# Z\_TILT improvements **URL:** \ **Category:** Features\ **Created:** [March 28, 2021, 2:58pm UTC](https://klipper.discourse.group/t/z-tilt-improvements/49 "2021-03-28T14:58:37Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![sensille](https://avatars.discourse-cdn.com/v4/letter/s/a87d85/32.png) [@sensille](https://klipper.discourse.group/u/sensille)\ **Post date:** [March 28, 2021, 2:58pm UTC](https://klipper.discourse.group/t/z-tilt-improvements/49/1 "2021-03-28T14:58:37Z")
I added 2 new commands to improve Z\_TILT accuracy. The goal is to reach the optimal bed position on the first run, without the need of retries. There are 2 problems with z\_tilt: - Often the exact positions of steppers/pivot points are not known. The branch adds a command Z\_TILT\_AUTODETECT to determine those positions. It is currently only implemented for 3 z-motors. - If the bed is not 100% level, probing only 3 points can lead to a (reproducibly) non-level bed. To address this, klipper allows to probe in more than 3 points and does a least-squares approximation to find the optimal positions. The downside is that this elongates the probing process. To address this, the branch adds the command Z\_TILT\_CALIBRATE. This probes a full set of points and determines the z-offsets of the main set of probes. When Z\_TILT\_ADJUST later probes the main probes, it uses the saved offsets to restore the positions reached with the full set of probes, thus speeding up the process significantly. I also only tested it with 3 motors, but it can potentially be useful also in other setups. It works well for me, but getting some more testes would be great. You can find it here: > Feedback would be greatly appreciated. ---
**Author:** ![EddyMI3D](https://yyz2.discourse-cdn.com/free1/user_avatar/klipper.discourse.group/eddymi3d/32/1981_2.png) [@EddyMI3D](https://klipper.discourse.group/u/EddyMI3D)\ **Post date:** [May 10, 2024, 9:06pm UTC](https://klipper.discourse.group/t/z-tilt-improvements/49/2 "2024-05-10T21:06:01Z")