Скачать презентацию Engineering Excellence Hardware Fundamentals Workshop Robust Design Prepared Скачать презентацию Engineering Excellence Hardware Fundamentals Workshop Robust Design Prepared

afd00f24507cf399a88f060ed8dc627d.ppt

  • Количество слайдов: 38

Engineering Excellence Hardware Fundamentals Workshop Robust Design Prepared By: The Engineering Excellence Institute Xerox Engineering Excellence Hardware Fundamentals Workshop Robust Design Prepared By: The Engineering Excellence Institute Xerox Corporation, Webster, New York 14580 Copyright 1996 Xerox Corporation. All rights July 1996 Revision #002

Objectives 2 At the end of this module you will be able to: • Objectives 2 At the end of this module you will be able to: • Describe the basic concepts of Robust Design • Describe the basic elements of Robust Design • State how Robust Design impacts the entire design process to improve TTM • Relate Robust Design to the Design Quality Process • List the process steps to achieving Robust Design • Recognize the Robust Design Quality Enablers and inspect for this robustness EEFW - Robust Design Revision #002

Major Connections To Time To Market 3. 1 Market & Product Strategy Vision 3. Major Connections To Time To Market 3. 1 Market & Product Strategy Vision 3. 2 Define Product Platform & Technology Define Product & Deliver Technology Readiness 3. 3 3. 4 Design Product 3. 5 Demonstrate Product Technology/Design Validation 3 3. 6 Deliver Product Delight Customers Product Launch Readiness The “Bottle Model”. . Design Latitude Variance ä A larger S/N ratio results in a robust design with the latitude to accommodate ‘Noises’ As we progress towards Robust Design crossover occurs sooner and the length of the Design and Demonstration Phases decreases EEFW - Robust Design Revision #002

Major Connections To Time To Market 4 The following considerations require Robustness of Function Major Connections To Time To Market 4 The following considerations require Robustness of Function which can be achieved by developing actual functions close to ideal, i. e. to close the gap between actual function and ideal function by applying • robust design Readiness Technology process. – Technology Readiness is confirmation that the hardware/software configuration supports the program’s overall objectives from a technical perspective • Reusability – Ability to handle a variety of signals and new or changing customer requirements • Reproducibility – Making technology insensitive to all sources of variation – Selecting good tuning factors for adjustment – Assuring that best conditions in labs are best for downstream customers. Revision #002 EEFW - Robust Design

Purpose of Robust Design • Develop low cost product designs, mfg process designs, measurement Purpose of Robust Design • Develop low cost product designs, mfg process designs, measurement system designs with consistent functional performance under a wide range of usage conditions through systems intended life. • Improve product quality within constraints of cost and time, through more efficient, customer focused, R&D activities. • 5 Achieve desired customer performance through optimization of designs, rapid commercialization of technologies, and reduced quality loss after shipment EEFW - Robust Design Revision #002

Product Quality Model: Quality achieved when VOC is met, and the enablers are process Product Quality Model: Quality achieved when VOC is met, and the enablers are process capable. MPSV 3. 1 3. 3 Design 3. 2 Define 3. 4 Demonstrate 3. 5 Deliver 3. 6 Voice of the Customer (VOC) Quality Function Deployment (QFD) Multiple Houses Market Analysis 1 Focus Interviews 2 1 a Customer Feedback 2 a Pre planning matrix 1 VOC 1 a Customer Importance Rating 2 Customer Competitive Evaluation 2 a XC Strategy and Quality Plan House 1 System Performance House 2 Spec Module Level Devlp. Critical Parameter House 3 Devlp. Subsystem Level Critical Parameter Devlp. House 4 Component Level Critical Specifications Devlp. House 5 Manufacturing Process Parameter Devlp. Production and Quality Control Planning Matrix Integrated System Design Reusability Technology, Modules, Manufacturing Process and Parts Critical Parameter Management Implementation Development • Data and Demonstration Optimum Concept Selected - Best of Breed Will Meet VOC at Lowest Cost Tolerance Design Phase Selection Process Qualification of Critical Specifications and enablers Optimum Nominal Value and Tolerance of Critical Parameters/Critical Specs Determined Parameter Design Phase Technology Readiness Critical Parameter Process Capability Verification On Line Quality Control Robust Design Concept Design • Ready Trade-offs - (Exceptions) Performance vs Cost Validation Quality: Design Quality • Planning / Customizing Manufacturing Quality • Plans/Strategy • Key Quality Indicators • Quality Initial Design Emphasis • Problem Discovery Process • Key Quality Indicators • Manufacturing Quality Plan • Design/Manufacturing Readiness • Status vs Plan • Verification, Plan customer satisfaction monitoring • Product Verification Test (PVT) Implementation • Supplier Process Certification • Manufacturing Process Robustness • Supplier Process Qualification • Factory and Plan Implementation • Customer Survey Software Process Management and Software Capability Maturity BAB 6/6/96

Definitions 7 Robustness: The ability of a product or process to function close to Definitions 7 Robustness: The ability of a product or process to function close to ideal customer satisfaction under actual conditions of use. A product or process is said to be robust when it is insensitive to the effects of sources of variability, even though the sources themselves have not been eliminated. Robust Design: A systematic engineering based methodology (which is part of a quality engineering process) that develops and manufactures high reliability products at low cost with reduced delivery cycle. The goal of Robust Design is to reduce cost and quality loss. Concept Design, Parameter Design, tolerance design and on-line QC are the 4 successive stages. Parameter design, most widely practiced, uses a two step optimization process --Maximizing S/N ratio and then Tuning to Target. Ideal Function: A mathematical relationship between input and output, which can be used for function or energy transfer optimization. Ideal function of resistor is Ohm’s law, to convert current to voltage. Revision #002 EEFW - Robust Design

Definitions 8 Control Factor: Factors affecting the function whose levels can be specified freely Definitions 8 Control Factor: Factors affecting the function whose levels can be specified freely by engineer / designer. Their settings are used to amplify sensitivity to signal, dampen sensitivity to noise plus tuning to target. Noise Factor: Factors affecting the function which cannot be controlled by the engineer/designer. Factors whose settings are difficult or expensive to control are also called noise factors. There are three noise classifications: 1) Internal 2) External 3) between product noises Response factor: Response factors are often called the response variables, measurands, quality characteristics, outputs. These are also selected by engineers to pick up information about control, noise, and signal factor effects. Some desirable properties for responses include: - fundamental (related to basic energy transfer mechanism of input/output relationship) - continuous, quantitative - monotonic with changes in control factor levels , unambiguous. - should be complete (cover important dimensions of the function) - valid, independent of imposed specifications - economical, timely Signal Factor: Factor whose levels carry the information to easily change output response, usually related to input power. It is based on the physics or engineering of the system. Revision #002 EEFW - Robust Design

Ideal Quality 9 Each customer expects that every product will deliver the target performance Ideal Quality 9 Each customer expects that every product will deliver the target performance each time the product is used, under all intended operating conditions, and throughout its intended life, with no harmful side effects. A B C Zero Defects CPk S/N Ratio Achieve Target Values “When a product’s performance deviates from the target performance, its q is considered inferior. Such deviations in performance cause losses to the of the product, and in varying degrees, to the rest of society. ” G. Taguchi (1992) EEFW - Robust Design Revision #002

Traditional Quality vs. Robust Quality Paradigm Shift Continuous Quality loss away from target Traditional Traditional Quality vs. Robust Quality Paradigm Shift Continuous Quality loss away from target Traditional Quality Good/No Good Loss $ Good No Loss No Good Loss $ No Good target Poor Good Functional Limits EEFW - Robust Design Poor Fair target 10 Good Best Functional Limits Revision #002

Summary / The Robustness Paradigm 11 • Robustness = Problem Prevention • Any deviation Summary / The Robustness Paradigm 11 • Robustness = Problem Prevention • Any deviation from target incurs a quality loss • Don’t need to always control / eliminate the root cause to improve design • High performance does not always require high cost • Robust Design process is to improve efficiency of R&D • Engineering focus using mathematical and statistical concepts EEFW - Robust Design Revision #002

Proactive Quality Indicators of Robustness Quality Indicator S / N Ratio Reliable Measures of Proactive Quality Indicators of Robustness Quality Indicator S / N Ratio Reliable Measures of Functional Limits / Latitude Important Noise Factors Range of Operation Power Consumption Benchmark Comparison Test Gap Design Complexity Tuning Factors Critical Parameter Nominal Values Cost Quality Loss System Integration Problems Understanding of Design Data Supporting Decisions EEFW - Robust Design 12 Desired State Gain Over Time Identified Expanded / Increased Identified Expanded Reduced Identified No Increase Improved Fewer Improved Unambiguous Revision #002

Definitions 13 Signal To Noise Ratio: • All engineering functions are transformation of energy Definitions 13 Signal To Noise Ratio: • All engineering functions are transformation of energy from one form to another, one place to another, one time to another • Variations in energy transfer cause functional variation • Therefore, Maximize Signal to Noise: Power of signal to create functional output Power of noise to create dysfunctional output Improve the function Useful Part of input while simultaneously S / N = Harmful Part of Input power reducing the Power dysfunction In Laymans’ terms What you want S / N =What you don’t want S/N= } EEFW - Robust Design Revision #002

Robust Design as Part of Concurrent Engineering Input Process Steps Design Concept Manufacturing Process Robust Design as Part of Concurrent Engineering Input Process Steps Design Concept Manufacturing Process Concept Parameter Design Mfg. Process Parameter Optimization Process System Verification Mfg. Process Verification Test Process Tolerance Design Process Manufacturing Tolerance Design Process Production / Field Readiness Test Process Robust Product Design 14 On-Line Quality Control Process Output Robust Mfg. Process Design Robust Product Process Capable EEFW - Robust Design Revision #002

Process Stages of Robust Design Concept Design Parameter Design Tolerance Design 1 In this Process Stages of Robust Design Concept Design Parameter Design Tolerance Design 1 In this innovation stage of the design process, the system and subsystem engineer examines a variety of architectures and technologies for achieving a desired function for a planned product and selects the most suitable one(s). 1 Once the concept is selected, the design engineer determines the best nominal values and tolerances for each of the critical parameters of the system / subsystem that will produce consistent output using low cost components and tolerances. 1 If at the end of the parameter design stage the output is not at or above benchmark , a third stage is introduced to improve quality by selectively adding cost by upgrading components and/or tightening tolerances, maintaining cost and quality balance. On-Line QC Design EEFW - Robust Design 15 1 Cost effective maintenance of quality Revision #002

From Problem Solving to Problem Prevention Problem Solving 16 Problem Prevention System Design Concept From Problem Solving to Problem Prevention Problem Solving 16 Problem Prevention System Design Concept Prototype Hardware Redesign/Fix Parameter Design Experimental Hardware/Modeling Subsystem Verification Tests Debug/PIT/System/ Characterization Tests No Build, Test, Iterate No Meets Product Specifications Robust? Subsystem Technology Development Yes Tolerance Design Intent Hardware SVT Release Design For Production EEFW - Robust Design Yes Cost Reduction N Verify Product Specs? Y Release Design for Production Revision #002

Process Stages: Tools, Outputs, and Measures Concept Design Parameter Design Tools Measures QFD Benchmarking Process Stages: Tools, Outputs, and Measures Concept Design Parameter Design Tools Measures QFD Benchmarking Noise Ratio Pugh/Combinex FMEA, DFA/DFM VA/VE function Modeling 17 Output House I Benchmark Defined Voice of Customer Signal to Selection Matrix Concept Chosen Failure modes Reliability Estimates FAST diagram (projected CP’s) Cost per Math. Model Validity of Model Critical Parameter Nom. Tolerance Design On-Line QC Design EEFW - Robust Design Noise Ratio Taguchi Methods Quality Loss Signal to and Ranges Defined, Technology Verified Sensitivity Cost & Performance Cost & Trade-offs Complete Revision #002

Robust Design as a Part of Critical Parameter Management 18 Critical Parameter Development Activity Robust Design as a Part of Critical Parameter Management 18 Critical Parameter Development Activity 2. 1: 1. 0 Concept Design On-Line Quality Control Critical Parameter Development Activity 6. 1: } Conduct Experiments to Define CP Range ( tolerance ) Tolerance Design / Analysis 1 5 4 3 Reset & Verify Nominal Set Points } Conduct Experiments / Analysis to Verify Performance at CP Nominal 2 6 System Robustness Demo Critical Parameter Development Activity 5. 2: Fixture & Measurement System Development Parameter Design Plans } Conduct / Analyze Experiments Design Hardware and Optimization Experiments } Critical Parameter Development Activity 5. 2: Conduct Experiments to Optimize CP Nominals and Verify all CP’s Identified Refer to slide 34 for detailed inspection list for each step EEFW - Robust Design Revision #002

Steps 1 & 2: Parameter Design Methodology Step 1: 6 5 4 Fixture & Steps 1 & 2: Parameter Design Methodology Step 1: 6 5 4 Fixture & Measurement System Development Step 2: Design 3 3 Parameter Plans CONTROL FACTORS Design aspects you control SIGNAL FACTORS Inputs 19 Primary Responses functional outputs Signal/Noise Ratio & Sensitivity Expected Quality Loss Benchmarking Failure Mode Counter Measure Main Function NOISE FACTORS Design aspects you can't control Secondary Responses dysfunctional outputs Failure Mode EEFW - Robust Design Revision #002

Steps 1 & 2: Design Decomposition & Flowchart: Dispensing Replenisher and Development Subsystem Step Steps 1 & 2: Design Decomposition & Flowchart: Dispensing Replenisher and Development Subsystem Step 1: 4 5 3 S / N ~ Beta 2 / Sigma 2 3 4 Signal Factors Primary Input Replenisher Dispensing Control Factors Trickle Charging Signal Primary Response * Develop Ideal Function for each subfunction Signal Factors Developer Mixing Signal Factors Primary Response Magnetic Roll Response Loading Noise Factors Signal Factors Primary Input Primary Response 6 5 Fixture & Measurement System Development Step 2: Parameter Design Plans 20 Signal Factors Primary Donor Roll Response Cloud Generation Response Photoreceptor Loading (Jumping) Primary Response Development rgf-4/96 EEFW - Robust Design Revision #002

Steps 1 & 2: Example for Toner Dispenser 4 4 5 Step 2: Design Steps 1 & 2: Example for Toner Dispenser 4 4 5 Step 2: Design 3 4 Parameter Plans 3 Signal Factor Auger Speed Function: Replenisher Dispensing Inputs P/C calls for toner (CAP) (BOTTLE) (AUGERS) LOWER-UPPER Spiral Pitch(1) Spiral Depth(2) Bottle speed ((9) # of fins (3) fin type(4) fin length (10) pitch(5) diameter (6) EEFW - Robust Design Signal Factor Carrier added Carrier Added (gms ) Auger RPM Primary Response Ideal Function: Convert Dispense Rate (Gm/Min) auger speed to dispense rate Control Factors S/ N ~ Beta 2 / Sigma 2 S / N ~ Beta 2 / Sigma 2 Trickle (gms) 6 5 Fixture & Measurement System Development Dispense GPM Step 1: 21 Main Function: Trickle Charging of Developer Ideal Function: Trickle out equals carrier added Primary Response Sump Mass & Charge Noise Factors toner flow (1) 4 -26 HF# tilt angle (2) + - 2 deg. dispenser (3) tolerances #1 vs. #2 Control Factors overflow location (1) overflow height (2) Mix Auger Type (3) Noise Factors Developer dispense rate* 2 -10 GPM (1) Dev. Housing tilt angle** + / - 2 Deg. (2) Toner Concentration Mix Auger 1 -4 % (3)*** Speed (4 ) * Replenisher dispense rate for 6%A/C at 40 PPM to 25%A/C at 65 PPM with a 3: 1 Ratio of toner to carrier by weight. Revision #002

Steps 1 & 2: Process Chart (P-Chart) 6 5 4 5 Step 1: Fixture Steps 1 & 2: Process Chart (P-Chart) 6 5 4 5 Step 1: Fixture & Measurement System Development Step 2: Parameter Design Plans 3 4 22 3 Levels 2 Noise Factor # Control Factor Units 1 Spiral Depth mm 4 8 2 Spiral Pitch mm 30 40 50 3 # of Fins 4 5 4 Fin Type type perp angle Auger Dia mm 9 10. 5 12 6 Auger Pitch mm 8 10. 5 13 7 Auger Opening mm sm med large 8 Bottle Speed rpm 4. 82 9. 75 14. 5 Level scoop 5 Units 6 Signal Factor Auger Speed (rpm) # 1 42. 5 1 Level 2 81. 5 3 1 2 Material Flow Rate & Angle 3 120 gpm deg 4. 0 -2 22. 0 +2 Dispenser # # 1 2 Ideal Function: Convert auger RPM to dispense rate Response Factor Dispense Rate (g/m)* Signal 1 y 1 , y 2 , y 3 , y 4 Signal 2 y 5 , y 6 , y 7 , y 8 Signal 3 y 9 , y 10, y 11 , y 12 * 3 signal levels * 4 noise levels =12 data per test EEFW - Robust Design Revision #002

Step 3: Optimization Plots 23 1 2 4 Step 3 : Conduct / Analyze Step 3: Optimization Plots 23 1 2 4 Step 3 : Conduct / Analyze Experiments g/m Dispensed 6 5 Dispense Rate vs. Auger Speed: 40 Data Points @ 4 Noise Levels Two Step Optimization ¶ Maximize S/N ratio · Tune to target beta 42. 5 EEFW - Robust Design Beta = Grams / Revolution = Mechanical Efficiency 81. 5 Auger Speed (rpm) 120 Revision #002

Step 3: Control Factor Classification 1 Affects Signal / Noise 2 No A B Step 3: Control Factor Classification 1 Affects Signal / Noise 2 No A B A C b Yes Step 3 : Conduct / Analyze Experiments No 4 Affects Response 6 5 24 EEFW - Robust Design Revision #002

Step 3: Factorial Effects Plot 25 1 6 5 4 2 S/N Ratio Step Step 3: Factorial Effects Plot 25 1 6 5 4 2 S/N Ratio Step 3 : Conduct / Analyze Experiments (Type A) Tuning Factor (Type B) (Type C) 2 3 1 2 3 1 2 3 Sensitivity 1 (Type A) Fin Type EEFW - Robust Design Auger Diameter 1 2 3 Auger Pitch Auger Opening Bottle Speed Revision #002

Steps 4 & 5: Subsystem & System Verification 26 1 2 6 3 Step Steps 4 & 5: Subsystem & System Verification 26 1 2 6 3 Step 5: System Latitude Step 4: / Reset & Verify Robustness Nominal Set Demo Points Predicted Actual Previous Condition S/N Ratio -21. 40 Sensitivity 0. 126 0. 14 0. 13 1. 48 1. 61 2. 41 Sigma -21. 21 -25. 36 Gain=4. 15 d. B System Verification : • After verification with both prototype and production intent subsystem hardware • Toner Dispenser integrated with total developer system for verification of developer requirements • Total Developer system integrated in Systems Verification Test for analysis of Image Quality Responses EEFW - Robust Design Revision #002

Step 6: Tolerance Design / Analysis Step 6: Tolerance Design/Analysis 27 1 2 5 Step 6: Tolerance Design / Analysis Step 6: Tolerance Design/Analysis 27 1 2 5 4 3 Tolerance Design • Trade-offs are made between reduction in quality loss due to performance variation and increase in manufacturing cost (selective reduction of tolerances, selective specification of higher grade material/components) • Performed only after signal to noise is maximized in parameter design • Sensitivity analysis and economic considerations are used to select the correct tolerances for drawings EEFW - Robust Design Revision #002

Key Messages 28 • The Robustness of Function ( Dynamic Signal to Noise ) Key Messages 28 • The Robustness of Function ( Dynamic Signal to Noise ) is used to achieve Technology Readiness. • Dynamic Signal to Noise will enable low cost Reusable and Reproducible Technologies for both design and manufacturing. • Ideal Quality equals Target Performance even in the presence of noise, throughout design life. • Robust Design is the process of developing and improving the design latitude. • Parameter Design minimizes the sensitivity to noises without eliminating the causes of the noise. • Signal to Noise is a proactive improvement metric. • Robust Design enables the prevention of problems. Problem solving minimized. EEFW - Robust Design Revision #002

Putting It All Together Key Quality Indicators Launch Quality 29 Performance Growth Curve Impact Putting It All Together Key Quality Indicators Launch Quality 29 Performance Growth Curve Impact Current Goal 1. Start with better performance • Parameter Design w/Verification • Dynamic S/N • Target Performance 2. Steepen the slope of the growth curve (Problem Prevention) Time To Market TTM Goal TTM Bench- Current mark Engineering Excellence Practices and Methods, in support of the Time-To-Market Process, enable establishing new benchmarks in schedule and/or quality. EEFW - Robust Design Revision #002

Robust Design – Selected Bibliography 30 Books Camp, Robert. Benchmarking: The Search for Industry Robust Design – Selected Bibliography 30 Books Camp, Robert. Benchmarking: The Search for Industry Best Practices that Lead to Superior Performance. Milwaukee, Wis. : ASQC Quality Press, 1987 Clausing, D. Total Quality Development, ASME, 1993 Phadke, Madhar. Quality Engineering Using Robust Design. Englewood Cliffs, NJ: Prentiss Hall, 1989 Mori, Teruo. Taguchi Techniques for Image and Pattern Developing Technology Mori, Teruo. The New Experimental Design, Taguchi’s approach to Quality Engineering Stein, P. Measurement System Engineering Suh, N. Principles of Design Taguchi, Genichi. System of Experimental Design: Engineering Methods to Optimize Quality and Minimize Costs. Dearborn, Mi: Unipub Kraus International Publications, American Suppliers Institute, 1987. Taguchi, Genichi. Quality Engineering Series Volumes 1 - 7, Japanese Standards Association (JSA), 1994. Taguchi. Robust Technology Development, ASME Press, 1993. Taguchi, Elsayed, Hsiang. Quality Engineering in Production Systems, 1989. Articles ASI Symposium Proceedings, ITT Symposium Proceedings, Xerox Symposium Proceedings Taguchi Center at Xerox Corporation: Taguchi Quality Engineering System for Robust Design. (MIT Videotape Series) EEFW -Engineering Robust Design Excellence Institute (EEI) Revision #002

Robust Design / Reference 31 Reference Slides EEFW - Robust Design Revision #002 Robust Design / Reference 31 Reference Slides EEFW - Robust Design Revision #002

Parameter Design Checklist (refer to slide 19) Step 1: Fixture & Measurement System Development Parameter Design Checklist (refer to slide 19) Step 1: Fixture & Measurement System Development Step 4: þ Flexible Designs with Adjustment þ Lowest Cost Tolerances Applied þ Measurement System Capability þ Decompose into Subsystems Step 2: þParameter Design Plans Determine benchmark system Main Function þ Identify 32 Reset & Verify Nominal Set Points þ Noise conditions repeated or stresses increased to validate previous results þ Critical parameter nominal values/latitudes verified þ Tuning factors verified for system integration test/model verification Step 5: System Latitude / Robustness þ Early Detection of Potential Downstream Demo þ Comparison to benchmark completed Problems þ System latitude demonstrated and/or þ Identify Side Effects and Failure Modes (FMEA) shortfalls identified þ Identify Noise Factors for Testing þ System integration tradeoffs identified þ Identify the Response(s) and Function that is to be Optimized þ System power reduction captured þ List Control Factors and Levels þ Identify Appropriate S / N Ratio and Ideal Function þ Run Current Case & Determine Benchmark Gap þ Estimate Experiment Time and Cost Step 6: Tolerance Design/Analysis Step 3 : Conduct / Analyze Experiments þ Cost/Quality Analysis to close gaps in þ Conduct peer reviews during experiments Have checkpoints and backups ready þ Identify response tuning factors and model sensitivities for simulation þ Summarize optimum parameter levels and tradeoffs where necessary þ Identify S/N ratio improvement þ Document lessons learned EEFW - Robust Design þ þ þ identified shortfalls conducted Confirmation experiment conducted Process optimization confirmed Cpk / latitude ratio tracking formalized On-line Q. C. feedback/forward process elements identified Revision #002

HJD Robustness LRB level @ April ‘ 96 Sheet #1 Signal Factor Auger Speed HJD Robustness LRB level @ April ‘ 96 Sheet #1 Signal Factor Auger Speed B C Primary input P/C calls for toner Function: Replenisher Dispensing Ideal Function: Dispense toner consistent and predictable rate. (CAP) (BOTTLE) (AUGERS) LOWER-UPPER Control Factors Spiral Pitch(1) Spiral Depth(2) Bottle speed ((9) # of fins (3) fin type(4) fin length (10) pitch(5) diameter (6) Control Factor Best Nominal Values C/F# LEVEL TYPE 1) 40 MM (A) 6) 9 MM (A) 2) 4 MM (A) 7) 12 MM (A) 3) 4 (A) 8) 9 MM (A) 4) Straight (C) 9) 5 RPM (A) 5) 10. 5 MM (B) 10) 40 MM (A) Trickle (gms) A A S / N ~ Beta 2/ Sigma 2 Dispense GPM Response No Yes Qualification Method Used Signal/ Noise Yes No Updates 4/96 J. Lioy S / N ~ Beta 2/ Sigma 2 Signal Factor Carrier added Main Function: Trickle Charging of Developer Auger RPM Primary Response Dispense Rate (Gm/Min) Noise Factors Ideal Function: Trickle out equals carrier added Control Factors toner flow (1) 4 -26 HF# overflow height (2) dispenser (3) tolerances #1 vs. #2 Mix Auger Type (3) Noise Analysis #1 - 18 GPM Max. #2 & #3 < 1 GPM Additional Analysis Bottle seal - Vacuum required added valve Mix Auger Speed (4 ) Primary Response Sump Mass & Charge Noise Factors overflow location (1) tilt angle (2) + - 2 deg. Carrier Added (gms ) Developer dispense rate* 2 -10 GPM (1) Dev. Housing tilt angle** + / - 2 Deg. (2) Toner Concentration 1 -4 % (3)*** * Replenished dispense rate for 6%A/C at 40 PPM to 25%A/C at 65 PPM with a 3: 1 Ratio of toner to carrier by weight. C/F Best Nom. Values C/F# LEVEL TYPE. 1) outboard (A) 2) 14 MM (B) 3)A 18 MM Psed. (A) 4) 400 RPM (A) ** Aug to wall< 2. 5 MM Noise Analysis Additional Analysis ** Added Auger to wall dist. (C/F ) *** Toner flow replaced T. C. noise 3)A Change from 12 H to 18 P mixing rgf-4/96 -hjdnid 2

Qualification Method Used Signal Factor Toner Gms. Added A A B C Primary Input Qualification Method Used Signal Factor Toner Gms. Added A A B C Primary Input Main Function: Developer Mixing (Add and Cross) Sump Mass level Ideal Function: Grams added Mixed and Charged Control Factors AUGERS Reload Mixing Signal Factor Trim Bar Gap S / N ~Beta 2 Sigma 2 Toner in dev Response No Yes Signal/Noise Yes No Update 4/96 J. Lioy Type (1) Main Function: Mag Roll Loading Toner gms. added Primary Response Mass @ T. C. % & Q/M Noise Factors Dev. Tilt Angle + / - 2 Deg. (1) Cutout (2) Speed (7) Toner add rate 3 -12 GPM (2) Type-Hel. 12 Speed (6) Toner flow rate 4 -12 HF# (3) Sump Mass (3) Dev. Asub T Mag Roll Spd. (4) <60 to >100 (4) Trim Gap (5) Noise Analysis Toner Entry Pt. (8) Tilt angle #1*** Toner Sensing Type Toner flow #2 (9) Control Factor Best Nominal Values Toner Disp. Rate #3 C/F# LEVEL TYPE Additional Analysis 1) 18 MM -Ps. (A) 6) 500 rpm (A) * #8 to Normal 2) with (A) 7) 400 rpm (A) 9)B Removed tc sensor 3) 900 gms (A) 8) Front * (A) *** Auger to wall A 4) 50 IPS (A) 9)B Long Snout (A) 4)A F [email protected] 40, F 2 @56 IPS Tuning for power &PQ 5) 0. 075” (A) Ideal Function: Mass on Roll uniform &controlled Control Factors Primary Response Dev. Mass on roll Dev. Tilt Angle + / - 2 Deg. (1) Mag/Donor spacing (2) Mag roll spd. (7) Mag field strn. (8) Trim Bar Gap Noise Factors Auger spd (1) Sump Mass (3) Mag angle (4) TBG (5) Donor Roll Spd. (6) S / N ~Beta 2 Sigma 2 Mass on Roll HJD Robustness LRB level @ April ‘ 96 Sheet #2 Toner flow rate 4 -12 HF# (2) Toner Conc. % 1 -4 % (3) Noise Analysis Position #1 noise Others minimal Control Factor Best Nominal Values Position on Roll in. / out. / ctr (4) 2 reads per position Additional Analysis 2)A - 0. 05” Rollback BC/F# LEVEL TYPE 3) 900 gms Mixing c 1) 500 rpm (C) 5) 0. 075” (B) 4) -0 deg. Auger Mks D 2)A 0. 045” (C) 6)D 15 IPS (A) 6) - F [email protected] 13. 6, F [email protected] 3)B 1000 gms (B) 7)F 50 IPS (A) 19. 2 IPS for PQ tune. 4)c +10 deg. (A) 8) Norm Lvl. (C) 7) -F [email protected] 40, F 2 @56 rgf-4/96 - hjdnid 2 IPSTune Power& PQ

HJD Robustness LRB level @ April ‘ 96 Sheet #3 Signal Factors Vdm D. HJD Robustness LRB level @ April ‘ 96 Sheet #3 Signal Factors Vdm D. C. Bias B C Primary Input Mag Roll Mass Main Function: Donor Roll Loading Donor to Mag. Primary Response Ideal Function: Constant Mass, Charge, and Size Control Factors Cycle number 1 st - 10 th - (1) AC freq. (2) Donor Roll Time cons. (3) Mass on roll (8) Donor roll toner Mass, Charge , Size Ideal Function: Cloud of uniform mass and charge Noise Factors AC P-P (1) Mag ang. (4) Donor spd (5) Mag. field (6) Donor /mag spacing (7) Main Function: Cloud Generation Vdm Bias (volts) Dev. Conduct 108 - 1011 - (2) Toner Flow 4 -12 HF # (3) Noise Analysis Cycle number -Large Control Factor Best Nominal Values C/F# LEVEL TYPE 1) 200 V (C) 5)c 14 IPS (A) A 2) 500 Hz (A) 6) Normal (A) 3 B < 2000 u. S ( C) 7) 0. 04”-0. 06” (B) 4) 0 - 5 deg (B) 8) 20+ (A) Control Factors V jump A. C. Primary Response Toner Cloud in Gap Noise Factors AC P-P (5) Bias Voltages Donor to Mag. to P/R A A S / N ~Beta 2 / Sigma 2 Dev. Mass Response No Yes Signal/Noise Yes No Cloud Density Qualification Method Used Update 4/96 J. Lioy Signal Factors S / N ~Beta 2 / Sigma 2 Vjump A. C. - P-P Cyclic State AC freq. (6) AC P-P (7) AC freq. (8) Toner properties Environmental Toner properties Donor Roll Props. Additional Analysis 2)A- 3. 25 KHz same as main P/S frequency 3 B-Remains undefined for function as well as measurement changes. 5)c- F [email protected] 13. 6, F [email protected] 19. 2 IPS for PQ tune Control Factor Best Nominal Values Noise Analysis C/F# LEVEL TYPE 5) 3 KV (A) Additional Analysis 6) 3 Khz (A) 7) 200 V (A) 8) 9 Khz (A) All @ 0. 014” GAP to P/R rgf-4/96 - hjdnid 2

HJD Robustness LRB level @ April ‘ 96 Sheet #4 Signal Factors Needs Discussion HJD Robustness LRB level @ April ‘ 96 Sheet #4 Signal Factors Needs Discussion A A B C Primary Input Toner Cloud Signal Factors Vdev = Vpr- Vdonor S / N ~ Beta 2 / Sigma 2 P/R Mass Response No Yes Signal/Noise Yes No Response Qualification Methods Used Update 4/96 J. Lioy Main Function: Non Interactive Development Ideal Function: P/R Mass not reflected back to the other color donor rolls Control Factors Jumping Biases P/R Mass Developed Cyclic State Jumping Electrodes Environmental Image Metrics for Lines , solids , halftones Control Factors Noise Factors Donor to P/R Gap (1) Toner Cons. & Tribo (Asub. T) (1) Toner Flow 4 - 12 HT # (3) Toner Cohesion Number (4) Toner Residence Time (5) Print Number and Area Coverage Combined Noise Analysis Donor to Mag Gap (2) Donor Roll Spd. (3) Toner properties Donor Roll Props. Control Factor Best Nominal Values C/F# LEVEL TYPE Primary Response Ideal Function: Reproduce image to match input signal Primary Response Noise Factors Signal Main Function: Photoreceptor Development Signal Mag Roll Spd. (4) Cloud Generation Factors 5, 6, 7, 8 Residence time key noise Noise Analysis Control Factor Best Nominal Values C/F# LEVEL TYPE 1)A 0. 014” (A) 5)E 3 KV (A) B F 2) 0. 045” (A) 6) 3 KHz (A) Additional Analysis 3)C 16 IPS ( A) 7) 200 V (A) 4)D 50 IPS (A) 8)G 9 K (A) rgf-4/96 - hjdnid 1 Additional Analysis 1)A- 0. 012” -PQ & Power 2)B- 0. 045” Rollback 3)C-F 1 / F 2 Speed Tuning 4)D-F 1 / F 2 Speed Tuning 5)E - 2. 2 -2. 4 KV P/S arcing 6)F - 3. 25 KHz PQ Tuning 8)G - 3. 25 KHz P/S Cost

Robust Design / Reference 37 Developer Housing Critical Parameter Diagram Replenisher Trickle entry point Robust Design / Reference 37 Developer Housing Critical Parameter Diagram Replenisher Trickle entry point port Auger to wall clearance < 2. 00 MM Mixing Auger Speed @ 400 MATERIAL FLOW Pickup Auger Speed @ 500 Mag field strength and position Mag Roll Donor roll time constant P/R DIRECTION IS UPWARD OUT OF THE PAGE Vm AC Donor Roll Vdm Photoreceptor OTHER DESIGN CONTROL PARAMETERS MAG ROLL TO DONOR ROLL SPEED RATIO - 3: 1 (WITH) Vd AC Vexp Vdb GROUND DONOR ROLL TO P/R SPEED RATIO - 1. 6 : 1 (AGAINST) Vdm ADJUSTABLE FOR DEVELOPMENT CONTROL (100+VLTS) S. Mordenga / R Faull April 96 EEFW - Robust Design AC JUMPING FIELD < 4. 5 VOLTS / MICRON AT 3. 25 KHZ. Revision #002

Robust Design / Reference 38 Critical Parameter Drawing for F 2/F 2 Replenisher Dispenser Robust Design / Reference 38 Critical Parameter Drawing for F 2/F 2 Replenisher Dispenser PADDLE WHEEL SECTION SUMP BOTTLE SPEED SET @ 5 RPM UPPER AUGER SPEED @120 RPM BOTTLE CRADLE FRICTIONAL FORCE AGITATOR BAR SPEED @24 RPM DEVELOPER HOUSING LOWER AUGER SPEED @100 RPM OTHER DESIGN CONTROL PARAMETERS BOTTLE FEED RATE TO UPPER AUGER - 4 GRAMS PER REVOLUTION UPPER AUGER FEED RATE TO LOWER AUGER - 0. 16 GRAMS PER REV. LOWER AUGER FEED RATE TO DEVELOPER HOUSING -. 23 GR. / REV. VACUUM IN BOTTLE WHILE RUNNING AT FULL RATE < TBD PSI TONER DISPENSER OUTPUT SPEC. FOR F 1/F 2 IS 21. 4 GRAMS PER MINUTE S. Mordenga/ R Faull April'96 EEFW - Robust Design Revision #002