18.2 - Project Prototype, Group 18

Kinematic Analysis:

The primary profiles of interest for our team is the position of the slider joint as a function of the input angle, as well as the force experienced at that joint as a function of the input angle.

• Position: Our plans are to have a total working space of around 8 inches above our bed/conveyor belt. We intend to achieve this by using a 4 inch crank that will revolve a total of an 8 inch swing across its full motion path, but this will be verified by our position analysis.

• Force: Our plans are to attach a probe at the slider joint that will hold a strain gauge or limit switch to determine when sufficient contact is made with the sample. Regardless of the equipment we opt for, we will still need our gauge/switch to be sensitive the the range of forces that we can expect the slider joint to be experiencing. This is where our acceleration analysis will come into play, since we can determine the force using the product of the mass of our probe (estimated to be around half a pound, generously) and the acceleration of the joint.

The results of our kinematic analysis is shown below:

Open

Figure 1. Kinematic Analysis of 4-Bar Slider Crank Mechanism

Open

Figure 2. Force Analysis at 4-Bar Slider Joint

From our analysis, we can confirm a total range of motion for our slider crank to be 8 inches, as shown in Figure 1, and from Figure 2 we can obtain the range of forces we need to be sensitive to in terms of contact forces on the probe.

• Geneva Drive: Also shown in Figure 1 is a plot of the input angle to our system as driven by the Geneva Drive as a function of time. This is important since we will be using the fact that a Geneva Drive inherently has a constant timing per cycle to calibrate our height readings. In other words, because the Geneva Drive’s dwell and motion periods follow a consistent pattern, we can determine the amount of time that has passed before our probe contacts the sample at a point and directly convert that timing into a distance reading by cross-referencing the position plot to our timing plot. It’s important to note that the period of the timing plot (i.e. time from dwell to dwell) is based on the angular velocity of the DC motor (from spec) since the DC motor will be a direct input to the Geneva Drive, but the slope of the motion interval will be based on the angular velocity experienced by the slider crank mechanism, which is 1/4 of the angular velocity of the DC motor due to the 1:4 gear reduction.

For our slider crank mechanism, 180 degrees of rotation would attribute towards moving the slider downwards, while the other 180 degrees of rotation would bring the slider back upwards towards the starting position. Currently, our design intends to use only 180 degrees of rotation (and then we will reverse the motor direction) in order to avoid the collinear alignment of all our links which would otherwise produce an indeterminate motion based on our Grashof Condition (Class III).

Open

Iteration Documentation:

Documented below are some of our previous iterations/design ideas.

The use of a Geneva Drive allows our slider to be split into distinct steps as the motor runs continuously. In our original concept drawing, the crank of the slider-crank system would be attached to the output of the Geneva Drive, and thus would only have an equal number of steps as there were slots in the Geneva Drive output. As a simple 4-slot Geneva Drive would only create 4 steps of motion in the slider, our initial design concept was to use two 4-slot Geneva Drives in series in order to increase the number of steps in our slider motion from 4 to 16. However, after thinking about how to implement this, we realized this would only translate the motion of the 4-step rotation across the Geneva Drives, rather than creating 16 individual steps as we desired. To navigate this problem, we pivoted to using gears to accomplish this instead. By fixing a 12-tooth pinion to the output of the Geneva Drive, and fixing our crank to a separate 48-tooth gear, we could create a 1:4 gear ratio that transformed the 4-step slider path to 16 steps. An image of the first-cut iteration of the Geneva Drive design is shown below in Figure 4.

Open

While the fundamental designs of our components were not modified during the manufacturing process, most of the iteration of our design went into adjustments of component alignment. When testing the fit of the Geneva Drive components and gears without securing them into a base plate, we initially believed the fit of the components was correct. After creating a testing board to secure the rotary bearings, we determined that the peg of the continuously-rotating Geneva Drive was not correctly aligned with the 4 slots of the Geneva Drive, preventing smooth motion. Because of this, we were required to recut the circular piece of the Geneva Drive after a better measurement of the alignment after the components were anchored, moving the peg hole towards the outer diameter of the circle by 3 mm.

In addition to the peg misalignment in the Geneva Drive, we encountered errors with the size of our shaft holes in the initial cuts of multiple components. The hole diameters were 0.02 mm too wide for our shafts and caused a lot of slippage between the gears and links, but this was easily remedied with recuts and superglue. Additionally, on the final design, we will switch from smooth shafts to a keyed shaft for all shaft/gear interfaces to limit the relative motion between the gears and the shafts. This will allow us to easily swap between gears should they wear faster than expected or need regular maintenance. The testing anchor plate can be seen below in Figure 5.

Open

The next component we iterated through was various link lengths to ensure the slider crank would operate as we expected. We utilized MotionGen to quickly test various link lengths that satisfy the Grashof Condition and still allow the necessary total length traveled. We arrived at the link lengths shown in our Grashof Calculations as seen in Figure 6.

Open

Figure 6. MotionGen animation of Slider Crank with Finalized Lengths

Our finalized prototype gear train is shown below in Figure 7.

Open

For our final version of the project, our team made the decision to change plans and replace the strain gauge with a limit switch for two reasons:

• Noise: A (cheap) strain gauge will experience a very large amount of noise in its reading. In order to account for this, we would need to filter the signal, which would add a lot of complexity, or apply a very tight cutoff value on the reading. However, if this cutoff value is too tight, we could potentially trigger a false positive due to even the slightest environmental changes (for example, people talking nearby or someone bumping our table would both cause additional vibration and lead to abnormal strain readings). Any false positive would then result in the wrong height being recorded, therefore contaminating our final result. A limit switch can avoid this issue since it is less sensitive to vibration.

• Fragility: Cheap strain gauges are very fragile and will break easily. This could lead to issues with lead times and potentially stall our progress. Additionally, even if it doesn’t break completely, repeated impacts can still lead to shifts in performance and may lead to the strain gauge reading a different range of strain values, which would require us to constantly re-calibrate our values and would not produce a consistent result. Limit gauges are typically much more sturdy and reliable.

Bill of Materials (Draft):