Backlash issue : large error detected

[marceau....@gmail.com]

Hi,

From several days I try to do the backlash calibration, but got same error each time :

 

Before I do the pixel per mm calibration, the Primary fiducial calibration, but same error when I try to do the backlash calibration.

Someone had an idea ? 

[bert shivaan]

first can you confirm if the 1.973mm is correct? If it is, you might want to try and fix it on your machine. If it is not, then the pixel size must be wrong I think.

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/79bd2e21-ce1c-4aad-a4e1-c1bf7424cb61n%40googlegroups.com.

[marceau....@gmail.com]

I doesn't exactly understand what correspond the 1.973mm.  So I don't know where I need to search ^^

I can say that when I ask the machine to move to 1mm, it is ok, I checked this with a ruler.
If I ask to moove the machine with the "drag" function on the picture, it is ok if the target point is near from the center (some millimeters), but if the target if far (1cm or 2), I had a little placement error.

[mark maker]

Strange. Just to double-check:

1. you did the preliminary Camera calibration more or less as shown in the first part of this video:
   https://youtu.be/md68n_J7uto
   Right?
   
   
2. and then - without changing anything on the machine or in the camera setting - you did the backlash calibration, using the same unmoved fiducial, i.e. the primary fiducial?
   Like in the second part of this video?
   https://youtu.be/md68n_J7uto?t=329
   
   

_Mark

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/d1edab7c-0253-4d8c-aaaf-f51316f6e4b7n%40googlegroups.com.

[marceau....@gmail.com]

Yes, I do this step in this order and nothing else, no moove of the fiducial, nothing :

After that I try to do the Backlash compensation, and again got the error...

[marceau....@gmail.com]

Hi,

Someone got the same issue ?

[mark maker]

Put a ruler or millimeter paper on a surface, that is at PCB/default Z, i.e. the same as the primary fiducial Z!

Then check the ruler/millimeter paper against the grid on the cross-hairs. Does it match?

_Mark

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/df64863f-1a0f-41fc-bc65-0d00bb04dc6dn%40googlegroups.com.