Two Issue: Wide Horizontal Waves and Layer Skip

Greetings all!

So I recently started printing a large project for someone and for the most part, everything has been going fine up to this point. Now, I'm running into two issues that seem rather strange.
Firstly, I had a serious layer shift at the exact same layer on the same print twice. I stepped down stepper current to 1A vs. the previously configured 1.2A (I had it at 1A by default and upped it for increased accuracy) and planned on slowing the print down since it may have been the area it was printing was causing a slip of sorts (although all belts are tight). I printed something that tall/that long of a print time with no issue a day or two prior, so this happening has me a little curious as to what is going on. My guess, though, is either something is corrupt in the g-code or maybe over heated stepper chips caused a thermal shutdown and then resumption of the print. This seems like the culprit since the shift, while at the exact same layer, was not the exact same amount of shifting (see photo one).

The second issue I just started running into is a weird WIDE horizontal wave that appears only on one side of the printer. It only affects prints or parts of prints that are on that side of the printer, however this also didn't start until the last couple days as other prints that were going across that same field of the print area have finished and haven't exhibited the same issue. While I can bypass the issue temporarily by avoiding that side of the printer, this doesn't solve the root cause. I initially was thinking trace for the bed heat, but that would affect more than one specific area and would also require some visual damage to the trace. The bed is heated using PID so it's not a Bang-Bang issue.The only potential thing I can think of outside that is maybe effector plate or maybe even firmware on the printer (I recently updated it). See photo two of the piece closes to camera (with hole) and you'll see the wide waves.

Any input into this would be super appreciated! I've listed the major printer specs below for reference as well. Looking forward to any/all suggestions!

Rostock Max V2
DuetWifi running 2.02 firmware (Dec-24-18 release)
PanelDue 7" touchscreen Interface
FSR Bed Leveling
Carbon Rods with Aluminum Ball Brackets
E3D V6 driven through HE280 accelerometer board (board is leftover from previous Rostock upgrade, fairly standard practice for Rostock tinkerers I believe?)
Currently using ZYLTech Ceramic White 1.75mm PLA filament through a .4mm nozzle at .32 layer height

As you are running a Duet, in the event that the stepper drivers overheat you will be informed via whatever of Duet Web Control, Panel Due or USB that are connected. So if you are not seeing any error messages, the layer shifts are not caused by overheating stepper drivers. The fact that the layer shift happens at the same point each time confirms this.

Have you checked that the layer shift isn't in the GCode itself, by running the GCode through an online GCode simulator?

It looks like your printer is a delta. If you get skipped steps, then as well as a horizontal layer shift, subsequent layers will get slanted. Do you see any evidence of this?

If you care to post your GCode file then I can try running it on my Duet-powered delta.

---

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].

DC42,

Thanks for getting back to me! I didn't see any error messages on Duet Web Control or the Panel Due, so that means it's safe to say it isn't the steppers overheating. Out of curiosity, is there any kind of timeout function on errors that would show up on the PanelDue or the Duet Web Control interface?

I just ran the same GCODE through two different online GCODE simulators and I didn't see anything out of the ordinary in the simulation. Unfortunately, I am unable to post the GCODE due to attachment restrictions (just barely though: 8.3MB!).

There was no slanting in the subsequent layers after the skip.

Additionally, do you believe the waviness I experienced from the 2nd photo could be related to what I'm experiencing as well? I just got home and had a chance to pull the prints off the bed and examined the one with the most waviness on it. It looks like both sides see this wave effect (when one side is moving out, the side opposite is moving in).

Thanks again!

Anyone else by chance have any potential insight? I'm replacing the idlers with a fresh set from SeeMeCNC and just replaced the belts. Running another print to see how it does. Maybe the skip will be gone, maybe not... I'll know soon enough though!

So I will keep this updated as things go along. The print continued overnight and when I awoke, I was greeted with, sadly, another random layer shift at the same layer on the print. I slowed the print down this time around just to see if maybe it was something too fast but to no avail. I'm beginning to believe the .GCODE is corrupted, regardless of the .GCODE simulation showing everything as okay. I come to this conclusion as the layer shift is always in the same direction, at the same layer, regardless of model rotation on the bed. I have had other prints that have printed at the same height or higher than where this shift is occurring AFTER the first shift had occurred on this print and they did not exhibit this shift...

Also, while the line waviness was much less distinct this time, it still appears to be present, which makes me believe it could potentially be the stepper toothed gear not being seated fully/correctly. I will investigate this today and return back with more information as it becomes available.

Likely my final post on this, but I did end up successfully completing my print. I'm not sure how easy it would have been without Simplify3D's multiple processes, but I ended up utilizing that feature to drop print speed to 30% of normal speed for the range of layers that the skip was happening in. Not sure why it didn't skew the print when it skipped originally, but that is something I'll look into at a later date.