Week 4 Assignment: 3D Printing

Design and 3D print a small object that could not be made subtractively.
Scan an object using a 3D scanner.

This week's group assignment entailed printing the design rules for the 3D printers in the EECS lab. Completing this task identified printer limitations in terms of various model characteristics such as overhang length, overhang angle, infill, etc. Below depicts a number of the design rules from the Prusa’s slicer program (left image) and the final products from both printers to the right (black are Prusa products and yellow are WOX products). I examined the products from both printers and will discuss observations below. Note, I have only included images of WOX products below, because the yellow PLA provided details that were more easily captured by photograph than the black PLA from the Prusa.

This design rule depicts the hole and wall thickness capabilities of the printers. The thinnest wall both printers were able to produce was 0.5 mm. Both printers began to create opening with a width of 0.2 mm, but did not produce an opening along the slit’s entire length until the width was increased to 0.4 mm.

This design rule depicts the degree of incline the printers were able to produce without supports. For both printers there was significant distortion visible at an incline of 0 and 10 degrees from the horizon and minor distortion visible at an incline of 20 and 30 degrees. The auto-support generation feature on both printers is far more conservative. The Prusa and WOX printers auto-generate support for any overhangs less than or equal to 45 and 60 degrees from the horizon, respectfully.

This design rule depicts the required clearance to ensure different components will print separately. Again, both printers performed similarly with a minimum clearance of 0.3 mm to ensure component separation. However, at this clearance the ring did not slide freely along the rail. For my design development I will plan to use 0.4 mm as my minimum clearance to ensure component separation.

This design rule depicts the finish appearance produced by each printer. The layering of PLA is clearly visible in the products from both products, but this can be addressed post printing with light sanding or exposure to acetone. What was more interesting to me is that the component from each printer fit tightly together. This demonstrates similar dimensional accuracy of both printers, and that PLA shrinkage while cooling is minimal and does not need to be accounted for during the design process.

In preparing the MakerGearM2 for printing I started by cleaning the glass bed with acetone and reviewed the automated print settings for the installed PLA medium. I noticed the bed and extruder temperatures were set higher than the values used for the Prusa and the WOX in the EECS lab. The MakerGearM2 was set at 75 deg C for the bed and 220 deg C for the extruder, verse 60 deg C and 200 deg C for the other 3D printers. After conferring with others that had used the printer before, I decided to proceed with the higher values. I recognized that to successfully print my Ball-N-Ball design support structures would be necessary. However, I was concerned with the ability to remove support structures from the interior of my design, and the undesirable appearance left from removing the quantity of support structures generated by the MakerGearM2's slicer program. For my initial print attempt, I opted to manually generate support structures. The below series of images depict the manually generated supports and the resulting print, which was ended early. The printing of the outer ball progressed smoothly, but during initial construction of the inner ball it tipped over on its supports and fell on its side.

All in all, my manually generated supports were insufficient to support the inner ball. However, I did find that the support structure from the lower portion of the outer ball to remove seamlessly. After discovering this I decided to give the auto support feature a try during the next print. This time there was success with supporting the inner ball throughout the printing process, but the final product was densely packed with support structures that seemed rather daunting to remove.

Now began the tedious task of support removal. As one would expect, removal of supports from the interior proved difficult, but doable if outfitted with a sound pair of needle nose pliers. I am fairly content with the final product, except the spiral design resulted in enough play of the outer structure that the inner ball could be removed. I guess I should have made the inner ball a bit larger. I would be hesitant to narrow the opening of the outer design spiral due to concerns of not being able to remove the inner support structure with needle nose pliers.

Trust auto generated supports; they are designed for seamless removal.

EECS lab is outfitted with a SenseV2 3D scanner. The system’s scanner is shown to the right. By moving the scanner slowly around an object it will create a 3D rendering of the item. The system is capable of imagining individual objects as well as an entire body. However, I experienced difficulty while attempting to scan objects smaller than a fist. The system’s imaging program allows images to be edited by setting boundaries to remove surroundings and fill/erase tools to manipulate the image. I am in the market for a new pair of boat shoes and decided to scan one of mine. The below image shows the scanning in progress and the final solidified model.

The 3D scanner performs poorly with smooth reflective objects.
Place object to be scanned on a contrasting, non-reflective background for best results.
Objects must of larger than a human fist.
Slow steady movements of the scanner are a must.
Recommend hanging smaller objects for a true 360 deg image.