Failure of a part is not an option, when the part in question is a piston used in the engine of a drag boat racer. Any failure often leads to an out of control boat and hence the possibility of severe human casualties. Failure of a part can be due to many aspects, such as wear, fatigue or even due to an out of tolerance part slipping through the inspection net in a manufacturing cell.
Racing Pistons Inc. has been manufacturing pistons for racing applications for 40 years and have achieved the highest quality levels in their field. Currently on their manufacturing cell of their racing piston production line, there is a sub-cell, which performs the functions of a quality control system.
The quality control system has two aspects to it, firstly a visual inspection system, this is designed to test each part and ensure every required specification is within tolerance. A piston not within the desired tolerances would be defined as a defective part and is removed from the conveyor belt before the next stage of the manufacturing process.
The second part of the quality control system is the removal of the piston from the conveyer belt. It is proposed to fabricate a one-degree of freedom robotic arm and control system to move the piston from the conveyor belt, to a recycling bin. The exact location of the piston is identified by the use of the visual inspection system.
It is this aspect of the problem that Lunar Consulting would like to submit tender for. This project proposal has been developed to highlight areas of the design process to be included. Lunar Consulting has achieved high standards of design; project management and commissioning on all projects worked on in the past and believes this project to be no different.
A prototype design is intended to be developed and tested on a test bed, which recreates the conditions of the cell used in the quality control system for the manufacture of the pistons. This is to ensure that disruption to the manufacturing process is kept to a minimum. For the purposes of this simulation several assumptions and simplifications have to be made. These are as follows:
Photographic representation of the test bed is shown in appendix B, figure 1. This series of photographs illustrates the test bed and allows easy interpretation of the problem.
2.1 Problem Definition
To design and fabricate a prototype (single unit only) robotic arm and develop a control system to be used to transfer a metal disk from several defined positions to a hole within the test cell. This procedure is performed using an electromagnet attached to one end of the robotic arm; the other end of the robotic arm is attached to a servomotor, which allows angular rotation along the arc of travel to the end of the robotic arm. The servomotor is to be controlled using a control system, which should be designed using Simulink.
2.2 Project Objectives:
This problem has three main objectives:
2.3 Scope and Limits
In order to simply the design process several constraints are applied to simplify the design process. They mainly fall into the three categories that are listed below. The lists situated below are not exclusive or exhaustive, but a detailed description of all the physical dimensions is included in appendix B1.
Physical Constraints: This aspect of the constraints is associated with the robot arm and incorporates many aspects including material selection, load applications, and characteristics, joining techniques, mounting options and the physical dimensions of the robot arm.
· The robotic arm structure is constructed using brass-brazing rod.
· The brazing rod can have a diameter of 1/16, 3/32 or 1/8 only.
· The brass-brazing rod can only be joined together by the use of 50/50 solder.
· The base of the robotic arm must be able to be mounted on the flange on the existing servomotor. The servomotor flange has an outside diameter of 4 and has 8 holes on a 3 bolt circle diameter.
· The robotic arm must be attached to the servomotor by a minimum of three Ό-20NC Ύ mounting bolts (with appropriate washers).
· The end of the robotic arm should have an arrangement to facilitate the attachment of the electromagnet with the use of a single bolt into a 5/16-18NC, ½ deep max and Ό deep min mounting hole.
Control Constraints: This aspect of the constraints deals with how the robotic arm should move with parameters such as period of travel, accuracy, reliability, and tolerances.
· A PC running the program Simulink should be used for control of the robotic arm.
· The controller must allow the metal disk to be dropped into the recycling hole 100% of the time. This process should be achieved efficiently in terms of time and cost.
· The controller must allow for different starting positions of the arm. It must be able to rotate through an angle of to. The precise angle of rotation is determined by the use of an encoder.
· The robotic arm must not overshoot the recycle slot by more than
System Constraints: This aspect of the constraints deals with the system as a whole.
The simplifications that are highlighted in section 1.0 are the main constraints to the system. In addition to these constraints several more are included:
· The robotic arm must not interfere with the protective Lexan shield situated around the test cell. The shield is situated at a radius of 24.
· The electromagnet must be in close proximity to the metal disk. This distance should within 1/8.
· The metal disk is placed on a stand, which is a distance of 0.375 above the height of the test bed.
· The face of the servomotor is a distance of 1.625 above the height of the test bed.
· The hole for dropping the metal disk in (recycle bin) is raised a distance of 0.25 above the height of the test bed, it also has a diameter of 2.75.
The component specifications, material properties of the rod and photographic representations of the test cell are included in Appendix B2 and B3.
2.4 Testing Procedure: This section is included in the background as it is important to be aware of the procedure for evaluating the design when considering the project structure. For this project several testing procedures will be used as detailed below.
Static Structural Integrity test: Successful completion of this test requires the structure to withstand the loads as described in project objective #1.
At the time of the structural integrity test the robotic arm will also be weighed and the weight recorded.
Dynamic test: This test requires both the use of the control system and robotic arm structure. The test consists of the metal disk being placed within an angular range of to at a distance of 20 from the centre of the servomotor. The test requires the robotic arm and control system to move the metal disk via angular rotation to the hole in the base of the test bed several times at different angular starting points. A successful test is defined as a demonstration of acceptable control of the robotic arm.
The design of the robot arm will encompass many differing technical areas and will require engineering theory and expertise to be put into practice. Initially the structural design of the robotic arm will be focused on, this involves many aspects of design practice and research. Once the robotic arm has been designed and constructed the system for controlling the robotic arm will be focused on.
3.2 Project Structure:
Project Initialization: In order to accomplish this project efficiently a breakdown of tasks must be defined in a logical order. This logical order is detailed below and also forms the basis of the project management section of this proposal. This is the approach to be used by Lunar Consulting to solve the design problem.
Strategy Development: Initially a discussion needs to be held to decide what direction the project will proceed. This is important as without a plan on how to tackle the problem the team will not be successful. This is also to establish rules for the team and a reporting structure; more detail on this section is included in the project management section of the report.
Preliminary Research: One of the first steps is to study the problem in depth to get a scope of what it entails and to make sure all design constraints and criteria are clearly understood. This might involve further correspondence with the client.
The purpose of the preliminary research is to kick-start the project and to familiarize the engineering team with design principles, and the engineering challenges that will be encountered. The first part of the research will involve gathering data on material properties of the solder and the brass rods, which will help in deciding suitable truss design. The next part of the research will be focused on truss designs, analysis and the engineering involved. This will lead to concepts of various structural designs that meet the specifications.
Basic Design Development: Initially several differing designs will be developed; this is to explore every aspect of the design and to expose a design with the potential to fulfill and solve the problem. This process will be achieved with the use of simple wire frame AutoCAD drawings. This method allows designs to be efficiently realized whilst giving a 3D impression of what the structure will look like. Simplified truss and beam calculations will be performed to evaluate the different designs.
Design Selection: This section will weigh the advantages and disadvantage of the various design concepts obtained from the basic design development. A consensus will be reached by all team engineers on the best design based on manufacturing, design functionality, design complexity, and strength to weight ratio.
Material Tests: In order to proceed into the detail calculations of the proposed design, the material properties of the brass rods and the solder joints will need to be verified. In order to verify the properties, tensile tests will be conducted on the brass rod to determine the tensile strength, yield stress, and ultimate tensile strength. Hardness tests will also be performed on the brass rod to test uniformity. A tensile test will also be performed on the solder to determine its defining characteristics. These tests are crucial in deciding the factor of safety to be used and hence are extremely important in the design process. This process will be conducted in the University of Waterloos material testing facilities.
Design Verification: The robotic arm selected in the design development section will then have a detailed analysis performed on it in order to optimize the design and to ensure that it meets all structural and safety concerns.
The detailed design verification will consist of stress, buckling, and deflection calculations, which will be performed to assess the structural integrity and functionality of the design. These calculations will take into account the joining techniques and apply a factor of safety. Based on these results the base design will be modified to optimize certain parameters of the design. The methods used for the verification are highlighted in appendix B4, it also includes the assumptions used in the different processes such as assuming pin jointed structures for the truss structure.
CAD Models/Engineering drawings: Once the optimal design has been finalized and verified, a 3D design representation can be produced using the AutoCAD system. From these CAD models component drawings and assembly drawings will be compiled. These will form the basis for manufacturing.
Manufacturing: This stage of the process is probably one of the most critical aspects of the project, failure to produce a well made prototype will result in a critical failure during the testing procedure, therefore in order to increase the reliability of the manufacturing process several procedures have been put in place. These are:
· Practice on the welding and cutting processes, to ensure that skills in those areas have been developed before the manufacture of the arm is too proceed.
· Breakdown of the manufacturing process into small chunks. This process can only be established once the final design is completed. This ensures that is manufacturing errors are present then only small sections of the design have to be re-manufactured.
· Visual inspection of solders at every stage, independently verified by members of the group not involved with the manufacturing process.
Static Testing: This process will be in two-sub stages, first internal team testing will be conducted to ensure that the design meets specifications, this will be performed using an identical procedure to that of the formal static structural testing. Measurements will be made on the deflections of the robot arm and readings will be compared with theoretical predictions. This stage will be performed in advance of the formal static structural testing in order to allow sufficient time for any necessary re-designs.
At this stage minor dynamic tests will also be conducted such as attaching the arm to the test station and manually adjusting the rotation of the arm to simulate loading under dynamic conditions, this is to establish that the design will pass both the static and dynamic tests at an early stage and give confidence to the teams solution to the design problem.
Structural Design Report: This stage of the project will be required for submission during the static testing; it will contain the following detail.
· Initial concept design, including basic calculations.
· Documentation of the design selection processes
· Development of the final design including selection reasoning.
· Analytical verification of the final design using the methods highlight in appendix B.
· Finite Element Models of the final design including differing stress distributions such as Von Mises stress, and displacement predictions.
· Results of internal structural tests including static and dynamic testing.
· Comparison between experimental and theoretical results.
· Full conclusion on the design process.
Control System Development:This stage of the process is conducted as the process is described in appendix B5, in summary it involves the procedure of determining the characteristics of a real system, then modeling the real system to provide an accurate representation of the real system. This model can then be used to optimize the control system to ensure that the fastest time possible for the transfer of the metal disk from one position to another can occur.
The control system will be developed using the computer program Simulink and several tests of the developed control system will be performed with the test rig to ensure the control system is efficiently optimized. The optimization process will be performed until it is no longer economically viable to optimize the control system.
Control Report: Control report will run along parallel lines to the structural report, it will include aspects such as:
· Results of experimental tests to establish the operating parameters
· Development of the linear model to effectively mimic the experimental results.
· Optimization of the linear model to include any non-linearitys that are present.
· Optimization of the control system to produce the most efficient transfer of the metal disk.
· Results of internal dynamic testing.
· Full conclusion on the development process.
Laboratory Facilities: The following facilities are available for the teams use.
All facilities form part of the faculty of Engineering at the University of Waterloo, a photographic representation of the facilities is shown in appendix B3.
4.1 Company Expertise:
Lunar Industries is a multinational company which operates on all continents of the world, however mainly their operations are centred around Europe and North America. The company was founded by three graduates of the University of Waterloo and one graduate from the University of Birmingham (England). The four graduates all met whilst the Englishman was an undergraduate student on exchange at the University Of Waterloo. They decided to found there own consulting company based in Waterloo, once all four graduates had a chance to experience real world engineering situations.
The company specializes in small industrial problems and because of the broad range of expertise of the company members, many different types of project can be facilitated, this includes the design of robotic arm end effectors and accompanying control systems.
Internal: Luna Industries have assigned a four man team to this project, they are
Member 4, Member 3, Member 2 and Member 1. Each member brings different degrees of specialization to the project although all members have a broad basic expertise in aspects such as structural design, control development and project management. Following is a list of the specializations for each member of the team.
Project Coordinator: Member 1
Structural Design: Member 3 and Member 2
Control Specialists: Member 4 and Member 1
Material Specialist: Member 4
Finite Element Specialist: Member 1
Appendix A includes a copy of the resumes for members of the design team, for example for the role of project manager Member 1 was selected for the role, as is evident from his resume he has experience in large scale project management, he is also selected as the finite element specialist after studying finite element methods under Prof. M.Worswick at the University of Waterloo.
Member 4 used to work for Johnson Controls Limited developing and optimizing control systems including the calibration of a new special arm system. This makes him ideal to fill the role of control specialist.
Member 3and Member 2 are both as shown by there resumes proven design experts with years of experience in the design of complex structures. They also have a proven knowledge of Cad drafting techniques and experience in the manufacturing industries.
The reasons stated above confirm that the team selected has a proven record of designing and manufacturing many different types of system, for these reasons it is evident to see that if they were selected for realization of the project then the project would be in safe hands and completion would be almost guaranteed.
Consultants: In addition to the in house expertise at our disposal there are several consultants whose valued expertise may be required to be used during completion of the project. They are:
Physical/Design: Professor Schneider, Professor Worswick,Jose Jimbert ,Professor Cronin
Controls: Prof Huisson, Arash Narimani, Omeir Ansari, Shahab Hasanzadeh, Imran Yousaf
Manufacturing: Clarence Wallace
4.2 Project Management:
Breakdown: The project will be broken down in smaller more manageable chunks in order to maintain focus and maintain the essential impetus and vitality needed to ensure the project is completed on-time and on budget. Initially the project will be broken down into two sections; the structural design and the control design. They will then be broken down into even smaller sections until an individual task is reached.
Each section will have a task leader (as defined above) who will be responsible for the completion of their task, and other team members will be assigned to assist accordingly. Table 1 shows the member of the team responsible for each stage of the total project. Also shown in figure 6, in appendix B6 is a Gantt chart illustrating the times taken for each stage. The timings for this section are based on the external report deadlines and the intuitive estimation of the time taken for each section to be completed, this chart was established during the project initialization phase and will represent a time line for section completion.
Table 1: Task Assignments
Project Manager: The project managers role is that of a facilitator for the completion of the project. In particular his role and tasks will be as follows:
· Organise meetings
· Producer of weekly project review report
· Production of timelines, task lists, and completion dates
· Monitoring and responsibility for adherence to timelines
· Compilation of major reports
Reporting Systems: Two reporting systems will be used during the duration of the project. They are listed in the following two sections.
External reports: These reports are submitted as part of the objectives of the project and should include all relevant documentation of the project. The project and submission is listed below:
a) Design Proposal - Due date: 31st January 2003
b) Static testing with static design report - Due date: 28th February 2003
c) Dynamic testing with control systems report - Due date: 28th March 2003
Internal reporting: This will take the form of several different type of reporting systems, detailed below.
Team meetings: Two weekly meeting are proposed to occur. The composition is as follows:
· Project design meeting: This meeting is when the main points of the project are discussed by the members of the team and important decisions are made by all members of the team. Important decisions could include a discussion of increasing the weight of the structure and hence structural integrity, and characteristics such as acceleration and response of the robotic arm. The production of other small reports will be discussed during this meeting.
· Progress review meeting :As many aspects of the project can change within a weekly period a progress review meeting will be arranged on a weekly basis. The purpose of this meeting is to ensure that tasks that are being pursued and to ensure accurate communication within the team.
The chair of both these meetings will be held by a different member of the team each week, this is to ensure that there is no one member of the team that stifles the development of the project.
Reporting Documentation: After each project review meeting a minor report will be produced illustrating the development of the project. This is essentially the minutes of the meetings and is arranged to ensure all members of the team have knowledge of what is happening with the project and what each other member of the team is in the process of achieving. It will be broken down as follows:
· Accurate position of project development in relation to initial plan.
· Weekly tasks to be achieved
· Description of current tasks being performed, including member of team assigned to the task and the deadline for submission to the team.
· Strategic development of the project.
· Deviations from the strategic plan and why.
· Date and time of next review meeting.
4.4 Project Completion: Completion of the project will occur once the conclusion of the dynamic testing has occurred and the submission of the control systems report has been achieved.
Budgeting is an essential ingredient in any successful project, both in terms of finance and resources. For this purpose a budget has been drawn up for the aspects included in this proposal. The budget shown below in table 2 is broken down into stages based on the tasks highlighted in section 3.0. There is one main assumption with cost analysis, that material and associated costs are not included in the budget, these are provided free of charge.
Therefore the major costs associated with the project are labour costs and therefore in order to keep costs under control the timeline and stage completion dates must be regimentally adhered to.
Table 2: Cost Analysis Breakdown
3 levels of cost per hour were established, they are as follows
The responsibility for the budget lies with the project manager. Providing the project goals and constraints are not altered in any way, the cost total project costs remain accurate.
B.1 Test Cell Specifications
Table 3 (a)(b)(c)(d)(e): Electromagnet, disk and support specifications
B.2 Material Specifications:
A sample of the brazing rod to be used for the project was examined using an energy dispersion X-ray analysis. It was found to have the following composition.
Table 4: Material composition (by weight)
Youngs modulus: 105GPa
Also taken was an average value for the yield strength for the three different diameters of rod. These are summarized below:
Table 5: Average yield stress of differing diameter rods
Individual values used to calculate the mean yield strength had a percentage deviation of 5% from the average yield stress. All of the results were taken from the same section of wire therefore the identification of wire to wire variation is not possible.
B.3 Photographic Representations
Figures 1(a)(b)(c)(d): Photographic Representation of the Idealized Test Cell
Figure 2: Workshop facilities used for the production of the prototype
B.4 Design Verification Methods
The following procedures are used in the structural design verification. Two main methods will be used however each method is broken down into different sub methods of differing complexity complexity. The main assumptions associated with this process are as follows:
· The soldered joints can be assumed to be pin joints, this is providing the centerlines of the members are concurrent at the joint.
· The externally applied forces are applied at the pin-joint.
Simple 2D truss calculations
Method: This procedure is very simplistic in nature and therefore can be performed, easily and quickly. It basically involves resolving forces into components in two directions then solving for these forces, the stresses and strains in the structure can then be calculated. This process will be used for initial design calculations and to prove aspects of the design, during the actual designing stage.
Buckling: This will be an important aspect of the verification stage. From Eulers formula, defined as where the terms are defined as:
Critical load for which the strut will buckle. E= modulus of elasticity, I= 2nd moment of area of strut about the neutral axis and L equals the length squared. This equation gives an important design criteria, because in order to increase the critical load , the length of each strut must be kept to a minimum.
Fatigue: Cyclic loading of the unit could be quite high, this is due to the repeated cycles of picking up the metal disk and releasing it. Therefore as part of the design verification fatigue should be taken into account, this should be performed by the use of the soderberg stress criterion and stress diagram which illustrates where fatigue failure is likely or unlikely.
3D Truss forces calculation using an analytical approach
· Structure is a simple 3D pin-jointed structure therefore no torque or bending is transmitted via the joints.
· Joint are ideal, i.e. perfectly joined structures.
· Each strut has negligible mass, compared to the force applied.
Method: This process uses the finite element approach of assembling a stiffness matrix for the entire structure then solving via the use of a mathematical solver such as MathCad. The advantage of this method is that the design can be optimized to produce a viable solution efficiently as the structures stiffness matrix and properties can be changed very easily to find the best solution. This method follows a 6 step process, each step is highlighted below, and however as the process is complex only the major aspects of each stage are included :
1) Discretize: This involves the breakdown of the robotic arm into its individual components (bar elements) and then recasting element forces and displacements in axial and perpendicular to each truss.
2) Interpolation: This involves introducing an interpolation function along the length of the bar element. From this linear interpolation, shape functions can be derived for the displacement of the bar element.
3) Element Properties: Using displacement shape functions it is possible to obtain strains and hence stresses in the bar element. These stresses can be placed in a matrix which relates the stress in the bar to the displacement of the ends of the bar element.
4) Assembly: This involves assembling the derived equations from each individual bar element into a series of global stiffness equations. These equations are all related to each other and hence will be placed in matrix form.
5) Application of Boundary conditions and Solve: This process applies boundary conditions, such as loading requirements to the global equations. These equations are then solved to determine the deflections and forces in each member.
6) Post process: Once the forces and displacements are calculated the strains and stresses in each member can be calculated.
The above process is extensive and exhaustive therefore it will be used as verification and optimization of our final design. Also because of the assumptions made safety factors will have to be applied to the final results.
3D Finite element models
a) Using Autocad: The choice of program for the 3-dimensional modeling of the robotic arm is Autocad, this is because there is easy access to the program and it also very easy to produce accurate 3D models using the process. AutoCAD Mechanical Desktop also has a simplistic finite element solver which will be used to solve for a solution to the problem. The following procedure will be followed:
· The designed truss structure will be inputted into the solver.
· Boundary conditions and loading requirements will be then added to the virtual model.
· The solver uses tetrahedral 3Delements to solve for the solution.
· Once a solution is obtained, then contour plots of Von Mises stress and deflection will be obtained.
The solver is very simplistic, which leads to solutions for complex problems taking a long time to perform, however because of the simplistic nature of the solver, it can perform solutions to very simple problems in a relatively short space of time. An example of a solution to a simple joint design is shown below.
Figure 3: Diagrammatic representation of AutoCAD finite element solution.
b) Using IDEAS: The following procedure shows an example of the method of using ideas 3Dfinite element solver to solve for the stresses in the object. It shows the modeling of a previous completed project which was a tubular component for use in the aeronautical industry. Although the geometry is different to that of the robotic arm, the procedure for finite element modeling is almost identical.
· Firstly a virtual robotic arm model is inputted into the ideas program.
· Boundary conditions, loading requirements and planes of symmetry are added to the virtual robotic arm model.
· The object is meshed using thin shell elements, an assumption is made that the cross section of the struts does not change.
· Finally the program solves the virtual model to establish the stress distribution and point deflection of the model.
· Plots of Von Mises stress, shear stress, deflection and many other aspects can be plotted.
The two contour plots of Von Mises Stress for the tubular structure are shown below to give examples of how the system can be used. This procedure highlights areas of weakness and adds verification to the manual calculation.
Figure 4: Contour Plot of Von Mises Stress Distribution within tubular structure.
B.5 Control System Design Methods
The control system consists equipment as highlighted in figure 5. The planetary gear head of the DC motor is connected to the test cell and controls the angular rotation of the robotic arm.
Figure 5: Diagrammatic representation of the control system setup
Controller design: The controller design process will consist of the following sequence of 7 steps:
1. Creation of a linear model of the system based on available data.
2. Evaluation of the real system using experimental testing procedures. These experimental procedures include the investigation of:
3. Refine the parameters of the linear model so that it best reflects the experimentally measured behavior.
4. Design a controller for the desired response using the refined linear model.
5. Reformulate the model based on known (or suspected) non-linearity in the system and adjustment of the model parameters so that the non-linear model exhibits the best match to the experimentally measured behavior.
6. Test of the controller designed in step 4 on the reformulated model, and tuning of the controller parameters / gains for the desired system response.
7. Application of the controller designed in step 6 to the real system, and experimentally fine-tune the controller gains.
B.6 Additional Project Management
Figure 6: Gantt chart illustrating section completion schedule.