I also saw a news story quoting the CEO to the effect that he was breaking established norms but with sound (supposedly) engineering behind it. Apparently referring to the hull design of carbon fiber shell mated to titanium end-caps for the pressure hull.
From personal experiences at work I've seen it is very difficult to bond composite to metal like that and have it leak-free under high pressure.
About 10 years ago we tested wrapped carbon fiber pressure bottles that had bonded-in threaded bosses on the top and bottom for tubing connection points. We tested it with liquid nitrogen (LN2) and pressurized each one till it burst.
The bottles leaked like crazy around the metal bosses. One or two of them we couldn't even pressurize high enough to burst it because it was leaking too much!
So yeah, submarine built with the same metal-to-composite interface?? NO THANKS NOT ME.
How about how they were spreading the epoxy onto the ends of the tube and grooves in the metal rings by hand? If the wreckage can be recovered, I bet one point of failure will be that epoxy was too rigid and it cracked.
What I'd choose for gluing a join like that would be a high durometer urethane rubber. Stiff enough to not deform under the weight of the end caps and rings yet flexible enough to handle the differing amounts of compression during dives. Smooth-On makes all kinds of urethane rubbers, resins, and foams, and has a urethane bonding surface prep which will make them stick permanently to just about any surface.
So Ure-Bond II prep the surfaces, apply the liquid rubber with a special caulking gun that mixes the two parts at point of use with a static mixing nozzle, slide the ring into place, clean up urethane squeeze out. You're looking for squeeze out all around and for it to be uniform. That would indicate the ring is all the way in place and not crooked. I'd make up a sheet of the urethane material to the desired joint thickness and tack them onto the joint face with freshly mixed urethane. Those pieces would only need to be like 1/4" diameter x the joint thickness and being the same material the new urethane would bond to them perfectly.
The part of the procedure that would guarantee a flawless seal would be a special inner and outer containment for the join to seal to the titanium ring and the carbon fiber tube. First pull a vacuum then apply pressure. That would evacuate air from the seal then ensure any bubbles left are squished to nothing so there's no flaws to work back and forth from dive pressure cycles.
Most submersibles have a plug hatch which is very thick, with several seals. They're designed to be latched and unlatched from the inside or outside. At the surface they close just well enough to be water tight. As the submersible dives, the increasing water pressure pushed on the hatch and makes the seal tighter.
They're designed that way so that as soon as the submersible surfaces the crew can pop the hatch open for fresh air, unless the waves are too high. I assume most submersibles also have a fresh air vent that can be opened on the surface in such conditions.
Reynolds Metals Aluminaut was originally designed without an elevated tower with the hatch. During an early test with the hatch open it got swamped and sank. (This was the early 1960's when these things were still new, and there were no "50 year old white guys" with long experience in submersible design.) So they added one. For the next test, Aluminaut was top heavy and rolled upside down.
Aluminaut has been sitting in a Virginia museum for 53 years, waiting for someone with the $$$ it would require to check out, refurbish, and update to put it back in the water. With a safe working depth up to 15,000 feet, it could easily visit Titanic, with up to 7 people and a payload capacity of 6,000 pounds. I don't know if the people are included in the payload weight.
If nobody wants to do a This Old Submersible with Aluminaut, copy it. I'm sure there are newer aluminum alloys that could meet its performance with a wall thickness less than 6.5" and at less than its 80,000 pounds, or exceed its performance at the same weight.
www.foxbusiness.com
