XY dimensional inaccuracy

Hi,

I am printing a 20 mm calibration cube and I measure 20,15 mm on both dimensions. I use a GT2 timing belt, 20 teeth pulleys, and the prusa calculator tells me to use 80 for XY steps. My extruder is calibrated. What can I change to correct this dimensional instability?

Thanks.

I have found about -0.5% error when I print ABS. I correct it by bumping up the steps/mm in x and y by 0.5%. I have found about -0.25% error when I print TPU.

I think it's due to plastic shrinkage because the error is consistent in x and y axes even though the x axis is belt driven and the y axis is driven by a ball screw in my printer.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I've thought about changing the steps/mm rate but I mostly print in PLA. Does it shrink as much as ABS? Also I think there is another variable, which is the distance of the nozzle tip to the bed. While the layer height is 0,16 mm, I think it would be possible to print even if the nozzle is closer or a little bit far, which would result in flatter (and wider) or narrower prints. Would that distance also cause a dimensional instability or is it too little to make a measurable difference?

The only way to know is to test it. You specify the line width or allow slic3r to choose it, so I don't think layer thickness or line width will affect the shrinkage because the volume of plastic extruded is controlled.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

When adjusting the X and Y scaling, you generally want to print a larger test object in order to dial in your printer the best.

That said, I just read somewhere else about the idea to print a second cube at 2-3 times the size of your first print. If the margin of error (0.15mm in your case) increases by the same amount, then it is an issue with your scaling. If it remains roughly the same though, it is likely over-extrusion by your hotend. Print a larger object and continue to calibrate your extruder's steps per mm.

Just a thought.

Cant remember where it is now, so Google for XY calibration, but there is a formula out there to help alter the steps numbers to correct for minor inaccuracy.

target size / actual size * old_steps/mm = new_steps/mm

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

There is also this guide, which has links to some STL files
[reprap.org]

*Beware* In above guide in the Measure section, at item 6 the formula shown is I believe incorrect, it's the wrong way round

It should look like this: new_steps_per_mm = (expected move length/actual move length) * old_steps_per_mm
do the bit in the brackets first, then times the result by the old_steps value

---

Supporting 3D Printers with Parts and Build services.
Printer: Ormerod 2 (528.4) Duel extruder set-up with Aluminium X-Rib, RRPro Firmware v1.11-ch (2016-04-08)

Um, that's the same formula I posted.

There's nothing wrong with the formula in item 6 of the calibration guide. Standard rules of algebra apply- when you do the calculation, you evaluate the expression in parentheses before the multiplication.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Yes, I was aware of that being the same formula.

Didn't understand the result I was getting from the formula in item 6 due to it not having the brackets shown and the old_steps value at the front.
If I was able to edit that page I would edit it to show the brackets and use the rule "workout the sum in the brackets first", then times the result using the items outside them which gives a steps value that to me makes sense and works.

algebra and me don't work well together.

Edited 2 time(s). Last edit at 04/08/2016 11:55AM by orictosh.

---

Supporting 3D Printers with Parts and Build services.
Printer: Ormerod 2 (528.4) Duel extruder set-up with Aluminium X-Rib, RRPro Firmware v1.11-ch (2016-04-08)