Xy accuracy.

Yesterday I managed to print a 295mm filled circle with absolutely perfect layer height, so I considered my calibration efforts complete and successful. Today I printed a 30mm cube which measured to about 30.5mm how might I go calibrating that out. It's within the range of over extrusion error, but my e steps are calibrated well, I just need to bring xy dimensions in a percent or two.

I suggest you try a larger print, because as you say 0.5mm could be the result of over extrusion.

If the XY dimensions are still too large, check the height of the print. If it is not accurate, adjust your steps/mm until it is. If it is accurate, then increase the configured diagonal rod length slightly, and recalibrate.

Alternatively, as I think you are running RepRapFirmware, you can use the M579 command make corrections to the XY scaling.

Another check you can do to make sure that you have the product of diagonal rod length and steps/mm correct is to print two parallel lines, from one edge of the bed to the other, one on each side of the centre and spaced one bed radius apart. Then check them for straightness.

---

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].

XY scaling can also be thrown out by effector tilt. Just another thing to check, if you're sure that your diagonal rod length is correct.

Ah nebbs, good call. My calibration has turned out great but I do rely on H parameters in the firmware to offset some effector tilt. The good news is I know what's causing it and have placed an order to fix it. Didn't consider how that'd impact scaling issues. I had assumed the h parameters during calibration would price that tilt into the equation and prevent it from impacting scale.

The effector still tilts and throws the nozzle out but it should be more obvious at the extremities than near the centre. Even with the nozzle based probe, tilt still affects the print, you're just compensating for it when you auto-calibrate,using an offset sensor and h values.

Makes me wonder David is there any way to make an effector tilt compensation algorithm? If you know the h values...

Edited 1 time(s). Last edit at 02/14/2017 04:31PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Calculating the correct kinematics when there is effector tilt would be very difficult. An easier task that I hope to do sometime is to be able to feed information such as tilt and calibration errors into an analysis program, and for the program to tell you exactly what sort of geometrical errors your machine has.

---

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].

That sounds interesting.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions