Titan Carbon failure(apologize if i used the wrong sub-thread)

[fastupslowdown]

@ 7:59. Speculation is a Titan system detected Carbon failure and attempted to ascend dropping ballast. I have limited knowledge of Carbon Fiber. Isn't it incredibly strong but when it fails it does so quickly and spectacularly?

Cameron who knows more about this than I is of the opinion the crew was aware the craft was delaminating -Mike G

[jclay]

Structures and pressure vessel design weren't my specialty but I think I'm on pretty firm ground in opining that carbon fiber reinforced plastic isn't intrinsically one of the better choices in extreme compressive applications. Pretty fantastic in tension and when design or layup schedules can get enough fibers into tension, shear. But in pretty straight up compression and ferocious service conditions, which is obviously the issue here, I think it likely a very poor choice.

As someone who, twice, has been utterly disoriented and lost while exploring some of Florida's underwater cavern systems, and for probably less than a minute, I can tell you it is terrifying; keep your shit together or you're dead. I was shot at, only once mind you, but it didn't even move the fear needle; my reaction was simply operational; get cover and get the hell out of there. Lost when on limited air supply....I'm here to tell you that's a WHOLE other world. And it makes sense; humans have been swinging swords at each other, and such as that, for millennia and have evolved accordingly; blind, disoriented and air supply limited? That ain't in our evolutionary background. It's a tragedy but at the point at which they were doomed, I'm glad it was catastrophic, essentially instantaneous hull failure and, doubtless, a very quick death. I don't like imagining what tens of hours of knowing that you're going to suffocate in a tiny tube would be like; or a fire, or loss of control.

[fastupslowdown]

Structures and pressure vessel design weren't my specialty but I think I'm on pretty firm ground in opining that carbon fiber reinforced plastic isn't intrinsically good in compressive applications. Pretty fantastic in tension and when design or layup schedules can get enough fibers into tension, shear. But in pretty straight up compression, which is obviously the issue here, I think it likely a very poor choice.

That sounds like what Cameron was saying, but when I ride a bike isn't the carbon fiber stress compression?

Tension forces pull and stretch material in opposite directions, allowing a rope bridge to support itself and the load it carries. Compression forces squeeze and push material inward, causing the rocks of an arch bridge to press against each other to carry the load.

[jclay]

That sounds like what Cameron was saying, but when I ride a bike isn't the carbon fiber stress compression?

Tension forces pull and stretch material in opposite directions, allowing a rope bridge to support itself and the load it carries. Compression forces squeeze and push material inward, causing the rocks of an arch bridge to press against each other to carry the load.

Some is, but you can design around it and low weight is important and achievable so it can work out. It's not that the plastic (whatever resins are used) won't handle any but in a loading situation that pretty much is all crushing/compressin the carbon in the areas seeing compression isn't really contributing; the other parts of the matrix are doing the work. I think there's a reason you don't hear about carbon fiber sub hulls in the Navy. And with a thicker hull I'd imagine a carbon fiber submersible could be designed to handle 5500 psi (plus a large factor of safety considering the gravity of the situation and failure); it appears that this one wasn't. And carbon fiber doesn't strike me as a wise choice from a structural perspective; more likely a choice made to keep costs manageable.

Folks hear "carbon fiber" and figure "strong as hell"; but not in compressive apps, at least not from the intrinsic properties of the carbon fiber. Think pre-stressed concrete; it works in tension and compression 'cause the re-steel handles the tensile loads, the concrete the compressive. My guess is that the resin (or perhaps some other element of the design) couldn't handle the compressive load.

If examined in more depth than I am able to do and it probably gets more intricate but from a first principle viewpoint I don't think I'm far off.

[Mark Kelly]

A cylindrical hull under compression will probably fail by buckling and the classical buckling criterion is related to stiffness rather than strength so CFRP isn't as poor a choice as it might seem.

On the other hand small local imperfections disproportionately reduce the buckling load, so manufacturing quality becomes paramount.

Think of like an arch: if one small section of the arch begins to flatten the stress in the rest of the arch will push that section further inwards so the buckling gets worse so it is pushed further inwards etc etc.

[72gmc]

On the other hand small local imperfections disproportionately reduce the buckling load, so manufacturing quality becomes paramount.

".. he was 'repeatedly told that no scan of the hull or Bond Line could be done to check for delaminations, porosity and voids of sufficient adhesion of the glue being used due to the thickness of the hull.'"

https://newrepublic.com/post/173802/...avel-oceangate

[sk_tle]

@ 7:59. Speculation is a Titan system detected Carbon failure and attempted to ascend dropping ballast. I have limited knowledge of Carbon Fiber. Isn't it incredibly strong but when it fails it does so quickly and spectacularly?

At these kinds of pressure, anything that fail fails spectacularly.


(fast forward to 47s)

The good thing is they didn't had time to suffer and were probably unaware of the situation mere seconds before they died. This is a better outcome than waiting 5 days to die from lack of oxygen.

[fastupslowdown]

At these kinds of pressure, anything that fail fails spectacularly.


(fast forward to 47s)

The good thing is they didn't had time to suffer and were probably unaware of the situation mere seconds before they died. This is a better outcome than waiting 5 days to die from lack of oxygen.

James Cameron thinks the passengers might have heard the laminate coming apart right before failure

[jclay]

A cylindrical hull under compression will probably fail by buckling and the classical buckling criterion is related to stiffness rather than strength so CFRP isn't as poor a choice as it might seem.

On the other hand small local imperfections disproportionately reduce the buckling load, so manufacturing quality becomes paramount.

Think of like an arch: if one small section of the arch begins to flatten the stress in the rest of the arch will push that section further inwards so the buckling gets worse so it is pushed further inwards etc etc.

Excellent points Mark and one would expect buckling to be a prime candidate for the initial failure mode.

Slender fibers offer little resistance to buckling; woven and without other support I'd expect similarly poor resistance to buckling; resistance to buckling failure (and associated matrix compressive failure) would then seem to depend on the supporting resins and fillers adhesion to the fibers, shear strength, resin elasticity and pure compressive strength). Any sort of macro discontinuity (hatchs, flanges, attachment points), thinner hull areas, voids and that sort of thing would be candidates for initial failure points and I'd expect that to occur on the OD of the neutral axis; on the other side of the NA is where I'd expect excellent performance from the carbon part of the equation. All in all, and I may be wrong, but an FRP matrix of any stripe doesn't seem like an intrinsically great performer in that context; certainly carbon fiber doesn't seem to me the superstar it can be in purely tensile loading conditions (and is commonly thought of by the general public).

I'd imagine we'll all get an education enhancement concerning FRP in the coming weeks and months.

[jclay]

At these kinds of pressure, anything that fail fails spectacularly.


(fast forward to 47s)

The good thing is they didn't had time to suffer and were probably unaware of the situation mere seconds before they died. This is a better outcome than waiting 5 days to die from lack of oxygen.

That tank car failure occurred at less than a 14.7 psi pressure differential; we don't know the depth of the Titan failure with precision but at an 1hr, 45 min into a 2 hour decent the pressure differential was in the neighborhood of 5,000 psi. I'd guess the crew knew there was trouble, even if only, hopefully, for a very brief instant.

[fastupslowdown]

so why wouldn't this work on the sub?

https://www.compositesworld.com/arti...fiber-bicycles

[Mark Kelly]

so why wouldn't this work on the sub?

https://www.compositesworld.com/arti...fiber-bicycles

Because each composite layer represents a reflective interface due to the difference in acoustic impedance* between the fibres and the matrix. With a few layers you can still see the signal from a discontinuity since basically a discontinuity is a large impedance difference which will dominate over a few small differences. With a large enough number of small differences the discontinuity will disappear into the background.




* Acoustic impedance is the ability of a material to accept an incoming sound wave from another material. If two materials have the same impedance, 100% of the incoming energy will be transmitted. This is rare. When the two materials have different impedance, less than 100% of energy will be transmitted and the difference will be reflected at the interface.

Impedance is the product of the density of the material and its propagation velocity (speed of sound). Propagation velocity is the square root of the ratio of stiffness to density, so these are the two things that affect impedance. Since the carbon fibres are more than 100 times as stiff as the matrix but in the same ball park for density, there is a very large difference in impedance between them

An example that might make more sense is your speakers: the cone is a lot stiffer and denser than air, so very little of the acoustic energy in the cone gets transmitted to air. A typical figure would be 2% transmission so 98% of the energy gets reflected back into the speaker mechanism where it causes resonance and dissonance. That's why a 100 watt amplifier produces a watt or so of actual acoustic energy in the room and the speakers sound so bad.

[fastupslowdown]

Because each composite layer represents a reflective interface due to the difference in acoustic impedance* between the fibres and the matrix. With a few layers you can still see the signal from a discontinuity since basically a discontinuity is a large impedance difference which will dominate over a few small differences. With a large enough number of small differences the discontinuity will disappear into the background.




* Acoustic impedance is the ability of a material to accept an incoming sound wave from another material. If two materials have the same impedance, 100% of the incoming energy will be transmitted. This is rare. When the two materials have different impedance, less than 100% of energy will be transmitted and the difference will be reflected at the interface.

Impedance is the product of the density of the material and its propagation velocity (speed of sound). Propagation velocity is the square root of the ratio of stiffness to density, so these are the two things that affect impedance. Since the carbon fibres are more than 100 times as stiff as the matrix but in the same ball park for density, there is a very large difference in impedance between them

An example that might make more sense is your speakers: the cone is a lot stiffer and denser than air, so very little of the acoustic energy in the cone gets transmitted to air. A typical figure would be 2% transmission so 98% of the energy gets reflected back into the speaker mechanism where it causes resonance and dissonance. That's why a 100 watt amplifier produces a watt or so of actual acoustic energy in the room and the speakers sound so bad.

Thank you . I find this thread very interesting. I imagine similar conversations are happening with not only carbon fiber bike buildings but at Boeing , Airbus, makers of yachts etc

[jclay]

Thank you . I find this thread very interesting. I imagine similar conversations are happening with not only carbon fiber bike buildings but at Boeing , Airbus, makers of yachts etc

Those (aircraft, yachts and pressure piping) are mature industries with scads of procedures, processes, mature and proven technology, regulations, testing protocols, accepted design and operation factors of safety, etc., etc.; they're long past the cowboy/adventurer/"hey, let's try this and no we don't need to follow too many rules" period of evolution. They're also very different loading/service environments. More likely, I'd guess, are heads shaking by those who understand what it takes to design, test, maintain, provide backup systems (etc.) and operate submersible vessels, at the design, operation and other aspects of the Titan project. Pipeline & pressure vessel design and operation in the thousands of psi are enormously serious businesses and in tension applications (like pipelines and aircraft pressure hulls), to my non-specialist mind, seems far easier to deal with; submersible hull design at similar pressures seems relatively off the charts by comparison. I'm wondering...can you put, say, cured epoxy under enough pressure to become fluid? Viscoelastic, if it sorta already isn't?

Mark - your info on that type of NDT was interesting and something I know zero about. The longer I live the less impressed I am with my little BSME; that degree is more a starting point for....well, the beginnings of specialization (and then only in a Newtonian framework) and then one realizes that the technical universe is mind bendingly enormous.

[j44ke]

I know the back story on James Cameron and deep sea exploration, and I also know that the materials expertise here at VSalon is considerable. I am not criticizing the presence of this thread or even really James Cameron. But for the news to go ask James Cameron to speculate on what happened in a disaster rescuers just found on the sea floor seems like classic US news idiocy and actually disrespectful of all who died in the accident and probably also a bit unfair to James Cameron to put him in this position where he has to speculate with very few actual facts.

Sort of feels like getting Stanley Kubrick (I know - he's dead) to speculate on Russian atomic threats in Ukraine.

The more respectful thing to do would be to get someone who builds these submersibles and knows the materials side of things and is a certified expert in their field. But I doubt anyone like that would be willing to comment at this point so early in the process, so they got James Cameron.

Okay, rant over. Everyone carry on here with the discussion. Like I said, I am not criticizing the thread. Onward with the materials science and the science of deep sea pressures.

[j44ke]

That tank car failure occurred at less than a 14.7 psi pressure differential; we don't know the depth of the Titan failure with precision but at an 1hr, 45 min into a 2 hour decent the pressure differential was in the neighborhood of 5,000 psi. I'd guess the crew knew there was trouble, even if only, hopefully, for a very brief instant.

Yes, take anything into a lab and put an external force over all surfaces of the object at 5000psi of pressure onto it (if that's possible in a lab, I don't know) and see what happens. As far as physics go, going to the bottom of the ocean is like going into outer space in terms of the challenges. Expected behaviors may or may not be consistent in that environment.

[suspectdevice]

I grew up under the impression that he Woods Hole Oceanographic Institute is a CIA front.

Isn’t the real story of the Titanic wreck discovery that it was found incidentally during a surveillance and reconnaissance operation?
also:
I’m pretty sure Cameron has security clearance into some dark programs, note in interviews that his contacts told him about the implosion pretty much in real time.

This whole “search” is such a sickening farce.

[fastupslowdown]

My big question after hearing from Cameron was that if the ship had sensors shouldn't those sensors have communicated with hits mother vessel?

[j44ke]

My big question after hearing from Cameron was that if the ship had sensors shouldn't those sensors have communicated with hits mother vessel?

Isn't that the definition of lost?

[bigbill]

As the retired submariner around here, I've been amazed at what I've read on social media in the last 4-5 days. As far as the implosion, at depth, the imploding hull moves inward at 2200' a minute, so we're talking less than a millisecond, less time than it takes for the brain to process what's happening. As far as the noises of delamination, all submersibles make popping noises as they descend due to the compression of materials. It's how we know an opposing submarine is changing depth. There were so many problems with this design that were never addressed, including the viewing port and communications. There was a seven-hour period after communication was lost before the support crew reported anything wrong. The fact that loss of communication wasn't seen as abnormal is also concerning. Regarding the use of carbon fiber, the choice of materials makes little sense. If it was done for cost, don't dive deep. You could build a steel or titanium vessel with neutral buoyancy if it were done for weight.

Submarine hulls are built to handle cycling compression stress. That's why steel is used. The Soviets had some Ti hull subs that could dive deep but eventually abandoned that design because of hull cracking. Even steel hulls have a stress life, typically 30 years of diving and surfacing (the numbers should always be equal). Service beyond the expected life of a hull requires expensive and time consuming non destructive testing.

Finally, my job when stationed at Pearl Harbor Naval Shipyard was non-nuclear repair work on submarines. The subsafe system was developed after the Thresher tragedy. Certain systems that keep the sea out or gages that show depth are part of the subsafe boundaries. Any work on those systems requires many levels of inspection and certification before the sub is allowed to go to sea. Even when the sub can leave port, when it dives, depth is held at certain increments while the repaired systems are inspected.

If any of you are in the PNW, the museum at the Keyport Base, north of Silverdale, WA, has most of the record breaking submersibles on display including the Trieste series.