Weird scaling artifact

I have Duet 2 Ethernet and the latest firmware. My z-probe is a custom piezo sensor/electronics that uses the nozzle tip.
I've gotten comfortable troubleshooting printing issues but this one is a stumper: my Benchy seems to have changed scale in mid print, see pics. I searched for the G-code scaling commands and even something that may have changed my calibration numbers but they weren't found in the file.
(sliced by PrusaSlicer)
Can't include the gcode. I keep getting 'too large' errors, even when I try to use a reply and just add a 1.5M zip of the gcode. You can request a copy from my email shadowphile@gmail.com if you want.
(update: I haven't tried to reprint this object because it's a longish print (.09mm layers) and it would be worse if it was ok the second time since I couldn't replicate the problem.)

thanks for reading and I hope you can help.

Edited 2 time(s). Last edit at 06/05/2021 12:01AM by shadowphile.

If the top is overhanging the bottom all round, then my guess is that it has skipped a small number of layers. Have you printed that GCode file before? If you haven't, then try slicing it again.

---

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 thought had occurred to me but the boat height is exactly correct.
Do you know of a good simulation that will graphically display my gcode? Obviously it was not visible in the PrusaSlicer display.
I could slice again but it's just a benchy so I don't need to 'fix' this gcode (and I'm troubleshooting problems right now anyway)
I was thinking it might be a slicer error but that would actually be a first for me.

addendum: I dropped the gcode into the online viewer [gcode.ws] and it does not show a lip. So either the hardware or the firmware seems to me.
(the coder in me suspects an unassigned variable in the firmware, and that the error that started with my print was a (incompatible) setting left from an earlier operation that was never updated when the newest job was run.)

Edited 1 time(s). Last edit at 06/06/2021 01:20AM by shadowphile.

Where on the build plate have you defined X0 Y0 to be? A scaling error would only cause an overhang all the way round if X0 Y0 is inside the bench.

---

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].

It's a delta so origin is in the middle. The model was approximate in the center. I was wondering myself where the origin of the scaling was based. It kind of disappears at the bow but hard to be sure, but it sure looks like it is even all the way around from the bottom view.

Any scaling applied is based on X0 Y0. The scaling is controlled by the M579 command, see [duet3d.dozuki.com].

---

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 is assuming that the scaling error was somehow implemented with the M579 command, which was the first thing I scanned for. If M579 got executed, it seems like a firmware hiccup. If I try to reprint this part, I will move it away from the origin to see if scaling stays the same. (if it does at all, which I'm afraid won't happen)