I am intrigued with your approach in the creative use of available commercial parts and how to keep down costs.

Rebuilding my own ROV because of the rapid advancement of camera technology, and I keep thinking of better ways to do everything.

Control: Old JR 72 mghz RC radios that had the DSC (direct servo control) jack in the back. Can control many channels through a 2 wire tether using intuitive joysticks and switches. Mine tests good using 300 feet of telephone cable.

Thrust: Ducted fans with in-runner brushless motors encapsulated in latex filled with light mineral oil. One on each side, reversible. Works sort of like a Skid steer tractor. Ascent and descent done with servo controlled diving planes and buoyancy control.

Buoyancy: 4" PVC tube with 3 chambers, Air supplied by scuba pony bottle controlled by servo actuated schrader valves. Bi-directional medical IV pumps move water fore to aft for CG adjust.

Video: Twin Mobius cameras mounted in twin 2" PVC tubes sealed at atmosphere. Streams live FPV to twin 7" HD monitors for 3D viewing. These tubes also contain the ESC's and electronics.

Power: Six high capacity LiPo batteries in separate 3" PVC tube.

Current challenge: Telemetry including leak detection, depth, Buoyancy chamber data, and other stuff.

Hello, It is apparent that you have a lot of knowledge about ROVs. I have built one that worked but the thrusters were not strong enough. It was controlled using a tether and a basic control box. I am a disabled veteran that needs something to do so I have decided to build another ROV. Not the most enlightened dude with electronics which is why I have joined this forum and need some help in understanding what I need to power 3 to 4 DT750 brushless DC outrunner motors by way of a tether. The challenge for me is the control system that will allow the DT750 to reverse and to be as simple as possible. I understand that each thruster requires an ECS in order to run. Any information about what I will need to build a control system would be appreciated.

Originally Posted by leeinMT

I am intrigued with your approach in the creative use of available commercial parts and how to keep down costs....

Thank you leeinMT! I haven't fully wrapped my head around how they do it (the whole tether communications/protocols) but Blue Robotics is using the Ardusub platform (the pixhawk flight controller system) and all that is available there.

Originally Posted by Whisky45

Hello, It is apparent that you have a lot of knowledge about ROVs. ...

Heh thank you, but not sure I would say that myself! I just get obsessed with an idea and start studying. So far your ahead of me with one already built!

If you're trying to get video up too, then you might be better off starting with a controller board like I did. I grabbed the whole developers kit from OpenROV. Otherwise not sure what to tell you except to redo what you already have!

So, what's up? Have you done anything with your box a parts?

I've been researching/ designing my own ROV setup. Probably going to build a small yet complicated version first, about the size of an ice chest, then build the larger one later. Larger being large-refrigerator/ small dumpster sized w/ twin 10-dof manipulator arms.

I don't like OpenRov's plexi tube design, too weak. Fine for the 50' depth you'd Play with it in. I'd rather have an aluminum tube with a clear camera dome on the front end, and potted epoxy pass-through terminals on the back end. With my larger heavier ROV's I want pan And tilt...If i see something move in my peripheral vision, I want to be able to look quickly left-right if i need to/ without having to use the thrusters (draining batteries) to pan the whole mass of the ROV. I found a company that will do Custom acrylic hemispheres for me at an affordable price. 10" hemisphere, 1/2" wall thickness for $140. That should survive to 2000' depth at least. Later if i get hired to do deeper work, i'll upgrade to 1" wall thickness to survive the 1400psi at 3000 ft down.

Gonna build my own tether using stranded Cat5 fished inside 1/2 poly rope.

I like Ardusub's idea of using ethernet extenders to link the top terminal to the ROV via VDSL over a single twisted pair. The pixhawk flight controller is a good idea to manage thrusters to keep the ROV stable and manuver it intuitively, but thier Qgroundcontrol software is too restrictive if i want to say, operate 2 10-dof manipulator arms. Or have dimmable PWM control over LED lighting.

So i'll use the ethernet link to control thier raspberry pi connected via I2C to ssc32 servo controller(s) so i can control lights, arms, active bouyancy, ballast ejection, etc via flowbotics studio software. Using the ethernet connection i can also piggyback on several IP cameras for more camera angles/ coverage. I can even use IP adapters to convert analog/low light/ security cameras to IP ones. Imagine having tiny FXT T80 WDR 15mm square cameras on the gripper/claw of each manipulator arm for a close-up view of what you are grabbing! To be able to reach inside a hole in a ships hull, then switch to the hand-cam and move the wrist to look around inside. https://www.amazon.com/FXT-Camera-CM.../dp/B0776WVCGN And you could install an 8-camera security DVR/multiplexer onboard the ROV and use it's ethernet link to view the camera feeds, while the DVR records the video.

Big dreams/ MAD Skillz/ No Money...lol.

Originally Posted by Oddmar

So, what's up? Have you done anything with your box a parts?

Beyond collecting the bits I need and soldering up what I can, nope... Life, a hurricane, and fixing the house up after, have taken up most of my time lately.

Originally Posted by Oddmar

Big dreams/ MAD Skillz/ No Money...lol.

I hear that, especially the no money lol! As you know (I am sure), I went with the OpenRov control system. It seems to be complete and flexible enough to get me down a ways. I can configure it for different systems and it even has slots I think for additional speed controllers. It "should" be flexible enough to pull off what I want it to do.

I agree with you on the OpenROV tube setup. And in fact, I posted a video in my first post of it failing at only 30m. I will be going with the Blue Robotics enclosures for mine. But first I have a bunch of things to do.

1) Get the main controller system powered up and communicating with the laptop. I may just directly power the BB up first, but I really want to power it up with the tether and everything. This will require figuring out a wiring scheme and get the schematics printed out.

2) Verify the joystick I have can be gotten to work. If not, try to get it to work. Program it.

3) Figure out if I have to use a USB camera, or if I can utilize one of the cameras I use for my drones. Get the camera working.

4) Once I have the base system online, I can then start to scrounge up parts for the thrusters or get complete thruster assemblies.

5) At this point, I should have a good idea what enclosures I need and can get them ordered.

Unfortunately, before I can make any progress on this project, I have to finish fixing my two cars and get both my workspace and workshop back in order. But seeing this post sure makes me want to clear the workbench off right now and dig my box of goodies out...

As you are a Ardupilot for boats Expert, I have a question.
To run a boat with two reversible motors, and the motor settings are using the thrust to steer, please recommend an inexpensive controller, and firmware app?

Sorry I haven't been on here in ages, and I unfortunately can't be too much help in this. I am far from an Ardupiolot "expert" but thank you lol. I have the basics of the ROV platform down, and have some (although now outdated) experience with drones in general. I went with the OpenROV platform to simplify this as it encompasses, solves, and integrates everything so nicely. In fact most of my knowledge base is documented here and in a couple other random threads in this forum where I have posted up a few thoughts.

Unfortunately, the status of the OpenROV project is kinda up in the air, or at least VERY confusing. I am not clear on what happened, but the Trident project (and website) seems to now be owned by a different company. For a while, I couldn't even access the forums. I can now and see a blog post that OpenExplorer is now part of National Geographic. Super cool. But finding related links is another matter.

But most importantly, I also access the old OpenROV store for more parts...

My hobby projects, especially those that cost actual money (and even worse, time), have taken a back seat for a while to house and car repairs. I was moving things around in my "craft area" the other day though and was staring longingly at the three boxes of parts for this project.

Since my original post, I have watched a few of these things come and go. I REALLY wish a company like DJI would tackle the underwater realm and create an affordable ROV that could be mass produced and readily accessible to the world. There are some real options out there for us now, but my magic number is $1k, and most of them start at $1.5 and then quickly go up.

The PowerVision almost fits the bill, but it is just ugly to me. And the design and comments on some of its odd behaviors kinda steers me away. Still, it seems to be the most available one out there. And it is very "DJI-ish" in its design and packaging. I believe it has been the long standing contender too.

There are some FathomOne drones still available it seems, but they have had issues with them failing. And it appears that they were potted for waterproofing. That design I like, but the potting means a failure is going to be hard, if not impossible to fix.

Fortunately, I bought the development package with all the control electronics right off the bat. I can still proceed with building my ROV as I envisioned it. As much as I would love to purchase a ready made option, I just don't think one exists that checks all my boxes, or even most of them. My biggest ones being price, and versatility. I still want to put a decent camera into the water, but I don't want to build the ROV around that. Would rather add it on like I mentioned above using a Go-Pro in a dive case.

Unfortunately, the BlueRobotics stuff isn't inexpensive. The main enclosure I configured to about $275. The BR camera and control bits add another $100 or so. And we haven't included thrusters, potting materials, tools, battery(s) and a bunch of other little bits.

My project here has been stopped by another big issue. 2018 and 2019 summers here the beaches have been unsafe from algae blooms and flesh eating bacteria infestation. Kinda takes the fun out of it. Still this is not a dead project, just badly stalled.

Progress (of a sort) finally! I have been studying this project, and working on it a little here and there for several years now. The stopping point has been the enclosure/shell. Every time I have gone to order parts from Blue Robotics, a decent portion of what I need is out of stock. Right now for instance, it is the acrylic domes. With the orders being fairly large, placing 2-3 orders at $200-300 each and no guarantee I will have every piece I need, is way too risky. And, that $600-900 is just a pile of parts that still needs lots of time to even possibly turn into a working ROV. Too many risks, so I never pulled the trigger again.

This leaves me with a decent pile of parts... I have the OpenROV developer kit. A NICE Pelican case for a "dock"station and (hopefully) ROV transport/storage case. The Joystick. And a bunch of random bits like LCD screens and support gear.

But still no "body" for the ROV.

I decided instead of spending 600-900 or more on more risky parts, I would just save up and get a complete ROV. But the ones I like are either expensive and ugly, really ugly and just mildly less expensive, or almost perfect but unreliable and no longer available. During all of this, the one platform I REALLY LIKED, also went under. Heh...

So I have been watching the market hoping for a new, or old, contender that checks all my points. The OpenROV Trident is easily available now. LOVE to operating system. But the price and COMPLETE LACK of customizability, not so much.

And then maybe 4-6 months ago, a slightly questionable Fathom One went up for sale on eBay. It is my FAVORITE design, looks wise (closely followed by the "tie fighter" Stealth 2 from Shark Marine). This one has some minor corrosion on the wifi buoy and random connection issues. I got a little bit of a discount on it with Best Offer, and finally pulled the trigger on it today.

So I got it cheaply enough that if it doesn't work well out of the box, I have some room to improve on it. I have seen mention somewhere that it was potted, but I can't find that now (other than my post above). Or pictures to prove one way or the other. I guess even if it is potted, I can probably "un"pot it. Thinking about it though, it doesn't make sense it would be FULLY potted inside.

Another concern is the software and platform to actually run it. It had Android and IOS software released for it. I can find the Android software, but I have "no compatable devices." As for IOS, it appears to have had the app pulled. Still if neither of those work, I do have the OpenROV controller system. And in the end, that is what I really want to do anyways.

I am studying up on the Yuneec ST16 for a complete controller package solution. They can be had cheaply enough now to be a serious consideration. It is an android tablet at the core. But it doesn't have the Playstore on it... Info is out there though to root it.

Whatever the case, I have a Fathom One on its way. If it works as is, cool! If not, I am still a lot closer. I can't leave anything alone, so it is doubtful it will be left stock for long, even if it does work correctly.

It is a good looking little ROV for sure! Here are some pictures from the eBay auction.

The tether looks like a mess...

I don't like the wifi buoy type of setup, but as I thought it over I realized it will make custom cable management MUCH easier to accomplish. A simple extension cord reel can be used since it won't need slip rings. If I can get the link reliable, that will do for starters.

I was reading through the Kickstarter manufacturing notes again to try and find the potting references.

The Manufacturer's first Fathom One. It's not totally done, but here's a first look at an assembled Fathom One minus the white front cap (We promise, it's not pretty on the inside once it's been potted).

This is one of the few pictures where I can sorta see inside.

This next one is even better, but no components installed.

I also found this note:

It also meant we waited to install the connector into the bottom of the Fathom One, and with no connector in the drone, there was no way for us to pot it.

From this and carefully direction of everything else they wrote, I was hoping once opened up everything is fairly accessible and that it is just the connection pass through points that are potted. But I just found this reading from the last post backwards...

Potting the drone is just about the final step of assembly and involves pouring a two part hardening foam into the Fathom One’s body. It’s a step that gives the unit incredible structural integrity (enough that you can stand on a drone without issue) but there’s also no going back

I am not sure what kind of foam they used, but it does give me decent hope I can "melt" it out with something like acetone if need be. I suspect it is some sort of open cell buoyancy type foam. But this shouldn't be a project stopping hurdle to overcome.

Alright, I have the Fathom in my hands now and am deep down the rabbit hole of problem solving... I also got a Yuneec ST16 controller with the hopes I can utilize it as a ground control station for this. Basically it is a controller with an android tablet stuffed into it. So it has lots of buttons and two nice (enough) control sticks.

But, this ST16 has to be side loaded (not a big deal there). Though I can only find two versions of the Fathom software, and neither one works on the ST16. It is currently running a locked down version of kitkat4.4.4.

Supposedly, IOS software exists, but I can't access a working download link or bring it up in the app store on any of my various Apple devices...

I have dissected the wireless buoy. It is simply an AP. And the tether to the ROV that is plugged into it, is actually terminated with a network cable plug. SO I suspect I don't even need a working Fathom app, and the app is simply a fancy access point gateway. I am not smart with networking kinda things though and not really sure how to proceed.

Unfortunately, there really are no ROV specific forums out there with lots of traffic. I am registered on the OpenROV and BlueRobotics forums. Honestly they are the only two that seem to have any traffic at all.

I have a SUPER detailed thread over on the Yuneec Pilots Forum. Advice for using/hacking ST16 for ROV This is going in depth on exploring how to use the ST16 controller.

Over on the BlueRobotics I asked Trying to get a Fathom One online, anyone seen this beast out in the wild?

Both of those go into some pretty deep details, and the BlueRobotics thread even supplied a link to the Fathom GitHub Depository with all the factory files including some additional .stl for GoPro holders, some lights, and a ballast holder for saltwater.

For now I have the Android app installed on a cheap Walmart tablet. The access point in the buoy is bad, so I haven't gotten connected to the Fathom yet. The tether is essentially just a network cable, but it is injecting power up to the access point. Not sure I can safetly plug it into my laptop directly. I might try with my Toughbook as I probably have a spare network card if it goes wrong... Or I may try another AP I have on hand. I have a new identical replacement ordered that will screw down back inside the buoy.

I may be able to repair that router board. It has corrosion issues, easy to solve and a lot of the time fixes the problem. I am going to try at least.

I got a really nice extension cord reel. It has a hub/cavity inside with four outlets. If I pull these out, I suspect I will have enough room to stuff a router board inside this and also have a nice tether management system. I will want to find another tether plug though so if I wanted to I could unwind, grab the buoy, and pack it in somewhere.

I ordered some dive filters for a GoPro Hero5 to modify (if needed) for my HiTec Aee action camera. That was before discovering the .stl files for the Session cameras.... I may just get a GoPro for this.

Finally making some REALLY GOOD progress on this!

Here is my little guy! Holy crap are these lights BRIGHT! This is powered on WITHOUT the tether. The tether has two terminals jumped on the ROV side plug to act as the power switch.

It turns out my connection problem, was actually a manufacturing defect. The cable from the topside tether connector to the access point router, was not correctly soldered. The POE wire was just slid in and heat shrunk. Not crimped, not soldered.

Had a MAJOR setback... I didn't realize that the SD card contained the operating system/image for the Fathom's Raspberry Pi. The card was glued in with some of the foam. I eventually managed to get it out moving it back and forth bit by bit, but the card was damaged.I had suspicions there were problems with the Fathom PDB (Power Distribution Board), so I wanted to get it all apart. I finally managed to do so, but with more casualties. The Pi was also destroyed. I wasn't worried there since it is an easy to obtain part. The image for the card though has proven impossible to find, or recover.

So I moved onto Plan B. Converting it to another operating system. But I had another good reason to do that too, what I found when I got the PDB out, was kinda scary. I managed to get it out with minimal damage. A bunch of the connectors pulled out, and the power filter capacitor had also pulled a leg out. Except on closer inspection, they hadn't. I have easily a few thousand hours of soldering together circuit boards. A lot of that was repairing ECUs that were poorly soldered together by beginners. These were hobbyists, not professionals. I have only seen one example soldered worse than this fathom PDB. The parts that are machine soldered or hot flowed, are fine. But all the via hole connectors, WOW. One connector fell off completely. Turns out it hadn't even been soldered. One leg of the capacitor, not soldered. A lot of the other holes were only partially filled, some cold soldered. The one connector that was decently soldered, the ESC power feed, wasn't soldered flush to the board. I found another wire in the topside tether connector, not soldered.

I get it, this was meant to be a $600 drone. And the feel and quality of most of the components are amazing! But the assembly, especially the CRITICAL soldering, not so much. Soldering the PDB boards correctly would have only taken 2-3 minutes a piece TOPS. The first few, maybe 3-5 minutes a piece. This is manualy placing and holding each connector. A simple placement jig could have been made with the materials they were using for assembly, like the foam, that would have sped this up, and made perfect boards with even less effort. I mean even being generous with time, I could easily knock out 30 an hour. Three hours, probably 100. Three hours straight would be about the max tolerable to do in one stint. But they supposedly made only 850 of these. 28hrs total.

The tether and plugs, that is a different story. That isn't so easy. But enough about all that speculation. It is really unfortunate that Fathom failed, and the solution was probably as close as to just charge $100-200 more. I think the Fathom is one of the best looking of the underwater ROV out there. The underlying quality is there, top notch components. Very nicely designed, simple, compact. And it looks GOOD!

Now there is a MASSIVE difference between the Fathom and my next favorite, the BlueROV. The BlueROV looks like a working class ROV. It isn't a "toy" rov. It needs a true topside computer, not a phone or tablet. In that regard, the Fathom might have been in a niche of an already niche market. It is a decent (enough) camera platform, but it can't carry anything more than a GoPro. It can't go super deep. Can't add a grappler. It can never be much more than a simple phone/tablet operated toy. Now if they could have gotten the price down to say $300-400, maybe even up to $500, but fixed the simple poor soldering issues, and added accessories for the rail underneath, it would have quickly taken over the market. It was just a little too expensive for the target market it seems to have been aimed for. For the $600-750 target to work, it would have needed to be more flexible. App control is nice, but the ability for real topside computer control would have been needed. And about $500, it needed to come with at least some sort of reel for the tether.

Whatever, moving on now LOL! Thinking over the above is good though as it makes me clarify how exactly I want to utilize this ROV. As a nice DJI-esque camera, a simple controller/tablet combo is the way to go. I plan on using it a couple steps up from just a smartphone operated toy, I want to take some decent pictures and video. I am not going to go deep diving with it or use it for research. I am more interested in getting an ROV in my hands to learn the platform.

So, Plan B. ArduSub. Kinda sucks that I have a full OpenROV setup sitting here, but the OpenROV PDB is way too big. ArduSub's components may still be a tight fit, but they are considerably smaller. It looks like I can retain the resoldered Fathom PDB, attached to the critical Pi, and also fit the Pixhawk. Fathom did something annoying, but SUPER smart. They filled the drone body with 624/650 6lb density Polyurethane Buoyancy foam. This stuff is what the Coast Guard uses to make buoys out of. Closed cell. What this does here, is it waterproofs and then adds an insane amount of strength to the body. But, it makes getting components out VERY difficult.

My solution to this is to fit an aluminum project box inside before refoaming it.. I cleaned out the old foam, that was a 3 day ordeal. I have the main box in and fitted. I am waiting for a pile of parts to arrive. Including some smaller boxes. I may need to weld on a couple pockets to fit stuff like the ESCs. New foam is on the way too.