spindragon

This model was inspired by a ride at the Royal Adelaide Show. The arms that support the central chair section may rotate in the same direction or in opposite directions. The chair section can spin slowly either way and may do so while the arms are stationary or moving. Passengers need to be securely held in place for the inverted part of the ride. On the real ride it is not uncommon for unsecured items (keys, wallets, money) to fall out of pockets.

The model posed a number of interesting challenges. Probably the greatest was to get the chair section balanced. Unfortunately I did not achieve this to my satisfaction. The result was that although the chair spun it did so in a rather jerky fashion.

	The counterweights are not merely decorative. They each contain a number of coins to properly balance the weight of the chair section. I also had to pack the containers with tissue paper to prevent the coins from rattling.
	The green axles act as safety bars to hold the minifigs in place. These axles are connected to a common shaft behind the chairs. The common shaft is sprung in such a way (see photo below) that the safety bar may be raised to remove the minfigs or lowered to hold them securely during the ride. Click on the photo for a larger image.
	This is the spring mechanism used to hold the safety bar in place.
	One of the challenges involved the chair spin drive train. I wanted the chair section to be driven equally from each end. The distances involved resulted in a significant amount of axle twist. Therefore to make the drive truely symmetrical, the drive train was brought to the centre of the model (near axle in the photo) and then split to each tower via the 16t gears and the rear axle.
	I felt that a standard Technic universal joint at each end of the chair section may not have been up to the stress levels so I devised this half universal joint. Because it only moves in one dimension the mid-arm joint is forced to flex on both arms when they not in the same positions (see the second picture above). The down side to this structure was a lot of friction.
	The chair drive motor and the first stage of the drive train. 8t:40t reduction followed by 16t:24t using a chain (visible behind the 40t gear).
	The RCX was programmed to put the model through a variety of movements. I also built a manual controller using 3 rotation sensors (sorry - no photo) but this proved disappointing. The main problem was the lack of consistency of readings from the rotation sensors. Consequently they often failed to deliver a value of zero when the control levers were brought back to their home position.
	The light sensor in the left of the photo was used to detect when the right arm was in the home position. A similar sensor was placed on the left tower. The sensors detected the arm position using the light reflected from the 1x1 white beam below it. I found that the light sensors were only reliable when facing downwards. Other orientations resulted in too much interference from ambient light. This position fitted very snugly and was more a result of good luck than good design. The light sensor in the centre of the photo was used to detect the home position of the chairs. The vertical axle is part of the chair drive train.
	Part of the chair drive train as it emerges from the turntable. Also note the two 8t gears providing symmetric drive to the turntable. (See "Symmetric Drives" on the "Solutions" page.)
	This photo shows as much of the arm drive train as is visible. The rest is buried out of sight. The first stage of the train had 5 gears in a row - 24t:16t:8t:16t:24t. The 8t gear in the centre was on the motor shaft. One of the 24t gears can be seen to the right of the yellow beam.