Background

This mechanism was tested on 30 Jul 2021

Testers: Ethan Cronier, Taylor McNabb

The purpose of this testing was to see how mechanism functions and NOT how well it was able to perform any of the chemical tests. The soil sensor was also not tested, and the boards used to control the mechanism were not tested either. Limit switches were disconnected and the lid to the mechanism was not tested.

The dirt used was gathered in the moment from outside of the SDC. The biggest challenge in gathering this dirt was to not look like a right weirdo stuffing soil into a McMaster-Carr bag... The dirt consisted of organic/fibrous matter and seemed to contain quite a bit more clay than sand. The dirt was also quite most and had the tendency to stick to parts.

To control the 2 DC motors and the shovel servo motor, the dual motor controller that was used in the initial design was used and the buttons on the board where used to control both of the motors. The servo was controlled using an arduino mega and some jank servo code. The mechanism was able to do its task as designed, whilst on life support only (requiring frequent human intervention). Many things jammed or broke or needed a nudge these problems will be discussed after the video of the mechanism working is shown.

Issues

Issues will now be listed in an unordered manner.

Belt

The belt seemed to perform well, we had no issues during testing (this round) one thing that was noticed was that the belt was twisted on one side due to interference. The belt contacts the main plate which causes the belt to twist due to the contact being at an angle. This issue was also noticed in the CAD where the belt model clipped through the plate model. This did not hinder the testing, but I found it interesting to see and thought good to mention it.

Dirt Dirt Dirt

When the dirt is picked up and released, not all of it will make it into the test tube with this design, as a result dirt will end up everywhere. This might not be an issue for future designed, but it looks incredibly unprofessional and risks cross contamination. One potential hazardous location for dirt to end up in is in the bearings or pulleys as shown in one of the images below. This should be considered for future designs.

Sticky Dirt

With the dirt tested, the dirt had a tendency to stick to the inside of the shovel. Otherwise this scoop worked extremely well and reliably, very impressive. In the image below, if you squint you can see that dirt is stuck in the jaws of the shovel.

Geneva Mechanism cam roller

The piece shown below has a setscrew in the side which is used to interface with a D-motor shaft. During testing this piece frequently popped off of the shaft. This is due to the set screw loosening in the plastic. A simple fix would be to switch this part to metal if this design were to be iterated. Another cause for concern would be why did the part pop in the first place? did the mechanism jam? it is difficult to differentiate these phenomena.

Test Tubes in Mechanism

This pertains only to the insertion of the test tubes into the mechanism. These were a pain in the ass to put into the mechanism. And once inserted, the tubes were extremely loose, as seen in the video below. An alternative to as what is shown in the first view videos at the top was to wrap tape around the test tubes and press them flush with the rotating carriage. This was sufficient for testing purposes. In future designs it will be important to ensure the test tubes are easy to insert into the mechanism.

Top Bearings

Bearings at the top of the mechanism were all loose, did not seem to impact performance, something worth noting.

Dirt into Test Tubes

This is potentially the biggest issue for the entire mechanism. Even when pulverized dry dirt is attempted to be put into the test tube by a pouring action, no dirt will enter the tiny hole in the plastic cap. This issue will need to be corrected for in the next design.