My ROV

[kenl]

I've gotten quite far on my ROV project and have received so much help from others on here I will try as best I can to document my build, so that it may hopefully help others.

My story starts with buying drop cameras to use on my boat, the main purpose was to help me to interpret the information that was displaying on my echo sounder screen. In my experience whenever a bottom feature came up on the echo sounder, there would be as many opinions as to what it was as there were people on the boat.

The first Camera, a rotating one was ok, but it leaked and it was hard to get the boat anchored in the exact right spot so that I could view the lump or ledge.

The second camera a bullet camera, just never pointed the right way......

Anyways, I bought some plans for an ROV from ebay, it was a University project and looked like this

IS1.JPG (590.22 KiB) Viewed 8250 times

I got all excited and ordered bildge pump motors and controllers. But as i researched it all a bit more and read just about all of the post 2010 part of this forum, I decided to go with brushless motors. I also really liked the simplicity of the single hull that Biker Bones has documented on this forum.

My new design is now based on the sea lion or sea otter ROV.

download.jpg (7.78 KiB) Viewed 8250 times

My basic hull is a 460mm (17ish ") long piece of water bore casing with an ID of 127mm (5"). The length is 460 due to the limitations of my lathe and I wanted room for expansion (380 would have done) The diameter is 5" because 100mm (4") was not big enough to fit all my electronics and batteries (6x 5000ma 3s packs) confortably, and bore casing because it was the only pipe type that I could find in that diameter, with the added bonus that bore casing is designed to withstand the pressure applied from the outside... apparently

I trimmed the pipe to size and skimmed it's entire length, I wouldn't bother with skimming it next time.

I then rebated the ends so as to reduce the OD by 1mm to provide a shoulder for the flange to sit against.

Using a CNC router I machined some flanges out of 10mm PVC flat sheet and glued them to each end using standard pipe glue and primer, primer everywhere!

hull5.jpg (521.61 KiB) Viewed 8250 times

The ends of the pipe and flanges where then machined flat on the lathe.

The front cap was cut from a sheet of 12.7mm (0.5") polycarbonate, a small shoulder was machined into it so 4mm sat inside the pipe, and a o-ring groove to coincide with pipe surface was also cut.

FW1.png (13.81 KiB) Viewed 8250 times

More to come

[jonnyhas]

Good post Kenl, keep up the good work

[rossrov]

Valuable post Ken. Good information and write-up.

[kenl]

The front window will eventually be a dome, but in the mean time a flat one will do.
I drilled the outside edged with 8 holes after printing a wagon wheel at a 1:1 scale as a template, I could have done this with the CNC, but it would have required a tooling change that I didn't want to deal with.

My CNC is old, very old for a CNC, 1984 old, and it is BIG 6500kg and has 19mm (3/4") collets, so tooling changes down to 3.2mm bits at 18,000rpm are not convenient.

wagonwheel.png (32.19 KiB) Viewed 8158 times

I then cut out the LED mounts for the front, it is just glued to the main pipe. The yellow line is the tool path, for a 12.7mm (1/2") bit.

LEDMOUNT.png (24.9 KiB) Viewed 8158 times

The LED's are set into a aluminium housing that is shown in section below. I have glued the LED to the housing using a heat sink glue, the lens (6mm perspex) presses down on an O-ring and is all held in place with a Circlip. The deeper it goes the more pressure on the O-ring. The light assembly is just held in place by an interference (tight) fit, i may machine the rebate in the light housing deeper and use a circlip to hold it all in place.

LEDHOUSE.png (17.5 KiB) Viewed 8158 times

Here is a photo with the LED's lit up.

hull7.jpg (184.38 KiB) Viewed 8158 times

[a_shorething]

Very cool design work. Clean and very well put together.

Major kudos!

[jonnyhas]

Very neat design and build kenl, one question how are you cutting your circlip groove?

[kenl]

Thanks for the comments guys, it's a lot fun building this, wish I did it years ago.

@jonnyhas I have ground a tool especially for the job, the tool is ground from a broken electric plane blade, which is about 3mm (1/8") thick HSS and say 1/2" wide. For light duty tools that require heavy grinding this stuff is great, any saw sharpening place would have broken thicknesser blades you could have.

CCtool.jpg (271.22 KiB) Viewed 8024 times

The back door of the ROV is made from a solid chunk of aluminium, I have machined it round and put 3 o-ring grooves in it. Two of the o-rings go inside the housing and 1 presses up against the end of the pipe. The reason aluminium is used is because I am concerned about heat buildup from the ESCs.
During some of my Thruster testing I managed to overheat an ESCs causing it to cut out until it cooled down, I don't really want that happening underwater. I have also ordered some small cooling fans for the ESCs.

endcap.jpg (348.42 KiB) Viewed 8024 times

My support slide that has all the electronics mounted on it is also aluminium, the cooling fin so to speak.

internals.jpg (553.52 KiB) Viewed 8024 times

With all the electrical bits and batteries attached to it it looks like the inside of a missile, posting photo's like this could spark a visit from the terrorist police.

[a_shorething]

Hi, we're from the department of military defense and we'd like to talk to you about your "R O V" or whatever you're calling it.

That looks awesome man!

Here's an interesting thought I had: The submarine guys have boat motors that are actually water cooled. I wonder if there is a way we could run some tubing through the motor housings using the outside water to just cool the motors without actually submerging the electrical components.

If they weren't so expensive I'd give it a test.

Alternately you could build a max input voltage limit on your software once you determine the most comfortable operating range. Kind of like a software rev limiter, you know?

[kenl]

The motors themselves are not a problem, I in fact do have them running in a solution of water and soluable oil.

It was the ESC that got hot, I do recall seeing water jackets of some sort for ESCs but would not want to risk a hose blowing off or bursting inside the main housing.

The software already has some limiting on it, I think it needs more, not due to overheating though. I've had a few test dives, and this thing is too fast...

[rossrov]

You certainly don't muck around, Ken - that's good news about the test dives. The thrusters sound like they are capable of moving a much draggier ROV, but better to have too much rather than too little power. I've considered a "turbo" button that selects flat-out speed range, with normal or low speed range being the default. High and Low range is another way to look at it. If you are working in currents then maybe a range of ranges could be the go. If you can do it for 2 speed ranges then more is not a problem, just have to get accustomed to what "gear" you are in. Do not know how the software you are using is written, but assuming it is Arduino should be pretty straightforward to add in an IF statement that checks whether the turbo button is pressed or not and send different numbers to the ESC part of the program. Things can be refined once you've done that.
If you find the ESC cooling an issue, maybe consider inverting the ESCs so that the fins are contacting the aluminium chassis, and use some heatsink compound, thermal paste of whatever if you think it required. Check whether the heatsinks are electrically isolated from the power and motor wires first. If checking for isolation scrape off the anodising as it is not conductive. You may know this stuff already, but lots of others read the forum too. Don't be afraid to ask for help with any coding, I know at least of 2 of us will be keen to jump in