We are still training our BlueROV2 deployments and during an afternoon of 4 dives between 50 and 95 m depth searching for corals, the “leak detected” alert popped up while the ROV was at the bottom of a cliff at nearly 90 m depth. After a “slight” panic ascent, we could not see any water in the ROV although the alert did not stop. I assumed that if there had been a leak at 90m, then the ROV should have been flooded by the time it surfaced and as it was not the case, my next assumption was that the detectors were getting old and maybe moisture from the air (Okinawan climate is super humid, we even started to get mold on some cables) triggered one of the sensor. So we decided to make a last dive “just to check” (and to have a chance to pull back the tether calmly and roll it properly this time). This last 65 m dive went very fine, although I got a bit overenthusiastic and spent a bit longer at the bottom than the “thouch and go” initially planned.
Back in the lab, we noticed that indeed a drop of water had soaked in the sensor the closest to the back cap, but with no other signs of humidity anywhere else.
Does someone have an idea how can a single drop of water enter, but nothing more, even with a subsequent dive (although slightly shallower)?
Also is there any recommended procedure if you get a leak alert but see no liquid inside? Of course the best is to stop all operations, but when you rent a boat several hundreds dollars for half a day, it is tempting to find a less radical option (although of course one drown ROV will cost more than a few days of boat rental).
Attached are images of the first leak alert, and of the coral community that kept me down longer than planned on the last dive.