Seeing as adding payloads and picking up objects from the ocean are going to change the ROV's buoyancy, it seems like it would be a good idea to start exploring ways to control the buoyancy on the fly. I was thinking of ways to use a CO2 bicycle tire cartridge as an air source.
To regulate pressure from the cartridge, maybe we could use an off-the-shelf compact tire reg like this one from Bontrager for $20, replacing the hand valve knob with a geared connection to a servo.
At first I was considering an inflatable bladder, but a rigid ballast tank might be even easier and more effective. The layout might be something like this:
Similar to how manned submarines work, the pressure from the air (CO2 in this case) would push water out of the bottom openings when you want to go up. Going down would involve opening the top vents to allow water pressure to push the air out. (Actually you would probably want the tanks connected somehow with only 1 top valve, as opposed to the way drawn, for the sake of simplicity).
You can get CO2 cartridges in 16 and 20 gram sizes (and probably others as well), using the ideal gas law (pV = nRT) we can calculate the volume of CO2 that can be produced from a cartridge at depth. Let's say we're using a 20g cartridge:
molar mass CO2 = (12 + 2*16) = 44 g/mol
n = 20/44 = 0.454545.... mol
R = 8.314 J/(K*mol)
ambient water temp T = 10 degC = 283.15 K
pressure @ 100ft P = 4 bar = 400,000 Pa
V = (0.454545)*(8.314)*(283.15)/(400,000) [m^3] = 0.002675 m^3 =~ 2700 cm^3
so ideally we could displace 2700 cm^3 of water with CO2 (there will be losses of course so we probably want to come up with an efficiency coefficient). That's 2700 grams minus the 20g of CO2, meaning the ROV effective mass could be reduced by 2680g during operation (maybe call this delta-M). Lemme know if any of this seems off...
From here we'd need to know the current displacement and mass of the ROV to try and come up with an ideal ballast tank size, anyone want to help with that?