Hi John,
From your picture it appears that the cable has broken at the active antenna. I have made a similar repair on a Meshtastic device which it worked fine. Unfortunately I can’t give you any details on where your connection should be. You can buy replacement active antennas if you don’t fancy the fiddly soldering. The antenna has a Hirose coax connector to a socket on the main board.
This is the stock one from the T Beam (I don’t use it any more, it’s rubbish). The wire is soldered under the metal shield and looks difficult/impossible to access.
This is an upgraded one off eBay. It looks like shield/braid of the cable is soldered to the copper track that the metal shield is also soldered to, and the core is soldered somewhere under the shield. The metal shield looks like it could be unsoldered and reattached after surgery.
@MW0KXN - I have ordered a pair of GPS antennas of the same size I had before. It does look like the cable has snapped off from under the shield. @MM0EFI - Thanks. I have new ones on order @GW4BML - not a lot of room in the 3D printed case for a bigger patch GPS @MW0PJE - I have a lillygo board. two new ones on order. @G6WRW - I think a bigger box for mine might be in order, I had to re-solder the antenna mount as well. @MM0FMF - thanks for the offer Andy, but new ones on order.
LoRa devices are low power because they are designed to work in licence exempt ISM and SRD bands. When used with an amateur radio licence why not use more power?
Would increasing the power for LoRa make a difference? If 100mW can easily go almost 200km would a modest increase in power to, say, 10W allow much longer range via troposcatter? Or would high gain yagis pointing in the right direction still be needed?