S5 X-Axis Offset Despite X/Y Calibration

[CTotten 18]

Late last week, I had a print where the X position between printcores 1 and 2 was significantly off.  This showed up as the prime tower looking like two interlocking rings instead of concentric rings.  I re-did the X/Y calibration, and printed another job requiring support material (Breakaway), and the same thing happened.  The only "change" before this happened is that I finally tried out the CC 0.6 printcore on a job using.  I used it for one job, then went back to the AA 0.4 printcore that I had been using.  Both of these instances occurred after that.

 

What else could be causing this?  In both cases, the "offset" distance was approximately the same.

 

Edited July 21, 2020 by CTotten

[UlrichC-DE 109]

This looks crazy!

 

In theory, XY is different for each print head combination. The values in cura the total distance. The offset on the printer makes the small compensation distances.  It's definitely not 5mm.

 

In Cura under Machine-Settings, Printhead-Settings -> Extruder 2 you find the big "Nozzle offset X".
For me (S5) this is 22mm (distance between printhead 1 and 2). Y is 0.

 

Different idea:
do you have the new firmware?

Edited July 21, 2020 by UlrichC-DE

[robinmdh 100]

XY calibration is indeed stored for a position locked pair of print-cores.

But this almost looks like the 22 mm difference Cura inserts between the positions isn't there...

The max the calibration can handle is 2.9268 mm around a center position so + or - 1.4634 mm.

 

To even begin to debug this I'll need more info, like the firmware version, the gcode file and the files duped to the USB stick when you select the dump logs to USB at least.

[CTotten 18]

I was running 5.7.2 when this happened.  I updated to 5.7.3 and it did not happen on the one print I have done since.

 

This happened on two different prints, on different parts, different slices, etc.  This was also the same printcore combination that I have been using for months, with the one exception of me using the CC 0.6 core one time, immediately prior this happening.  I did not touch the X/Y calibration when I used the CC core because it was a single core print.

[robinmdh 100]

Good to hear it hasn't happened again!
It's odd though, no changes on any of that code in the 5.7.2 -> 5.7.3 change.
I wouldn't want that to happen to anyone, If you can send me the logs and gcode or UFP file I'd be glad to take a look and make this doesn't happen to anyone else.