Week 9:

The design of the project has been finished; the code works, the LEDs work with the sensor, the return system is in place, and the project looks fantastic. However, there has been trouble with the actual motion control of the board, especially with the first batch of servos being too weak to move the board. Upon ordering the stronger servos, time was lost simply just waiting for the servos to be delivered. Once here late week 8, many ideas were tried for movement of the board. Beneath the board, there are two dowels attached to a half of a plastic circular ball which were to be attached to the servos directly for a "Pull" style of motion using rope to attach the servo to the post. This failed, but ideas from this were implicated for other designs. Next, a smaller dowel was attached to the servos for an extended leg of the servo which would allow for more distance when rotating. The problem with this is that the dowel beneath the board and the dowel attached to the servos differ in size greatly, thus loosing a lot of torque from the servos, and basically putting the rotation of the servos to waste as every turn was killed by the longer dowel. This idea was also scrapped, and ideas were thrown out for a third attempt. Finally, the idea of removing the star attachment and drilling a hole in a dowel to be attached directly onto the gear. Using this, rope was attached to each side of the dowel, and the ends of the wire were attached to each end of the project. The idea behind this is a rotation "Left" will pull the left end of the board down, and a rotation "Forward" will pull the board down in the back, and so on. The motion mimics the motion of the Wii Nunchuck based on the orientation of the board and the Nunchuck. This idea worked initially; the board rotated left and right, forward and back with the nunchuck, however the range of motion proved to be not enough for the project to move enough for the ball to go from start to end. This idea seems to have great potential in actually working, but due to time constraints, rigging this idea up to work correctly seems impossible, but attempting to fix this will happen before lab for sure. Possibly attaching more wires to the board and the dowel will allow for a stronger pull in all sides of the project. If this is to fail, we will bring an actual Labyrinth board in to show an example of what the project should have been. 

Figure 1) Shows the board balancing with a rubber band attachment. This is used on the left and the back sides of the board, which holds the board level.

 Figure 2) This is just another picture showing the board being held up by a rectangular pattern of strings, which allow for less contact with the ball, which in tern allows for easier rotation.

 Figure 3) Here is a video of the board being rotated with the Wii Nunchuck and the third attempt at rotating the project with the elongated gear attachment.

Week 8:

This week, trouble struck. But before this news comes up, positive information will be relayed. Most of the codes are functional, including the motion control for the Nunchuck, and most of the Lifebar sensor. The intent of the lifebar is to take a bar of life off each time the ball roles over a sensor back to the first level. With these functional, the problem of assembling the motion of the board has really stumped the group. The board has been weighed so that is rests evenly on the pivot point, yet the servos must be attached in a better location than they are now in order for the rotation to actually work. Once this is fixed, the project can be covered with plexiglas and tested with an audience, which can occur very shortly after lab this week. Also, the servos that were ordered have a much stronger motor within with will actually be able to move the board.

Mid Week 8:

There has been a hitch in the plan. The servos that were supposedly to be used for the project are too weak to move the board. Newer and stronger servo motors have been purchased which move vertically rather than rotationally. As soon as these arrive, more work on the physical board will ensue.

Week 7:

This week, the group worked on much of the physical board. The gluing of the board was done, which created a solid and sturdy game board with the use of biscuits. Once glued, a reset was discussed. Something simple and out of the way was needed, which wouldn't prohibit movement and would allow for the installation of a motion detector at an exit hole. The idea devised was multiple hole drilled into the support beams of the undercarriage of the board with a plywood bottom. This way, when a ball is to fall through a hole, a catcher beneath is utilized to bring the ball back onto the first level beneath the first ramp. The detector will be placed within the final hole in order to take life away from the ball if it happens tho have to go through the reset.

The next aspect to the game that was created was the pivot point. Utilizing half a plastic ball from a group in the juggling engineering lab, a round pivot cup was created. The idea devised was to gather a good bit of sand and put it within the plastic ball and create a cover to trap the sand within. This is because the ball is rather fragile, as it tends to crack easily. By filling the ball to the brim with sand, it helps counteract the constant inward pressure due to the weight of the board. Next, the actual pivot point was created. A simple wooden box was created and lined with packaging tape and then filled with sand. The reason for this is simple: the plastic ball rotates very easily with the soft mound of sand around it, and once within the box, the box becomes very heavy. The weight is needed to keep the servo motors in place while controlling the movement of the board. Though the servos have yet to be attached, ideas have been devised, and they soon will be put through trial and error to see which will work the best, as the motion control has finally come into fruition with a working code for the Wii Nunchuck to be utilized.

The Ball used to pivot the board.

 The Box and the Ball in action.

The Nunchuck-servo code in action.

Week 6:

This week entailed more woodwork, and more coding. In class, corner joints of the labyrinth were glued together. The technique used to do this was biscuiting, which can be found in the tutorials tab. However, before the other side was glued for the opposite corner, it was discovered that there were measurement errors prohibiting the cuts to actually work with the boards if glued together. The problem was some of the pieces were too long, which would allow the actual board to fall through the base of the project. Upon revisiting member Tyler Fox's house to redo and fix the boards, the project looks better than ever. The intent for this week is to glue to boards together and finish the coding aspect of the project due to the arrival of the final materials needed for the project to work, followed by an outer casing of the board.

Side View 1 Prior to Glueing

 Top View Prior to Glueing

 Side View 2 Prior to Glueing

Week 5:

During this week, coding for the Arduino was attempted. Many hours of work had taken place with trying to get the Arduino to intake the motion of an accelerometer and translate it to the two servo motors controlling the x-axis and y-axis of the game, however none of the efforts yielded useable code for the game. However, by the end of this week, the code should be completed. Once written, the setup will be created, as most of the electronics have arrived such as the breadboard, wires, and sensor. The only problem has been the package office at one of the Drexel Housing facilities; most of the orders have arrived, dating back to last Wednesday, yet the package office has yet to go through the shipments. Also, on Sunday, members Tyler and Jesse went out and cut wood for the outer box of the board, which will hold the board together and allow for a plexiglas top to cover the game itself so that the ball within will not be allowed to be tampered with while playing.

Week 4:

After class on Tuesday, members Steven and Jesse went to the machine shop to get wood cut for the actual board. The bases of the maze were cut into three pieces of equal sizes (8" by 12") and numerous pieces were cut to form the inner walls of the maze which each vary in size greatly. The Arduino program was downloaded, but the code has yet to be finished due to lack of knowledge of Arduino language. The pieces that had been cut earlier in the week were then taken to another woodshop to make them look better for the final product. With this done, the inner walls and the connecting ramps were attached, completing the interior design of the project. Another thing to note, extension cables and another Servo motor arrived in the mail, which will help us figure out the set up for the design and for the Arduino code.
 The refined design

 attempting to code the Arduino

 After cutting the wood

 Extension Cables