Delta auto-calibration - changes in bed tilt calculations
garyd9 last edited by
Assumptions (please tell me if these are wrong.)
My understanding of the delta auto-calibration is that the S parameter provided with G30 in bed.g can be 8 to request that the duet board try to find the proper values for bed tilt (which result in changes to M666's A and B parameters.) The "Calibrating a delta printer" wiki page warns that "Unfortunately, auto calibration can't easily distinguish between bed tilt and differences in tower separation."
I think the tower separation values are X,Y, and Z parameters of M665.
If the G30 in bed.g uses G6 (instead of S8), then bed tilt calculations are not done (and M666 A/B parameters are not modified.) In this case, the tower separation values are still re-calculated, however.
Request (and reason for request)
I'd like to request a new G30 parameter that calibrates bed tilt (M666 A/B) and assumes that the tower separation values are already correct (in a manner similar to how S6 assumes that bed tilt is already correct.) My hope is that doing this would allow for more accurate bed tilt calculations if/when the tower separation values in M665 X/Y/Z are already correct.
My reasoning here is that I think that it's much more likely for the bed (especially a removable bed) on a delta to slightly change it's tilt when compared to the likelihood of the vertical towers somehow moving. This is especially true in a delta where the corners and towers are all metal constructed, but the bed is moved/removed/replaced fairly frequently for cleaning, etc. Each movement has the potential to change the bed tilt.
Sounds reasonable! I'll add this to the wishlist.