CAD Modeling and Material List
While designing the device in Solidworks, we realized that the system would be easier to manage if there were two cams rather than a single cam with the inner profile mapping to one direction and the outer profile mapping to the other direction (as in the initial concept). Therefore, the initial CAD model shown below depicts stacked cams. The planned follower was a 1 mm diameter shaft. The working principle was the smaller the follower size, the higher the resolution of the output sketch of the end effector.
The final CAD assembly can be accessed here: Sketching Mechanism CAD.
Initial CAD Model
Final CAD Model
Design Choices
In order to make rapid, iterable prototypes, we 3D printed and laser cut most of our parts. Additionally, we used a mechanical protoboard to quickly change the layout of the parts without having to build or reconfigure the structure. As we iterated through design and experimentation, we learned piece by piece how to fix problems and improve performance. For example, we increased the force transmission between the cam and the follower by switching from low-friction bushings to linear ball bearings. This change, in addition to using two pillow blocks and linear ball bearings per follower rather than just the one shown in our initial CAD assembly, greatly reduced the flexure on prismatic joints and mitigated the risk of binding up as the cam rotates through steep angles.
Next, we saw that the rod linking the elbow joint to the end effector was bending as the follower moved back and forth, so we added gussets to the elbow joint to ensure that the end effector would stay in line with the joint and more precisely transmit the deflection experienced by the follower. The end effector performance was further improved when we changed from the single block design to the flexible two-link design where each link is connected to the two end effector rods by linear ball bearings. These two links meet in the center and are connected by a rotary shaft with its own rotary ball bearings. This two link design allows the end effector to overcome vertical misalignment in the supporting rods by reducing constraints on the mechanism. Essentially, the flexibility of the end effector tolerates and adapts to irregularities instead of locking the linkage system in place as the single, rigid end effector would.
Finally, we moved away from the 1mm rotary shaft as it was too prone to catching in the cam. The rotary shaft slid on the cam (rather than rolling) because 1mm rotary bearings are not easily accessible. We switched to a 5mm shaft supported by 5mm rotary bearings. This decision reduced the achievable resolution of the sketch, but it allowed the system to function more smoothly.
Purchased Material List
This list contains all the materials incorporated into our final design that required external purchase. All other materials used in the design were either provided through the class (Arduino, motor driver, encoder, etc.) or self-sourced from the Maker Studio and personal resources.
Material | Qty | Source |
---|---|---|
MPU-6050 accelerometer | 1 | Amazon |
Spring assortment kit | 1 | Amazon |
8mm ID linear bearings | 12 | Amazon |
8mm diam, 200 mm long stainless steel linear shaft | 4 | McMaster-Carr |
5mm diam, 200 mm long stainless steel rotary shaft | 2 | McMaster-Carr |
5mm ID rotary bearings | 10 | McMaster-Carr |
1/4" thick 12"x12" acrylic sheet | 1 | McMaster-Carr |
Welcome to the University Wiki Service! Please use your IID (yourEID@eid.utexas.edu) when prompted for your email address during login or click here to enter your EID. If you are experiencing any issues loading content on pages, please try these steps to clear your browser cache.