User Tools

Site Tools


mdcpickandplace:fiducials

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
mdcpickandplace:fiducials [2010/06/11 19:00]
ladyada
mdcpickandplace:fiducials [2016/01/28 18:05] (current)
Line 48: Line 48:
 Make sure you set the Bias point first so that you have that correct, we put the fid's as the first two elements in the PCB parts. Make sure you set the Bias point first so that you have that correct, we put the fid's as the first two elements in the PCB parts.
  
 +
 +====== Offsetting ======
 +
 +One problem that may occur with fiducials is the harsh direct light from the top vision may create '​bumps'​ that show up dark, confusing the vision detection. This is due to the solder surface being uneven, the shadows show up particularly well as seen here:
 +
 +{{:​mdcpickandplace:​arduinomark.jpg|}}
 +
 +Try reducing the **Level** to 0.8 or 0.9. If that doesn'​t help, a solution is to program in the fiducial with a detection offset, say 4mm in one direction. This way the light is angled, more diffused:
 +
 +
 +{{:​mdcpickandplace:​arduinomark4mmoffset.jpg|}}
 +
 +Put **4** into the **dx0** column to shift over the vision recognition.
 +
 +^ Part ^ Nozzle ^ dx0 ^ dy0 ^ st ^ da9 ^ X ^ Y ^ Pt ^ W ^ Thick ^ LunaVis ^ TS ^ Level ^ Type ^ Voff ^ Remark ^
 +^ 1mm fiducial | 2 | **4** | 0 | 0 | 0 | 2 | 2 | 0 | 0 | 1 | 1 | 50 | 0.9 | MARK3 | | 1mmfi |
 +
 +{{:​mdcpickandplace:​4mmoffset.jpg|}}
 +
 +You can also try Teach mode, press Shift+S and enter "​3,​2,​2,​1,​0.9,​4"​ for test (MARK3 type mark, 2mm by 2mm window, 1mm dot, 0.9 Level, 4mm offset)
  
/home/ladyada/public_html/wiki/data/attic/mdcpickandplace/fiducials.1276282815.txt.gz · Last modified: 2016/01/28 18:05 (external edit)