Engineering assessment of the OceanGate Titan failure
captainsjournal.com
captainsjournal.com
Some things I came across, which are in line with the article and the first spurce video:
- The view port was not rated for the intended operating depth, if true I am out of words...
- the pre-preg material was apparently beyond shelf life, if true almost as bad as the view port. When pre-preg is beyong shelf life, the resign doesn't reliable bond anymore. That means basically delaminated layers before the "vessel" saw water for the first time
- 480(!) layers of carbon fibre, that carries the risk of prematurw dryong of the resin during production. Impact, see above.
- not curing and no auto-clav, again with the same results as above
- The mating rings for the emd caps have been glued on. Nothing wrong with that per se, but a bond supposed to sustain repeated dives to 4,000 m needs to be very precise. Ocean Gate applied the glue manually by hand, with inconsistent thickness in a non-clean enviroment. No idea to which tolerances the CFC hull and rings have been machined, but based the video of the bonding (they used chains attached to gate crane) they didn't seem to be too precise. The results are a high potential of foreign objects and debris being glued in, non-defined wiggle room between the CFV hull and the titanium rings and end caps and yet another strucural weakness of the vessel.
- they also used titanium fibre. If trued, it doesn't matter because the preasure was applied from the outside, all the tensile streangth of carbon and titano fibers only worls against preasure from inside the vessel.
- No back-ups, exposed wiring at the outside, flamable material at the inside, no telemetry, no emergency beacon, only wireless (!) controls (and they didn't even do enough testing to find key mapping errors of their Logitech controller...)
- Despite claiming otherwise on their homepage (archive.org still has it) they did attach equipmwnt directly to the carbon hull on the inside by screwing (!) to the hull. Result, all the fibers cut by the screws don't take any load whatsoever anymore, increased risk of delaminations (the hull is compressed, the screw on the inside isn't)
- Safety culture: completely absent, to the point of being proud about it. Unable to distinguosh between vessel certification and proper operations (there is a reason that aerospace certifies the design, production, maintenance and operations organisation seperately), firing people raising concerns, I am out of words here..
- The vessel can only be opened from the outside (!), so even a perfect dive can lead to suffocation is the vessel isn't located fast enough...
I could go on, but calling this abomination "engineering" is an insult to the universe.
Finally, some potential failure modes:
- the view port (obviously)
- the hull (delamination, stress, bad material, intentionally induced defects...)
- the bond between the rings for the end caps and the hull (weakened bond due to different deformation of the two materials, weak bond due to foreign objects, inconsistens bond due to bad application of the glue...)
All in all, from design over manufacturing to maintenance and operations, this thing was a death tral whos only upside was a quick death in case of an implosion.
This one is actually the most mind-blowing to me if true.
That and the fact that it actually made it to Titanic a few times and back without catastrophic failure.
I pitty the other four passengers, Rush got what asked for, and rightly so.
There is nothing gleeful in any of my comments so, quite the opposite, I am really sad. Amd getting the facts straight requires a proper investigation, one that is happening. What I did was a superficial "FMEA" based on whatever, incomplete, information was publicly available at the time. Some is wrong, some is missing and some comes straight from OceanGate / Rush himself.
i've no idea what i'm talking about but my guess would be that they were sort of aware of the design shortcomings and then just tried to compensate by overengineering the static components. and then they just pretended the material wouldn't progressively deteriorate towards the inevitable demise.
I've seen this claim repeated over and over with this single photo [0] with a display mount circled.
It was immediately obvious to me that the pattern of lights on the roof in that photo isn't coming through the 5" hull, but that the monitor is instead screwed into a liner of some sort. Additional evidence [1] seems to agree.
[0]: https://twitter.com/williamlegate/status/1672335515004162050
[1]: https://twitter.com/ohshutupskeeees/status/16723781759707504...
Edit: My Pixel 2 trained muscle memory is just so slightly off at the new Pixel 7, you should see initial versions of my posts lately...
The crazy part is that I'm certain detection and text correction weren't always this bad.
I support bringing it up when it's frequent enough to cause frustration. Putting some effort once into the writing - typing correction occurs almost effortlessly on almost any device these days - benefits every reader.
Wow. What kind of hubris could look past that huge red flag?
Apparently the kind that thinks it is interesting that it deforms by several inches and not scary
It’s all relative. Good engineering doesn’t mean higher number.
No one knows what the FOS for this sub was. Given the information from the CEO, things like that were just slowing down innovation!
EDIT: Not all rockets, just using an example.
The "bad-boy" disruptive innovator kind that doesn't play by the rules and that's worshiped in SV because he makes his wealth playing fast and loose around old stuffy bureaucratic regulations made by "evil old men" to gatekeep innovators like him from entering the naval market held by centuries year old corporations who actually know what they're doing.
The problem for him was that, unlike consumer tech, the laws of physics can't be disrupted. He'd tell you that himself if he'd still be alive. Shame that four other people had to perish to prove him wrong.
No need to generalize. (In fact as I understood it he haunted the Pacific Northwest.)
I hope it changes, bacause as we have seen it kills people, litterally. Titan was actually worse than Titanic, because the engineering of Titanic was actually sound for the time, saftey procedures and seamanship were not. The latter changed significantly after the Titanic, lets hope that this SV stained engineering culture will, too, after Titan.
underground traffic jam
Count me out, I'm telecommuting.
Perhaps a lawyer for a plaintiff with deep pockets will call for such a demonstration to be performed.
The Titan made it a few times.
The kind that would scrap ASICs for a Walmart PC using 100% Python because "those damned quant geeks don't know REAL innovation".
I don't think it was outright hubris. I think it was commercial necessity, driven by the business model that OceanGate used: Rush needed rich tourists to fund the business, and he got rich tourists by offering them a close-up view of a famous shipwreck. To give his passengers a view that they were willing to pay for, he needed an oversized viewport.
He had already proven this business model with his earlier submarines, so it is understandable that he wanted to carry it on as he built vehicles intended to go deeper. Less understandable is that he set aside the reality that a large-enough viewport could not actually be safely procured. That's not really hubris though, just plain old profit-driven idiocy. The whole thing feels like it has less in common with the Titanic and more in common with the Ford Pinto.
The same kind that is proud of saving money by buying carbon fiber for the craft at a significant discount from Boeing because it is surplus that is too old to be considered safe for aviation use, says “safety is pure waste”, etc.
The video was absolutely right to mention lack of dust-free environment (vs open warehouse), lack of temperature control, and lack of degassing equipment. We also have no idea about the properties of the epoxy used to bond it, whether it has the suitable amount of compliance, properties under temperature change, etc.
Worse yet, I saw no precautions being taken to avoid entrapped air in the bond, in fact, it seemed quite likely, and the sections with entrapped bubbles would be NOT bonded. Same for any sections with entrapped dust, debris, or whatever else was floating around that warehouse. I would want to see, along with the all of the above steps, either or both applying very substantial pressure to the Ti/Carbon parts while curing to force out any air, and/or doing the whole process in a vaccum environment.
The out-of-date carbon fiber? Sure, it is fine, for some months, for noncritical applications. For critical applications? NO WAY. And the CEO's claim that aerospace needs to be more stringent than undersea 5000psi is just beyond stupid. You need absolutely the best bond between layers that you can get.
Also, the 480-odd layer layup and cure is hugely experimental by itself, and I've read elsewhere that the shop was given six weeks to design and fabricate the hull. Insane. I'd feel like six weeks to design and run a few initial proof-of-concept tests of a 5-inch thick layup would be too quick.
I could keep going for pages just on the carbon-fiber crimes I've read about.
My absolute disgust for the CEO grows every time I read more. He got what he deserved, but the passengers he suckered into taking the trip did not.
Would you expect carbon fiber to actually do anything useful in compression?
I fail to see how what is essentially magic string does anything helpful in the case of a submarine hull. It's the opposite of most applications, where the carbon fibers are in tension.
The only compressive strength I can see is from the matrix the fibers were embedded in, and given all the things you've noted... that's going to be way below optimum, and there are going to be voids for crack formation everywhere.
I'd certainly be VERY skeptical of a compression design until it was THOROUGHLY TESTED not only on the bench but in real-world scenarios.
In fact, fiberglass often has better compressive strength and we sometimes add fiberglass outer layers for that purpose, where there is local compressive load.
>>most applications, where the carbon fibers are in tension.
Exactly!
Carbon fibers are astonishing in tension. They take a point load and spread it out through large distances of matrix. Aligning it with tension or bending loads we can make amazing lightweight things.
The matrix is indeed the weak point and the and the shear between layers and between strands is the key failure mode I'm always trying to overcome or evade. Check out this 2min vid on the ASTM Open Hole Compression test [0]; it'll tell you a lot.
I'm not saying that carbon fiber composites are weak in compression, they are actually quite good. But the compression area is usually the point of failure.
The real failure mechanism goes back to the matrix failing between the fibers or layers. A great example was from one product we made using cored construction, an oblate carbon fiber tube with a structural foam core. The goal was to be substantially lighter than aluminum. The standard for that bar was 700 Lbs of load. Our structure was half the weight of the equivalent weight of aluminum, and didn't crack until 1200 Lbs of load. But what cracked was the top layer, in compression. The cool thing was that the rest of the structure held together at much higher loads, because basically the crushed upper load bearing surface was now suspended by the lower carbon fiber in tension, like a hammock.
There is no such opportunity with that sub — the core is the air and the gooey humans inside.
More importantly, the kinds of failure that happen in excess flexion and compression are cumulative.
A key advantage but also very tricky point is that carbon fiber composites do not have the kind of fatigue that metal does. A homogeneous metal still has grain structure, so if it is repeatedly flexed and never approaches the yield point, such as on an airplane wing, micro-cracks will grow and it will eventually fail (they know how to do inspections for this).
Carbon fiber composites with long-chain polymer matrices can flex nearly forever, as long as it never exceeds the yield point at any microscopic point. However, exceed that point, the matrix gets a micro-crack. Those micro-cracks WILL add up, and will accelerate because the entire trick of composites is for the reinforcement carbon fibers to spread the load through the matrix. That failed point of the matrix now no longer contributes to the load-bearing. So, all the other points take the load, and another will fail. This process will accelerate.
The thing is, these failure points can be detected by ultrasound or X-ray examination, but typically not visually, whereas metal fatigue shows up on the surface, especially with the right kinds of paint and/or dyes, so inspection can be easy.
Perhaps most fatally in this catastrophe, is that failure in carbon fiber composites is not gradual or ductile. Metals will show signs of weakness a good amount of time before the failure, and when it fails will tend to bend, not fracture. Carbon composites will fail almost explosively and catastrophically; again, see the ASTM video, the failure at 1:48 is totally typical.
I am actually surprised to read that James Cameron heard from his sources that the Titan crew had dumped their weights and were ascending, trying to manage an emergency. If that was the sensors signalling cracks, I'd expect only milliseconds between the first micro-crack and the catastrophic failure. It seems they had at least seconds, possibly due to the unusually large 5" thick layup.
I could keep going, but yes, you are on the right track.
But I always thought composite nonmetal materials are hard to test nondestructively, which is the main limiting factor for their use in aerospace. Metals can be X-rayed to assess their condition, carbon not.
Is that wrong? What am I missing? Just curious.
https://www.nytimes.com/2023/06/20/us/oceangate-titanic-miss...
> Years before OceanGate’s submersible craft went missing in the Atlantic Ocean with five people onboard, the company faced several warnings as it prepared for its hallmark mission of taking wealthy passengers to tour the Titanic’s wreckage.
> OceanGate’s director of marine operations, David Lochridge, started working on a report around that time, according to court documents, ultimately producing a scathing document in which he said the craft needed more testing and stressed “the potential dangers to passengers of the Titan as the submersible reached extreme depths.”
> Two months later, OceanGate faced similarly dire calls from more than three dozen people — industry leaders, deep-sea explorers and oceanographers — who warned in a letter to its chief executive, Stockton Rush, that the company’s “experimental” approach and its decision to forgo a traditional assessment could lead to potentially “catastrophic” problems with the Titanic mission.
It's not like carbon fiber was any cheaper than just making the thing out of steel or titanium like literally every other submersible. It just doesn't make any sense. You have none of the weight penalties that exist in aviation, so the heavier and the thicker the better.
They can’t use foam for flotation at that depth. There is a concrete like “synthetic foam” if you want unpowered rise to surface but it’s expensive. I read early on they wanted carbon to be light enough to get back up by releasing weights on powerloss.
From what I have seen of the hull, it was a tube. I can’t remeber seeing a view with a complex or even drafted shape to it.
I suspect that spinning carbon by fiber would take far too long, and think they probably used narrow weave to do it. It would have been cheaper than the required thickness of titanium, which is the same strength as steel, but 40% the weight and still a little heavier than aluminum.
Who knows!? Nothing they did showed that they ever chose the more expensive option. Carbon was likely just the cheapest thing they could use for their goals.
Fun experiment: Take a roll from some toilet paper, squeeze it somewhat evenly at the ends. It is kond of stable. Then peel the layers apart, ans squeeze again. The same would happen if the composite tube delaminated at one of the end cap rings. This whole thing is beyond infuriating.
I haven’t read that. But, that’s not an option. Carbon won’t just cure on its own, ever. It will be a shitty sticky mess forever. It had to be cured somehow.
As to the carbon expiration. I don’t know what that was. I suspect it could be epoxy like in prepreg, but I doubt prepreg here. And it could be shelf life of the carbon itself as it can be hydrophilic in some cases.
It's so absurd. Deep sea tourism is going to be something only for the .1% for a while.
They can afford a vessel that's 1000x more expensive. Your market is about 50 people. They've got unlimited money, basically.
Why cheap out and die? I just don't get it.
I'm assuming your engineering requirements for an unmanned sub would be substantially easier for a number of reasons (not needing to worry about killing someone, mainly).
This would make sense. The first bathyscapes (FNRS-2, Trieste) used the same iron ballast method via electromagnets, but bouyancy was provided by an external tank filled with a liquid of lower density than water -- gasoline.
I hadn't heard of the foam before, but I looked to see how modern deep submersibles provided bouyancy. The first one I checked was the one I'm most familiar with from my childhood fascination with the Titanic -- DSV Alvin, commissioned in 1964 and taken by Robert Ballard on the very first expedition to the wreck -- and it uses such a foam developed around the same time. A minor correction, though: the foam is called "syntactic foam"; the term shares its etymology with "syntax": from Greek σύν (sún, “together”) and τάξις (táxis, “arrangement”). These foams are composite materials containing what they call "cenospheres" or "microballoons": tiny, rigid, hollow spheres. Why use a large steel ballast filled with flammable hydrocarbons when you can use thousands of much smaller ones made out of an assortment of materials from ceramic to titanium to polymers?
The word you are looking for is "syntactic foam", which is an epoxy or other material typically with embedded hollow glass microspheres. This material is used for the buoyancy components of undersea vessels. It comes in a broad variety of densities and depth ratings. It is typically machined into shapes on the inside of the part fitting around the metallic components and on the outside the shape of the outer surface of the vessel. It is also somewhat expensive, more for higher depth ratings (on the scale of dozens to low thousands of dollars per cubic foot). Here's a product from one of the suppliers I worked with for more info [0]
Typically, the goal in designing the vessel is to account for the mass and buoyancy of the metal parts and design in the right final volume of syntactic foam so that the vessel is neutrally buoyant at the operating depths, and doesn't have to work to stay up or stay down.
For more moderate depths, there are also various structural foams, which use a variety of chemistries to achieve low densities and high compression strengths. These are also used in sandwich panel construction for composites (instead of honeycomb, and for the same effect), e.g., [1].
Source: I own & run a composites design & manufacturing shop which does work on, among other things aerospace and undersea projects.
[0] https://www.trelleborg.com/en/applied-technologies/products-...
Well, they were kind of proud [0] of buying carbon fiber from Boeing with a steep discount due to it being too old for aviation use, so maybe it was cheaper than steel or titanium, even if carbon fiber in general would not be.
[0] long before the accident; I doubt anyone at OceanGate would brag about that now.
That it was all a mess at OceanGate, that bit I got :)
That said, we are seeing much more CF (and CF composites) being deployed into commercial aerospace (famously the 787 wings, but also the A350,
Strutural repairs are tricky so.
https://tecscan.ca/ultrasonic-testing-defect-sizing-in-compo...
The money phrase: >>take off their engineer hats and put on their manager hats
This is what kills people.
To boil it down to one short course in engineering ethics:
Whenever anyone tells you to do act like a manager instead of an engineer when making a technical decision, tell them, in no uncertain terms, and as politely or impolitely as needed to make the point: to fuck off; physics is physics, and neither knows about, cares about, nor yields to human concerns.
(Unless, of course, you are OK with living decades of your remaining life after the accident knowing you accepted a blood-money paycheck to do bad engineering.)
Is your current job, or your company's immediate future worth possibly killing people?
Dear lord even if it is/was supposed to do that, this would be disturbing to look at.
It's quite possible it could lead to new regulations that would make similar stupid decisions illegal.
It might be time, it sounds like the whole human-occupied submersible industry is entirely unregulated.
That's gonna be a no^∞ from me.
It takes an incredible level of density to ignore this once, but multiple times?
Can someone tell me what happens to a human body when suddenly exposed to ~5000 psi?
Edit: I looked it up. The human body almost instantly incinerates. It takes a matter of milliseconds.
Mercifully you'd be turned to goo within a split second.
This is most likely BS- consider that a shotgun shell is made from a low melting-temp plastic and handles higher pressures and temperatures without issue (and repeatedly if you're reloading them).
Probably your hair and light clothing would burn somewhat during an implosion, but would be extinguished as soon as the water hit it. Then it would stop being hair and light clothing just due to the kinetic energy of water hitting you at cannon-ball velocity.
Yeah the "vaporizing" part is definitely wrong, as is the C4 comparison. It's pretty easy to figure out how much energy is involved. It's just a matter of multiplying the volume of the interior by the external pressure, in this case I don't know the exact dimensions, but the inside looks like at least 48" wide by 10ft long.
You'll end up with a number around 100 million foot-pounds, or 135MJ. (Or, for a more meaningful number, about the energy released from burning ~3 liters of gasoline).
Still an enormous amount of energy, and enough to turn people into tiny little pieces, but it isn't going to literally convert you to ashes like the article implies. It might barely be enough to boil 1 "person's worth" of water under ideal conditions.
Is that normal for a submersible viewport? I feel like that's probably not normal.
More succinctly: It's probably "normal", but OceanGate didn't know if it was "normal".
Does anyone know of a good documentary about famous engineering failures? Bonus points if they actually go into detail about the scientific explanation, as opposed to focusing on the human failures
In all seriousness, the series about Chernobyl is great in driving that point across.
Sure, titanium or appropriate steel would be more expensive, but it's hard for me to imagine that we're talking about an amount of money that would simply make the project unprofitable, since there are plenty of other submersibles out there built without carbon fiber hulls, and those companies are making money. Even if we were talking about a difference of some hundreds of thousands of dollars (can't imagine it would be more) between the two materials, we're still talking about life-or-death here (and one of those lives was Rush's). And while obviously the overhead costs of operating a submersible business are very high, these customers are paying $250k/dive. It seems like if Rush had said to people, "We went with ti over carbon, which would have saved money, but now we can be much more sure that the material won't fail catastrophically," anyone with that kind of money would have been happy to pay an extra, say, $10k or $20k or $50k/dive to make sure that they didn't die.
My take on the situation is that the money was less important in his calculus than the ego aspect of wanting to be seen as a maverick and showing the rest of the sub engineering community that he was right against their conventional wisdom. Being seen an 'innovator' seems to have been a powerful motivation for Rush. Often the actual amount of money matters much less to people than the 'animal spirits' (to use Keyne's phrase) underlying those decisions. Maybe I'm just trying to give Rush the benefit of the doubt, since putting people's lives at risk just so he could further enrich himself (and it doesn't seem like we're talking here about a guy that could barely afford lunch) is just evil. Maybe he was greedy and cheap (which doesn't necessarily mean that he didn't have a lot of good qualities).
Thoughts? I'd love to see someone run the numbers.