. Tires (radial versus bias belt), engine-transmission brak ing, suspension, and aerodynamic shape also influence vehicle braking to some extent. During this test program, emissions were measured under the following sets of conditions: Seven different friction material formulations made by five different manufacturers. , Disc versus drum brakes. New friction materials versus used friction materials. Freshly machined rotors versus polished (used) rotors. Moderate use versus abusive use under seven driving categories. Emissions were examined for asbestos content using the best available techniques and confirmatory tests were conducted at two additional laboratories (Battelle and Johns-Manville). The program results for asbestos emissions were slightly higher and were chosen ex clusively to calculate a maximum asbestos emissions estimate. 10-17 HWBUI0002504 10.3.2 Weighting Factors 10.3.2.1 Distribution Weighting Factors 10.3.2.1.1 Location of Wear Debris Disc brakes are designed to allow air to pass through them freely. Because of this open design they retain less wear debris than drum brakes. Consequently, the surfaces samples for the disc brake are essentially emitted samples. As indicated earlier, the disc brake emission collector collected all disc brake wear debris. It was frac tionated into the following samples: airborne: these are the light particles which remain in the air-stream as it progress through the collector - in this program this material was collected on filters - and the particulates were presumed to be representative of disc brake wear debris which may remain airborne in the atmosphere for a period of time. surfaces: these are the heavier agglomerates which quickly fall to the ground or nearby surfaces - in this program this material was deposited on the bottom and inner surfaces of the shroud and on the brake parts. sump: these are the particles of brake debris remaining in the brake on the lining surfaces, in the rivet holes, and in the caliper. The drum brake emissions collector also collected all of the drum brake wear debris. It fractionated into the following samples: airborne: there are light particles which remain in the air-stream as it passes through the collector in this program this material was collected on filters - and the particulates were presumed to be representative of disc brake wear debris which may remain air-borne in the atmosphere for a period of time. surfaces: these are the heavier agglomerates which quickly fall to the ground or nearby surfaces - in this program this material was deposited on the inner surfaces of the shroud and on the brake parts. sump: these are the particles which remain on the brake drum and lining surfaces and in the rivet holes. 1:. I i:;;. h, i L, 10-18 HWBUI0002505 10.3.2.1.2 Comparison of Normal versus Shrouded Brakes The effect which the shrouding produced on the quantity and distribution of the emissions for the right front disc brake and the right rear drum brake can be estimated as follows. The data given earlier in Tables 9-9 and 9-10 for vehicle tests 1 and 2 show that the individual sump samples of wear debris collected after each schedule for the normal disc brake are com parable in amount to those of the shrouded disc brake. The totals are also comparable. On the other hand, the data given in Table 9-11 show that the cumulative sump sample, taken at the end of vehicle test 3, is significantly less than the total obtained from the summation of each individual schedule. Further, the total sump sample for vehicle test 3 is comparable in value to the average value collected for each schedule alone of vehicle tests 2 and 3. Consequently, the sump sample appears to achieve and maintain an average value independent of the mileage. When the debris is removed, the equilibrium value appears to reestablish itself. The surfaces samples for the open brake behaves in a similar manner. The surfaces samples of the open brake were smaller (only 2.7 to 4.5 percent of the shrouded brake samples). The total surfaces sample for the left front brake, taken after vehicle test 3, was comparable to the maximum single value (for the Detroit traffic test the longest mileage test), when individual values are taken after each schedule. Here again, an equilibrium value appears to establish itself. The best estimate for the distribution of the disc brake wear debris when the brake is new can be made by considering the Burnish data of Table 9-13. After 174 miles, the distribution of emis sions can be estimated. 9% returned in sump 2% on surfaces - calculated from open brake 4% airborne - measured at shrouded brake 85% road dropout - balance The best estimate for the distribution of the disc brake wear debris for a use brake can be made from the data from vehicle test 3. After 2632 miles, the distribution from the open disc brake is as follows: 1.5% retained in sump 1.4% retained on surfaces - evaluated from open brake 7% airborne - average from shrouded brake 91% road-dropout - balance As the brake is used, the amount retained in the brake remains constant, but the percent values for retention decrease to less than one percent. 10-19 HWBUI0002506 The average for the airborne samples for 8100 miles is six percent. Thus the distribution for the well-used disc brake can be estimated as follows: 1% retained in brake 6% airborne 93% road drop-out No major change in these relative proportions is expected for continued operation beyond the 8100 miles of the test to wear-out. The data given in Table 9-9 for vehicle test 1 show that the amounts of sump sample debris collected after each schedule for the normal (unshrouded) drum brake are somewhat comparable to those of the shrouded drum brake. The totals are also somewhat comparable, al though the total for the shrouded brake seems slightly low. The surfaces samples are slightly less for the normal drum brake and the overall totals for the entire test show that approximately 53 percent of the surfaces samples remained in the brake. The negative numbers for the sump samples are real. Since 80 percent of the sump sample from the previous schedule was reintroduced onto the brake drum prior to each following schedule, it was possible for the sample to redistribute in such a way that the sump sample decreased to a value less than the original amount added. The surfaces sample showed a corresponding increase so that the total debris generated was never a negative value for the shrouded brake. The normal drum brake could and did occasionally show a negative value for total debris. The rear brake data given in Table 9-10 for vehicle test 2 was for the continued testing of the used original equipment rear brake linings from vehicle test 1. Thus each rear wheel started with the 80 percent sump debris addition (from vehicle test 1). The total debris collected from the shrouded brake for vehicle tests 1 and 2 were comparable. The total debris from the open brake was less for vehicle test 2 than for vehicle test 1. The only explanation feasible is as follows. The amount in the sump debris of this test is at an equilibrium level and there is no more physical room for the material. Thus it falls out of the wheel immediately and less debris is distributed to the sur faces in vehicle test 2 as compared to vehicle test 1. The best estimate for the normal distribution of drum brake wear debris when the brake is new can be made from the Burnish data of Table 9-13. After 174 miles, the distribution of emissions can be estimated as follows: 16% retained in sump 29% retained on surfaces - calculated from open brake 1% airborne - measured at shrouded brake 54 road drop-out - balance j; E 10-20 HWBUI0002507 The best estimate for the distribution of drum brake wear debris for a used brake can be made by extending the data in Table 9-11 for vehicle test 3. For 2632 miles, the open drum brake accumulated the following: Sump debris: 0.97 gm (9% of weight loss) Surface debris: 1.91 gm (17% of weight loss) Supplementary data from employees' vehicles and vehicles submitted for brake relines in local shops indicated the following trends for friction materials with an average life of 40,000 miles: After 5000 miles: Sump debris: ^2 gm (vl0% of friction material wear) Surfaces debris: ^4 gm (vL8% of friction material wear) After 25,000 miles: Sump debris: ^3 gms (^5% of friction material wear) Surfaces debris: ^10 gms (''d.2% of friction material wear) After 40,000 miles: Sump debris: ^4 gms (^3% of friction material wear) Surfaces debris: vl2 gms (^10% of friction material wear) The trends are graphically illustrated in Figure 10-2. Not all of the wear debris found in a brake is from the friction material itself. The analysis of a few samples for iron content indicated approximately 5-15 percent by weight is iron oxide which came from the rotor wear. In this program, no driving or braking took place on dusty or salted roads. However, the average consumer would drive under these conditions at times and this would slightly alter the com position of the retained brake debris. For purposes of estimation, the average vehicle has linings that are 50 percent worn (at the 20,000 mile point on Figure 10-2) and the contributions to the atmosphere can be estimated in the following manner: 13% retained in sump 6% retained on surfaces 2% airborne 79% road drop-out 10-21 HWBUI0002508 Figure 10-2 - Variation of Wear Debris Retention with Mileage (for Rear Drum Brakes) 10.3.2.1.3 Summary As a result of the above analysis, it is proposed that the test results he weighted in the following manner for the determination of asbestos emissions distribution from a passenger vehicle: Simulation Test Schedule Required Disc Brake: New friction couple After 2632 miles After 8100 miles After 20,000 miles Burnish Composite Composite Composite Drum Brake: New friction couple After 2632 miles After 20,000 miles Burnish Composite Composite Airborne (Atmosphere) 0.04 0.07 0.06 0.06 0.01 0.02 0.02 Road Vehicle Drop-Out Retention 0.85 0.91 0.93 0.93 0.11 0.03 0.01 0.01 0.54 0.72 0.79 0.35 0.26 0.19 10-22 HWBUI0002509 10.3.2.2 Severity Weighting Factors 10.3.2.2.1 New Versus Burnished Friction Materials The vehicle test results indicated that rerunning the rear drum linings from vehicle test 1 on vehicle test 2 gave lower asbestos emissions. The A. B. Baseline and Detroit Traffic schedules of vehicle test 2 showed lower asbestos than the corresponding schedules of vehicle test 1. The same trends were shown for the clutch. Consequently, it is necessary to include a certain percentage (weight) of Burnish schedule results in the calculations of total asbestos emissions. Burnish tends to be completed after approximately 200 miles of test driving schedules, but stability of surface finish may not occur until somewhat later on private passenger cars. Assuming that the maximum is 400 miles and that the brake friction materials have a 30,000-mile average life, the fraction of driving is: 400 30,000 0.013 This is the severity factor used for determining the extent of Burnish asbestos emissions. During the next period of driving, the friction materials are still moving toward their stabilized level for light to moderate duty. The test vehicle used approximately 350 miles. Assuming that up to 400 miles are required, the fraction of driving is kept the same as for the Burnish factor above. Thus the A. B. Baseline factor is also 0.013. 10.3.2.2.2 Severe or Abusive Braking The results presented previously show conclusively that total emissions increase during and immediately after severe (abu sive) braking (the fade tests). While it is necessary to test brakes under these abusive conditions so that they will function in emergencies, it is not common to operate vehicles in the abusive manner that generates high emissions. Data reported in Section 10.2 indicate that severe braking occurs less than 10 percent of the time, nationwide. Most of the severe braking takes place in comparatively remote or lightly populated mountain areas where 11.5 percent of all driving takes place (Table 10-10). With experience, most of the local drivers tend to drive in a manner where they refrain from fading the brakes. Assuming that the severe braking encountered under these condi tions occurs a maximum of 10 percent, the fraction for severe braking can thus be calculated: 0.10 x 11.5 = 1.15 percent of all driving The amount of hot-rodding followed by severe braking, is not known. Nor is the amount of driving known when a foot is kept on 1. 10-23 HWBUI0002510 the brake at all times. A further action which causes emissions caused by severe brake friction material dragging is on the unreleased emergency brake. To account for all of these actions, the above fraction of driving should be increased to a 2.0 percent maximum. Thus the factor for the 10-Stop Fade consideration is 0.020. 10.3.2.2.3 Renewal of Friction Surfaces The data given in Section 9.6 show that the wear rate of friction materials is increased after high-temperature operation. Consequently, the calculations of asbestos emissions requires the inclu sion of an A. F. Baseline contribution. The test vehicle used approxi mately 350 miles; under less than moderate duty, approximately 600 miles would be required. The fraction of driving which gives the A. F. Baseline factor can be calulated: 600 30,000 0.020 10.3.2.2.4 Normal Brake Operation Most braking takes place under normal or low tempera tures (less than 300F). As indicated earlier, the Detroit Traffic Test is an accelerated vehicle test and is more severe than normal driving throughout most of the country. The temperatures actually encountered would be typical of warmer climates and hilly areas such as those found in the Los Angeles area. In any event, the Detroit Traffic Test used in this program gives the maximum asbestos emissions expected for normal brake operation. 10.3.2.3 Summary As a result of the above analysis, it is proposed that ; test results be weighted in the following manner for the determination maximum total asbestos emissions for a passenger vehicle: Simulation New friction couple Surfaces preparation Moderate braking Severe braking Renewal of friction surfaces Test Schedule Required Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline Weighting Factor 0.013 0.013 0.934* 0.020 0.020 *Determined by difference 10.3.3 Asbestos Emissions Per Vehicle 10.3.3.1 Total Asbestos Emissions Estimate The total asbestos emissions for the test vehicle can m: | ! 10-24 i i HWBUI0002511 be estimated from the test results given in Section 9 and the severity weighting factors. The calculation is summarized in Table 10-19. The asbestos emissions for the disc brakes, drum brakes, and clutch from each vehicle test were averaged and used with the Burnish, A. B. Baseline, Detroit Traffic, 10-Stop Fade, and A. F. Baseline schedules. The cor responding severity factors were used to calculate the weighted emissions. The sum of the weighted emissions is the total asbestos emissions expected from an average vehicle. 10.3.3.2 Asbestos Emissions and Their Fate The total asbestos emissions per vehicle as calculated above, are not merely deposited onto the ground or entirely emitted into the atmosphere. In general, the asbestos emissions can and will probably end up in any of three different places: retained in the brake and/or disposed of during servicing. deposited on the ground floated into the atmosphere The calculation given in Table 10-20 shows that the largest part of the asbestos emissions produced is emitted from the brakes and clutch as drop-out material. The fate of the total emissions calculated earlier for the test vehicle is as follows: Road-dropout: Airborne: Retained in Brake: 23.34/28.51 x 100 = 81.9 percent 1.05/28.51 x 100 = 3.7 percent 28.51 - 24.39/28.51 x 100 =14.4 percent 10.4 ESTIMATED ASBESTOS EMISSIONS 10.4.1 Passenger Vehicles 10.4.1.1 Total Asbestos Emissions The estimated total asbestos emissions per year can be calculated from the following information: Total asbestos emissions per vehicle: Number of miles per year: Number of pass'enger vehicles: 28.51 micrograms/mile 9,978 96,400,000 Estimated asbestos emissions in pounds per year 28.51 yg mile-vehicle x 10 "6 yg x 9.978 x 103 SiiML x yr lib. x 454 gm 96.4 x 10 vehicles = 60,400 pounds per year. 10.4.1.2 Distribution of Asbestos Emissions The distribution of the total asbestos emissions cal culated for the test vehicle can be determined by applying the percent distributions given in Section 10.3.3.2. The fate of the 60,400 pounds 10-25 HWBUI0002512 T a b le 10-19 - T o ta l A sbestos E m issions C a lc u la te d From th e T e st V e h ic le (Average fo r the Three V e h ic le T ests) CO /-N X3 a) C ci <D rH 4-1 T 1 -H Aoo ttoo S \ H *r 1 00 (U B 3- CM CM in 00 rH O oo O r^* CM rH o rH O ooo o o CO r-*. r-* O av O i--1 rH rH co CM UO CO O CM CM o co CM CM U0 o oOO uo 00 CM 00 cS CO 4-1 T 1 u H 4-J o CO rH co rH Mcol* O CM O rH CM cd Ma) Aoo 4o-> > ca/3) Va-)1 3 CO ft O o O o a> O oo O 4-1 O oH u d) rH 0J ft <! 0) iH a) H 00 B cd i- i 00 a) a Sw' < o H CO co CO H B CO CO <D o H CM M CO a) f-<HD JO a CO H <5 s<Di > 1--1 00 00 rH o CM Mt VO i i a> OV rH o CO rH o CM rCHO i--1 r--1 CM vO rH i--1 1 i in CO CO Ml* vO Mt CO CM i i rO-v Oh o VO rH CTv uo i i MJ- Ml* CT\ o uO Oh 1 1 r^* CM rH CTv in rH CM Mt rH vO 00 CTv Oh rH rH | i rH 00 r-* CO mi* CM 1 i cO CM CM.* CM CM Ov 1 i 00 co vO Ml* o rH rH i--1 1 Ov rH Cr-'O*. rH rH m uo rH 00 co Mt vO oo 1 1 mo r-. uo CM uo i--1 1 | mco P". o CM rH rH 1 i Oo vO CO in vO CM rH rH rH 1 1 ooo uo rH rH rH CM 00 i--I vO in CM CO 00 O CO m | \ CO Mf rH rH CTV vO Ml* rH | 1 O vO 00 uo o CM o CM rH | 1 Ml* oo in 00 vO UO CM I i r^* Ml* rH CM rH co CM rH tH rH CM m CO CO o | 1 OV oo VO CM VO CO | | rH rH vO CM ov UO CM | 1 oG CD H rH 4-1 O 3& rl U uO Fh a> 0) cd U UCTJ PQ CQ A! a O CO H s3U 4-1 3 rH ft p QJ CD cd cd PQ PQ 4a3 a s 4-1 CO 3 3 H p 1*4 Q rH CJ> a> a) 4c<di 4cd u i-i pq PQ 43 a a co H 34 u 4-1 3 rH npo a) Q) 4c*di 4cdi u 1-4 pq PQ 43 a a co H sU3 4-1 3 rH ppo a) ai 4c4d 4c4d 1*4 u pq PQ 4U3 a CO eU3 43-1 ppa o a) *H a) C3 m G a) r--1 MH cd a) TJ r-- a) 1-4 cd ai 3 *0 cCdO H ccdo #a) Si CO pq 4-1 H o& pq a H CO cu PQ UO 4-1 CO 4-1 3 PQ < <D p o rH <ti 10-26 HWBUI0002513 Q) &U I u pH TO *a0H CO CHO 1 CO o CO <D rQ -aU> rbC0 H a) &I a a) o t--i o 4J H d a) %H > U 4J CO H Q o CM oI CU 1--I *3 H u C c r V u *r* 6 1--O)1 a n: a 60 4-1 =L 42 v-/ 460 r- a & 0) G o 42 H <3 4-1 G O 1 CU o P CN t--l O OoO OoO <1- O <r -xf c ooo Ooo Oop Ooo vO r-. o Ooo Ooo 00 i--1 i-4 m CN O oOO CN m O r> <r i-4 Ov co O mO i-4 O o OOo CO -a- o CO co o CN i-4 o CO o o Ooo Ooo ov o o <1- o ooo in o p p 60 > G A 1--14-1 H U co CO -cr O o T 4J O i--l p CN CN U 42 U o o Ov O O 4-1 a 60 CJ o > H G a QJ F*H o o o O oH CO & 60 G 'I4. 42 CO 60 ^ ra 4-1 33 O G P rl C 0) G M O 42 V4 H <3 4-1 G O a O P4 P t-4 ooo ooo mo CO m ooo ooo ooo m -4- o CO in ooo vO CN i-4 oOO o O CJ CO OV O av r-. ooO vO CN oOo O Oo CO Ov o OV r- Ooo vo CN oOo oOo CO OV o P- ooo a V- t-4 <0a X pH CO CO i-4 r O CN <t VO 1 si- Ov o m Ov i 00 vO co vt o 1 10J f--1 ) *r4 30 aG 60 <u G- > < G O H CO CO CO H u Cl) CO H o 4-1 a) CN CO i-4 d> a 42 *H CO 42 <3 ai > 1--1 ov Ov t-- o P i-4 o i--1 i-4 CN CO i--I r-. CN vo r- i-4 1 in CO cn G- sO CO CN | o Ov r-. O vo OV fH m 1 r- CN Ov in t-4 CN <f t-4 vO CO Ov ov r4 t-4 | rH CO r-x CO Hj* CN 1 CO CN CN CN CN OV 1 OV CO rH i--. i-4 mm 00 p vO oo 1 mo r-. in CN m | in p> CO o CN i-4 i--1 1 Oo VO co m VO CN --i t-4 1 1o m co t-4 -- CN CO vO m CN co CO o 00 m 1 CO 1 ov vO r-. r> 00 vo in o CN o CN i--l <r 00 m 00 vO p CN i r- CN ov t-4 P CN CN p Po 1 CO VO vo p vo CN in CN G c 0) H 1-4 CU G H O In CJ pc- <1) <i> X 4*3 GG GU P P 42 a U 4-1 CO G G *H G t-4 PPU (1) (11 4*3 44 GG UU P pq 42 a <J CJ 4J CO G G r4 U P P CJ <D (1) 4*3 4*3 GG UU P P 42 U aO 4-1 CO G G H U P CJ (11 44 4*3 GG UU P P 42 o ao 4-1 10 G G U pPO a) 4*3 4*3 UU P P ,G o O GG U PPo a) 42 G CO TJ H UG 2M JG CO P <D G r4 0) P CO <c pGq 4-1 O *i4 H O 14-1 U 4-1 4-1 G Cl) u PH CU o U 0) 1 TO oG i-4 Fn CU (0 <3 p 10-27 HWBUI0002514 calculated for the test vehicle is as follows: Road-dropout: Airborne: 49,470 pounds 2,230 pounds Brake retention: 8,700 pounds ,, (.0> ? * ^ ; 10.4.2 Trucks and Buses 10.4.2.1 Estimated Severity Factors Truck brakes usually tend to operate at higher tempera tures than passenger cars. Thus a greater proportion of high-temperature use in severity weighting can be obtained by using a larger fraction of the 10-Stop Fade asbestos emissions. For trucks and buses, the test re sults are arbitrarily weighted as follows: Simulation Test Schedule Light Truck: New materials Moderate braking Moderate braking Abusive braking Surface renewal Medium Truck and Bus: Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline New materials Moderate braking Moderate braking Abusive braking Surface renewal Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline Heavy Truck: New materials Moderate braking Moderate braking Abusive braking Surface renewal Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline Weighting Factor 0.013 0.013 0.874 0.050 0.050 0.013 0.013 0.774 0.100 0.100 0.013 0.013 0.674 0.150 0.150 10.4.2.2 Estimated Emissions Factors In addition to the severity factors, it is necessary to estimate the total average asbestos emissions for a light truck, a medium truck (or bus), and a heavy truck. The value for the test car was 28.51 micrograms/mile. Assuming that light trucks have a result proportional to the ratio of the friction materials weights. the asbestos emissions for light trucks are larger than the test car by the following factors: Front drum brake: Rear drum brake: Clutch: 2.0 3.0 2.0 It ; 10-28 i I ' HWBUI0002515 A slightly larger factor is selected for the clutch as it is used more often in trucks than in passenger cars. Following the same line of reasoning, the asbestos emissions for medium trucks and buses are larger than the test car by the following factors: Front drum brake: Rear drum brake: Clutch: 3.0 10.0 4.0 The factors for a heavy truck are as follows: Front drum brake: Rear drum brake: Clutch: 5.0 30.0 6.0 10.4.2.3 Calculations of Total Asbestos Emissions for Trucks and Buses The estimates for the total average asbestos emissions per vehicle are shown in Tables 10-21, 10-22, and 10-23, for a light truck, a medium truck or bus, and a heavy truck, respectively. Table 10-21 - Estimate for Total Asbestos Emissions Expected for a Light Truck Schedule Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F, Baseline Friction Couple Asbestos Emissions Estimate Per Axle (ug/mile) Asbestos Emissions (ug/mile) Severity Weighting Factors Weighted Emissions (ug/mile) Front Drum Brake Rear Drum Brake Clutch 62.78 x 2.0' 62.78 x 3.0 1.11 x 2.0 125.56 188.34 2.22 0.013 1.63 2.45 0.03 Front Drum Brake Rear Drum Brake Clutch 9.04 x 2.0 9.04 x 3.0 1.11 x 2.0 18.08 27.12 2.22 0.013 0.24 0.35 0.03 Front Drum Brake Rear Drum Brake Clutch 11.46 x 2.0 11.46 x 3.0 1.11 x 2.0 22.92 34.38 2.22 0.874 20.03 30.05 1.94 Front Drum Brake Rear Drum Brake Clutch 116.30 x 2.0 116.30 x 3.0 1.11 x 2.0 232.60 348.90 2.22 0.050 11.63 17.44 0.11 Front Drum Brake Rear Drum Brake Clutch 5.88 x 2.0 5.88 x 3.0 1.11 x 2.0 11.76 17.64 2.22 0.050 0.59 0.88 0.11 Total 87.51 10-29 I HWBUI0002516 Table 10-22 - Estimate of Total Asbestos Emissions Expected for a Medium Truck Schedule Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline Friction Couple Asbestos Emissions Estimate Per Axle (pg/raile) Asbestos Emissions (ug/mile) Severity Weighting Factors Weighted Emissions (ug/mile) Front Drum Brake Rear Drum Brake Clutch 62.78 X 3.0 62.78 X 10.0 1.11 X 4.0 188.34 627.80 4.44 0.013 2.45 8.16 0.06 Front Drum Brake Rear Drum Brake Clutch 9.04 X 3.0 9.04 X 10.0 1.11 X 4.0 27.12 90.40 4.44 0.013 0.35 1.18 0.06 Front Drum Brake Rear Drum Brake Clutch 11.46 X 3.0 11.46 X 10.0 1.11 X 4.0 34.38 114.60 4.44 0.774 26.61 88.70 3.44 Front Drum Brake Rear Drum Brake Clutch 116.30 X 3.0 116.30 X 10.0 1.11 X 4.0 348.90 . 1163.00 4.44 0.100 34.89 116.30 0.44 Front Drum Brake Rear Drum Brake Clutch 5.88 X 3.0 5.88 X 10.0 1.11 X .4.0 17.64 58.80 4.44 0.100 1.76 5.88 0.44 Total 290.72 Table 10-23 - Estimate of Total Asbestos Emissions Expected for a Heavy Truck Schedule Burnish A. B. Baseline Detroit Traffic 10-Stop Fade A. F. Baseline Friction Couple Asbestos Emissions Estimate Per Axle (pg/mile) Asbestos Emissions (ug/mile) Severity Weighting Factors Weighted Emissions (ug/mile) Front Drum Brake Rear Drum Brake Clutch 62.78 X 5.0 62.78 X 30,0 1.11 X 6.0 313.90 1883.40 6.66 0.013 4.08 24.48 0.09 Front Drum Brake Rear Drum Brake Clutch 9.04 X 5.0 9.04 X 30.0 1.11 X 6.0 45.20 271.20 6.66 0.013 0.59 3.52 0.09 Front Drum Brake Rear Drum Brake Clutch 11.46 X 5.0 11.46 X 30.0 1.11 X 6.0 57.30 343.80 6.66 0.674 38.62 231.72 4.49 Front Drum Brake Rear Drum Brake Clutch 116.30 X 5.0 116.30 X 30.0 1.11 X 6.0 581.50 3489.00 6.66 0.150 87.22 523.35 1.00 Front Drum Brake Rear Drum Brake Clutch 5.88 X 5.0 5.88 X 30.0 1.11 X 6.0 29.40 176.40 6.66 0.150 4.41 26.46 1.00 Total 951.12 30 HWBUI0002517 Table 10-24 Distribution-Weighted Asbestos Emissions From A Light Truck Schedule Burnish A.B. Baseline Detroit Traffic 10-Stop Fade A.F. Baseline Friction Couple Front Drum Brake Rear Drum Brake Clutch Front Drum Brake Rear Drum Brake Clutch Front Drum Brake Rear Drum Brake Clutch Front Drum Brake Rear Drum Brake Clutch Front Drum Brake Rear Drum Brake Clutch Severity Weighted Emissions (mg/mile) 1.63 2.45 0.03 0.24 0.35 0.03 20.03 30.05 1.94 11.63 17.44 0.11 0.59 0.88 0.11 87.51 Distribution Weighting Factors Weighted Emissions (mg/mile) Drop out Airborne Retention Drop out Airborne Retention 0.89 0.89 0.10 0.90 0.90 0.10 0.90 0.90 0.10 0.90 0.90 0.11 0.90 0.90 0.10 0.02 0.02 0.02 0.03 0.03 0.02 0.03 0.03 0.02 0.03 0.03 0.02 0.03 0.03 0.02 0.09 0.09 0.88 0.07 0.07 0.88 0.07 0.07 0.88 0.07 0.07 0.88 0.07 0.07 0.88 1.45 2.18 0.00 0.22 0.32 0.00 18.03 27.05 0.19 10.47 15.70 0.01 0.53 0.79 0.01 76.95 0.03 0.05 0.00 0.01 0.01 0.00 0.60 0.90 0.04 0.35 0.52 0.00 0.02 0.03 0.00 2.56 0.15 0.22 0.03 0.02 0.02 0.03 1.40 2.10 1.71 0.81 1.22 0.10 0.04 0.06 0.10 8.01 Table 10-25 - Summary of All Brake and Clutch Emission (lbs per year) Vehicle Passenger Cars Light Trucks Medium Trucks and Buses Heavy Trucks Miscellaneous* of Vehicles 96,400,000 17,100,000 Total Asbestos Emissions 60,400 32,300 Drop-Out Distribution Ai rborne Brake Retention 49,470 2,230 8,700 28,420 940 2,940 2,600,000 16,300 14,330 470 1,500 1,200,000 32,900 6,615,000 16,300* Totals 158,200 Percent of Total 28,920 14,330 135,470 (85.6) 950 470 5,060 (3.2) 3,030 1,500 17,670 (11.2) * Estimated equal to medium trucks as weights of friction material used for both categories are almost equal (Table 12). Includes motorcycles, trailers, etc. 10-31 HWBUI0002518 The estimated total asbestos emissions per year for light trucks can be calculated from the following information: Total asbestos emissions per vehicle: Number of miles per year: Number of trucks: 87.51 yg/mile 9,807 17,100,000 Estimated asbestos emissions in pounds per year = 87.51 x 10-6 _gm x 9.807 x 10 3 miles x 1 lb x 17.1 x 10 vehicles mile Pg yr 454 gms = 32,300 pounds per year The estimated total asbestos emissions per year for medium trucks and buses can be calculated from the following information: Total asbestos emissions per vehicle: Number of miles per year: Number of trucks: 290.72 yg/mile 9,807 2,600,000 Estimated asbestos emissions in pounds per year = 290.72 -HfL. mile x 10-6 _gm Pg x 9.807 x 10 3 miles yr x 1 lb 454 gms x 2.60 x 10 vehicles = 16,300 pounds per year The estimated total asbestos emissions per year for heavy trucks can be calculated from the following information: Total asbestos emissions per mile: Number of miles per year: Number of trucks: 951,12 yg/mile 9,807 1,600,000 Estimated asbestos emissions in pounds per year = 951.12 --vj?-- x 10 6 x 9.807 x 10^ e- x .^-- x 1.60 x 10 ^ vehicles mile yg yr 454 gms = 32,900 pounds per year 10.4.2.4 Total Truck Asbestos Emissions The estimated value for the total asbestos emissions from trucks and buses is as follows: Light trucks Medium trucks and buses Heavy trucks 32,300 pounds 16,300 pounds 32,900 pounds 81,500 pounds I' 10-32 I HWBUI0002519 10.4.2.5 Distribution of Truck Asbestos Emissions 10.4.2.5.1 Estimated Distribution Weighting Factors The distribution weighting factors for passenger car brakes are discussed in Section 10.3.2.1. Trucks contain drum brakes which are designed to be more open than passenger car drum brakes. In many instances, no splash shields are used or they are very open. More over, trucks run with hard tires and stiff suspensions; these lead to a hard ride and more shaking at the wheels. Consequently, the retention of wear debris in truck drum brakes is estimated to be approximately 25 percent of that for passenger vehicle drum brakes. From the data given in Section 10.3.2.1.3, it is possible to estimate the following distribution weighting factors for truck drum brakes: Simulation Test Schedule Airborne Road Drop-Out Vehicle Retention New friction couple Burnish After 2632 miles Composite 0.02 0.03 0.89 0.90 0.09 0.07 The airborne figure was adjusted upward. The openness of the disc brake gives rise to a higher airborne fraction than for a closed drum brake. The value estimated for the truck drum brake takes this into account. 10.4.2.5.2 Truck Asbestos Emissions and Their Fate The total asbestos emissions calculated for the light truck, medium truck and bus, or heavy truck will be distributed into three places: o retained in the brake and/or disposed of during service o deposited on the ground o floated into the atmosphere The typical calculations for the light trucks are given in Table 10-24 and can be summarized: Road Drop-out: Airborne: Brake Retention: 76.95/87.51 x 100 = 87.9 percent 2.56/87.51 x 100 = 2.9 percent 8.01/87.51 x 100 =9.2 percent 10.4.3 Estimated Asbestos Emissions The summary of the total asbestos emissions estimates and their distribution is given in Table 10-25. i- 10-33 HWBUI0002520 Table 10-26 Calculation of Weighted Average Unconverted Asbestos Percent for Sump Sample Schedule Burnish A.B. Baseline Detroit Traffic 10-Stop Fade A.F. Baseline Vehicle Test i 2 3 Average 1 2 3 Average i 2 3 Average 1 2 3 Average i 2 3 Average Front Brake 0.813 0.402 0.218 0.478 1.650 0.236 0.106 0.664 0.247 0.304 0.018 0.190 0.463 0.244 0.045 0.251 0.656 0.443 0.158 0.419 Rear Brake _ 0.484 0.063 0.274 0.179 0.268 0.063 0.170 0.129 0.240 0.046 0.138 0.280 0.199 0.102 0.194 0.419 0.138 0.034 0.197 Overall Average 0.376 0.417 0.164 0.223 0.308 Weighting Factor 0.013 0.013 0.824 0.075 0.075 Weighted Percent 0.005 0.005 0.135 0.017 0.023 0.185 Table 10-27 - Calculation of Weighted Average Asbestos Percent for Surfaces Sample Schedule Burnish A.B. Baseline Detroit Traffic 10-Stop Fade A.F. Baseline Vehicle Test 1 2 3 Average 1 2 3 Average i 2 3 Average 1 2 3 Average 1 2 3 Average Front Brake 0.398 0.968 0.101 0.489 0.472 0.311 0.024 0.269 0.137 0.548 0.013 0.233 0.162 0.261 0.041 0.155 0.507 0.173 0.073 0.251 Rear Brake 0.865 1.416 0.046 0.776 0.415 0.098 0.056 0.190 0.373 0.113 0.057 0.181 0.071 0.619 0.055 0.248 0.242 0.237 0.022 0.167 Overall Average 0.633 0.230 0.207 0.202 0.209 Weighting Factor Weighted Percent 0.013 0.008 . 0.013 0.003 0.824 0.171 0.075 0.015 0.075 0.016 0.213 10-34 I: HWBUI0002521 10.4.4 Alternate Estimate for Total Asbestos Emissions An alternate method to calculate an estimate for total asbestos emissions is proposed. Tables 10-26, 10-27, and 10-28 list the data and the calculations required to produce a weighted average asbestos percent for the sump, surfaces, and airborne samples using the weighted portions of the Baseline, A. B. Baseline, Detroit Traffic, 10-Stop Fade, and A. F. Baseline schedules. The data given in Table 10-16 indicated that the percent ages of annual consumption of friction materials is as follows: Passenger Car Brakes: 56 percent Truck Brakes: 44 percent The data given in Section 10.3.2.1.3 and Section 10.4.2.5.1 show that the average distributions for most of the brake mileages can be summarized as follows: Brake Airborne Road Drop-Out Vehicle Retention Passenger Car Brakes Truck Brakes 0.03 0.03 0.77 0.90 0.20 0.07 The calculation for the weighted average asbestos found in all brakes is given in Table 10-29. The average wear debris recovery for each vehicle can also be calculated from the data given in Table 9-14: Fronts . Rears Test 1 2 3 51.2* 60.8 60.2 67.8* 81.6 81.3 *Low because of incomplete, difficult recoveries of Fade samples. The drum rear samples are known to contain approximately 10 weight percent material such as iron oxide from drum wear. Thus an average recovery of approximately 65 percent, based on the debris from the friction materials for vehicle tests 2 and 3, is used in the calculation. Therefore, the asbestos percent based on the friction material worn in the test vehicle is calculated as follows: 0.206 x 0.65 = 0.134 10-35 HWBUI0002522 Table 10-28 - Calculation of weighted.average.asbestos Percent for Airborne Samples Schedule Burnish A.B. Baseline Detroit Traffic 10-Stop Fade A.F. Baseline Vehicle Test 1 2 3 Average 1 2 3 Average 1 2 3 Average 1 2 3 Average 1 2 3 Average Front Brake 0.144 0.186 0.099 0.143 0.290 0.211 0.125 0.209 0.052 0.234 0.025 0.104 0.154 0.107 0.055 0.105 0.134 0.165 0.047 0.115 Rear Brake 0.349 0.087 0.060 0.165 0.493 0.344 0.044 0.294 0.092 0.115 0.037 0.081 0.132 0.194 0.014 0.113 0.079 0.180 0.003 0.087 Overall Average 0.154 0.251 0.093 0.109 0.101 Weighting Factor 0.013 * 0.013 0.824 0.075 0.075 Weighted Percent 0.002 0.003 0.077 0.008 0.008 0.098 Table 10-29 - Data and Calculation of.Weighted.Asbestos Percent for All Vehicles Friction Materials Passenger Car Fraction of Friction Material 0.56 Fraction of Debris Airborne Drop-Out Retention 0.03 0.77 0.20 Weight Percent Asbestos 0.098 0.213 0.185 Weighted Asbestos Percent 0.002 0.092 0.021 Truck 0.44 Airborne Drop-Out Retention 0.03 0.40 0.07 0.098 0.213 0.185 0.001 0.084 0.006 0.206 ! i 10-36 HWBUI0002523 The total weight of friction material worn away annually in brakes is 117 million pounds. The total asbestos estimate from brakes can now be calculated: 117,000,000 x 0.00134 = 157,000 pounds The total weight of friction material worn away annually in clutches is 6.6 million pounds. The total asbestos estimate from clutches (assuming a retention of ^75 percent) can be calculated as follows: 6,600,000 x 0.00047 x 0.25 = 800 pounds Thus the total is approximately 157,800 pounds, which agrees with the figure reported in Section 10.4.3 above. 10.5 SUMMARY OF ASBESTOS EMISSIONS The results indicate that more than 99.7 percent of the asbestos used in friction material is converted to a non-fibril form. The total asbestos emissions for all vehicles in the United States is estimated to be 158,000 pounds annually. Of this total, a small part was found to remain airborne. The contribution to the atmosphere is estimated to be 5060 pounds, or 3.2 percent of the total asbestos emissions. 10-37 HWBUI0002524 10,6 REFERENCES (10-1) National Highway Traffic Safety Administration and Federal Highway Administration preliminary estimates, private com munication, January 18, 1973. (10-2) R. L. Polk & Co., private communication, January 12, 1973. (10-3) "Truck Sales Jumped 24 Percent in 1972 to 2.6 Million Units," Wall Street Journal, January 8, 1973. (10-4) "1972 Automobile Facts and Figures," Motor Vehicle Manufacturers Association, 1973. (10-5) "Motor Truck Facts," Motor Vehicle Manufacturers Associaation, 1973. (10-6) W. J. Zechel, et. al., "Automotive Brake Evaluation Under Customer Usage Conditions," SAE Special Publication SP-338, Society of Automotive Engineers, April 1968. (10-7) R. G. Mortimer (University of Michigan), "Hard Braking is More Common Than You Might Think," Automotive Engineering, August 1971, page 32. (10-8) N. Carpenter and E. G. Goddard, "The Measurement of Braking Performance for Compulsory Motor Vehicle Inspec tion," Journal, Institute of Motor Industry, Vol. 4, No. 10, 1955. (10-9) Private Communication, January 1973. (10-10) Private Communication, January 1973. (10-11) C. F. Harwood (IITRI), "Asbestos Air Pollution Resulting From the Wear of Braking Linings," Illinois Institute of Technology Research Institute, April 1972. (10-12) T. C. May and R. W. Lewis, "Asbestos," A Chapter from "Mineral Facts and Problems," U.S. Bureau of Mines Bulletin 650, 1970 edition. (10-13) Private Communication, January 1973. 10-38 i i i HWBUI0002525 SECTION 11 SUMMARY The specific objectives of this program were stated earlier. The end objective was the documentation of asbestos emissions from brakes and clutches and the assessment of their overall contributions to the atmosphere. These objectives were met by means of the program as planned and completed. For the results to be representative of average consumer-type vehicle usage, a vehicle intermediate between compacts and luxury vehicles was selected. The vehicle was equipped with front disc brakes, rear drum brakes, and a dry clutch. All friction material used contained asbestos. Three vehicle tests were run. The first vehicle test was run with the original-equipment friction materials. The second vehicle test simulated a partial reline - fronts only relined - while the rears were repeated to give a replicate test, and an indication of emissions trends for friction materials with continued use. The third vehicle test simulated a complete brake reline, and included turned discs and drums. The friction materials selected for the program were representative of those supplied by the industry - five different manufacturers produced the original equipment and aftermarket disc pads and drum linings, which were selected in part because of their high-volume usage. The collection of particulate and gaseous emissions from any vehicle friction couple has never, to the authors* knowledge, been reported in the literature. To define the extent of these emissions, a collection system was devised which combined the functions of separation and storage. Unique emissions collectors for both disc and drum brakes, and for a clutch, were conceived, designed, and built as the main embodiment of this instrumenta tion. The collectors separated the wear debris into three different frac tions: a sump sample which included the wear debris on the lining surfaces, in the rivet holes, and on the brake drum; a surfaces sample which in cluded the wear debris on the brake and collector shroud surfaces; and an airborne sample collected on membrane filters. The front disc brake emission collector was tested on an inertia dynamometer to check out its operation. The most critical part of the collector operation was the rotating seal. The final design selected was made from graphite-filled Teflon which rotated against a hardened steel ring. Thermal response tests were made, the results indicating that for normal operation the shrouded disc brake would operate at a slightly higher temperature than the normal brake configuration. A method was devised to determine the leak rate of the collector. The rotating seal was largely responsible for the residual leak; the leak rate was reduced significantly by use of a very small amount of hightemperature grease. The rotating seal underwent durability testing and HWBUI0002526 survived seven hours at 50 mph. A second series of dynamometer tests gave reproducible emissions and temperatures. Finally, a test was made in which the collector, and especially the rotating seal, were shown to be relatively leak-tight to external liquid and particulate contaminants. The vehicle was equipped with standard brake test instrumentation. A front disc brake collector and a rear drum brake collector were in stalled on the right side of the vehicle. The clutch was sealed by closing the few holes in its casing. The left wheel brakes were left in their normal configuration and were used to monitor the operation of the shrouded brakes. Wear debris was taken from the left brakes as well as from those on the right. The amounts of debris formed and their com positions were used to demonstrate that the brake shrouding did not significantly affect the operation of the brakes on the right side. Although there are many brake and friction material tests suitable for measuring a specific condition or combination of conditions, no one test adequately matches driving conditions which would be representative of all vehicles. Consequently, it became necessary to devise a rational and meaningful vehicle test. Seven schedules were chosen, and each was followed by a measurements procedure (take emissions samples, measure wear, inspect the systems, and replace worn parts). The first three schedules - Burnish, After Burnish (A.B.) Baseline, and Detroit Traffic constituted low to moderate temperature testing. The final four sched ules - 10-Stop Fade, After Fade (A.F.) Baseline, 15-Stop Fade, and Final Baseline - constituted high temperature and abusive testing. A detailed step-by-step procedure was prepared for the removal of brake emissions from the collectors. The amounts of debris collected at the various locations were used later to calculate the mass balances. A detailed step-by-step procedure was also prepared for the sampling, handling, and analysis of the asbestos emissions. The analysis of brake wear debris indicated the presence of 20-30 weight percent polymeric and carbonaceous material, which in turn pro duced a fluffy appearance when the material was examined in the scanning electron microscope. The organic portion was therefore removed by low temperature ashing. Two representative samplings were used: the first in sample selection for low temperature ashing; and the second in sample selection for distribution on a filter membrane for subsequent analysis. The only suitable method of analysis for asbestos was microscopy. At the beginning of the program, one optical and three transmission electron microscopy methods had been developed by others. Two of the electron microscope methods were not suitable because they changed the fiber size distribution of the asbestos. Initial experiments indicated that the asbestos in wear debris varied in size from 0.1 micron in diame ter up to 2.0 microns in diameter by 50 microns in length. To preserve the particle size distribution, a combination of optical microscopy and transmission electron microscopy was selected. Optical microscopy with phase contrast optics at 400 diameters was used to detect fibers with 11-2 HWBUI0002527 diameters of 0.5 micron and above. Transmission electron microscopy at 22,000 diameters was used to detect fibers with diameters of 0.5 micron and below. The projected fiber diameter and length were used to determine the fiber volume. The volumes were summed, converted to weight, and normal ized for the entire sample weighed onto the filter. A computer program was subsequently developed to minimize computational time and to elimi nate errors in the processing of these calculations. The possible sources of error associated with sample processing and analysis were minimized through the use of statistical methods. The largest effect found on the asbestos analysis was the occasional occur rence of a large fiber. The extent of error increased as the percent asbestos in the sample decreased. The data showed that the maximum deviation for results with asbestos contents above 0.10 percent averaged 12-15 percent and had a maximum at approximately 20 percent. The maxi mum deviation increased for asbestos contents below 0.10 percent, where errors for the small amounts of asbestos emissions become less signifi cant from the standpoint of air pollution. A gas handling system was set up to separate, measure, and transfer samples to the gas chromatograph. A step-by-step procedure was prepared. The non-condensable gases were analyzed by separating the gases on a Porapak Q column and then detecting the separated components with a thermal conductivity detector. In some instances, a flame ionization detector was also used. The condensable gases were analyzed by separa ting the components on a W-98 silicone gum rubber column and detecting with a flame ionization detector. The vehicle tests were carried out during the period May 1 to November 17, 1972. Only one major problem occurred during the vehicle testing: during the 10-Stop Fade of vehicle test 1, the hardened steel ring,on which the rotating seal slides, became loose. The ring was re anchored and no further problem was encountered during the 15-Stop Fade or in any further test schedule. This result showed the durability of the collector and rotating seal design and construction. For the 8100 miles of testing, only seven front and eight rotating seals were required. This was better than anticipated at the start of the program. The mileage accumulations for each schedule of each vehicle test showed that the. mileages were reproducible. Prior to vehicle test 1, a Burnish schedule was performed and the temperature for the right front disc brake was found to be near 400F. The temperature difference, AT, between the right and left front brakes was 200-250F. This was unacceptable. The addition of several thermal control aids reduced the difference to the range 115 to 160QF. Most important was the reduction of the right front brake operating tempera tures from about 400F to about 300F. The brake shrouding did not present any problems on hot days. During the Detroit Traffic Test, the AT between the front brakes remained between 100 and 140F while that for the rear drum brake remained between 10 and 30F. 11-3 HWBUI0002528 During the fade tests, the right front brake was programmed accord ing to known temperature versus stop-number curves obtained for the same friction materials previously run on an inertia dynamometer and on a normal test vehicle. After the seventh stop, the time increment between stops was increased to prevent the right front brake from overheating the friction materials within the shroud. The 15-Stop Fade produced an addi tional effect within the brake. The heat generated was retained within the shroud and caused the brake fluid within the right front caliper to heat. The resultant fluid boil produced a loss of pedal for the last two fade stops and persisted through the first eight recovery stops. The recovery of particulate emissions from both front brakes yielded a comparison of the amounts of emissions of the normal with the shrouded disc brake. For both brakes, the sump sample appeared to reach and maintain an average amount of debris which was independent of the mileage. When the debris was removed, the steady-state value appeared to reestablish itself. The surfaces sample for the open brake showed the same behavior. The recovery data were used to calculate the distri bution weighting factors used later to calculate the distribution of brake emissions in three categories: retained in the brake, released as road drop-out material, and airborne material. The amounts of particulates collected in the three samples were compared with the friction material weight losses. In all cases, the data for the right front brake ranged from approximately 42-percent recovery in the initial schedule of each vehicle test to approximately 70 percent for the final schedules. The low initial recovery may have been caused by excess gassing during Burnish, resulting in a lower recovery of particulates. The brake emissions recovered from the fade tests were "tarry", and the surfaces samples were recovered only with difficulty. The material recovery data for the rear brakes showed wide vari ation. The problem in part was due to the 80-percent sump carry-over to the following test. During certain vehicle tests, the sump sample was redistributed within the brake and registered a lower value after the test. Thus, a negative weight was recorded. The other samples for that brake did increase and a positive overall value was always recorded. The range of asbestos content in the brake emissions for vehicle tests 1 and 2 went from a high of 1.65 percent to a low of 0.03 percent for the 90 analyses; only three were above 1.00 percent. The overall average for vehicle test 1 was 0.38 percent and for vehicle test 2 was 0.25 percent. For vehicle test 3, the range of asbestos content was from a high of 0.51 percent to a low of 0.003 percent. The overall average was 0.07 percent. To provide a systematic independent check on the asbestos analyses, the EPA Project Officer arranged a second contract for Battelle Columbus Laboratories to analyze twenty-four samples generated during the program. Johns-Manville also provided three analyses. The Johns-Manvilie and 11-4 HWBUI0002529 Battelle data agreed very well. In 19 of the 24 analyses, the Program results were slightly higher than the corresponding Battelle results. The Program average was 0.25 percent, while the Battelle average was 0.17 percent or 68 percent of the Program average. The Program data were expected to be higher for the following reasons: all larger fibers (bundles of fibrils) found were assumed to be cylinders of projected diameter and length, and a fiber with an elliptical cross section was actually smaller in volume than reported; and all fibers were assumed to be 100-percent asbestos (partially degraded asbestos or olivine fibers were weak and mechanically reduced to non-fibrous material by the other analytical techniques, hence only the remaining asbestos fibrils that remained were counted). Analysis of the gaseous emissions was made difficult by the large quantities of water and carbon dixodie which by-passed the gas condition ing trap and entered through the rotating seals. Most samples showed traces of to C3 hydrocarbons only as lower concentrations of higher molecular weight gases could not be detected,. When the gases were con centrated from the fade.tests, C4 to Cs compounds such as butanes, ben zene, toluene, phenol, and cresols were detected in the gas traps. The corresponding shroud extracts showed C12 to C30 materials, including cardanol. To compare the emissions properly, it was found necessary to cal culate the asbestos emission factors for each sample in micrograms/mile. The values of these factors were obtained by multiplying the weights of brake debris generated by the appropriate asbestos percents, then divid ing by the miles per test schedule. Thus, either a high asbestos content or a large sample gave a high factor. The following observations were made: asbestos emissions were higher for new friction surfaces and decreased with use; the drum brake produced more asbestos emissions than the disc brake initially, the difference decreasing as the friction materials continued in use; heavy (abusive) duty did not necessarily give a higher percent ' asbestos, however, the large amount of debris produced gave a significant rise in asbestos emissions; asbestos emissions from the brakes were found to decrease from fade or heavy-duty stops (highest) to burnish to moderate braking (lowest); for the disc pads only, there was an increase in asbestos emis sions with increased asbestos content in the friction material (there was no such trend for the drum brake materials); 11-5 HWBUI0002530  both the front disc pads and the drum linings of vehicle test 3 have wear comparable to that found in the other two vehicle tests, yet the asbestos emissions were significantly lower (both pads and secondary linings contained brass chips which may have been in part responsible for the more complete con version of the asbestos in the brake emissions); and for both disc and drum brakes, the surfaces sample was the largest of the three (^92 percent), the sump sample was next (^7 percent), and the airborne sample was the smallest (VL percent). Each fade schedule included the 12-stop recovery and a 35-stop reburnish. It was expected that the recovery and reburnish stops would remove the heat-damaged rubbing surfaces of the friction materials. However, both the A.F. Baseline and Final Baselines show an average increase of 250 per cent more wear over their approximately 350 miles or approximately 1500 stops. These data indicate that many stops were necessary to produce a renewed surface. The end objective of this report was to develop the best possible estimate of total brake and clutch emissions per year for the U.S. popu lation of braked vehicles in use. A data base was obtained from a wide variety of trade publications and from industry and government contacts. The number of vehicles given and the other data presented indicated a wide and varied usage so that some assumptions had to be made in prepar ing the asbestos emissions estimates. The asbestos content in brake friction materials was given by the two most knowledgable industry sources as 90 and 120 million pounds for 1972. In the program our calculations, ^118 million pounds was used. These ^118 million pounds of asbestos were incorporated in 171 million pounds of brake friction materials. A calculation gave the total brake friction material worn away as 117 million pounds per year. The program calculation for the asbestos content in clutch friction materials was 4.38 million pounds in 8.76 million pounds of friction material. A calculation gave the total clutch friction material worn away as 6.56 million pounds. Two series of weighting factors were developed. The distribution weighting factors were calculated from the material recoveries in the shrouded and unshrouded brakes as indicated earlier. Estimates were made for both disc and drum brakes from the test vehicle for the amounts of the different samples distributed from the brakes. These estimates were made for the Burnish schedule, for the end of a complete vehicle test, and for the end of all three vehicle tests. For the drum brake, estimates were made for up to 40,000 miles. The severity weighting factors were calculated from the percentages of different braking modes: for new versus burnished friction materials, for severe (or abusive) versus moderate braking, and for the renewal of friction surfaces (post fade) . 11-6 HWBUI0002531 The total asbestos emissions estimate for the test vehicle was calculated from the asbestos emissions factors and the severity weight ing factors. The average asbestos emissions factors for the disc brakes, drum brakes, and clutch were used along with the Burnish (new surfaces), A.B. Baseline (surfaces preparation), Detroit Traffic (moderate braking), 10-Stop Fade (severe braking) and A.F. Baseline (surface renewal) weight ing factors. The value obtained for the test vehicle was 28.51 yg/mile. The total asbestos emissions were not considered to be merely dumped onto the ground or thrown into the atmosphere. The fates of the emissions were calculated with the use of the distribution weighting factors as follows: Road Drop-out 81.9 percent Airborne 3.7 percent Brake Retention 14.4 percent The annual total asbestos emissions estimate for 96,400,000 passen ger vehicles travelling.an average of 9,978 miles per year was found to be 60,400 pounds distributed as follows: Road Drop-Out 49,470 pounds Airborne 2,230 pounds Brake Retention 8,700 pounds 60,400 pounds Truck brakes usually tend to generate and to operate at higher temperatures than passenger cars. To estimate asbestos emissions for light, medium, and heavy trucks, it was found necessary to adjust the severity under which the emission would be expected to occur. The severity weighting was obtained by using a larger fraction of the 10Stop Fade (and A.F. Baseline) asbestos emissions according to the following progression: Passenger car 2.0 percent Light truck 5.0 percent Medium truck (and bus) 10.0 percent Heavy truck 15.0 percent In addition, it was necessary to estimate the asbestos emissions factors. Assuming that these factors were roughly proportional to the increased friction material weights of the appropriate vehicles compared to the test vehicle, the following estimated factors were determined: Light truck Medium truck (and bus) Heavy truck Front Drum Brake 2.0 3.0 5.0 Rear Drum Brake 3.0 10.0 30.0 Clutch 2.0 4.0 6.0 |s II 11-7 HWBUI0002532 The annual total asbestos emissions estimates for light trucks, medium trucks (and buses) and heavy trucks were found to be: Light trucks 32,300 pounds Medium trucks (and buses) 16,300 pounds Heavy trucks 32,900 pounds 81,500 pounds The distribution weighting factors for trucks were calculated based on the following considerations. The truck drum brake is designed to be more open than the car drum brake and in many instances no splash shields are used. Based on the estimates, that only about 25 percent as much debris remains in a truck drum brake as compared to a passenger car drum brake, distribution weighting factors for trucks were estimated. The distribution of the total asbestos emissions estimate was as follows: Road Drop-Out 87.9 percent Airborne 2.9 percent Brake Retention 9.2 percent The summary of the estimated asbestos emissions for all vehicles in pounds/year was calculated as follows: Total Asbestos Emissions Distribution Drop-Out Airborne Retention Passenger Cars Light Trucks Medium Trucks (and buses) Heavy Trucks Miscellaneous 60,400 32,300 16,300 32,900 16,300 49,470 28,420 14,330 28,920 14,330 2,230 940 470 950 470 8,700 2,940 1,500 3,030 1,500 158,200 135,470 5,060 17,670 Distribution by percent: 85.6% 3.2% 11.2% These estimates should be considered as maximum values. The total asbestos emission estimates, and especially the airborne emissions estimate, are considerably lower than the 540,000 pounds esti mated by IIT Research Institute. (10~H) 11-8 HWBUI0002533 APPENDIX A WEAR DEBRIS CALCULATIONS A.1 INTRODUCTION This appendix discusses friction material reactions used to estimate amounts of wear debris which were expected to have been collected during various cycles of the "Brake Emissions" program. Consideration is given to both disc and drum brakes. A.2 FRONT DISC BRAKE A.2.1 Composition From previous studies, the typical composition of a disc pad can be approximated: Resin }Cashew Rubber Asbestos Zinc ^32 wt % M34 wt % v4 wt % A.2.2 Burnished Products From various studies, the following amounts of burnished products can be estimated. Particulates Gravimetric Factor Resin }Cashew Rubber ^32 wt % Carbonaceous Binder ^16 wt % 0.50 (from TGA) Asbestos M34 wt % + Olivine ^55 wt % 0.86 (from TGA) Zinc ^4 wt % ZnO ^5 wt % 1.25 (from TGA) Total 76 wt % Consequently, the expected amount of particulate wear debris expected is 76% of the weight lost. Although disc wear does occur, it cannot be estimated at this time. Gaseous Products The average organic molecule can be assumed to have the structure: A-l HWBUI0002534 OH I ---- CH, C9H10ON --CH = CH-NH which undergoes the following oxidation reaction: C9H1QON + 11.5 02 + 9 C02 + 5 H2Q + NO MW = 148 MW = 32 From the above calculation, only 50% of the organic portion oxidized: 32 wt % x 0.50 = 16 wt % which gives 16/148 = 0.108 mole of organic oxidized. This requires: 11.5 x 0.108 = 1.24 moles of 02 1.24 x 32.0 = 39.7 gms of 02 Total amount of material reacting: 16.0 + 39.7 = 55.7 gms Conversion factor is: 55.7 16.0 3.48 1 A-2 HWBUI0002535 Since not all 16.0 gms of the organic portion will be completely oxidized to gases, this value represents a maximum amount of material expected. The expected amount of gaseous wear products is 56% of the weight lost. A.2.3 Typical Disc Brake Wear Rates Burnish: Detroit Traffic: SAE Fade: 7 Stop, 90 mph: MJ.007" MD.020" ^0.030" MD.015" or v0.018 cm or ^0.051 cm or ^0.076 cm or ^0.038 cm A. 2.4 Disc Pad Area 2 Inner: 8.50 in Outer: 11.25 in2 19.75 in2 ' 1.9.7_5_ i.n^2 x (2.54 cm/in)^2 = 127.5 cm2^ A. 2.5 Calculated Wear During Burnish Area x Thickness - 127.5 x 0.018 = 2.30 cc Typical S.G. for disc pad is 1.70 Calculated weight loss: 2.30 x 1.70 = 3.91 gms o Particulate Portion 3.91 x 0.76 = 2.97 gms which will be collected in four areas sump brake surfaces 8.0y filter 0.2p filter 1 iSS '.' JlS I' !& SS; if :,;s lift': I! j-'.'.V: '.'M A-3 HWBUI0002536  Gaseous Portion 3.91 x 0.56 = 2.19 gms which will be collected in three traps: one condensable gas trap two noncondensable gas traps filled with charcoal Total Sample Expected 2.97 + 2.19 = 5.16 gms. A.2.6 Summary of Typical Wear Expected Test Burnish Detroit Traffic SAE Fade 7 Stop, 90 mph Wear (inch/cm) 0.007/0.018 0.020/0.051 0.030/0.076 0.015/0.038 Particulates (gms) 2.97 8.47 12.70 6.36 Gases (gms) 2.19 6.26 9.38 4.68 Total 5.16 14.73 22.08 11.04 A. 3 REAR DRUM BRAKE A.3.1 Composition From previous studies, the typical compositions of primary and secondary linings can be approximated: Resin ^ Cashew I Asbestos Primary 28 72 Secondary 31 69 A.3.2 Brake Parameters Designation: 11 x 2-1/4" DSSA Primary Area: 2.25" x 8.8" (average arc length) = 19.8 in^ Secondary Area: 2.25" x 11.5" (average arc length) = 25.9 in2 A-4 HWBUI0002537 higher torque stops. (The total lining length is 8.8 + 11.5 = 20.3". The actual drum circumference is 3.14 x 11" = 34.6". Thus the lining sweeps 20.3/ 34.6 x 100 = 59% of the available drum surface.) A.3.3 Burnished Products Because the secondary is programmed to absorb more torque and essentially wear more throughout the lining combination's usage, and because the primary and secondary wear debris contributions are always combined, a more typical starting composition for the combination must be slanted toward the secondary's composition analysis: 30 wt % Asbestos 70 wt % The following amounts of burnished products can be estimated: Particulates Gravimetric Factor Resin Cashew 30 wt % + Carbonaceous Binder ^15 wt % 0.50 Asbestos 70 wt % -> Olivine ^62 wt % 0.90 77 wt % Gaseous Products The average organic molecule can be assumed to be OH -- CH 3 0 II A-5 HWBUI0002538 which undergoes the following oxidation reaction: C H H10 102 + 11,5 2 * 10C02 + 5 2 MW = 162 Since only 50% of the organic portion is oxidized, i.e., 15 wt %, this corresponds to 15/162 = 0.093 mole oxidized which in turn requires 11.5 x 0.093 x 32 = 34.3 gms of oxygen. The total amount of reactants is 15.0 + 34.3 = 49.3 gms which represent a gravimetric factor equal to: 49.3 15.0 = 3.29 A.3.4 Typical Drum Rear Wear Rates Burnish: Detroit Traffics SAE Fade: 7 Stop, 90 mph: ^0.005" 0.010" MD.008" 0.005" or M).013 cm or 0.026 cm or v0.020 cm or 0.013 cm Drum Brake Lining Areas Primary: Secondary: 2 19.8 in 25.9 in^ 45.7 in^ 45.7 in^ x (2.54 cm/in)^ = 297 cm^ A.3.6 Calculated Wear During Burnish Volume = Area x Thickness = 297 x 0.013 = 3.86 cc Typical S. G. is 1.70 Calculated weight lost: 3.86 x 1.70 = 6.56 gms I | i | J j| I A-6 HWBUI0002539  Particulate Portion 6.56 x 0.77 = 5.15 gms Gaseous Portion 6.56 x 0.49 = 3.,21 gms Total Wear Deb ris 5.15 + 3.21 = 8,.36 gms A.3.7 Summary of Typical Wear Expected Test Burnish Detroit Traffic SAE Fade 7 Stop, 90 mph Wear (inch/cm) 0.005/0.013 0.010/0.026 0.008/0.020 0.005/0.013 Particulates (gms) 5.15 10.30 8.24 5.15 Gases (gms) 3.21 6.42 5.13 3.21 Total (gms) 8.36 16.72 13.37 8.36 Mi li Ml A-7 HWBUI0002540 APPENDIX B NUCLEPORE FILTER FLOW CHARACTERISTICS FOR BRAKE LINING WEAR DEBRIS B.l INTRODUCTION This appendix describes a series of experiments used to define the filter flow characteristics of Nuclepore membranes with varying loadings of brake lining wear debris. Figure B-1 is a schematic of the experi mental setup. B.2 FLOW VERSUS PRESSURE DROP Table B-1 is the compilations of flow versus pressure drop data for both unloaded filters and loaded filters containing approximately 1.0 gram of wear debris (from a 1970 Chrysler Imperial rear drum brake). This data is plotted in Figure B-2. B.3 FLOW AND PRESSURE DROP VERSUS LOADING Table B-2 is the flow and pressure drop variations versus filter loading data at a fairly high initial pressure drop setting while Table B-3 is similar data for a lower initial pressure drop. The data are plotted in Figure B-3. The initial and final (with approximately 1 gram loading) pressure drops and flows for this data are identified as Runs 3 and 4 on Figure B-2. B.4 MATERIAL BALANCE DATA Additional Table B-2 Data: Amount of debris added to system: Amount of debris collected on filters: Amount of debris lost in system : * some spillage, most remains on walls of tube leading from funnel to filter holder. Removal of 8y membrane from under debris after light shaking: Original weight: Amount remaining on filter: Removal of 8y membrane from under debris and air blow: Original weight: Amount remaining on filter: 1036.0 mg 1000.6 35.4 mg 15.3 mg 14.1 1.2 mg 14.7 mg 14.1 0.6 mg igss Si B-1 HWBUI0002541 Figure B-l - Sketch of Experimental Setup for Defining Filter Flow Characteristics of Nuclepore Membranes B-2 HWBUI0002542 Table B-l - Flow Versus Pressure Drop Data Unloaded Filters Run 1 Run 2 Flow (CFtl) AP (cm Hg) Flow (CFM) AP (cm Hg) 0.9 2.3 3.8 5.9 7.7 9.7 11.0 12.0 1.0 3.4 6.2 11.0 16.6 23.9 28.9 33.4 0.9 2.3 3.8 . 7.7 9.7 1.0 3.7 6.5 17.1 24.2 Loaded Filter Run 1 Flow (CFtl) AP (cm Hg) 0.9 2.0 3.0 4.0 5.0 7.0 9.0 11.0 12.0 1.0 3.1 4.1 7.2 9.6 15.2 22.7 30.6 35.6 P-82-99-3 Figure B-2 -- Plot of Flow Versus Pressure Drop B-3 HWBUI0002543 Table B-2 - Flow and Pressure Drop Versus Loading Data Wear Debris Reservoir Weight (gn>) 13.0160 13.0025 12.9744 12.9528 12.8883 12.7911 12.5590 12.3547 11.9800 AW (mg) -- 13.5 28.1 21.6 64.5 97.2 232.1 204.3 374.7 Filter Loading (mg) Pressure Drop (cm Hg) Flow (CFM) 0.0 52.7 - 27.6 = 25.1 13.5 51.4 - 28.8 = 22.6 41.6 51.3 - 28.9 = 22.4 63.2 51.1 - 29.2 = 21.9 127.7 50.7 - 29.6 = 21.1 224.9 50.4 - 29.9 = 20.5 457.0 50.3 - 30.0 = 20.3 661.3 50.1 - 30.2 = 19.9 1,036.0 50.0 - 30.3 = 19.7 9.4 9.0 9.0 8.8 8.7 8.4 8.3 8.2 8.0 PRESSURE DROP (CM HG> FLOW (CFH) Figure B-3 - Plot of Pressure Drop and Flow Versus Filter Loading B-4 HWBUI0002544 Table B-3 - Flow and Pressure Drop Versus Loading Data Wear Debris Reservoir Weight (gm) 13.7892 13.7741 13.7590 13.7416 13.7012 13.5552 13.3928 13.5198 12.8014 AW (mg) -- 15.1 15.1 17.4 40.4 146.0 162.4 233.0 358.4 Filter Loading (mg) 0.0 15.1 30.2 47.6 98.0 234.0 396.4 629.4 987.8 Pressure Drop (cm Hg) 43.3 - 36.9 = 6.4 43.3 - 36.9 - 6.4 43.2 - 36.9 = 6.3 43.2 - 36.9 = 6.3 43.2 - 36.9 - 6.3 43.2 - 36.9 - 6.3 43.2 - 37.0 = 6.2 43.2 - 37.0 = 6.2 43.2 - 37.0 = 6.2 Flow (CFM) 3.8 3.7 3.7 3.7 3.7 3.7 3.7 3.7 3.6 Additional Table B-3 Data: Removal of 8y membrane from under debris after light shaking and air blow: Original weight: Amount remaining on filter: Weight of funnel plus tubing plus debris: Weight of funnel plus tubing: Weight of debris trapped: 14.9 mg 13.4 0.6 mg 43,187.0 mg 43,144.8 42.2 mg B.5 CONCLUSIONS The flow rate through the filters is fairly reproducible for different membranes. Gas flow is proportional to pressure drop. 2 Initial 1 to 2 mg/cm loading reduces flow slightly (^10 percent). 2 Higher loading (up to 100 mg/cm ) reduces flow an additional 10 percent. Wear debris can be separated from membrane and stored in a vial without any significant loss. B-5 HWBUI0002545  Minimal losses will occur in the heated transfer line between the shroud and the filter holder. Over-all filter setup chosed for this project appears to operate satisfactorily for approximately 1 gram of collected debris. Larger amounts will require a pre filter of several filter changes. B-6 HWBUI0002546 Sequence 1 2 3 4 5A 5B 5C 5D 6 APPENDIX C COLLECTION OF BRAKE AND CLUTCH EMISSIONS Description Preparation of Collection Systems (Table C-l) Installation of Collection System on Test Vehicle (Table C-2) Preparation of Vehicle for Test (Table C-3) Vehicle Test (Table C-4) Removal of Particulate Filters and Gas Collectors from Vehicle (Table C-5) Removal of Wear Debris from Disc Brake Emissions Collector (Table C-6) Removal of Wear Debris from Drum Brake Emissions Collector (Table C-7) Removal of Wear Debris from Clutch Emissions Collector (Table C-8) Transfer of Samples to Analytical Laboratories (Table C-9) C-l HWBUI0002547 C-2 HWBUI0002548 Table C -l - P re p a ra tio n o f C o lle c tio n System fo r V e h ic le T est - Sequence CNJ C-3 HWBUI0002549 Table C-2 - In s ta lla tio n o f C o lle c tio n Systems - Sequence T able C-3 - P re p a ra tio n o f V e h ic le fo r T est - Sequence C-4 HWBUI0002550 T able C-4 - V e h ic le Test - Sequence C-5 HWBUI0002551 C-6 HWBUI0002552 T a b le C -5 - Rem oval o f P a r t ic u la t e F i l t e r s and Gas C o lle c t io n fro m V e h ic le - Sequence 5A T able C-6 - Removal o f Wear D e b ris from D isc Brake E m issions C o lle c to r (1 o f 2) C-7 HWBUI0002553 CM 4O-1 k CM 08 HWBUI0002554 T a b le C -6 - Rem oval o f Wear D e b ris fro m D is c B ra k e E m is s io n s C o lle c t o r - Sequence 5B T a b le C-7 - Rem oval o f Wear D e b ris From Drum B ra k e E m is s io n s C o lle c t o r - Sequence 5C (1 o f 2) C-9 HWBUI0002555 C-10 T a b le C-7 - Rem oval o f Wear D e b ris From Drum B ra ke E m is s io n s C o lle c t o r - Sequence 5C HWBUI0002556 T a b le C-8 - Rem oval o f Wear D e b ris From C lu tc h E m is s io n s C o lle c t o r - Sequence 5D C-ll HWBUI0002557 C-12 T a b le C-9 - T ra n s fe r o f Samples to A n a ly tic a l L a b o ra to rie s - Sequence HWBUI0002558 APPENDIX D ANALYSIS OF BRAKE AND CLUTCH EMISSIONS Sequence 7A 7B 8A 8B 9 10A 10B 11A 11B 12 Process Particulate Filters (for Material Balance Considera tions) (Table D-l) Redistribution of Particulates for Microscopy (Table D-2) Particulate Analysis by Optical Microscopy (Table D-3) Particulate Analysis by Electron Microscopy (Table D-4) Asbestos Emissions and Size Distribution (Table D-5) Noncondensable Gaseous Emissions Processing (Table D-6) Analysis of Noncondensable Gaseous Emissions (Table D-7) Condensable Gaseous Emission Processing (Table D-8) Analysis of Condensable Gaseous Emissions (Table D-9) Total Gaseous Emissions (Table D-10) sp $3 D-l HWBUI0002559 D-2 HWBUI0002560 T a b le D - l - P ro c e s s in g P a r tic u la te Samples - Sequence 7A T a b le D -2 - R e d is t r ib u t io n o f P a r t ic u la t e s f o r M ic ro s c o p y - Sequence 7B c o H 4-1 3 *H U 4-> CO H CO Td 44 ua cd r-4 B& 0) 6 P4 cd (0 a) > Ml Q) co a) ph a) 43 CO 4-1 u> *H & T) 4-1 a> G CO <u G > *H r--J u o CO Ml 'd 0) a) 44 P4 cd a) a) 4-1 4i rH H G MM cd 1 d) \ 0) H 4-1 u CJ G PM <U ^> a U iH CO 01 O > CO Sn 4-1 H c i--1 cd H *rf 43 o H H CO G G 42 OU cm 0) CO H 0) Pi! G cd H O H G 43 a a) H gU oo H 4-1 4-> cd cd Mi ao o 43 cd 1-3 z o 4-1 cd o 43 cd Sn cd G& O H cn E 4-1 3 G 0) G 43 CO o *rl PD cd Ml > 4-1 . CO tH rH H Ml o T3 Td M4 4-1 s OG o 4J o T3 O C r* cd o O Mi --1 4-1 0) cd 44 >o H 4-1 O 01 o o CO G PM 0) O Ml o 4-1 PH Q CO Jn cd 1 cn n a) G CO G O cd > Mi o TJ Mi 4J OG 4-1 O O' P5 5O rH 4-1 cd 44 O 4-1 n oo G PM o O 4J pa CO o CO c o cd 01 5 G 4-1 a) u 43 CO PM 4-1 0) G 4-1 O CO 4-1 fl) M4 > o Td 43 cd CO PM G Q) *H ai (0 44 T3 bU a> C Ml CO H a> G 43 > CO o a) cd CJ 43 s rH Mi a) PM o Ml PM "G 0) 0) 43 cd i-M H 0) rH CM CO Q) Ml ' G CO G M GG cd cd H H OO H H GG 43 43 Oa 0) a) HH cG cd cd H CJ O H H GG 43 43 OO 0) fl) HH G cd o H G 43 o 01 H G cd o H G 43 O <11 H G` cd H O G 41 O 0) H > Mi Mi Oo 4-1 4J cd cd Mi U o "O X 43 cd - cd 1-3 4 > > Mi Mi OO 4-4 4-> cd cd Ml Ml oO 43 43 cd cd i-3 r-l Mi O 4-> cd Mi o 43 cd H-l . >* Mi 0 4-1 cd M4 O 43 cd hJ Mi O 4-1 . cd Ml o 43 cd hJ- a o *rM 4J cd U . 0) cm o bo Cd 6n 0) LO > o a) 03 o <U bOrH +ca H E m 44 Cd in cd 03 H O <J >i H 43 43 -a bO *H 4-1 B 0) cd o & o Ml pa *4--1 a) rH 6 CO ag agO Cd *H *rH 03 4J 4-1 cd O <J rH cd H PM Mi Pm 43 G bO O *H fc} cd H 0) m oo U 4-1 O C3 o Pm o c* G H H rH O G cd 43 4-> a) rH rH Mm O O O O oU H a) 44 G cd H a> pa Td Go 0) o a CO o Gn ,03 04 a) H o PM 01 rH a G G. <N O Q O E E r<r o 4J G O Q) G Mi cd a) Mi 4-1 43 r-i E H (U ph s o 4-1 G M Mi G O PM rH T3 H PM cd 03 Td G cd G O H 4-1 G iH O 03 rH rH 0) Ml a *H G &G 03 Pm G o 4-1 G Ml (1) O 4-1 03 rH H PM Pm o 4-1 a) 03 > o G 43 0) cd 43 HB o 7d G CM cd rH a G 4-> GG OO cd > s 014-4 Mi cd > 4-1 (1) 03 hH 0) H G 01 G> & (1) O PM 4-1 O 4-1 xi 03 T) <G H ' cd bo M4 cd < 0) 4-1 Ml Td 1--1 0) G H 44 cd Pm cd <1) a a) m G> Go O 43 & cd cd o> QB 4-4 o 43 Ml CO cd CM cd 4-4 & 03 rH u G a) > o 03 o 43 4-1 4-1 Me s 01 4-1 CO rH <u H nd Pm 03 O 4-1 rH . ai 43 G CO G cd G Z2 4-1 G G CO O !% cd O t-4 4-1 Q a) 43 o PM CO rH n cd rH o & < C_3 .G CD ' rH G uG o cd cd > 43 4-1 rH rH G PM a) Td GG C3 cd G Pm PM O CO Mm H <D O 4J 43 G PM CO G' O *H iH HQs Ml H in . ai Ml > o 4-1 fl) u o u PM Mm 43 CO n M4 4-J a> PM G 0) Ml o 4-1 03 X) * cd . Ml <D 4J Pm <U >' O' E ai p4 D-3 HWBUI0002561 D-4 HWBUI0002562 T a b le D-3 - P a rtic u la te A n a ly s is by O p tic a l M icro sco p y - Sequence 8A (1 Of 2) T a b le D-3 - P a r tic u la te A n a ly s is by O p tic a l M ic ro s c o p y - Sequence 8A (2 o f 2) II Hi D-5 HWBUI0002563 D--6 HWBUI0002564 T a b le D -4 - P a r t ic u la t e A n a ly s is by E le c tr o n M ic ro s c o p y - Sequence 8B (1 O f 3) T a b le D -4 - P a r t ic u la t e A n a ly s is b y E le c tr o n M ic ro s c o p y - Sequence 8B D-7 HWBUI0002565 c*> +OH cn I D-8 HWBUI0002566 T a b le D -4 - P a r t ic u la t e A n a ly s is by E le c tr o n M ic ro s c o p y - Sequence 8B Table D-5 - Asbestos E m issions and S ize D is tr ib u tio n - Sequence D-9 HWBUI0002567 - -T a b le D -6 N o n co n d e n sa b le Gaseous E m is s io n s P ro c e s s in g Sequence 10A D-10 HWBUI0002568 -T a b le D -7 A n a ly s is o f N o n co n d e n sa b le Gaseous E m is s io n s - Sequence 10B D-ll HWBUI0002569 R em arks C le a n v e s s e l w ith s o lv e n t and re g re a s e s to p -c o c k . I<--1I <u o C<U3 3 o<u' u 14 Jh 14 3 3 a) 42 .<3 43 43 Pp P P cd cd 3 J-i 14 J-i J-i 00 bO 60 b0 60 oo O O JJ 4-1 4-J 4-1 a cd H CO a Fs E a CO o o o o CO M p 43 J-I .3 14 43 J-I 43 <ou CO CJ u O O o PCj CO CO CO (0 u Ph Ou O O CO aoH CO CO *rl 4^ 42 42 42 cd 3 W0 rJ tJ P P o CO 3 cd 4-1 a a cd CJ 3 O 3 3 *H P O 4J 4-1 OP 0)CO cd P cd 3 <d c Cl 3 O c -3 <C -3 14 14 33 42 42 PP 33 14 14 bO bi) OO 4-1 4-1 33 aa Oo 14 14 42 42 U l_> 33 33 OO 42 3 P P 3 O P 4-1 >> rH 3 -33 42 3 P rH 3 O rH 4-1 3 3 <! 14 M 14 33 3 42 42 42 PP P 3 14 14 14 bO bO bO Oo O 4J 33 3 aa a O 14 14 ,<3 42 42 o 33 3 33 3 OO O r 42 33 3 PP P rH P i--1 33 3 OO O rl *H P 4-1 P f>> T--J P T--1 33 3 Q3 < < <3 3O CO 3 O P P 3| 3 3 P <d (3 a) > O cd U 3 CO 33 'O rH o 3 3O 42 4-1 3 P OO3 rH 3 3 *T3 p 4-> e 4-j a 3 3 3 3 e3 (0 3 <O 3 P 3 14 o Cd 4-1 p 33 3> 3 a 3 O co 3 P 333 P 3P 3 42 > 14 P H rH 3 33 33 O *H u6 3 3 3 T3 3 O 4-> p 0) CO H P 43 4J o3 CO CO cd P a3 P3 4-1 O 3 O O oo i-i QH 42 bO Cd 3 CM 14 3 33 3 3 33 3 3 U 14 14 E-4 3 4-1 14 P O 42 i 3p CO P (-4 4-> 3 4-1 t-H TJ 14 3 P p cd 4-> T3 3 P 3 3O P33 u CO 3 3 Td 5Q 3 rH 3O a Po P 42 3 14 H uO 'd 3 O O 3 'O 3 H i-- 3O bo bO O mm P *3 O P3 42 P 4-1 3 3 H O o p3 o JXJ 3 42 U 3 3 3 3O 3 P cd zn o 3O 3 b0 JH rH 3 14 4-1 & 3 4J O 9 3 H 14 a) <u i--i 42 n3 3 T3 33 O3 MO CO P3 OU co 3 A Q) 3 J-i >4-1 O3o P 4-1 3P O3 !3 4-1 3r 3HP 3 4J H > 3 4-1 4J. i-H 3 3 O O 3 -W 4-1 H s 3O *H P i--) 3 C H a a O 14 42 U o <J J-i 4-1 P 4-1 3 3 4-1 3 22 3 14 3 3 < 3 O CM 33 CO 4J P 3 14 O P 3P P P 4J b0 rH 3P H 3 W3 g3 TJ 3 3 H 3 TJ O* 3 3 P 3 3 3 O T3 M !?; c <fl h a P 42 PQ 3 i-H 3 3 rC 3 P (3 1 i-4 O l- 4-1 3 3 3 3 3 OO hJ O > OH CO -1 O H ts 33 *3 42 ftCJ Vj *3 H P rH a3 P 43 Q 4J 3P a) cd P bO 3O O3 3 m> 3 4J 3 rH 3 > 3 42 PO Po <o 3 3 3 N a) H 3 14 3 o 3 *H 3 3 I 42 3 3 oP 4-t O 3P rH O O 3 4-1 3 3P 60 3 P c P co 14 pH C0 U P 43 QH 4-1 <4-1 4f2i C< 3 3 P3 3P pa co 3 H 3 3 > 42 42 3 P H 4-1 *3 3O 42 P 3 CO P3 33 3 O- 3 3 < D-12 HWBUI0002570 -T a b le D -9 A n a ly s is o f C ondensable Gaseous E m is s io n s - Sequence 11B D-13 HWBUI0002571 D-14 HWBUI0002572 Exhibit D-I MICROSCOPY COUNT SHEET Vehicle Test: _________ Schedule: ______________ ]____________ Collector: Count Weight: Sample: _________________ ___________ Date: Total Sample Weight:.mg LTA Weights (I/F): ./ _______ Comments: _____ mg .mg xooo'zz xoov Date Counted: OM Field Size: No. of Fields: ___________ 88^ x 88ii Microscopist: D--15 HWBUI0002573 Sample: Run No: Location: Box No: Comments: Exhibit D-II - Gaseous Emissions Data Sample Date: _____ Analyst: _____ Analysis Date: ____ Noncondensable Emissions P= V= T= n= CO CH4. C2 C2H6 Condensable Emissions: Sample Plus Tare: Tare: Sample: D-16 HWBUI0002574 w g FQ *3 CO co SW3 g o PM S3 CO O CO CJ M g PW O CO CO S3 MO CO W >H CO g <1 <3 u <1 dd g(On13) a po o /---s 1--1 g a) CO a) 0) 4-1 e CO o co o CO rH a) -- f--j o d CO H cd d uo co 00 03 S3 P P o Pi O O CO --00^ i--1 O rH g-U H o u g VD & O d 4-1 o cd H U d cd cu N s H gd cd o uM 00 o 0) pi a P cd --1 CJ p d) o gOg O cd d CM d H P aO 4J ao a CJ 0 oO o un O CM . co CO d rH o o rH d 4-1 <D 4-J cd gdU Pm <u g CO > g rH r*H <T H o cd CJ <D 0) CO CO CO CO U d d <U t--1 pH d O 0) <11 d o o d M0 O 4J M d 0) a CO cd !--1 CM d 1"1 1"1 (0 CO 0) o So g 0) CJ CJ Pi g 1 cj PQ H X p CJ p d d w g H QI CPQ CO gW CO S3 po -U g rl O CO O CO rCrl wX gb Og S3 W Op co cSOWo3 M CO c>Ho <O1 3 dg0O) cd p rH o 4(UJ CO CO CO 1--<UI O' ah gd cd p P cd CO u o p P g o CO 0) e 00 ""v. o i-H o rH - --^ oo t--1 00 d a d oa> j-i 60 O Pi p o o CU O g vO rH d O 6 4J o CJ oo O o . 00 <r 1 H > 4HJ U d d d o cj a) g H pH cd d p CJ 0 O p rH (U d a) H d X H oX do o H -CO dd a) 0) gP 0) <r CJ 00 00 d d d d <D o o 0) o cd o d Pi o Pi u 00 P g P cd 0) 4-1 d 4-1 P Pi 4-1 Pi g 4-1 P H X cd <D cd 4-1 cd CO g g o CJ g CJ w S3 o CO O g M EH Eh H CO Eh gM M g 3 d P r--1 HO CO CJ g M <D a Pi cd d Pi 4-J 00 cd o Pi Pi (U PP a <U 0) g Pi H w d 4-1 Pi Rp cd O o Pi 4-1 <U o g Cu <u o o rH a <u 4J M <u H P H pP g w $/ S':;;:. W;' D-17 HWBUI0002575 APPENDIX E ELECTRON MICROSCOPY ANALYSIS OF BRAKE EMISSION SAMPLES E.l SUMMARY The previously developed technique for preparing brake emission debris for viewing with the electron microscope was found to be unsat isfactory for test car samples. That procedure was developed with standards and employee vehicle wear debris. When applied to the test vehicle brake emissions, excessive film breakage and instability occurred. Consequently, the technique was modified so that it was reasonably sure that no further difficulties would be encountered with the remaining samples to be analyzed. Basically success was achieved with (1) use of a finer copper grid, (2) double carbon film, and (3) reduced debris loading on the film. The details of the procedure, as modified, are given in Exhibit E-l. E. 2 PROBLEM The difficulty with the previously developed procedure was manifested by the following: (1) Fracture of the carbon film occurred during the cleaning operation employed to remove the filter. The fracture was so severe that carbon film coverage of the copper grid varied from 0 to 10% of the area originally covered. (2) The large wear debris particles were unstable under the electron beam (expansion and contraction). This particle movement tore the film and released the wear debris so that analysis was impossible. (3) The filter and carbon film appeared to shrink under the action of the cleaning solvent. This raised the question as to whether concentration of the wear debris was occurring because of the shrinkage and in this way creating a discrepancy between the real and observed number of particles per unit area. E.3 PROBABLE CAUSES It was assumed that the first two difficulties were created by the large nonasbestos wear debris particles. These particles appeared larger than those previously found in wear debris samples from the employee vehicle drum surfaces. The large particles apparently cause film tears because they created thin spots or openings in the carbon film during evaporation attributable to shadowing effects. These weak E-l HWBUI0002576 spots in the film and the bulk of the particles then contributed to film tearing during the strains of the cleaning operation. The particle instability under the electron beam could be attributed to heating and charging effects. The third difficulty occurred because the filter shrank under the action of the solvent before dissolving. Since the carbon film and filter were initially attached, the filter stressed the carbon film and also caused it to distort or crack. E.4 SOLUTION ATTEMPTED The following procedures or materials were tried in an attempt to eliminate the film breakage problem. (1) A range of film thicknesses was investigated from thinner to thicker than that previously used. (2) Silicon monoxide films were used in the hope that they would provide additional strength. (3) Different films and film combinations were tried. These included chromium, platinum-carbon, chromium on carbon and collodion on carbon. (4) Extreme care was exercised in laying the film on the grid prior to cleaning in order to ensure no handling breakage. (5) The time to clean the filter from the film was shortened in an attempt to prevent breakage during cleaning. (6) The cleaning solution was diluted or saturated in an attempt to reduce the severity of the cleaning operation. (7) (8) (9) (10) Carbon film was cleaned both while on a rough and on a smooth side of carbon grid. Vacuum drying of the film following cleaning was tried to remove all traces of residual solvent in case this was the cause of particle instability under the beam. A finer mesh grid (400 instead of 200) was tried to provide greater support for the carbon film. A second carbon film was vaporized over the wear debris and onto the original film following the cleaning operation. This sandwiched the wear debris between two carbon films. It was hoped that the additional film would provide additional strength and reduce particle instability. ! j E-2 HWBUI0002577 E.5 SUCCESSFUL APPROACHES Only the last three of the above mentioned techniques proved significant in combating film breakage and instability: fine mesh grid, double carbon film and lower debris loading on the film. The 400 mesh grid apparently provided greater support for the film and did reduce breakage during cleaning. The second carbon film applied after clean ing measurably strengthened the original film. Although there still appeared to be some particle instability under the beam the double film Was strong enough to resist tearing. The effect of the degree of loading on film integrity following cleaning is shown in Figure E-l. The debris was an air-borne sample taken from the right rear brake following the baseline test. All samples were on a 400 mesh grid and received a double carbon film. The debris loadings on the submitted filters were 1, 0.50, and 0.30 mg. Thfe last two show good film coverage while the 1 mg loading shows extensive breakage. Also the 1 mg loading showed more stability under the beam than did the other two. Therefore, even though the counting time was increased because of the decreased loading, the filter loading was reduced to less than 0.60 mg. Another consideration in selecting the debris loading for the film was the possibility that the large particles would mask the asbestos fibers. Considering the extreme, if the debris were concen trated to the extent that particles covered the entire grid opening, few fibers would be seen. The debris distribution for several initial filter loadings is shown in Figure E-2. The possibility of erroneous results due to film shrinkage with attendant concentration of the debris was investigated. It did not seem reasonable that debris concentration would occur because the debris was attached to the carbon film and the carbon film itself would not shrink under the action of the solvent. Since the periphery of the carbon film was changing it was assumed that this could only occur by fracture or distortion of the film. Optical microscopy confirmed the assumptions. The film did not undergo shrinkage, but rather accomplished the dimen sional change by forming folds as shown in Figure E-3. These folds were confined mainly to the periphery and would not pose any difficulty to the measurement as long as the folded area was avoided. In other films the folds were not confined entirely to the periphery, however, these areas could easily be avoided during the counting as they were readily detectable on viewing with the electron microscope. Therefore, the film folding under the action of the solvent posed no serious diffi culties in the analysis. E-3 HWBUI0002578 Figure E-l - Photomicrographs of Carbon Films on 400-Mesh Grid 50X [Films were made from filters with original debris loading of 0.3 (upper), 0.5 (center) and 1 (lower) mg. Debris was airborne collection from A.B. Baseline schedule of vehicle test for right rear drum brake] E-4 HWBUI0002579 Figure E-2 - Electron Micrographs of Debris Distribution on Carbon Films Made from Initial Filter Loadings of 0.3 (upper), 0.5 (center) and 1 (lower) mg. (Debris was airborne collection from Baseline test) 6000X E-5 HWBUI0002580 f. f Figure E-3 - Photomicrograph of Carbon Film on 400-Mesh Grid Following Cleaning-Operation. 50X (Shrinkage was due to film folds near periphery of film. Film was made from filter with 0.5 mg debris loading.) ' E.6 CONCLUSIONS 1. The modified technique for preparing electron microscopy samples should provide a trouble-free method for analyzing all samples required for the program. 2. Loading of the filter was limited to less than 0.6 mg. This was required to prevent film breakage and, in addition, there was less chance of the large particles masking the asbestos fibers. 3. Film folding during cleaning did not introduce error into the analysis. E-6 HWBUI0002581 EXHIBIT E-I PROCEDURE FOR ELECTRON MICROSCOPY ANALYSIS OF BRAKE EMISSION SAMPLES 1. Receive and log in sample. 2. Cut out 1/2-inch filter circle with special cutter. 3. Mount filter circle (debris side up) on glass slide with tape. 4. Vacuum evaporate carbon film on filter circle (4 mm length of carbon rod; 1 mm in diameter). 5. Remove from evaporator and cut out four 1/8-inch circles with special cutter. 6. Prepare cleaning bath to remove filter from carbon film by the following steps. Put polyurethane foam block in a petri dish and fill with chloroform to about 1/8-inch below the foam surface. Put 400 mesh copper grids on the top surface of the foam. 7. Place 1/8-inch carbon-coated circles on grids in the chloroform bath with carbon side down. 8. Let sample clean in bath at least 3 hours. Bath is covered during cleaning to prevent contamination. 9. Remove the carbon film on the grid from bath and carbon coat the debris side in the vacuum evaporator. 10. Remove from evaporator and view in transmission electron microscope. E-7 HWBUI0002582 APPENDIX F COMPUTER PROGRAM FOR COMPUTATION OF ASBESTOS CONTENT USING ANALYTICAL MICROSCOPY TECHNIQUES FORTRAN IV G LEVEL 20 MAIN DATE = 730?4 15/19/51 0001 .******** *,.*** ************************************************** ****** *********** C IN TWO DIMENSIONAL ARP'AYS, l NDP X I CORRESPONDS TO DOWS 11 - 14,. C 15* 2 5 16*26., 27 -31 ON MICROSCOPY COUNT SHF.FT AND INDEX J C CORRESPONDS TO COLUMNS 1 - 8' ON micp-OSCOPY COUNT SHEET. C************************************************************************ ******* DIMENSION BOX (49) ,VOL(12,8). I VOL (12 , P U'FCTP (2 1 ,VPLT<49> 1 , INM(8),XOL<12 * B1, 1 IVH(5),ISCH<20),ICOM20),ISAMP<20),10ATF(TO),T0T(2I C******************************************************************************* C ARRAY XOL CONTAINS VOLUME * 10**12. C******************************************************************************* 0002 DATA XOL/.077E-3,.2F-3,.53E-3,9*0.0, 2 .L92E-3,.56-3,1.316-3,10.156-3,8*0.0, 3 .51E-3, 1.34E-3,3.5E-3,27.1F-3.123.E-3,.982,6*0.0, 4 1.66E--3,4.25E-3,11.3E-3,38.06-3,401.E-3,3.19,12.8.33.2,63.6, 4 3*0.0, 5. 3.83E-3,10,-3,26.3E-3,203.E-3frto0.E-3,7.3,29.4,76.7,l47., 5 241.,436.,0., 6 7.776-3,20.6-3,52.66-3,406.6-3,1780.E-3,l3.5,53.8,L41,,270., 6 433.,805.,1.62E3, 7 4*0.0.2960.F-3,23.3,93.,243.,466.,765.,13R0.,2.89E3, 8 4*0.0,4740.6-3,39.3,157.,409.,784.,1290.*2.3363,4. 69E3/ C**** ** ** ** ** ************ ************************ ******** ****** ft#**#*#*#******** C **** *F*C**T*R*I *1*) *M**U*L*T*IP**LI*E*S**P*O*W*S***1*1*-1*6*.***F*C**T*R*(2*)**M**U*L*T*IP**LI*E*S**R*O*W*S**2**5*-3*1*.************* 0003 0004 0005 0006 0007 0008 DATA FCTR/110.E6,.12784F6/ DO 190 1-1,12 PRINT 999,I.{XOL(I,J).J = 1 .8) 999 FORMAT!I5,8E14.5) = 190 CONTINUE STEP=1.0/6.0 C* *********************************** ******************************************* C COMPUTE ARRAY BOX POP L OG-LOG' PLOT.; BOX = 10**-16 TO 10**-R IN . C STEPS OF l_0**l/6. C**** ** ************ ** **** ****-********************************************.******* 0009 0010 0011 DO 10 1=1,49 BOX(Il = STP*( !-l )-4.0-13.0 10 CONTINUE C******** *********************************************************************** C COMPUTE VOLU.JI = THE LOG OF XOL(I.J) AND MULTIPLY XOL (I,J) BY C 10**-12 C******************************************************************************* 0012 0013 0014 0015 0016 0017 0018 DO 20 1=1,12 DO 20 J=1,8 IVOL<I,J)=0 VOL(I,J)=0.0 lF(XOL(I,J).60.0.) GO TO 20 VOL(IJ)=ALOG10(XOL(I,J))-12.0 XOL<I.J)=XOL<I,J)*l.E-12 C******************************************************************************* C COMPARE VOL I I,J) TO ARRAY BOX. ASSIGN NUMBER OF CLOSFST BOX TO C IVOL(l,J>. c**** *************************************************************************** 0019 0020 0021 0022 0023 DO 30 K=1.49 IF(VOL(IJI.LE.BOX(K))GO TO 40 30 CONTINUE K=49 40 IVOU I, J)=K PAGF 0001 F--1 HWBUI0002583 FORTRAN IV G LEVEL 20 MAIN DATF = 73024 15/19/51 0024 0025 0026 0 02 7 0028 0029 0030 0031 0032 0033 0034 0035 0036 0037 0038 0039 0040 0041 0042 0043 0044 0045 0046 0047 0048 0049 0050 0051 0052 0053 0054 0055 0056 0057 0058 0059 0060 0061 0062 0063 0064 0065 0066 20 CONTINUE DO 50 1-1.49 BOX(l >-lO.O**BOX( I> 50 CONTINUE DO 191 I 11 12 PRINT 999*1.(VOL(IJ),J-1,8) 191 CONTINUE C C start ooing stuff co*** *************************************************************************** c REAO HEADING CARD. c*** *************************************************************.************** 100 REAO ( 5.900.END*500 1 II l ! ICOL(K),K*1,20),(IS 900 FORMAT!14,IX# 75 41) REAO (5,901r lO.TOTWT, 901 FORMAT!14,6 X,4E10.0) IF! IO.NE.101)GO TO 501 T0TWT-T0TWT+.001 XIXI*.001 XF*XF*.001 XCNT-XC NT *.001 DO 200 1*1,49 VPLTIIMO.O **2*00 *C*O**N*T*IN*U*E************************************************** ********** ******* C****T*O*T*(*l*1**=**S*U*M**O*F**R*O*W*S**1*1*-*1*6*.***T*O**T*(2*l**=**S*U*M**O*F**R*O*W*S**2**5*-3*1*.******************* TOT!1)*0.0 TOT(2)*0.0 TOTAL^O.O 110 READ (5,920) ID,I,J,(INM(K),K=l,8)lTOT IF(10.NE.ID11 GO TO 501 920 FORMAT!14,21l,4X,813.6X,15) IF! I.E0.2)J=J + 1 IF!I.EQ.3)J-J+ll IF!I.E0.3)1-2 DO 120 K=18 TOTH )*TOT( l)+2.56*X0L( J,K)*INM(K)*FCTR( I) / ITOT IF!IVOL!J,K).EQ.O)GO TO 120 VPLT! IVOL!J,K))*VPLT(1VOL(J,K))lNM(K)*FCTR(I)/ITOT o*1*2*0 *C*O*N**TI*N*U*E******************************************************************* C TEST FOR LAST DATA CARD. ************************************* ****************************************** IF(I.E0.2.AND.J.E0.12IG0 TO 130 GO TO 110 130 CONTINUE TOTAL=TOT(l)+T0T<2> PCT=T0TAL*XF*100.0/IXCNT*XM PRINT 911 911 FOR MAT(1 45 X * **** ASBESTOS ANALYTICAL RESULTS *+**) PRINT 912.TOT(lJ.TOT(2l 912 FORMAT!'O'.105X,FM SUR TOTAL =.El 1.4./106X.OM SUB TOTAL 1 Ell.4) PRINT 910.! IVH< M,1-1,5).tlOATE(J).J=l10),XI .TOTAL, 1 (ISCHIKI,K*l,20),101,XF.XCNT, =* , PAGE 0002 F-2 HWBUI0002584 o o <4- * X 0-4 CO - ITi or *4 - o II * I- * --- * O m t-- Uj * hOC _/ z o - - II X X h rX -H in 2 uj * - - <_> --< * a < ~-l UJ o < v. o. --O* ofvj <t W * h Ho- CO 111 * UJ UJ h Ui C <T __I CO - OO D II < fUJK * X Ix X fofiouio-ino - tu * -1*3-* 3 c -- t/5 * O UJ -- CM I-- < o * xa 3 O *< > </) * x 11 i- -J -- X * at lu i o l X CO I oO w u<or ----o -- -- > < > O -*-j ar o i-- a u jO 0.0 J < O' | --' o -- mO' O0s' OWK _i o-i 1- O' O --< < O < k- < -j o o' o c-> w -- x or x oO < I- H JhQ.IZ(L c M0U<O0^HLL0Ot-flh/lUZJ m- oo -- o in o in O' in oocod aooo1 fooo~ h noo oorru* (n noo 't oo " o co <j k CO O' < o - o kk^t < in co - <4- <k- *O' XW> HcN--Xako- X -1 c > CL O CO > * o *- O >1- co k CD C-< <f ^ <-4 _r o <-< CM k- f<1 < O' o -* E O Xo o a; a I X. k- o > O 00 to 11. -- -J o O CM CO O' o -- o o -t k- h- o- < <i --I --I z o -- sT O O k u. uj o < fC k O' z o ~ k- CD CM vO -a O' < O cc a. _J < XU k- > k- o o W W Mk x a x k- >>>o --1^0 O o in CO O' O' cl y< to X to o X--o <f k IA Of < 1-t ID O oo CL -- CC O o <1 k- r*- co < X < *-4 o ~-4-0'4- O- k- h- -0 U <r c rv nO cc X l_> o > _J k- O < -1 O CL <n a_ XX >- CL tO u. X -j -J < >- o o o co > x *-h Joo OO CO O' O' X > to z o -* o I<- >-tl u > -J * ox CO O z o k k CD O' < _4 _< V) N J X o hOO^ k CO -t O' < ^ <o co XXX< kOZW to CO M -N z c *-- -t CM J O' M O O' ^ tD O' O' .X to O P TIO N S IN E F F E C T * I D , EBCD IC , SOURCE , NOL I ST , DEC K , NOLOAD , MAP OPTIONS IN E FFE C T* NAME = MAIN , LINECNT = S TA TIS TIC S * SOURCE STATEMENTS = 7 4 , PROGRAM S IZ E 60 - 49T4 S T A T IS T IC S * NO D IA G N O S T IC S GENERATED `&M iit! f-3 HWBUI0002585 APPENDIX G VEHICLE TEST DATA REPORTS Because of the large number of test reports for different phases of the vehicle tests and subsequent collection of brake emissions, the individual data sheets were assembled into a separate volume for each vehicle test. The original test data sheets were retained by the Principal Investigator at Bendix Research Laboratories. The volume for each vehicle test contains seven major sections, one for each test schedule; and each section is broken down further into three secondary sections for the vehicle operation results, the brake emissions data sheets, and the analytical results: Vehicle operation data reports: Figure G-l is the Vehicle Log Sheet which remained in the vehicle to record the overall vehicle operation throughout the test schedule. Figure G-2 is the Burnish report. Figure G-3 is the Detroit Traffic data sheetused duringthe Baseline and Detroit Traffic schedules. Figure G-4 is the 10-Stop Fade data sheet. Figure G-5 is the 15-Stop Fade data sheet. Figure G-6 is the Recovery data sheet used inconjunction with the 10- and 15-Stop Fade data sheets during the Fade schedules. Figure G-7 is the Reburnish data sheet used in the 10- and 15-Stop Fade schedules. Brake Emissions data sheets: Figures G-8 and G-9 are the Wear Sheet-Front Axle and Wear Sheet-Rear Axle used to record both weight and thickness losses of all friction materials on the vehicle. Figures G-10 and G-ll are the Collection of Brake Emissions (Right Side) and Collection of Brake Emissions (Left/Control Side) which were used to record and calculate the brake wear debris collected in the sump, on the surfaces, and the airborne samples. Figure G-12 is the Test Summary Sheet used to summarize the wear data and to check out the material balances. G-l HWBUI0002586 Analytical Results: Figure 8-13 is the Microscopy Count Sheet onto which each fiber found by microscopy was classified according to its dimensions. Figure 8-16 is the Computer Printout which gave the asbestos percent and the size distribution. Figure G-13 is the Asbestos Emission Summary Sheet used to summarize each test schedule. G-2 HWBUI0002587 1/ b H i c l B og-L S t iv e r F ig u re G--1 - V e h ic le Log S heet So . S J* {2 V,LS^So 5 e \Ssi 5! a -4 . Ct ft* 41 I ^ Q> S<s0. <U VS V'w<o < 1 Cfl/p. ^ ^V- 5 S: Qt psfe Wi G-3 HWBUI0002588 Test No. Burnish (Brake Performance Test Weight): Odometer DateTemp. 200 Stops, 40 MPH, 12 FPSPS in Highest Gear, 250 IBT or 1 Mile Maximum 40 MPH Cooling Speed, 80% Burnish Stop Initial Temperatures F L.F. R.F. L.R. R.R. Max, L.P. psi Max. P.F, lb. P.T. in. Comments 1 20 40 60 80 100 120 140 160 180 200 Comments Figure G-2 - Burnish Report G-4 HWBUI0002589 START 01)0M. DETROIT CITY TRAFFIC TEST .END ODOM .TEST NO DATE CAR NO. START ** MICH. & 3RD. ** ADAMS & WOOD. ADAMS & WOOD. 12 MILE ** ADAMS & WOOD. ADAMS & WOOD. ADAMS & WOOD. 12 MILE ** ADAMS & WOOD. ADAMS & WOOD. ADAMS & WOOD. 12 MILE ** ADAMS & WOOD. ADAMS & WOOD. ADAMS & WOOD. 12 MILE ** ADAMS & WOOD. ADAMS; & WOOD. ADAMS & WOOD. 12 MILE ** ADAMS & WOOD. ADAMS & WOOD. END ** TIME AMB. L.F. R.F. L.R. R.R. Figure G-3 - Detroit Traffic Data Sheet COMMENTS HWBUI0002590 Test No. First Baseline: (Brake Performance Test Weight) OdometerDateTemp, 3 Stops, 30 MPH, 10 FPSPS in Highest Gear, 140-150 IBT Each Stop Stop Initial Temperatures F L.F. R.F. L.R. R.R. L,P.*PSI Init, Sust. Fin, Max, P.F. lb. P*T* in. Comments 1 2 3 First Fade: OdometerDateTemp.Wind 10 Stops, 60 MPH, 15 FPSPS in Highest Gear, 0.4 Mile Interval, 140-150 IBT For Stop I Req't.: 200# Maximum P.E. First Four Stops Stop Initial Temperatures F** L.F. R.F. L.R. R.R. L,P.*PSI Init. Sust. Fin. P.F. lb. Init. Sust. Fin. P.T. in. Comments 1 2 3 4 5 6 7 8 9 10 Total Elapsed TimeSec. k Must include maximum line pressure. kk On cooling cycle* Figure G-4 - 10-Stop Fade Data Sheet G-6 HWBUI0002591 First Baseline: (Brake Performance Test Weight) Odometer Date Temp. Test No. Stop 1 3 Stops , 30 MPH, 10 FPSPS in Gear, 140-150' IBT Each Stop Initial Temperatures F L.F. R.F. L.R. R.R. L.P.*PSI Init. Sust. Fin. Max. lb. P.T. in. 2 3 Comments First Fade: OdometerDate _____________Temp.Wind 15 Stops, 60 MPH, 15 FPSPS in Gear, 0.4 Mile Interval, 140-150 IBT For First Stop Req't.: 200?/ Maximum P.E. For First Eight Stops Stop i Initial Temperatures F L.F. R.F. L.R. R.R. L.P.*PSI Init. Sust. Fin. P.F. lb Init. Sust. Fin. P.T. in. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Total Elapsed Time Sec. Comments Must include maximum line pressure. Figure G-5 - 15-Stop Fade Data Sheet G-7, HWBUI0002592 Test No, Recovery - One Mile After Last Fade Stop, at 40 MPH Cooling Speed 12 Stops, 30 MPH, 10 FPSPS in Highest Gear, or Max. Decel. for 200 lb. PF if 10 FPSPS Cannot Be Obtained, One Mile Interval Req't.: 5 FPSPS Minimum for 200// P.E. for First 5 Stops, 150# Maximum P.E. for Stop #6 Stop Initial Temperatures F L.F. R.F. L.R. R.R. L.P.*PSI Init. Sust. Fin. P.F. lb. Init. Sust. Fin. P.T. in. Comments 1 2 3 4 5 6 7 8 9 10 11 12 Driver . .Observer. (1) If warning light is activated, which M'cyl. piston bottomed out? Front Rear CH (2) Center differential valve piston. (3) Check front/rear pressure differential: }Front line pressurePSI Rear line pressure . .PSI Read simultaneously at approx. 1000 PSI Figure G-6 - Recovery Data Sheet G--8 HWBUI0002593 Reburnish (Brake Performance Test Weight): OdometerDateTemp. 35 Stops, 40 MPH, 12 FPSPS, 250 IBT of One Mile Maximum in Gear Stop Initial^Temperatures F L.F. R.F. L.R, R.R, Max, L,P. psi Max, P.F, lb. P.T. in. Comments 1 10 20 25 35 Figure G-7 - Reburnish Data Sheet HWBUI0002594 Vehicle Test No. Test Sequence: WEAR SHEETAXLE MAKE OF CAR:______________________ MODEL: LINING:______________________________________________________ COMMENTS:____________________________________________________ BTGHT Date Mileage Total Stops 1 2 3 4 5 6 Weight Loss RIGHT Date Mileage Total Stops 1 2 3 4 5 6 7 Weight Loss T.F.FT Date Mileage Total Stops 1 2 3 4 5 6 Weight Loss LEFT Date Mileage Total Stops 1 2 3 4 5 6 7 Weight Loss G-10 , Figure G-8 - Wear Sheet - Front Axle HWBUI0002595 Vehicle Test No. Test Sequence:__ WEAR SHEET AXLE MAKE OF CAR:MODEL: LINING:____________________________________ ' ______________________ COMMENTS:____________________________________________!__________________ RTGHT RRAtCE (PRI. OR FORWARD) Date Mileage Total Stops 1 2 3 4 5 6 Weight Loss i RTGHT RR AtCE (SEC, OR REVERSE) Date Mileage Total Stops 1 2 3 4 5 6 7 8 Weight Loss i# LEFT RR AtCE (PRI. OR FORWARD) Date Mileage Total Stops 1 2 3 4 5 6 Weight Loss T.F.FT RRA1<E (SEC. OR REVERSE) Date Mileage Total Stops 1 2 3 4 5 6 7 8 Weight Loss It Figure G-9 - Wear Sheet - Rear Axle G-ll HWBUI0002596 Coiieefiatl OF 0RAK5 EMISSIONS (Right sne') Febide Test No.______ Test Sequence:_____________________________________________ 7est Dates Start: Finish'. Might NRont 0fi.fiKE Sump Sample: CahpAt Picton Rivet Hole's Filter* debris'. Filter . Surfaces SftmpiE-FiHeMjttns: Stroud do-tor Filter: 'BaeKiNfj PlBte fliPdoPNE Sample Filter * debris total oh S-Om. total on 02 m. total on Filters: gm- Total SAMPLE r~ -------------- Suma: i i------------T Surfaces: 09hi--. J 1 i i i 1| ____ 1 r*~ r--------> Airborne: 1 J Comments: i 1 | i i i 1 1 i i 1 Might dpum brake Sump Sample Utear TraoK Fillet Holes Filter *debns: Z Fitters: ioial: -Xn% Pet urn: 2-0%Bdtiin: SuPFACES SAMPLE.-filter-idebris: 'BraKe Pa<ts Filter: Drum Fhp&tppE Sample FiUer <t-debris: Filter: titaL or S ou: totalon k%MiiotsL on Flier: g.OM. c.iu. ftn Corrected Suup Sample lAjeioht This test ! Previous test j i----------*------- r1i 1 r-1 _J Total Sample. L -------------- >. Sump: r 1 1 _J r~ -------------> Surfaces: ------------ =. Airborne : Comments: r _l Figure G-10 - Collection of Brake Emissions (Right Side) G-12 HWBUI0002597 CoLLecTioU oFgeam emissions (Left 91^ Vehicle Test Ho-. Test f&^uenae: _ LEFT f/eoNT 8#ME Sump Sample. Caliper faster Hiuet fates Filterv-debris 'filter: . Total Sample: i--------- ^ Sump m --------1 i- - Surfaces'. Surface SflmPLE: filters debris: Shroud * Cttae BaaHwff Phrte -filter:. tm----------------- 1 Comments: LEFT VfUM BROKE Sump Sample : filter sdebns: Wear TrgcK 2 filters: _ Hiuei Holes ~8o% reiurtr- 2.0% retoin: SuRFPU% SftmPLE: filter +debris; Brake Ports filter : _ Drum Corrected Sump Sample UteifthF i------- -> This Test : i Previous Test: ________ l i----------- <-------- > ii ji r' ______ i yn fm Total SAmpLE:* I --------------------------------------- > c-- 1II _____ I Sump Surfaces: Comments: Cl u tch filter -^debris: filter: told oh %o: total on 0. %' totaloh filters: 8-Om. 0.2m. Jm Figure G-ll - Collection of Brake Emissions (Left Side) s G-13 HWBUI0002598 Test SummaM SHEET Vehicle Test To._____ Ted: Sequence- _______________________________________________ Th'/eS; fiedaL Applications: Complete Stops'. Comments-. Test hates ; Start :m Finish . Ffoht T)tsc B&flKE:S PiGUT/sUdoubep Left/Contzol Temperature fiance Cf) '- Samples Collected (qnS)'Sump ' Surfaces: Aid borne ' ToTaL ' Lihina Ujear \ Fo C TotaL CReCjoveN Vo) Cepr Drum BRpkes- Temperature TSm^e Cf) : Samples Collected Sump Surfaces \ fhrhome: TotaL : Lihine Uleab". TA * Tola! (RetoverUf*) Clutch' Samples Collected T T>tFF Figure G--12 - Test Summary Sheet G-14 HWBUI0002599 flsOBsTos Emissions SummARV SHEET. Vehicle, Test No. Test Schedule : (Note: Debris data are repotted as Debris generated per step) Clutch- Debris', Asbeste s\ tty % Right Front brake: Sump Debris: Asbestos: Surfaces : Debris'. Asbestos.: AiRBorne: VebriS: Asbestos\ Right Rear. Brake: Left Rear BeaRe SumP debris: fisbestos. Surfaces: debris; Asbestos: E/rBoms Debris: Asbestos: Figure G-13 - Asbestos Emissions Summary Sheet it: G-15 HWBUI0002600 APPENDIX H ASBESTOS EMISSIONS ANALYSIS REPORTS FROM BATTELLE COLUMBUS LABORATORIES AND JOHNS-MANVILLE RESEARCH AND ENGINEERING CENTER The following information has been excerpted from the final summary report "Asbestos Emissions Analysis" by Battelle Columbus Laboratories (C. W. Melton and W. M. Henry) to the Environmental Protection Agency (Contract No. 68-01-0416). Dr. Joseph H. Somers was the Project Techni cal Officer who supplied this information to Bendix Research Laboratories. The objective of the project was to analyze 24 clutch and brake lining emissions for asbestos content. The samples were generated, col lected, and submitted for analysis by the Bendix Research Laboratories. The analytical method used for asbestos in brake and clutch lining emissions was a modification of one developed for the analysis of asbes tos in collections of airborne particulate. In general, results indica ted that sump samples contained, by far, more asbestos than airborne or surface samples. Experimental Procedure and Calculations: The analytical procedure employed was as follows: (1) An aliquot of the brake or clutch debris was weighed. (2) The aliquot was put into a centrifuge tube and was low-temperature ashed for three days. (3) The ash residue was suspended in 10 ml of an aqueous solution to which 3 drops of 1.0 percent Aerosol OT was added and then was treated ultrasonically to separate the asbestos fibers into colloi dal fibrils. (4) The extraneous material was separated by centrifugation from the asbestos which remained in colloidal suspension in the supernatant liquid. (5) The supernatant liquid was filtered through a Millipore filter (HAWP 025, HA 0.45 p, 25 mm) to deposit the asbestos fibrils uni formly over the surface of the filter. (6) A carbon film was vapor-deposited over the asbestos fibrils on the filter and a 3-mm x 3-mm square was cut from the resulting carbon-coated specimen. (7) The filter was dissolved in acetone to leave the carbon film bear ing the asbestos fibrils. (8) The carbon film was picked up on a 200-mesh electron microscope specimen support grid. mM 'll ;?; i i:;5 jvi;-:':- H-l HWBUI0002601 (9) (10) (11) The support grid and carbon film were loaded into the transmis sion electron microscope and the fibrils within several (usually ten) grid openings were counted and the results recorded as num ber of fibrils per grid opening. The number of fibrils per grid opening was translated into micro grams of asbestos in the aliquot taken by using a standard curve which was generated from known standard samples. Percentage asbestos in the sample was calculated by dividing the weight of the detected asbestos by the weight of the aliquot. H-2 HWBUI0002602 Experimental Results Results of the analyses of the 24 samples submitted by Beridix are as follows: BATTELLE SAMPLE NO 29897-1-1 29897-1-2 29897-1-3 29897-1-4 29897-1-5 29897-1-6 29897-8-1 29897-8-2 BENDIX SAMPLE DESCRIPTION Vehicle Test 1 Detroit Traffic Schedule Right Front Brake Airborne Sample Vehicle Test 1 Detroit Traffic Schedule Right Rear Brake Airborne Sample Vehicle Test 1 Detroit Traffic Schedule Right Front Brake Sump Sample Vehicle Test 1 Detroit Traffic Schedule Clutch Airborne Sample Vehicle Test 1 Detroit Traffic Schedule Right Rear Brake Surfaces Sample Vehicle Test 1 Detroit Traffic Schedule Right Rear Brake Sump Sample Vehicle Test 1 10-Stop Fade Right Front Brake Airborne Sample Vehicle Test 1 10-Stop Fade Right Front Brake Sump Sample PERCENT ASBESTOS IN SAMPLE 0.00055 0.00057 0.207 0.0068 0.0061 0.025 0.031 1.22 H-3 HWBUI0002603 BATTELLE SAMPLE NO. 29897-8-3 29897-8-4 29897-8-5 29897-8-6 29897-8-7 29897-8-8 29897-14-1 29897-14-2 29897-14-3 BENDIX SAMPLE DESCRIPTION Vehicle Test 1 10-Stop Fade Right Rear Brake Airborne Sample Vehicle Test 1 10-Stop Fade Right Rear Brake Sump Sample Vehicle Test 2 Burnish Schedule Right Front Brake Sump Sample Vehicle Test 2 Burnish Schedule Right Rear Brake Sump Sample Vehicle Test 2 Burnish Schedule Right Front Brake Airborne Sample Vehicle Test 2 Burnish Schedule Right Rear Brake Surfaces Sample Vehicle Test 1 Detroit Traffic Schedule Right Front Brake Surfaces Sample Vehicle Test 2 Detroit Traffic Schedule Right Front Brake Airborne Sample Vehicle Test 2 Detroit Traffic Schedule Right Front Brake Sump Sample PERCENT ASBESTOS IN SAMPLE 0.0074 0.216 0.275 0.103 0.100 0.376 0.011 0.0057 0.62 4 HWBUI0002604 BATTELLE SAMPLE NO 29897-14- 4 29897-14- 5 29897-14- 6 29897-14- 7 29897-14- 8 29897-14- 9 29897-14- 10 BENDIX SAMPLE DESCRIPTION Vehicle Test 2 Detroit Traffic Schedule Right Rear Brake Airborne Sample Vfehicle Test 2 Detroit Traffic Schedule Right Rear Brake Sump Surface Vehicle Test 2 Detroit Traffic Schedule Right Rear Brake Surfaces Sample Vehicle Test 3 Detroit Traffic Schedule Right Front Brake Airborne Sample Vehicle Test 3 Detroit Traffic Schedule Right Front Brake Sump Sample Vehicle Test 3 Detroit Traffic Schedule Right Rear Brake Airborne Sample Vehicle Test 3 Detroit Traffic Schedule Right Rear Brake Sump Sample PERCENT ASBESTOS IN SAMPLE 0.015 0.057 0.061 0.016 0.135 0.044 0.0104 H-5 HWBUI0002605 &0&01 jonns-Manvtne Research & Engineering Center P.O.Box 159 Manville, N. J. 08835 (201) 722-9000 September 6, 1972 Dr. M. G. Jacko Materials and Processes Department Bendix Research Laboratories Southfield, MI 48075 Dear Dr. Jacko: We have finally completed the analysis of the brake lining wear debris you sent to us. These samples were described as follows in your letter to Dr. Speil of July 5, 1972. Sample No. 1 - Vehicle Test 1 J-M No. 4123-58-1 Detroit Traffic Schedule Right Front Brake Surfaces Sample Sample No. 2 - Vehicle Test 1 J-M No. 4123-58-2 Detroit Traffic Schedule Right Rear Brake Surfaces Sample Sample No. 3 - Vehicle Test 1 J-M No. 4123-58-3 Detroit Traffic Schedule Right Rear Brake Airborne Sample The fiber content of the samples and the weight loss on ignition are given below. The fiber content was determined by the rub-out procedure which I believe we explained to you at the time of your visit to our laboratory. No. 1 Initial Sample Weight (mg) 50.78 Weight of Ash After Ignition @ 400C(mg) 43.50 Weight Loss on Ignition 14.4% No. 2 46.50 31.10 33.1% No. 3 45.70 30.20 33.9% Total Fiber in Sample (micrograms) Percent Fiber in Original Sample 2.54 0.005 0.98 0.002 2.73 0.006 We were unable to determine the element distribution because our emission spectrograph has been disassembled for our move to Denver. H-6 HWBUI0002606 Dr. M. G. Jacko September 6, 1972 Page 2 If you have any questions regarding these samples or pro cedures you can contact Dr. Speil. He will be available through the month of September. After that time you can contact me in Denver. Our mailing address and telephone number in Denver will be: Johns-Manvilie Corporation P. 0. Box 5108 Denver, Colorado 80217 Phone: (303) 770-1000 Very truly yours J. P. Leineweber mp cc: Dr. S. Speil H-7 HWBUI0002607