Having a real issue getting proper sized prints

Still working on a Prusa i2, looking to get prints as accurate as possible so I can print parts for it's replacement.

Details of the machine.
Prusa i2
Ramps w/Marlin
GT2 16t Pulleys
GT2 Belts


I've measured the carriage and bed movement, if I tell it 100mm it moves 100mm, and if I tell it to go back 100mm it goes back to the exact same place (measured).

The issue I'm having is that I can print a calibration cube. Say a 20mm x 20mm and it comes out to 19.8mm x 19.8mm
Which I'm thinking is probably that the carriages don't move exactly 20 mm, or measuring error, or material shrink
I can adjust for that

But the problem is, that if I print anything real
Say for example a part that has a measurement that is 17mm, it comes out oversized at 17.45mm

Not really sure what course of action to take

I wish I new what to tell you, althou you have been 3D Printing 3 years longer than myself.

I know you have tried all the X,Y and Z Stepper adjustment calibrations. which is strange as the test cube was printed smaller and then the sample part was printed larger.
This could be print speed, loose GT2 16t gears, loose X and Y-Axis belts, old stepper motors, Low Stepper current, loose nuts on the frame assemble, 3D Printed parts on the frame flexing, ect...
Iit could be difficult to trouble shoot this kind of 3D printer problem, so posted everything logical that could think of that could be a probable cause....

Edited 1 time(s). Last edit at 02/21/2018 09:26PM by Roberts_Clif.

I aim for a 20.4mm cube every time, that way when I print an object which need to sink a M3 nut in to a print, it go without to much fuss. Could it be the variation of .2 - .4 within the errors of the calliper's?
personally I not get to hung up on those reading if the prints look good more so if the axis are even.

There are several sorts of scaling error:

1. Constant scale errors, i.e. if you double the size of the test cube then the error doubles. On a Cartesian printer, these are caused by part shrinkage during cooling and/or the steps/mm setting being not quite right.

2. Constant amount scaling errors, i.e. if you double the size of the test sube then the absolute error remains the same. Caused by the actual extrusion width not being quite what the slicer intended, or if it is under-size then backlash in the drive system may also be a factor.

3. Scaling errors that vary with where on the bed you print. Generally not a problem with Cartesian and CoreXY printers. On Delta, Scara and Polar printers, can be caused by various geometrical or configuration errors.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Think it would be beneficial for me to mount a digital caliper an axis (lets say X for example) extend it to say 120mm, Move the carriage to just touching the caliper and then command a 100mm move, Allow the carriage to push closed the caliper and see how close to 100mm it gets?

Edit: looks like it's not a new idea. https://www.thingiverse.com/thing:2112176 and https://www.thingiverse.com/thing:2115894

Edited 2 time(s). Last edit at 02/22/2018 03:32PM by FA-MAS.

Went and measured the carriage movement, and did so fairly accurately. Seems the carriages weren't moving as much as they should. Steps per mm were correct as per the calculations from pulley size, teeth and belt type. Turns out it was the belt tension, the X wasn't tight enough and the Y was too tight. Got them both to measure accurately. Printing a test cube now.

So after making sure the carriages moved exactly the amount requested, I printed this thing. https://www.thingiverse.com/thing:12820
Measured, adjusted the axis, reprinted and repeated until it was measuring square.

Test cube came out perfectly at 20x20 plus or minus 0.04
So that is good

When I print an object, its not as far off as it was, but still slightly larger
I'm thinking at this point it's just the wobbly frame, allowing a little slop and probably the best I'm going to do with the machine.

Edited 1 time(s). Last edit at 03/02/2018 03:15PM by FA-MAS.