Ok, I’ve gotten the updated test version and it is working much better! Very good work.
However, as I’ve now got some rather reliable creepage tests, I’ve been improving my board, and I have come across another issue that I think is not the intended result.
It appears that at least on an inner layer (oh I went ahead and increased my board to 4 layers to help with routing which was getting tight in too many places), slots can get ignored. I have double checked this slot on the 3D view, and it does show it to be cut out of the PCB, so I suspect that this creepage error should be a different location:
I also think I’m going to improve on that slot and make it look a bit more like a tab so that the creepage is forced to take a different route on the In2.Cu layer as well. It is currently hitting in that spot on the left with the red marker. I’ll strip out the fancy stuff on this version of the board and upload it if you like. … I did so, and this time the error did not occur in the same place. That error is now just missing. Perhaps the error in the DRC is older than that slot. I’m sorry I can’t be certain of that error at the moment. If it happens again at this point, I’ll know for sure.
I went ahead and added the tab, like a u shape around the pin, and now all creepage errors for that pin are gone, but the minimum creepage test condition is 20.5mm, and using the measure tool, I estimate that moving from the PTH to the In2.Cu layer’s HVH1 trace is about 7.5ish or so mm. There should probably be some sort of error there, but there’s nothing. Maybe it’s because the PTH doesn’t connect on that layer? I doubt there’s an air gap around the PTH just because there’s no copper connecting to it. Then again… if that were the case, it wouldn’t have given the error before either.
I’ll go ahead and upload this version of the test board and you can play around with it if you like. These results are enough to make me feel a bit more confident about my design, so I’ll probably send this off and build the prototype. I only need around 1.5mm of creepage distance on an inner layer, and I’m now around 5mm on most errors. That’s probably good enough as long as the relays hold up.
Just as an update since I suddenly abandoned this thread. Holidays and life have gotten in the way a bit as usual. I have found a few more bugs and have preliminarily reported them in existing bug reports that seemed similar, but I have been asked to make new bug reports for each issue. I’m not exactly sure which bugs are related or not, but I intend to make those bug reports as soon as I can.
It’s fairly easy for us to triage bugs in to groups if they look related, so don’t worry about that, but it’s a lot of work to split things out from existing bug reports. Thanks for reporting!
