Slic3R issues?

I've been trying to print myself a spare extruder body using Greg's Mk3 design (Thing 8252). The SCAD files compile fine, and looking at the STL output from OpenSCAD in Netfabb Basic that looks fine as well. However, the Gcode that Slic3R produces (using Pronterface to review it, and having actually printed one!) seems to have some errors. The idler nut trap nearest the gear side of the body is blocked by a too-high print where there should be clearance for the idler bolt, and the support block for the idler hinge does not actually meet the hinge projection. I'm quite prepared to accept that this is operator error, but I can't find what I'm doing wrong! Any thoughts or suggestions, anyone?

I've also had problems finding a herringbone gear design that will print, either from STL or SCAD. Again, suggestions welcome! I'd like to have some spares in stock before something breaks...

I don't think that it is a printer problem as I have been able to print something like the Panelololu case with its thin webs and internal walls with no problem.

Thanks,

I noticed the same in the modified wade's I got. Solution:

1 generous helping of curse-words
1 part xacto fiddling
1 part acetone/abs slurry to fill the gap once captive nuts were in place.

Down the road I'll be importing the stl into my cad program and modifying it further, or punting and downloading a newer design.

Glad I'm not the only one, then! In my innocence, I thought that this was such a standard and well-used design, and OpenSCAD and Slic3r such well-tried and highly recommended applications, that someone else must have noticed and fixed it. But maybe not...

Unfortunately, I've never tried STL editing but given that the STL seemed OK and Netfabb claimed to have repaired any errors it found anyway, I'm not sure how that would help. In my case, at least, the problem seems to be in Slic3r, as far as I can tell.

The saving grace for me is that I am using my own design "hobbed" bolt, which is actually a machine-cut set of teeth which are very sharp compared with the somewhat rounded teeth that hobbing with a tap gives. This means that you need surprisingly little idler spring tension to reliably grab the filament, and that in turn makes it easier to just pull the idler (actually, a guidler in my case) back to release the filament for changing. But it's still frustrating not to be able to get reliably from SCAD to Gcode.

Out of interest, have you also been able to print herringbone gears? That seems to be a bit of a black art, and it's a bit beyond a sharp knife and a pot of abs slurry to fix after printing

I'm ashamed to say that I'm still building my first printer, so I can't comment experientally on it.

At a guess?

Careful control of bed and extruder temperature,
Careful adjustment of slicing parameters,
And slow print speed.
Possible need for external cooling jet of some sort.

I don't know yet, but that IS one of the first things I'll be printing right away, along with a new extruder shell, once I've done my initial calibrations, and I'll be employing all four of those suggestion myself.

Run the Stl through cloud.netfabb.com . For some reason that design often gives bad Stl files.

Hmm...

1. Repair the STL in NetFabb Cloud (Assuming you did this already though)

2. Fine tune your Slic3r extrusion multiplier (https://github.com/alexrj/Slic3r/wiki/Calibration)

3. Post this as an issue on the Slic3r GitHub page-> Give lots of specifics! https://github.com/alexrj/Slic3r/issues?state=closed

Read here how to write a good slic3r bug report-> https://github.com/alexrj/Slic3r/wiki/Quick-guide-to-writing-good-bug-reports