Forced position for time lapse

CBX_Micha

I played around a bit with the setting for forcing the time-lapse position.
Of course, this makes the recordings much calmer and more balanced overall, but with one glaring catch: the head zooms over the printed parts without lifting the Z-axis a bit. This can sometimes cause problems or, if there is already stringing, this problem can be made even worse.

My idea:
Wouldn't it be possible, when printing a layer, to determine a point for the next layer, which is, for example, at the extreme end of the part currently being printed (Xmin/Ymax or Xmax/Ymax)? If you could then select one of these points (use top/left top/right top/middle || ... as a snapshot position), depending on where the camera is installed, it would always be the optimal location without additional movements (over the parts).
(I hope I was able to get my idea across clearly?)

Repetier

In Printer Configuration->Tools->General Extruder Parameter you can define a small z lift which is then also used for timelapse to prevent this.

I understand what you mean, but that is too complex for this. You can manually mark positions for snapshot and only use these. Would mean havin a preprocessor which adds them according to some rules. When we support dynamic preprocessors this could be implemented, but this is not coming soon.

CBX_Micha

Ok, I thought it wouldn't be that difficult to fill 4 variables with the min/max values per layer for the next layer and use it as a snap position.
But as I said: I have no idea about programming. I just have my ideas sometimes... B)

Repetier

Problem is we operate on a stream of g-code commands not knowing what will come next and then it is too late. That is why you need to analyse complete g-code before and add marker (snapshot commands). But the idea is good to reduce it and it is not that visible that a layer is not complete. Even better is when you do it only during infill so you get smooth shell. Stopping in outer perimeter would be quite visible as well.

CBX_Micha

Well, I thought otherwise...
When a stream comes, you can read which X and Y values are passed through; At least I suspect so.
You don't have to send the maximum to the CAM for the current layer. It would be enough if, for example, the maxima of layer X were taken and used for the snapshot position for Layer X+1. From layer to layer, the maxima of X and Y do not change so massively so that they can be used for the snapshot position of the next layer.
That would at least make preprocessing unnecessary...

At least that was what I thought in my confused brain ;)