Hi everyone! Here are the GBX R&D updates from the past month:
Vibration Motor Revisions
Last month, the new vibration motor feed throat enabled printing with HID Global rPC flake, and multiple 1 hour vases were printed. However, when longer prints were attempted, the vibration motor feed throat design had some issues. The mount that holds the motor in place was held by an M3 screw on either side that fastened into heat set inserts in the feed throat, creating two halves that clamped together on either side of the motor. During printing, the vibration of the motor would slowly loosen the screws over time, which caused the sound of the vibration motor to gradually increase as the two halves of the clamping mechanism separated and vibrated against each other. Eventually, this loosening progressed enough for the motor to spin within its mount, causing the wires to tangle and pulling one of the spade connectors loose. This interrupted the circuit and the vibration motor stopped running in the middle of a print.
Vibration motor with twisted wires after clamp came loose (left) and Vibration Motor Feed Throat Rev2 with inserts for nyloc nuts (right)
To prevent the screws from loosening over time, the vibration motor feed throat was redesigned to accept nyloc nuts instead of heat set inserts to hold the clamping screws in place. Additional design changes were added, including:
- Two nubs inside the clamping housing to snap into the motor to better prevent rotational movement
- A cover for the wires and connectors at the top of the motor
Assembly with Vibration Motor Feed Throat Rev2 with Vibration Motor Mount Rev4
The new design was printed and installed a Gigabot X. With the new design, the motor no longer rotated within the mount, and the screws no longer loosened over time. However, a new problem emerged: the vibration motor was much louder than it was in the previous design.
During previous testing, the vibration motor became louder as the screws loosened, and the two halves of the motor mount were able to vibrate against each other. Therefore, it was hypothesized that sections of the new design were vibrating against each other and causing the loud noise. To troubleshoot, decibel measurements were taken after thin foam was added in different areas of the design. The decibel readings were:
- The ambient noise in the room, without the vibration motor running, was 47 decibels
- With the vibration motor running with no adjustments, the sound was 56 decibels
- After adding foam between the bottom sections of the clamp, the sound reduced to 54 decibels
- In addition to the foam at the bottom of the clamp, the connectors at the top were adjusted to point away from the cover. This reduced the sound to 53 decibels.
As a result of the testing, a new design was made to add tolerance between the two halves of the clamp at the bottom and top. The top cover geometry was also altered to give more space for the wires and connectors inside.
Side view of Vibration Motor Feed Throat Rev2 with Vibration Motor Mount Rev5 with spaces between the two components.
The next step is to test the new design and take some decibel readings to validate if the design changes reduce the noise of the vibration motor.
The hopper acts as a reservoir for printing material, enabling print times of 24 hours or more between refills of the hopper. While the current hopper design works reliably for most materials, especially pellets, some materials with general flow issues can become stuck while flowing from the hopper to the feed tube below. Most efforts to mitigate flow issues have been focused at the feed throat, where the material must flow from the feed tube into the extruder. This is what led to the development of the crammer and the vibration motor feed throat systems. However, some of the materials that have issues flowing through the feed throat (rPET water bottle flake, TPU pellets) have also had issues flowing through the hopper.
In response to this, an alternative version of the hopper was developed with a steeper internal angle and a more gradual constriction from the internal diameter of the hopper to the internal diameter of the feed tube. Previous testing of flow issues in the feed throat revealed that material flow was determined by:
- How smooth and fluid the internal geometry is
- The degree and rate at which the internal geometry constricts from one cross sectional shape to a smaller one.
Therefore, to maximize flow, the internal geometry of the hopper must be as smooth as possible, and the constriction of the material from the hopper to the feed tube must be as gradual as possible. The hopper’s internal geometry was already modeled as a smooth, curved surface, so the second factor was focused on. Since the feed tube’s internal diameter was already set, the hopper was instead altered to more gradually constrict the material to the ID of the feed tube. The internal wall angle of the original hopper was 25 degrees, and the new one was reduced to 15 degrees.
This resulted in an overall smaller hopper, reducing the volume from 8680 to 3170 cm^3, which is 37% of the previous volume. However, this smaller volume is still sufficient to hold enough Ultrafuse rPET pellets to print for 17 hours if printing with a 0.8mm nozzle at standard layer width (0.96mm), layer height (0.6033mm), and print speed (60mm/s) settings.
The new hopper was tested with a TPU pellet that had shown issues with flowing from the hopper to the feed tube during a 7 hour print. With the new hopper, there were no flow issues for 3 prints that were 7 hours each.
The new 15 degree hopper will probably not replace the previous due to its smaller volume, but it can serve as an alternative for use with materials with flow issues. It also serves as an example of how to optimize internal wall angle and hopper volume to meet a customer’s specific material and print setting needs.
GBX hopper with a 25 degree internal slope from the vertical (left) and with a 15 degree internal slope (right).
Hopper Gantry Channel Covers XLT
A few months ago, the hopper gantry channel covers for the Regular GBX machine were updated with the following improvements:
- Fastening points were strengthened with fillets due to a couple of these parts breaking during shipping
- Heat set inserts were changed from M4 to M3. This allows for a lower necessary tolerance to attach this printed part to the hopper gantry channels. This was previously an issue because this part is printed from polycarbonate, which shrank enough over the length of this part to prevent the fastening points from correctly aligning with their counterparts in the hopper gantry channel.
The covers were printed and successfully installed on a printer last month. After the successful testing, the design changes were replicated on the XLT channel covers.
Hopper Gantry Channel Cover for XLT Rev4 with design changes to strengthen fastening points and accept M3 heat set inserts.