Printed Gear ALWAYS has a "rib" created relative to the hub recessed area. All prusa slicer versions and ALL MK3 and Mk4 printers

[kevinmichaeldudley]

Description of the bug

All circular DISC objects with a recess in the dish/disc cause a "rib" to be created AT the same level as the transition from the "recessed area flat" to the fillet. This has occurred for YEARS and I have reported before but no response. I print gears but ANY disc shape with a recessed hub can cause it. But, LARGER DIAMETER AND DEEPER RECESS AND SHARPER FILLET cause it worse.

Project file & How to reproduce

Prusa slicer bug on gears.zip

Checklist of files included above

• [X] Project file
• [X] Screenshot

Version of PrusaSlicer

Happens all versions from 2.3 and up, latest is 2.8.0

Operating system

Windows 11

Printer model

Mk3 Mk3+ and Mk4. 8 different printers.

[u89djt]

EDITED to remove reference to custom gcode bcs it's just a different firmware indicator. You've been stuck for years, so presumably you'll welcome a fresh take on it from a fellow user. First question: does it happen with all default parameters? I see you've got parameter changes. The defaults are chosen to work well together. Maybe you've asked for too much plastic to be squished together in circumstances that lead to the ridge. The gcode you've posted confirms that the extrusion paths per se don't indicate a ridge. You've concluded that there's a bug with the printers. I've never seen anything like this on a MK3S and an XL, but I haven't tried your parameters. If I switch to MK3S (not plus) and keep all the parameter changes, I get this: There's no sign of a ridge. The top hypothesis might be that it doesn't happen with a wedge cut out of the gear. Cutting out the wedge seems unlikely to be important but that might need chasing. You've said the problem worsens at larger diameters, so a straight edge - an infinite radius - would show it. You could make a short rack with the same context and see if it happens to that. I haven't thought of any reason why a closed disc could cause a glitch yet. If a closed loop extrusion using an arc substitution was involved, that could be worth looking at, but these are gear teeth. It would have to be an arc in the wrong place causing plastic squish laterally to be sustained all the way out to the surface. And arc substitutions are recent - you're looking at years. Say and ask whatever comes to mind.

[neophyl]

Isn't this the benchy hull line problem ? Many many posts all over about it, some here, more on the prusa forums but its not limited to just prusa's.

[u89djt]

@neophyl you'll see in the gcode that there's no comparable infill transition in this model, and in the photo that there is a conspicuously protruding line that seems to be comparable with an extrusion width, rather than a minor, mostly cosmetic ripple. Note the way the illumination catches it.

[neophyl]

I does look more extreme than any usual hull line problem. Only time I have seen something as bad is when the stepper on Z has failed to move it up a full layer on a printer. Given this is on multiple printers though it would tend to rule that out. Have you tried setting external perimeters first ? With this geometry there should be no problem doing so as a test to see if it makes any difference.

[u89djt]

@kevinmichaeldudley @neophyl I'm printing the whole disc on the MK3S with the altered parameters preserved for the avoidance of any doubt. To reduce print time/material consumption, I cropped it up to 9mm above the base so there are a few layers of infill before the solid surfaces, and it'll stop just above the inflection point between the inside and outside fillets behind the teeth. I'll collect that in the morning.

[u89djt]

@kevinmichaeldudley no sign of disturbance around the start of the fillet for me. I have an aggressive first layer offset so there's an elephant's foot. Your rib would be expected around 3.802mm up where the inner surface and fillet start (as opposed to the 12.802 in your original). If you can execute a print on a MK3S that has the ridge and send me that exact 3mf file, I'll print the whole thing. Do me a favour and crop it down to the minimum height that causes it so I'm not throwing unnecessary plastic at it. Here's the 3mf of the stunted ring: shorter.zip And here's the wedge I printed previously full height wedge K1050 SB-13 80T Standard Offset V6 MK4.zip

[u89djt]

@kevinmichaeldudley see if you can identify a relatively small object that exhibits the problem on a MK3S and we can compare results. Do you have any objects of this general form that don't show a ridge? I've deleted previous comments as they didn't really add anything.

[kevinmichaeldudley]

Printing wedges: will always print perfectly, it only occurs on full “circle” prints.

MK3: I only have “new” MK4’s now, all my MK3’s burned up with the house in June ;>(

Smaller diameters: The issue NEVER shows on diameters much below 75 mm

Your slice: I’ll print the same 3.8 mm up to see if it does not show for me. (not surprised as “thickness” does seem to have a part)

Your comment: Assuming that's yours, what did you do differently in item 125739486192 ? That doesn't seem to have the ridge.

Yes that is mine. Just MUCH larger fillet (blend towards the inner hub). ½ wide gear vs .75 in (seems to play a part in severity).

The gear model I sent you was 4x worse than when printed on the MK3. On the MK3 it is just enough to show a mark.

Here is another gear, thinner gear 116T which has the more typical that I see. Just enough to be a “blemish” but not enough to actually impact part service. I just hold my breath and no customer has complained.

I’ll work up a thinner sample such as this gear to send a model.

I have studied the gcode as it is a copy layer after layer (on the perimeters at least) and I see no variance.

FYI: it is always slightly, slightly shifted, higher on one side than the other but ALWAYS returns back to normal position and finish.

I appreciate the attention and help. I will put more together and send to you. Could be a few days.

Thanks so much! This drives me crazy. I ran TWO of the large gears ($10.00 gone) before I noticed the “witness mark” was MUCH more severe on the MK4.

Also, I did flatten the fillet AND ran as “structural” on the MK4 and got a near perfect part.

I will be in touch soon.

From: Dave T. @.> Sent: Saturday, August 3, 2024 4:15 AM To: prusa3d/PrusaSlicer @.> Cc: kevinmichaeldudley @.>; Mention @.> Subject: Re: [prusa3d/PrusaSlicer] Printed Gear ALWAYS has a "rib" created relative to the hub recessed area. All prusa slicer versions and ALL MK3 and Mk4 printers (Issue #13191)

@kevinmichaeldudley https://github.com/kevinmichaeldudley no sign of disturbance around the start of the fillet for me. I have an aggressive first layer offset so there's an elephant's foot. Your rib would be expected around 3.802mm up where the inner surface and fillet start (as opposed to the 12.802 in your original). If you can execute a print on a MK3S that has the ridge and send me that exact 3mf file, I'll print the whole thing. Do me a favour and crop it down to the minimum height that causes it so I'm not throwing unnecessary plastic at it. 20240803_100011.jpg (view on web) https://github.com/user-attachments/assets/84129cbd-55e9-423c-9a6a-13a1dce308c3 Here's the 3mf of the stunted ring: shorter.zip https://github.com/user-attachments/files/16478849/shorter.zip And here's the wedge I printed previously full height wedge K1050 SB-13 80T Standard Offset V6 MK4.zip https://github.com/user-attachments/files/16478850/full.height.wedge.K1050.SB-13.80T.Standard.Offset.V6.MK4.zip

— Reply to this email directly, view it on GitHub https://github.com/prusa3d/PrusaSlicer/issues/13191#issuecomment-2266649784 , or unsubscribe https://github.com/notifications/unsubscribe-auth/AQ4HZHMYGQWZWIQSCGXSJCLZPSNPZAVCNFSM6AAAAABL3L3LFGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENRWGY2DSNZYGQ . You are receiving this because you were mentioned. https://github.com/notifications/beacon/AQ4HZHLRUCO2U63D5BMIT7LZPSNPZA5CNFSM6AAAAABL3L3LFGWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTUHDJKLQ.gif Message ID: @. @.> >

[u89djt]

No worries - this is a great way to learn stuff before it gets in the way of my own projects :) It would be good to see what's happening physically during deposition of the ridge layers. Maybe stop the print when you've got a ridge and it might be possible to see what's gone awry. It would be handy for someone with a MK4 to chime in, really. I haven't had any plausible ideas.