II. Gear Geneva Mechanism

One of our design goals was to have a mechanism that would provide a "snapping" motion which triggers a fast jaw biting motion after the user has placed a coin on the mouth of the coin bank.

Additionally, we needed a modular way to power the Jaw mechanism without relying on knowing what the final design would turn be. This design philosophy led us to develop an integrated Gearbox and Geneva Timing mechanism that provides extra gearing for torque plus the mechanism needed to provide the desired snapping motion 

The gearbox needs to accomplish many different goals:

  1. Provide the Jaw Mechanism power in a flexible manner
  2. Provide extra torque if needed
  3. Give the jaw the ability to have regular jaw motions and to "snap" at a particular configuration.
  4. Enable an easy way to reload the jaw to prepare for the next "snap" 
  5. Have a flexible way to control the snap strength

Version 1: Naive Approach with Issues

To accomplish (1),(2) and (3), a modular gearbox implementation seemed obvious.

Since the crank of the Jaw mechanism only needs a torque input, any power transmission that gives this torque would be acceptable. (1) is accomplished by making the gearbox modular, and it powers the jaw mechanism via a timing belt.

The gearbox can be mounted anywhere so long as the L-brackets can be fixed. The belts can be tension by simply moving the gearbox mounting point on a slot. The output of the gear is coaxial with the pulley of the gearbox. 

(3) is accomplished using a unique type of geneva mechanism. In essence, the final output of the gear is modified such that a) it is spring loaded somehow and b) it has missing gear teeth to decouple the input from the motor and let the spring "fire" the output gear.

This unique gear configuration enables the mechanism to "snap" at any desired configuration since the start and end of the "snap" motion is dictated by the position and number of missing  gear teeth.

The idea was that (4) was accomplished by the virtue of the design. After the mechanism snaps, the mechanism is reloaded again since the teeth of the gears would re-engage with the input gear. However, the design did not enable this action.

(5) is accomplished In this design by having a wheel with holes for the extension spring. The idea was to have multiple holes that go out radially which the extension spring can mount onto. This enables changing the preload force of the spring to provide more snapping force if needed. 

Manufacturing 

Only requires a band saw, a laser cutter, and a lathe. The band saw is to cut the aluminum shafts, the laser cutter is to cut the two gearbox plates and gears, and the lathe is to make the standoffs that rigidly connects the gearbox together

Design Issues

There are two main design issues. The first is that (3) and (4) had design issues so the snap motion and its reload capability was not achieved at all.

The second was that even if (4) worked, the snapping motion would cause high impact on the output and input gears.

 

 

Version 2: With Full 360 degree Implementation

 

In this design, we tried to address the issues of (3) and (4) not working. We still use the missing teeth method to control the starting and ending snap points of the mechanism. However, the way we load the mechanism is completely different.

We now utilize a lever (light wood color in the image) that has a toggle position. When the toggle position is passed, the mechanism is ready to provide the snap force and motion. The snap force is provided by the extension spring.  The extension spring is connected to one end of the lever and its other end can be mounted to different locations on the gearbox.

Manufacturing and Assembly

We still use the same manufacturing method as in version 1.

To ease assembly, we locate the gears and fix the different components using shaft collars. We also use 3/8" delrin bushings on the shafts to alleviate friction.

Finally, the laser cutter provides the accuracy we need to axially locate the shafts for minimal friction misallignment.

Design Issues

While now all design objectives have been obtained, the output and input gear are still undergoing high-impact after the mechanism has snapped.

 

Version 3: Gear Integrity Preservation  

In this design, we address the issue of the gear integrity. We also improve and simplify the manufacturing and assembly of the gearbox.

To address the gear integrity we separate the spring loaded output and the missing gear teeth into two separate gears.

The missing teeth is now an idler gear. The output gear is always loaded and will fire when it is no longer engaged with the gear with the missing gear teeth. Since we have full control of everything but the preloaded output gear, we can choose to engage the output gear whenever we desire.

The last final design change is using T-slots as standoffs. The laser cutter is quite accurate so here we use it to create custom and accurate standoffs for the gearbox.

Assembly Issues and Fixes

Wood Warp

The main assumption of all of the gearbox designs is that the laser cutter will cut out the exact piece we desire. However, the laser cutter has no knowledge of warped/bent wood. This warping can cause inaccuracies on the final hole cutouts as well as the expected width and length of the part. It is important to keep this in mind when designing.The main fix was to find very flat sections of the wood that will be used for laser cutting.

Axle Alignment

We had to use washers and spacers on the standoffs to align the gearbox plates such that axle misalignment was minimized. We were successful in locating this and had low friction shafts.

D-shaft and Wooden Gears incompatibility

To mount fix the rotation of the gears on the shaft, we used a D-shaft which required that we had D-shaped bores on the gears. Initially our gears were made out of 1/4" wood. However, the wood we were using was processed so the aluminum D-shaft "ate" out the bore over time. Thus the D-shaped eventually was more circular after a few test runs.

To solve this problem, we started using acrylic gears since the acrylic material was denser than wood.

Gear Manufacturing

When we were initially manufacturing the gears, we only had access to 1/8" acrylic. This was too small. So we first tried stacking the gears together to make 1/4" thick gears. This provided problems as now the mechanism required the alignment of twice as many gears. This caused some gears to get stuck and many teeth were broken. Additionally, since we were using D-shafts, there was also the issue of bore play if the laser cutting tolerances are not correct which worsen the gear play.

In any case, this was solved by using a 1/4" acrylic from the beginning, which prevented the need of stacking gears.

Final Design Implementation

The gearbox has a gear ratio of 40:16.  With the Polulu motor's 131:1 gearing, this provides an overall ratio of 327.5:1. 

From the matlab's conservative calculation estimates this gear ratios are more than enough for our application. When we tested version 3 with the rest of the mechanism, the gearbox was able to provide the reloadable snap motion required as well as the power required to power the mechanism.

In the end, the snap motion was very powerful and the design of the tongue caused the coin to be thrown back out to the user. Due to time constraints, the gearbox had to be simplified to simply provide power to the mechanism. We removed the springs as well as the missing teeth so in the end, it was simply a gearbox with a belt transmission that provided torque to the crank of the jaw mechanism.

From the design goal perspective, this implementation was a complete success.