What’s been going on?
Hi everyone,
Sorry that it’s been so long since our last post. There has been a lot happening behind the scenes at re:3D.
Helen has moved on to a different company, so we will try to live up to the amazing precedent that she set for these updates.
We will continue to post GBX updates here (on the Pellet Extruder page), but after today, these updates will focus more on engineering changes, GBX upgrades, community feedback, etc. We're also working on a lot of true research behind the scenes, so we’ll post those updates on the Gigabot Labs page.
The pellet extruder platforms (GBX, TBX, EBX) are becoming more commercializable, which allows us to use them for long-term research projects. Patrick posted a few months ago about a GBX for high temperature materials. We’re also working on the Gigalabs (yes, more than one) for Puerto Rico, the US Army and the US Air Force Academy, as well as printing from space waste and thermosets with NASA.
R&D
Since January 2022 we’ve had a GBX at The University of Texas’ Center for AM Design and Innovation as a testbed for R&D.
We added load cells under the heated bed to measure the rate of extrusion. This system uses (4x) 5 kg load cells that mount to the existing screws on the bed.
Figure 1: The load cells and enclosure mounted to GBX
Each of the load cells is connected to its own amplifier (HX711) and then to an Arduino Uno. Here is a great article that explains how this type of system works. Having each load cell connected to its own amplifier and pins on the arduino allows for easier troubleshooting.
Figure 2: Early CAD prototype of load cell circuit (left), real-life prototype (right)
The original Arduino code by Bogde was modified to include (4x) individual amplifiers and an output to microSD. Each amplifier output is summed to get the total mass being deposited onto the bed.
Figure 3: Individual load cell outputs (blue, orange, green, and red) and the summed output (purple)
Active feed assist (crammer) and rPET flake
We have doubled-down on our focus to print from recycled PET water bottles. This is the most frequent request from our customers and stakeholders. It is also one of the most difficult materials to print from because 1) it comes in all shapes, sizes, and compositions, and 2) it transitions back and forth between amorphous and crystalline depending on time and temperature [reference].
This post will focus on #1. The wide range of morphology makes it difficult to feed rPET flake into the GBX extruder with gravity alone. re:3D has experimented with a vibration motor and an active feeder (“crammer”) with varying degrees of success. Because of the figure below, we have decided to focus on the crammer rather than the vibration motor.
Figure 4: The vibration helps with consistency in extrusion (salmon), but not in achieving the throughput of pellets (green).
Using the crammer that Helen designed, we were able to print with rPET flake at the same rate as printing from analogous rPET pellets. The “1:1 Ratio” in the legend below just means that the crammer is forcing material into the extruder at the same rate as is being extruded.
Figure 5: rPET flake can print with throughputs as high as rPET pellet using the crammer
This shows huge potential for printing from flake and other materials that don’t feed well with the standard GBX hopper/tube/feed throat. We have aspirations of printing from flake at 1-2 kg/hr, and we think the active feeding solution is a step in the right direction.
GBX Hopper to Feed Tube Coupler
We’ve had an ongoing issue with the coupler that connects the feed tube to the hopper breaking. The associated part numbers are [12180] and [12302]. The have been multiple failure modes, including the coupler breaking at the hose clamp, and the coupler pulling out the threaded inserts in the hopper.
Figure 6: Broken GBX hopper hose clamp
We are replacing these parts with a new metal assembly. Stay tuned for more details! If you're experiencing issues with this part breaking, please put in a service request or email support@re3d.org for the new parts.