This may sound crazy.  How about a quadcopter designed for underwater use.  Brushless motors operate perfectly well submerged and the rest can be waterproofed quite easily.  Idea being that one would use small props, basically the quad would operate in a different medium but essentially it should still stabilise as it does in air.  I thought about escs (which don't like water) getting hot but they could be exposed, covered with tectyl or some other waterproofing spray.

As a fail-safe, waterquad would be slightly buoyant so in case of power loss, it would merely float to the surface.  One would then have keep throttle on all the time to stay submerged and power up to sink as opposed to normal operation.  I know that 2.4ghz does not work underwater but the older 35mhz systems do.  For fpv, 1.3ghz could be used ??  Also I wonder which flight controller would work best, kk2 maybe for cost-effectiveness. 

Has this been attempted before and if so, any success?   Couldn't seem to find anything on this concept.  I'm itching to start building but somebody please stop me if I'm wasting my time. 

Views: 12356

Reply to This

Replies to This Discussion

http://openrov.com/ sort of thing you mean? Not a quad, but submerged brushless motors, usually slight negative bouyancy as it uses an umbilical for comms which can double up as emergency retrieval.

It'd be really curious to know if this would work. I'm not sure how much of a quads flight characteristics come from it balancing off against gravity. Inverting the system so it's opposing buoyancy may well achieve the same thing.

I use to work around proper worker and eyeball class ROVs, they all use electric (or hydraulic) thrust. Each thruster is fixed but vectored, so to go forward you drive thruster pointing South West and South East.

Would be fun to try actual rotors in a flat quad configuration though!

Also, 2.4ghz might work if you have the TX antenna under water too?

Hi Anthony, that looks interesting but I'm thinking of a regular quadcopter adapted for water use by adding rc boat props for example.  I may actually just give this a go at some point.

Chris, I've googled this concept but nothing seems to turn up, think I should just build the thing and see what happens.  Regarding tx underwater, one could possibly extend the aerial by detaching it from the tx whilst adding say 3 meters of wire and have the actual antenna floating on a small buoy.  I wonder if that could work.  I'll wait for more feedback maybe somebody knows of the limitations of these flight controllers to the point that it simply would not work.


Most of the ROVs that I have had experience with use ducted thrusters that can provide both forward/reverse and vertical thrust. However most big ROVs use ballast tanks to maintain neutral buoyancy and use the thrusters to move forward/backwards and up/down. Without a ballasting system, your ROV will sink and you will continually have to use the thrusters (propellers) to keep it at a given depth. This will use a lot of battery power.

Just at thought.


TCIII ArduRover2 Developer

I'd like to clarify something Tom has said about ROVs regarding buoyancy and propulsion systems.  I should probably add that I spent 23 years designing and operating Remotely and Autonomous Underwater Vehicles prior to joining 3D Robotics.  I am still actively consulting in the ROV industry.

>>>most big ROVs use ballast tanks to maintain neutral buoyancy 

While most maned vehicles use some form of variable ballast, less than 1% of ROVs use variable ballast systems.  VB systems are heavy and expensive so there has to be some mission specific reason to have one fitted on a vehicle.  It is much more common practice to add or remove ballast weights from dive to dive as equipment is changed on the vehicle.

Some operators prefer to use neutral tether and trim an observation class vehicle to be neutral so they are not using any power to hold a depth, but most work class vehicles are trimmed buoyant so that a) they will float to the surface in the event of a power failure, b) you can use your buoyancy to help pick something up and not sink,  and c) the vertical thrusters are always pushing the vehicle down so the thruster wash is always above the vehicle and not being directed at the ground.  This way when you are operating close to the bottom you can turn off the verticals to ascend rather than having to direct thrust down and causing silt to blow everywhere and ruin your visibility.

One of these days we will make a version of the APM code that will control an ROV.  I just need to write a new motors library to run 4 horizontals and 2 verticals rather than the 6 motors on a hexa. I have a few other things to tackle first.




Points well made and taken. Thanks for the observations and insights.



Quadcopter for underwater use? Well, I don't think it is a bad idea. Be careful with the design of copter orientation system. 

Read more at http://robotics.stackexchange.com/questions/1466/quadcopter-applica...

What I understand from it is that the quadcopter for underwater use is a possible thing. And hope you know the importance of hydraulic lubrication systems. Repairs should be done immediately.

something you may have trouble with in regards to slight buoyancy is the deeper you go (and this is assuming your using air as the medium to obtain buoyancy) is it can get harder to obtain depth as you go further down, kind of like a pool noodle is hard to keep down when you want it down (its a bit like drag, and will require more power). Now I know the following idea may sound more complicated, but you could use a very small can of compressed air (or maybe co2) with a small water proof inflatable sack thats programmed to inflate on command or when fail-safe activates. just a thought, Good luck and I hope to see your results!

I would be curious to know if you have done any progress on this project. I was just thinking about something very similar.

I don't think this sounds crazy at all.  

I have done some testing of 2.4ghz under water and there is no way to get it to travel any where near enough distance to control and underwater quad. In fresh water you may get up to 16cm if everything is perfect but less than 10cm is more likely. The lower frequencies do work thought the antennae get bigger but if you are intending to get video from the quad then the bandwidth on these become a problem.. 

I am looking in to the same thing as this but will tether mine. at present I'm intending to use a fiber optic to command and get data back..

I'm tempted to get hold of a real cheap quad and convert it by swapping the air props for boat props and sticking a raspberry pi on it to handle the fiber optic conversion and camera stuff.  Water proof the whole thing and see how it goes. I would opt for neutral buoyancy if possible if anything a slight negative would be advantageous. 

Did you get anywhere with this  ?  


Reply to Discussion


© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service