BLT offset wizzard when homing is with limiters

[Grzesinski]

I've been changing nozzles a lot lately and noticed that the BLT offset wizard (from the TFT touchscreen) is not working properly.
It should be noted that in my printer homing is performed with Z end-stop, BLTouch is used to correct the bends of the table only.
The whole procedure proceeds as expected, but the calibration result is incorrect.

Steps to reproduce

1. Manually set any reasonably BLT offset (optionally)
2. Run BLT offset wizzard
3. Check result and compare it with Marlin ofset wizzard

Expected behavior
TFT BLT Wizzard offset should gives results similar as Marlin Wizzard .

Actual behavior
The whole procedure proceeds as expected (homing, testing), but the calibration result is incorrect.
The Marlin BLT offset wizard calibration gives a result of about -2.5mm and this is the correct value, verified on many printouts.
The calibration result from the FTF display gives strange, incorrect results (-0,x -1,xmm, etc.).
The end result of the wizard depends on the current stored BLT offset, which should not be the case.

Hardware Variant

SKR MINI E3 V2.0 (TMC2209)
TFT35 E3 V3.0 + SKR MINI E3 V2.0 (TMC2209)
Endstop homing + BLTouch

TFT Firmware Version & Main Board Firmware details
BIGTREE_TFT35_V3.0_E3.27.x (2023-01)
Marlin 2.1.2 (2023-01)

Discussed in #2733

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[digant73]

try to change the value for parameter "probing_z_offset" in config.ini (e.g. set it to 1).

[Grzesinski]

That's not the reason.
I had it configured in config.ini all the time
probing_z_offset:1
As per my hardware setup description:
"- Enable it in case Marlin firmware is not configured to use the probe for Z axis homing
(e.g. USE_PROBE_FOR_Z_HOMING disabled in Marlin firmware)".

[digant73]

Yes, I provided that param because I also had a similar issue in one of my printers. I previously opened a bug report on Marlin repo and info about the offsets applied here and there by Marlin were provided but at the end it was more a bug in Marlin side than something wrong in TFT fw. At the end (for the issue I had in my printer) I could solve the issue in TFT fw with that workaround (enabling probing after homing with endstop). But is seems it is not enough for you or something was changed in Marlin now. I will try to find and link here my PR (it should report also the link of the discussion on Marlin repo).
Probably it is better to report your issue on Marlin repo just for an update about the logic applied by Marlin now.

EDIT: here the link to my PR #2146. I see that it was also a fix for a bug report you provided. Was that fix working in the past for you?

[Grzesinski]

Does this feature work for you?
I use Marlin 2.1.2 (2023-01) due to TFT FW requirements.
The changes to the config files have been synced with my previous builds. I also made the changes required by the latest release of TFT FW.
In Marlin, BLT z-offset calibration works fine.

[digant73]

yes it was (although not exactly your setup) but I don't have that printer (setup) now.
Do you remember if it was working for you in the past (before you upgraded Marlin and/or TFT fw to latest versions)?
Do you also use Home Offset in your setup?
Also, please answer to question on #2742 just to have a clear picture of that other issue

[Grzesinski]

Unfortunately, the feature also failed in my previous build Marlin 2.0.9.1+TFT (2022-02).
I was hoping the new version was fixed.
I don't have any Home Offsets (M206) set.
I will check #2742 today. TFT FW update required?

[digant73]

Ok. I will check in the future if your issue could be fixable in TFT side (it is more related to info not available to TFT).

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.