All 3D printers print in a series of 2D planes stacked up to form a 3D shape, never moving along the 3 axes at the same time.
In fact, it only moves in 1D (while moving upwards to start the next layer) or 2D (while printing in the same layer)...
On the other hand, non-planar 3D printing is based on 3D paths, which gives a much smoother 3D print.
Check the results yourself:
Don't think I should mention which is non-planar:
The slicer doesn't make 2D layers anymore, but 3D curves:
What do you think of it?
I am very interested to see where this advancement in the technology will take us, for it is a change in one of the very basic parts of the principle, and thus potential for large ramifications (in a good way)!
I'm very eager to see it, and even more eager to try it and mess around with it when I can have my hands on a 3D printer again!
I think applying this technique with a machine purpose-built for it could make for very promising work.
Where instead of the machine being optimized and mostly limited to linear motion, perhaps adding the possibility of rotational motion of the printhead or printed object, could make for very differing and perhaps better (in some aspects) prints...
I am very curious and excited to see!
~Sjaak
Certainly an improvement. But, it is limited to some parts as evidenced by the measurements taken of the print head at the start (short description and video link here).
The technique basically duplicates the cutting path a CNC mill would follow when cutting a similar shape from a block of material.
The remaining hurdle is mainly the FDM machine on a whole. FDM equipment extrudes a rather thick "stream" of material, and traditionally has a thicker Z layer thickness compared to other 3D printing technologies.
A secondary problem is the stacking of extruded FDM layers. If the machine is outputting a cylinder of plastic, it makes sense that the resulting shape is like a figure 8 in profile. For a better visual, consider stacking logs to make a log cabin. The "layer lines" in a cabin are part of the problem with FDM equipment. Convert the nozzle to extrude a flatter layer of material, and the problem can be reduced.
There are advantages in speed when depositing more material in thicker layers. But the drawbacks include more obvious layer lines, and the possibility of de-lamination along those layer lines.
Check out those "layer lines" from stacking round logs:
Make the layers flatter, and the problem starts going away.
This is a terrible image, but it's the first I found. Instead of a round FDM nozzle, use a flat FDM nozzle:
What do you think of a "six-axis 3D printer"?
More nicely smoother finished sides. And prints without overhangs and adding support.
Do you like yourself?
The nozzle has always the same position. Just like the finisching layer on the non-planar object. No overlaying layers.
Movement object is more in the 3D space, while the head stays in de 2D space.
When I print 4legs table model so that top was against the board and legs goes up. And I make typical layer by layer print gcode where was so many empty travel paths between the legs. Then I wondering why is software so primitive.
If software knows printhead , moving frame and nozzle dimension. Then it's possible calculate faster paths. So that nozzle will not collide printing object. Example when printheadnozzle height is 15mm then you can print one leg with circular path to height 13mm and then go print next leg 13mm and so on. And when all legs is 13mm height then software start next 13 step.
The G code on the printers we own can be changed to make the airfoil in the above example. I looked at the model and found that only the top few layers were printed with all three axis changing. Another problem when you go to the referenced study that made the wing is that the printing heads are not build to have Z travel. The maximum slop can not exceed the physical slope of the printing heat cone which for most heads is shallow. This is because they want to conduct heat to the tip.
The real problem is adhesion between layers. CNC Cook on youtube does studies of 3D printing. Adjust parameters and measures the strength results. He has looked at layer adhesion. Could not present an approach to improve it.
The only way to improve layer adhesion is to heat the plastic already laid down just prior to adding more plastic, but not heat up beyond just the surface layer. Hot plastic will creep due to gravity away from it's original position. I have been thinking how I can heat just the surface prior to laying down new plastic.
If you don't receive the email within an hour (and you've checked your Spam folder), email us as confirmation@grabcad.com.