Tilt correction in confocal scanner behaves strangely

[NicolasStaudenmaier]

What is affected by this bug?

When using tilt correction in the confocal scanner the z-position behaves strangely in the position optimization (optical refocus) when changing from the initial position (where the tilt correction has been turned on) to another. The z-scan of the position optimization might show a nice result but the actual z-position that will be set after the optimization is somehow different. I observe this because the fluorescence after the optimization is lower than the maximum.

When does this occur?

Occurs when tilt correction is used and you do an optical refocus in a different position than the initial one.

Where on the platform does it happen?

Probably some bug with the interplay of tilt correction and the optimizer logic. No error, warning etc. is given.

How do we replicate the issue?

Activate tilt correction, go to a different spot, do position optimization.

Expected behavior (i.e. solution)

The optimizer logic has to account for the effect of the tilt correction. So the tilt has to be either included or excluded (I don't know how it is implemented) in order to go to the correct z-position in an optical refocus.

Other Comments

[Neverhorst]

Tilt correction is very obscure in the current confocal implementation since responsibilities about positioning are not clearly defined. This should be better in the new core, although there is currently no tilt correction implemented for the new scanning architecture. But there is a single place where it can be injected to affect everything (scanning and optimize) equally.