GBX Update May 18 2022

Hello everyone. Here are the R&D updates for GBX from the last month.

Open Hardware Summit - Material Testing Procedure

On Apr 22, 2022, Helen gave a presentation on an open source material testing procedure at the 2022 Open Hardware Summit. For a recording of the presentation, see this link. The presentation followed the Material Testing Procedure for Pellet Extrusion Knowledge Base article, and it covered the following topics:

• Background information on pellet 3D printing
• The challenges behind developing a material testing procedure
• Extrusion testing
• Optimizing print settings
• Tensile testing
• Documentation resources

GBX Klipper Configuration

Klipper configuration has continued for the GBX 2 beta with Klipper, a touchscreen, and a 32 bit board. The first beta unit finished its 100 hours of testing and was shipped out to the customer, and the second beta unit is currently undergoing pretesting. During the testing of the first and second beta units, the below additional issues were found and resolved. For a list of issues previously identified, refer to last month’s forum post 

The testing for the second GBX2 Beta is expected to have fewer issues, since the fixes for the issues identified with the first beta unit have already been applied to the second unit.

GBX Klipper Quick Start Guide

To support the future roll out of GBX2, a new version of the GBX Quick Start Guide was created to incorporate the Klipper touchscreen interface. The Klipper-specific Quick Start Guide is available for download on re:3D’s website here.

EMC2 PQR rPET Testing

After the success with other rPET samples from the Engineering Mechanics Corporation of Columbus (EMC2), GBX material testing was continued with EMC2’s PQR rPET flake. This flake was created from PET water and beverage bottles from forward operating bases (FOBs) that were processed through EMC2’s containerized rPET recycling process. If the material could be printed on GBX, it would serve as a proof of concept for a fully circular process of converting Army PET waste into valuable products.

Particle analysis was conducted on the sample, and the results show most flakes between 1 and 5mm in length. While this particle size distribution in other regrind samples has led to successful prints, it doesn’t guarantee successful material flow into the extruder.

Particle analysis image for EMC2 PQR rPET, with a 2” long calibration cylinder.

ImageJ particle area analysis graphs for EMC2 PQR rPET

Tests were conducted to determine maximum extrusion rate and extrusion consistency following the protocol outlined in the Material Testing Procedure for Pellet Extrusion Knowledge Base article. The results revealed a decent extrusion rate of 0.207kg/hr, but a high standard deviation of 0.38g across five extrusion trials with an average mass of 1.92g. This indicates inconsistent extrusion.

Further testing was paused on the PQR rPET flake because our collaborators at EMC2 communicated that they have a sifting protocol for their flake samples. Their flake is sifted first through a screen with ⅛” (3.175mm) holes, then through a screen with 1/12” (2.12mm) holes. The flake that passed through the ⅛” hole screen, but did not pass through the 1/12” hole screen, is to be used for further material testing.

Board Temperature and Fan Duct Testing

Last month, various methods were explored to reduce the temperature of the Archimajor board components during the printer’s operation, including a fan duct and altering the firmware. Those changes brought the board’s steady state temperature down to 57C during operation, which is below the maximum 65C temperature recommended by the manufacturer. While this allows the printer to operate in the short term, for the long term we want to reduce the max board temperature further for two reasons:

1. Electronic components have a shorter lifespan as they are exposed to higher temperatures over extended periods of time
2. Issues associated with shifting and motors losing steps can be reduced by increasing the current supplied to the motors, but increasing the current increases the temperature of components on the board. To increase the current we can supply to the motors, we need to increase the cooling at the Archimajor board.

Therefore, a testing plan was created to test different methods of decreasing the board temperature. A FLIR camera will be used to quantify the effect of implementing any of the below approaches

• Position the fan duct to blow across the mosfets at a 45 degree angle.
• Use a larger fan with a higher CFM
• Add heat sinks to the hottest board components (the mosfets and motor drivers), and test heat sink orientation both parallel and perpendicular to the direction of airflow.

So far, initial testing was only conducted for the first method. The testing was conducted on a Terabot with an Archimajor board, which has overall lower board temperatures compared to GBX because it has different motors that require lower currents. However, initial data for the Terabot board temperatures can indicate which cooling method may be effective on GBX, too.

Mosfet and motor driver temperatures for Terabot with an Archimajor board given different fan duct positions.

Unfortunately, altering the position of the fan duct and the airflow direction didn’t have a significant effect on the final steady state Archimajor board temperature. The next steps are to get initial data on the impact of the other cooling methods.