Printed Part Dimensions Not Accurate

I have calibrated my printer with a bunch of 20 x 20 mm test cubes. The issue come with parts I designed in Autodesk Inventor such as I specify a nut hole to be 7mm wide and comes out to like 6.32mm. File gets exported with the resolution set to High. It's accurate with external dimensions, but never with internals like holes. Any suggestions?

Just a guess, you're using Slic3r.
Step one: ditch Slic3r. Infamous for having undersized inner circles.
Step two: get familiar with better slicers like KISSlicer, Cura, Simplify3D [not for the light in budget], etc.

---

Realizer- One who realizes dreams by making them a reality either by possibility or by completion. Also creating or renewing hopes of dreams.
"keep in mind, even the best printer can not print with the best filament if the user is the problem." -Ohmarinus

Also you should never use a print to calibrate your X and Y steps per mm. Use a set of digital calipers and measure the actual movement of the axis. NEVER change your X and Y steps per mm based on a printed object dimensions.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Given that the filament width extruded is fairly large compared to the desired resolution, small nut holes are inevitably going to be difficult to get right. A simple solution is to print a test set of holes around the desired range to find one that fits!

I came up with this OpenScad file that creates a test object with 3, 4, 5, 6, and 8mm bolt holes and nut cutouts in both the horizontal and vertical directions. You can adjust the sizes until they come out right with your particular printer/slicer, etc. and then use those values elsewhere. The values that are in there are what worked for me in both Slic3r and Cura. I printed at .25mm layer height.

To save filament, you can print only until the bottom flat part finishes, then kill the print job. Once you have the horizontal values correct, then you can plug those in as a starting point for the vertical values as for me at least, they are somewhat close in most cases.

For full disclosure, the box and hexagon functions in the file came from another OpenScad file I had downloaded for something, but I modified them a little bit.

Bruce

Update:

My printer has been calibrated as accurately as I can get it with a caliper and my eyes. Steps/mm are set where they should be to travel a measured distance. Right now, it is off by about 0.5mm. Is there anything in the marlin firmware that also controls steps/mm besides the one being calculated for such as a scaling factor or something?

Edited 1 time(s). Last edit at 07/31/2014 09:43PM by chuttney1.