I've been given an assignment at work. Get the fire water storage tanks inspected. I can either hire an inspection certified diver or build an ROV with corporate funding... I think we all know what the right answer is here...
I realize what a wretched starting point this is, but for my first ROV I need more than just a basic observation class model. To do a complete inspection I need video, lights, an arm to poke with a pit gauge and ultrasonic probe, and since my entire area of interest is focused on the bottom and shell of the tank, I'd like it to be more of a magnetically attached crawler than a swimmer. I only need to go 30' to 50' in depth, operating in clean fresh water locally and if upper management takes notice the ROV (and possibly myself) might be shipped around to inspect the fire water tanks in other locations within the corporation. Should that happen, the ROV might find itself in brackish water instead of fresh.
Of course, I'll be starting with just a very simple submerged crawler and build on it in phases until it's meeting all my needs.
I plan on starting with a sea perch framehttp://seaperch.mit.edu/docs/seaperch-build-october2011.pdf
and modifying it with makeblock tracks http://makeblock.com/track-with-track-axle-40-pack
instead of thrusters. I've read a lot of back and forth concerning running brushed motors wet. Those who have actually tried it have found only minor corrosion from fresh water dives if not shaken out after use. If more research has been done on the matter that I may of missed, i'd appreciate a heads up. All the build logs I've found failed to provide more than a few days of follow up. Currently, i'm leaning towards simply replacing the drive motors occasionally since the tank is on a 3 year inspection interval. Though, if I am asked to inspect one of the brackish tanks, i'll probably be regretting that decision.
For lighting, i'm planning on using 1000 lumen 12V cree lights wired up to topside led dimmer switches. I'll need one light placed low in the body to "shadow" out pits and other lights higher up on the body to flood the area.
When I get around to adding the arm I plan on using a couple of HiTec 32646W HS-646WP Water Proof Analog Servo servos since they're IP-67 rated. I want the arm to move like a crane - up/down, left/right, extend and retract. Has anyone tried running a stepper motor wet in freshwater? I'll need the finer movement in the extend/retract motion and they're cheap enough to be counted as disposable so long as they'll work for a day or two. Since it wont have the proprietary gear box that the drive motors have, so i can try sealing one of those up if even freshwater is an insta-fail. It only needs to poke the ground with a pit gauge or press the tip of an ultrasonic probe to the wall/floor. I might be building the arm out of legos when it comes time.
For video I plan on using a simple usb webcam.
It's all high level goals right now. I'm sure there are a ton of finer details and aspects that I'm not even aware of yet. I've been researching motor controllers, servo controllers, H bridges and such. Is there anything i'm missing? Pit falls everyone runs into that'd be worth looking out for?