Dear all,
Following up on my last post “Grasshopper Gcode to 3D printing”.
I got a problem printing this Gcode I created with Grasshopper:
test2_gcode.txt (3.8 KB)
2D_pattern.gh (20.4 KB)
I got a problem regarding the extrusion:
Does somebody know how to fix this ?
Thank you !
Hi @Leblond
just from looking at your generated gcode I see two potential problems:
I made a comparison with some gcode generated by cura, you can see the E value constantly increasing, as well as the first layer actually being printed at a height of 0.2mm
Dear @lando.schumpich,
Thanks a lot for your advice !
I’ve corrected my file. However I found out a difference between “Relative” and “Absolute” positioning.
It seems Ender 3 printer works under relative positioning. It basically means it is depositing a certain amount of material between each points. On the other hand, I noticed a “M82” commands on a file made in Cura which apparently switch from Relative to absolute. In that case, what you describe seems to be the right method…
Thank you for your precious help !
I have only ever used absolute mode for custom gcode and I have not worked with ender 3 yet, but rather mostly with ultimaker printers. It is generally allways a good idea to compare the output of a known and trusted slicer source with your custom gcode.
Did you get to work with only changing the Z height, as the nozzle might be to close to the build plate to properly extrude material your original code example?
Got it ! Yes, I changed the Z height and it is definitely better. However, I do have a problem of flow, the printer is struggling (The wheel turns in a vacuum and makes noise as if I asked it to load too much material while printing). I tested it on an other curve and here is the result:
test2_gcode.txt (76.1 KB)
It is now a matter of flow and extrusion that I don’t know how to solve… any ideas ?
Here is my Grasshopper file:
2D_pattern.gh (36.1 KB)
I’m not sure to understand the link between the nozzle size, the flow rate, and the parameters I’m playing with in the grasshopper file…
Hi, great to hear that it works better 
Extrusion is dependant on a multitude of things:
I used this formula with good results in the past:
Thank you for the link ! So now, I’m not sure how to play with these parameters in my Gh file… for example, at no time did I mention the size of my nozzle or the flow (%)… any leads ?
The flow % is a value between 0.0 and 1.0, which you need to play with a bit in your experiments. Typically 60% (0.6) is quite ok as a start, but I don’t know your machine, so YMMV.
I put the variables in your script and hopefully wired it up correctly, have a look:
2D_pattern.gh (38.5 KB)
Looks great !!! It is, I suppose, exactly what I wanted 
I’ll do some printing and will get back to you quickly.
Thank you so much !
Hey @lando.schumpich,
After making few prints, I still have the same problem, the wheel turns in a vacuum and makes noise as if the flow rate was to high… I tried between 0.05 and 0.9… but the problem is the same… it feels like the prints are very thick.
Do you have an idea on how to solve this ?
Again, thank you
Do you think it could be due to the Arcs ??
I would need to look at your file again how you handle arcs.
I think i messed up the formula a little by not including the filament diameter as written in the link. Although thats interesting you stillbget tye same result even when setting flow percentage to 0.05.
Maybe the problem is rather the speed of the move? How does the movement look? ‘Regular’ or too fast? Does this only occur on arcs or also on straight line segments?
Yes, it is very similar. I’ve tried with different setting but it’s not changing that much in any way.
From what I can see, the printed layer looks very very thick and quite raw, I never saw that before with a regular slicer.
When printing, what I see when printing is an irregular deposit as if the the motor was trying to extrude even more. I also see that every arcs are all spread out. Indeed, the speed might be too fast. Even thought I do not have so many straight lines, I think the printer appreciates them more.
So I updated the E formula in the script, it should give more reasonable values now, just make sure you put in the filament diameter you are actually using.
Since you convert all your arcs to line segments anyways, there shouldn’t be any issues with that.
A good test always is to just print a one layer rectangle and also slice that in a known working slicer, so you can compare the 2 gcodes, which might give you further hints.
From the photos the prints definitely look overextruded, I hope that this fixes itself with the new formula
2D_pattern.gh (35.1 KB)
The values are changing quite a lot ! I’m testing it right now
Thank you !
