I’ve been looking into it more closely and both R14 and R16 are populated. In fact, someone asked me the same thing on Reddit where I also posted about this, and it’s actually quite curious because pretty much everything on the board is populated, which honestly feels unusual nowadays.
As for the antenna, I was thinking about installing a really good quality 3 cm antenna. Using a larger one could increase the range significantly, but then some devices might not be able to respond properly, which would be counterproductive for a mesh network. Still, if I can slightly improve the reliability of both nearby transmission and reception, that would be my preferred approach.
Thanks a lot for the information about the F-antenna. Taking into account what you mentioned, it might actually benefit me since it will be horizontally aligned with several router-level devices, so mounting it vertically higher up instead of horizontally at desk height like I’ve had it until now might improve its performance. Still, it’s always good to explore what options are available.
Also, thanks for the advice about working with circuits. It’s one of those things people sometimes don’t give enough importance to, even though it really matters. I already have an ESD mat, but it ended up buried in one of the boxes I still haven’t unpacked after the move, so right now I’m kind of working in a pretty improvised way. Sorry for setting a bad example 
Regarding what you mentioned about the other chip, that’s actually one of the things that concerned me as well. That chip gets even hotter than the CM4, which usually stays well below 50°C (122°F) outside the enclosure. That’s why, besides the airflow channel I made for the CM4, I also added some ventilation slots near that chip.
I was just taking some more photos and noticed it’s an ESP32-U4WD. Supposedly its maximum temperature is 105°C (221°F), although in some places I see its operating temperature range listed as between -40°C (-40°F) and 85°C (185°F).
Using the thermal camera, with the board outside the enclosure and in a room with the air conditioning set to 24°C (75.2°F), the ESP32 usually doesn’t even reach 50°C (122°F). What really surprised me, though, was the white LED. That thing easily goes above 60°C (140°F) within just a few seconds of turning it on, which honestly surprised me quite a bit.
To be fair, the temperatures outside the enclosure actually seem pretty normal to me, except for the LED temperature. Of course, it’s not the same thing testing it uncovered in an air-conditioned room versus having it inside a very small room with other equipment generating heat and enclosed in an unventilated case. That’s precisely why I designed the enclosure with ventilation, so that once mounted vertically on the wall, it can cool itself more efficiently through convection.
I’m leaving a few more detailed photos here.
