A solution to an abl problem
The whole 3 years i had this 3d printer i denied that abl was giving me false data. I just thought that my bed was just that crooked that abl just failed.
Then i had to print something that was almost the same size as my print bed. And obviously i had major issues with the first layer because abl was actually giving klipper false data.
On the first layer, the front of the bed was too close, the middle was almost perfect and the back part was too far.
Usually i just compensate by moving the z offset up or down but in this situation, i had mo idea where to move. I couldn't move down because the nozzle is already kind of scratching the front of the bed. I also couldn't move up because the back part of the plate was already barely sticking to the plate.
By the way i made sure to manually level the bed to be as flat as possible. I went through the 4 bed scews like 10 times just making micro adjustments to make sure the feeler guage has a consistent feel for each corner
THE SOLUTION:
The solution was a zero offset cr touch mount.
Turns out my x axis is kind of twisted. Its such a small amount to tell that it was but i just assumed that maybe it was twisted, in order to fix my abl problem.
The problem was that my x axis was actually twisted and since the cr touch was 30mm in front of the nozzle and the nozzle is the zero reference for the cr touch offset, every time it would do the abl, the twist would get multiplied at the cr touch part.
The zero offset cr touch mount solved it because instead of the twist getting multiplied at the cr touch part which is 30mm in front of the nozzle, the mount puts it aligned with the nozzle so the cr touch twists with the nozzle, giving klipper accurate abl results.