Rounding error on copy/paste
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
KiCad |
Fix Released
|
Unknown
|
Bug Description
When laying out multiple copies of complicated or dense components (eg: large BGAs), I'll often break out one, or just a section of one of the components and then copy/paste the resulting tracks/vias to the rest of the part and other similar parts. On a recent design, I ran into an odd problem where pasting the tracks/vias resulted in a very small rounding error on the coordinates (0.000001 mm) of just a couple of the pasted vias/tracks. This then led to endless errors with trying to select and route tracks, determine track lengths, perform length matching, odd 0-length stubs created when connecting to these vias, etc.
I was using a pretty course metric grid when copying and pasting, on the order of 0.25 mm, and this problem only affected a few of the hundreds of tracks/vias that were pasted. If it had been every pasted via I would have noticed it much sooner, but since it was only a handful of the ~500 that were pasted it made it much more elusive.
Here's an excerpt of the .kicad_pcb file for some of the vias that were affected:
(via (at 192.4 94.750001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDD5))
(via (at 188.4 94.750001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDF9))
(via (at 186.8 106.000001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDB0))
(via (at 194 106.000001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDBC))
(via (at 192.4 106.000001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDD5))
(via (at 188.4 106.000001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDF9))
(via (at 186.8 117.250001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDB0))
(via (at 194 117.250001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDBC))
(via (at 192.4 117.250001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDD5))
(via (at 188.4 117.250001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDF9))
(via (at 186.8 128.500001) (size 0.4572) (drill 0.2032) (layers F.Cu B.Cu) (net 2) (tstamp 5DD3DDB0))
I was using KiCAD version 5.1.4 in the standard js-reynaud/
Changed in kicad: | |
importance: | Undecided → Unknown |
status: | Expired → Fix Released |
Does the problem still exist in 5.1.5? Please copy the full version info from About KiCad -> Copy Version Info (this is required information for all bug reports).