Lego Racer

Challenge:

With a partner (someone different from the bottle opener and windlass assignments), design a 

vehicle with a single motor, powered by a PicoCricket, that can carry a 1.0 kg weight as fast as 

possible on a 4 meter course. You will use one of the old gray rectangular motors that does not 

have internal gearing, which will force you to experiment with building your own gear trains.

Design Process:

First, we tried placing the motor at the back of the vehicle near the wheels, adding a 40t gear on the back axle and an 8t gear on the front axle. We hoped that minimizing the number of gears while still creating the largest gear ratio possible (8:40 = 1:5) would increase speed and reduce friction. We connected the axles by wrapping a chain around the gears. A picture of our first iteration appears below. 

Our vehicle went pretty fast without the weight, but when we added the 1 kg, it wouldn't move. We learned a few things from our first iteration. First, the belt added a significant amount of friction with little benefit. Second, we didn't actually need to connect the two axles. If we built a gear train from the motor to one of the axles, the motion of the motor-driven wheels would turn the other set of wheels once the vehicle was placed on the ground. Lastly, while we thought we needed to create the largest gear ratio possible in order to maximize speed and win the race, we actually needed to find a gear ratio that was a good compromise between torque and speed. Without sufficient torque, the vehicle would not be able to haul the weight.

                          

After these realizations, we decided to place the motor in the middle of the vehicle and build our gear train from it to one axle. On the other side of the motor, we built a fence to prevent the weight from falling off the car. During the journey to our final iteration, the design of our car did not change much, but the gear ratio did. We tried several different ratios, documenting the speed of each in a chart. A picture of our final gear train (ratio 1:25) appears to the right. 

Above and to the left, you can see the fence for the weight. Pictured to the right is the entire vehicle with the weight and motor attached. 

Here is a video of our lego racer in action. 

Engineering Analysis:

All of the gear ratios we tried (after deciding to make a two-wheel drive vehicle) are shown in the chart above. With the first two ratios, the car would not move.  With the last ratio, the car was very slow. Our final gear ratio was 1:25.  We chose this ratio because it was the fastest of all the ratios we tried. 

                                              

Well Windlass

Here are all of our sketches that we made during the brainstorming process. We decided to go with the cubic structure because we thought it would be sturdiest. 

Here is our first physical mock-up. After building this and trying to wind the string around the pencil, we decided that the rod and holder needed some more work. We decided to put the rod through holes in the taller sides of the top of the cube in order to make sure we met the requirement that 10 cm of the water bottle be above the table. Putting the rod through holes instead of just setting it on curves in the frame would also ensure that it would not come out of the structure when we were cranking it.  We also decided that since torque = radius * force, we could increase the torque of the rod and pull the water bottle up faster by increasing the radius around which the string rotates. We decided we would make some kind of a spool to slide onto the middle of the rod in order to accomplish this. 

Here's a SolidWorks drawing of the base of our windlass. One thing we had to account for when designing our parts in SolidWorks was how different the dimensions of pegs and slots would be after the laser cut them out. We therefore designed the slots to be 0.2 mm smaller than we wanted to ensure we would have a tight fit. We also printed small testers for our pegs and slots to check the fit before printing the whole part. 

Here is the first iteration of our base. We soon realized after printing this that, since we had changed the side through which the rod would go, we needed to rotate the slots in our base. Otherwise, we would have to put the short side of the rectangle (11 cm wide) over the 12 cm gap between the tables, and that wouldn't be very sturdy. 

Here is the first complete iteration of our well windlass. At first, it didn't work because the rod was rotating inside the handle and the spool and the spool (made of 6 pieces) was sliding apart and the string would get stuck between two pieces. We drilled two horizontal holes through the spool and added piano wire to hold all 6 pieces together, and then we drilled a vertical hole through one spool and the rod and inserted piano wire to ensure that the spool and the rod would turn together. We also drilled a hole and added piano wire to the handle and the rod to cause the rod to turn when the handle was turned.  We bent all the ends of the wires sideways so that they wouldn't come out.

A close-up of our final design sitting over the "well." I worked pretty well, but the base moved quite a bit while we were lifting the water bottle. We decided to make a stabilizer to help with this. 

This is the part we made to stabilize the base. It went inside our windlass, between the tables, and clipped over the slides of the windlass. It provided a little stabilization, but was a little too loose to really help. With some more iterations, it would probably work really well. 

Here is our calculation of how much Delrin was used for a final product, approximately 490 squared centimeters. 

Engineering analysis: As mentioned above, the main physical consideration was how to increase the torque of the rod in order to lift the water bottle within the time limit. We accomplished this by increasing the radius of the rod where the string was by adding the spool. Another consideration that was problematic for some was the deflection of the rod. Since deflection is proportional to L^3, the length played a big roll in determining how much the rod would deflect. Our rod was pretty short, since it only had to be a little longer than the 12 cm gap, and therefore it did not deflect enough to cause problems for us. 

If given more time, we would work to make sure the stabilizer fit securely between the tables, and if we had a higher Delrin budget, we would add a second stabilizer. We also might work on making an equally sturdy design while decreasing the amount of Delrin used, perhaps by using a triangular design for the sides of the windlass instead of rectangles, like some of the other students did. 

Mechanisms

Pictured below is the slotted yoke drive. Starting from the position pictured, the disk rotates counterclockwise with a seemingly constant speed, and the peg pushes the slot and rod to the right. The rod then pauses momentarily as the peg travels up the slot. The peg pushes the slot and rod to the left. Then, the rod pauses as the peg travels down the slot. The motion repeats. Thus, the rotational motion of the disk and peg is translated into the back-and-forth linear motion of the rod. I like this mechanism because at first sight, its movement was confusing; however, after a minute of watching the video, I understood how it worked and thought it was awesome that purely rotational motion could create purely linear motion. This mechanism has been used in many internal combustion engines. I can also imagine an application for it in which bike pedals power the turning of the disk in order to provide linear motion for some specific purpose.

"The location of the piston versus time is a sine wave of constant amplitude and constant frequency, given a constant rotational speed." (Wikipedia)

Sources:
http://kmoddl.library.cornell.edu/model.php?m=446&movie=show
https://en.wikipedia.org/wiki/Scotch_yoke

Fastening and Attaching

The biggest advantage to heat staking is that it's permanent, so if you need to make sure that something absolutely will not move or disconnect, heat staking is the way to go. For example, if I wanted to build a rocket or a house and only had these three options of attaching the walls to the ceiling, I would use heat staking because I don't want my house to collapse on me or my rocket to fall apart or allow air inside (such as it might if you connect the ceiling with loose pegs and slots or piano wire) in space. The drawback of heat staking is also that it's permanent. If you realize you mess up, there's no way to disconnect the pieces and start over. Another drawback that my partner and I realized while working on our well windlass is that it may not work for complex pieces. You may not be able to fit certain pieces into the machine because of its structure.

Piano wire is useful for connecting things because it allows movement. You can use it to make hinges so that a part can swivel back and forth. It would be useful for making things like doors and windows. A possible drawback is drilling too large of a hole or a hole too close to the edge of a part and needing to start over.

Slots and pegs are very useful if you have parts that you want to be able to continually disconnect and reconnect. For example, my phone case snaps together so that I can take it off to clean my screen or remove my battery.  One downside to slots and pegs is that the laser cutter doesn't actually cut the dimensions you enter in SolidWorks, so the fit may be looser than you expected and the pieces disconnect unintentionally. This is something you have to account for when designing your part in SolidWorks.

Tight bushings would be useful if you want the bushing to move very little on its own, but still be able to remove it with a tool. For example, if you put a toilet seat cover on a toilet and attach it with a screw, you might put a tight bushing on the bottom of the screw because you don't want your toilet seat rotating all over the place. Loose bushings are nice if you want the bushing to move up and down the rod easily. If I have something like a piston that needs to move quickly and easily, I'd put it in a loose bushing.

Press bushings measured approximately 6.29 mm +/- .10 mm. Snug bushings measured 6.40 mm +/- .15 mm. Loose slots measured 0.138 in +/- .001 in. Snug slots measured approximately 0.1315 in +/- .0005 in.

Slots that were supposed to measure 0.135 in. measured (on average) 0.138 in. Slots that were supposed to measure 0.125 in. measured 0.1315 in. Slots that were supposed to measure 0.115 in. measured 0.121 in.  The peg measured 0.128 in. and fit best in the "0.125 in." (actually 0.1315 in.) slots. It's important to note that a 0.128 in. peg, for example, would not fit in a 0.128 in. slot. You have to allow a little extra room when deciding the width of the slot. You also must keep in mind that the laser cutter will cut slots somewhere between 0.003 in. and 0.007 in. wider than you specified in SolidWorks.

Using the drill

Using the thermal press

Bottle Opener 2/2

My engineering analysis of our bottle opener indicated that its max stress would be 69 MPa, which is one MPa under the yield strength for acetal. It makes sense that our opener chipped when we applied a lot of force. According to the equations for area moment of inertia and max stress shown above, in order to decrease max stress in our opener, we needed to decrease the applied force (which is harder to control) and the length while increasing base and height. Height is the most important factor since it is squared in the denominator of the resulting max stress equation. On our first iteration, we used the thinnest plastic, 1/8". For our next design, we went with the thickest option, 1/4". We needed to decrease the length of the opener without compromising ease of use, and we found that a 6.5 cm handle seemed like a good length. We also needed to increase the base, the length of the opener that comes into contact with the cap. Our previous design barely came into contact with the cap, so we decided to try a new design. 

This design has a small curved hook that goes under the cap and a long arm that provides leverage on top of the cap. 

A sketch with exact dimensions

The drawing of our part in SolidWorks

The final product

Yay! It works! (Video of me opening a bottle.)

The final step of our process was filing down the sharp corners so that they wouldn't hurt the user's hands. 

Our final design was very quick and easy to use, but if given more time, we could improve on its aesthetics by adding engravings or shaping it like something more interesting than a bottle opener. We could also brainstorm ways to make the handle more comfortable (which would probably involve it  becoming more cylindrical instead of rectangular and couldn't be done with the laser cutter) and add some convenience features, such as a keychain. 

Bottle Opener 1/2

The first step of our bottle opener creation process was brainstorming. We came up with several very different ideas. 

Here are a few ideas on one paper. The top left design has a circular hole on one end of the opener. The hole would go over one end of the bottle cap, hooking under the edge, and then the user would push down on the solid end to remove the cap. The bottom two designs have slightly different shapes, but are both meant to be inserted under the edge of the bottle cap at one end and then pulled up or down on the other end. The top right design is similar to a wrench. One end, a semicircle with the same diameter as the bottle neck right underneath the cap, would slide onto the bottle neck. The user would then exert a downward force on the handle to remove the cap. 

This design has two arms with teeth in order to grip the cap. One arm fits under the edge of the cap and the other goes on top of the cap. The user pulls up on the handle.

 This design is an expansion of the above design. It has a longer top arm in order to provide more leverage. 

This design is a third variation on the above two designs. The top arm curves around the top of the bottle and hooks under the other side of the cap. 

Ultimately, we chose the wrench design. After thinking it over, we decided that it would be very difficult to to get the measurements for the semicircle just right so that it would fit snugly on the neck of the bottle.

We modified the wrench design, changing the semicircle to a V. The V would allow for variations in bottleneck diameter and be easier to slide onto the bottle. Here is a foam model of our design.

After checking how the foam model fit on a bottle, we narrowed down all of the dimensions in order to draw our part in SolidWorks.  We decided that a wider handle would be less likely to break. 

Here is our first bottle opener. It was unsuccessful. It couldn't open the bottle, and if we applied too much force, it would begin to chip, We believe that it did not have enough contact area with the bottle cap and that it needed a stabilizer on top of the cap.