Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
1
P14471 Vibration Testing Apparatus II Final Review
5/13/2014
Brett Billings
Jacob Gardner
Nick Greco
Ron “Sparky” Jimbo
Claire Kobal
Ryan Selig
Ashley Waldron
Agenda
2
Revisit of Requirements
Build
Test
Issues
Lessons Learned
Project Management
Appendix
3
Revisit of Requirements
Customer Requirements
4
Engineering Requirements
5
Number Requirement Raw
Score Units
Ideal
Measure 1 Displacement of luminaire 69 in 1/32
2 Vibration of luminaire 69 cycles/min 2000
3 Duration of vibration test 69 hours 35
4 Isolate motor from oil spills 45 Binary Yes
5 Maintenance Documentation 44 Survey (easy to follow) 80%
6 Mount pendant configuration 35 Binary Yes
7 Operation Documentation 33 Survey (easy to follow) 80%
8 Machine won't operate if crankshaft enclosure is open 25 Binary Yes
9 Completely stop machine with Emergency Stop 25 seconds < 10
10 Maximum voltage of Motor 25 V AC 240
11 Minimize number of operators 21 people 2
12 Max weight of mounted luminaire 20 lbs > 150
13 Visually display settings and status of test 18 Binary Yes
14 Steps to set up 16 steps < 10
15 Setup Time 16 seconds < 120
16 Low Sound 15 dBA < 85
17 Minimize pinch points 15 Count < 3
18 Total materials cost 15 $ < 4,000
19 Mount stanchion configuration 7 Binary Yes
20 Mount yoke configuration 7 Binary Yes
21 Mount trunnion configuration 7 Binary Yes
22 Machine footprint 5 in2 < 34X48
House of Quality
6
UL844
7
2000 RPM
1/32’’ displacement
35 hours
8
Build
Frame Assembly
9
Steel was assembled and welded in shop
All welded joints were annealed to relieve hardness caused by the heat during welding
C-channel was machined to fit on top of frame and to hold flange assembly
Baseplate for crankshaft mechanism was modified and mounted to frame
Crankshaft Modifications
10
Modifications to improve adjustment:
Added access holes to connecting rod
Replaced 3/8-24 set screw with a 3/16-100 set
Re-machined T-block for bushing
Replaced springs with weaker ones
Access hole added to rear guard
Additional Work
11
Dial Gauge
Added magnetic back plate
Created alignment block to ensure
gauge is perpendicular to box (cosine
error)
Top guard was re-machined in order to
ensure adequate overhang for bolts
Painted steel frame to resist rusting
Belt Guard Mount There was a misunderstanding in the
machine shop and half of our original belt
guard was thrown away
A new, rectangular guard was fabricated
Elongated holes were drilled in the L-
brackets to allow for vertical adjustment
Issue: Seems like there might be a lot of
vibration. Never tested with machine on.
12
Encoder Mount
Original design:
Attach mount to bottom of belt guard
Issue:
Belt guard adjustment requires encoder to be adjusted as well
Final design:
Metal arm extending horizontally from L-bracket with rubber
pad to prevent excess movement
13
Control of Motor
Variable Frequency Drive added to control motor frequency
VFD spec’d based on:
Remote start/stop
Remote ramp up or ramp down the speed of the motor
External faults that could be used as an E-Stop
14
Feedback from Apparatus
Closed loop control system
RPM read by rotary encoder and sent to
MCU
Raw feedback from the encoder is processed
and displayed on LCD
Encoder needs to have high enough
resolution to keep the error within 1% of
2000 RPM
15
Microcontroller: Programming
16
Microcontroller Unit (MCU) is a product of Texas
Instruments (TI) under the MSP430 family
MSP430G2452 – Main program and LCD communication
MSP430G2553 – VFD switching logic
TI Launchpad board used to reprogram MCU
Microcontroller: Interfacing
17
Supply voltage is 3.3 V DC from voltage regulator unit
MSP430G2452 – Drives control and logic for the test LCD (10 pin connections) Feedback from Encoder Output A (1024 PPR)
Control switch to manually pause/resume test (10 kΩ pull-up) Primitive serial data to MSP430G2553 (2 pin connections)
MSP430G2553 – Directly controls VFD through switches Toggles transistors (switches) used to send digital data to VFD
Additional control switch to power on and off both MCUs and LCD
Display Unit
18
20x4 HD44780 (Hitachi) Liquid Crystal Display (LCD) with
LED backlighting
Implements ASCII printable characters
Character display: 4 rows, 20 columns
Commands entered as parallel data
Data [7..0]; Register Select; Enable
Write mode; fixed contrast and lighting
Microcontroller: Logic/Implementation
19
Initialization
Wait for about a second for display to power on before writing
Set internal clock frequency to 16 MHz Varying clock speed essentially varies all delays and timing involved
Configure input and output pins
Write static items to display
Initialize integer variables
Configure and enable Timer and Edge Triggered interrupts
Interrupt handlers Timer A: Increments a time-keeping variable
Edge 1: Increments frequency counter from Encoder output
Edge 2: Toggles a pause variable from pause/resume control switch
Microcontroller: Logic/Implementation
20
Display routines Send Command Simple routine to send commands in form of parallel data
Command specified as binary string
e.g.: SendCommand("1101010010"); sets character position to the start of line four
Enable automatically set low and then high to send the full command
Write Character Converts inputted character to binary equivalent and sends the command
Direct ASCII compatibility due to HD44780 controller
Write String Writes a sequence of characters; e.g.: WriteString(“Hello world!”);
Microcontroller: Logic/Implementation
21
Main algorithm (simplified)
Test paused (or stopped)
Signal VFD to stop motor
Hold state, including timer and latest status message on display
Test resumed (or started)
Signal VFD to start motor and set to default speed (~2000 RPM)
Resume state
For each passing second…
Update time both internally and on display
Count up to a specific frequency from encoder signal
Use specific frequency and time passed to accurately calculate RPM
RPM within 1% of target: normal operation; else within 5%: ramp motor speed up or down through VFD; else outside 5%: pause test
Pause test if no encoder signal detected or 35 hours elapsed
22
Test
Current Testing Status
23
# Engineering Requirement
Raw
Score Importance
Ideal
Measure Units Tolerance Test Plan Status
1 Displacement of luminaire 69 5 1/32 in ± 3% 1A, 1B, 4 NOT TESTED
2 Vibration of luminaire 69 5 2000 cycles/min ± 2% 2A, 2B NOT TESTED
3 Duration of vibration test 69 5 35 hours 2A, 2B, 3 PARTIAL
4 Isolate motor from oil spills 45 4 Yes Binary 2A, 2B PASS
5 Maintenance Documentation 44 2 0.8 Survey 8 NOT TESTED
6 Mount pendant configuration 35 5 Yes Binary 5 PASS
7 Operation Documentation 33 2 0.8 Survey 8 NOT TESTED
8 Machine won't operate if crankshaft enclosure is open 25 4 Yes Binary 7A, 7B PARTIAL
9 Completely stop machine with Emergency Stop 25 4 < 10 seconds 7A, 7B PARTIAL
10 Voltage of motor 25 5 240 V AC N/A PASS
11 Minimize number of operators 21 3 <= 2 people 6A, 6B PARTIAL
12 Max weight of mounted luminaire 20 5 > 150 lbs 2A, 2B NOT TESTED
13 Visually display settings and status of test 18 4 Yes Binary 3 PASS
14 Steps to set up 16 3 < 10 steps 6A, 6B PARTIAL
15 Setup Time 16 4 < 120 seconds 5A,5B,6A, 6B PARTIAL
16 Low Sound 15 3 < 85 dBA 7A, 7B PARTIAL
17 Minimize pinch points 15 3 < 3 count 7A, 7B PARTIAL
18 Total materials cost 15 3 < 4,000 $ N/A PASS
19 Mount stanchion configuration 7 5 Yes Binary 5 NOT TESTED
20 Mount yoke configuration 7 5 Yes Binary 5 NOT TESTED
21 Mount trunnion configuration 7 5 Yes Binary 5 NOT TESTED
22 Machine footprint 5 3 < 34X48 in2 5 PASS
Encoder Testing
Initial testing done independent of apparatus with a DC motor
to evaluate following components:
Output signal’s amplitude
Output frequency
Power and control connections
Allowed encoder to be integrated with apparatus and
connected to MCU to display speed on LCD
24
VFD Control Testing
The VFD control signals were also tested independently on a
proto-board for isolation purposes
The VFD was programed for the motor used as well as for
the digital outputs needed for the control signal
To remotely control the VFD power transistors were used as
switches
The control of these switches would be coming from the
MCU
25
Safety Test
26
Failed at RIT – 6 of 14 items are “No” Labeling, lockouts, 2nd E-stop
Will pass at CCH in Syracuse Ordered additional lockout & lock
Setup Test
27
Did not meet original engineering metrics
Target: 2 minutes, 10 steps
Actual: 16 minutes, 14 steps
Should have used previous team as benchmark
Drastic improvement vs 49 minutes & 31 steps
Can be done with one person though
Setup Test
28
Standard Work for CCH – steps, time, visual aids
Method to reach displacement in 2 adjustments
Use of dial gauge
Quick reference table
1 turn = .019’’
Reliable to +/- 3%
Current Adjust to Current Adjust to Current Adjust to
> .03130 Backoff 0.0210 0.02613 0.0100 0.02063
0.0310 0.03113 0.0200 0.02563 0.0090 0.02013
0.0300 0.03063 0.0190 0.02513 0.0080 0.01963
0.0290 0.03013 0.0180 0.02463 0.0070 0.01913
0.0280 0.02963 0.0170 0.02413 0.0060 0.01863
0.0270 0.02913 0.0160 0.02363 0.0050 0.01813
0.0260 0.02863 0.0150 0.02313 0.0040 0.01763
0.0250 0.02813 0.0140 0.02263 0.0030 0.01713
0.0240 0.02763 0.0130 0.02213 0.0020 0.01663
0.0230 0.02713 0.0120 0.02163 0.0010 0.01613
0.0220 0.02663 0.0110 0.02113 0.0000 0.01563
29
Issues
Luminaire Vibration
30
Luminaire vibrates more than machine
Extra conduit was threaded
Set screw was tightened
CCH not able to help by email/phone
Resolution:
Try plumber’s tape
Consult CCH in Syracuse during final testing
Motor Mount
Excess vibration in motor mount at 1/3 speed
Mount was reinforced with t-shaped brace
Top View
31
Electrical
VFD damaged during initial testing
Root cause analysis performed
VFD replaced by CCH
All future wiring done by licensed electrician
32
Electrical
33
Encoder mounting
Coupler needed to be modified
Coupler broke right before shipping
Enclosures
Shipped apparatus without
LCD stuff
Control changes from CCH at demo
34
Lessons Learned
Experience Gained
35
Combined structural mechanics with vibration
Required learning more in-depth features of structural
mechanics & natural frequency
Required learning of ANSYS Workbench to analyze the stress
at the operating frequency and the natural frequency of the
frame
Electrical component selection & wiring
Lessons Learned
36
Review the previous MSD team’s work more
thoroughly before starting
Verify existing parts match CAD models
Communicate effectively between both team
members and others helping with the project
Weigh the risks of using free help vs. hired
professionals
Lessons Learned
37
Iterative design process
Escalate issues when you can’t solve them
yourself
Compromise on differences of opinions
between team members
Work together to meet internal deadlines
Clearly define team member roles
e.g. Document owners
38
Project Management
Budget
39
Original Budget = $4,000.00
Current Costs = $2,978.88 Parts = $2,540.33 S/H = $198.55 Expected Travel = $240.00
Items saved on: CCH donated VFD & replacement ($290x2) CCH paid for shipping ($600) No welding fees ($400) No electrician fees ($200)
https://edge.rit.edu/edge/P14471/public/WorkingDocuments/Detailed%20Design/BOM_rev6_tracking.pdf
Risk Analysis
40
Risk Effect Cause Severity Likelihood Score Action Item Owner Due
Poor design 3 2 6 Make it adjustable Brett DONE
Disp. changes during test 3 2 6 Calculate torque required on bolts so displacement won't slip. Brett DONE
Poor calibration 2 1 2 Define calibration procedure and schedule Jake 4/15/2014
Crankshaft slipping 3 2 6 Evaluate previous team's design Nick DONE
RPM changes during test 3 2 6 Design closed loop feedback and VFD to adjust RPM Ryan DONE
Belt slips 2 2 4 Use v-belt and belt guard (for oil drip) Ashley DONE
Motor inefficiencies 3 1 3 Accept Risk --- DONE
Belt is too loose 1 2 2 Design motor mount to be adjustable Ashley DONE
Frame misalignment 2 1 2 Write procedure to verify conduit is 100% vertical Claire 4/15/2014
Motor burns out Costly to replace, test fails Motor burns out 3 1 3 Calculate required torque and RPM, select appropriate motor Ron DONE
Crankshaft improperly lubricated 3 1 3 Define maintenance procedure and schedule Ron 4/15/2014
Motor is too loud 2 1 2 Add motor enclosure, sound damping, hearing protection Brett DONE
Vibration of frame 2 1 2 Add sound damping, add hearing protection Brett DONE
Luminaire breaks and rattles 2 1 2 Accept Risk - test will be ended early --- DONE
Operators don't wear earplugs 1 2 2 Accept Risk - exposure duration is still low --- DONE
Fixed guard not in place 3 1 3 Place warning labels on apparatus Brett 4/15/2014
Machine not properly guarded 3 1 3 Install polycarbonate guard and belt guard Brett DONE
Energy source not isolated 3 2 3 Install location to lockout machine Brett DONE
Lockout not observed 3 1 3 Accept Risk --- DONE
Button is broken 3 1 3 Define maintenance schedule Ryan 4/15/2014
Electrical failure 3 1 3 Have CCH check work with certified electrician Ron DONE
Pinch points exist Operator injury could occur Existence of pinch point 2 3 6 Guard all pinch points Brett DONE
Setup requires awkward positioning 2 2 4 Design conduit to be assembled first, then attached all at once Claire DONE
Luminaire is too heavy 2 1 2 Specify use of lift assist or 2nd operator, Design for use Brett DONE
Project scope 2 2 4 Use current fixtures Claire DONE
Fixture doesn't work with frame 1 1 1 Change frame and/or fixture design Claire DONE
Oil spill on motor connection Motor failure/smoke Leakage 1 2 2 Offset motor so oil can't drip on it Ashley DONE
Vibration is same as natural frequency 3 2 6 Perform ANSYS Analysis, add supports to change natural frequency Nick DONE
Improper construction 3 2 6 Use certified welder Nick DONE
Poor material selection 3 2 6 Perform ANSYS Stress Analysis Nick DONE
Luminaire is too heavy 3 2 6 Perform ANSYS Stress Analysis Nick DONE
Project runs over budget
Customer is dissatisfied,
engineering requirements must be
sacrificed
Unforeseeable expenses 2 2 4 Proper budgeting/compare prices/buy standard sizes Nick DONE
Design decision 1 2 2 Accept Risk --- DONE
Changes in technology compared to original 1 2 2 Accept Risk --- DONE
Correction of mistakes 3 2 6 Participate in milestone design reviews ALL DONE
Delay in Tool Shop 2 2 4 Schedule machine time 1 week prior - not required, drop in OK Ashley DONE
Materials are delivered late 2 2 4 Order during Winter Break Nick DONE
Poor planning 2 1 2 Maintain accurate project plan and adjust as needed Brett DONE
Member doesn't finish assigned work 2 1 2 Refer to Norms and Values - seek help or adjust workload ALL DONE
Electrical shock from apparatus Operator injury could occur Improper wiring 3 2 6 Have CCH check work with certified electrician Ron DONE
RPM can't be measured accurately Apparatus doesn't meet UL844 Encoder vibrates too much 3 1 3 Add damping Jake DONE
Unable to integrate measurement devices with
display
Machine is just as difficult to setup
and monitor as beforeLack of knowledge 2 2 4 Research integration requirements and prepare LabView backup Ryan DONE
Can't get items (Motor, VFD, etc.) from Eaton in
available time
Need to purchase items instead,
increased cost of projectLead time is too long 2 2 4
Request items by specified date from CCH and ask for delivery date. Have
backup supplier prepared.Ron DONE
Apparatus fails test plan after shipping to CCHHard to fix apparatus once it's in
Syracuse
Can't fully test apparatus at RIT due to safety
and power supply requirements3 2 6
Check on shipping LT to Syracuse, investigate using U-haul to deliver it
ourselves, ask if CCH has facilities on site we can use for modificationsBrett DONE
Severity Scale: 1 = Minor, 2 = Noticeable, 3 = Severe
Likelihood Scale: 1 = Improbable, 2 = Possible, 3 = Very Likely
Customer is dissatisfied, Students
receive poor course grade, RIT's
reputation is negatively impacted
Operator injury could occurApparatus is not ergonomically safe to setup
Frame falls apart Test fails, costly to fix
Can't mount all configurations typesCustomer is not able to test some
configurations
Apparatus is larger than existing oneCustomer is dissatisfied, new holes
must be drilled in floor
Project deadline is not met
Operator injury could occurMachine starts during maintenance
Operator injury could occurEmergency stop doesn't work
Machine is too loud
Operator injury could occurMachine operates when moving parts exposed
Hearing conservation or
engineering intervention is
required
Apparatus doesn't meet UL844Vibration is not 2000 cycles per min
Apparatus doesn't meet UL844Displacement is not equal to 1/32in
What’s Left?
Remaining work:
Finalize technical paper
Travel to Syracuse for final testing – Test Plan B’s
Fix issues detected during testing
Solve excessive luminaire vibration
Deliver maintenance & operation documentation
If we don’t finish by 5/22:
Enough work for a Phase III?
Work for a Co-op?
41
Questions?
42
43
Appendix
Room Layout
44
Selected Design
45
Interchangeable Conduits
Motor with V-belt
VFD
Digital Dial Gauge
Encoder
LCD, Microcontroller
Polycarbonate Guards
E-stop
Paint for Rust Protection
Test Plans
46
Refer to Excel document
https://edge.rit.edu/edge/P14471/public/WorkingDocuments/MSDII_B
uild%20Test%20Documentation/P14471 Test Plans_Rev2.xlsx
or EDGE
https://edge.rit.edu/edge/P14471/public/Build%2C%20Test%2C%20Do
cument
Full Risk Analysis
47
See EDGE
https://edge.rit.edu/edge/P14471/public/ProjectManageme
nt/Risk%20Analysis.pdf
VFD Damage Incident
48
See EDGE
https://edge.rit.edu/edge/P14471/public/WorkingDocumen
ts/VFD%20Damage%20Incident/VFD%20Damage%20Incide
nt.pdf