Project Number: P18310

Transcription

1 Multidisciplinary Senior Design Conference Kate Gleason College of Engineering Rochester Institute of Technology Rochester, New York Project Number: P18310 TUNED VIBRATION ABSORBER DISPLAY Stephen Giordano Electrical Engineering Connor Jensen Mechanical Engineering Nathan Ehnot Mechanical Engineering Kyle Kelly Electrical Engineering Shayne Hollands Mechanical Engineering Malik Wallace Computer Engineering ABSTRACT LORD Corporation manufactures Tuned Vibration Absorbers (TVAs) for a variety of industries. TVAs reduce vibrations and decrease wear and tear on machine components. LORD Corporation has multiple trade show displays, however, none of them are able to show the difference a TVA can make on a moving system. LORD requested a TVA trade show display to highlight the difference a TVA can make when compared to an identical system. The trade show display will be accepted by the customer complete with all included materials for correct and safe operation by nontechnical personnel. The display provides an eye-catching product to assist LORD s demonstrations at trade shows in the future. This paper will outline the model, design, and manufacturing of the system. Test plans and data collection ensure successful demonstration of the final trade show display. BACKGROUND The project is to develop and build a trade show display. The project will fulfil the deliverables set by the customer in figure 4. The project will educate potential customers about the elastomer technology in a visually aesthetic format. To understand how to engage a potential customer. The team attended a local trade show to view current trade show displays. Sale representatives were interviewed about the displays and presentation skills. Afterwards, the team came together to brainstorm designs for the project. Customer and engineering requirements were developed by the sponsor and team. The engineering requirements serve to determine how successful the final product will be. Below is figure 4 and 5 the customer and engineering requirements are shown. It is important to review the stakeholders and the effect success will have on the project. The sponsor will benefit from the display. The project also represents RIT, the Senior Design program, and our guide. The team always worked with respect, confidence, and a desire to learn and have fun. SYSTEM MODEL Design Choice - Technical Feasibility The major design decisions, such as the masses of the systems, the mass of the TVA and the spring and damping constants of the mounts are all based on TVA design theory. The fundamental theory behind the operation and design of a tuned vibration absorber is the mass spring damper equation: Figure 1: Fundamental mass spring damper equation Determining Motor Size and Piston Force For the TVA to function properly and mitigate the desired vibration, the resonating frequency of the TVA will match the frequency of the vibration that will be damped. The design utilizes a motor driven piston to generate the

2 vibratory force in the systems. The first step in designing the system is to determine the force, F in figure 1, which the piston will apply to the system. The force output of the oscillating piston will be approximated as a sine curve. The mass of the piston and the rotational speed of the motor at the target frequency of 15 Hz is used to determine the force that the moving piston is imparting on the system. The proper motor for the system was selected from the required torque output of the motor. A full derivation of the process for solving for this force is available in the System Model Derivation document on the team s EDGE site. The calculations were programmed into Matlab to simulate the changing piston force over time as the piston oscillated. Figure 2 shows a Matlab graph of the piston force and the required motor torque over time. The piston force is used as an input to the mass spring damper equation to accurately model the displacement of the two systems and the TVA. A full derivation of the equations is available in the System Model Derivation document on EDGE. The model can vary the design parameters and view the output vibration. The CAD model is used to approximate the required mass of the two systems. The final estimated weight is higher and was used as an input into Matlab. The increased weight ensured the system will be below the target weight. Ballast masses were used to reach the proper mass for operation. Figure 3 shows the displacement output of the two systems with the final selected parameters. The graph on the left shows a high TVA displacement, but no displacement in the system, meaning the TVA is completely mitigating all vibrations. The graph on the right shows the vibration of the system without a TVA. Figure 2: Piston force (left) and motor torque (right) over time for each system Figure 3: Comparison of the TVA system and non-tva system displacements Determining LORD Isolator Mounts LORD vibration isolator mounts were chosen based off the selected spring and damping constants from the Matlab simulation. The mount parameters are as close, but are not exact. Additional tuning abilities are added into the design: frequency tuning through the GUI, TVA mass adjustment via steel plates, and piston force adjustment via a

3 Proceedings of the Multi-Disciplinary Senior Design Conference Page 3 slot on the rotating wheel. The modifications allow the system to be tuned and compensate for the slight variation in the given parameters. Choosing Materials Material selection for the systems, TVA and electronics box are based on the projected weights from the CAD models. Aluminum was selected for all MECE system components. Aluminum is light, strong, and least expensive metal available. Based on the required mass, the TVA will be made of steel, and machined to size. The electrical box was initially designed using aluminum, but after reviewing the CAD model the weight was determined to be too high. Instead ABS plastic was used, which is lighter and less expensive than aluminum. The ABS electrical box will be assembled with high strength ABS epoxy. Multiple strength tests on the glued joints were completed to ensure the joints were strong enough to withstand vibration. More information on this is available in the testing section. Figure 5: Engineering Requirements Figure 4: Customer Requirements SYSTEM DEVELOPMENT Sub-systems were developed to separate the project into manageable sections. The project was broken into mechanical, electrical, and GUI sub-systems. Shown below in Figure 6 is the system overview block diagram. The diagram shows how the isolated sub-systems come together and how each piece is interconnected.

4 Figure 6: System Overview Block Diagram MECHANICAL DESIGN DEVELOPMENT Key design features were set by the given customer requirements: two vibrating systems, easy to handle and ship, attract customers to booth, and be safe. The provided customer needs are condensed into measurable engineering requirements: Rugged design, guarded moving parts, aesthetically pleasing, and set up time less than 15 minutes. Engineering tests were developed written to validate each of the engineering requirements. Once the requirements were set the team began to brainstorm ideas for possible designs. A morphological analysis was used to study possible designs. The chosen evaluation criteria were dynamic display, under budget, easy setup time, rugged, easy to manage/troubleshoot, safe, energy efficient, complete in two semesters. The final three ideas were fleshed out and compared with a pugh chart to measure the pros and cons against each design. The initial design concept Figure 7 was reported to the customer. LORD gave feedback to mount the TVA on top of the system instead of in line with the force. LORD stressed that the spring system would be replaced with a provided LORD rubber mount. The TVA must be % of the system mass force. Final design is shown in Figure 8. Figure 7: Initial design concept Figure 8: Final design CAD drawing

5 Proceedings of the Multi-Disciplinary Senior Design Conference Page 5 Once the final design was completed parts to be purchased within the system were benchmarked. LORD mounts are analyzed for the size, shape, and spring coefficient. Metals, motors, and power sources are studied for range of strength, torque, and durability. Post-post LORD elastomer mounts are chosen to simplify the build of the system. Each purchased part was compared from multiple vendors for price and shipping time. The TVA size and weight is predicted from Matlab. The final 3D model is developed with Autodesk Inventor. Individual parts to be machined in house are carefully designed with datums and tolerances. Purchased parts required no drawings. If the purchased part required edits a 3D model and drawing is created. All engineering drawings can be found in the EDGE site. Final engineering drawings are reviewed by the team and RIT shop operators. Shop operators checked for design and manufacturing feasibility. The CAD total assembly model is used to predict the size, weight, and possible movement of the system. GD&T standards were utilized while creating the engineering drawings. Weight estimate of the ME system weight was handcalculated to be 7 pounds. Inventor predicted 6 pounds and the physical system weighed approximately 5 pounds. MECHANICAL BUILD DEVELOPMENT Following the engineering drawings, the ME systems were machined in house by the team. Aluminum is used for the the mechanical parts. Individual parts were machined from a 24x24x¼ inch aluminum sheet. Parts are cut to size by bandsaw and machined square, parallel, to allow for further machining. Parts are zeroed on the corner and machined to the correct size. Corresponding to the drawings aluminum parts are cut to desired length to the nearest thousandth of an inch. Large parts, including the base plate, are clamped to the mill and then machined per their respective drawings. If the drawing required it the holes are counter sunk and tapped. The shaft supports and connecting rods are very similar in shape. Both parts are machined to a rectangular shape and two holes are drilled for a press fit with the bearings. The holes for the bearings were machined by drilling a small hole in that location, then expanding the hole with an endmill by milling in an expanding circular pattern. This was accomplished by programming the mill to cut in a circular pattern. When finished, the hole was inches undersized to allow for a good press fit with the bearings. The piston housing support is machined out a large half circle. The mill cut a circular pattern and with each pass the cut went deeper until the desired shape is achieved. The circular piston wheels are the most difficult part to machine. The wheel hub was attached first and used to secure the wheel to the milling machine. This allows the part to be cut in a circular pattern without removing it from the mill. This ensured the wheel stayed centered and would spin true in the ME system without wobbling and causing wear issues. ELECRTICAL BUILD DEVELOPMENT To fulfill the customer and engineering requirements, the electrical system was designed with specific components. The main functionality of the electrical system is as follows: Provide a consistent (with the ability to change) rotation frequency on two separate systems in the form of a motor-shaft output, measure the acceleration each system is under with maximum accuracy, be self-contained, only requiring plug and play from the user, run for long periods of time over multiple days. From the engineering requirements in figure 4 and 5, each system is designed with the idea of being isolated from each other. This assured that there will not be any interference from one system on the other. A single system has the following requirements: An electric motor with rotary encoder built in for ease of access. Microcontroller with the ability to interface with a computer, and eventually a graphical user interface, and any other peripherals needed, accelerometer, motor controller etc. Motor controller with ability to easily interface with a microcontroller so the frequency can be adjusted from the computer. Accelerometer that can interface with the microcontroller to display system acceleration on computer. These above requirements led to the full design of a single system, which is easily doubled to allow for two identical vibration systems. The first component chosen was the motor. The geared, planetary DC brushed motor with rotary encoder was chosen due to its torque output at the desired speed 15 [Hz] = 900 [RPM] and built in rotary encoder. The torque speed characteristics of the motor chosen are found in figure 9 below.

6 Figure 9: Torque-Speed Characteristics For safety and protection, fuses are incorporated into the positive line for each motor. The specifications for the chosen motor is that when stalled, the motor draws 20 Amps. 15 Amps fuses are chosen to allow for current spikes which is less than a stall saving the motor from excess damage. If the motor stalls the current will increase until the fuse blows, shutting the system down. This protects the motor and whatever causes the stall, equipment malfunction, human interference, etc. The maximum current the motor controller could possibly need to drive is therefore 15 Amps, and to ensure sufficient supply, a 30 Amp motor driver was chosen. The driver is designed as a shield, meaning it can be plugged directly into the microcontroller. This allows for ease of installation, replacement, and reduces the amount of jumper wires going to the microcontroller. The Arduino MEGA development board was chosen as the microcontroller. This decision was made due to the large quantity of open source libraries and scripts available. The Mega contains many digital and analog pins, allowing capacity for any additional needs that may arise. The 13.8 V/17A power supply was chosen due to its high current output, small size and auxiliary 5V output. This removed the need for a separate 5V source, used in powering the LED s. The power supply was designed in a fashion that allowed it to be easily mounted in the electric drawer, granting access to the outside. This meant that the system can be set up and shut down without ever opening the drawer. Simply plug in the power cable to the wall and the USB to the computer, set up the GUI and flip the motor switches. The system will run as intended. As the design progressed, many other parameters and features were considered and eventually integrated in. For instance, the ability to switch a system on or off was not in the engineering requirements, but is an added benefit. This led to the design of individual on/off switches for each motor. A second feature added was the exhaust fans. The system was predicted to generate a significant amount of heat, especially after running for several hours. Cooling was deemed necessary for system longevity, and two fans were incorporated into the system. The fans reside in the electrical drawer. This allows for cooling without visually seeing the fan in the ME system. A third feature added is the LED s. Physical attraction is very important in a trade show environment. By incorporating LED s into the system attention will be pulled to the display. Adding in the LED s required bringing in a third microcontroller. The Adafruit Trinket was chosen to minimize the current demand and space required. SOFTWARE DEVELOPMENT The software was required to retrieve serial data from USB ports and graph it. The biggest constraint was being unable to purchase a license for the software, so programming in Matlab was not an option. QT was chosen, a crossplatform framework used to develop desktop applications. QT uses the language C++. To effectively display the data from the systems, dual modes are necessary, technical and non-technical displays. The software has been configured to switch between the two modes where the technical display shows plots of the accelerometer data and the frequency of the motor for each system as well as giving the user the ability to change the frequency if desired. The non-technical display shows the plot with accelerometer data and a visual readout of the frequency for each system.

7 Proceedings of the Multi-Disciplinary Senior Design Conference Page 7 Figure 10: GUI screenshot The GUI s main problems is lag. When plotting all four graphs the GUI would lag behind the ME system. The first solution was to make the GUI multi-threaded. Multi-threading allowed the receiving serial data process to be done on a different thread, while updating the GUI occurred on another thread. However, that did not fix the issue. Another bottleneck occurred when scaling the graph. Instead of constantly scaling all four graphs every insertion, the best solution was to scale every couple hundred of points. The problem was still not completely fixed. The final bottleneck occurred because the software constantly added data to the graphs and the objects holding the data would began to use too many resources. The solution was to make the software delete points as it scaled. Once those problems were fixed, the GUI had smooth operations. ENGINEERING TESTS Engineering tests shown in figure 11 are developed to ensure the project meets the engineering requirements shown in figure 4 and 5. Each test has set ideal limits for testing to reach and accomplish. Engineering requirements are color coded to easily understand. Green indicates that the test was passed satisfactorily, yellow signifies that the test marginally passed and is acceptable, red shows that the test failed. Status shows the current position of the test.

8 Figure 11: List of tests performed on the system BOM The bill of materials was tracked through a large excel sheet. The BOM recorded all parts, purchases, date ordered, shipping time, and calculated the cost. Parts are purchased during winter break to allow for shipping time. The project continued with all parts delivered on time. LORD provided a budget and the team managed to keep the project under the budget. The project saved approximately $716 of the budget for reserve. SYSTEM RESULTS While the system was stacked the vibration could not be mitigated. This resulted in a design change to separating the systems during trade shows. The systems still fit together for shipping and handling. The final result is shown below in figure 12. Mechanical system stands alone attached by an extension cord to the electrical box. The electrical box may sit on the table or may be placed on the floor. The mechanical system alone remains inside the 18x18 inch foot print constraint. The laptop will be paired with a larger screen for viewing. The GUI can provide multiple windows for different displays. The plexiglass case provides protection for the mechanical system and the customer. Figure 12: System set up for trade show viewing

9 Proceedings of the Multi-Disciplinary Senior Design Conference Page 9 CONCLUSIONS AND RECOMMENDATIONS In conclusion the system performs well. The vibration in the TVA system is reduced by 37% of the non-tva system. The vibration difference between the systems is visible from a distance of 4-5 feet away. The LORD logo is highlighted by red LED lights that breathe. The system was designed to capture the audience and draw them in. The moving parts and lights are an effective tool. The system will be used immediately by LORD in the manufacturing trade show in Huston Texas. The system will be stored in Erie, PA for trade show and educational use of LORD customer and employees. For future renditions it would be recommended to keep mechanical and electrical systems separate. Ensure each engineering drawing is checked and validated by a team member. Use CAD modeling to its fullest to ensure clearance of movement and parts. REFERENCES [1] ACKNOWLEDGMENTS A very special thanks to LORD Corporation, Keith Ptak, Zach Fuhrer, Zack Leicht, William Humphrey, Kathleen Lamkin-Kennard, Sergey Lyshevski, and the RIT ME machine shop staff for without their support the project would not have been possible.