Spent 3 hours calibrating for my print to look like this

[user_3026326371]

What can see in your middle row is over-extrusion, in the left row I see gaps between the infill lines.
I can also see that the nozzle is smearing the filament over the wall loops.

The sandpaper look is the result of excessive filament build up.
Try an infill test like this:
Small cube, 3 bottom layers, 100% rectilinear infill at 0 degrees so it won’t diagonal in fill up the corners if too much.
Observe the infill building up with the layers.
With the extrusion factor too low you will see gaps between the lines.
With too much you will see a build up of filament in on the free side of the infill line.
This will be worst once reaching the opposite wall.
You want two values to note down somewhere:
1: The one where the gap between the lines seems to disappear.
2: The one where you first notice build up and roughness.
The ideal will be somewhere in between.
If you start with a vase mode print and tune the extrusion factor to result in a correct wall thickness you get the IDEAL case.
With more wall loops or infill this value often needs to be dialled down a bit because the filament is no longer able to expand around the nozzle IN ALL directions.
What can’ be pushed out one side has to be more on another side
Printing the same vase mode cube without infill and 5 wall loops results in the nozzles queezing the plastic properly.
Considering the different thicknesses for inner and outer wall their combined thickness in the printed model should match what you measure.

So you can get three different calibration results for the ‘perfect’ flow rate…
If you go for the average value, add them up and then divide by three, you should be pretty much spot on.
Comes down to the FDM balance between accuracy, strength, layer adhesion and surface finish.
From this baseline you adjust the other parameters, like k-factor to get the pressure advance right.
But also little things like how far infill lines go over perimeter lines or by how much you might have change for spare infill of low density in order to get these infill lines to properly stick together.
Some users prefer easy to work with line widths in order to design their models with better print results in mind.
Other users might increase these values quite a bit to get more strength.
Either way a change here will always result in changes in regards to how wide the nozzle has to push the filament to match this width.
If the flow ration is 1.02 while it SHOULD be 0.98 this would result in unavoidable build up somewhere.

The temperature is a vital factor as well…
Hot means soft, means easy to push out wide and fast.
Also means that where a line is added hot filament as it easier to creek into tiny gaps and uneven areas.
Cold means some or many of these gaps might not be filled and the plastic pushes out wider on the free side.
Quite often users are on the very low end for the print temp and optimised for a good layer bond of the wall loops.
Things like infill, gaps or inner walls however print much faster, sometimes we even combine infill layers…
Not all temperature tests account for all these speed changes …
For example, by default temp for PLA is 228 degrees Celsius and for going fast up to 235 while small or delicate models are printed slower and between 215 and 220 degrees.
All with matching and individually calibrated profiles…

[Olias]

I feel you pain. However, you may be expecting too much out of the technology. Bambu’s speeds are it’s own Achilles heal in some scenarios. Clearly you have prior experience but that may have colored your expectation. An Ender as an example doesn’t have the speed that Bambu does.

When I have unsatisfactory prints, the first thing I do after cutting the model down to test size is to put the printer in quiet mode to reduce the speed to 50% of all movements. Note that you can go into the speed menu and micromanage each movement type, but the quiet mode is a quick way to do the same without spending hours tweaking.

Here’s an example of what I mean. Each one of these test prints are a cube primitive scaled to 25.6x25.6x1mm. This is similar to the flow rate calibration tile as far as height goes but in a smaller footprint. The upper right is the only one that uses defaults settings. However, you can see that changing the layer heights and ironing affects have huge differences.

100% Speed

50% Speed Quiet Mode Same Model

Coincidentally, I too am using Overture filament for this exercise. But as you know, Overture offers 4 different black PLAs so it is real easy to confuse the products. This is the one I’m using. I found that the Bambu default profile for Overture does not work well so I tuned my own.

https://www.amazon.com/dp/B07PGY2JP1/

It should also be pointed out that there are other factors outside of tuning which can also greatly impact output. In this case, I have 5 different cube primitives but this time I set them from 1-5mm in height. Look at the difference even though all of them are using the identical tuned profile.

The one thing I might suggest if you aren’t already doing so. I picked up a tip from one of the YouTube channels. Apparently, after each calibration parameter, one should start off with an entirely new project. The reason for this is that if you forget to save your profile, it will not carry over to the next calibration. Once I figured that out, it explained to me why I had to recalibrate so many times when I thought I had already calibrated.

[95Prints3D]

I was hoping you were going to comment as ive seen you give really good information. It must be an overture thing, using the same filament you linked. Ive printed at 220mms and 50mms for the print to look identical on top. To me none of those look good. I can get sunlu to look decent but overture does not like my machine. I make sure to change to new project after any adjust is made. I just bought an x1, hopefully it can lead me to a path clearer path instead of guessing. I try everything that is suggested but always leads me to step one. I can spend a week calibrating overture black but i get the same issues with overture space grey and overture matte black. The matte gives me the opposite effect instead of the valleys and gaps being shiny they are matte while the rest of the finish is shiny. If its not underextruding its overextruding and vice versa. I cant even close pa calibrations. Ill run a test and itll give me a range of .16-.02 but ill run it again with no changes and im sitting at .28-.03. I have an ams, ive ran it though the spool holder just to see and no change. Chasing my own tail here.

[DWdesigns]

@95Prints3D

Its been 9 months of just calibrating a single filament. Wanted to see if there was anything i havent tried.

Wow, you have patience. Why waited so long before asking for help here?

You don’t want to do ironing, and it seems you can’t have that side facing down on the built plate. But you need a path to move forward, not going in circles here.

Maybe the next step is to see if there is someone in your area who has a Bambu Lab machine (P, X, or A series) and go see if theirs can produce the result that you want. This is just in case that what you got happened to be a bad one.

If you can’t do that, then maybe stick with the Ender that you have found satisfactory results? Or maybe reevaluate if 3D printing is the right technology for your purpose at this stage?

[schauer42]

Silly question, when you send the prints to the printer after slicing, are you unchecking the box for flow calibration? Otherwise all your timing will be for naught.

[Nebur]

P1 series - there is no flow autocalib

[Thermal_Runaway]

Glad to see someone else mentioned print speed already

It is possible to generate surfaces that are rough like sandpaper (like OP described) or reflective and glossy just by adjusting speed and no other variables. I made an example print a while back showing this. This is 1 layer thick PLA with no ironing and no other settings than default.

The speed that you need to use to achieve “glossy” finish is a function of temperature, so it may be difficult to calibrate this in a meaningful way using OEM speed profiles

I shared a script in another thread a few days ago that allows you to adjust speed to any value you want. The script handles all of the MQTT connection info and payload formatting, but feel free to modify any way you like https://forum.bambulab.com/t/custom-print-speeds-script/75875

[95Prints3D]

Theres just so much, facebook and reddit are the easiest. You cant make a post as a new member on these forums. So had to go else where. Unfortunately i dont know anyone in 3d printing and just like on fb everything that gets recommended ive tried. I had greg the overture rep help me, he used his printer on my models to try to help. Definitely had some differences. Got into 3d printing for action figure display, huge community with all the top modeler use a bambu lab so i know what to expect.

[95Prints3D]

P1s, doesnt have that option

[95Prints3D]

Yeah i thought speed is the biggest factor but people pull perfect prints off using the generic profile and the exact same filament. Already tried from 200mms in intervals to 50mms the same as my ender. The effect is the same even at those speeds, also just makes it look worse. Bambus speeds are nice because it turns everything matte.

[Thermal_Runaway]

There’s a temperature dependence on this effect. I am able to get that sandpaper-like texture with PLA if I print at 220C and vary speed between 50-120mm/s

When I get home later, I will share the model I made for testing this. It uses layer change gcode to apply speed changes to the default PLA profile, so you just set a temperature and from the print you should be able to identify the range of speeds that produce smooth surfaces.

Also, I read your first post again and your flow ratio is quite low. I would move it back up to 97-98. Surface texture is strongly affected by flow ratio, and as soon as you’re able to remove some of these variables from the picture, it should become much clearer how to better control surface finish or texture

[95Prints3D]

You say 50-120mms but its defaulted to 200mms. Honestly dont even bother, my x1c will be here tuesday. I tried for 9 months on this thing. I appreciate it.

[Thermal_Runaway]

That was not my point though! That was the range of speeds I looked at, but surface finish is highly dependent on not just print speed but temperature and layer height. My result above was with 0.18mm layer height at 220C, but if you print with a 0.12-0.14mm layer height you could easily get a glossy finish at 150-200mm/s. I will do some testing tomorrow and see if I can narrow down a temperature range + layer height that will work at 200 mm/s for a shiny surface.

Also you need to remember that only the outer wall matters here! Say you used a 0.18mm layer height with PLA at 220C like I did. If you print the outer wall somewhere between 50-80 mm/s, you are very likely to get a glossy finish. Your normal print speeds can stay the same, so it only slows down for outer walls.

You are very close to a solution here. I’m putting together a couple different models which I’ll post here.

[Thermal_Runaway]

You can actually predict what the surface roughness aka surface finish of an FDM print will be based on operating conditions of layer height, print speed, and temperature. You need empirical data to do this, but my point is it’s highly repeatable if the right variables are controlled.

One of many papers on this topic: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10878859/

Someone a while back even made a slicer modification that allows you to spatially control the surface roughness of your prints. Nothing like this exists in Bambu Studio or Orca, but in time there will! https://www.repetier.com/velocity-painting/

[ProtoSpyre]

There’s a few parameters that coincide to achieve top layer results. I understood flow was your major doing but did you run a bed tram? see how you’re alignment was to the bed from the TH? New Hotend?

If you only print models around that size of the flow tests than you won’t really run into issues with surface quality. Otherwise you’ll really never get consistent settings to achieve a perfectly flat top layer or completely smooth… too many variables and inconsistencies with the machine parts unlike a CNC lathe or mill that are built to maintain tolerances.

[95Prints3D]

This is how my bed was from factory. Tried manually leveling it following wiki. I heard this isnt a good way to see if its flat. Dont have the patience to see if its square. I print all kinds of stuff, super small to entire bed length.

[odingalt]

I have X1C for ABS prints and to print toys quickly. Quality is meh. My ABS prints are weak because default speeds don’t lend to good interlayer bonding. Warping is always extreme no thanks to my banana shaped heat bed (it’s not flat) and top and bottom layer quality is always horrible. Yup, wasted like tens of hours on calibrations and various settings to no avail other than counterintuitively I cra k the cooling up to 85% as my default for ABS prints despite conventional advice saying to disable or lower cooling.

I have two MK3S for PETG and PLA and when I need the item to look good or I need strength from better layer bonding. They print beautiful prints albeit slowly every single time without fail.

[user_3026326371]

As some pointed out, I AM damaged goods when it comes to 3D printing.
I started during the early days when all we had was 3mm filament of questionable quality LOL

I am actually quite happy with my print results now, nut it took quite some time to get things right…

I really like to calibrate every new colour or material I add - from scratch and totally ignoring what Bambu has in mind as I find their defaults, how shall I say it? Unsatisfactory…
Have get the camera out one day to upload some pics I guess…

Take the dreaded top surfaces…
I have seen an endless amount of corresponding topics, all discussing how to get this vital layer look perfect…
Ironing, slight changes here and there but in most cases the result only gets a bit better but never reaches a quality I would consider good enough to sell a model like this.
I found that Bambu makes it way too easy for the user while at the same time making it quite difficult to get around certain issues.
Had to change my calibration and tuning routines because of this…
A top surface is NOT just the last layer of a model, it IS ALL top surfaces, with that whatever foundation they might have.
For some reason this seems to have got lost when we switched to faster and more capable machines/slicers.

The FIRST layer was always the most important one for a print, if you ask me then it still is.
This layer defines how well or bad the second layer goes down.
Since we usually print those first layers with 100% infill any minor issues keeps building up.
Becomes quite obvious when you try to print a test cube with 100%infill.
I can also find well meant replies where users recommend to NEVER use a 100% for inside a model - because it tends to turn into a mess… LOL
If you can’t get a clean and smooth surface with layer 3 then you never will…
And realistically already the 2nd layer should be perfect.

My flat top surfaces are smooth, clean and free from these bump lines around the outer walls.
There is no gap between the lines, no rough lines either…
No need for scarf seems on my end either…
An aligned seem IS visible but only a slight issues for tight push fits, for which I just sand over the seam if required.
With random start points I have a hard time finding these points and have to either feel them or grab a magnifying glass.
How did I get there ? Through ignorance and hard work LOL

Flow ratio and print temp come first on my list.
Then the K-factor, overlaps, retraction and so on.
Only after I am happy with the results I go and adjust the calibration factors for holes and outer dimensions in order to get the required accuracy.
I can design a M8 by 1.25 bolt and nut to DIN specs, print them and they fit together the same way they fit on the metal originals.
I don’t design with FDM tolerances in mind, I design to actual dimensions and print the parts like this.

Bambu has ease of use and high volume printing in mind, preferably only using their own filaments of course.
They did not re-invent 3D printing though and whatever we had to do in the old days to get good quality and accurate prints still applies the same way.
Yes, the defaults are MUCH closer to being usable than a few years ago but if you really want the lot you won’t get around proper MANUAL calibration.
And that only gets you as far as the flimsy hardware allows…

[henry140]

Happy you’ve found the solution!

[bgechan2010]

hey @95Prints3D il take ur p1 printer lol where u live