I have had a P1S for several months now. I was printing a few holders for my drill bits (Bambu PLA in Red and Blue) and the holes weren’t fitting even with a little bit “extra” diameter (all created in Bambu Slicer).
I then started looking into X/Y/Z calibration methods (and skew compensation per an article here). While the X/Y skew was slightly off, I was able to follow the process to improve it using temporary G-code.
However, careful measurement of 20mm cubes and a 100mm xyz test print (Make of 100mm x 10mm Axis calibration piece by Janis Rode | Download free STL model | Printables.com) reveals that the Y dimension is a little off and X is worse. Z is spot on. I found a G-code procedure that would allow “fixing” this using M92, but that requires being able to get the current values before modding, which would require console access. And that’s where I’m stuck.
Any way to do this with Bambu printers? OctoPrint won’t work, bambu_cli is not supported and may not work at all, …
Thanks
Welcome to the locked-down world of Bambu. CLI or SSH access into the shell is one of Bambu’s closely held secrets. It’s not likely you’ll find it without their direct help and so far, they have expressed belligerence in sharing anything that they consider sensitive. You could open a ticket and try your luck. Maybe you will get a newbie who doesn’t know better and somehow answers your question.
I have a P1S am having a x axis xcalibration and zaxis issue currently also. 20mm cube measures x=19.93 and y=20.00 and z=20.10. If you figure this out can you post so I might be able to figure out how to fix my issue (not sure about skew working on that). With the Ender3Pro you just adjusted the step but that does not seem to be possible with Bambu P1S
How much are those off? Typically, differences between X and Y in a CoreXY printer come from uneven belt tension and/or skew.
Any relative error (prints are too small by 0,x%) in both directions more often than not doesn’t come from inaccurate mechanics, so you shouldn’t touch esteps. Instead the root cause is material shrinking when it is cooling down. As this is individual per material, you better use shrinkage compensation in the material profiles. Orca Slicer has it for quite some time, but BambuStudio has integrated it too now.
those 20mm calibration cubes are a terrible tool to identify scaling issues.
Here is an excellent video that explains why:
If you are interested, have a look at my take on a better suited model:
For more info on skew compensation, there is an extensive thread at:
Thanks for the reply. I already stated that I was able to dial in the skew per the Bambu Lab model/blog. I am not relying on the 20mm cube. Instead I am using the 100mm calibration model I listed. X was off by .27%, Y by 0.20%. Scaling the same model with those values gave me close to perfect accuracy.
I will check out the video and your model and see what I can make of the shrinkage when my current prints finish. I left shrinkage at the default for the Bambu Lab PLA Basic that I have been using the last few days.
Sorry, I didn’t want to be ignorant. The cube information was more for @CrazieBird and I wanted to give a complete answer if somebody else reads the thread.
I printed your model and made a (very) slight adjustment to the shrinkage value for one filament. I’ll do a few others later today and also try to check the effect.
It might be painfully slow, but it seems like if you can’t get access to the data you can get there by trial and error. You’ve already calculated the error. Do more test prints with the compensations based on those. Keep repeating until you’ve minimized the error and you’ll be pretty darn close. I’ve never felt that 3D printing was supposed to be as accurate as some of the other manufacturing methods. But maybe I’m selling it short.
Prints from my “desktop” X1C are at least ± 0.22%, and often in the “industrial” range or better (± 0.15 mm), especially if I’ve taken the time to calibrate filament shrinkage. Better accuracy than that on a single print is possible, but not worth the effort, and will probably not be consistently repeatable. Errors are usually on the plus side.
Still, I do not expect much precision on small holes. They are close enough if they are for screws and I usually plan to drill to size if more precision is needed.
For the most part, I have been satisfied with “out of the box” precision on my P1S. Where the size becomes a problem is (1) printing parts that need to fit together (often multiple plates/filament colors) and (2) holders for drills and sockets where I don’t want them to fall out. So far, with the limited experience I have, case 1 hasn’t impacted me much other than a few tight fits (holes likely).
In the latter case, xy hole compensation might be sufficient to get a reasonably tight but not too small fit, but it can be tedious. Out of the five gridfinity bins I printed in the past week, three were too tight (shanks, not business end of the bits), one was nearly exact but kind of tight (it will loosen a little) and one was way big (after calculating shrinkage). I thought I applied the calculated shrinkage but I didn’t set it properly in the filament settings. But that would have made the hole even bigger. And I did apply xy hole compensation but maybe too much. And for those, the XY dimensions are less important as long as the bins fit the base and side-by-side (they do).
For now, I know what I know and will apply that when needed. Thanks all!
