User:Medelen8/ENES100/APW Design B

Problem Statement
To design a number of subsystems in order to make a "power wheel" autonomous. It is imperative that these systems be able to work together without any difficulty. In doing so, it is hoped that at some stage the vehicle will be able to drive itself without any operator instructions. However, as it stands now, several subsystems must first be designed in order to replace the interactions a human driver would have. These subsystems must be both simple to build, simple to repair and improve upon. At this stage an improved steering subsystem needs to be designed and a mechanism and subsystem for depressing the accelerator must be constructed.

Requirements for each element or component derived from system level goals and requirements
Each subsystem must be simple in the sense that any future group can pick up and continue the project. This means making each subsystem simple and efficient. While the code is not yet complete each devices actions must not interfere with any other system. Besides simplistic designs, the parts themselves must be simplistic. This will ensure that they can be easily replaced should the need arise.

The requirements for each subsystem are as follows:

The Pedal Pusher
 * 1)  Must be able to withstand a minimum of 3 Newtons' of force
 * 2)  Must be able to interface with a lead screw attached to a motor
 * 3)  Must be able to operate in the forward and reverse directions (no locking mechanism)
 * 4)  Needs to be durable enough to last 30 cycles a minute, with at least 50 hours of total use
 * 5)  Weighs 3 ounces or less to avoid putting stress on the lead screw and motor.
 * 6)  Needs to switch between depressed and open state within 3 seconds
 * 7)  Must be simple to construct and under $10

Motor Mount
 * 1)  Needs to be able to hold a variety of motors (universal design)
 * 2)  Needs to be able to support at least 2 pounds
 * 3)  Stable enough to withstand 5 Newtons of force pushing the opposite direction of the pedal
 * 4)  Has to attach to the frame in the rear of the vehicle (No new holes may be drilled into power wheel)
 * 5)  Simple to construct within a maximum budget of $15.

Frame
 * 1)  Fit on the preexisting base-plate that was screwed onto the seat. Frame can extend over the base-plate lengthwise because of extra space in front of the seat. Max dimensions Length 10" Width 20”
 * 2)  Able to support 20 pounds of hanging weight (Weight of the steering wheel motor, pedal motor, and arduinos)
 * 3)  Compartment big enough for a car battery. Length: 10”, Width: 6.88” Height: 7.94”
 * 4)  Mount location for the steering subsystem and pedal pusher motor mount
 * 5)  Able to be constructed with materials under $25
 * 6)  Stable enough to handle accelerating up to 5mph in forward direction and 2.5mph in reverse
 * 7)  Weigh less than 100 pounds (Max weight power wheel supports is 130 pounds)
 * 8) The arduino box should be mounted on the inside of the wooden box so that it is in a more secure location
 * 9) The top and back of the "box" should be open so that access to the components in the frame is easy
 * 10) The frame should be secured with L-brackets in the existing 4 screws on the floorboard of the box
 * 11) The walls of the frame should be held together with L-brackets to maximize the available space inside the frame

Alternatives in design
Throughout this project several changes were made in the designs of the systems and subsystems. Images of these changes can be viewed below.

Frame Initial Design
The initial design of the frame for the power wheel was constructed with untreated wood. The wood was beginning to splinter and crack due to drilling screws into the wood with the grain. We decided it would be best to redesign the frame since it was beginning to fall apart. With the new redesign, we thought it would be beneficial to increase the surface area that the steering wheel could mount to. This would give us a wider range of motors to use for the steering wheel and allow for a wider range of motion for some motors. Here is an actual picture of the original frame:

Frame Final Design
For the final design of the frame several modifications were done. First the arduino box mount was moved to the inside of the wooden box so that it is in a more secure location. This was done so that it won't be hit while turning corners in the highway. Next the top and back of the "box" were opened so that access to the components in the frame is easy. The frame wall would have to be assembled using L-Brackets instead of a pieces 2x4 and wood screws so that there is more room inside of the frame. This would allow us to be able to house whatever battery we choose to add to it. The frame should be secured with L-brackets in the existing 4 screws on the floorboard of the box. The frame is to be removable from the floorboard for easy disassembly and maintainability. The original steering mechanism will be mounted directly on the center of the steering wheel because it is easier to turn the steering wheel in the middle than on only one of the sides. To increase durability and longevity, we decided to coat the wood in Plasti Dip. This would prevent splintering and allow it to operate in adverse weather conditions. The other option would be to construct the frame out of plexi glass but that would increase the cost immensely. Plasti Dip is about $7 a can (1 can would be more than enough) where as the appropiate amount of plexi glass would put our expenses for the frame alone around $90. Having the wood coated in plasti dip would also increase the aesthetic value.

Concept #1
The initial design for the pedal pusher was a design that would require the user to screw two pieces together to secure the hex nut in place. This would have made it fairly easy to secure the hex nut into the part. However, the screw holes to secure the 2 pieces together would be a possible point of failure after many uses. Majority of the torque from the hex nut would be concentrated around the screw points which severely increases the chance of failure and breakage. Also, ensuring that the screw holes line up perfectly on both pieces would sharply increase the complexity of the design. Even with proper measurements, theres a chance the 3D printer wouldn't be as accurate as we need it to be. Finding screws that won't strip the plastic away would have been another challenge to overcome since metal would easily strip the plastic threads. The upside to this design is that it would have provided the most secure fit of the hex nut and could have accommodated hex nuts of similar sizes.

Concept #2
Another design was a pedal pusher that attaches to a clamp that goes around the pedal. This would allow the pedal to respond instantly to the motor without any wait time for the piece to come into contact with the pedal. The clamp would have a piece that would attach to the nut and consequently to the lead screw on the motor. This would have been the most responsive design since their wouldn't be a wait time for the pedal pusher to make contact with the pedal. The negative to this is that the angle at which the piece should push the pedal would change as the pedal is depressed. Since the piece between the clamp and nut would have to flex every-time it extends or retracts, it would weaken over time and could possible snap after repeated use.

Concept #3
Concept #3 was a design that would allow for tool-less assembly and be easier to align to the pedal. Without being physically attached to the pedal we wouldn’t have to worry about the piece breaking or aligning the piece to the appropriate angle. Since it would be one solid piece, the possibility of parts becoming weaker over time is reduced. The screw points are no longer a possible point of failure and the design would be easier to print. In this design, the hex nut slides into the hole on the side. The space cut out for the hex nut will provide a tight fit so that when the hex nut turns, the conical piece turns with it. As the hex nut turns, it moves up or down the lead screw. Consequently, the conical pieces moves with it causing the pedal to be depressed or not depressed depending on the position of the hex nut on the lead screw.

The two points of possible failure that we identified on this design are the front of the piece thats in contact with the pedal and the 'walls' securing the hex nut in place. The front of the conical shape may become worn down since its in contact with a harder plastic and friction from the centrifugal motion will wear out the plastic. Over time the walls securing the hex nut could become worn out which will make the hex nut loose, further stripping away the plastic material. The tighter the fit on the hex nut, the less likely the fit will become lose. Both of these scenarios will happen well beyond the initial requirements we laid out, if they occur at all.

Concept #1 Motor mount
This was concept #1 for the pedal motor mount. It was designed specifically to fit the curves of this motor. The curves would have absorbed most of the torque the motor would output and evenly distributed the force along the piece. This mount would have only worked for motors of similar cylindrical dimensions. Creating the curved insets would have been a difficult task to complete without having the dimensions of the motor. Even with accurate measurements, construction still would have been a challenge to get it to be a completely snug fit. The added complexity would have required several prototypes and more material, increasing the cost of the project. After doing all of this, the design would be obsolete if a future team decide to use a different motor.

Concept #2 Motor Mount
This is concept #2 for the pedal motor amount. The flat surface allows for a U-Bolt to clamp around the motor and the narrow piece of the platform. This is the most universal design as it allows for a wider variety of sizes and shapes of motor to be used. It also allows us to have more control of the position of the motor. This is important incase we alter the design of the pedal pusher and it becomes longer or shorter. We can move the motor to any position along this mount to accommodate different sizes of pedal pushers. The flat design is also the simplest to design and the easiest to mount to the new frame. For the motor we decided to use, it will provide more than enough support. However, if a future team does decide to use a heavier motor, a support beam can easily be attached to the underside of the flat platform and connected to the frame. The adaptability of this design allows for multiple uses.

The initial design
As you can see above, each subsystem has several initial designs. These designs are all practical and were worth consideration. Instead of prototyping each design, we were able to list the pros and cons of each design and fill out a design matrix. Since our requirements and solutions were relatively simple, prototyping each design wouldn't have been cost effective or an efficient use of time.

Experimental prototypes and testing conducted during design
Since we are in the design phase, none of these subsystems have been prototyped yet. However, the pedal part does have a 3D Replicator 2 file ready to be printed. The only thing that could be tested was the code and Arduino. As far as the coding goes, testing was done using a basic DC motor. Wires were soldered onto the motor leads and plugged into the Arduino Uno. During the testing, a few ideas came to mind. Mainly, different voltages and different current will affect the speed of the motor. This was confirmed by using comparing the speed of the motor spinning from the 5V supplied via the USB cable, to the speed of the motor attached directly to a 9V source. Since our motor will be running on 12V, the variations in Voltages isn’t a concern. However, as the battery ages and is used, its current output will drop causing the motor to spin slower. This will affect our code since the code is designed to turn the motor on for a specific amount of time (1.35 seconds). If it spins slower, the pedal may not be depressed, throwing off the rest of the code.

Since the 5V power source we used (Computer) never runs out and we didn’t have a way of measuring current for the 9v battery, the current detection portion of the code couldn’t be tested yet to see if it can accurately predict how quickly the current will drop or how much of an impact that will have on the speed of the motor. Also, since we won’t be using a 9V battery, the data collected from the 9V test wouldn’t be useful or applicable to our project. Thus we weren't able to completely optimize the code but using math (outlined below) we were able to estimate that the motor would need to run full speed for 1.35 seconds to completely depress the pedal. The experimental code is pasted below:

Appropriate optimization in the presence of constraints
As seen in the surrounding images all of the systems had to be edited to improve compatibility a well as efficiency. It was decided early on that the systems involved needed to be simple, moderately light weight, and efficient. Given these basic constraints, the group set about creating systems and subsystems to control aspects of the vehicle. A major change had to be done to the steering mount in order to improve its efficiency.

In regards to the pedal pusher design, 4 imperative design features required optimization while adhering to the constraints initially set forth.
 * 1) In the code that controls the pedal pusher, we implemented the use of the brake pin (pin 9) in order to quickly turn off the motor. The original method would be to just stop supplying the motor with power, but the motor would still spin until is gradually comes to a stop. By using the brake pin, we are able to stop the motor more quickly, making it easier to control how far the pedal pusher travels and allowing us to cut the motor off quickly in case of an emergency.
 * 2) The shape of the front end of the conical piece. Originally it started out as a flat surface (see concept #2 of the pedal piece) but in order to maximize the effectiveness of the force output of the motor, it was decided that most of the force should be applied to a small surface area. This would allow a great deal of force to be applied directly in the top center of the pedal instead of 'wasted' on the sides. By having a small shape, it would also be quicker, cheaper, and easier to print.
 * 3) To ensure the motor wouldn't 'bottom out' or extend itself too far into the pedal pusher, a buffer zone of 4.23 inches was created. The lead screw has a length of 4.125 inches, which leaves a safety net of 0.105 inches. It was determined this was the most optimal buffer zone to create inside the pedal part to maximize integrity and ensure durability. Giving too much extra space would have made the part weaker, and not enough would have caused the motor's lead screw to dig into the pedal part, possibly creating a fracture point or damaging the motor.
 * 4) The size of the pedal pusher part itself was optimized given the possible mounting locations and the size of the lead screw. We decided on 5 inches instead of a larger length to maximize the transfer of energy from the motor to the pedal. A longer piece would weigh more and require more energy to turn. By using a part thats only 5 inches long, we reduce the amount of materials used, reduce the amount of energy required to turn it, and maximize the torque efficiency. Going smaller wouldn't have been an option since the actual cone part is only 0.77 inches. Any less than that would have made the front end very fragile and susceptible to damage under load. There also needs to be solid material behind the acting force to be effective on the pedal. Not enough mass or a hollow area would cause the piece to collapse under stress.

Iteration until convergence
During this project both the motor platform and the steering mount/frame changed many times. At first the systems started out as overly complex and in need of simplification. The steering frame/mount was redesigned to ensure that the steering mechanism could not only turn easier, but that the steering mechanism itself was at a better angle of contact. This would remove stress that was placed upon the bolts holding the steering mechanism to the frame and to prevent it from breaking or fracturing under pressure. The motor platform for the pedal pusher changed in design three times. (The initial and final design can be seen above here ') The first design did not have much thought as to function and focused more on form. The second was between the initial and final designs in an attempt to keep both form and function. However, the difficulties concerning looks were removed in the final fully functional design. Being a simple subsystem this platform is deigned to do its job, and do it well without interfering with any other systems. In addition to the improvements described above in the optimization section, there were 2 areas in which iterative design was applied to the pedal part.
 * 1) First was in determining the hole for the hex nut on the side of concept #3. Originally we had both sides cut out with a cover that could be screwed on. After the first design we decided to close one side and just have one side open with a cover. During the 2nd redesign phase, we decided to get rid of the cover all together since the screw holes would have weakened the structural integrity of the piece.
 * 2) The 2nd area in which iterative design was proved monumental in our design is the actual shape of the front end. When we changed the design from a flat front end to a conical shape, it was still a large surface area. We went through about 4 designs, each decreasing the surface area of the conical shape until we thought we may have reduced it too much. We decided to test the smallest possible diameter by using different sized dowel rods to press the pedal and seeing which one depressed the pedal the easiest without flexing or bending. After a certain size, the effectiveness of increasing the diameter stops becoming beneficial. The diameter of the dowel rod was 0.2 inches which is what we made the tip of the front end.

The final design
The final designs for the three systems designed in this project cycle are as follows:

Steering Mount/Frame
The final design for the frame can be seen here:

Pedal Pusher:
Concept #3 for the pedal pusher was chosen as a final design due to having the best reliability, simplest construction and easiest installation.

Technical and scientific knowledge
Since this project had the goal of finishing the design of the power wheel, most of the different calculations and scientific research has already been done in the first part of this project ( design A ) but there was something forgotten : the control of the gas pedal. We have the different parts needed to control it, to push it forward or to release it to make the power wheel go faster or slower, we have the software code necessary to control the power wheel and its motor , but we didn't know how much time the motor had to be on so that the traveler nut moves forward enough to push the pedal. In order to fix that we have done different calculation using the RPM of the motor ( 178 ), the pitch of the bolt on which the traveler nut is ( pitch = 1/8 inch) and the distance the traveler nut needs to run through before the part can push the pedal forward ( 1/2 inch)

First : find the number of rotations per second $$ 178rotations/min = 178 rotations/60s = 2.96666667rotations/s $$

Second : find the number of rotations needed to obtain the distance of 1/2 inch

$$ 1 rotation = 1 pitch = 1/8 inch $$ $$ 4 rotation = 4 pitch = 1/2 inch $$

Third :find out by how much we need to multiply 2.96666667 to obtain 4 $$ 4/2.96666667 = 1.34831464 $$    $$ 2.96666667 $$ x $$ 1.34831464 = 4 $$

fourth : we know that we have 2.96666667 rotations every second, but we need 1.34831464 times the rotation for one second thus finding the time can be resumed to this

$$ 2.96666667 rotations $$ x $$ 1.34831464 = 4 rotations $$ $$ 1 second $$ x $$ 1.34831464 = 1.34831464 seconds $$

Now that we have done these calculations, we know that in order to have the gas pedal pushed so that the power wheel advances forward , we have to make the motor spin 1.34831464 seconds ( roughly 1.35 seconds ).

Creativity, problem solving, and group decision-making
Throughout this project cycle, the group decided to go three directions. This was done to maximize efficiency and increase overall productivity. A major aspect was that besides the designing of new systems, the steering mount was completely redone. These designs as seen in the above pictures have changed drastically throughout this project cycle. The focus was to make the systems both more simple and function fluidly.

Creativity
An example of creative problem solving can be found in how we monitor the amount of current to determine how long the motor should spin. When the motor is operating at full speed it takes 1.35 seconds for it to depress the pedal. However, when the battery gets low and can't provide as many amps, the motor spins slower. To combat this, we developed code that will read the current coming into the Monster motor shield (using pin A0) to see if the current has dropped. Once the amperage drops below a certain threshold, the code increases the amount of time the motor spins to make up for the slower speed. This ensures that the pedal will also be fully depressed, regardless of the amount of current the power source is putting out. The additional positive side of adding this code is that we now have a way to measure how power efficient our entire design is and we can make an alarm to sound when the battery is very weak. This will also aid in trouble shooting the system and finding faults, short circuits, lose wires, ect. The amount of current is printed to the console window for logging purposes.

Group Decision-Making
There were several instances of group decision making, but the most prominent we deciding on which concepts to use. To determine which design to go with, we created a design matrix for the pedal part as well as the motor mount.

For the pedal part we determined that the most critical characteristics we needed to evaluate are Design Simplicity, Responsiveness, Reliability, and Ease of Modification. Design simplicity refers to how much effort is required to create the model in CAD, print the model on the 3D printer, and install the part on the power wheel. Responsiveness is how much time it takes for the force of the motor to be transferred to the pedal. Reliability is determined by evaluating all the possible points of failures and how likely they are to occur. And Ease of Modification is how hard it will be for future teams to modify our original CAD model to work with a different type of motor. The results are outlined below:

For the motor mount, we used a similar design matrix to pick which concept to go with. The variables we compared this time were Design Simplicity, Adaptability, Stability, and Compatibility with the Frame. Design simplicity is how much effort is required to create the model and install the part on the power wheel. Adaptability is determined by how easily the mount can be modified to work with other motors, Stability is how much opposing force the mount can handle since a strong motor will have a lot of torque, and Compatibility with the Frame is how easily can the mount be attached to the frame in the rear of the power wheel. The results are in the table below:

Prior work in the field, standardization and reuse of designs (including reverse engineering and redesign)
We redesigned the frame that every subsystem needed to attach too because the original frame done by the previous group had some flaws in it. The new frame is based off of the one a previous group created, but with much needed improvements. The same Arduino, Arduino casing (blackbox), and wires are also being reused. The new frame also mounts to the existing holes that a previous group drilled into the seats. We are also reusing the math that the previous group did to calculate the amount of torque needed to depress the pedal.

Performance, life cycle cost and value
Since we are still in the design phase, real world test haven't been completed to evaluate the performance of the Autonomous Power Wheel. However, life cycle cost should be relatively low, with the biggest expense coming from replacing batteries. Value wise, most of the parts are made with wood or can be 3D printed, with the exception of the motor costing $15.

Aesthetics and human factors
Since this project is an attempt to remove the human factors from the power wheel itself, this does not apply to the project. On the other hand, one of our goals was to streamline our designs in order to make the systems more simple and efficient. In doing so, we made it very simple for future teams to implement our designs and maintain them. The motor mount for the pedal will accommodate a large selection of motors, and the sizes of the holes on the pedal part can be adjust by changing one value in solid works. The variables that should be adjusted in the code are declared globally at the top for easy modification. The idea of spraying plasti dip on the wood frame will prevent future team members from getting splinters while handling the power wheel and adds nice aesthetic value by matching the colors to the power wheel, giving it a 'finished' look. However, aesthetics are not a priority for us since we are still in the design and implementation phase.

Implementation, verification, test and environmental sustainability
After testing performance of each sub-system, the implementation phase can begin. First the new frame will have to be constructed and installed. The motormount should be next the piece installed (attached to the frame). After that the motor can be bolted down onto the mont and the pedal pusher can be attached the the hex nut on the motor. Once everything is lined up properly, it can be tightened down to ensure it doesn't move around during operation. The Arduinos can then be mounted and the wires ran to all the devices. The battery should be installed last. After everything is checked, testing can commence. During the testing phase, monitoring braking performance is critical. Environmental sustainability isn't a concern for us since we aren't using consumable parts (other than batteries) or pollutants.

Maintainability, reliability, and safety
Given the simplistic nature of the systems created a level of reliability can be established. None of the parts should wear prematurely and maintenance is relatively non existent other than replacing the batteries once the current drops below a certain threshold (The code will tell you when its time to change them).

Maintainability should be easy if the frame is build according to the Multi-view Drawing and the requirements

The reliability if the frame is also maximized due to the fact that it will be coated in Plast-Dip and it will be held together with L-brakets. The Plasti-Dip also improves the safety rating of the design since its resistant to fires caused by sparks and will eliminate the chance of splinters.

Robustness, evolution, product improvement and retirement
As each system was developed, it grew to fit the proper levels of complexity for its function. Also, several systems were redesigned multiple times to improve not only function but efficiency. As the project is very much still in the design phase, more care is being placed on the function of the systems themselves. This will ensure in the long run that the final product will be efficient and reliable.