Something is wrong with G3 d3 or fusion/pp

chernogorsky

Hi,
I found some curios bug if Im using G3 operation with f360 + universal pp 1.5 + enabled arc/circle

It creates oval (X axis is the most affected one)

The greater cutting feed the worse the "circle". That one on a picture if I remember correctly was made with 2400 mm/min

And here the result when I disabled arc movements (same feed rate)

c-inner_failed.gcode c-inner.gcode

M115
FIRMWARE_NAME: RepRapFirmware for Duet 3 MB6XD FIRMWARE_VERSION: 3.4.1 ELECTRONICS: Duet 3 MB6XD v0.1 FIRMWARE_DATE: 2022-06-01 21:10:05

3DPMicro

Can you post the program with the arc moves. I can run it through my cad/cam which will back plot the gcode to geometry

chernogorsky

@3dpmicro Hi, not shure I follow
I attached c-inner_failed (which is with arc, g2/g3) and c-inner (which is g0/g1 only)

alankilian

This post is deleted!

alankilian

@chernogorsky Boy, that's strange.

I concatenated your G1 path and your G1/G3 path and rendered them using THIS online rendering program and they plot preciously the same path in the X-Y plane.

Using a previous version of RRF on a Duet2 I was able to make prefect objects using millions of G2 commands so it has worked in the past. I was using small G2 arcvs however, so maybe I just didn't see a small variation in X-Y dimensions.

I can try updating my system to whatever version of RRF you are using and test your file and see what shape I get.

chernogorsky

I have not seen any issue with a printer (and ASFAIK it utilizes g2/3 at some points)
but as you can see from the pictures with CNC it looks different

It may be still post processor issue, but g2/g3 code looks ok-ish