centroid file top and bottom orientation differ by 90 degrees
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
pcb |
New
|
Undecided
|
Unassigned |
Bug Description
I am using PCB version 1.99z. I have attached a small double sided example pcb with one component on each side. The file->export->bom xy-file is:
# RefDes, Description, Value, X, Y, rotation, top/bottom
# X,Y in mils. rotation in degrees.
# -------
RN314,"
RN315,"
The locations are correct but the top is rotated 90 degrees wrt the
bottom.
I think the "heuristic" approach to determining the device orientation in the gEDA pcb code is mistaken (see src/hid/bom/bom.c and pcb.pdf section "E.3 Computation of Centroid and Rotation"). The orientation is assumed to be that between pins 1 and 2. Unfortunately, IC manufacturers are not consistent in the orientation of these pins. For example in one of my designs I have DIL pin-outs like:
1 2
3 4
5 6
and
1 6
2 5
3 4
and 4-sided SMT packages like:
3 2 1 20 19
4
5
6
and
144 143 142
1
2
3
ScreamingCircuits gives a guide to its expectations of device orientation in a centroid file in "http://
To do centroid output reliably, gEDA pcb should maintain an internal record of the rotation that has been applied to each footprint.
Changed in geda-project: | |
importance: | Undecided → Wishlist |
tags: | added: xy-export |
Hi,
You might want to have a look at:
http:// www.fed. de/downloads/ The_CAD_ Library_ of_the_ Future. pdf
AFAICT, this one follows the IPC rules and "defacto standards" of the industry.
Kind regards, Bert Timmerman