search

sparkfunX / Desktop-PickAndPlace-CHMT36VA

Tools and conversion scripts for the CHM-T36VA Desktop Pick and Place from SparkFun
GNU General Public License v3.0
66 stars 32 forks source link

Placement data not coming through for bottom parts #13

Closed nseidle closed 6 years ago

nseidle commented 6 years ago

I think it has to do with the outline auto-detection but bottom parts are not lining up as expected.

The solution for now (because paste is sitting) is to flip the board in eagle, and output top placement data per usual methods.

nseidle commented 6 years ago

I should be specific: the placement data is coming through it's just wrong and it's just the X coordinate.

berniwa commented 6 years ago

Could you send your file? Do you have a proper dimension layer outline? Have you tried to specify the board width/height manually?

THX2112 commented 6 years ago

How wrong? Our X placement is off by a fraction of a mm, but it's close enough that the paste pulls the part in during reflow. I had assumed this was due to sloppy PCB alignment on our part.

Checking "Flip Horizontal" works for us (except for the slight misalignment). All of our PCBs have components on the bottom and we had to flip the PCBs in Eagle before using the ULP came along.

nseidle commented 6 years ago

Here is a pic of the BRD:

image

Here is the BRD: magarray-halleffect-Panel.zip

I believe the issue is that that our v-score dimension lines are causing the "Automatic offset" feature to incorrectly believe the width of the board is wider and taller than it actually is. I am away from our machine so I can't verify. The automatic offset feature is really nice, it's our panel files that are the problem.

I think the solution is to change our panelization ULP to draw the v-score lines on a different layer (not dimension, maybe mill layer) and then change our CAM ulp to then pick up the mill layer and export it to the GKO gerber.

I'm going to close out this issue. Let me know if anyone else has similar issues.