Error when printing cylindrical objects

Hi, I’ve had zero problems with my P1S until now hence my non appearance on this forum. As a result I published before I finished writing. My apologies.
Any help will be gratefully appreciated. Cylindrical or round objects have become noticeably indented. While there is often a slight pattern in rounded objects this is worse, creating a series of concavities between the print points. This might be a fault in the printer rather than a resolution issue in the slicing software. I’ve searched online and found no reference to this.



Fish3B v16.3mf (49.0 KB)

It is funny how you copied the instructions how to post a good problem description and didn’t follow a single step :stuck_out_tongue_closed_eyes: sorry, you posted a foto which us a very good first step.
Anyway, it would be very useful if you wrote some words about what is wrong in the print. It looks to me like a nicely printed model

Sorry about that. Hope my latest edit makes this more clear.

Can we have the .3mf file?

Struggling to get my head around this forum software. Hope this does the trick.
Fish3B v16.3mf (49.0 KB)

I had a few prints marred by this indentation so I created a sample cylinder and tried to find a settings solution in Bambu Lab. No improvement. Thought it was a resolution problem but then noticed the printer was weaving inward between the dot points. Appreciate any wisdom you can bring to this puzzle.

Been a fantastic printer until now.

It’s printing the cylinder as it should. The cylinder model has very low resolution so it’s not a cylinder but a series of vertical planes.
Yo can see in the slicer capture here how the preview shows it will not be printing a circle but a series of segments:

1 Like

This is a real puzzle.

I printed the first 8mm of the file without seeing that concavity in the segments, and verified that the segments were flat between vertices with a straightedge.

I do not think this is a slicer problem, and I cannot think of a hardware fault that would have this result.

I considered the possibility that the infill was shrinking and pulling in the walls, but I do not see any correlation. The low spots are evenly spaced but the infill attachment points do not match the low spots.

I would still change the infill pattern - the nozzle will collide at every intersection on each layer using the grid pattern. Gyroid or crosshatch might be preferable.

Incorrect pressure advance can cause bulging at corners, but I would not expect such a large effect at such small angles.

1 Like

There is something about the 3MF file(s) that you uploaded that does not agree with your photo. Your photo Shows a rectilinear infill pattern whereas your 3MF files show crosshatch. So that we can be evaluating the same issues, can you slice your model and show us an image that looks like this:

What we are looking for an a sliced image that your slicer is sending to the printer to verify and match up what your photo is showing.

However, two suggestions come to mind. If you do confirm that your infill is set to rectilinear, try changing it to crosshatch or some other 3D infill such as Cubic, Gyroid, 3d Honeycomb, tri-hexagon etc. This will provide a more even and less regular wall pattern which it appears your photograph may be suffering from.

The other thing you can try is to increase the wall loop to your current setting which appears to be the default of 2 and increase it to 3 or 4. Here’s the impact that will have

What this will do is create a smoother transition between supports.

The last thing you may try is to change the default wall order from inner/outer to outer/inner. This will allow for the walls to be set up first and allowing them to harden before the infill is laid down.

1 Like

A good test is to print your object and a similar “reference” object in the same print. I’ve added a similar part next to your part using the “Add primitive”, cored out the center with an “Add negative part” and then boolean meshed them together. Here is the file. Your part is on the left, and the one I made is on the right.

Fish3B v16 Modified.3mf (105.1 KB)

It is very interesting to note that your part has 280 Triangles, while the near identical part that I generated in Bambu Studio has 5760 Triangles.

image

image

If you print these parts at the same time it should give you a really good idea if your problem is model related or printer related.

1 Like


You’re a wizard Jon. As the comparison pictures show, your altered file worked perfectly. It has no striations. The faint stripes on the revised cylinder are just reflections of my shirt and a nearby window. The print times were surprisingly identical despite the difference in quality, with both pieces taking 17 seconds for each layer. Needless to say, I’m impressed, not to mention grateful that this is a software fix rather than a printer error. My question now is how you did it.

1 Like

Thanks for your advice Olias. Much appreciated. I anticipated an infill problem and tried some others which also didn’t work. This explains why the file in the photo was different to the 3mf file.

I’ll take your advice on the infills you mentioned. Your suggestion about the wall order sounds good. I’ll try it and see. Is this similar to Jon’s advice? His altered 3mf file worked perfectly, but I’m not quite sure how he did it.

My guidance focused specifically on your model. Ideally, access to the raw CAD or mesh source allows for finer control. @JonRaymond demonstrated this concept by creating a native geometry and by using the slicer’s built-in primitives to create a solid cylinder, then hollowing it with a smaller cylinder as a negative, resulting in a hollowed cylinder but with a much finer mesh than your model could achieve.

His example highlights a key issue with your model: its low polygon count, which contributes to the “ribbing” effect. With fewer triangles, the slicer is instructed to render lines spaced further apart.

To understand the slicer’s behavior better for self-diagnosing issues, it’s helpful to know that STL files represent models as a mesh of triangles. We see a circle, but the slicer interprets polygons. More triangles yield finer surface geometry. You can check triangle count in the slicer’s prepare screen (bottom right) or by selecting “Show Wireframe” in the simplify dialogue, which reveals the model’s underlying structure.

Here’s JonRaymond’s 3MF file with your original cylinder on the left. It has only 280 triangles, which is low for a model of this size.

Now, compare this to JonRaymond’s version with 5760 triangles. The difference is stark; his model has finer polygons than the filament width, resulting in a much smoother surface.

1 Like

Thanks Olias. I understand the concept of polygon count, I just don’t know how you increase it with Bambu Lab’s software settings.

I’m a design artist, not a technician. While it works I’ve no desire to look under the hood. On that note I’ve been enjoying the PS1’s ease of use over the last year, it’s like a kitchen appliance in comparison to my previous printer, but I’m at a loss with things like this. Any suggestion gratefully received.

In my previous post I point to the low resolution of the original cylinder.
About increasing the definition, you can’t. You have to generate a new cylinder in a CAD software package or use the primitives Bambu Studio has.
How did you design the original model?

The cylinder was just a test piece. My actual designs are a bit more complex, and were created with Fusion 360.

Try exporting your designs as .step files instead of printing direct from Fusion or exporting STL files.

3 Likes

What was your path for getting the model from what you generated it in to Bambu Studio? Was your original model from Fusion?

I design the model in fusion, export it as an STL file, then import that file into Bambu Studio.

I don’t recall having any problem until recently, so maybe I’ve unwittingly changed an export setting.