I am a part of the NSERC Canadian Field Robotics Network (NCFRN: http://ncfrn2.cim.mcgill.ca:8090/NCFRN/en) and we are currently testing out the OpenROV at our annual field trails in Barbados. We have noticed a flaw in the pressure relief system that causes one of the endcaps to slowly pop open. When the OpenROV is sealed up and then turned on for use in ~4-6 feet of water the change in heat inside the camber increases the pressure enough that the one of the endcaps get popped off and the pressure relief plunger does not perform as expected. Needless to say the OpenROV took on enough water to cause some problems.
We have dried everything out and have checked the components one by one for a loss of functionality. It seems that everything works (cockpit, lights, motors, IMU, etc.) when powering the beaglebone of mains via the 5V barrel connection or the mini-usb port. However when providing power via the topside adapter the beagle bone does not power up. The PWR and ETH led come on in this case and we have narrowed down the problem, which is that the beagle bone is only being provided 2.5 volts via the header pins from the OpenROV controller board.
Our quick fix is to trim the camera usb to make some room for an internal batter which provided power to the beagle. However, I was wondering if you had any suggestions as to where to look for any shorts that might be causing this issue as it is an odd failure mode. Or anything else that might remedy the problem.
P.S we noticed prior to this that without a significant charge on the onboard batteries the topside adapter doesn't power the robot. Conversely the 5V adapter and mini-usb port can power the motors which suggests that these systems are not isolated which I expect there were intended to be.