I am working on a couple of issues so cannot film or weigh in water which is one of the issues. I know it is slightly heavier than neutral buoyancy.
If anyone has a 3-D printer and would like to collaborate I can send my drawings.
You MUST have a lathe/mill combo at minimum to build one of these. I now see a 3D printer can make many parts and simplify the construction.
The end-cap design solved many of my problems of making the sections waterproof while having easy access for maintenance. My original design took over an hour to take apart and re-assemble, the new design comes apart in 30 seconds and does not require any calibration to re-assemble in less than a minute.
I use dual shaft motors from Tsiny (Ebay) but they are heavy. I can increase buoyancy by making the tubes longer or wider but this makes them heavier and there is very little gained. There is a limit to the metal I can remove. The big issue is the motor weight and since I only need the dual shaft motor at the base, and only to lift the arm in air, the solution is to replace the other motors with lighter ones which I have not been able to find in dual shaft. My target is to have an arm robust enough to work in air so it does not self-destruct in testing. It does not have to lift anything heavy in water because the ROV will simply go the other way. The thrusters have limited ability to hold the ROV on the bottom. The solution to lifting, say an outboard motor, is to take a cable down from the surface and hook it on. My gripper is $20 from Ebay and there were two problems. It was designed for a servo drive which has numerous problems. I solved this with a right-angle gear driven by a gear-head motor making it smaller, easier to seal, and easy to add grip strength setting. The metal grips do not hold hard objects well and the solution is to slip surgical tubing over the ends.
My interest in 3D printing is to make a single shaft version end-cap to solve my buoyancy problem.
In my ROV I am also replacing the compass and pitch and roll clinometers with a single I2C module (working), replacing the wire-wrap boards with PCBs (made), and trying to get the Bar30 sensor software working (see my other post).
I am not sure if I have hi-jacked someone else’s thread with this post and apologize if I have. Posting on Forums is not my strong point. I hope my Bar30 post is on a new thread.