All-metal, portable, high-speed, low-cost advanced multi-tool desktop factory derived from LulzBot Taz and TazMega. Specifically designed as an easy upgrade path for existing Taz machines. Personally, this upgrade will be applied to my old Taz3 machine, to provide redundancy and additional throughput for the TazMega.
- All-metal, no plastic mounting brackets.
- Belt-drive X/Y, ACME threaded rod Z.
- Composite metal structure with timber reinforcement. Maximum rigidity at minimum cost.
- Simple assembly, 8 wheel minumum, no carriage plates, no spacer blocks.
- Support for piercet’s bed plate, among several others.
- Gantry plates are the only custom hardware required.
- Specifically designed as a low-cost upgrade for existing Taz machines.
- Specifically intended for use with MightyTool.
Please examine the design closely. Alternative subsystems and innovative simplifications have been included.
GitHub Repo: https://github.com/mirage335/TazStiff
Total BOM cost is <$650 from, mostly from OpenBuilds as follows.
- ~$100 - Gantry plates from eMachineShop.
- $156.90 - Cast Corner Bracket, Black Angle Corner Connector, 90 Degree Joining Plate. More expensive than 3D printed components, and possibly other metal parts sources, but rigidity is especially important here.
- $105 - Wheels and eccentric spacers. Some of the wheels and eccentric spacers are probably unnecessary, being included as extra margin for the first prototype.
- $41.50 - ACME lead screws. Probably the easiest place to save money, as a variety of other leadscrews could be made to perform equally well for the Z-axis given appropriate high-compression anti-backlash nuts.
Are you planning on using a RAMBo or some other controller? What software will you be using to slice models for the MightyTool (subtractive operations vs. the additive 3D printing operations)?
Planning to keep the RAMBO electronics, with this build. Might upgrade to a custom Duet/Duex inspired design later though. Similarly, planning to use Slic3r to start with, though it would be great to finally compare other modern slicers. Especially MatterControl, which the SeeMeCNC folks recommend highly.
Although the rigid table underneath should compensate, using corner brackets for the Y-axis results in a slight twist that has proven difficult to eliminate. TazMega uses plates for this, which should definitely be used for the TazStiff as well in future builds.
Assembly mostly complete, held up by lack of tee nuts. Also, plastic and metal 90deg joining plates have been directly compared. - https://youtu.be/4F2My8FK1iI .
Mechanical assembly complete. Belts and electronics assembly to go.
Again, the lack of large flat joining plates on the Y-axis has bit me. Future TazStiff builds really should use the same plates as TazMega. Now, Marlin does not correctly calculate the bed compensation matrix.
Initial bed topography, as measured by differential optical probe.
Like TazMega, and despite the lack of flat joining plates mounting the Y-axis, initial alignment is actually quite good. Bed is flat to ~0.25mm.
Still, twisting is the dominant error term, which MarlinFirmware cannot compensate. TazStiff’s rubber foam and deck screw mounting to the table underneath should be able to correct this.
Finally starting to get really decent prints.
Photos attached. Video posted on YouTube - https://www.youtube.com/watch?v=hoJDu2Lae4E .
Print speed for the second part was 120mm/s.