The 3D printer has a compact size so that it can fit in a large backpack. It was essential for me that the printer has a reasonable print volume despite the requirement to be portable.
The outer dimensions are 300x330x105mm (folded), and the build volume will be 160x220x150mm (XYZ).
As you can see in the renderings, the printer will unfold during printing. The printer will be built out of CNC machined aluminum parts and 3d printed ones. I chose aluminum in the first place because I wanted to build the printer as rigid as possible (proof of concept).
To obtain rigid joints, the printer uses a combination of axial and radial bearings. Three screws then preload each axial friction bearing.
Einsy Rambo Board
small, low profile board
MGN9 linear guide
UHP-350-24 Mean Well power supply
350W 24V power supply which will be integrated into the printer
Thanks to Platformio I switched back to Marlin 2 because the compile times are much better than those of the Arduino IDE.
The printer now has a 12x12cm silicone heatbed, and I was able to install the OLED display.
The whole display controller assembly (3D-printed part) still needs some modifications because, at the moment, not all components fit.
Reinforced Z-slider I reinforced the Z-slider with another 3mm aluminium plate which I screwed onto the old Z-slider. Unfortunately, my tapped screw threads weren't very precise so that I couldn't use all screw holes to join the two parts together. With the new reinforced Z-slider, the printer is now at all Z-heights above the printbed rigid (no Z-flex when applying extra load in Z-direction). When The Nozzle moves under the printbed, the Z-flex starts to increase. Which for the printing process is unimportant, but can be a problem when homing to Z min (could result in homing position inaccuracy).
Backlash The Z-axis has backlash, which I didn't aspect to happen because the axis has a preload due to the scissor lift and the weight of the upper construction. The backlash caused the print artefacts in the first layers, which I mentioned in the last log because the standard PrusaSlicer Gcode moves the printhead at first some millimetres above the printbed before the printer moves downwards and then starts printing.
To solve the problem, I have changed the Gcode so that after homing, the printer moves only upwards, which is a short term solution, but backlash limits the possible features of the printer. Z-hop and auto bed leveling will be not usable.
Even homing is hard with backlash, for more accuracy I'm pushing the upper construction manually downwards before homing, to eliminate the backlash.
I tried the backlash compensation feature in Marlin, but it didn't work very well since the printer moves slightly upwards during the compensation process. The next thing that I want to try is to improve lubrication because I think there is already enough preload to prevent the backlash. Maybe some of you have any ideas on how to solve that problem.
Print quality looks good. Some test parts are getting weird print artefacts from which I don't know where they come. But I think this is not a Z-axis problem. I increased the belt tensions, which improved the problem. When I print with slower speed, the artefacts are mostly gone. But I'm still not sure what exactly is causing these artefacts. Feel free to write a comment If you have any ideas.
I managed to wire everything up. It is not the best wire management, but it works. I added a LED-strip at the front for better looks and usability. The power supply and EinsyRambo are mounted under the 3d printer between the Z-slider and the printbed.
In the back, there are a power switch, a USB-port and a power connection + fuse.
First print :
The first prints are looking very promising. I printed the test cubes with a 0.2mm layer height.
On the other hand, the first layers are still pretty bad. I lifted the printbed 6mm as I thought this would improve it, but it didn't. Without a heated bed, the print adhesion ist not good, which could be a factor as well. The Z-slider has some flex which leads to some looseness in Z-direction. It is made out of 3mm Aluminum(7075), but it is still not rigid enough. I want to try to reinforce it, maybe this will help.
Here is a video of the homing procedure. X and Y are homed sensorless and Z with an regular endstop.
With the help of Uzair Patel I was able to solve the non-linear Z-Axis firmware problem. I decided to use marlin 1 instead of marlin 2, because marlin 2 takes way longer to compile in the arduino IDE than marlin 1. I added the extruder and the Bowden tube to the printer.
And here is a picture of the "folded" Bowden tube.
On the bottom side are holes for high-temperature magnets which are required for the removable printbed. At the rear right corner is a cutout for the printhead, because the nozzle needs to be below the print surface so that different nozzles or print surfaces can be attached. Later I want to add a heatbed, but because of the cutout in the printbed, I can't use a standard one. I plan to build one by my self out of copper tape and capton tape.
I use 3mm MDF to insulate the printbed from the base frame.
All sidearms are mounted. The main structure is almost done. In the video, you can see the folding process, which looks very promising. The printer weighs already 4,3Kg, but the aluminum parts pay off because the construction is quite rigid.