Rutgers University just put out a video on a “drone” that can fly and then drop into a body of water, using its propellers to move around. This isn’t the first time we’ve covered a university making sure Skynet can find us even in the bathtub, but this one is a little more manageable for the home experimenter. The robot uses a Y8 motor combination. Each motor pair on its four arms spin in opposite directions, but provide thrust in the same direction. Usually this provides a bit more stability and a lot more redundancy in a drone. In this case we think it helps the robot leave the water and offers a bit more thrust underwater when the props become dramatically less efficient.
We’re excited to see where this direction goes. We can already picture the new and interesting ways one can lose a drone and GoPro forever using this, even with the integral in your toolbox. We’d also like to see if the drone-building community can figure out the new dynamics for this drone and release a library for the less mathematically inclined to play with. Video after the break.
Thanks [Keith O] for the tip!
now what I really want to know is how they made the motors waterproof
Motors are just coils and magnets, and occasionally brushes. The only thing in them that is damaged by water are the bearings, and with proper lubrication practices, this can be avoided.
What about the varnish insulating the wires for the windings? They soften with exposure to moisture. A dunk is one thing and longer term exposure is another.
I grew up in a rain forest and the humidity was Hell on motors. My father was an electrician and would need dry out stored motors before putting them into service. He’d check things with a megger and use a welder as a power supply to warm up the windings and drive off moisture until he had an acceptable insulation resistance reading; the process could take days. If you ran them when they were still damp they would short internally.
Definitely a valid point, but it could be easily accounted for ahead of time by dipping the motor coils in a hydrophobic varnish or epoxy of some sort.
I suspect that if they haven’t figured this out already, they will soon.
These are brushless motors. The rotor is a magnet and the coils around it are potted in epoxy.
Very cool. I have no experience with brushless motors.
Most hobby brushless motors do not have the coils potted and are open to air to help with cooling.
they’re brush-less dc motors so they’re already “waterproof” and this is fresh water so that makes it easier.
Brushless motors are already mostly waterproof. Maybe a bit of sealant on solder connections.
Brushless motors don’t care. They will however seize do to contamination sooner or later. The guys at Flitetest have a waterproofing for electronics video. The other much larger problem is 2.4ghz won’t penetrate any useful depth of water. No modern RC equipment. No video. Rc Sub guys still use the old surface Rc gear.
Good point about RF. I noticed the drone is tethered. That’s probably the only way to give it enough power and communicate.
The motor might have a nonconducting fluid inside but that would reduce the efficiency so I would bet they motor is just sealed well.
The copper windings in the motors have enamel on them.
Here’s a video that might interest you: https://www.youtube.com/watch?v=s4z8QMgTEA4
they didnt, and navy wouldnt care one bit, spare set of motor/rotor combo would be what? couple of $ thousands ($200 times military spending multiplier)? compared to ~10K for ONE navy gun shell and ~100K for one rocket? this is peanuts to the military.
Hmmmm, https://www.youtube.com/watch?v=UaRpTclSwX0
Tom Swift invented a full size version with enclosed blades in 1956. He called it a Diving Seacopter. Later in the 1960’s the design was adopted by Admiral Nelson for use on the Seaview nuclear submarine when they voyaged to the bottom of the sea.
Voyage to the Bottom of the Sea was set in the then distant future of the 1980s! You really start to feel you age when you start outliving the science fiction you enjoyed in your youth.
Most ROVs use the same hobbyking motors that model multicopters and airplanes use, there’s no problem with water but you need to remember to wash them after swimming in saltwater.
The video is so last year though ;)
There’s been many waterproof (floating) multicopter buildes before this. The reason submerging is difficult, apart from the RC signals, is that underwater the propeller will put up a much higher load on the motor and the motors get very hot. I’m not sure how they solved this in this video — normally a ROV will use a 4″-5″ prop where an aircraft would use a 10″ prop on the same motor. Or do they just use the motors for very short periods? This would make sense as you don’t need continuous thrust to keep the vehicle from crashing (with the right overall density)
Well, lets see. If you look carefully in the video , you will see they are in fact submerged in water. If I had to bet on it , I would say that’s a mighty fine cooling mechanism.
they’re using the Y layout to allow idling the center rotors, and only run the perimeter ones; also, these are probably “outrunner” motors, which are pretty open for ventilation in air
Looks more like an X8 configuration. They could alternate between the top and bottom rotors to reduce the duty cycle while underwater.
Now if you were to go 100m to 200m I think that these blade configuration would not work.
Like everyone is saying there would be way to much power being used.
And you cant just let them sink on there own. If there are thinking of the Navy. They just cant sit there for a hour or two waiting for it to sink. They need to go as fast as possible.
And of course we still don’t have good anuff batteries to do the job yet.
I wonder if it’s possible to control this thing underwater using VLF.
Antenna size might be a problem.
Why? You can’t imagine a small multi-rotor pulling a 200 meter long antenna? ;)
For short range missions like underwater bridge survey inspections (as mentioned in the video) it might be best to use hydrophonic communication. The operators could use RF signalling to a floating surface buoy which contains the hydrophone transeiver and relays commands and data back and forth to the rover underwater. When the rover comes out of the water again the buoy turns off, gets picked up by the drone and flown back to the operator over RF as usual.
That would probably give a good compromise of transmission bandwidth / system complexity and power budget.
Close to the surface, RF or optical comms would probably be easiest to communicate and control, but hydroacoustics could be used to control it for longer ranges (RF is usually good up to a few metres, optical to a few hundred).
Is it time to dig out my 300 baud acoustic modem?
“The robot uses a Y8 motor combination. Each motor pair on its four arms”. Then I would call it a X8 motor combination (4 arms). I just checked in the video, it’s an X8
> In this case we think it helps the robot leave the water and offers a bit
> more thrust underwater when the props become dramatically less efficient
On the contrary! The propellors become about 1000 times more efficient as water is about 1000 times heavier than air.
That however does mean that in the air the props spin at a speed where back-EMF sensing works great, but under water the RPM drops to a level where standard Back-EMF sensing no longer works. Thus you may need an expensive motor with hall sensors or an advanced motor controller (FOC, HFI).
i dunno why you would need such precise motor control. you would probably just directly measure the inertial change
Air vs water efficiency may be, but doesn’t need to be a problem. We already know that loons fly under water, and quite well, from what I’m told. It’s a fishing technique that they use. This is just another case of learning from nature and applying it.
A properly tuned loon?
great, so now I gotta register with the FAA AND the Coast Guard? :P
good one.
i loled.
Can I have the circuit diagrams of this drone
I can foresee a lot of snapped props from the water entry stage of this. Unless there is some builtin protection that helps prevent water entry with throttle up effectively dropping it in the water.
One interesting thing to note is that while underwater the motor bodies seem to spin, and the arms are quite thick. It looks almost like they spin the entire motor assembly underwater using a different drive train to get around the fact that the direct drive motors are geared for high speed low friction operation in air…
It was pretty ingenious using opposite direction motors to assure a more stable ride!
That remainds me a RC toy ( http://s.click.aliexpress.com/e/UZjEUz7 ) a mix between a boat, hovercraft and a quadcopter !