X5SA Bed Leaveling

Created by: Lester Caine, Last modification: 10 November 2025

The blog explains the aggro that I went through here!

The first problem is the simple fact that while springs make life easier when levelling the bed, they only work if the bed is perfectly flat in the first place, and while the original heater plate is better than the mains powered one, both have slight bows in them. So the springs have been replaced with extra nuts on the aligning bolts, so that the middle of the plate can be adjusted to be parallel to the corners, something that the springs would never be strong enough to achieve. This also involved the addition of extra rails on either side of the plate to correct the bow front to back. The advantage of the nuts is that they can be corrected relative to one another and accurately set. Prior to adding this extra element, the test patches tended to have varying thickness across the patch, something which was improved after tightening the mid point down against the two ends and completed once the extra mid point adjustment was added.

But this sort of jumps the gun a little as the very first step was to roughly align the x axis while power is off. Something that I missed was that while the z axis motors move in sync, and problem can kick them out of sync and mess up alignment. When this happens in the future, the trick will be to do the paper test on either side of the bed while power is off, and manually adjust both sides of the bed by turning the z axis screw threads. Homing perhaps should have separate home switches for each side of the bed but since it seems to be the right hand motor which stalls, after homing the left side is too high, and simply winding that down only gets one roughly level again. Both sides need adjusting to bring the nozzle to paper thickness away from the bed, before power is restored and hopefully homing will produce the correct result.

While the test print only produces 5 patches, it would be much better if it also provided patches in the 4 mid points so that these can be fine tuned. For the time being I have an extra pair of prints which reduce the area to the front left corner ro back right, and this was where I was when I screwed up. So now to get the main four patch test printing as it was yesterday.