ABS print lifting at the corners

[doobes]

I’ve got a fairly new P1S with AMS.

Using Bambuslicer, I went to print a tray I designed for my homelab.

Looks like this in SolidWorks:

I’m using ABS with the basic settings, but with no brim.

[small rack tray.3mf|attachment]

(upload://lfCdWNUty3nqYvGMM9W7fsFuzuR.3mf) (56.5 KB)

When I took it off the printer, the top surface of the cutout was very rough in comparison with all the other top surfaces:

When looking at the tray, it’s obviously deformed, as the edges have lifted and it isn’t flat. I think that is what contributed to the rough surface.

Looking for some thoughts on settings so that it prints flat.

Currently running the extruder at 260C and the bed at 90C.

Appreciate your ideas.

chris

[byMH62]

First you need to figure out if the model is lifting from the plate or is it the plate lifting at the corners with the model. If the plate is lifting at the corners on large models, I use bed clamps. Not sure which ones I have but there are many variations out there. Here’s one: https://makerworld.com/en/models/39464?from=search#profileId-38669

If the plate stays flat, do a search on this forum.

[Olias]

FYI. Your cut and past for the 3MF file got spit between lines so I am repasting it for those not familiar with the forum software.
small rack tray.3mf (56.5 KB)

Frist, you have brims turned off. They need to be on.

Thanks for uploading the actual model and also the quality screengrab from Solidworks. This really helps understand your design-intent and saves a lot of guesswork.

My first question. When you dried your filament. What was the before and after weight? This sill confirm that moisture was present and then removed. If you stated that the filament was straight out of the factory seal. Then that’s the first issue. Don’t trust any filament you have not personally verified has been dried which is the reason for the before and after weights.

This looks like classic under extrusion which could point to either miscalibrated or moist filament.

Based on what I see in the photo. There is something going on at this transition layer(layers 36-40 at 0.28 layer height). However, when I view your 3MF file, I see nothing that would help me understand why. That was until I looked at the filament profile.

About your 3MF file. Tell the truth now. Was this the actual file you printed from or did you make changes before you uploaded? I ask this question because the next thing to investigate was the filament settings. However, you have four filaments in this 3MF file and filament #4 is set to generic ABS. Using Generic filament settings is fine if you just want a test print but you should be tuning your filaments for large/expensive builds like this. If that was done, it was not captured in your 3MF file when you uploaded it. If you wouldn’t mind doing a screengrab of the filament settings that also would help.

However, I did investigate the layers and found something odd that I cannot explain. What I was looking for is fan speed transition.

The first place I looked was in the filament profile. Here is what I found in the settings. This may not mean anything but it would be a good diagnostic step to turn off all fan settings to zero just to confirm that there is a change in transition.

I say this because right at the first internal bridge transition point(layer 36) where the solid top layer for that one section where you get rough under extrusion occurs, something is happening with the fan speed. Is this the cause? Unknown. And that’s why I suggest turning fans completely off just to see if it modifies the behavior.

This could also be adjusted by using a different infill pattern such as rectilinear. Currently the 3MF file you uploaded was monotonic.

You may also experiment with the bridging setting but I sense that is only a bandaid if the filament issues are not addressed first. Note these settings may only appear in Orca Slicer 2.2 with developer or advanced settings on. Note that many of the really granular features are not present in Bambu Studio. You can use both slicers on your same machine.

[doobes]

Holy moly.

Um. This was a 3MF of a smaller segment of the model I was using to test the size of the opening against the CPU that’s going through it. It was the same setup though as for the model shown.

The spool of ABS that’s in there I got from MicroCenter about 3 years ago and opened recently, and it went directly in the AMS. Don’t have a means to dry filament, but I’ve got a bunch of desiccant in the AMS. Yes, I know it won’t dry the filament.

You are probably right. It might not have been dry from the outset. My experience level with this is obviously nowhere near yours.

The comment about transition is confirmed by this picture:

Based on your suggestions, I’ll do some experimentation to see what results.

My printing has been from from BambuSlicer because I’m doing all this on a Linux box (Elementary OS v7.1, an Ubuntu 22.04 variant). The viewer on OrcaSlice is a problem as GSStreamer complains about the wrong libraries. I’ve fiddled with it, but I’m not smart enough to get it working. I like the display, so back to BS.

When Pop!_OS COSMIC Beta is released, I will upgrade the distro on the box. I’ve tested OrcaSlicer on a laptop that has an Ubuntu 24.04 variant on it, and it’s fine.

Thank you, I greatly appreciate you taking the time to respond in such a detailed manner.

chris

[doobes]

Interesting.

Will have to do some investigation to see if this is an issue. Based upon the difficulty getting the print off the bed when it’s hot, the plate lifting could certainly be a culprit.

Based on your remarks, and Olias’ below, I’ve got some things to work with.

I’ll report back with status as it makes sense.

Greatly appreciate the feedback.

chris

[Olias]

@doobes

Don’t have a means to dry filament,

If you have a print bed and an empty spool box, you have a dryer. Check out this video. Truth be told, I was the original skeptic as to whether or not a dedicated filament dryer was needed and I didn’t want to spend money that would have later sat collecting dust.

So I viewed this video which allowed me to test out filament drying without spending any money. Check it out.

This is an example of wet PETG that really forced me to consider a dedicated dryer.

Here was my first effort where I tried to get fancy with a USB 40mm fan($3) to exhaust humidity in the box and a cigar box circular humidor($6) to measure the humidity. Did you know that at 50C, monochrome LCDs turn black?

The downside of course is that it will tie up your printer and of course, with ABS, you’ll have to elevate the temperature a lot but do not elevate it too much. Because you may become an owner of one of these abstract modern art pieces.

[doobes]

LOL

I’ve got one of these in the attic:

I’ll get it back out of the attic and give it a try.

Awesome.

If you are ever in the Williamsburg, VA area, ping me. I’ll buy you an adult beverage (assuming, of course, you are of legal age…)

cheers,

chris

[byMH62]

And once you manage to print it flat, wait for the model to completely cool down before removing it from the plate. I’ve seen large flat models warp after removal if pulled too soon.

[doobes]

So,

I needed to tweak the model a little bit as the guides on the corners were a wee bit tight on my aluminum posts.

Then I reprinted, doing nothing more than adding a brim. Here is what it looks like with the CPU that goes in it.

All the finishes are fine, and there is no weird line present that Olias was talking about. Dunno. I’ve had the slicer do weird things before. At work, we used to call this a “spurious anomaly”…

It’s much flatter, although not dead flat. It will be fine as it’s warped across the entirety, not just the corners. Maybe 2mm. As I’m stacking up a bunch of these, it will be fine.

The guides were still a bit tight at the bottom methinks because the bottom layers have swelled a bit. I opened up the slot another 0.1mm and it should be fine.

I can’t tell you how much I appreciate all the help here. I’ve really not had the need to reach out to the community, but I’m gobsmacked at the wonderful response.

Kudos!