Ok guys, after months of inactivity due to general disillusionment I'm trying to get

Ok guys, after months of inactivity due to general disillusionment I’m trying to get this NorcalReprap Prusa I3 calibrated so we can finally get some decent prints out of it.

This is the standard 20mm test cube sliced with Slic3r at 0.2mm layer height. There’s significant distortion on the two right hand corners, along with excess height in these corners. Dimensionally the part is not bad on the two good corners, being around 20.05mm at the top, and 19.95mm at the bottom, the other corners and distortion make it more like 19.5mm so significantly out there.

You can also see that the top has some weird artifacts in it, besides just the grooves in the layer there appear to be some holes of some kind.

Any advice on where to go from here?

Cura settings are:
Layer height: 0.2
Shell thickness: 1.2
Bottom/top thickness 0.6
Fill density: 20%
Print speed: 40mm/s
Printing temp: 230c
Bed Temp: 110c
Diameter: 1.78mm
Flow: 100%

Retraction speed: 20
Retraction Distance: 4.5

Initial layer thickness: 0.2
Travel speed: 150
Bottom layer speed: 15
Infill speed: 60

Marlin settings are:
#define DEFAULT_AXIS_STEPS_PER_UNIT {80,80,4000,727}
#define DEFAULT_MAX_FEEDRATE {100, 100,2, 45}
#define DEFAULT_MAX_ACCELERATION {9000,9000,10,10000}

#define DEFAULT_ACCELERATION 2000
#define DEFAULT_RETRACT_ACCELERATION 1000

#define DEFAULT_XYJERK 20.0 // (mm/sec)
#define DEFAULT_ZJERK 0.4 // (mm/sec)
#define DEFAULT_EJERK 5.0 // (mm/sec)

You mention the printer has been sitting for a while. Is your ABS (judging by the temps) dry? Those holes look like they may be from steam “popping” out the nozzle. Are they anywhere else?
Otherwise, this looks a little hot (this is ABS, right? otherwise, VERY hot), and a bit overextruded. Is your extruder calibrated, and did you measure the filament and enter that in your slicer?

I’ve heard Cura tends to leading top layer artifacts like that, unless the part is actively cooled.

Hey Carlton,

Well, yeah, the ABS has been out for a while so that could be a factor. But we’re in Pasadena, CA here, and its pretty dry, never much above 20% humidity here!

Honestly though even when this filament was fresh we had similar problems. But I just remeasured the filament and if anything it appears to have shrunk, looks like 1.73mm now so I’m running another print now at 220c and will report back in a few.

So that didn’t seem to help any, made it worse if anything.

https://plus.google.com/u/1/114825475221343681660/posts/5NdJPLxEJqy

Looks now like the errors are on both sides, and the issue where the right side seems to tall is even worse than before. Going to try re-measuring the e-steps again to see if that’s the problem.

Any other suggestions?

Just had a thought, could this be some sort of pressure build up issue?

As I understand it the printer slows into the corners, giving built up pressure in the melt zone more time to splurge out.

Lowering the temp on the hot end would have actually increased the pressure wouldn’t it? If so, then that would explain why the corners are far worse now than before.

Finally…3 cubes later and I got some decent prints out of it.

Turned on the fan and the corner issues went away, also modified e-steps, for some reason the gregs wade extruder is outputting about 14% less plastic than the calculated settings indicate it should. Still, this seems better now, time to move onto overhangs and bridges.

Your height seems way lower than 20mm. Is your Z properly calibrated? If not you need to that first before sorting out extrusion issues.

I print one of those “cubes” for each type of filament and slicer I use. I print them with 100% infill and a layer height of 0.2 mm (since it is a factor of the “cube” height of 10 mm). I make sure the first layer is also 0.2 mm and that the nozzle is just touching the bed when at z0 and at the center of the bed. I don’t use a fan.
Then I measure the actual height with calipers and divide that into 10 mm to calculate the flow adjustment, which I set in the slicer settings.
Different slicers seem to need different flow tweaks. E.g. for my 2.85 mm diameter blueberry ABS, the tweaks measured this way turned out to be 101% for Cura 14.03 and 0.95 (95%) for Kisslicer 1.4.1.4, so Kisslicer appears to over extrude relative to Cura.

@Jasper_Janssen I honestly didn’t notice that but you are right the cube is coming in around .5mm too small. I’m confused though since its threaded rod on the z axis and the calculations in marlin are correct. Wondering where I went wrong.

@Kit_Adams so you adjust flow to adjust z height?

@Ben_Delarre Are the photos very distorted? It looks like it’s closer to 10 mm high than 20.

Oh its a 20x20x10mm cube. So not actually a cube at all! Sorry should have said!

@Ben_Delarre
No, I know my z axis steps per mm are exactly right (calculated and measured results agree).
I use the measured height of the 100% infill “cube” to calculate the flow adjustment for each slicer and filament.

I find measuring the filament diameter is difficult to get accurate enough, and also the results still depend on the slicer you use. With 100% infill, if there is over extrusion the extra filament has to go somewhere and in a “cube” that shape it results in a domed top.

I measure the height to the top of the dome (if present) and find that gives good results when used to calculate the flow tweak. At least my print quality increased a lot after I started doing this.

Huh interesting idea. My z couplers failed today do I’m going to have to start over entirely but I will try this when I get to that stage. Thanks for the tip!