Up camera offset / nozzle calibration

[robert@serve101.org]

On a recent batch of boards that needed building, I decided to give nozzle calibration another try.

First, I moved a nozzle to the 'up camera' with a zero rotation and set the 'up camera' offset values. But then, I did a calibration of the nozzle, and captured the results. I then calculated the average X and Y offset for the nozzle's various angular positions. Since these did not indicate an average of near zero, I adjusted the up camera offset values. I repeated this process a few times, until the average X Y calibration offsets of the nozzle was very close to zero. I hoped this would be a more accurate offset to the up camera.

After doing this, the placement of parts was quite a bit more accurate. So perhaps this new method of determining the 'up camera' offset made a difference.