Parts don't fit together
Greetings,
We bought an i3 for rapid prototyping and I've spent the last week learning and printing. Kudos to Prusa Research for their efforts with chat support but this question seems to be out of scope for chat. Incidentally, since our application for the machine is for building parts, so my efforts have been primarily focused on dimensional accuracy over art so far. I'm using latest stables of Slic3r on windows, and driving the machine directly with Repetier with out the LCD screen. (Screen not functioning.)
Filament Settings
Filament Diameter: 1.77 (actual from measurement)
Extrusion Multiplier: 1
Extruder: 200/195
Bed: 55/52
Print Settings:
Layers and perimeters:
Layer height: 0.32
First layer: 0.32
Extra Perimeters: unchecked
Avoid Crossing perimeters: unchecked
Detect Thin Walls: checked
Detect bridging perimeters: checked
Seam position: Nearest
External perimeters first: unchecked
Advance Settings
Default Extrusion Width: 0.4
First Layer: 200%
(all else default)
With these settings, I can print the 0.4mm thin wall cube within about 25 microns of fluctuation, but a 20mm cube comes out about 250 microns too big in the x axis width . As you can imagine, when I print interlocking test pieces, they don't fit. I've tried various calibrations methods with out success, but I have yet to find one that accurately talks about how to deal with this and are just too vague in one way or the other.
Thanks in advance for reading!
EDIT: I did verify my steps/mm against my belts, gear teeth and threaded rod. They are correct from the factory.
EDIT2: The cube I was measuring was a cube I had printed a bit ago so I decided to print a fresh one in case anything had changed and it's off the 250 microns in X AND Y. Before it was only off in X. I'm not sure what adjustment was changed to produce the difference but square is better, right? At least the discrepancy is equal between the axis's.
Re: Parts don't fit together
First thing; print your first layer much slower than the rest.
In order to get accurate dimensions, you need to print the outside perimeter first, then the loops and then infill.
Best to run at least 1 or 2 loops inside the perimeter.
Unfortunately, I can't help further than this as I don't use Slic3r.
Peter
EDIT
Keep your layer heights to 0.2mm or less if you want accuracy; the "squish" will not be as precise with taller layers. And don't expect it to be "rapid" - the slower you print, the more precise the outcome...
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…
Re: Parts don't fit together
Peter, thank you for your suggestions!
I applied your suggestions and got a prettier box of the same size. Pretty is good too, so that improvement is certainly a help also. 😀
I also printed another .4mm thin wall cube and the resultant model is within a respectable tolerance, (75 microns) though all measurements were on the high side or right on. The interesting thing came about when I measured it's x/y dimensions, (20mm in it's design size, which I hadn't before as it wasn't part of it's intended test,) they were spot on!
So what I finally adjusted to affect change in the with of the 20mm box's X/Y sizes was in Print Settings -> Advanced -> Perimeters: 85% and External Perimeters: 85%. With so many factors in play that affect each other so adjustments I've made may be throwing each other off, I have to guess here that Slic3r's automatic values were causing the perimeters to be too wide. Time will tell but, focusing on the perimeter walls was the key to affecting any change in the overall width for my settings.
Thanks again!
Re: Parts don't fit together
Daniel
You have to remember that printing layers, the lines will never be an exact rectangle, with the outside edges being rounded.
This rounding will depend on a few factors (speed, temp etc) and the slicer makes a best guess at correcting for the round edges.
So, bottom line is that you will find it very difficult to arrive at exact dimensions. And when you have, you change the filament and you have to start again...
Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…