University of Portland Donald P. Shiley School of Engineering LETTER OF TRANSMITTAL DATE: April 17, 2014 TO: Dr. Munro, ME 482 instructor (CC: Dr. Khan, Mr. Dave Kammeyer) FROM: Ben Bruns, Stephen Christensen, Cody Fast SUBJECT: Adjustable Wheelchair Senior Design Report The purpose of the Adjustable Wheelchair Senior Design Project was to create a functioning, adjustable rugby wheelchair to serve as a means for collecting size dimensions for players looking to purchase a new chair. The project in its entirety took place on the University of Portland campus, the design team s off-campus work site, and the Vancouver VA gymnasium where the Portland Pounders practices were held. This memorandum is a summary of the project and its procedures, results, conclusions, and appraisal of the project. To accomplish the designing and fabrication of a functioning adjustable wheelchair, design criteria were first established to guide the project. The most essential of these criteria were adjustability of seat width, dump angle (seat angle CW from horizontal), easily manufactured, and within the budget provided by the engineering school. Once the criteria were adequately defined, design features were established through brainstorming and sketches from which a prototype was built. This prototype served as a method for improvement and simplification and eventually led to a second prototype. Using the second prototype, sketches, and the previously created criteria table the model was imported into SolidWorks with minor design changes as required. Stress analysis was performed on areas of concern and further design revisions were made. The resulting SolidWorks model was then analyzed by the engineering school s shop technicians to evaluate the manufacturability of the chair, parts were revised within the computer model, and the stress analysis was performed on revisions. Drawings that were derived from the CAD model were then used in material ordering and fabrication of parts. The chair was assembled and the project was presented on during the University of Portland Founders Day. The chair that was produced by the previously mentioned process meets the design criteria initially established at the beginning of the project. The chair can be manipulated to adjust seat width, dump angle, and seat back height. Further improvements are suggested before the chair is fully functional and can be effectively used as a measuring device. The chair which was manufactured still requires the installation of a footrest and a moveable axle in order to permit full adjustment of the seat dimensions. Also, it is recommended that the shaft collars used in the final prototype be phased out in favor of quick release collars to remove the need for a properly sized wrench. The ergonomic knobs should also be used in place of bolt heads to allow for seat width adjustment without the use of wrenches. The final prototype is capable of being freely adjusted. However, further design iterations are necessary to permit for ease in adjustability and to accommodate a larger range of users. Thank you for your time. If you have any questions, please do not hesitate to contact any of us at bruns14@up.edu, christen14@up.edu, or fast14@up.edu.
University of Portland Donald P. Shiley School of Engineering ME 481 - ME 482 Mechanical Engineering Project I - II Adjustable Wheelchair Senior Design Report Submitted by: Ben Bruns, Stephen Christensen, Cody Fast April 17, 2014 Project Duration: Fall 2013 - Spring 2014
1 Table of Contents List of Figures and Tables... 2 Acknowledgements... 3 Executive Summary... 4 I Introduction... 4 II Background... 4 III Discussion... 6 III.A Design Considerations and Criteria... 6 III.B Sketches... 10 III.C Prototyping... 11 III.D Computer Aided Design (CAD) Model... 12 III.E Stress Analysis... 13 III.F Sheet Metal Part Revisions... 13 III.G Ordering Materials for Prototype Construction... 14 III.H Fabrication... 14 IV Conclusions and Recommendations... 16 V References... 18 Appendix A... 19 Appendix B... 29 Appendix C... 36
2 List of Figures and Tables Figures Figure 1: Dimension reference diagram... 5 Figure 2: Action Item log... 6 Figure 3: Initial design sketches... 11 Figure 4: LEGO Technic prototype... 12 Figure 5: Main items for stress analysis... 13 Figure 6: Completed SolidWorks mode... 15 Figure 7: Completed final prototype... 15 Figure 8: Sketches of design for fully adjustable axle... 16 Tables Table 1: Design Considerations... 7 Table 2: Design project criteria table... 8 Table 3: Milestones... 9 Table 4: Project budget; estimated and actual... 10
3 Acknowledgements The design team would like to thank the following for their support of this project: Dave Kammeyer Gordon Johnson Allen Hansen Jacob Amos Dr. Khan Dr. Lulay Dr. Munro Jason Ball (Ran-Tech Engineering & Aerospace) Industrial advisor Portland Pounders technician UP shop technician UP shop technician Faculty advisor Senior capstone instructor Senior capstone instructor Laser cutting sheet metal
4 Executive Summary The Adjustable Wheelchair Senior Design Project had the objective of developing a functional, adjustable rugby wheelchair to serve as a means for collecting size dimensions for players looking to purchase a new chair. The project in its entirety took place on the University of Portland campus, the design team s off-campus work site, and the Vancouver VA gymnasium where the Portland Pounders practices were held. This memorandum is a summary of the project and its procedures, results, conclusions, and appraisal of the project. Design criteria were initially established to guide the project. The most essential of these criteria were adjustability of seat width, dump angle, manufacturability, and within the budget provided by the engineering school. Once the criteria were adequately defined, design features were established through brainstorming and sketches from which a prototype was built. This prototype served as a method for improvement and simplification and eventually led to a second prototype. Using the second prototype, sketches, and the previously created criteria table the model was imported into SolidWorks with minor design changes as required. Stress analysis was performed on areas of concern and further design revisions were made. The resulting SolidWorks model was then analyzed by the engineering school s shop technicians to evaluate the manufacturability of the chair, parts were revised within the computer model, and the stress analysis was performed on revisions. Drawings that were derived from the CAD model were then used in material ordering and fabrication of parts. The chair was assembled and the project was presented on during the University of Portland Founders Day. The chair that was produced by the previously mentioned process meets the design criteria initially established at the beginning of the project. The chair can be manipulated to adjust seat width, dump angle, and seat back height. Further improvements should be made before the chair is fully functional and can be effectively used as a measuring device. I Introduction The purpose of this project was to design and build an adjustable wheelchair for the sport of wheelchair rugby. This project was one of several ideas pitched to the University during the spring 2013 semester by Dave Kammeyer, a University of Portland engineering school alumnus and player for the Portland Pounders wheelchair rugby team. The project was motivated by the need for an adjustable wheelchair to aid in the sizing of new chairs for wheelchair rugby players. This report will provide background on the sport of wheelchair rugby and discuss some of the key variables in selecting an appropriate chair. The design process and the features of the final prototype will also be discussed. Finally, the project results and recommendations for future work and improvements will be provided. II Background Wheelchair rugby is a team sport in which two teams of four players each try to advance a ball to an end zone at the opposite end of the playing field--generally a basketball court. It is a contact
5 sport, and the player s chairs often collide during play as the players attempt to prevent the opponent from advancing the ball. Rugby wheelchairs take one of two forms. The first is the attack chair, which is designed to play an offensive role. Attack chairs are fitted with a curved bumper on the front for ramming and to make them harder to trap. Defense chairs have a pick bar instead of a bumper on the front. The pick bar is designed to trap the wheel of an opponent s chair, preventing them from moving. Both attack and defense chairs share a similar frame structure, are braced and reinforced, and are generally fabricated from aluminum [1]. Rugby wheelchairs are expensive, costing upwards of $3000. Each player has to have a chair custom-sized to fit them, which is then custom-built by a company such as Vesco or Melrose. The chair order forms are complicated, and may require the customer to specify upwards of 10 different dimensions, as shown in Figure 1 below [2], [3]. Figure 1: Dimension reference diagram from Vesco Metal Craft rugby wheelchair order form [2] The process of determining the optimal dimensions for a new chair is tedious, especially for new players who are not used to rugby wheelchairs. Currently, the best method for determining the dimensions for a player s chair is to have them sit in other players chairs in order to get a feel for which dimensions may be carried over to the new chair and which dimensions may not. While workable, this process is time- and labor-intensive and does not yield enough quantitative data to provide much accuracy in dimensioning new chairs. As such, many chairs must be modified post-delivery. This modification process involves cutting and re-welding the chairs to fit the player better. This trial-and-error process is inconvenient and expensive, prompting the need for the design of an adjustable chair. The project will directly benefit the Portland Pounders wheelchair rugby team, and aims to provide them with an adjustable chair for sizing. This will help to save them money as well as, hopefully, providing them with a better-fitting chair overall. This project also has the potential to make wheelchair rugby more accessible to paraplegics and quadriplegics by lowering the initial cost, which is a significant barrier to entry. It will also help to promote fitness in an oftenoverlooked segment of society.
6 III Discussion Considering the problem of determining a player s dimensions, the project result was to create a means for gathering this information. Measuring a player outside of a rugby chair would be a good solution; however, it would not capture key aspects of a rugby chair. If arbitrarily measured, there would be no way to account for the feeling, comfort, and functionality that accompanies a well-fitting chair. In light of this, it was deemed necessary early in the design process and through interactions with rugby players that a full sized, adjustable wheelchair would be the best solution. This would allow for players to determine how the incremental adjustments affected the feel of the chair as well as to experience these changes in conjunction with their own controlling of the chair. This adjustable chair would also allow for a player to choose features or sizes they desire rather than purely using the quantitative sizing numbers. The end result and goal is that players playing experiences will be enhanced because their individual needs and wants are met by their chair. Figure 2 below contains action item logs detailing the procedure undergone during this project. Figure 2: Action Item log for Fall 2014 (right) and Spring 2014 (left) III.A Design Considerations and Criteria Prior to all design processes, the tasks of designing and building an adjustable wheelchair were examined in light of various factors. The adjustable chair produced through this project would have environmental, consumer, societal, economic, and manufacturing effects, in addition to
7 others, that would need to be considered. These considerations were examined and are listed with their assessments in Table 1 below. Table 1: Design Considerations Design Considerations Performance Serviceability Economic Environmental Environmental Sustainability Manufacturability Ethical Health and Safety Social Project Application Parameters for device function were determined. Repair is not a concern, as the device will not undergo prolonged use. Steel was selected as the primary material for economic reasons. The design will have no foreseeable effects on the environment. Sustainability was not considered, as the device will have limited environmental impact. Manufacturability will influence material selection and final geometry. We have not knowingly violated ethical rules, and do not intend to do so. The device will benefit those with disabilities by ensuring their chairs fit properly. We intend on ensuring that the device is mechanically safe before allowing human testing. The device will be designed to avoid pressure sores. By easing the process of purchasing a new chair, this device will help to improve the quality of life for disabled athletes. Relevant Section in Report III., Appendix N/A III.F. N/A N/A III.F., III.G., III.H. N/A III.E., IV. II., III. Political Politics were not a factor in this design. N/A The design team went to a Portland Pounders practice in September 2013 to obtain preliminary information and to gain an understanding of the scope of the project. Dave Kammeyer and Gordon Johnson, the Portland Pounders equipment technician, were able to provide valuable input and recommendations. As a result of this initial interaction, the design team was able to order several aspects of the project by priority and compile a criteria table to guide the design process. The criteria table is shown below in Table 2.
8 Table 2: Design project criteria table # Criteria Priority Description 1 Dump Angle Essential Ability to adjust the angle of the seat bottom. 2 Width Essential Ability to adjust the width of the seat bottom. 3 Manufacturability Essential Must be able to be built with the resources available. 4 Low Cost High Must not cost more than the budget provided. 5 Easy to Use (Adjust) High Must be adjustable while the player is seated in the chair and the player must be able to maneuver the chair. (The player does not necessarily have to be able to adjust the chair on their own.) 6 Center of Gravity Medium Ability to move the axle forward and backward. 7 Footrest Medium The ability to adjust the angle and height of the footrest. Among the many things learned as a result of the initial meeting with Dave and the Portland Pounders were that the design team should place primary importance on adjusting the seat width and dump angle (angle of tilt of the seat) and secondary importance on an adjustable axle, footrest, and camber angle of the wheels. In order to meet these goals, the team made some further decisions regarding the construction of a full-size prototype. First, mild steel would be used instead of the aluminum usually used for rugby wheelchairs. This would keep the cost of materials down, as well as allowing for ease of welding, though the chair would likely be heavier than if it were made from aluminum. Second, the wheelchair prototype would not be intended for play, only for sizing. This would allow the team to eliminate many of the reinforcing members needed for stability during play, and also minimize the effect of the added weight of steel. After the initial consultation with Dave Kammeyer, Gordon Johnson, and the Portland Pounders, the design team began to brainstorm ideas and create a framework for the project. The team planned out a list of milestones and a preliminary budget to keep the project on schedule, as shown in Tables 3 and 4. The initial budget included items such as prefabricated hinges, as well as casters, wheels, and axle components that would be purchased from a rugby wheelchair manufacturer. The list of milestones was primarily structured around the due dates and project requirements of the ME 481 and ME 482 classes.
9 Table 3: Milestones Milestones Description Date Due Project Plan Project plan completed. 9/27/2013 Design Decision Prototype Constructed Prototype Analysis Decide between alternative designs for use in prototyping. Completed construction of prototype device. Have taken pictures of prototype and evaluated results. 9/27/2013 10/11/2013 10/23/2013 Prototype Demo Memo Demo Memo completed. 10/25/2013 Status Report Memo Have Memo completed. 11/22/2013 Final Oral Report Slides Completed PowerPoint for final report. 12/6/2013 Final Design Purchase Materials Construction Order Final Product Assembled Device Testing Player Testing Final Presentation Final Report Determine precise design for final device including material selection. Purchase materials for construction of device. Complete assembly plan and begin assembly process. Complete construction of full-sized device. Complete quantitative testing of device components and assembly. Complete qualitative testing by rugby players. Complete powerpoint and organization for final presentation. Finish writing and compiling final report on project. 12/6/2013 12/6/2013 1/24/2014 2/14/2014 2/21/2014 3/7/2014 3/22/2014 4/8/2014
10 Table 4: Project budget; estimated and actual Component Estimated Cost ($) Actual Cost ($) Steel Tubing 105 87 Hinges 32 Shaft Collars 120 144 Casters 180 Axle 62 Wheels 550 Nylon 40 Prototype Material 100 Total Cost $ 1229 $ 271 III.B Sketches During the initial planning, the team discussed options such as whether or not to purchase and then modify either a standard wheelchair or a rugby wheelchair. Ultimately, it was decided that the team would have the most freedom building a full-size prototype chair from scratch. The team also began to sketch initial design concepts for adjustable features. Some of the first sketches are shown below in Figure 3.
11 a) b) c) d) Figure 3: Initial design sketches showing a) whole chair, b) front sliding joint, c) hinge joint, and d) telescoping pipes with quick-release shaft collar III.C Prototyping Based on these preliminary sketches, the team built a miniature prototype out of LEGO Technic parts. This small prototype allowed the team to move beyond the two-dimensional sketches and model the chair in three dimensions. For the first time, the team was able to evaluate the feasibility of many of the hinges and brackets, and to consider what kind of frame would have to be built. The LEGO prototype is shown below in Figure 4.
12 Figure 4: LEGO Technic prototype The team gained valuable information from the initial, small-scale prototype. First of all, the frame that had been considered and used in the prototype involved a single member running from front to back underneath the center of the chair. The LEGO prototype demonstrated to the team that one central member would likely not provide enough stability for the chair. This motivated the team to change the design from one central member to two members, one located on either side as explained in Appendix A, DDD1. The second major piece of information gained from the prototype was that the side members were not fully constrained. The team had focused on making the chair adjustable, but had failed to consider that the structural rigidity of the chair had been compromised in the process. The identification of this problem led to further design decisions that constrained the side members at the back of the seat as explained in Appendix A, DDD2. One further change that was made at this point in the design process was to design sheet metal parts to use for the hinges, brackets, and sliding joints. III.D Computer Aided Design (CAD) Model To further the design process after initial prototyping, the ideas gathered from establishing design criteria, prototyping, and the initial sketches generated were combined and implemented into a SolidWorks model. During the transfer of the model s many characteristics into the CAD software, joints were modified to more accurately represent reality than their LEGO counterparts. In addition to joint revisions, problems with adjustability were also addressed. Initial seat dimensions were gathered from a sample chair provided by Dave Kammeyer and were increased so as to allow for a range of adjustability. The seat width was made adjustable from 0-16 while the dump angle had an allowance is 17-31. There is some variability in these numbers as the complex nature of the chair may cause interference between the axle and the seat base depending on the dump angle and seat width set. Once the chair was fully dimensioned, the axle and wheels were added to the assembly and final dimensional changes were made as necessary.
13 Continuing updates were made to the SolidWorks model of the full chair assembly as the design process continued. After consulting with the shop technicians in the Donald P. Shiley School of Engineering, further design revisions were made to account for manufacturability, a criteria previously noted as essential. With the wheelchair design meeting functional and manufacturing requirements, the materials were ordered and manufacturing was allowed to commence. Appendix C contains all of the engineering drawings for the final assembly. III.E Stress Analysis To better understand the stress magnitude and flow within the design, stress analysis was performed using Autodesk Simulation Mechanical (ASM) on many of the wheelchair s components. Because the chair assembly in its entirety would be rather difficult, tedious, and time consuming, the analysis was performed on sub-assemblies and constituent parts in order to increase efficiency. The main areas of concern that were to be addressed were the front, middle, and rear joints of the seat connecting this entire assembly to the frame at two points, this was the main area of concern. These areas are highlighted with red arrows in Figure 5 below: Figure 5: Main items for stress analysis Stress analysis was also performed on the tubing segments used for the base frame which are highlighted in Figure 5 by a green arrow. The procedure and results for the stress analysis are detailed in Appendix B. III.F Sheet Metal Part Revisions Following the primary completion of the SolidWorks model of the full chair assembly and collaboration with the shop technicians, there were a few portions of the chair assembly that required addressing. Three parts in particular or similar geometries contained 90 bends that were too close to bend using the conventional, hand, press brake included in the shop. This required redesigning these parts in order to allow for ease in manufacturing, and allowing the UP shop to fabricate them without added time or hassle. In order to address the problem, the design team proposed to split each of these parts into two separate pieces. For the rear sliding bracket this was accomplished by replacing the bracket with parallel plates to be bolted over the frame
14 for support and rigidity, which is discussed in DDD4 of Appendix A. The middle bracket was simply split with the bends on the middle of the base, as discussed in DDD5 of Appendix A. This was a suitable solution in that it would be welded next to its counterpart in the same way as it was previously intended with the addition of a weld bead to hold it together. The last design revision made was a redesign of the front articulating joints. This part was decomposed into two similar parts with the ability to nest one on the other in order to share concentric hole relations. Similar to the rear sliding bracket, the front sliding bracket was also replaced by two plates clamped over the front frame member by three fasteners, as discussed in DDD6 of Appendix A. III.G Ordering Materials for Prototype Construction For all of the circular, square, and rectangular tubing used in the construction of the final prototype, 1018 mild steel was used. AISI 1018 CD steel has a tensile strength of 64 ksi and yield strength of 54 ksi. It also possesses a Brinell hardness of 126 [4]. For the selection of material, standard piping sizes were examined. Using the sample chair provided, a pipe diameter was estimated for the steel chair. For the seat base, it was discovered that, given the standard schedule 40 wall thickness, nesting pipes would have a large air gap. In order to minimize this, steel piping with a schedule 10 wall thickness was chosen for the outermost nesting pipes. Schedule 40 piping was used for the remainder of the chair components. In the end, three varieties of steel tubing were used in construction. 1 inch, schedule 10 piping was used for the outer nested piping for the seat base assembly. ¾ inch, schedule 40 piping was used for the inner nested piping for the seat assembly. Finally, 1 inch schedule 40 piping was used for the seat back assembly, axle, and caster assemblies. Dimensions taken from the sample chair were also used to determine the necessary size for the square and rectangular tubing used in the construction of the chair frame. For the square tubing, 1 inch by 1 inch tubing was used. For the rear rectangular member, 2 inch by 1 inch tubing was used. Additionally, ⅛ inch gussets were purchased to support the axle assembly. The final components purchased were the shaft collars. Initially, the exact outer diameter of our nested piping was measured. Steel, quick release shaft collars were then ordered in the appropriate size. These components took several weeks to arrive and, when they did arrive, they were not the size they were listed as. In order to complete the chair on time, new steel shaft collars were ordered, though these were not quick release. The new shaft collars required an allen wrench to make adjustments which was not ideal but still fulfilled the desired function. III.H Fabrication To begin the construction of the prototype, the large portions of pipe were cut down to just over their respective lengths to allow for notching and angles. The frame of the chair and the seat back assembly were welded as designated using gas metal arc welding (GMAW). Sheet metal parts were cut using a Mitsubishi HV laser cutting machine. The resulting product was then assembled and the 24 inch wheelchair wheels and the castors were attached. The resulting chair is shown below in Figure 6 and Figure 7.
15 Figure 6: Completed SolidWorks model of chair assembly Figure 7: Completed final prototype
16 IV Conclusions and Recommendations The device that was constructed was able to meet all of the objectives that had been established. In the final prototype, the rear brackets could be slid along the rear rectangular member of the chair frame to adjust the width of the player s seat. Additionally, through the use of nested piping and shaft collars, the dump angle of the seat could be adjusted. The front sliding brackets could also be moved to adjust the front width of the chair. These adjustable features fulfilled our two primary objectives; to adjust the seat width and dump angle. Through design iterations, described in the attached design decision documentation, the criterion of manufacturability was fulfilled. The final prototype was also completed under the allotted budget and significantly below the initial estimated cost. Despite the fulfillment of the primary objectives, there were a number of secondary objectives that were not fulfilled. The other two primary adjustable features which are not incorporated in this device are the adjustable footrest and the adjustable center of gravity. Even though they were not included in the final prototype, designs and means of incorporating these features were discussed during the design process. The inclusion of an adjustable footrest would be fairly straight forward. Hollow tubes could be mounted on the inside of the chair s frame that would allow a footrest to be raised or lowered. A method as simple as a threaded pipe and a nut could then be used to adjust the height of the footrest. Allowing the adjustment of the center of gravity would be much more difficult to incorporate, but it is also a major factor which players need information about before purchasing a chair. During the initial design and prototype stage, possible designs for an adjustable axle were discussed. Sketches of the initial concept design can be found in Figure 8 below. Figure 8: Sketches of design for fully adjustable axle
17 The design shown above was originally formed for a chair design that used a single central member rather than the more traditional trapezoidal design used in the final prototype. This design called for two brackets attached to the central member, which could slide forward and backward. The axle would then be attached to these sliding components using two i shaped posts and shaft collars. Each end of the axle would be capable of rotating, allowing for the adjustment of the camber angle of the wheels. It would also include a telescoping shaft that could be adjusted with the seat width adjustments. In order to lock the angle of the axle, sheet metal brackets would be used. These would be attached via pins to the two axle ends. However, this design would only allow for a predetermined list of axle angles. Even though the final prototype did not include these final two adjustable features, the project was still a success. The primary goals of adjustability and manufacturability were met. Additionally, the scope of the project remained consistent throughout the process. There were also some challenges encountered. The first major challenge was when the initial LEGO Technic prototype was converted into a SolidWorks model. The imprecise nature of the prototype meant that a number of changes and modifications needed to be made as the design was recreated. Part selection and ordering were also difficult. An initial design was selected and a materials list was created, but many of the components then needed to be redesigned for improved manufacturability. This caused a delay in the ordering of materials, pushing back the schedule. Though the schedule was pushed back several times, this allowed for further refinement and iteration of the design prior to construction. This helped ensure that, when the chair was actually constructed, the components all worked together properly. Future teams should first focus on incorporating an adjustable axle into the design. This was the primary piece that was not included in the final prototype. Additionally, an adjustable footrest should be added to improve the usefulness of the sizing chair. The chair should also be redesigned to use more common piping sizes. The schedule 10 piping used in the final prototype worked well, but it was difficult to find and conflicting information was found during research regarding their actual wall thickness, which could have caused major design problems. The initial design called for quick release shaft collars to allow ease of adjustment for the chair. Proper quick release shaft collars for the piping size and material used were not found. In the future, along with using more common piping sizes, piping should be used for which quick release shaft collars are readily available. Future chairs should also be built from aluminum or chromoly steel. These are materials that actual rugby wheelchair are built from, so they would give the player a more accurate feel for what their chair would feel like. This will be particularly important when an adjustable axle for center of gravity is incorporated, as the steel used in the construction of the final prototype is considerably heavier than the aluminum commonly used. Finally, the chair design should be modified to include positional markings for more accurate sizing. While, in its current state, measurement data can be taken from the chair, it requires the use of a tape measure. Additionally, there is currently no method for ensuring that the adjusted front and rear sliders are centered about the axis of symmetry of the frame. In the future, dimensional marks should be added to the chair so that members can quickly and easily be adjusted to known positions. One method that was considered for doing this was to etch or otherwise mark positions on the chair at specific intervals to allow quick adjustments to be made.
18 V References [1] International Rules for the Sport of Wheelchair Rugby. iwrf.com. International Wheelchair Rugby Federation. 2012. Web. 26 September 2013. [2] Custom Wheelchair Order Form. vmcrugbychairs.com. Vesco Metal Craft, 2013. Web. 26 September 2013. [3] Rhino Script Form. melrosewheelchairs.com. Melrose Kiwi Concept Chairs USA, 2013. Web. 15 November 2013. [4] Budynas, Richard G., J. Keith. Nisbett, and Joseph Edward. Shigley. Shigley's Mechanical Engineering Design. 9th ed. New York: McGraw-Hill, 2011. Print.
19 Appendix A Design Decision Documents DDD1 (DESIGN DECISION DOCUMENT 1) DATE: November 22, 2013 TO: Dr. Lulay, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Ben Bruns, Adjustable Wheelchair capstone project team Purpose Assess structural members. Given LEGO technic prototype. Assumptions Dynamics of the prototype will scale. Solution Analysis of pros and cons for designs. Conclusion Switch to two structural frame members. Purpose: After examining the prototype closely, it was decided that further analysis must be done to determine if one center structural member provides sufficient stability and proper mounting for the main wheels. Given: The LEGO Technic small-scale prototype was constructed with one center member. Assumptions: A full-scale steel prototype may or may not behave like LEGO prototype. It was assumed for the sake of this decision that this is not the case. Solution: Both sides of the argument were analyzed One center member Pros: design is simple, minimal, and would require less shop time Cons: may be prone to twisting especially when primary wheels are mounted, more difficult to balance, gets in the way of the foot rest, untested design Two structural members Pros: allows for stable mounting of the wheels, allows room for a foot rest is more stable overall, provides some resistance to torsion in the members, proven design Cons: adds more parts and manufacturing time, increasing complexity Conclusion: It was determined that the designed should be altered to include two front-to-back structural members instead of one. When the pros and cons of each design were weighed, it became apparent that a design with two members would be superior to a design with only one member in terms of stability and practicality of construction.
20 DDD2 (DESIGN DECISION DOCUMENT 2) DATE: November 22, 2013 TO: Dr. Lulay, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Cody Fast, Adjustable Wheelchair capstone project team Purpose Incorporate casters and axle into design. Given Rugby Wheelchair provided by Dave Kammeyer. Assumptions Design used for sample chair will be compatible with new chair. Solution Casters from sample chair will be used on new chair and similar mounting will be used. Conclusion Must ensure these additions do not cause problems with the design. Purpose: A sample rugby wheelchair was provided by Mr. Kammeyer. This chair included all of the necessary hardware, including casters and axle pins, but does not include the primary wheels. The caster assembly on this sample chair is housed in a portion of metal tubing that has been welded to the frame. Within this section of metal tubing, there is a secondary metal tube, welded in place, to support the bearings for the caster assembly. This chair also includes a main axle, made from a bent piece of metal tubing, to which the axle pins are directly inserted to hold the main wheels. Figure 1: Sample caster assembly Given: The sample chair provides estimated dimensions and geometries that are being used as a basis for the new chair design. Assume: Competition chairs are generally made out of aluminum; however, since steel is stronger than aluminum, it has been assumed that the member dimensions of the sample chair may be safely mimicked without the risk of fracture.
21 Solution: The casters from the sample chair will be removed and used as casters in the new chair. Many of the dimensions from the chair, such as the caster assembly and axle dimensions will be used in the new design. Conclusion: In order for the reuse of parts to be effective, it must be ensured that the sample chair design does not interfere with the adjustable features of the new chair design. Analysis must be done to ensure that the square frame design being used for the new chair does not cause the casters to interfere with the primary wheels. Likewise, the positioning of the axle must be done in such a way that the chair width can still be freely adjusted.
22 DDD3 (DESIGN DECISION DOCUMENT 3) DATE: November 22, 2013 TO: Dr. Lulay, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Stephen Christensen, Adjustable Wheelchair capstone project team Purpose Redesign of wheelchair seat back member mounting to prevent seat width variability. Given LEGO technic prototype. Assumptions Joints will readily slide and if needed, plastic washers can be added. Solution Replace the previously proposed joints with joints that are easier to constrain. Conclusion The proposed joint should eliminate all width variability while conserving adjustability. Problem: After discussing the current design and examining the prototype, it was found that unneeded adjustability, requiring additional constraint, could be removed while design features remained intact. The previous joint, as shown in Figure 1, allows for rotation over both the horizontal and vertical axis as indicated by the dotted lines. This requires constraining the vertical axis at both the front and back of the wheelchair as illustrated in Figure 2. Because there are two joints on the end of dump angle members, a great deal of clamping force would be required to overcome the moments generated at each of these joints, as illustrated in Figures 2 and 3. The current assembly allows the seat width to vary from the front to back; this feature is unneeded and impossible given the fabric seat to be used. Figure 1: Previous joint assembly. Figure 2: Initial prototype with areas of concern noted with red and blue arrows. (Note: In Figure 2, the red arrows highlight the joints which allow for rotation in the clockwise and counterclockwise directions.)
23 Figure 3: Moments developed due to weight of user. Given: LEGO prototype, partial SolidWorks model, full size rugby wheelchair (for comparison) Assume: The seat width should be constant from the back to the front of the seat sling. Solution: In order to remove the ability for the seat width that vary from front to back, the following system was designed. This joint system will allow for dump angle variability, and translation along the bottom member to adjust seat width, as shown in Figure 4, while preventing the previously mentioned rotation. The location of this new system will be as shown in Figure 5. Because of this revision, there is no potential for rotation in the axis of the upright tubing, and the final design requires less constraint than the prior. Figure 4: Adjustability at the various joints. Figure 5: Placement on frame. Conclusion: The revisions to the design should remedy concerns about under-constraint as well as remedy the problem of an inconstant seat width. The proposed design will require slightly more manufacturing time than the previous design; however, it is necessary to prevent these unnecessary movements of the design.
24 DDD4 (DESIGN DECISION DOCUMENT 4) DATE: April 12, 2014 TO: Dr. Munro, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Cody Fast, Adjustable Wheelchair capstone project team Purpose Redesign rear sliding member to increase manufacturability. Given Previous design using a C-channel with two close 90 degree bends. Assumptions Loading conditions will not change appreciably with revision. Solution Replace the former design with two parts connected via fasteners. Conclusion The new design will avoid the close bends of the previous design, increasing manufacturability, while still performing the same basic function. Problem: After discussing initial designs with the Shiley School of Engineering shop, it was discovered that the design for the rear sliding bracket would be difficult to manufacture given its geometry. The initial design, which can be seen in Figure 1, called for a C-channel that would slide over the rear rectangular member of the chair frame, held at the rear by a single fastener. Creating the two close 90 degree bends, however, would prove to be very difficult and would likely require cutting and welding; a process that would introduce roughness to the surface and possible material warping, both of which would have reduced the member s ability to slide easily. As a result of these issues, it was determined that the part should be redesigned. Figure 1: Initial design utilizing a C-channel Given: The press brake machine used to bend sheet metal cannot be used to create two 90 degree in the same direction if the resulting flanges are to be longer than the distance separating them. Additionally, alternative methods for manufacturing the C-channel would have caused additional problems. Assume: The same sliding motion can be achieved through the use of sheet metal plates connected via fasteners.
25 Solution: In order to remove the need for close 90 degree bends and to avoid possible problems resulting from the manufacturing process, the part was fully redesigned. The new design, which can be found in Figures 2 and 3 below, would consist of two rectangular sheet metal parts. These plates would be held together through the use of three fasteners; two in the front and one in the back. In order to improve the ease of assembly, the fastener size designated for these parts was the same as was used in other parts of the assembly. Figure 2: Redesigned assembly Figure 3: Redesigned sheet metal part Conclusion: Though more sheet metal parts and fasteners must now be used in the assembly, the new design does not require any sheet metal bending. This fully addresses the design flaw noted in the original device and simplifies the overall design and the assembly process.
26 DDD5 (DESIGN DECISION DOCUMENT 5) DATE: March 25, 2013 TO: Dr. Lulay, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Stephen Christensen, Adjustable Wheelchair capstone project team Purpose Redesign middle seat pivoting joint in order to increase manufacturability. Given Previous C shaped section. Assumptions Loading conditions will not change with revision. Solution Replace the previously proposed joints with joints that are cheaply manufactured. Conclusion The revised joint will be easily manufactured with the resources contained in the shop. The revision will provide the same fit, form, and function as the previous part. Problem: While consulting with the shop technicians in the Shiley School of Engineering shop, it was made known that the current design for the middle joint on the seat, shown as Figure 1 below, would be difficult to manufacture with the given resources. This difficulty was a direct result of there being two, 90 bends 1 apart with long, extending flanges. If the part was to be formed using a press brake machine, this would be both difficult and expensive. If formed by hand, the part would be difficult to form and of less accuracy than desired. Because of these problems, the design needed to be revised to accommodate for forming operations. This problem is illustrated in Figure 2 below. Note that the far right of the illustration shows the current part problem. Figure 1: First iteration of the mid-seat bracket.
27 Figure 2: First and second design press brake illustrations. Given: The press brake machine to be used cannot bend two 90 in the same direction if the resulting flanges are to be longer than the distance separating them. Assume: Press brake machine will be capable of forming a small flange (approximately 1/3 of other flange length.) without the use of wrapping to prevent blowouts. Solution: If the press brake machine can bend the flange described in the assumptions section. The original C bracket can be split between the two 90 bends resulting in two parts of relatively easy manufacturability. These parts will fit exactly the way the C flag part fit, but because they are L (4) will be need altogether to account for the (2) C brackets. Also, appropriate measures must be taken so that these (4) parts will fit as required on the end of each of their respective tubes. The revised part forming process includes the first and second portions of Figure 2. This revised part is illustrated in Figure 3 below: Figure 3: Revised mid-seat bracket. Conclusion: Although the number of parts required to create the assembly has increased, the manufacturability of the assembly s constituent parts has also increased. This allows for parts to be manufactured using the machines in the Shiley School of Engineering shop.
28 DDD6 (DESIGN DECISION DOCUMENT 6) DATE: April 13, 2014 TO: Dr. Munro, Instructor Dr. Khan, Faculty Advisor Mr. Kammeyer, Industrial Advisor FROM: Ben Bruns, Adjustable Wheelchair capstone project team Purpose: Redesign front fork joint part in order to simplify fabrication Given: Previous fork-shaped design (Figure 1-a), which cannot be bent on UP s press brake, for the reason shown in Figure 2 Assumptions: Loading conditions should not change significantly with revision Solution: Replace single part having two bends with two parts having one bend each (Figure 1) Conclusion: The revised design will be significantly easier to bend, while retaining the same fit, form, and function of the previous part. a) b) c) Figure 1: a) Initial joint design, b) and c) revised joint design Figure 2: Sketch illustrating the difficulty of fabricating the initial design on a press brake
29 Appendix B Stress Analysis Detail The first portion of the chair which was analyzed was that of the frame with the first revision of the seat back assembly installed. Within ASM the assembly was held together by bonded contacts, and constrained on the bottom surface of the castor housings at each of the four corners of the frame from x,y,z translation as shown in Figure 1. The mesh was chosen as brick, and the material was chosen as ANSI 1018 CR. 150 lbf vertical force was then applied at the seat bracket in order to simulate the stresses developed by a ~300 lb player sitting in the chair based on estimations calculated as shown in Figure 2 and Figure 3. This analysis was repeated multiple times with refinement points being added at areas of high stress before the chair was re-meshed and analyzed. Each time the max von Mises stress was recorded along with the approximate mesh size at the area of concern being noted. Once the error between the max von Mises stresses of two consecutive trials was ~1%, the iterative process was stopped and the respective data recorded. This process was repeated twice more for horizontal and 45 forces at 150 lbf in probable directions in order to understand the full range of loading conditions possible. Figure 1: ASM constraint diagram. Black arrows indicate areas with added mesh refinement. For this particular model, the green areas on the far left and right indicate translational constraint in the x, y, and z directions
30 Figure 2: Frame, stress analysis with 150 lbf applied vertically at each set of tabs Figure 3: Closeup view of area of maximum von Mises stress The results of frame and seat back stress analysis revealed that there were areas of concentrated stress near the caster mounts and the square frame components. This value was found after 3 mesh refinement iterations in which it converged to ~2% between iterations. The iteration details are included in Table 1 below:
31 Table 1: Tabulated stress iterations of main frame and seat back assembly Analysis Iterations Maximum Stress (von Mises) (psi) Estimated Mesh Size (in) Stress Error 1 19391.08 0.0928334 2 27416.03 0.066329 0.2927 3 28073.59 0.0664058 0.0234 Although the stress in this portion was relatively high, it was still within the yield strength of the material to be used, ANSI 1018 CD, which has a yield strength (σ ys) of 54.0 ksi. In addition, this stress analysis does not take into consideration the effects of welding and the material that will be added in this junction to decrease the stress concentration. This analysis, however, was useful in that it yielded reasonable results for possible stress levels. The second item that was analyzed was that of the seat back assembly in order to understand their behavior under load. For this analysis, the seat back assembly was treated as a single part with bonded contact between all constituent parts and similar to the prior analysis, the meshing style was brick. The portion of the upright that was constrained was the bottom C bracket and which was constrained as described in Figure 4 below. Vertical forces of 75 lbf were applied at the tab s holes (of which there are 2 tabs on each of the 2 seat back assembly) in order to simulate the weight of 300 lbf sitting in the chair. Figure 4: Forces and constraint details for seat back assembly
32 The resulting analysis of this assembly revealed that the von Mises stresses developed by the 75 lbf forces was close to 23,763 psi after 6 meshing iterations. This was once again within the bounds of the material properties and provided the assembly with a reasonable factor of safety assuming the passenger s weight does not exceed the 300 lbf assumed weight. The results of this analysis are detailed in Figure 5 and Table 2 below: Figure 5: Resulting stress of 75 lbf. load applied to tab s holes Table 2: Tabulated mesh refinement iteration with resulting stress Analysis Iterations Maximum Stress (von Mises) (psi) Estimated Mesh Size (in) Stress Error 1 5121.087 0.280128 2 16967.22 0.0932523 0.6982 3 18595.12 0.0664396 0.0875 5 23723.11 0.0538551 0.2162 6 23763.8 0.0470859 0.0017 The third area of concern that underwent analysis was the front articulating joints. This joint is illustrated in Error! Reference source not found. by the farthest right, red arrow. These joints
33 were analyzed using similar methods to those previously mentioned as illustrated in Figure 6 and Table 3 below. From the analysis it was determined that under the proposed loading conditions the von Mises stress are within the yield strength of the part for the majority of the part. The only areas which lie above the acceptable stress magnitude are at the junction of the washer s cylindrical face and the bracket. Because this stress is highly localized and the model contains idealize sharp edges for the washer, this stress concentration was not seen as a reason for concern; however, as a precautionary step, the washer was removed to reduce the stress concentration. Stress analysis was performed again on the revised bracket (a product of increased manufacturability) in order to verify the stress due to loading remained within the yield strength of the material. This analysis indicated the stresses experienced were over the material s yield strength; however, these concentrations were highly localized while the nominal stress within the model was significantly below the yield strength. This being the case, and because this part was to be loaded statically, the engineers decided to approve the part for manufacturing. The constraints and results are shown in Figure 7 and Figure 8 as well as Table 4 below. Figure 6: Constraints placed on front articulating joint
34 Table 3: Tabulated mesh refinement iteration with resulting stress Analysis Iterations Maximum Stress (von Mises) (psi) Estimated Mesh Size (in) Stress Error 1 15906.39 0.0533 2 38468.75 0.0141 0.5865 3 38407.09 0.0124 0.0016 Figure 7: Constraints applied for stress analysis
35 Figure 8: Location and magnitude of maximum stress Table 4: Tabulated mesh refinement iteration with resulting stress Analysis Iterations Maximum Stress (von Mises) (psi) Estimated Mesh Size (in) Stress Error 1 15860.16 0.1075 2 17519.34 0.0678 0.095 3 31917.54 0.0345 0.451 4 40993.07 0.0158 0.221 5 60143.19 0.0085 0.318 6 82598.12 0.0047 0.272