Hacker News Reader

Truly bizarre. Failures like this happen in milliseconds, it doesn't matter how realtime your realtime monitoring is.

I'm not arguing in favor of the classification system, but, I think if you went to them and framed things as simply as they are here "you require validation every 3 years, we're doing it on every single dive" they would say "that's more than once per 3 years, APPROVED!" so the situation is clearly far more complicated than they are indicating.

Yes, I don't understand this acoustic monitoring concept... I'm curious what the logic behind it was. It just doesn't seem like there would be enough time to take action once the monitor starts to detect failure.

It depends on the material. It might work for certain tours of steel, since the failure state is gradual.

Carbon fiber, however, is not known to gradually fail.

It seems they felt this was one of the novel innovations that enabled this design - that the monitoring system was so good, they'd never be at risk in the pressure hull at depth.

James Cameron said it looks like they had dropped their emergency ascent ballast before the implosion, so they may have gotten enough warning to take action. But I would guess that system was never tested to failure in real life.

It will be interesting to see which of the novel innovations supposedly enabling this unconventional design failed, if it can be determined. Rush also lampshaded the fact that his company had "successfully" bonded the titanium endcaps to carbon fiber despite that being contradictory to conventional materials science wisdom, reading the patent for the monitoring system it looks like it may have been intended to provide warning about those bonds failing too.

It seems like it would be trivial to test how much early warning such a system gives by bringing smaller/cheaper test hulls to failure, either in the ocean, or in a pressure testing chamber. I wonder if they did that?

Do you know in what way it was 'acoustic'? I.e. was it listening for sounds, or was it more like sonar?

I looked over the patent yesterday, I think someone linked it in this thread. It would “listen” for the sounds of impending structural failure, by comparing the sounds in the hull to a dataset of previous test dives. One of the passengers on Titan’s 2nd dive said it made cracking and popping noises as it got deeper but the patent intended to detect a change in acoustics (in conjunction with stress sensors) far enough in advance for corrective action.

I think the confusion about tensile strength comes from imagining a single carbon fibre. It very obviously has little strength in compression, like a bicycle spoke. But the fibres don't act alone.

Even under tension, you need the epoxy to transmit loads between fibres, or else the carbon fibre is pretty useless which is why the finished material CFRP (carbon fibre reinforced plastic) is used. Unlike metals which are generally equal strength in compression and tension, CFRP is roughly half as strong in compression. But it's very light so from a strength point of view it's not automatically the worst material choice. I get Cameron's point but if it really had "no strength in external compression" then it would be a useless material in general.

Regardless of material the main failure mode of a tube with external pressure is buckling, so the strength in compression is much lower than what you expect from a hand calculation that doesn't consider buckling anyway[1]. But CFRP is a risky choice for at least 4 reasons: analysis is not as straight-forward/easy-to-trust, signs of fatigue are harder to detect, delamination is difficult to control[2], and interface to other materials is tricky, especially when temperature changes and water are involved.

[1] You can use hand calculations to check buckling as well but they depend on analytical models which, if they even exist for CFRP, would be quite difficult to trust because there are so many more variables compared to metals.

[2] Delamination must be really nasty when coupled with buckling. Local weaknesses greatly affect buckling, and a even a small void between plys is a local weakness where the plys don't transmit load to each other.

When I heard Cameron speaking about this recently, I thought he was referring to two different materials being used, rather than the carbon fibre per se.

Thanks! Very informative reply, greatly appreciated.

The epoxy in carbon fiber makes it stiff. Stiffness helps reduce buckling loads.

The buckling load is the load at which a structure buckles. I think you mean that stiffness helps resist buckling.

Err yeah I mean that stiffness reduces the internal stress caused by external compression force. Stiffness increases the buckling load of a material.

Honestly, the "real time health monitoring" felt like almost every tech "innovation" over the last decade or so - a half-baked excuse not to do things correctly. And when you can't switch your lights on because the cloud is down, it's inconvenient, but clearly the results of applying this thinking to critical safety are self-evident.

I too am perplexed. Perhaps the compressive strength is all derived from the epoxy not from the carbon fiber.

Yes, this doesn't make sense to me either. I suppose it could help prevent bucking and crack propagation, because as the walls of the sub started to bend, one side would have to suddenly be in tension? Still, I wonder if a pure plastic/epoxy hull would have actually been stronger than composite in this situation.

From the design they published of it, I do really think they intended to have the carbon fiber tensioned between a pair of titanium rings.

I understand epoxy doesn't deal well with that, but with the right plastic, I believe it could work very well.

It is a carbon fiber tube with a hemispherical titanium cap on each end. The caps push inward under pressure putting the carbon fiber hull under compression, not tension.

That would only be possible if the rings were fixed relative to each other, which they are not.

Yeah, you are right. It's not tensioned between the rings then.

The craft survived several launches - that proved it can do it.

Whether it can do it reliably / survive fatigue accross repeated cycles is a different question. Imagine we'll find out as the investigation progresses.

We already found out, I think.

Maybe some orcas played with it.

> The craft survived several launches - that proved it can do it.

Exactly, that's my point. I didn't realize that (this present catastrophe notwithstanding) that carbon fiber could even begin to have such compressive strength. How is the fiber formed in such a way to maintain that strength?

align layers of fibers in different directions and you can get strength in different directions. But it is tricky! The glue holding the metal end caps to the carbon tube is also a possible failure point. Imagine a metal ring slid over a carbon tube held on with glue. If pressure deforms the metal differently than the carbon, you get a gap at the interface and BOOM

or if sea water degrades the glue, or if the carbon fiber degrades after each trip and eventually isn't strong enough any more, etc.

A point raised by James Cameron (and others) was the titanium bands mated to the carbon fiber hull. Apparently in the material science community, that's a huge no-no, and a well known one to boot. Galvanic corrosion or something like that...

It’s a dumb idea for numerous reasons. The titanium and carbon fibre will expand and contract in relation to temperature and pressure at different rates, they are two different materials with totally different material properties. So you’ve added a second force on the vehicle which is the pulling and pushing of the carbon fiber and titanium from one another. And all you have to hold that together is fancy glue, the glue is taking all that force. A 5 year old could do better.

It’s counter-intuitive, but the so-called hoop stress on a cylinder being compressed from the outside actually results in tensile stress not compressive. This is because the material as it resists shrinking to a smaller radius is being pulled apart.

Is this accurate? All of the examples I found online about hoop stress had diagrams where the pressure inside the cylinder was higher (as in a water pipe for example).

Tensile strength would make sense in that scenario, but I don't see how it would apply when the cylinder is empty and being crushed by outside forces.

James Cameron gave an interview where he claimed the carbon fiber hull concept was tested by a team competing to get down to Challenger Deep at the same time his team was - and the concept failed in testing. He also claimed many within the deep submersible community wrote letters to OceanGate warning them their plan was flawed and would result in catastrophic loss.

Specifically, the failure mode was delamination.

I can't vouch for how credible this all is - but, given how much disregard the OceanGate team had for safety, it doesn't seem too far fetched.

https://www.youtube.com/watch?v=e9YB31ElEFQ

That was a great interview IMO, thanks for posting. I think some people might find it hard to both convey sympathy and compassion for the deceased and their families as well as being adamant that the submersible design was nuts (especially since the man primarily responsible died), but I think Cameron in particular did that very well.

Challenger Deep is exponentially more difficult than this - 10km Vs 4km so over double the pressure.

While it may still be an entirely unsuitable material for 4km, not being suitable for 10km doesn't prove that.

As far as I'm aware, and please correct me if I'm wrong, they had never dove to this depth before.

So, sure, it was launched before and recovered, but it never experienced the same conditions which caused the failure.

They did: http://web.archive.org/web/20200814020351/https://www.oceang...

(At least twice, and probably more times).

Apparently it's at least 13 dives since 2021: https://www.geekwire.com/2022/beyond-the-titanic-oceangates-...

I saw someone on Twitter suggest that this was probably WHY it failed this time. Each dive would weaken it.

i think we found out when it imploded

>Whether it can do it reliably / survive fatigue accross repeated cycles is a different question. Imagine we'll find out as the investigation progresses.

Seems like we already found out the answer is “no”. The overall ship design failed.

The ship failed. What we don't know yet is whether the carbon fiber hull is why.

All evidence so far points at implosion, which is kinda by definition a hull failure.

const int well_before_in_milliseconds = 50;