Looking interesting. The stinking big OCXO gives it away .
As I say, just mucking about at the moment. When I started i didn't know if I was laying out a breakout/dev board for the STM32F411 like the filename suggests, or a n F411 + FPGA dev board, or doing a dry run on what it obviously became. I've got a couple of very nice Trimble Resolution-T GPS
timing boards, one of which is the PPS source for my NTP server and the other which I've earmarked for a GPSDO. They have a "
difficult to find a mating part for" hard-metric header on and I thought the discipline of adding a footprint for a daughter board seemed like a useful learning tool - so the GPSDO route was taken.
I do have to say one thing about kicad and that is most of the problems you are seeing are endemic to the Mac version of it. There’s something about Qt and Cocoa which is just entirely fuckity wherever has been portrd. But I’m using 5.1.9, on windows, and it’s absolutely fine. I’m doing all the EDA stuff on windows due to the “principle of least buggery”.
Except they use wxWidgets, so Qt can't take the blame for that one. I found the same UI problems in earlier versions of KiCad on both Linux and Windows before there was a Mac version. There's just a wild lack of consistency some of which which screams "For god's sake refactor this thing so that the same actions use the same bit of code" and the other half of the scream is "... and learn something about how people use things - a five to ten key press workaround is
not the same as doing something right in the first place".
Edit: I’ve looked at some of the UBlox modules recently and I reckon they are hand solderable as well. That brings the BOM cost and board size of such things down considerably. And yes I have the same thing planned
The Resolution-T (see below) limits how small I can spin this so I thought I might as well go for the whole 100mm x 100mm and get my money's worth.
BD139 for scale.