did my first proper cell tower inspection last week and ran into some wierd signal issues about 2/3 of the way up the structure. im running an M30T and the RC Plus controller, so range shouldnt have been the problem at all ? i was maybe 80 feet away from the tower horizontally and the thing is only like 180 feet AGL. lost video feed for about 8 seconds, drone held its hover fine but i couldnt see anything which is obviously useless for documentation purposes.

my guess is RF interference from the active antennas on the tower but im not totally sure. the site was live, carrier hadnt done a shutdown for us. has anyone dealt with this on live towers and is there a standard distance you try to maintain from the antenna arrays themselves? i dont want to get closer than i have to but some of the mounts and hardware i needed to document are basically right next to the panels.

also the client wants thermal alongside visible light and i wasnt running the thermal at the same time ? wonder if that had anything to do with it, probably not but throwing it out there

yeah thats almost certainly the RF environment messing with your occusync or whatever link the M30T uses. live cell towers are brutal for this, especially if theres LTE and 5G NR running simultaneously on the same structure. the 2.4ghz and 5.8ghz bands the controller uses are basically in the same neighborhood as some of that traffic and you get all kinds of multipath and interference artifacts. some guys switch the link to 900mhz mode if the controller supports it for exactly this reason, cuts through better in RF-dense environments though you lose some bandwidth obviously.

maintaining distance from the active panels is more of a judgment call but ive heard people say no closer than 10-15 feet from the face of the panel arrays as a rough starting point. the stuff behind the panels, structural welds, mounting hardware, cable ladder runs ? you can usually get tighter on those since youre not directly in the beam path. thermal dropout I doubt had anything to do with it, the sensors share the same video chain but the processing side shouldnt affect the link itself.

we always request a maintenance window on live towers before flying close, even if the client pushes back on scheduling. had a situation about two years ago where we got video corruption and a compass error warning on a Phantom 4 RTK about 15 feet from an active panel cluster and it was not a fun few seconds. the drone was fine but the documentation photos from that section were garbage and we had to redo the whole upper tier. not worth it to rush it on a live structure honestly.