I am working with a Bambu P2S (Studio 2.4) and trying to achieve a specific level of first-layer quality on the official Smooth PEI plate. While my prints are perfectly smooth to the touch, they are lacking visual uniformity.
The goal is to produce a first layer that looks like a single, solid sheet of molded plastic where the individual nozzle paths are invisible. Currently, I can see a distinct directional “grain” from the nozzle’s pathing, and on wide flat areas like coaster rims, I’m seeing slight ridges or “plowing” where the plastic isn’t laying perfectly flush.
Current Settings in Studio:
Wall Generator: Arachne
Only one wall on first layer: Enabled
Bottom Surface Pattern: Monotonic
I am looking for advice on how to get the plastic to “wet” and coalesce into a completely unified, mirror-like surface. Specifically:
Is there a recommended Initial Layer Flow Ratio adjustment to help the lines blend together without causing the nozzle to plow?
Is it possible to slightly adjust the Z-offset on the P2S to get more “squish” on that first layer to eliminate the visible paths?
Does the high-speed cooling on these machines need to be turned down for the first layer to allow the plastic more time to settle flat?
Any input from users who have successfully achieved a “molded” look on a smooth plate would be greatly appreciated.
I think you need to calibrate overall flow rates and pressure advance for those filaments. The gaps between black lines indicate a need for more flow, and PA for both colors needs to be adjusted. Holes where the nozzle reverses direction show that black PA is definitely too high. Yellow flow and PA can also be slightly improved. I would activate Develop mode in Preferences and use the top-line Calibrations for fine tuning presets. (I actually use OrcaSlicer, but Studio borrowed the same tests.)
I would expect a first layer with fairly well calibrated filament to be at least this smooth:
I think the best Initial Layer flow will vary so there will not be a single “recommended” adjustment that works for all filaments. Find your preferred Initial Layer Flow with a simple test. Create a primitive cylinder, scaled to a convenient size (40x40x1), then make a few clones. Assign each object a different initial flow. Compare the results.
(I forgot to check all my settings, so these test samples have too many wall loops)
I would find the central portion of any of these samples acceptable with a single outer wall. If I were still concerned about the tiny flaws in the thick walls, I would re-test with 1.15 and 1.20 Initial Layer Flows.
I have never found a need to adjust Z-offset on my X1C, nor change cooling defaults.
Thanks for taking the time to respond and your suggests base on your test.
After analyzing the results of several test prints, I’ve reached a logical conclusion: achieving a truly unified, “molded” visual finish on a smooth plate may be a fundamental limitation of FDM technology.
Even when a part is perfectly smooth to the touch, the visual “grain” of the nozzle’s path remains. Since FDM is essentially a process of laying down individual “ropes” of plastic side-by-side, those lines will always exist as physical and optical artifacts.
Logically, no matter how much you refine the flow or the symmetry of the pathing, the human eye is still going to pick up on the directional sheen where those beads of plastic meet. On high-speed machines like the P2S, the plastic is cooled almost instantly, which “freezes” the rounded geometry of each line before it has a chance to flow into the next one and create a single, uniform sheet.
In my opinion, someone with average eyesight will be able to see those patterns starting at 3’ or 4’ away, regardless of how well-calibrated the machine is. The textured PEI plate seems to be the only logical solution to this because it’s the only surface that physically breaks up the light and masks the nozzle’s signature entirely.
On another note, it’s to bad that Bambu doesn’t develop a “Fine” or “Satin” textured plate. While that wouldn’t solve the issue, it would get a lot closer to what I want to achieve. The current textured build plate is to aggressive for my project, but definately has its place.
It’s important to manage your expectations up front. What you are asking cannot be done with the Bambu plates available in the Bambu store. There is no slicer checkbox, no magic build plate, and no filament that will give you molded-plastic results out of the box. That said, there is a way to get close - but it comes with real effort, iteration, and tradeoffs.
If you are the kind of person who wants to push a button and get instant gratification, or you would characterize yourself as being a member of the TL;DR club, this long post is not for you and you should skip the rest of this post.
What follows is not a recipe, it is a process. It involves sourcing non-standard hardware, modifying surfaces, wasting filament, reprinting parts, and accepting that some results require manual finishing. That is the reality of FDM today, there are no shortcuts I am afraid.
Here is an “extreme” benchmark example of what is physically possible using a combination of methods. I realize this is well beyond what you are aiming for, and most people will never reach this level. I am including it deliberately so you understand where the upper bound actually is. If you can approach this, everything below it becomes easier by comparison.
This is one I did in 2023 during a very good exchange with fellow forum members. You can adapt elements of these methods to get closer to a “molded plastic” appearance, but doing so requires specialty build surfaces, surface modification, repeated tuning, and patience.
This thread is long, but it documents how optical transparency was achieved and the techniques the group used to get there. In all candor, this is from what I consider “the good old days” of this forum, when the majority of contributions were focused on how to achieve results rather than asking for shortcuts. It shows what sustained collaboration and experimentation can accomplish.
Steps to get there
1) Build surface
Forget about most of the Bambu Lab smooth plate or any of today’s Bambu stock plates, they are “ALL” overpriced and underperform, you can get much better build surfaces from Amazon or Ali Express than than the Bambu store for 1/3rd less. The sole exception is the engineering build plate which frequently is unavailable. https://us.store.bambulab.com/products/bambu-engineering-plate?id=657433241362993164
I maintain nine 255mm plates and have one copy of each Bambu smooth or engineering plate. One man’s smooth is another man’s rocky road. If what Bambu is presenting as smooth, then asphalt is polished marble. Let’s just be grateful Bambu is not in the optics business, because what they call smooth is, by any reasonable definition, a matte finish.
If you want to search intelligently for smooth alternatives, the terms you are looking for are “specular” or “mirror” finish. These are not common anymore. I was able to find exactly one plate two years ago that disappeared almost immediately. It does exist, and you might still find one on AliExpress, but do not expect availability or consistency. If you find one, please post it here.
If that plate is unavailable, your next-best option is a smooth engineering plate or in a pinch, you can use their Smooth PEI plate, assuming you can still find one. Once again, Bambu giveth and taketh away. The engineering plate was one of several popular plates available in the early days and is now gone. Other manufacturers still make equivalents, but you will have to search, test, and accept that not all of them are equal.
2) Modify the build surface using Kapton (polyimide) film
Once you have sourced the smoothest build plate you can reasonably obtain, surface modification becomes unavoidable. This is not optional if your goal is to minimize visible nozzle paths.
Here is another long post that goes deeper into this subject:
Below is an example using a combination of methods, including a strip of Kapton tape, with filament similar to yours. No, this is still not molded-like. But it is materially closer, and that difference did not come for free.
3) Camouflaging the bottom surface through texture and filament manipulation
If you are chasing a solid color, you need to accept an uncomfortable truth: you cannot completely eliminate filament lines on the first layer without post-processing of some kind. What you can do is camouflage them.
This can be done through silk filaments or first-layer pattern manipulation. The downside is that silk filaments are mechanically weaker, more brittle, and less forgiving to tune. This is a tradeoff, not a win.
Here is a post from the other day where I showed a member how to do this:
This example comes from an experiment using three-color silk filament and a Hilbert curve pattern. It does not meet your stated objective, but it demonstrates how perception can be manipulated even when geometry cannot.
This is the step many people hope to avoid - some might even call it heresy - but it is also the step that actually works and works extremely well with patience and practice
Post processing using wet/dry sandpaper is unavoidable if you want to remove nozzle artifacts rather than merely hide them. This is a manual, time-consuming process using progressive grits of emery cloth or wet/dry sandpaper. The optical cube example above was achieved this way. I started at 400 grit and worked up to 1,000 grit. There are no shortcuts here.
The final cheat:
This will sound like heresy to some, but Rust-Oleum Triple Glaze has no equal when it comes to producing a glossy finish. I’ve tried others but this one performed the best. For your application this is overkill, and you still must sand the surface first to remove nozzle lines. That said, my experience has been that I can stop at 400 grit wet sanding, apply Triple Glaze, and achieve a result that most people outside this forum would assume was injection molded.
Note: For optimal clarity, spray one coat and let it cure for 48-72 hours, then add a second coat and let cure for one week at room temp. Three coats won’t achieve any improvement and will add a “gooey” look.
I appreciate all of the replies and suggestions so far. I’m looking for advice on dialing in the visual quality of my first layer bottom surface when using the Bambu Smooth PEI plate.
I’ve been 3D printing since the original Robo R1 was released, so I’m very comfortable with ‘tinkering’ and fine-tuning settings. While I am satisfied with the general smoothness and touch of the bottom surface on my X1C and P1S, I am trying to eliminate the visible ‘filament path’ (the distinct lines where the nozzle lays down the plastic).
My goal is a completely uniform, visual finish that looks like a single solid surface rather than a series of lines. I’m aware that sanding and clear-coating are options, but I would like to achieve this result directly off the build plate through slicer calibration. Is that a possible goal? If not, I’m fine with that too.
Has anyone successfully ‘dialed in’ their settings to make these toolpaths invisible on the smooth plate?
Let’s just be grateful Bambu is not in the optics business, because what they call smooth is, by any reasonable definition, a matte finish.
