various updates

Published on 4 July 2024 at 20:28

It has been a while since the last update but a lot has happened. 

Flow optimization

Since the last update I have done multiple iterations of the flow converter inside the laser. This was done for several reasons:

  • Improve cooling of the laser
    Before I had to reduce the air flow of the laser fans to prevent that the powder got blown away, this was a compromise between the cooling of the laser and the powder that was being blown away. During the winter and spring this was no problem because it was relatively cold but now that summer is starting the laser would overheat with the low flow.
  • Reduce powder disturbance
    The higher and not optimal flow was disturbing the powder bed and this was especially problematic during larger surfaces as the powder bed would no longer be perfectly flat at the end of the layer.
  • Keep good welding gasses removal
    Having too low flow around the point where the laser is melting the powder will cause the loss of laser power and increases the risk of contaminating the laser optics (in this case the protection glass).

Change the coreXY motors to nema23

The current nema 17 0.8nm motors were struggling with the high RPM combined with the higher temperatures of the summer. I already knew that the motors were very critically mentioned but i was hoping i could postpone the change to the next version of the machine. I'm currently almost done with the transformation to the new motors. For reference the nema17 motors had only 0.35nm left at 900RPM while the new motors will have still 1.2nm left at the same RPM.

Experimenting with Varying layer heights

With 0.05mm the machine produces already quite nice details and surfaces but since the acceleration is limited it is not utilizing the laser power optimally.
With 0.1mm the machine produces quite rough surfaces but the laser utilization is a lot better.

Luckily OrcaSlicer has a feature that allows you to print the fill and walls with different layer heights. This allows for the walls to be printed every layer at 0.05mm while the fill is only printed every 0.1mm layers. I have found that this is a very nice compromise that results in the surface quality of 0.05mm but almost with the same speed as the 0.1mm layers.

What is next

First i have to finalize the transition to the nema 23 motors.
After this is done I will start to do some more process development to improve the top skin of the prints. The walls are very nice and overhang skin is quite OK already but i will elaborate more about this in the next update.