SPACECAL Research Project Report

Design for Manufacturing, Taylor Labs, University of California, Berkeley

Authored by Tristan W Schwab, Undergraduate of Mechanical Engineering April 30, 2021

The scope of this report is to discuss the progress of the SpaceCAL Project aimed towards the Zero-Gravity flight test in the Spring of 2022 and my contributions to the project and Design for Manufacturing Group during the Spring semester of 2021.

Figure 1

NASA Tech Flight Proposal

The purpose of SpaceCAL is to develop a compact enclosure containing 5 parallel computed axial lithography (CAL) printers to be tested in suborbital testing with the Zero-Gravity Flight Demonstration. The system is projected to fly in the Spring of 2022 and complete several varying viscosity resin prints.

SpaceCAL is inherently a technology demonstration of the current CAL technology developed at the University of California, Berkeley in microgravity; however, alternative scopes of research may apply. Suggestions have been raised to the enhancement of research in additive manufacturing for in space manufacturing, including: 1) Part strength versus fluid velocity through shadowgraph analysis in resin vials. 2) In situ automated post-processing of CAL prints. 3) Testing of a temperature control system in a space environment for low-gravity resin printing. 4) Tracking low-viscosity polymerization in low-gravity. 5) Performance of CAL for micro-gravity bioprinting.

The SpaceCAL project can ultimately demonstrate not only the unique abilities of CAL AM technology but shows potential to provide invaluable research in the growing field of in-space manufacturing.

As mentioned, the SpaceCAL project is scheduled to fly in the Spring of 2022 (next year). During the launch of the project in January, primary focus has been development of an enclosure to contain the CAL system, electronics, optics, and hardware considerations, and the design of a compact and exchangeable vial stack (fig. 1, right). The planned design has evolved since the initial Flight Tech Proposal. First, original documentation discussed three sets of vial stacks, containing 22 hydrogels, 11 high, medium and low viscosity resins. The current design now contains 5 vial stacks of 5 vials each. Change of the number of vial stacks and vials affects the number of resins which will be used in flight. Second, the project will no longer consider the use of “Schlieren imaging to record video data of the refractive index history”, but rather, Shadowgraph methods due to the high sensitivity of Schlieren imaging.

Current State of SpaceCAL and Contributions

SpaceCAL exists in an early design stage on Solidworks. Full purchase orders for the primary assembly include hardwire, including but not limited to 8020 beams, linear guide rails, projectors, optics, and electronics.

The author was established as the Mechanical sub team lead, tasked with rebuilding the first iteration of the SpaceCAL system on Autodesk Fusion 360. The author rebuilt an 8020 enclosure (Frame Assembly) and performed finite element analysis on Fusion 360 to analyze g-loads on the frame specified by Zero-Gravity.

The author used these analyses to confirm the material selection of the frame assembly and add triangulated supports the frame.

Progress then shifted to rebuilding the SpaceCAL system on Fusion 360 (due to the limited capabilities of Fusion 360 and parametric modeling). The author collaborated with Grad Student Lead, Joe Toombs, to develop a parametric sketch in Solidworks and develop the second version frame assembly. Current focus has now turned to the development of optic setups for particle tracing, continuing summer research positions in the Design for Manufacturing group, assembling the SpaceCAL system, and collaborating with fellow future Cal Grad Student, Taylor Waddel, on mechatronics and software. Future project objectives are expected to shift towards resin formulation and characterization.

The SpaceCAL project certainly arrives at a remarkable time for the College of Engineering at Berkeley to be involved in the growing excitement for exploration and industry in space.

Visualizing Stress Distribution in 3D Printed Lattices

The first portion of this article showcases my final project in PDF format. My first prototype is shown below the PDF.

Click here to view the final prototype video.

TristanWSchwabFinalProject

First and Second Prototype

Project Description:

For the initial prototype of this project, I demonstrate the unique complacency of lattice structures designed and optimized through NTopology and manufactured on an Ender3 Pro and Formlabs3 using elastic materials. I showcase my design process, my thought process in building lattices in NTopology, and my process to build an interface to visualize force distribution through a lattice.

Click here to see the prototype demonstration. 

Designing a Lattice:

There are multiple platforms for designing lattices. I selected NTopology, a software used in the industry and readily available on a student license. NTopology is unique for its interface with AM and easy “block” UI, which proved to be very efficient when learning about different lattice structures and adjusting parameters.

Figure 1: Fluorite and Body Centered Cubic Lattice Structures generated in NTopology after importing a CAD Body

Printing Process:

The first CAD bodies that I designed had small voids which I envisioned could house the force sensitive resistors. This idea would have likely worked, but printing on my Ender3 in TPU, an elastic filament, proved that printing any of the lattices with the support structures for the voids was not easily scalable, and unnecessarily overcomplicated the design. Ultimately, those prints did not turn out to be the best, and I decided that generating a simple rectangular prism lattice without voids would be the best solution. 

Unfortunately, the files I sent to the Jacobs Center to print on their FormLabs SLA printer were unusable. I wish I had given more thought to the initial design so that I could use an SLA print for a comparable demonstration, but I submitted my initial design prematurely. This print was made in elastic 50A resin, which may have been a bit too elastic for the purposes of this project.

Figure 2: Example of TPU failed lattice print with voids
Figure 3: Failed SLA print.

It turns out that the best print is the simplest design. This cannot be more true when it comes to printing elastic lattices, which fundamentally behave like springs. I tried a variety of lattice designs, Weire-Phelan, Kelvin Cell, Isotruss, Fluorite, etc… all of these lattices are nearly impossible to manufacture without support structures. I prioritized visualizing the cells themselves, so printing at a low density was a heavy consideration. The best lattice for my purposes was the Body Centered Cubic Design, which does not present overhangs greater than 45 degrees which would necessitate printing supports. 

Circuit Design

To begin my circuit design, I set up one FSR embedded in a lattice which I had on hand. Once I generated the right print, I tried embedding two FSRs accompanied with LEDs. I was having some trouble with getting the LEDs and FSRs to stay connected with the jumper wires, so I decided to solder end tips to all of them and plug them into the female-male jumper wires and wrap them in electrical tape for final presentation. 

Figure 3: Front and back faces of body-centered cubic lattice embedded with LEDs and FSRs.

Final Setup

When I was initially applying a force to the front face of the lattice, it was actually expanding outward as a result of the Poisson’s Effect. It turns out that this effect was so great that the corner FSRs were not picking up a force, since the upper face of the slots I cut out from the lattice were lifting upward. To counteract this, I built a foam board frame to contain the lattice. This made the entire setup a bit more portable and pleasing to look at, though it took away the side view of the lattice and made it more difficult to see the bottom face force response. 

I’ll be honest, this setup looks like a jumble of wires and trying to make the wires clean and orderly was an extremely difficult aspect of this project. After a few tries, I decided to separate the wiring for the left and right sides of the lattice, and this effectively cut my odds of miswiring in half. I also color coded the jumper wires as practice to improve visibility for myself and viewers.

Figure 6: Side view of foam board slots for FSRs and LEDs.
Figure 7: Split wiring from left and right sides.
Figure 8: Arduino board using all analog ports. Voltage source from battery pack.

Figure 4: Of course, the code.
Figure 9: Side view of foam board slots for FSRs and LEDs