Xerox Bearing Tester Detailed Design Review [1] 8 February 2013 - PowerPoint PPT Presentation

Xerox Bearing Tester Detailed Design Review [1] 8 February 2013 P13505 1

Project Team and Stakeholders Customer Erwin Ruiz Technical Specialist Project Manager Xerox Corporation Guide William Nowak Principle Engineer Xerox Corporation Faculty Consultant Jason Kolodziej RIT ME Faculty Project Team Stephen Rugg Project Manager Will Craig Co-lead Engineer Andrew Shuman Co-lead Engineer Kevin Albino Team Facilitator/Project Engineer Lauren Kaczor Project Engineer Tyler Hill Project Engineer 8 February 2013 P13505 2

Meeting Agenda - Project Background 1:05-1:10 - Updated Customer Needs & Specs (Lauren) 1:10-1:20 - Test Plan (Tyler) - MSA Plan (Tyler) - System Assembly 1:20-2:10 - Overall Design (Kevin) - Motor Design (Andrew) - Motor Torque Requirements (Will) - Mandrel (Andrew) - Alignment System (Lauren) - Clamp (Andrew) - Frequency Analysis of Rails (Steve) - Data Acquisition 2:10-2:30 - Accelerometer Mounting (Steve) - Signal Conditioner (Steve) - DAQ (Will) - LabVIEW (Will) - Budget/BOM (Tyler) 2:30-2:40 - Risk Assessment (Lauren) 2:40-2:45 - Questions 8 February 2013 P13505 3

Project Background • There are 2 Bearings in the iGen Printer System supporting the fuser roller • Every 200,000 prints the printers come in for remanufacturing • Bearing is reused a maximum of 5 times • Currently, there is no quantitative test to qualify bearings – Current test is to hold the bearing up to your ear and see if you hear or feel anything – Potentially leads to many good bearings being scrapped • Customer wishes to test performance of other bearing manufacturers against current product 8 February 2013 P13505 4

Project Background 8 February 2013 P13505 5

Updated Customer Needs 8 February 2013 P13505 6

Specifications 8 February 2013 P13505 7

Test Plans • SP1: Time To Train User Ideal Value < 1 hour Plan: Gather a group of 10-15 people and teach them individually how to use the system. Document how long it takes each person to gain competence with the system. Ideally, the group will contain a broad range of knowledge bases with regards to vibration and mechanical systems. This will help to get a better estimation on the time it would take to train someone with no engineering background. The individuals who were trained will be asked about areas that were unclear or difficult to understand. From their feedback, efforts will be made to streamline the process. This will occur whether or not the specification is met. • SP8: Use Xerox’s Machine to Get a Pass Fail Readings on Test Bearings Ideal Value: Pass/Fail Plan: Xerox’s Acoustic machine will be setup and run according to the manual specifications. Pass or fail values will be collected for all of the test bearings. Four tests will be conducted on each bearing to make sure the result is repeatable since the machine has failed an MSA. The purpose of collecting these readings will be to compare them against the results from the vibration test stand that is being developed. 8 February 2013 P13505 8

Test Plans • SP9: Measurement System Analysis on Vibration Machine Ideal Value: Pass Plan: A measurement system analysis (MSA) will be conducted on the vibration machine. If the vibration machine fails the MSA, efforts will be made to identify areas where there are inconsistencies. These inconsistencies will be fixed and the MSA will be conducted again. This process will iterate until the vibration machine passes the MSA or until we run out of time. • SP10: Assess Health of Bearing Ideal Value: Pass/Fail Plan: The results of the vibration collection will be dependent upon the pass and fail criteria that is developed using the test bearings. Vibration measurements will be conducted on each of these bearings and they will be correlated to the tests that were already conducted by Arnprior. The developed pass and fail criteria will then be input into the LabView interface and a pass or fail reading will be output at the end of each test. 8 February 2013 P13505 9

MSA Plan • 16 Samples, 6 Operators, 3 Replicates – Typically 10 Samples, 3 Operators and 2 Replicates is used – As we do not have historical data to compare to, more samples would be better if we could obtain more bearings in various states of wear – More operators will result in an improved precision for the reproducibility estimate [2] • Every operator will test each of the 16 samples, 3 times • The order of the tests will be done based off of a randomly generated schedule • Minitab or another statistical package will be used to analyze the data using the ANOVA method • The output value, pass/fail or some number, will be determined based upon the failure criteria which is still to be determined 8 February 2013 P13505 10

System Assembly Motor Assembly Alignment Assembly Arbor Press Assembly 8 February 2013 P13505 11

Motor Pugh [6] 8 February 2013 P13505 12

Motor Mount Motor • Motor will be mounted on a large aluminum block • The aluminum block will rest on a rubber gasket • This is done in an effort to attenuate Aluminum Block any vibration Motor Clamps Face Mounting Plates 8 February 2013 P13505 13

Motor Clamps • These will wrap around the motor so that it is not cantilevered • The top bolt can be used to further tight the clamp once it has been bolted down • Trying to make the structure as rigid as possible 8 February 2013 P13505 14

Motor Torque Requirements • Tested the required motor torque for our bearings • Preliminary drop test predicted 37 in*lb at 1800 rpm based on Newton’s Viscosity law – extrapolated from 43 rpm • New test with motor at 2000 rpm showed required torque of 2.7 in*lb for high resistance bearing 8 February 2013 P13505 15

Motor Test Results Torque Required To Rotate Test Bearings 3 2.5 2 Required Torque (in*lbs) 1.5 High Resistance Bearing Low Resistance Bearing 1 0.5 0 0 500 1000 1500 2000 Angular Velocity (RPM) 8 February 2013 P13505 16

Xerox Acoustic Machine • Xerox machine has a slip fit mandrel • Found that the bearing slipped on the mandrel when the motor rotated • Show video 8 February 2013 P13505 17

Two Mandrel Designs Slip Fit Mandrel Press Fit Mandrel 8 February 2013 P13505 18

Arbor Press Assembly • ½ Ton Press • 4.5” working range • Cost: $40 8 February 2013 P13505 19

Arbor Press Assembly 8 February 2013 P13505 20

Clamp Alignment System Accelerometer (2) Bearing Brake (2) Bearing Clamp Rail Adapter (2) L-Bracket (8) 80/20 Rails (4) Carriage (2) 8 February 2013 P13505 21

T-adapters • Aluminum – Easy to machine – Higher damping coefficient than steel • Transfers holes from horizontal to vertical position • Connects lower half clamp to the carriage 8 February 2013 P13505 22

Outer Race Bearing Clamp • Will be made out of steel to improve vibration transmission • Because of tight tolerances the Brinkman Lab is going to manufacture it 8 February 2013 P13505 23

ANSYS Clamp Analysis • ANSYS analysis shows a very small deflection: 0.0002 in • This proves that displacement control is not a feasible option and a force control method should be implemented. 8 February 2013 P13505 24

80/20 Rail Natural Frequencies 8 February 2013 P13505 25

80/20 Natural Frequency Calculations Natural Frequency v. Length (Mass of Center) 5000 4500 4000 Natural Frequency (Hz) 3500 3000 2500 2000 Assumptions: 1500 • Applied mass is larger than the beam mass 1000 • Applied mass is located at 0.80*L 500 • Both ends are clamped 4 5 6 7 8 9 10 11 12 Length of Beam (in) • Constant loading can be simulated as additional mass 8 February 2013 P13505 26

80/20 Natural Frequency Calculations 4 Natural Frequency v. Length (No Mass) x 10 2.2 First Mode 2 For the first mode: Β 1 l=4.7300 1.8 1.6 Natural Frequency (Hz) ρ=mass per unit length 1.4 A=cross-sectional area L=beam length 1.2 E=Elastic modulus 1 I=Moment of Inertia 0.8 0.6 • No mass applied other than the 0.4 beam mass • This is the highest natural 0.2 4 5 6 7 8 9 10 11 12 frequency that could be reached Length of Beam (in) without stiffening the beam 8 February 2013 P13505 27

Accelerometer Mounting • 100 mV/g • ICP Accelerometer • Frequency Range: 0.5-10,000 Hz • Attach with wax or glue • If needed the accelerometer could be stud mounted [3] 8 February 2013 P13505 28

Accelerometer Mounting - Feasibility • Outer support accelerometer mounting has been successfully applied in industry and other test stands [4] [5] 8 February 2013 P13505 29

SKF Condition Monitoring • Accelerometer Signal Conditioner • Power ICP Circuit 8 February 2013 P13505 30

Signal Conditioner - Schematic 8 February 2013 P13505 31

Signal Conditioner – Board/Enclosure 8 February 2013 P13505 32

LabView 6210 DAQ • 16 analog inputs • 250kS/s • 4 voltage ranges • Maximum voltage range accuracy: 2.69 mV • Minimum voltage range accuracy: 0.088 mV [7] 8 February 2013 P13505 33

LabView Feasibility Test Current LabVIEW code has been tested in the systems lab 8 February 2013 P13505 34

LabView 8 February 2013 P13505 35

LabView Block Diagram 8 February 2013 P13505 36