3 ms·

The number of stupid decisions that went in the design and construction of the Titan is astonishing. One of my favorites was that, after putting on the carbon f

by bambax 11mo ago

The number of stupid decisions that went in the design and construction of the Titan is astonishing. One of my favorites was that, after putting on the carbon fiber around the tube, they would sand imperfections to make the surface perfectly smooth, severing layers in the process! It shouldn't require an engineering degree from MIT to recognize this as ill-advised.

lazide 11mo ago
Even without that, the material is just wrong. It’s strong in tension, not so much compression. Tends towards sudden brittle fractures. Doesn’t like impacts, as it tends to have issues with delaminating. It’s just not what you ever want as a sub hull. It’s dumb. And weight is not even a huge issue for a sub!
- bambax 11mo ago
  Yes, using carbon fiber was also a very bad decision; it was known for a very long time that it was only good for single-use sub, because after the first dive it was too damaged to continue. In 2014, Virgin Oceanic, which had similar plans with similar technology, closed shop because it didn't make economic sense to build a new sub for each dive. But weight is absolutely an issue; the basic and tried-and-true metal sphere design allows for only three people. Since size and thickness grow exponentially, making a sphere for more than three people becomes more and more difficult. And it should also be possible to lift the vehicle with a crane. But if you want to carry paying passengers (like Oceangate did), having only two per dive is very limiting. That's why they went with a tube design, and carbon fiber to limit weight. But it couldn't work, and it didn't.
  - schiffern 11mo ago
    >size and thickness grow exponentially It's a [reverse] pressure vessel, so it follows pressure vessel scaling. Mass scaling is linear with internal volume.
    - FabHK 11mo ago
      It’s funny how “literally” often means “figuratively” now, and “exponentially” means “polynomially”.
      - bambax 11mo ago
        Yep. https://books.google.com/ngrams/graph?content=exponentially%2C+polynomially&year_start=1900&year_end=2022&corpus=en&smoothing=3&case_insensitive=true https://books.google.com/ngrams/graph?content=exponentially%...
    - bambax 11mo ago
      Ok yes "exponentially" was hyperbolic. Mass scales linearly with volume, but volume is proportional to the cube of the radius (not linear). Also, in practice, small imperfections can have a disproportionate impact on the resistance of the sphere so design codes typically apply conservative reductions that can have a big impact on actual thickness requirements.
      - wkat4242 11mo ago
        Did this thing meet any design codes though? I doubt it.
        donavanm 11mo ago
        I read the report when it come out. From memory, no. It never had any components or certification for human pressure vessels. IIRC theres no existing regs for carbon fiber and it would have cost like $50M to do the design and test work. They did buy some things, like the viewport, from companies who do certified parts, but instead opted for the same design minus any test certs to save money. The craft was never certified or inspected by the uscg. It did have a registration for a while, but they had to play find-a-new-district-sign-off shell games for a while, then… just stopped bothering.
        wkat4242 11mo ago
- owenversteeg 11mo ago
  “Strong in tension, not compression” is a meme, and obviously wrong. It is certainly stronger in tension, but it is also remarkably strong in compression. That’s why it’s used - yes, in compression - in modern passenger aircraft. You don’t even need to know that, though; the simple fact is that the Titan had a double-digit number of deep dives. If it was weak in compression it would not have survived diving to 3.7 kilometers deep or even a fraction of that depth _once_. That said, yes, it’s a difficult material to use properly, and they were a bunch of cowboys slapping things together. It’s no surprise that they missed several critical steps and created a sub doomed to fail. N.b. all of this was kickstarted by James Cameron saying that carbon fiber has “no strength in compression” in a New York Times “science” article, which the Times printed directly.
  - lazide 11mo ago
    Aircraft fuselages are typically loaded in tension. It’s a key part of the design. Carbon fiber compressive strength is only ~ 30-50% of it’s tensile strength because of the way the fibers and the epoxy interact. When compressed, the carbon fibers don’t do as much. [https://www.sciencedirect.com/science/article/abs/pii/S026382232201025X https://www.sciencedirect.com/science/article/abs/pii/S02638...] But don’t believe me, actually read a useful paper on the subject. In fact, it’s a major factor limiting it’s wider use. As is it’s fatigue behavior, which would probably also explain why it eventually imploded! I never followed James Cameron’s interview, but it sounds like he knows what he is talking about!
    - buildbot 11mo ago
      Speaking of which I heavily recommend reading interview the prime ancestor comment to this chain linked. It’s really clear he knows what he is talking about.
    - owenversteeg 11mo ago
      James Cameron certainly knows a lot about submarines, but if he says something factually incorrect then it’s factually incorrect, period. Carbon fiber does not have “no strength in compression” and it is used in compression in countless applications, for example airplane wings. Again, the fact that the sub - built at absurdly low cost for its size, built by a bunch of cowboys that didn’t know what they were doing - DID survive to 3.7 km deep on several occasions is proof sufficient. If CF had no compressive strength than the whole thing would have failed at a tiny fraction of that depth. If CF had no compressive strength then what kept the sub together during the successful dives? Hopes and dreams? I’m not here to defend the decision to use carbon fiber, and as I’ve said I completely agree that there are many issues with using it in this application. Delamination, water ingress, bonding the titanium to the carbon fiber, difficulty of manufacture including varying layer thickness and voids, sensitivity to impact, the list goes on. But _those_ are the issues, not the compressive strength.
franktankbank 11mo ago
I don't like this interpretation of things. Its worthwhile to experiment and try things. They were basically mentally ill as a group and rejected genuine concern. Everyone wants to shit on the build but it was the human relations that killed it.
- nativeit 11mo ago
  Also, honestly, the build. That “genuine concern” they ignored was that the build was critically flawed. I don’t think anyone here would have these takes if a small group of curious engineers tried their hand at a composite submersible, it was when they kept doing it after all the qualified engineers had said, “This is crazy, I’m out.” The build was kind of dumb, and I’m hardly an engineer. It’s common sense. Carbon fiber composites are interesting because they’re strong relative to their weight. Remove either of those features and they become pointless. Who cares if a submarine is heavy?
  - potato3732842 11mo ago
    A submarine needs to be light to be neutrally buoyant in order to fly though the water properly. Otherwise you have a bathyscaphe, which has some other not nice failure modes (in addition to the "implodes if you f it up" one) and are much less maneuverable and arguably the whole system is less durable more costly for high tempo operations. "Just build a tube strong enough and big enough to not need all that" is a better answer, if you can pull it off.

by bambax 11mo ago

lazide 11mo ago

Even without that, the material is just wrong. It’s strong in tension, not so much compression. Tends towards sudden brittle fractures. Doesn’t like impacts, as it tends to have issues with delaminating. It’s just not what you ever want as a sub hull. It’s dumb. And weight is not even a huge issue for a sub!

bambax 11mo ago
Yes, using carbon fiber was also a very bad decision; it was known for a very long time that it was only good for single-use sub, because after the first dive it was too damaged to continue. In 2014, Virgin Oceanic, which had similar plans with similar technology, closed shop because it didn't make economic sense to build a new sub for each dive. But weight is absolutely an issue; the basic and tried-and-true metal sphere design allows for only three people. Since size and thickness grow exponentially, making a sphere for more than three people becomes more and more difficult. And it should also be possible to lift the vehicle with a crane. But if you want to carry paying passengers (like Oceangate did), having only two per dive is very limiting. That's why they went with a tube design, and carbon fiber to limit weight. But it couldn't work, and it didn't.
- schiffern 11mo ago
  >size and thickness grow exponentially It's a [reverse] pressure vessel, so it follows pressure vessel scaling. Mass scaling is linear with internal volume.
  - FabHK 11mo ago
    It’s funny how “literally” often means “figuratively” now, and “exponentially” means “polynomially”.
    - bambax 11mo ago
      Yep. https://books.google.com/ngrams/graph?content=exponentially%2C+polynomially&year_start=1900&year_end=2022&corpus=en&smoothing=3&case_insensitive=true https://books.google.com/ngrams/graph?content=exponentially%...
  - bambax 11mo ago
    Ok yes "exponentially" was hyperbolic. Mass scales linearly with volume, but volume is proportional to the cube of the radius (not linear). Also, in practice, small imperfections can have a disproportionate impact on the resistance of the sphere so design codes typically apply conservative reductions that can have a big impact on actual thickness requirements.
    - wkat4242 11mo ago
      Did this thing meet any design codes though? I doubt it.
      - donavanm 11mo ago
        I read the report when it come out. From memory, no. It never had any components or certification for human pressure vessels. IIRC theres no existing regs for carbon fiber and it would have cost like $50M to do the design and test work. They did buy some things, like the viewport, from companies who do certified parts, but instead opted for the same design minus any test certs to save money. The craft was never certified or inspected by the uscg. It did have a registration for a while, but they had to play find-a-new-district-sign-off shell games for a while, then… just stopped bothering.
        wkat4242 11mo ago
owenversteeg 11mo ago
“Strong in tension, not compression” is a meme, and obviously wrong. It is certainly stronger in tension, but it is also remarkably strong in compression. That’s why it’s used - yes, in compression - in modern passenger aircraft. You don’t even need to know that, though; the simple fact is that the Titan had a double-digit number of deep dives. If it was weak in compression it would not have survived diving to 3.7 kilometers deep or even a fraction of that depth _once_. That said, yes, it’s a difficult material to use properly, and they were a bunch of cowboys slapping things together. It’s no surprise that they missed several critical steps and created a sub doomed to fail. N.b. all of this was kickstarted by James Cameron saying that carbon fiber has “no strength in compression” in a New York Times “science” article, which the Times printed directly.
- lazide 11mo ago
  Aircraft fuselages are typically loaded in tension. It’s a key part of the design. Carbon fiber compressive strength is only ~ 30-50% of it’s tensile strength because of the way the fibers and the epoxy interact. When compressed, the carbon fibers don’t do as much. [https://www.sciencedirect.com/science/article/abs/pii/S026382232201025X https://www.sciencedirect.com/science/article/abs/pii/S02638...] But don’t believe me, actually read a useful paper on the subject. In fact, it’s a major factor limiting it’s wider use. As is it’s fatigue behavior, which would probably also explain why it eventually imploded! I never followed James Cameron’s interview, but it sounds like he knows what he is talking about!
  - buildbot 11mo ago
    Speaking of which I heavily recommend reading interview the prime ancestor comment to this chain linked. It’s really clear he knows what he is talking about.
  - owenversteeg 11mo ago
    James Cameron certainly knows a lot about submarines, but if he says something factually incorrect then it’s factually incorrect, period. Carbon fiber does not have “no strength in compression” and it is used in compression in countless applications, for example airplane wings. Again, the fact that the sub - built at absurdly low cost for its size, built by a bunch of cowboys that didn’t know what they were doing - DID survive to 3.7 km deep on several occasions is proof sufficient. If CF had no compressive strength than the whole thing would have failed at a tiny fraction of that depth. If CF had no compressive strength then what kept the sub together during the successful dives? Hopes and dreams? I’m not here to defend the decision to use carbon fiber, and as I’ve said I completely agree that there are many issues with using it in this application. Delamination, water ingress, bonding the titanium to the carbon fiber, difficulty of manufacture including varying layer thickness and voids, sensitivity to impact, the list goes on. But _those_ are the issues, not the compressive strength.

franktankbank 11mo ago

I don't like this interpretation of things. Its worthwhile to experiment and try things. They were basically mentally ill as a group and rejected genuine concern. Everyone wants to shit on the build but it was the human relations that killed it.

nativeit 11mo ago
Also, honestly, the build. That “genuine concern” they ignored was that the build was critically flawed. I don’t think anyone here would have these takes if a small group of curious engineers tried their hand at a composite submersible, it was when they kept doing it after all the qualified engineers had said, “This is crazy, I’m out.” The build was kind of dumb, and I’m hardly an engineer. It’s common sense. Carbon fiber composites are interesting because they’re strong relative to their weight. Remove either of those features and they become pointless. Who cares if a submarine is heavy?
- potato3732842 11mo ago
  A submarine needs to be light to be neutrally buoyant in order to fly though the water properly. Otherwise you have a bathyscaphe, which has some other not nice failure modes (in addition to the "implodes if you f it up" one) and are much less maneuverable and arguably the whole system is less durable more costly for high tempo operations. "Just build a tube strong enough and big enough to not need all that" is a better answer, if you can pull it off.