Disc Launcher - Senior Design Project Lab | ECE 445, Study Guides, Projects, Research of Electrical and Electronics Engineering

Download Disc Launcher - Senior Design Project Lab | ECE 445 and more Study Guides, Projects, Research Electrical and Electronics Engineering in PDF only on Docsity!

Disc Launcher By Daniel Kang Keith Orlowski ECE 345, SENIOR DESIGN PROJECT FALL 2002 TA: Paul Leisher Project No. 13

ABSTRACT

The idea behind this project was to create a machine that would be able to simulate the ability of a human to throw a disc, often called a Frisbee. It will corner the market providing instructional and recreational use at low cost to consumer. The design uses two facing belts rotating in the same forward direction. Each belt is independently driven with its own dc-motor. The signals sent to each of the motors are initial digital signals that must be first converted to analog signals and then amplified. The final product was hoped to be able to throw a disc up to 50 yd at 10 r/s. Figure ii – Proposed Platform Model ii

1. INTRODUCTION

CNN Sports Illustrated reported “more Frisbees are supposedly sold each year than footballs, baseballs and basketballs combined” [1]. More and more people are using discs in one form or another. One rapidly growing use is in the sport Ultimate Disc. The Ultimate Players Association, founded in 1979, is one of the main organizers of the sport and already has over 15,000 members. Other flying disc organizations have over 100,000 members [2]. The market for a disc-launching machine definitely does exist. 1.1 Review and Update Material The project had to be modified due to time constraints. The function of the original design incorporated step motors to control the pitch, yaw, and roll of the disc. This portion of the project needed to be cut in order to allow for enough time to design, develop, and test the remaining portions. In the proposed plan there was no consideration of heat dissipation. Due to the amount of power required to drive the motors a power amplifier needed to be implemented. This solid-state device generates a significant amount of heat that needed to be displaced. In order to handle this drawback, a heat sink needed to be added. 1.2 Function and Block Diagram Figure 1.2 – Block Diagram Figure 1.2 shows the basic layout of the design. iv

1.3 Subprojects Referring to figure 1.2, the design can be broken up into two main components. The user interface, logic, and display make up the first half that is mainly derived from digital components. The second subproject is the digital to analog converter, power amplifier with heat sink, and the variable dc-motor that is mainly comprised of analog components. I. Digital i. User interface ii. Logic iii. Display II. Analog i. Digital to analog converter (DAC) ii. Power amplifier with heat sink iii. Variable DC-motor iv. Disc platform v

Many platform designs were discussed. The idea of using slingshot or cross bow to give the disc initial velocity after somehow bringing the disc up to the required rotational speed was eliminated because of the instability and lack of electrical components in the design. A robotic are mimicking human movement was quickly discarded due to its extreme complexity. Next came the idea of one rotating belt pushing against one edge the disc while the other edge slid along a frictional surface. This idea allowed us to give the disc forward velocity and rotation at the same time, however the two would always be dependant upon each other. The final idea of having two rotating belts pushing the disc on opposite edges was discussed. This would allow the two degrees of freedom needed to control the velocity and rotation independently. It also allow for electrical control of the system. vii

3. DESIGN DETAILS

3.1 Components Motor Driver (1 motor): 1 DAC chip 1 Power Amp 2541 (half of the component) User Interface: 4 Hexadecimal displays 8 D flip flops – store memory for the Hex displays 2 Dual 4-input NANDs – AND together several signals for flip flop clocks 1 Hex inverter chip – invert the dipswitch signals for input into the NANDs 1 4-switch dip switch – determine which motor and which digit are selected 1 Hex open collector inverter chip – stabilize Data Available from key encoder 1 16-key encoder – encode keypad output to hex numbers 1 keypad – user input 3.2 Diagrams and calculations I. User interface The user interface is composed of the keypad, the logic components, the memory, and the displays. A dipswitch controls which of the four displays is currently being changed. The D flip flops are initially set to a known state by another dipswitch that toggles the CLR bit. This is done by the use of two dipswitches that use binary numbers to keep track of which display is selected. For example, when the user has selected a certain combination like 0 = D and 1 = M, the lower digit of the right motor is selected. The keypad feeds directly into the 16-key encoder, which outputs a hexadecimal number in binary. This binary number is fed directly into the D inputs of the D flip flops. The data available pin on the encoder goes to high whenever a key is pressed. The output from this pin is ANDed with the D and M bits from the dipswitches to determine which of four clocks for the D flip flops will rise. The Q outputs of the D flip flops are fed into the corresponding pins of the displays, so that the correct hexadecimal digit is shown on the selected display. The blanking and latch strobe inputs on the displays are tied to high, so that the display is always taking an input. One sixteenth of the final design can be seen in Fig. 3.2.I. viii

Using equation 3.2.II we calculate the minimum resistance required. Since this device can output a maximum of 4.2 mA [3], and VREF+ = 5 V the minimum resistance value is 1.19 k. The resistors were chosen to be 5.6 k to allow for ample tolerance without worry of saturation. Figure 3.2.II – Digital to analog converter layout x

III. Power Amplifier The power amplifier takes the current output of from the digital to analog converter and amplifies the power to allow the ability to drive the 24 V motors. As shown in figure 3.2.III the setup of the amplifier is an inverting gain configuration to allow us to invert the signal from the digital to analog converter. The gain can be calculated using the following formula [4]. Vo = VIN * (RF / RB) (3.2.III) If RB is chosen to be the same as RREF+ then maximum input voltage will remain VREF+. RF must now be chosen to provide the appropriate amount of gain to allow the output voltage to be the 24 V maximum allowed by the motor. RF calculates to be 26.93 k. After testing the system a value of 27.4 k works best because the voltages are not exact. Figure 3.2.III – Power amplifier layout A heat sink is needed on the amplifier due to how much power the amplifier is required to dissipate. xi

TABLE 4.3 VRMS OF THE DAC (IO IS CONNECTED TO GND, R = 5.6k) Input (Hex) Voltage (V) Vp-p (mV) I (mA) 00 -0.032 312.5 0 01 -0.043 468 -0. 02 -0.108 468 -0. 04 -0.224 468 -0. 08 -0.433 468 -0. 10 -0.748 312.5 -0. 20 -1.49 312.5 -0. 40 -3.06 312.5 -0. 80 -6.05 312.5 -1. FF -11.91 312.5 -2. TABLE 4.4 OUTPUT OF AUDIO AMPLIFIER Input (Hex) VAVG (V) Vp-p (V) I (mA) 00 -0.66 10.94 --- F0 13.60 --- 26 When we received the new amplifier, we immediately tested the input node of the amplifier and found that there was less than a 0.5V peak-to-peak swing. This was good news indeed, as this was 5% of the voltage peak-to-peak swing of the previous amplifier. In addition, the output node of the amplifier exhibited a nearly clean DC signal at all voltage levels, as shown by Table 4.5. TABLE 4.5 OUTPUT OF POWER AMPLIFIER Input (Hex) VAVG (V) Vp-p (mV) I (A) 00 0.05 312.50 0.88 m 20 3.14 1.50 1. 40 6.37 2.20 1. 80 11.91 2.50 2. F0 19.27 2.90 2. 4.2 Conclusions The main stumbling block was our motor driver, which we spent several weeks debugging. Unfortunately, the time we spent debugging the driver took time away from the other parts of the project that we had been hoping to implement, such as the tachometer readout and the pitch and yaw controls. We realized that at times one must bite the bullet and spend some resources in order to get past stumbling blocks. In one way or another, those resources come down to one thing: money. In addition, we had to sacrifice some extra features in order to get the basic functionality working properly. In most engineering projects, the verification portion of the project is often the most important part, usually more time-consuming than the design portion of the project itself. Amplifier aside, the testing of the DAC and the user interface was rather straightforward. The order that we tested the project was the amplifier first, then the DAC, then the user interface. However, this was not the optimal method of debugging the circuit, as the amplifier was dependent on the DAC. If we had been wise to check the DAC before moving onto the amplifier, we definitely would have saved a lot of xiii

time trying to figure out why the output of the amplifier was acting as oddly as it was. Our final result however was a linear response of output voltage across the motor with linear digital input. Our voltage noise, measured using peak to peak values, was also relatively small as show in figure 4.2. Digtal Input vs Motor Voltage R^2 = 0. 0 5 10 15 20 25 0 50 100 150 200 250 300 Digital Input Voltage (V) Vavg Vpp Linear (Vavg) Figure 4.2 – Linear voltage output xiv

5.3 Schedule TABLE 5.3 – WEEKELY SCHEDULE OF TASKS AND DUE DATES Week Due Tasks 09/16/02 Project Proposal Research Parts – DK KO Work on project proposal – DK KO 09/23/ Select and Order Parts – DK Design Frame – KO DK 09/30/02 Design Review Submit Frame to Shop – KO Prepare for design review – DK KO 10/07/02 Design tachometer – DK 10/14/ Build and test motor driver – KO Design input and motor driver – DK 10/21/02 Design and test motor driver – DK KO 10/28/ Design user interface – DK Integrate variable speed motors and frame – KO 11/04/02 Mock-up Demos Prepare for Mock-up Demos – DK KO Design tachometer – DK 11/11/ Address comments from design review – DK KO Testing of user interface – DK 11/18/ Final testing and debugging – DK KO Redesign user interface – DK 11/25/ Final testing and debugging – DK KO Work on Final Paper – DK KO 12/02/02 Presentation Prepare for Final Presentation – DK KO Work on Final Paper – DK KO 12/09/02 Final Papers Due Finish Final Paper – DK KO xvi

6 CONCLUSIONS

Although we were unable to attain our original goal of throwing a disc 50 yards, we believe that our efforts were definitely a step in the right direction. By the time we were finished, our project could only throw the disc 5 yards, but there was definitely several areas where a few fixes would go a long way. 6.1 Accomplishments We were able to get our disc thrower to throw the disc about 5 yards, which was well short of our original goal of 50 yards. The main contributing factor to our shortcoming was money. In engineering, the two main limiting factors are time and money. Although we were not short on time, we did try to minimize the cost of our project. This was done by using the parts from the Parts Shop whenever possible. The only two parts that we could not obtain from the Parts Shop were the motors and the power amplifier. Although we ended up having to pay for the amplifier, we were able to borrow the motors from the Power lab. The parts that were incorporated into performing this task were the user input and the motor driver. The user input system used a keypad to take input and retained the last input value in memory. We were also able to construct a motor driver that amplified the input signal and output a voltage range of zero to approximately 24 volts. 6.2 Uncertainties We were not sure if the disc was going to gain enough acceleration in the short time span between the two belts. The first of two factors that contributed to this condition was the fact that the disc was in contact with the belts for only 30 cm. This would mean that the disc would have to experience a significant acceleration in a very short period of time. If this condition wasn't met, then the disc would not leave the belts at the desired velocity. The second factor was the friction coefficient of the belts themselves. Due to the fact that the belts were only in contact with the disc for a short period of time, we had to make sure that the belts had enough grip to prevent slippage once the belts came in contact with the disc. 6.3 Future Work / Alternatives We came up with several ideas that could be implemented in future versions of our project. The first change would be to locate belts that still had the riveting on the one side of the belt that came in contact with the pulleys, but have a softer, solid rubber texture on the side that came in contact with the disc. This would allow for a large decrease in the amount of slippage that occurs in our current model. Once the issue of the belt was taken care of, we would move onto obtaining better motor drivers that would more efficiently disperse the heat caused by the amplifier. Also, we would obtain larger power amplifiers, in order to allow more power to be run to the motors. xvii