Attempted to shim the toolhead-to-carriage interface
However:
Any shim thick enough to correct the angle prevents the X gantry belt from engaging
The design does not allow enough adjustment range
Photos / files
Photos showing printed lean
Used a lego plate here as I didn’t have a square available at the time of taking the pictures. Their tolerances are pretty good so I figured it would be good enough to show the lean. I did end up checking the prints with the square I later bought and it showed the same lean.
Just to be on the safe side: Is your table also level?
Not that it’s exactly the other way around:
Bed: NOT level
Toolhead: level
Why am I asking such a strange question? Even if the tool head is tilted, it shouldn’t affect the print in this way. Quality? Yes, but it won’t produce tilted cubes. Why? It’s like holding a pen at an angle - yes, you might write worse, but the position doesn’t change in terms of height.
There are really only two possible reasons for this error: either the height axis is misaligned, which can be ruled out if you have already replaced it, or it is the print plate axis.
As mentioned, imagine a pen: even if you tilt it permanently, it will still move to the correct positions, only with a fixed incorrect offset. But in your case, the position changes in height. If you say you’ve already replaced the frame, I would check the printer plate again, as I said, to make sure that your table or the printer feet aren’t uneven and have interfered with your measurements.
I could be completely wrong! It’s just a thought I wanted to share! It reminds me of CNC machines that are set up incorrectly. If a cube is leaning, it’s rarely because of the tool head, but rather because of the axes. With a tilted tool head, only the central point is shifted, but it does not produce leaning objects; rather, an axis must be at an angle.
If I may guess: Your printer plate axis does not match the alignment of the printer height axis. Your tilted tool head is certainly also a problem, but not the cause. But hey, as I said, just a guess.
Many words, explained much more simply. I have now drawn it in with a tilted toolhead. Sorry, I couldn’t draw it more precisely, but I think you can see that an angled tool head doesn’t tilt the cube, but rather an angled axis.
Thanks for the detailed reply and for taking the time to think it through. The pen analogy and diagram were genuinely helpful and made me re-evaluate some of my earlier assumptions.
After going back over everything, I agree with your main point: a pure toolhead tilt on its own shouldn’t create a growing positional error with Z. A lean that increases with height pretty much has to come from some axis not being perpendicular to another.
At this stage though, the whole Z frame has already been replaced and the results are completely unchanged. That makes me think whatever is off is either deeper in the axis-to-axis alignment or something that just isn’t realistically diagnosable or adjustable on the user side.
I think I’ll need to go back to Bambu support and see if they’re willing to do a deeper inspection of the machine themselves rather than continuing with more trial-and-error part swaps.
