Automated Work Cell
|Render of Final Workcell Design|
|Team Name||Do-All Robotics|
|Duration||Summer - Fall 2016|
|Faculty Advisers||Dr. Steven Beyerlein|
|Mentor||Dr. Mike Maughan|
Team Do-All Robotics was tasked to improve the existing Robotic Workcell located in the Senior Design Suite. This project includes a second generation design of the workcell itself with hopes of making a sturdier, more modular work-cell that is able to handle tasks of higher precision and of a wider variety. The ME Department also assigned the team to create a standalone instruction manual that incoming teams could use as a resource to better understand the full capabilities of the DENSO Robotic Arm and be able to program the workcell to meet a wide variety of manufacturing needs
In Spring 2013 The Boeing Company donated DENSO robotic arms to The University of Idaho College of Engineering. That fall Mechanical and Electrical Engineering students, as a part of Team Roboshow, were tasked to learn basic programming of the robotic arms and create a work-cell for the arms. Their work-cell incorporated multiple safety features as well as a clear poly-carbonate enclosure for public demonstration. The team was able to successfully program the robot to use a dry-erase marker to create logos and patterns on a white board.
The following year another design team, known as Team Vandalbot, was formed to design a manufacturing process which could assemble various rivet and nut-plate assemblies in order to automate a repetitive task currently done by factory workers. The team was able to demonstrate this process and provide documentation for future teams.
Previous Design Issues
There are a few critical problems with the existing work-cell design. When the robotic arm operates at high speeds the cell tends to wobble which decreases precision in manufacturing. The way the current work-cell is set up only allows for a limited number of configurations when developing a manufacturing process. Another problem facing the robotic arm work-cell is lack of documentation on operating the DENSO arms using included peripherals and developing programs using WINCAPS software. Previous teams' documentation includes many references to manuals that are difficult to find or include non-critical information.
The Mechanical Engineering Department at the University of Idaho requires certain updates to be made to their current Robotic Arm Work-cell. A 2nd generation work-cell needs to be developed to improve the cell’s flexibility in manufacturing capabilities such that it: incorporates a modular design, eliminates movement & vibration under load, and is able to include a second robotic arm. Standalone documentation of robot setup and operation should be compiled for future students.
Project Design Goals
1. Create a workcell for an automated manufacturing process with a Denso robotic arm.
- - Improve upon the stability of the workcell to limit vibrations and motion that might cause errors during manufacturing.
- - Create a modular workcell to allow for any kind of work surface or manufacturing process.
2. Incorporate a second Denso robotic arm into the work cell.
- - Have both arms be controlled by one program.
- - Have both arms work on the same process simultaneously without interfering with each other.
3. Explore the limits and possibilities of manufacturing processes with two Denso robotic arms.
- - See if the addition of a second arm adds any value to the process or just makes the process needlessly complicated.
|Pedestal||Maximum deflection in x/y/z/rotation||0.01||in||1|
|Pedestal weight w/out arm or brain||130||lb||2|
|Does not tip during operation||-||-||1|
|Resonance frequency not achieved during operation||-||-||2|
|Tables||Minimize use of 8020 Aluminum||≤300||ft||2|
|Table Surface height is ergonomic||36||in||1|
|Time to add/remove table from cell (reconfiguration)||2||min||2|
|Enclosure||Standard size||12 x 12||ft||2|
|Walking space in cell||≥2||ft||1|
|e-Gard cuts power to bot (safety lockout)||-||-||1|
|Instruction Manual||Length of manual||20 - 25||pages||2|
|Target user||Junior||class standing||2|
|Extensive use of pictures and diagrams||-||-||2|
|Deliberate progression through sections||-||-||1|
|FEA on Loading Modes|
|FEA on worst case scenario loading modes to ensure our pedestal would not deflect or tip during use.|
One of the main issues with the original workcell is that the method in which the robot was mounted in the workcell was not effective in keeping it steady. During operation at high speeds the robot would cause the base to rock back an forth which caused repeatably issues and made precision operation almost impossible. In order to fix this the DENSO installation manual was consulted and used to develop a pedestal out of ¼ inch square steel tubing. To ensure the pedestal would not move or deflect during use a finite element analysis was done considering the worst case torque loading from the arm swinging around and the worst case vertical loading where a mass on the arm could cause it to tip. Pictured to the left are the results of the FEA done in SolidWorks.
The maximum torsional deflection that the pedestal will see is 0.114mm. This is the linear distance from the original corner to the new position of the corner as it twists.
The maximum vertical deflection that the pedestal will see is 0.013mm.
After the pedestal was assembled, 8020 T-Slot aluminum was fastened to the outside edges so work surfaces could be attached. This T-Slot aluminum was used for most of the workcell construction.
Modular Table Work Surface
The new workcell needed to be reconfigured easily so future design groups would be able to change the work surface to best fit their needs for their particular project. A modular work surface was designed so that 24"x24" square tables could be added or subtracted in a grid to create the optimal working area.
Each work surface is 36" in height with adjustable feet to account for imperfections in the floor. This surface height has optimal ergonomics for workers to interact with the workcell for long periods of time in a typical manufacturing setting. The tables have an 8020 T-Slot aluminum border around them so they can be attached to other tables an so accessories can be added to the work surface.
For the enclosure space a permanent wall was created around the robot and its work surface. Located in GJ 108, this enclosure is made from 80/20 aluminum tubing and clear polycarbonate to allow the robots to be viewed from the outside of the cell. There are two doors, one reuses the electronic lockout (e-gard) from the old workcell and the other is manually locked out from the inside in case of an electronic failure.
The workcell comes equipped with a shelf unit for the RC7M controllers, an electrical panel with terminal blocks for any sort of electrical connection a project may require, and a rolling cart for transporting material in and out of the cell.
An instruction manual was being developed to accompany the redesigned workcell. Previous documentation for the workcell was lengthy and contained many references to outside sources that were difficult to find or more detailed than need be. The new workcell manual that has been developed is much shorter and will not require outside sources to teach the reader use of the workcell. This current version was completed in September of 2016.
Instruction Manual Main Sections
I. Introduction with Project History
II. Safety Requirements
III. Teach Pendant Operation
IV. WINCAPS Software & Programming
For the Fall semester, after completion of the workcell, three programs were developed to each display the robots' different capabilities. Basic operation and movement, I/O with an air supply, as well as communication between the controllers for two of the arms. A quick change end-effector system was also developed to make switching between each of these projects quicker and easier, each of the end-effectors used for these projects can be changed in only a few minutes.
Drawing (Single Robot)
For this project the robot was programmed to draw the University of Idaho "I" on a clear piece of polycarbonate using a dry-erase marker. A video of the robot doing this can be found here. For this the team needed to develop a new end-effector that could hold a dry-erase marker and keep it spring loaded much like a human hand does when writing. This project was completed in September 2016.
Much like Speed Stacking this project was developed to control the robot so it could stack cups as quickly as possible then take them down. A gripper end-effector was developed for this project using a Trimantec HFZ series gripper and 3D printed claws shaped for picking up one or more cups. The robot's air solenoids were used to control the gripper. Unfortunately the cable that was needed to control the I/O for the gripper was not delivered soon enough for this project to be completed at this point in time.
Drawing (Double Robot)
The purpose of this project was to demonstrate communication between the robots and show they could work together to create more complex processes. The robot's were programmed to be able to draw one of the enemies from Space Invaders. One robot would hold the pen while the other held a drawing surface. First the robot with the pen would draw part of the image, then the other robot would move the drawing surface around to finish. This project wasn't able to be completed due to difficulties in wiring both controllers to the safety lockout (the robots are not able operate if they are not connected properly to the safety lockout).
|Team "Do-All Robotics"|
|Ryan Chapman||I am a senior studying Mechanical Engineering and planning to graduate in December of 2016. Tinkering with cars in my driveway in high school is what got me interested in engineering initially. After graduation I plan to seek out a career in Aerospace Engineering and obtain my PE certification as soon as I am eligible. Some other hobbies I enjoy doing are camping, boating, trying exotic foods, and talking about politics with people that disagree with me.||Mechanical Engineering|
|Jacob Dostie||I have moved around quite a bit, from New Jersey to Oahu, however I would say my home town is Port Angeles Washington. The reason I am studying Mechanical Engineering is because I have always had a passion to learn how stuff works and why. I am always looking to try and challenge myself be it academically or physically. In the future I plan on getting my P.E., then maybe even getting a Masters or even a Ph.D. Some things that I enjoy doing are working out, cooking, building/disassembling things, playing magic and video games.||Mechanical Engineering|
|Adam Grebil||I am originally from Boise Idaho and engineering has always peek my interests since I could rememeber. I have always been taking things apart and figuring out how they work. I am currently finishing up my Masters in Mechanical Engineering and excited to see what the future will bring. I enjoy everything outdoors and working on cars.||Mechanical Engineering|
|Bridger Hopkins||I am originally form Grand Rapids, Minnesota. I have wanted to be a mechanical engineer for as far back as I can remember. My first two years of school were done at Itasca Community College before transferring to University of Idaho to finish up my degree. I enjoy spending time with family, gardening, throwing pottery, and a good movie.||Mechanical Engineering|