[SOLVED]: Nozzle calibration always fails

[Mark81]

Hi! I'm trying to run the nozzle calibration but it always says "can't see nozzle - nozzle calibration failed".
Surely I'm not correctly processing the images. But I need some hints to understand what the machine needs to see.
Here a screen shot without any filter, and then the filters loaded for the nozzle function.

Please, would you help me to understand what I'm doing wrong?

[ameli]

I ran into some trouble with the calibration failing as well. I believe you need to play around with the "Vision Parameters" for each nozzle you're trying to calibrate. Set the min and max size which corresponds to the acceptable radius measurements for each nozzle.

For some reason, the defaults in the instruction manual didn't work on mine, I had to tweak them until I could get the calibration to pass.

http://www.liteplacer.com/nozzles-setup/

[Mark81]

Thanks!
Still it's not clear to me how the calibration works. Which circle has to meet the radius constraints? If I take a "measure" it selects the smaller one (in the log window). Does the calibration routine do the same?
Last question: what's the meaning of the color in the camera window? Circles are drawn in different colors - I bet they have some meaning!

[ameli]

Which circle has to meet the radius constraints? If I take a "measure" it selects the smaller one (in the log window).

I'm not sure which circle (if multiples are found) the calibration routine will choose. But you want to select the radius values that correspond to the circle that actually touch the part to be placed. You can either look for the inner or outer circle of that surface.

I haven't figured out what the multiple colors mean yet, either.

[JuKu]

It uses the smallest circle that fits in the size limits.

The colors: If closest and smallest circle are the same, it is drawn in magenta (pink). If not, smallest is aqua (light blue) and closest is lime (green). Other circles are drawn in orange.

[Mark81]

It uses the smallest circle that fits in the size limits.

Hence, if I understand correctly, i.e. I need to target the inner hole of the largest nozzles because it will be the smallest circle visible. Instead I was trying to get the actual nozzle shape.

The colors: If closest and smallest circle are the same, it is drawn in magenta (pink). If not, smallest is aqua (light blue) and closest is lime (green). Other circles are drawn in orange.

Yeah, I've just figured it out form the source code right now!
So, the goal is to have the magenta one for correct recognition.

Thanks for your support!

[Mark81]

After understand how the optical calibration works, now it rocks! I was able to calibrate all the nozzle, with a different setup of the video processing effect and slighly changing the vision table (I didn't use the alternate function, so I inserted smaller diameters).

[sebastian]

After understand how the optical calibration works, now it rocks! I was able to calibrate all the nozzle, with a different setup of the video processing effect and slighly changing the vision table (I didn't use the alternate function, so I inserted smaller diameters).

Would you be so kind and share your vision settings for all nozzles

[sebastian]

I shared my own findings here: https://wiki.apertus.org/index.php/Lite ... hine_Notes

In summary the nozzle offset with the samsung nozzles is neglectible though