P14031: Jib Transfer Bench - Rochester Institute of...

33
P14031: Jib Transfer Bench Matt Brunelle Nicole Conway Mike Kennedy Katy Wurman Sub-System Design 1

Transcript of P14031: Jib Transfer Bench - Rochester Institute of...

P14031: Jib Transfer Bench

Matt Brunelle

Nicole Conway

Mike Kennedy

Katy Wurman

Sub-System Design

1

• Critical Subsystems

• Design to Pursue (old and new)

• Customer and Engineering Requirements (updated)

• Key Constraints to Watch

• Chosen Movement Method and Movement Assist

• Calculations

• ANSYS Models

• Selected Lazy Suzan

• Risk Analysis

• Project Schedule

• Chair Benchmarks

• Questions

2

Agenda

P14031 Problem Statement

• Current State o A jib transfer bench was created in Spring 2013 (P13031) that is heavy,

expensive, labor-intensive to assemble, and accommodating of a strict size constraint present at the time.

• Desired State o A jib transfer bench that enables a jib trimmer to move transversely across the

width of the sailboat, without the use of their legs or core muscles.

• Project Goals o A fully-functional prototype that is:

• lighter

• cheaper

• easier to assemble and manufacture

• Constraints o Designed for use with the Sonar class of sailboats

o Completely mechanical solution

o Designed for ease of reproducibility

o Cannot require alteration of the boat or cause damage

3 1-page Project Summary (EDGE)

• Why? o Most technically challenging

design elements

o Directly effect many of the

most important customer and

engineering requirements

4

• Movement method o Linear vs. rotational

• Movement assist o Gravity?

o Assist bar?

o Pulley or other systems?

Critical Subsystems

5

• Customer likes: o Moves side to side

o Didn’t look intimidating

• Customer dislikes: o Doesn’t span over benches

o Binding potential

o More moving parts

Design to Pursue (10/3)

6

• Advantages: o Simplicity

o Goes out over benches

• Disadvantages: o Moment is HUUUUUGGGE,

Rochester, HUUUUUUUGGE!!!!

Design to Pursue (10/15)

Importance Scale: 1 = High Priority, 2 = Medium Priority, 3 = Low Priority 7

Customer Requirements

Engineering Requirements

8 Updated HOQ on EDGE

• Simple design! o We received feedback from Caitlyn at Piers Park that all previous

iterations were WAY over-engineered

• Easy installation

• Use with a wide range of users o Chair design/dimensions

o Movement assist devices

• An un-intimidating design

• Do not limit feeling of freedom by securing user too

much

• Budget decreased from $2,500 to $1,000

9

Key Constraints to Watch

• Rotational Lazy Susan

design

• Similar to Team

Paradise design

(Paralympic Sport Club

out of Miami, FL)

10

Movement Method Chosen

http://www.teamparadise.org/adaptive_equipment

11

Calculations – Movement Method

Base Tear-Out (Worst Case Load)

Number of

Bolts

Plywood

Thickness

(in)

Washer

Diameter (in)

Load

Required to

Tear the

Turntable Out

of the Base

(Lbs)

Minimum

Turntable

Bolt-Circle

Diameter (in)

4 0.75 1 1178 13.57

8 0.5 0.75 1377 13.74

8 0.75 0.75 2062 9.43

Boat Deck Tear-Out (Worst Case Load)

Part 1: Hand Calculations

Load on

Clamps (A1)

(Lbs)

Load on

Deck from

Base (A2)

(Lbs)

Pressure on

A1 (Psi)

Pressure on

A2 (Psi)

714 962 155 2.7

Bearing Requirements (Worst Case Load)

Turntable

Diameter (in)

Lower Load

(Lbs)

Higher Load

(Lbs)

Minimum

Dynamic

Load

Capacity

(Lbs)

12 370 1842 517

16 370 1412 398

18 370 1269 359

Overall Assumptions:

Load at end of arm = 350 lbs

Arm radius = 22"

Arm height = 14"

Assumptions: Note: Worst case load is a heel

angle of 45 deg, the user starts

at the higher side and swings

unimpeded to the lower side.

Plywood is an Isotropic material

Load is Evenly Distributed Over all Bolts

Safety Factor of 1.5

Materials are Rigid

Assumptions (Hand Calcs): Assumptions (ANSYS):

Load is Applied Evenly Over Entire Area Balsa and CSM are Isotropic

Materials

Materials are Rigid There are 4 Layers of Roving

Pressure Constant Over Area Static Analysis

Gravity is Ignored

Pressure Constant Over Area

Roving Layers 0.030" thick

Assumptions: CSM 0.045" thick

Safety Factor of 1.5 CSM Poisson's ratio is 0.3

Ball Bearings

Desired Life of 13000 Cycles

Cyclic Loading

Standard Rated Life of 106 Cycles

12

• Lazy Susan: $15.46 o Need an 8-hole pattern

o 1000 lbs capacity

o 5/16” thick

o 12” outside diameter

o 2 lb weight

o VXB Part # Kit8999

• Marine-grade plywood

to mount upon o ¾” thick

Selected Lazy Susan

http://www.amazon.com/Capacity-Bearing-Turntable-Bearings-VXB/dp/B0045DV04I

• Gravity!

• Stability bar

• Lock-in position on

each side?

• Braking method?

13

Movement Assist

14

Transfer Speed

Heel angle (deg) Starting Position Ending Position Speed at ending

position (mph)

Speed at ending

position (ft/s)

Centripetal

acceleration (ft/sec2)

45 Side Other side 4.00 5.875 18.56

45 Side Middle 3.45 5.054 13.96

• Max transfer speed

• Moment on lazy Susan

• Will fiberglass floor be

damaged?

Calculations – Movement Assist

Assumptions:

22" (1.83ft) radius

Frictionless

Conc. of energy

15

Positions of Loads

16

ANSYS – BC and Applied Loads

17

ANSYS – Material Layers

18

ANSYS - Deflection

19

ANSYS – Stress Analysis

Risk

Number Risk Cause (why it happens) Effects Severity Probability

Hazard

Score

Actions to reduce failure

mode

1 User impacts bulkhead

during transfer Inadequate/dysfunctional

braking mechanism User uncomfortably jostled 2 2 4

Extensive testing of braking

mechanism

2

Customer does not like

some parts of the

design

Mis-interpreted customer

requirements

Device is not used by the

customer 2 1 2

Keep customer informed

throughout the design

process, seeking feedback

and checking design with

customer

3 System damages boat Device has sharp corners Unhappy boat owner 2 1 2 Break edges of all

components

4 System damages boat

floor Unreliable stress analysis Potential injury 3 1 3

Double check stress

analysis & increase factor of

safety

5 Seat scuffs bench Unwanted seat deflection Seat hits top of bench 1 3 3 Design robust vertical

support(s)

6 Team becomes

unproductive Incompatible personalities

Incomplete/unreasonable

project 3 1 3

Discuss issues with

group/advisor

7 Environmental

Deterioration Weather Device no longer useable 1 3 3

Use cover when in

storage/use appropriate

materials

8 Improperly secured in

boat

Improperly constrained Unsafe for user to use

(device may fall out) 3 1 3

Properly model boat’s inside

dimensions

Poor install Unsafe for user to use 2 2 4 Clear install instructions

9 Device is overweight

Overdesigned components

Unreasonable

expectations

Installation team becomes

frustrated/injured

Unable to use in races

1 2 2

Select light materials

Keep weight in mind while

designing

10 Person does not fit in

device comfortably

Device does not

accomodate wide enough

range of body types

User is uncomfortable and

may choose to not use the

device

1 2 2 Ensure device is

ergonomically designed

20

Risk Analysis

Risk

Number Risk Cause (why it happens) Effects Severity Probability

Hazard

Score Actions to reduce failure mode

11 Device does not fit in

boat

Improper initial sonar

measurments Unusable device 2 1 2 Measure twice

Improper tolerancing Unusable device 2 1 2 Measure several boats

12 Cannot access jib

lines Poor design Unusable device 3 1 3 Design with anthropometric data

13 Boom impacts user's

head Seat is too high Potential injury 2 2 4

Maximize the possible distance

between boom and seat

14 User has poor

visibility Large vertical footprint Unsafe for user 2 1 2 Small vertical profile

15

Does not comply

with ISAF/IFDS

regulations

Distance between bench and

device seat is more than

200mm

Device cannot

be used in races 2 2 4

Keep regulations in mind while

designing the system (Mechanical

solution, Non-permanent install, No

modifications)

Device is permanently

fastened or requires

modification of the boat

Device does contains non-

mechanical component

16 Complicated

construction

Lots of manufacturing time Decreased

Reproducibility 1 1 1 Minimize the number of parts

Lots of custom parts High cost 1 2 2 COTS parts

17 Complicated to

install

Lots of components Unhappy

installers 2 1 2 Minimize subsystem breakdown

Poor instructions Unhappy

installers 1 1 1 Clear install instructions

18 ANSYS model

inaccurate

Incomplete published data for

material properties Boat deck failure 3 2 6 Obtain boat floor sample to test

21

Risk Analysis - Continued

Project Schedule

22

Task Description Planned Start Planned End Planned Duration

Actual Start Actual End Actual

Duration Longer than

planned? If yes, why?

Design of Rotational Movement Mechanism

10/4/2013 10/4/2013 1 10/16/2013 10/23/2013 7 YES

With the changed design the Rotational movement

mechanism became a more important subsystem and

needed more focus.

Also, calculations done twice (static & dynamic loads)

Design of Lateral Movement Mechanism

10/4/2013 10/11/2013 7 N/A N/A N/A N/A N/A

Device Interface with Sonar 10/9/2013 10/17/2013 8 10/16/2013 10/27/2013 11 YES

The interface is more complicated and a more critical

part of the design with the switch to the cantilevered seat

design

Design of Movement Assist Mechanism

10/14/2013 10/17/2013 3 10/16/2013 10/21/2013 5 YES Scheduling conflits with team

members

Risk Analysis (Update) 10/18/2013 10/21/2013 3 10/24/2013 10/25/2013 2 NO

Detailed Test Plan for high risk subsystem

10/18/2013 10/21/2013 3 10/28/2013 10/28/13 1 NO

Subsystem Design Phase Presentation & Peer Reviews

10/24/2013 10/24/2013 1 10/29/2013 10/29/2013 1 NO

23

24

25

http://www.designflick.com/2012/06/cantilever-chair-by-choi-byung-hoon.html -priceless

http://www.contextgallery.com/shop/panton-cantilever-chair-standard/ -$255

http://www.ssbtractor.com/page40.html -$143

Seat Benchmarking

http://www.ssbtractor.com/page40.html -$80.29

26

http://photo.americascup.com/25-09-2013-San-Francisco-USA-CA-34th-America-s-Cup-,en,igf1724p96n53.html

Questions?

Lateral Seat Movement Cantilevered Arm

Cu

rved

Track

Linear Track over Bench

-x2 Track

Flo

or

Tra

ck

28

Pugh Charts – Lateral Seat Movement

Design 1 Linear track above the benches

Design 2 Linear track on the floor between benches

Design 3 Curved track above benches

Design 4 Cantilevered arm

Design 5 Coaster (negative x-squared)

Design 6 Original (Current Design)

Criteria Design 1 Design 2 Design 3 Design 4 Design 5 Design 6

Access to jib lines 0 - + + 0

DATUM

Estimated material cost + + - - -

Track complexity 0 + - + -

Weight capacity 0 + 0 - 0

Device weight (less is better) + + - + 0

Footprint 0 + - + 0

Linear range of movement 0 - 0 0 0

Transfer time 0 0 - - -

Safety (pinch points) 0 + - + +

COTS Components + + 0 0 0

Ease of construction 0 + - + -

Sum + 3 8 1 6 1 0

Sum - 0 2 7 3 4 0

Sum 0 8 1 3 2 6 0

Total 3 6 -6 3 -3 0 29

Pugh Charts – Lateral Seat Movement

Design 1 Linear track above the benches

Design 2 Linear track on the floor between benches

Design 3 Curved track above benches

Design 4 Cantilevered arm

Design 5 Coaster (negative x-squared)

Design 6 Original (Current Design)

Criteria Design 1 Design 2 Design 3 Design 4 Design 5 Design 6

Access to jib lines +

DATUM

+ + - +

Estimated material cost - - - - -

Track complexity 0 - + - -

Weight capacity 0 - - 0 0

Device weight (less is better) - - + - -

Footprint - - + - -

Linear range of movement + + + + +

Transfer time 0 - - 0 0

Safety (pinch points) - - + - -

COTS Components 0 - - - -

Ease of construction - - - - -

Sum + 2 0 2 6 1 2

Sum - 5 0 9 5 8 7

Sum 0 4 0 0 0 2 2

Total -3 0 -7 1 -7 -5 30

31

Movement Assist Design 2 – Ratchet and Lever Design 1 – Block & Tackle w/

Tensioner and Cleats Design 3 – Stability Bar w/

Carabiners

Design 4 – Hand Cranks (double boat winch)

Design 6 – Block & Tackle w/ Tensioner and Carabiners

Design 5 – Center-Mounted Crank

Pugh Charts – Movement Assist Criteria Design 1 Design 2 Design 3 Design 4 Design 5 Design 6 Original

Possibility of pinch points 0 + + + + 0

DATUM

Time to transfer between

sides 0 - + - + 0

Effort to transfer between

sides + - - + - +

Estimated cost - - + - - -

COTS components 0 0 0 0 - 0

Feasibility of design + - 0 0 0 0

Sum + 2 1 3 2 2 1 0

Sum - 1 4 1 2 3 1 0

Sum 0 3 1 2 2 1 4 0

Total 1 -3 2 0 -1 0 0

Criteria Design 1 Design 2 Design 3 Design 4 Design 5 Design 6 Original

Possibility of pinch points - -

DATUM

- - - -

Time to transfer between

sides - - - - - -

Effort to transfer between

sides + - + + + +

Estimated cost + - - - + +

COTS components + - + 0 + +

Feasibility of design - - 0 0 - -

Sum + 3 0 0 2 1 3 3

Sum - 3 6 0 3 3 3 3

Sum 0 0 0 0 1 2 0 0

Total 0 -6 0 -1 -2 0 0

*** assume that the user does

not have a stability bar***

Movement

mechanism

Locking

Position

Design 1 Block and tackle with

tensioner Cleats

Design 2 Ratchet and lever None (self

locking)

Design 3 Stability bar Carabiners

Design 4 Hand crank (double

boat winch)

None (self

locking)

Design 5 Center mounted crank Carabiners from

chair

Design 6 Block and tackle with

tensioner Carabiners

32

• What’s more important? o Transfer speed

OR

o Minimal effort to transfer

• Is a stability bar a viable option as a movement

method?

Questions for you:

33