Final Report F4U Corsair Flight Simulator Spring 2014 CT Corsair Senior Design Team Zachary Mosch, Randy Bertrand, Lauren Bradley, Arthur Podkowiak, Amanda Sweat, David Tartaglino Michael Turner uconn.ee.corsair@gmail.com uconn.me.corsair@gmail.com Project Advisors Rajeev Bansal Stephen Stagon ITE, 463 Engineering II, 311 (860) 486-3410 (860) 486-5088 rajeev@engr.uconn.edu stephen.stagon@engr.uconn.
CT Corsair Final Report May 2, 2014 TABLE OF CONTENTS 1. Summary and Overview…………………………………………………………………...….3 2. Introduction………………………………………………………………………………..…..3 2.1. Sponsor Background………………………………………………………………….…..3 2.2. Simulator Background……………………………………………………………………3 2.3. Problem Statement……………………………..…………………………………………4 2.4. Design Goals………………………………...……………………………………………4 3. Design Details……………………………..……..……………………………………………5 3.1. Design Deliverables………………………………………………………………………5 3.2.
CT Corsair Final Report May 2, 2014 8.5.2. Finite Element Analysis: Model Selection and Criteria…………………………36 8.5.3. Analysis Results…………………………………………………………………36 8.5.4. Tensile Test Validation…………………………………………..………………37 8.5.5. Finite Element Analysis Results…………………………………………………38 9. Motor Mount Design…………………………………………..…..…………………………39 9.1. Motor Mount Overview…………………………………………………………………39 9.2. Parametric Model Development………………………….…………..…………………39 9.3. Finite Element Analysis……………………………………..………..…………………40 9.3.
CT Corsair 1 Final Report May 2, 2014 Summary and Overview Connecticut Corsair has sponsored an interdisciplinary team of senior engineering students to restore a damaged Gyro IPT flight simulator, developed by Environmental Tectonics Corporation (ETC), with obsolete components and software to working condition. The ultimate goal of this project is to have the simulator respond to user input via standard airplane controls such that the simulator mimics the flight patterns of the F4U-4 Corsair aircraft.
CT Corsair Final Report May 2, 2014 models. The simulator provided to the team is equipped with three induction motors and accompanying VFDs intended to control the pitch, roll, heave, and vestibular movements of the simulator. These motors are not capable of running the simulator for an extended time period due to their tendency to overheat. To combat this problem ETC installed cooling fans on the motors which act as a quick fix.
CT Corsair Final Report May 2, 2014 To accomplish motion restoration, the team needed to replace the lower scissor arms and successfully select, calibrate and install system appropriate motors, drives and gearboxes. Accompanying requirements included the construction of auxiliary control circuitry, the mapping of IO requirements from hardware to software and an exploration of the proposed Prepar3D simulation software. 3 Design Details 3.
CT Corsair Final Report May 2, 2014 The ninth deliverable was the addition of user-controlled three-axis motion using adjustable joystick inputs. The final deliverable was a study and preparation of the Prepar3D flight simulation software. This software was outside the scope of our project goals and this deliverable is intended to assist future design teams. 3.
CT Corsair Final Report May 2, 2014 The following points outline the general workflow of the primary servo/induction installation and design process.
CT Corsair Final Report May 2, 2014 Servo drives receive command signals from a controller. They amplify the signal and transmit current to an actuator to produce a proportional motion. The command signal typically represents velocity but can also represent torques or positions. Unlike induction, servos have an integrated feedback sensor which reports the shaft’s status to the drive. The drive compares actual status with the commanded status.
CT Corsair Final Report Criteria Servo Advantages Encoder Encoder integrated Accurate position control Size Approx.
CT Corsair Final Report May 2, 2014 3.3. Induction Motor Rebuild: the How and Why The Nord induction motors and gearbox accompanying the simulator are the original pieces of hardware installed while the simulator was being developed at ETC. Both natural and unnatural wear and tear have resulted in damages to the motor housings and several internal pieces. The gearboxes were also in pieces, and one of three VFDs only worked in turning the induction motor in the positive direction.
CT Corsair Final Report May 2, 2014 by the microcontroller. Their purpose is to protect the machine from exceeding its limits by providing positional analog voltage information from the two induction motors. Figure 8. Induction Motor Encoder Feedback via Potentiometer This custom feedback device was selected for use on the simulator for several reasons. First, the encoders are very old and no support exists for their model and datasheets are nearly impossible to find.
CT Corsair Final Report May 2, 2014 Pitch and roll are dependent on the center of the simulator platform having a fixed pivot point. The pivot point is obtained by connecting the center of the simulator platform to a spring and universal joint combination. The universal joint is then connected to the platform and the guide rod ring which eliminates translational linear movement of the simulator platform along the x and y directions but will allow for rotational movement about the x and y axes.
CT Corsair Final Report May 2, 2014 4.3 Final Motor Specifications Although final motor specifications are inherently dependent on gearbox ratio selection, the calculations performed in Section 5 dictate the systems overall speed and torque requirements. Table 4 contrasts the original Gyro IPTTM specs with the new Corsair simulator specs. Criteria Gyro IPTTM Corsair Simulator Motor Type Induction Servo Speed 26 rpm 85 rpm Torque Output 1212 in-lbs 3500 in-lbs Power 1.5 hp 5 hp Table 4.
CT Corsair Final Report May 2, 2014 An F4U-4 Corsair Plane Model was purchased from A2A; a software add-on company. The file is from their “Aircraft Factory” line and includes F4U-4 A, B, C and D models. By installing the provided executable, a 3D Corsair airplane is loaded into the program with skins, dashboards and expected flight characteristics. 5 Kinematic Analysis for Motor Selection 5.1 Theory The original prototype’s induction motors are 1hp.
CT Corsair Final Report May 2, 2014 (Equation 1) The spring force is defined as: (Equation 2a) Where k represents the spring constant in N/m and y is the linear displacement of one side of the spring. This equation, becomes Equation 2b due to the shared spring force between the three arms. (Equation 2b) The free body diagram reveals the relation between the angle Φ, and the displacement of the simulator in the y-direction.
CT Corsair Figure 13. Components Contributing to Static Weight of Simulator Final Report May 2, 2014 Figure 14. Plot of Weight vs. Displacement for Spring Coefficient 5.3 Free Body Analysis: Torque Requirements To determine the motor torque requirement, it was necessary to find the total static simulator weight the motors have to overcome. This weight includes pushrods, the central spring, the universal joint and the platform. These components are shown in Figure 13.
CT Corsair Final Report May 2, 2014 develop the equations, the free body diagrams were drawn to include the cam, pushrods, platform and central spring joint. The diagram depicts the analysis of one motor, as the loading for each motor is assumed to be identical.
CT Corsair Final Report May 2, 2014 The cam represented in green, the pushrod is red, and the platform blue. Using this model, the cam was driven by in the positive and negative z-direction from -100° to +90° in increments of 10°. This range was selected based on the range of motion the cam can reach. At each increment the angles and were measured in the 3D model.
CT Corsair Final Report May 2, 2014 The next simulation analyzed was the simulator pitch and roll. For this analysis, it is still assumed that all three motors provide equal force to induce. The free body diagrams were derived and are shown in Figure 19. Figure 19.
CT Corsair Final Report May 2, 2014 As with the previous calculation, it is important to note that the angles do not vary linearly with each other. To understand the relation between , , , and a 3D model was created of just the cam, pushrod, and platform. Figure 21 shows this model with green representing the cam, red representing the pushrod, and blue representing the platform. The model again used 10° intervals between -100° and 90° for . For each , the angles , , and were recorded.
CT Corsair Final Report May 2, 2014 5.4 Torque Requirements After evaluation of the three main simulator movements, the total torque requirements are 3484.1 in-lbs and 2894.3 in-lbs. The larger torque value was selected and rounded up to 3500 in-lbs. The validity of this number was assessed by two means. First, when comparing the two calculated torque values, they were within 600 in-lbs of each other.
CT Corsair Final Report May 2, 2014 This equation states that for every 10° θ travels, the platform will have rolled a distance represented by . The equation was multiplied by the required roll rate of 81°/sec. Roll rates based on are displayed in Table 1-3 in Error! Reference source not found.. The greatest angular velocity was chosen from Table 1-2 (also in Appendix 3) to be 85 RPM. As a result, the simulator requires a gearbox-motor combination that operates at 85 RPM. 5.
CT Corsair Final Report May 2, 2014 6.2 Construction and Functionality The first prototype iteration used an Arduino Uno and LEDs to symbolized motor speed with light intensity. The second prototype implemented hobby motors. The microcontroller used was an Arduino Uno. Arduino technology has numerous open source libraries available, many of which are servo motor specific. Log files were created to take the potentiometer inputs and communicate the data to the controller.
CT Corsair Final Report May 2, 2014 The functionality of this prototype contributed to the proof of concept of the simulator programming by providing a means to enforce device data flow. Variable voltage inputs, representing joystick position, and encoder feedback position inputs were fed to a bounded, two dimensional mapping function which then resulted in motor position output. The range of motion was represented using 226 bit locations in both the x and y directions.
CT Corsair Final Report May 2, 2014 Figure 25. Model Setup Window for Model Selection Rotational Actuator (Actual) G405-1034A Linear Actuator (Same Parameters) G415-800A PTC Table 6. Actuator Model Numbers After the motor model is selected, the software parameters could then be set and adjusted using its internal menus which include System Setup, I/O Setup, and Motion. By default, the actuator was then launched in Commissioning Mode.
CT Corsair Final Report May 2, 2014 Finally, the drive is controlled using the GUI’s Control Panel. This panel is represented in Figure 27. Figure 27. Control Panel The Drive Enable is a software drive enable input, which enables the functions on the software’s screen. For actual operation, a hardware enable input at the drive terminal is required for operation. When operating normally, the hardware input enables both the hardware and the software.
CT Corsair Final Report May 2, 2014 The software also provides a useful Scope Function which was used to observe motor behaviors. Function 28. Scope Function The drive is controlled by the Arduino UNO microcontroller via a 9-pin analog male connector. The drive continuously reads inputs from the microcontroller and communicates these commands to the actuating shaft. It then provides adjustments by reading encoder feedback data and sending this information to the microcontroller for processing. 7.
CT Corsair Final Report May 2, 2014 9. The drive/motor will most likely fault. Adjust the Velocity Error Fault Limit which is found in the Control panel tab > Diagnostic Window > Select. 10. The Static to Dynamic Threshold should be somewhere near the Velocity Error Fault Limit. Refer to the manual for exact specifications. (file: CDS7323-A EMA Commission Software Manual) 11. The drive might still spin on its own without any move parameters.
CT Corsair Final Report May 2, 2014 20. Turn off 24Vdc hardware enable 21. In System Commands, Enable Model 22. Verify status LED 2 is blinking again on the drive 23. Select Save Parameters 24. Close System commands and Diagnostics windows 7.
CT Corsair Final Report May 2, 2014 pinMode(dirPinFwd1, OUTPUT); pinMode(dirPinRev1, OUTPUT); pinMode(dirPinFwd2, OUTPUT); pinMode(dirPinRev2, OUTPUT); pinMode(speedPin1, OUTPUT); pinMode(speedPin2, OUTPUT); delay(50); } // variables for mapping int speedRoll_1 = 0; int speedRoll_2 = 0; int readX = 0; int readY = 0; int testWrite = 0; int MAX_PWM_OUT = 200; int MIN_PWM_OUT = 25; void loop() { // put your main code here, to run repeatedly: // read and map the values from a joystick for testing purposes re
CT Corsair Final Report May 2, 2014 in the z-direction. The scissor arm system is shown in Figure 29. The upper scissor arms are shown in green, the lower scissor arms are shown in red and the shock absorbing springs are shown in blue. 8.2 Purpose of Redesign The previously manufactured lower scissor arm was incorrectly designed and over-engineered. The existing upper scissor arms from the Gyro IPTTM consist of welded aluminum plates and internal pin structures.
CT Corsair Final Report May 2, 2014 expected. Therefore, the arm will either fail in buckling or at the bearing holes due to yielding5. 8.4 Upper Scissor Arm: Finite Element Analysis 8.4.1 Model Selection Prior to analysis in ABAQUS, it was crucial to ensure the upper scissor arm drawn in Solidworks included every detail of the member including the internal support pins.
CT Corsair Final Report May 2, 2014 8.4.2 Analysis Criteria The program used for analysis was ABAQUS CAE1. When compared to ANSYS, another option for FEA, ABAQUS provides better meshing capabilities and a user-friendly interface more intuitive to a new user than ANSYS8. In order to achieve accurate and usable results, it was imperative that appropriate inputs were provided including material properties, boundary conditions, loading scenarios and meshing and mesh convergence.
CT Corsair Final Report May 2, 2014 Figure 37. Bearing stress loading Since the pin was omitted from the analysis, the bearing stress was assumed to distribute uniformly over the bearing hole. As a result, during the FEA the load was applied to half of the bearing hole as shown in Figure 37. This caused deformation only to occur on the z-axis and created an accurate output stress concentration in the bearing holes.
CT Corsair Final Report May 2, 2014 8.4.3 Analysis Results The FEA was completed using the aforementioned criteria. Using 9038 as the number of elements in the mesh density resulted in a stress distribution shown in Figure 41. It can be clearly seen that the member will fail at the bearing holes before any other member deformation. This failure occurs at approximately 2400 N, or 540 lbs, for each upper scissor arm.
CT Corsair Final Report May 2, 2014 8.5.2 Finite Element Analysis: Model Selection and Criteria After the lower scissor arm was designed, it was necessary to perform finite element analysis on it in order to confirm that it can support the weight of the simulator as well as other combining forces. To begin the analysis, a free body diagram was used to visualize how the forces would act upon the simulator.
CT Corsair Final Report May 2, 2014 As can be seen, the two end bearing holes are being pulled in tension while the bottom two bearing holes are used to fix the arm in place. It can be noted that a mesh convergence analysis was completed for this arm, just as one was completed for the lower scissor arm earlier in the report. A high mesh density was chosen after reviewing the mesh convergence analysis.
CT Corsair Final Report Once the physical tensile test was performed, the results were acquired. From the tensile test, it was found that the quarter-scale arm failed in shear at a 45 degree angle to the hole, as can be seen in Figure 48. This showed that the highest concentration of stress was in fact not at the tip of the bearing holes, but rather at a 45 degree angle to the side or the top of the holes.
CT Corsair 9 Final Report May 2, 2014 Motor Mount Design 9.1 Motor Mount Overview The existing motor mounts are shown in Figure 50 and Figure 51. The existing motor mounts are sufficient for securing the induction motors, but they are inadequate for securing the servo motor and gearboxes that will eventual replace all the induction motors. The mounts are critical components as they hold the motors in place and support the entire simulator weight while in operation.
CT Corsair Final Report May 2, 2014 Figure 52: SolidWorks Model of Motor Mount 9.3 Finite Element Analysis 9.3.1 Model Selection Once a free body analysis was carried out on the motor mounts, two main forces needed to be analyzed: the simulator weight force and the motor spin torque. These forces have been illustrated in Figure 53. What this meant was that we would need to simulate in ANSYS the main face of the motor mount having both these types of forces acting on the bearing holes at the same time.
CT Corsair Final Report May 2, 2014 One thing to note about this FEA is that the simulator weight may not always exclusively act in the downward direction, as the motors move and cause the force to be applied in slightly different ways. Therefore, eight different simulations were performed. In these simulations, the direction of the weight acted in the north, northeast, east, southeast, south, southwest, west, and northwest directions.
CT Corsair Final Report May 2, 2014 9.3.4 Motor Mount Analysis Conclusions From the obtained results, assuming manufacturing errors, steel 8620 was used for the motor mounts. This material surely accommodates for any errors in manufacturing, ensuring the motor mounts are be safe. Using Aluminum 6061-T6 would be satisfactory, but is ultimately too much of a risk to use this material as the team is custom machining the parts. 10 Conclusion 10.
CT Corsair Final Report May 2, 2014 by linear relationships with each other, forcing the development of two CAD models. The first CAD model was of the vertical motion profile and modeled the relationship between the cam, pushrod, and simulator platform. The second CAD model was of the pitch/roll motion profile and modeled the relationship between the cam, pushrod, and simulator platform.
CT Corsair Final Report May 2, 2014 diagram of the entire scissor arm assembly. The free body diagram allowed the simplification to be noticed that the failure loading of the upper arm should match the failure loading of the lower arm. The reason behind the simplification is if the lower arm has a higher failure loading point than the upper arm, then the lower scissor is overdesigned. In reverse, if the lower scissor arm has a lower failure loading point then it becomes the weak link in the system.
CT Corsair Final Report May 2, 2014 simulator to function correctly. Once the gearbox and motor were constrained a new motor mount was designed utilizing the attachment points of the old motor mounts. Then an FEA simulation in ANSYS was performed that applied the maximum torque and maximum force onto the motor mount. The lowest safety factor on the motor mount was 2.4 signaling that the design is sufficient.
CT Corsair Final Report May 2, 2014 serial encoder feed-out pins have been studied and their outputs documented for future closed loop positional control. The servo motor drive runs on 208VAC and is enabled from a computer, via an RS232 cable, using a custom GUI software developed. The Arduino controller can be connected to the Servo drive’s general I/O port. The third deliverable was the integration of an F4U-4A Corsair airplane model into the Prepar3D simulation software.
CT Corsair Final Report May 2, 2014 actuator at the right time and the actuator shaft would spin without user commands. This phase was re-calibrated using high level Moog access codes as provided by their application support engineers. 10.3 Future Recommendations After meeting all of the deliverables for the 2013-2014 senior design project, the path forward for the 2014-2015 team is laid out as follows. The mechanical engineering team’s goals next year will consist of four distinct parts.
CT Corsair Final Report May 2, 2014 installation of safety features and future work on the wiring, installation and IO requirements of the cockpit and control panels. The next recommendation is the refinement of the simulator’s motor control. The motors should operate using positional feedback. They should also operate with a fourth degree of motion; spin. Finally, two more servo motors should be integrated into the simulator by the culmination of the design process.
CT Corsair A1 Final Report May 2, 2014 Nomenclature Actuating Arms – A set of three arms bolted to the primary motors of the motion base, with one actuating arm per motor. The actuating arms are adjusted by their respective motors, affecting the pitch and roll of the cockpit. Actuator – Linear or rotational motor used to control a system. Cockpit – The aircraft flight deck, containing the aircraft’s flight controls and monitoring systems.
CT Corsair Final Report May 2, 2014 Instrument Panel – A display of multiple gauges, sensor displays, gyroscopic displays, and other flight data displays located in front of, or near, the pilot. The instrument panel provides the pilot with all of the available information about his or her aircraft at any given time. I/O – Stands for Input/Output. IO is communication between a processer and the world. Inputs are received data and outputs are sent data.
CT Corsair Final Report May 2, 2014 Yaw angle – designated as the angle of the simulator’s nose with respect to its forward-facing zero position. Zero-position – The home position of the flight simulator while raised for operation, where = = = 0. A2 References 1. "Abaqus CAE." Finite Element Analysis. N.p., n.d. Web. 15 Nov. 2013. . "Advanced Pilot Training." ETC Corporate. Environmental Tectonics Corporation, n.d. Web. 30 Sept. 2013.
CT Corsair Final Report May 2, 2014 30. 31. "Young’s Modulus." Engineering Toolbox. N.p.. Web. 15 Oct 2013. CA73477 SUPPORT FILES FOR 902B SOFTWARE. 1 April 2014. DS2110 COMMUTATION OFFSET PROCEDURE. Moog. 10 April 2014 A3 Supplementary Analysis Data 29. Weight (kg) 2.27 4.54 6.8 9.07 11.34 13.61 15.88 18.14 20.41 22.68 27.22 31.75 36.29 40.82 45.36 49.9 54.43 58.97 63.5 68.04 72.57 77.11 81.65 86.18 88.04 92.57 97.11 101.65 106.18 108.
CT Corsair -20 -30 -40 -50 -60 -70 -80 -90 -100 70.98 62.18 53.82 45.83 38.18 30.79 23.6 16.53 9.51 Final Report 89.02 87.82 86.18 84.17 81.82 79.21 76.4 73.47 70.49 Angle of Platform Φ 22.59 21.86 20.2 17.79 14.81 11.41 7.73 3.89 0 -3.85 -7.86 -11.12 -14.39 -17.33 -19.9 -22.05 -23.77 -25.03 15.01 12.16 9.53 7.15 5.06 3.28 1.84 0.73 105.45 113.98 126.24 143.56 168.61 207.03 272.39 408.54 May 2, 2014 17.57 19 21.04 23.93 28.1 34.5 45.4 68.09 Table 1-3: Roll Rate vs.
CT Corsair A4 Final Report May 2, 2014 Final Wiring Diagram 54
