Page 1 of 1

Print quality debugging

Posted: Wed Jul 02, 2014 10:12 pm
by benavery
I recently started with my M2, and have been experiencing a few minor print quality issues - but I have no idea where to start tweaking to remove them. Reprap has a troubleshooting page with pictures, but it extremely sporadic about what it does, and doesn't cover. Anyone know of good, and slightly more comprehensive troubleshooting guides (with stuff like zits -> increase retraction, poor layer adhesion -> increase temp (or??))

Failing some other material, here is the things I have found - and wonder if people have suggesions:
- Poor layer adhesion with ABS. I printed the motor mount with white MG ABS, using the slic3r profile found on the wiki, at 230degrees. But cable mount part that extends broke off very easily, cleanly along the layers. Is the solution here to simply bump up extrusion temp ?

- Overextruded surfaces. The final cap layer on some pieces look overextruded - there are many raised blobs - kind of indicating too much plastic. I tried the extrusion test though (measure width of single-wall with extrusion multiplier at 1.0) and it measured at precisely 0.39 which is what the slic3r file also claimed it should be. Do I trust the measurement, or back the multiplier off a bit too make the surfaces prettier

- Vertical waves. In the below photo - you can see vertical waves through the piece. I noticed the same on the factory printed parts - so obviously I'm not alone in this. I looked at the gcode, and the piece looked to all have perfectly straight edges. Two weird observations: They didn't occur for about the first 1/4 of the piece (roughly the first 3 'facets' from the right edge are all razor straight). They don't seem to align to consistent absolute x/y positions (which would have suggest stepper rounding weirdness) - but instead look to be relative to the start of the edge. In some spots you can see the waves all align, zig-zagging parallel to the endge through multiple facets. At a guess I would say extrusion inconsistency ? Maybe distance from the heated-bed (or just raw time the hot-end was hot for to reach 1/4 way up) could be related ? Is this then a temp thing?

- This isn't really a print quality question - but how hot are the steppers expected to get ? I checked the X/Y ones about 1hr into that vase print, and they were very warm. Not so hot I couldn't touch them, but probably not pleasant to hold for too long, which I suppose puts them somewhere around 50C.

Image

Re: Print quality debugging

Posted: Wed Jul 02, 2014 10:44 pm
by jimc
the waves are called ringing. its a resonance in the bed or or extruder right after a direction change. its normal and what you have there isnt excessive at all. it always looks much worse with black filament as well. for layer adhesion, yes bump up the temp. abs is usually around 245. if your overextruding then measure your filament and double check that the diameter is set correctly. if it is then lower your extrusion multiplier. both the dia and multiplier control flow rate.

Re: Print quality debugging

Posted: Thu Jul 03, 2014 2:54 am
by jsc
To control ringing, try turning down the acceleration. You can set context-dependent accelerations in Slic3r. The default is something like 8000 or 9000, try turning it down to 1000 for exterior perimeters.

Re: Print quality debugging

Posted: Wed Jul 30, 2014 2:47 pm
by JCWilliams
How would I do the deceleration in S3D? :?:

Re: Print quality debugging

Posted: Wed Jul 30, 2014 3:01 pm
by jsc
In the profile settings, Scripts tab, add M201 X1000 Y1000 to the starting g-code. You can experiment with the effect of changing the values. Someone ran off a large experiment on acceleration settings in a monster thread in the Google forum and determined that they got diminishing returns below 800. Also, you may wish to try different numbers in the X and Y axes as they will have different resonant behaviors.

Re: Print quality debugging

Posted: Wed Jul 30, 2014 9:16 pm
by JCWilliams
Thanks jsc I will give that a try when i come into work tomorrow.