I’m aware that the petg profiles in bambu studio are way off, and I have taken some time reading through vartious posts and have managed to get the temperature tower to print almost good enough. Is there any particular order for carrying out ther rest of the calibration tests, since I expect the results of one effects the calibration of others?
That’s a good question. I’ve asked it both here and on the SoftFever Github site, but never found a definitive answer.
Some say flow rate should be first, then temp tower and PA.
My usual procedure for any new filament is to use menu order - temp tower, flow rate, PA. I’m then usually quite happy with the results, so I’ve only used the other calibrations to fix problems as they occur.
I often wonder if a second pass through all of them, using a limited range of values, might be produce even better results. Maybe I’ll experiment someday when I’m bored.
1 Like
Thanks. I’ve also tried the esun json files, but they seem little different than the generic Bambu files, and fail miserably. I’m beginning to think that the 0.4 nozzle would be easier, I’m sort of fiddling around with too many unkowns. I fitted an 0.6 hardened nozzle to my old modified ender 3, a year or two back, experimenting with cf, and found it printed as good as the 0.4 for the functional stuff that I was doing, and was faster. I fitted the 0.6 to this p1s, but I’m not sure if it would print faster than the 0.4, since it will be limited by the rate of melt, not the printer speed. I used Cura before, and was more familiar, I guess, as to where to find the settings that needed changing. Maybe I should start again with Orca, from first principles.
I’ve read where some have suggested using the bambu lab PETG profiles as your starting point rather than the generic PETG profiles.
Anyhow, it seems a shame we should have to start with either one, because by now it’s very likely that someone has already dialed-n and confirmed most PETG’s on the market. If only there were an organized collection point to support people sharing their full, detailed PETG profiles, then maybe we wouldn’t have to waste time repeating work that someone else has already ]done.
“Starting point” profiles will usually provide a decent print, but there will always be room for improvement. It’s mostly up to you to decide if you want to look for improvements.
There is no single profile that will provide optimum prints for everyone because there are too many variables involved. Temperature, humidity, printer, nozzle, build plate, slicer settings, manufacturing batch, filament color, etc. all have potential effects. I can guarantee that someone printing in a climate controlled office is using different presets than I do in a damp, just-barely-not-frozen garage workshop.
Anyone getting great prints has probably calibrated their filament in at least some respects, but I doubt that you will find even three such people that came up with the exact same settings. I can dial in the absolute best settings for my printer, but they may be unusable or just OK when used on your printer.
Don’t blame the filament makers, or the printer manufacturers, of the software developers for this variability, it’s just the way it is. It’s why generic and manufacturer profiles don’t always work. They are not going to publish a profile unless it worked great for them, on their equipment. Failures can happen because there is something different about your print conditions that made the profile unsuitable. I don’t think a database or collection of “dialed in” presets can ever exist. Some one, some where, will always have a need to make some tweaks to any profile from that collection to get better results.
Personally, I’d rather invest a little time to perform my own calibrations with a new filament at the start than to use someone else’s settings. Theirs might not work very well, so I have a failed print, and I have to calibrate anyway. Or if their presets do work, there will always be a voice in the back of my head that saying that I might be able to do even better (I often can), so I end up calibrating anyway.
2 Likes
PETG is the only material that did not print at all for me using either Bambu or Generic profiles. A temperature tower finally showed me that I needed higher temperatures than either Bambu or the manufacturer recommended, at least for some colors. Some Sunlu PETG colors need up to 265°C on my X1C with 0.4 hardened nozzle.
Wow, my head is spinning. I had been under the impression that the main selling point for buying Bambu Lab filament is that the profiles for it had been perfected, that each roll arrives with the appropriate identifying code on an RFID tag, and that pretty much all we (the end user) would need to do is load and go. Last week I even purchased 3 rolls of Bambulab PET-CF with that as my assumption. If I’d knownthere was no truth to itk, I probably would have purchased it at lower cost from a different source…
Bambu filament in a Bambu printer using the Bambu profile can print very, very well, because Bambu controls more of the variables. If you are happy using the Bambu profile, that’s great and easy. Some people don’t have the same success using some Bambu filaments and profiles. PETG seems to be a bit problematic, maybe it is more sensitive to the printing environment.
You may or may not be able to improve the results with your own calibrations. You will not know you have the best result unless you try other settings, tested individually over a range of values. That’s where the OrcaSlicer calibration tests shine.
I was happy with the small sample of Bambu PLA that came with my printer, using the Bambu profile. It is my first 3D printer and I knew nothing about calibrations at the time.
Then Sunlu PETG was a mess using either Bambu or Generic presets. Google led me to temperature towers (needed greater nozzle temp), and that led me to investigate other types of calibrations. Most of which have improved my prints enough that I now calibrate a new filament immediately after opening the bag and drying it.
Bambu offered me two very cheap spools of Bambu PLA-CF last summer. It’s nice stuff but a higher flow ratio than the Bambu preset works better for me, on my printer.
Those are the only two Bambu filaments that I have used.
Bambu filament is not my first choice because, without an AMS, the RFID tag is useless, their shipping can be slow, they are often out of stock, and I can obtain quality material quicker and much cheaper elsewhere.
1 Like
Thanks for all your responses. iirc, for the cura/ender settings, I think I may have started with the pla settings, altered retraction, temperatures, cooling, and that was about it. Of course, not enclosed, and slowerr print speed.
The best bridging on the temp. tower is at about 215 deg, which is a bit low. I think I may have a mislabeled new spool of pla, not petg, since manufacturer’s petg range is higher than that, and I’ve usually printed it at about 255. I have some older petg, I’ll try that with the bambu original settings. I’ll let you know.
not sure if this effects this, but it doesn’t ‘look right’. I’ve a p1s, but all the system presets for process say x1c. Are there any p1s system presets, or is it just a name to group together printer/nozzle, filament settings? I can add in user presets, but where to find the system presets for p1s?
Yes, it is different, since this topic is about filament presets.
The X1C and P1S are similar enough that they use the same Process profiles. Bambu does not provide P1S-specific Process defaults.
1 Like
thanks. I found the folder with the json process files, They are named for x1c and p1p. However, I’m not quite sure how Orca slicer relates to Bambu. There are two sets of profiles, one under each program. I thought Orca was a sort of wrapper for Bambu, and would not duplicate/copy it, but it would use the Bambu resources folder, for example.
OrcaSlicer is a complete stand-alone slicer on its own, you don’t need Studio for it to be fully functional. It is developed as “fork” of Studio, just as Studio is fork of PrusaSlicer, which is a fork of Slic3r.
SoftFever started with the Studio open source code and added features from Prusa and others. Bambu, in turn, has added Orca features to Studio.
OrcaSlicer and Studio use their own configuration folders - in the help menu they both have a link to their folder.
They are somewhat connected, through the Bambu cloud, if your slicer Preferences (Ctrl+P) have the Auto sync box checked:
Preset changes you make in one slicer will appear in the other, if the setting is supported by both slicers.
The sync option is helpful if you like different features that are in one slicer but not the other, or use more than one computer.
In particular, Orca stores K values (pressure advance) in the filament presets. Studio ignores those values, using the values stored on the printer and displayed on the Device tab. OrcaSlicer uses the value in the filament preset and overrides the value on the printer.
Studio offers MakerWorld connections, but I think that is the only feature that Orca does not have, while Orca offers a more complete set of filament calibration methods and many more slicer options.
1 Like
Thanks for the detailed explanation. I was confused by the fact I had also installed Bambu. The new json profiles I generate are stored under appdata/roaming. Not sure if I like that. I’ve ‘adjusted’ the fan speeds. If that doesn’t work, I think I’ll temporaily give up on the temp tower for petg. I found a guy on youtube, calibrating Orca on an x1c, who never used the temp tower, because ‘it fell over on the build plate’ .https://www.youtube.com/watch?v=lDqUBymbjJ8
I have to correct my earlier post, I learned something new today!
The bolded sentence is incorrect. When presets from OrcaSlicer have pressure advance enabled and values set, Studio will not display them, but Studio will use the values, and will also override any K value set on the Device tab.
I highly recommend this much better explanation from @ExpectDeer.
2 Likes
Thanks again. Not got far enough with petg to be concerned with pressure advance, so far. Not sure about him saying towards the end about both being stored in the cloud. Hoping it is not if I’m printing in lan mode and making changes, unless the cloud he is referring to is something else.
I’d like to try your way, in part because you say it takes only a little bit of time, but also because I get the impression you really do know what you’re talking about.
What is your method exactly? If you’ve already written a good summary description your algorithm elsewhere, then a link would suffice.
We are getting away from your original questions but…
There are default printer, process, and filament profiles from Bambu that are installed with either slicer.
If you update the program, these are also updated.
If you modify a default preset it can be saved as a User Preset with another name, and will then be available in the dropdown boxes. These are saved on your computer.
IF you also have Auto Sync selected in Preferences, these presets are also saved to the Bambu cloud, and will be available in Studio on any internet-connected computer when you log into Studio. Any preset changes made on one computer will transfer automatically to the rest (and be saved there as well).
In LAN-only mode, or without Auto sync selected, new presets and preset changes will only affect the computer you are using.
If you choose Auto Sync, OrcaSlicer and Studio will share the same presets, through the Bambu cloud, but will store them in separate locations on your computer(s).
@lkraus When I was calibrating my old modified ender 3, iirc I started from pla settings. More or less out of desperation, I’ve set bambu studio to generic pla, and the temp tower from 260 to 220, and the petg printed pretty well flawlessly for all temperatures, very few bridging problems. I’ve been assured by the filament manufacturer it was petg that was sent, and their test prints on an x1c were fine. I will see what happens when I print pla at that temp range, and some petg which I’ve used before. I did check the g code for m104, and that changed as it should. If the other petg prints ok, and the pla does not, then I will try and compare the values in the profiles.
Boy do I have you fooled!!! 
I’ve only been 3D printing for a year, with an X1C,no AMS. Printing is just for my own purposes, nothing for sale, mostly one time prints, maybe seventy-five completed items so far. Mostly repair parts and functional prints; I avoid decorative items as a waste of plastic. I recognize when I have defects, think about solutions, but if a part is functional and looks 85% OK, I’ll often let it go rather than reprint. I’d guess my initial failure rate was about 60%, went down considerably after I started calibrating my filament, dropped some more after I learned to track moisture in my filament. Nearly all of my prints are good now (and >95% on appearance), reprints are most often due to design issues, or forgetfulness/stupidity.
I do think I have pretty good grasp on the basics of using slicers, and what works for me. There are lots of variables to a getting a good print, but for me there are two rules that are most important:
Filament must be dry. Fresh out of the bag, weigh it, dry it until it stops losing weight. Record the weight before and after every use, re-dry if it has gained more than a few grams of moisture, store in vacuum bags with desiccant. Two-three grams of moisture, I’ll take a chance; over 6 grams means re-dry.
Calibrate every filament, at least the first spool for every type, color, or manufacturer, and test at least nozzle temperature, flow ratio and pressure advance. It’s only way I know that I have the correct settings. I use OrcaSlicer calibrations, but I’m still uncertain how the order of calibrations affects results. The rest of the calibrations I might use if I have a particular issue, a retraction test for excessive stringing, for example. The maximum volumetric capacity test let me boost speeds with some filaments, but speed is not a big concern for me and I don’t always run that test.
Any methodical calibration is better than none. Test one parameter at a time. I see people saying, “my print is bad, I boosted this by this random amount and I lowered that and changed the plate and it’s still bad!” This is madness. Don’t test on your full-plate castle. Use a small model (calibration model), test a single variable over a range of values, pick the best. Temp-Flow-PA are pretty straightforward to dial in and apply to every print. If you know you have those fundamental parameters correct and still have a problem, you can concentrate on the hundred other options that are harder to test, or less clear-cut.
“Only a little bit of time” is relative to the time lost to failed prints with uncalibrated filament; calibration does take some time. A hour print that fails is a loss of time and the filament, and you still don’t know what to change or by how much.
OrcaSlicer estimates the print time for these main three tests using default values at about 120 minutes. You can reduce this time substantially if you can reduce the range of values tested or if you monitor the tests and stop them when the values used start to cause worse results. New color of the same material and manufacturer that you’ve already calibrated? Start with that temp, test just two steps up and two down instead of the default four up and down. If the best is at either extreme, print just two or three steps more in that direction.
I notice the Orca wiki lists the calibration in Flow-Temp-PA order; I’ve read to today that PA should precede flow. Seems to me flow can’t be right if temp is high or low, PA will be wrong if flow is wrong, and the temperature tower will less than optimal if either flow or PA is wrong. Maybe two passes through all three is really needed to be absolutely certain we have the best settings, with a smaller range of values for the second pass?