[floxy]

Here is an article with a few more details on the construction of the Cyclops vessel:

https://www.compositesworld.com/articles/composite-submersib...

...still not seeing what the carbon fiber is buying here, other than historical reasons / "that's how we do composite pressure vessels". Maybe it adds some slight rigidity when it is being raised up out of the water?

[pge]

From the linked article: "OceanGate CEO Stockton Rush says the company had been evaluating the potential of using a carbon fiber composite hull since 2010, primarily because it permits creation of a pressure vessel that is naturally buoyant and, therefore, would enable OceanGate to forgo the use — and the significant expense — of syntactic foam on its exterior."

[mcpackieh]

My first thought was that expanding foam from the hardware store is very cheap, but obviously that sort of foam would be crushed by the water pressure.

> Syntactic foams are composite materials synthesized by filling a metal, polymer,[1] cementitious or ceramic matrix with hollow spheres called microballoons[2] or cenospheres or non-hollow spheres (e.g. perlite) as aggregates.[3] In this context, "syntactic" means "put together." [...] The compressive properties of syntactic foams, in most cases, strongly depend on the properties of microballoons.

[selimthegrim]

I wonder how similar its properties are to air-entrained concrete.

[jacquesm]

Not very because air-entrained concrete has very little resistance to crumbling and doesn't handle tensile stress well (which you could fix by adding fibers or some other element that handles tension better, hence reinforced concrete). It has plenty of applications under normal atmospheric conditions but for submersibles you will want something with entirely different properties.

[andbberger]

clowns. the pressure hull of a deep-sea submersible is not the place to go looking for cost savings, especially when the target clientele is cost-insensitive billionaires.

[CarVac]

Carbon fiber is used because it is rigid.

The failure mode for a member in compression is buckling, which is resisted by stiffness, not strength.

It's also a lot easier to make a huge thick tube out of composites than out of titanium. That's why only the spherical ends are titanium, and the middle tube part is CF.

[floxy]

>The failure mode for a member in compression is buckling, which is resisted by stiffness

Hmm, this is probably the right answer. When things start to buckle, you'd be putting part of the surface in tension, which would be resisted by the fiber. I would definitely be very interested to see the plots of strain gauges embedded throughout the thickness of the wall as it goes to depth (in all three axes, hoop, radial, and axial). My intuition completely fails here. Good thing I'm not making submersible vehicles.

[CarVac]

Once buckling occurs, it hardly matters what's in compression and what's in tension: the applied force has a tremendous mechanical advantage over the material strength resisting it once the stiffness fails to prevent it from crossing a threshold.

Secondly, CF does substantially improve compressive strength over neat resin, which would fail in shear.

The fibers individually may not withstand compression, but embedding them in the epoxy resin prevents them from buckling and the composite material exhibits substantially improved performance over either base material.

The exception is tensile stress that causes delamination, for which there is no benefit over the neat resin.

[jacquesm]

It would be a good thing if the people who did make this contraption also didn't make submersible vessels based on the bits and pieces that have made it into the news so far. This sounded like an accident waiting to happen. At those pressures if something goes wrong it will crumple like a tin can and having the hatch sealed from the outside means that even if they didn't die at depth they may not have a way out if their comms have failed. The whole thing strikes me as beyond irresponsible. I wonder if the people that built the sub would take it to depth.

[patrec]

In what scenario would not being sealed from the outside make a positive difference? If they managed to surface, how would being able to "get out" in the middle of the ocean be helpful? If they didn't manage to surface, under what circumstances would have opening the hatch helped?

[jacquesm]

Air.

[patrec]

True, although fortunately the air supply is meant to last for 4 days. As a passenger I would probably be more alarmed about the apparent absence of some kind of distress beacon than the inability to open the hatch without help.

[mustacheemperor]

I think part of the issue there is designing a hatch that can be opened on the surface to exchange air without immediately swamping the sub in the sea chop. Alvin sank very fast when it was accidentally dropped into the water with the hatch open.

[actionfromafar]

They could get air.

[Yeroc]

Well it seems like the President and CEO of the company was onboard the craft so I suppose the answer is "yes".

[jacquesm]

I highly doubt he built the sub.

[wnkrshm]

They copied the glass fiber + titanium rings idea from the Deepflight Challenger that Virgin wanted to dive with. [0]

Virgin deemed the design not safe enough for more than a single dive and quietly scrapped the diving events: "Based on testing at high pressure, the DeepFlight Challenger was determined to be suitable only for a single dive, not the repeated uses that had been planned as part of Virgin Oceanic service." [1]

[0] https://www.compositesworld.com/articles/composite-submersib... [1] https://en.wikipedia.org/wiki/DeepFlight_Challenger

[marksmithhfx]

Poor design choice. Too many materials expanding and contracting at different rates leads to friction stress and ultimate material failure. Maybe not at first, but over repeated compression and expansion cycles.

[stickfigure]

I would guess weight? From pictures, the launch platform doesn't look like something that could manage a tube fabricated from 5" thick steel.

[floxy]

It seems like the epoxy in this composite structure is sustaining all of the compressive loads. The carbon fiber would seem to be just along for the ride, so to speak, since fibers are only useful in tension. Am I missing something?

[saltcured]

Wouldn't this be a bit like reinforced concrete? In practice there are not just pure compressive loads, but flexing and shearing that put some areas under tension, so the embedded reinforcements fight that and help keep the bulk material in the right places.

Also, think of the bill of a toucan or fiberglass surf board. A lot of rigidity comes from the tensile strength of the skin wrapped around an enclosed volume of relatively weak stuff that mostly keeps the skin in the right configuration.

[ohazi]

Here "in practice" actually means "in sheets." If you have a flat sheet of concrete / carbon fiber composite supported at both ends, and then you apply a downward force in the middle, the top of the sheet will be in compression while the bottom of the sheet will be under tension.

In a negative pressure vessel that's a cylindrical shell, all of the force is being applied uniformly inward from the outside surface of the cylinder, so every part of every fiber will be in compression. There's no part of the structure that outside pressure would deform in a direction that would stretch the fibers -- you would need to have greater pressure on the inside for that to happen, and in that case, the fibers would again all be in tension.

One failure mode to consider is shear failure (i.e. if the changes in circumference due to compression on the outside vs. the inside of the hull are too different, resin between layers could start to fail.

The other would be micro-buckling of tiny sections of fiber in areas where the resin has already failed, or isn't as strong, or in layup defects.

If either of these cause enough deviation from the nominal structure shape to allow buckling to start, the hull will fail catastrophically.

> Also, think of the bill of a toucan or fiberglass surf board. A lot of rigidity comes from the tensile strength of the skin wrapped around an enclosed volume of relatively weak stuff that mostly keeps the skin in the right configuration.

It is like this, but as another comment [1] pointed out, in the case of a negative pressure vessel, the configuration is unstable -- If you develop just one defect, the entire thing implodes. You might be able to buy yourself a little more headroom with something like pre-stressed / post-tensioned concrete (not sure if there's a practical equivalent for carbon composites), but this is a pretty extreme negative pressure vessel...

[1] https://news.ycombinator.com/item?id=36392856

[interner_reader]

>Also, think of the bill of a toucan or fiberglass surf board. A lot of rigidity comes from the tensile strength of the skin wrapped around an enclosed volume of relatively weak stuff that mostly keeps the skin in the right configuration.

So in other words carbon fibers are a bit elongated (or straindd) in CFRP? What increases shear strength of the composite but could weaken its compressive strength?

[pomian]

Interesting. There must be interior bulkheads(walls), your idea of shear could very well happen where the bulkheads prevent tension of the inside surface.

[marksmithhfx]

This makes me think, why didn’t they build in composite reinforced bulkheads every 12 inches or so. That definitely would have had substantial impact on compressing the cf hull and preventing delamination. So obvious.

[marksmithhfx]

Like ribs in an airplane wing, which provides the bulk of compression strength and transfers loads to spars.