So I made an antenna based on a design on thingiverse/blog post. IIRC this is some Yagi variant. I did the math and did a bunch of tedious stuff to try and make it a proper project at the time. I do not have a network analyzer, smith charts are for shaman/wEEzards, and radio is the work of the devil. I was looking to extend the range of the 2.4G channel across my house just slightly using directionality. Empirically, it worked at first. I'm here half trying to diagnose, and half trying to understand some fundamental circuit design basics. The 2.4G output has two channels coming from the same peripheral radio chip. Normally these go to the dipole antennas. I used the same micro coax connector and proper coax throughout, with proper grounds/solder station/flux/etc. The continuity tests good still. The router automatically tries to move the transmit and receive signals between the two separate 2.4G channels under normal operation, but it drops one channel regularly now. I can see this on the network status of a connected device, and the drop happens regardless of proximity. I have not physically tested that the dropped channel is the one that had my DIY antenna. It could be a software problem. My curiosity is could this asymmetry cause a common failure mode, other than something obvious like ESD?
I etch, design, KiCAD, toner transfer, photolithography, and Arduino my way around in this space as a hobbyist, but like, I'm too lazy to go on the eevblog forum to ask this formally and radio is magic IMO.
This is the output stage in a bad hand held half ass pic with no attempt to do things like clean off the residue on the chips from the thermal pad that rests over them.
That seems most likely. Takes some tuning to get the impedance match right.