Weird separation shift

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, and others.

jondilly1974

Speed running apprentice, motor Slayer 🤘
Lifetime Premium!
Premium Member
Excellence Award
Build Thread Contributor
Messages
26,134
Reaction score
38,839
Location
Toledo, OH
Arrma RC's
  1. Granite
  2. Infraction
  3. Kraton 6s
  4. Kraton EXB
  5. Senton 3s
  6. Typhon 6s
  7. Typhon 3s
I was printing some @suchtragedy @magnetik skirts and had a weird shift at a layer that I have never had happen before. It’s like something heated my prints and pushed most things around. I had ironing on so maybe that was the cause.

4A93C904-7A4E-4AF7-A680-DE43972B4C6A.jpeg
568AB920-EE75-4C2E-B6CA-87AEA89E026F.jpeg
 
Do 3D printers ever have multiple coordinate systems that they use, like G54 and G55, etc. in the code? Looks like an axis zero position was reset mid-print. But I don't know 3D printers, so I could be wildly wrong.
 
Looks like she got snagged up on the Y and skipped some steps.. z-hop on or off?
Do 3D printers ever have multiple coordinate systems that they use, like G54 and G55, etc. in the code? Looks like an axis zero position was reset mid-print. But I don't know 3D printers, so I could be wildly wrong.

The only thing a printer knows for offsets is based off home 0, 0, 0... if the belt skips and the motor skips steps it has no idea..?.. on the other hand more expensive units have trinamic drivers, if it senses a skipping load it will re-home and start again..
 
Last edited:
Well it’s off in opposite directions only in certain areas. The far edge that keeps the body tucked isn't off. And some of the areas that got pushed went opposite directions. I’m running the same code again to see if it happens again.
 
I would look at missed steps.

What's the hardware (board and drivers) and settings?

My TMC2209 drivers don't miss steps thanks to the ability to control the currents from the lcd.

My old drv drivers would miss steps if they weren't tuned to the proper currents and got hot enough.
 
Well it’s off in opposite directions only in certain areas. The far edge that keeps the body tucked isn't off. And some of the areas that got pushed went opposite directions. I’m running the same code again to see if it happens again.



???? ...Totally missed that, has to be in the slice for one to be opposite of the others... corrupt sd possibly?.. pc got sad for second while slicing? ???
 
I would look at missed steps.

What's the hardware (board and drivers) and settings?

My TMC2209 drivers don't miss steps thanks to the ability to control the currents from the lcd.

My old drv drivers would miss steps if they weren't tuned to the proper currents and got hot enough.
Creality Silent Board 2208 drivers
???? ...Totally missed that, has to be in the slice for one to be opposite of the others... corrupt sd possibly?.. pc got sad for second while slicing? ???
Well I was tweaking the flow rates during that print so maybe I made it mad. Second run of the same gcode is running and seems ok so far but it’s got many hours left to go.
 
Are you running your 2208 in uart mode? Do you have the ability to set your currents via console/lcd?

I feel driver current is irrelevant in this situation.. 3 parts are layer shifted in one direction and the last part is the opposite direction.. I too thought it was skipped steps, until 8s dilly pointed out the opposite layer shifts.. and to be honest out of all the printers I've had over the years, skipped steps= a stall/ snag.. I've never seen random skipped steps due to low or high current on a prebuilt machine.. not saying it couldn't happen.. it's just not likely at all..

There really should be no need to increase/ decrease driver current unless it's a home built setup and is low/high.. increased current, increases stepper/driver heat, not really a good thing...

I think something went crazy in the code wether it was while it was sliced, saved to the card, or so forth.
 
Does the preview mode in Cura look normal? Like it you go to the layers it shifts and play through the nozzle movements is it fine? If its fine there then I would think the machine just wigged out reading the g-code or maybe a stepper motor is acting funny. If it does it again the same way maybe try a different model sliced with the same settings as a test.
 
Are you running your 2208 in uart mode? Do you have the ability to set your currents via console/lcd?
Not sure. This is running TH3D Marlin. I’ve never made changes to current, so whatever they have in the base firmware is what it’s set at.
I feel driver current is irrelevant in this situation.. 3 parts are layer shifted in one direction and the last part is the opposite direction.. I too thought it was skipped steps, until 8s dilly pointed out the opposite layer shifts.. and to be honest out of all the printers I've had over the years, skipped steps= a stall/ snag.. I've never seen random skipped steps due to low or high current on a prebuilt machine.. not saying it couldn't happen.. it's just not likely at all..

There really should be no need to increase/ decrease driver current unless it's a home built setup and is low/high.. increased current, increases stepper/driver heat, not really a good thing...

I think something went crazy in the code wether it was while it was sliced, saved to the card, or so forth.
It’s about 5 hours from completing the second run. My guess is when I adjusted flow it was at that level that looks melted or under extruded. But you would think whatever did that to the misplaced areas would/should have affected the raised edges too.
Does the preview mode in Cura look normal? Like it you go to the layers it shifts and play through the nozzle movements is it fine? If its fine there then I would think the machine just wigged out reading the g-code or maybe a stepper motor is acting funny. If it does it again the same way maybe try a different model sliced with the same settings as a test.
Cura looked normal. This second print looks normal so far. I will reply back tonight when it finishes.

Also I do have a security camera viewing the printer that I could replay footage from to see if we can tell what happened. I will spend a few minutes later and see if I can see that level of detail in the shot or not. The purpose of the camera was to see when prints finish, not to see layer by layer detail. It’s only a 1080p camera.
 
Is this print over sd card? I've had prints go weird on me midway and turns out my card got corrupted mid print.
Yes it’s a cheap SD as well. I didn’t change anything this go around so we will see I guess. It’s only an 8gb card so not a big loss if I have to pitch it.
 
Second print turned out better but still not perfect. The weird drift didn’t happen, but one piece had some under extrusion on the top layer. Might turn off ironing next time. Good enough for a speed runner.

B6DFE80E-0070-488F-95B2-6CB97F09BF68.jpeg
1270CE05-1645-4EF1-9CE4-594610C0D6E0.jpeg
3B0ABE3E-8208-4434-AE52-5A0801BD55B4.jpeg
43B253A1-A7DC-4CF9-AF2A-6A342B1FD0D4.jpeg
 
Wth..that is so strange.. never seen anything like that before.. to go from opposite side layer shifts to a print that completed..?.. there is some ghosts somewhere...
 
Is that PLA and how long did that take to print?
It was PETG and I’m not sure how long. Probably 16-24 hours depending on a bunch of factors.
 
For me it looks like it's a feeding problem, not enough material from time to time.
Are you direct drive or bowden ?
Maybe check your feeder gear ?
 
Old Thread: Hello . There have been no replies in this thread for 90 days.
Content in this thread may no longer be relevant.
Perhaps it would be better to start a new thread instead.
Back
Top