Is anyone still using 5-point bed compensation?
-
...because I'm thinking of removing it from the RRF 3.01 release. The recommended replacement is 3x3 mesh bed compensation, although that doesn't allow you to set trigger height corrections (G30 H parameters) on each point.
-
@dc42 is this something like G30 S5 where its supposed to execute (at least on non-deltas) adjustments to the bed plane (live levelling)? I was planning on giving it a try (because it covers more of my large bed vs just 3 point i'm currently using) but if there's an option to use a 3x3 grid (9pts) instead of 5pts to tram my bed, no worries.
-
@Luke-sLaboratory said in Is anyone still using 5-point bed compensation?:
@dc42 is this something like G30 S5 where its supposed to execute (at least on non-deltas) adjustments to the bed plane (live levelling)? I was planning on giving it a try (because it covers more of my large bed vs just 3 point i'm currently using) but if there's an option to use a 3x3 grid (9pts) instead of 5pts to tram my bed, no worries.
Yes.
-
@dc42 Neat! I have no issues with dropping 5 point if it means we can use a mesh to retilt the bed.
Is the mesh option for live-level a current feature, or something that will be added in 3.01 under G29 (or similar)?
-
@Luke-sLaboratory said in Is anyone still using 5-point bed compensation?:
@dc42 Neat! I have no issues with dropping 5 point if it means we can use a mesh to retilt the bed.
Is the mesh option for live-level a current feature, or something that will be added in 3.01 under G29 (or similar)?
I think you are asking whether it will in future possible to load a height map, then probe a small number of points and have RRF superimpose a tilt on the height map according to the result. If so, then yes I am considering adding such a feature, which would make use of the existing code for bed compensation based on 3- and 4-point probing. But 5-point probing doesn't fit into that scheme.
-
@dc42 said in Is anyone still using 5-point bed compensation?:
I think you are asking whether it will in future possible to load a height map, then probe a small number of points and have RRF superimpose a tilt on the height map according to the result. If so, then yes I am considering adding such a feature, which would make use of the existing code for bed compensation based on 3- and 4-point probing. But 5-point probing doesn't fit into that scheme.
I'm not looking for a superimposed tilt on the mesh, I'm looking for using a mesh to inform the live-level feature (could be a coarse 3x3 or 5x5 mesh) before a detailed "surface map" mesh is used. Just to be clear, instead of using individual points like in the tutorials here: https://duet3d.dozuki.com/Wiki/Bed_levelling_using_multiple_independent_Z_motors I'd like to be able to specify the mesh to use for calibration like
M557 X15:600 Y15:600 P5:5 ; Set mesh for bed auto-tilt G29 S0 A1 B3 ; Probe New Mesh, use for autocalibration, 3 motors,
This (in theory) would use the same technique as current bed levelling w/ multiple independent z-motors to calulate the movement needed, but would greatly simplify user inputs.
for a cherry on top, could build in similar z-probe features like results must be within XXX variation and YYY maximum amount of tries to get that, similar to what conditional scripts are doing today.
-
We were talking about different things. It will still be possible to use 5 points when doing bed tilt correction using independent Z motors.
-
@dc42 Thanks for the explanation, sounds good.
Would be pretty neat to be able to use a mesh for live levelling though!
-
@Luke-sLaboratory said in Is anyone still using 5-point bed compensation?:
@dc42 Thanks for the explanation, sounds good.
Would be pretty neat to be able to use a mesh for live levelling though!
Why?
-
Greatly would simplify setting up the bed.g macro and aid in understanding (at least for me!) - once you know how the mesh works, you know how both live-levelling and mesh compensation work.
Purely convenience. I haven't touched my bed.g in a few months.