Spend more time trying to calibrate this machine than i do printing anything. Its getting old and im ready to get rid of it. Can anyone give any advice? When i calibrate my calibrations are different than my printed results. Built profiles from top to bottom. Once i get it once print to look good the next one looks terrible. Everything i print with looks like this. Sunlu, bambu, amolen, polyterra and overture. Ive tried rolls for 6 hours and 12 hours. Ive done just about everything at this point except get rid of it.
When i run flow calibration my .98 square or 0 for either pass looks exactly how that does. I chose -5 which is .931 and reran pass 2 to get .931 again. Why do my prints look different than the calibrations, isnt that the whole point?
Heres a great example, when i use 0 in this picture that is .98. Ill choose it but when i print ANYTHING else with a top layer it looks exactly how the .95-.96 sqaures look.
Here is the exact same thing but using overture space grey pla. Results are the exact same, why?
You know what? Donât expect too much. All FDM printers are like this. Active flow rate compensation might help, but if you donât like the way the top layer turns out turn on ironing. Currently my new P1S prints like this. And I donât think itâs a problem at all.
Yeah i get that but my $200 ender machine put out better results than this.
Dunno, then. Bambu canât be too good at everything, youâve just got to adapt. Use ironing. Use a soldering iron to flatten your top layer. If you care so much, FDM might not be good. Resin is AWESOME and produces BEAUTIFUL prints. I myself donât care about the top layer because I print functional parts.
1 Like
I donât think calibration can improve much. Try ironing and play with the various ironing settings to see if you can get the surface that you want.
If the model is oneâs own design and a particular surface quality is essential, one can take that into design consideration to have that side designed to be facing down the build plate. Then that side will have a smooth or patterned surface.
1 Like
Is it overall bad quality or are you just struggling with top layers ?
There is such a thing as over-calibratingâŚ
If you have an issue with the print result you of course adjust parameters.
Sometimes these changes affect other things and through that can cause other problems.
When I started there was just temperature, extrusion factor and spring speed to worry about.
No slicer or firmware could handle more.
The results were accordinglyâŚ
Today though our machines can compensate for a lot of stuff, not always with the success we expect.
But it wonât change the fact that if you basic calibration is not top all else you adjust will fail to get the best possible results.
One specific print you calibrated for might come out great while a different model keeps failingâŚ
You are frustrated, I get thatâŚ
But did you actually ever tried from scratch with the stock profiles and from there a manual calibration?
Print temp, flow ratio, K-factor - one after the other.
Once you found the best values for those you can check and if required adjust the max flow rate.
If then test prints still come out unacceptable you MIGHT have other issues.
If not you can then fine tune on the results.
That however should all be done with the grandfather principle in mind.
Meaning you KEEP the profile from the initial calibration and create new profiles from there.
If one of them brings better results you make it a keeper as well - only ever tweak one parameter at a time.
Like that you always âbackupsâ in case a change turned bad and you have to revert.
Just noticed that you mentioned you got better result from an Ender machine. Try reducing the print speed on Top surface to what you used on the Ender and see if the result is similar.
2 Likes
I have tried at 50mms and lowered acceleration as well.
Yep built an entire profile from the ground up using orca slicer. The one i made is in the middle and i think generic profile with default settings is left and right side is overture profile. Only adjustment was K. My calibrated one had only temp change from 220 to 200 and its the worst of the 3
That probably leaves ironing the only thing left to try. It may improve but I wouldnât expect too much on that either. If a near perfect surface is needed, making that side facing down on the built plate is the way to go.
As I understand it, âIroningâ involves extruding a layer of filament on the surface. One may need to adjust the vavirous settings under ironing to get a decent result. I would use the default Ironing setting first though, before adjusting any.
Appericate that, just wanna add what ive tried going off that. Nozzle has been changed. Its been this way since the day i got it. Both nozzles it comes with leave my prints this way. I have reduced speed of top layer all the way down to 50mms and have tried different variables inbetween, 75mms, 150mms. I have lowered acceleration to match said speeds as well. Ive tried different Volumetric speeds as well from bambus suggestion of 8mms to 22mms or close to what bambus filament is. As far as ironing sometimes i print something and it has build up of filament. Ive tried different infills and % but neither change anything. Ironing wont give me what im looking for, im trying to get the top surface good looking, not most of it with defects that i already have.
I gave up a long time ago. Its been 9 months of just calibrating a single filament. Wanted to see if there was anything i havent tried. Think im just gonna try to sell it and get an x1c. Even if lidar doesnt work at least it will for some filaments⌠cant get anything to work on this machine.
Thats what i want and cant even have it on this machine. I cant even get close to good enough, if i spent 9 months in the range instead of guessing id probably not be in this situation. I already made an expensive disappointment lol
1 Like
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 wonât 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 canâ 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 âperfectâ flow rateâŚ
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 wellâŚ
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 layersâŚ
Not all temperature tests account for all these speed changes âŚ
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 profilesâŚ
2 Likes
I feel you pain. However, you may be expecting too much out of the technology. Bambuâs speeds are itâs own Achilles heal in some scenarios. Clearly you have prior experience but that may have colored your expectation. An Ender as an example doesnât 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.
Hereâs 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 Iâm 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 arenât 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.
2 Likes
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.