6 ms·

Building a radio-controlled submarine with automatic depth control

152 points by mik3y 4y ago

robotguy 4y ago
As an EE for an ROV manufacturer for 20 years, this was really interesting to see how he attacked some of the design issues that plagued my entire career. Great stuff!
BugsJustFindMe 4y ago
It doesn't matter much if you only ever want to poke around in a swimming pool or very shallow stream like shown here, but if you get into this hobby and ever want to run in deeper water without a tether (whether autonomously or acoustically) you'll either want it to be positively buoyant or have an independently powered emergency surfacing mechanism (eg. drop a weight). Otherwise if you lose power or control you will lose your sub.
- blacksmith_tb 4y ago
  Seems like it's pretty common to use a tether (like the OpenROV) for yanking it back out in case of failures also - I assume any self-contained failsafe like dropping a weight could itself fail?
  - mbreese 4y ago
    Yes, any failsafe could fail, but there are ways to reduce that risk. For example, if you use an “default safe” system where you have to actively keep a weight from falling, then when you lose signal or lose power, the weight drops, you know that in the “default” state, you’re safe. And example would be if you held a weight in place with an electro magnet. As soon as you lose power, the weight would drop, and the vessel would return to the surface. This is the same idea behind semi truck brakes. The brakes are by default engaged. When the truck is running, there is a build up of pressure to disengage the brakes so that the truck can drive. If something goes wrong (loss of power on the truck), the brakes are engaged by default to stop the truck.
    - magicalhippo 4y ago
      This was basically the first lesson in automation class, which made an impact at the time. A safety switch should be normally closed, energizing a normally open relay, such that any broken cables or connections would be like hitting the switch, causing the relay to de-energize, subsequently cutting power to the main circuit.
    - jacquesm 4y ago
      Same in industrial control systems that are used to control machinery where a failure could lead to injury or loss of life (or property). The default is to stop the machine in a state where it is known to be safe, this can lead to some pretty difficult problems if you want to get out of that state to re-start the machinery but the alternative is to continue to move when you really shouldn't be. Even something as simple as an emergency stop switch and the subsequent re-energizing of the system can be a very hard to solve problem from a control perspective. Especially with Servo systems with incremental rather than absolute encoders.
    - mcswell 4y ago
      Default engaged brakes were patented by George Westinghouse in 1869. The use was of course on trains.
  - dylan604 4y ago
    Common should probably give way to expected. The Navy uses tehtered control on its torpedos for the utmost in stealthy attacks. If the best the Navy and military budgets can come up with is tethered control, not likely something in the hobby world will be better.
- changoplatanero 4y ago
  How well does radio transmission work under water?
  - eimrine 4y ago
    Typically 3-4m, 7m peak depth.
  - yardie 4y ago
    Doesn’t work on any unlicensed frequencies. 2.4Ghz and 5Ghz are absorbed by water. Militeary submarines use VLF frequencies. With a mile long antenna. So tether is the only real practical method.
    - jacquesm 4y ago
      No idea why this got downvoted, it is 100% accurate.
      - BugsJustFindMe 4y ago
        It's not 100% accurate to say that tether is the only practical method. Untethered AUV/UUVs are quite common in the industry.
        yardie 4y ago
        Please point them out. All the UUVs/AUVs I've found are either tethered to a buoy or completely autonomous and only transmit when the surface. In all practicality there is no completely untethered ROV on the market. Acoustic controlled UUVs are still line of sight.
    - vbezhenar 4y ago
      What about sound?
      - yardie 4y ago
        They operate at 100s of bits per second not fast enough to transmit live video so your back to line of sight.
rektide 4y ago
Magnetically coupling the proph-drive through the hull was such a neat trick! Really nice build period.
- pxmpxm 4y ago
  As was the full PID setup for the depth control.
- dylan604 4y ago
  Yeah, I always liked that feature from the first builds when I saw them too. Very elegant solution
  - jacquesm 4y ago
    It serves as a clutch of sorts as well in case the prop ever gets jammed.
megraf 4y ago
I've watched the BrickExperiments YouTube channel, but I always assumed that the channel was drivel, and that much like social media of today, the details of the build would be omitted, and lost forever- but this changes the game for me. A 10 part post is beyond what I'd consider most 'good' creators to put together. The author is obviously technical, and is wise to the nature of the videos. I'm impressed.
rpaddock 4y ago
Some might find the system by James H. Rogers, developed in the era of World-War-One of interest. The late T.E.Bearden claimed that Rogers system was rediscovered and 'lost' at least seven times from then until our times. "James H. ROGERS Underground & Underwater Radio ( Static-free Reception & Transmission Underwater & Underground )" http://www.rexresearch.com/rogers/1rogers.htm http://www.rexresearch.com/rogers/1rogers.htm

152 points by mik3y 4y ago

As an EE for an ROV manufacturer for 20 years, this was really interesting to see how he attacked some of the design issues that plagued my entire career. Great stuff!

BugsJustFindMe 4y ago

It doesn't matter much if you only ever want to poke around in a swimming pool or very shallow stream like shown here, but if you get into this hobby and ever want to run in deeper water without a tether (whether autonomously or acoustically) you'll either want it to be positively buoyant or have an independently powered emergency surfacing mechanism (eg. drop a weight). Otherwise if you lose power or control you will lose your sub.

blacksmith_tb 4y ago
Seems like it's pretty common to use a tether (like the OpenROV) for yanking it back out in case of failures also - I assume any self-contained failsafe like dropping a weight could itself fail?
- mbreese 4y ago
  Yes, any failsafe could fail, but there are ways to reduce that risk. For example, if you use an “default safe” system where you have to actively keep a weight from falling, then when you lose signal or lose power, the weight drops, you know that in the “default” state, you’re safe. And example would be if you held a weight in place with an electro magnet. As soon as you lose power, the weight would drop, and the vessel would return to the surface. This is the same idea behind semi truck brakes. The brakes are by default engaged. When the truck is running, there is a build up of pressure to disengage the brakes so that the truck can drive. If something goes wrong (loss of power on the truck), the brakes are engaged by default to stop the truck.
  - magicalhippo 4y ago
    This was basically the first lesson in automation class, which made an impact at the time. A safety switch should be normally closed, energizing a normally open relay, such that any broken cables or connections would be like hitting the switch, causing the relay to de-energize, subsequently cutting power to the main circuit.
  - jacquesm 4y ago
    Same in industrial control systems that are used to control machinery where a failure could lead to injury or loss of life (or property). The default is to stop the machine in a state where it is known to be safe, this can lead to some pretty difficult problems if you want to get out of that state to re-start the machinery but the alternative is to continue to move when you really shouldn't be. Even something as simple as an emergency stop switch and the subsequent re-energizing of the system can be a very hard to solve problem from a control perspective. Especially with Servo systems with incremental rather than absolute encoders.
  - mcswell 4y ago
    Default engaged brakes were patented by George Westinghouse in 1869. The use was of course on trains.
- dylan604 4y ago
  Common should probably give way to expected. The Navy uses tehtered control on its torpedos for the utmost in stealthy attacks. If the best the Navy and military budgets can come up with is tethered control, not likely something in the hobby world will be better.
changoplatanero 4y ago
How well does radio transmission work under water?
- eimrine 4y ago
  Typically 3-4m, 7m peak depth.
- yardie 4y ago
  Doesn’t work on any unlicensed frequencies. 2.4Ghz and 5Ghz are absorbed by water. Militeary submarines use VLF frequencies. With a mile long antenna. So tether is the only real practical method.
  - jacquesm 4y ago
    No idea why this got downvoted, it is 100% accurate.
    - BugsJustFindMe 4y ago
      It's not 100% accurate to say that tether is the only practical method. Untethered AUV/UUVs are quite common in the industry.
      - yardie 4y ago
        Please point them out. All the UUVs/AUVs I've found are either tethered to a buoy or completely autonomous and only transmit when the surface. In all practicality there is no completely untethered ROV on the market. Acoustic controlled UUVs are still line of sight.
  - vbezhenar 4y ago
    What about sound?
    - yardie 4y ago
      They operate at 100s of bits per second not fast enough to transmit live video so your back to line of sight.

rektide 4y ago

Magnetically coupling the proph-drive through the hull was such a neat trick! Really nice build period.

pxmpxm 4y ago
As was the full PID setup for the depth control.
dylan604 4y ago
Yeah, I always liked that feature from the first builds when I saw them too. Very elegant solution
- jacquesm 4y ago
  It serves as a clutch of sorts as well in case the prop ever gets jammed.

megraf 4y ago

I've watched the BrickExperiments YouTube channel, but I always assumed that the channel was drivel, and that much like social media of today, the details of the build would be omitted, and lost forever- but this changes the game for me. A 10 part post is beyond what I'd consider most 'good' creators to put together. The author is obviously technical, and is wise to the nature of the videos. I'm impressed.

rpaddock 4y ago

Some might find the system by James H. Rogers, developed in the era of World-War-One of interest. The late T.E.Bearden claimed that Rogers system was rediscovered and 'lost' at least seven times from then until our times. "James H. ROGERS Underground & Underwater Radio ( Static-free Reception & Transmission Underwater & Underground )" http://www.rexresearch.com/rogers/1rogers.htm http://www.rexresearch.com/rogers/1rogers.htm