Alpha Prototyping

To test the performance of the roller bed conveyor, an apparatus was assembled using cardboard, plywood, paper, a 9V battery, and a 6V, 15RPM DC motor. The conveyor tread turns as the rollers turn due to the amount of friction between the belt and the roller wheels. This turning motion of the belt causes both rotors to move in the same direction. This setup allows the team to confirm and evaluate important factors such as incline angle and the time to deliver a LEGO up and off the belt. A 3mm wooden dowel was run through a cardboard roller bearing and secured on the ends with additional cardboard that was cut in circles. The roller was then attached to the cardboard frame and strung with a conveyor tread made of poster border to simulate a slotted, rougher surface. A cardboard membrane was applied between the middle of the frame to provide support to the treading and prevent slack. After completing the frame and attaching legs, the DC motor was then attached to the dowel at the top of the conveyor. Both the motor and the 9V battery were connected to an on/off switch to toggle the circuit on and off. The motor was held in place during testing to keep rotation about the dowel instead of the motor itself. Additional foundation was added to the bottom of the system during testing to evaluate the conveyor at about a 20 degree angle from the floor, the maximum operating angle that the team calculated. A picture of the conveyor apparatus is shown above.

The biggest challenge the team faced with this prototype is the frictional interferences that occurred between belt surfaces, The initial rollers that were used for this design consisted of two bottle caps layered with rubber bands. Ultimately, these rollers paired with a cardboard frame that was not perfectly straight caused a lot of interference with the movement of the belt. This often made it difficult for the motor to apply enough torque at certain points along the belt, sometimes failing to move the belt entirely. To counter this, the team revamped the rollers using thick, cylindrical rolls of cardboard and establishing some separation between the rollers and the frame. By eliminating these tight fits and faulty components, the roller bearings became much more free to move and required little torque to spin. Removing this frame contact allowed the motor to successfully move a group of LEGO parts up the belt and into a bin below. Test runs with this system made it evident that small to medium sized LEGO pieces would take approximately 8-10 seconds to travel from the bottom to the top of the belt. The team concluded that this is a suitable time frame for each piece to give the remainder of the device adequate time downstream as to not cause an overflow of parts. Since the apparatus was tested with an approximation angle of 20 degrees off the ground (using a protractor), the team concluded that a 20 degree incline was achievable while near the maximum allowable angle. It is likely that the team will plan for a more optimal angle about between 10 to 15 degrees to increase part stability on the belt and add additional rise time to further prevent an overflow of LEGO.

To test the performance of the vibration plate, an apparatus was configured using cardboard, a vibration motor, and a 9V battery. The vibration plate forms a 90 degree V-shape at a slight angle between 5 and 10 degrees to utilize gravity and motor vibration to string out large groups of LEGO (single-file) without overflowing the camera bay. The plate rests on a horizontal area that is cut away from the rest of the foundation with room between the area and the supports. This allows the vibration plate to vibrate independently from its foundation and the rest of the device, preventing the vibration from affecting other parts of the system. This allows the team to evaluate critical features of the subsystem, such as incline angle, plate length, travel time, and number of required motors. For this apparatus, a piece of cardboard was cut and bent in the middle to form a 90 degree angle. The piece was then placed into cardboard supports that would connect the plate to a flat surface. A vibration motor was then taped to the underside of the resting plate. This motor could then be attached to the 9V battery to simulate how the LEGO parts would move and interact on the vibration plate. A picture of the vibration plate apparatus is shown above.

The team ran into no problems with the construction and testing of this alpha prototype, but had some key takeaways. The team concluded that the majority of small to medium sized LEGO took approximately 3-4 seconds to enter and exit the plate assembly. This travel time is slightly quicker than expected, so the team proposed to decrease the incline angle to about 5-10 degrees to slow down the LEGO parts and ultimately increase the time through the subsystem. In the testing with one vibration motor on one side, the group found that the parts entering the system took longer than anticipated to separate and form a line. With this in mind, the team tested and confirmed that a second vibration motor on the other side of the plate would help to separate the pieces even quicker. Doing so allowed the team to keep the length of the plate, since adding the second motor gave the parts sufficient space to organize themselves before entering the camera belt. In upcoming tests and prototypes, the team would like to test the behaviors of the motors against different materials such as plywood or ABS plastic to evaluate how much this affects plate vibration.

Originally, the inclined conveyor belt was going to be smooth and the LEGO pieces would be moved up the belt with friction. Now, given how circular the profile of many of the LEGO pieces are, it has been determined that it would be better to have an inclined conveyor belt with ridges on it so the LEGO pieces would get stuck in the ridges rather than having to rely solely on the friction between the LEGO pieces and the belt. The original prototype for the conveyor belt, seen earlier in Alpha Prototype 1, was made of cardboard with a paper belt. This new alpha prototype is made of cardboard. The material does not matter as much any more for the alpha prototype because the normal force of the belt, not the friction force, will be moving the LEGO. It should be noted that the actual beta prototype will have a proper flexible plastic conveyor belt.

In this test a belt was constructed with one long, rectangular piece of cardboard and many small strips of cardboard glued on top of it and equidistant intervals to simulate the ridges. This was held up against a wooden board that would be one of the hopper walls. On this board are five marked angles at 15°, 20°, 25°, 30°, and 35°. A small, cardboard back wall was made to hold the LEGOs at the base of the conveyor. Additionally, a small flap made of paper added to the bottom of the back wall as to prevent the LEGO pieces falling down the conveyor due to the gap caused by the ridges on the conveyor belt. The cardboard was then pulled upward at an angle corresponding to the ones mentioned earlier while LEGOs were pooled at the bottom.

It was found that between the angles of 15° and 30° there were no issues with pieces all pieces moving along the conveyor, but at 35°, some pieces would continually fall down the conveyor and not reach the top. The reason the group wants the conveyor at the steepest angle possible is to save on space. The less inclined the conveyor is, the wider the hopper is going to have to be and the more difficult it will be to fit this device on an average desk. Additionally, the steeper the incline, the more the parts separated as gravity made it so they did not stack on top of each other as much. This would help to separate them more as they went through the machine.

Testing was conducted using plastic food storage containers as the collection bins. Two containers were used, one with dimensions 5.5” x 5.5” x 2”, and another with dimensions 5” x 3” x 2.5”. Alpha prototype testing with the arm had the pieces dropping from a height of about three inches and resulted in minimal bouncing as the pieces fell into the bin. Further tests of dropping pieces showed that pieces bouncing out of an empty bin only became a problem at a drop height five inches above the container. As such, the team determined that LEGO pieces bouncing out of their collection bins was not a serious concern as long as the disparity between the height of the conveyor belt and the collection bins was minimized.

Testing to determine the desired dimensions of the collection bins was conducted using the 5.5” x 5.5” x 2” container, as it has greater storage capacity than the other container (60.5 sq in. and 37.5 sq in. respectively). The container was able to hold 134 assorted LEGO pieces sampled from across all six categories. With six collection bins planned for use, this would allow the device to hold approximately 804 pieces. However, the EV3 core and expansion sets contain 523 and 853 pieces respectively, so the device would only be able to sort one set at a time before needing to be emptied. To improve upon this, the team decided to find alternative containers with similar lengths and widths but greater heights to improve upon the storage capacity without needing to extend the belt length.

Additional Prototyping Models and Modifications to be Introduced in Phase V and Beyond!