5 ms·

What little I know about composites/carbon fiber STRONGLY suggest it's a terrible approach for a submersible. Am I wrong?

pvaldes 3y ago
The results speak loudly by themselves. This is not the first submarine build and many of the other are still on duty after decades.
dan_quixote 3y ago
Not necessarily. CFRP is plenty strong to handle the task. But it's a complicated material - though not necessarily new. I think one complication people underestimate when comparing to steel or titanium is that CFRP is NOT isotropic - meaning that its physical properties are NOT the same regardless of orientation. Another is that CFRP has so many dimensions to its strength (e.g. orientation of resolved force vs fiver orientation, fiber diameter, fiber length, fiber material, epoxy material, etc). Yet another is the concept of fatigue life. Steel is very forgiving in this regard - as long as you don't strain it enough to reach plastic deformation, you're good almost forever. CFRP (and even aluminum) have a lifetime no matter what. In that case, you need decent non-destructive testing techniques of which I'm certainly not an expert. I have one major unanswered question about it though...is there any reason they needed the weight savings one gets from CFRP? Every sub I've ever heard of requires additional ballast to dive even when using tried-and-true steel hull. If they didn't need the weight savings, it was completely absurd to use CFRP.
- jcalvinowens 3y ago
  The metal pressure vessels are very negatively buoyant. The bathyscaphe Trieste used 32000 gallons of gasoline to achieve positive buoyancy. Modern DSVs use incredibly expensive and fragile syntactic foam. A CFRP pressure vessel can be naturally positively buoyant, or much closer to it. Titan's dry weight was less than DSV Limiting Factor, despite having much greater internal volume and carrying 3 more people. Also, being able to load and launch in the water, rather than loading on the deck and needing a human rated crane to launch as most DSVs do, probably saved a lot of money operating Titan.
msandford 3y ago
I never understood what all the people who were going on and on about "water intrusion" were saying, because it was never explained. But I think I understand now. Boring old A36 carbon steel is good for like 30ksi (30000psi). The water pressure is only say 6ksi which is significantly less. That means that the steel is much stronger on its own, in every direction, than the water pressure pushing on it. The epoxy in most epoxy composites isn't rated for 30ksi, it's generally less than 10ksi and perhaps even less than 5ksi. Epoxy on its own is capable of standing up to less pressure than the composite will see at 13000ft deep. So the concern is that it's not if but when water manages to wiggle it's way past some of the epoxy between the carbon strands. Sure the bulk composite material can stand up to 6ksi, but if the glue that holds it together is weaker, well, the water can push it out of the way. However deep the water gets it has reduced the effective thickness and this strength of your composite at that point, if it goes deep enough it shoots through and that's the start of a very quickly evolving collapse.
- ubermonkey 3y ago
  I'm a cyclist. People DO worry about the failure mode of carbon fiber when riding, and the stakes in bicycling are FAR LOWER -- I mean, you'll almost certainly live. From the factory it's gonna be fine, but the worry people have is getting a nick or dent in it in such a way that it compromises the strength in an important dimension. Carbon fork failures used to be a thing, and that can end REALLY REALLY BADLY for a rider on a road bike. So going deep enough that a vessel failure means effectively instant disintegration in a carbon sub? F*ck no, like, on the face of it.
- everforward 3y ago
  The other part is that carbon fiber is good at handling expansive pressure (pressure coming from the inside of the vessel) and bad at handling compressive pressure (pressure outside the vessel). It kind of intuitively makes sense. Carbon fiber is roughly like a fabric. If you make a tube out of a piece of fabric, there's almost no resistance when you try to compress it between your hands. There's far more resistance if you try to move your hands apart inside it, up until the fabric rips. It was a terrible material for this to begin with between the risks of delamination and carbon fiber being specifically bad at the kind of pressure they were dealing with.

pvaldes 3y ago

The results speak loudly by themselves. This is not the first submarine build and many of the other are still on duty after decades.

dan_quixote 3y ago

Not necessarily. CFRP is plenty strong to handle the task. But it's a complicated material - though not necessarily new. I think one complication people underestimate when comparing to steel or titanium is that CFRP is NOT isotropic - meaning that its physical properties are NOT the same regardless of orientation. Another is that CFRP has so many dimensions to its strength (e.g. orientation of resolved force vs fiver orientation, fiber diameter, fiber length, fiber material, epoxy material, etc). Yet another is the concept of fatigue life. Steel is very forgiving in this regard - as long as you don't strain it enough to reach plastic deformation, you're good almost forever. CFRP (and even aluminum) have a lifetime no matter what. In that case, you need decent non-destructive testing techniques of which I'm certainly not an expert. I have one major unanswered question about it though...is there any reason they needed the weight savings one gets from CFRP? Every sub I've ever heard of requires additional ballast to dive even when using tried-and-true steel hull. If they didn't need the weight savings, it was completely absurd to use CFRP.

jcalvinowens 3y ago
The metal pressure vessels are very negatively buoyant. The bathyscaphe Trieste used 32000 gallons of gasoline to achieve positive buoyancy. Modern DSVs use incredibly expensive and fragile syntactic foam. A CFRP pressure vessel can be naturally positively buoyant, or much closer to it. Titan's dry weight was less than DSV Limiting Factor, despite having much greater internal volume and carrying 3 more people. Also, being able to load and launch in the water, rather than loading on the deck and needing a human rated crane to launch as most DSVs do, probably saved a lot of money operating Titan.

msandford 3y ago

I never understood what all the people who were going on and on about "water intrusion" were saying, because it was never explained. But I think I understand now. Boring old A36 carbon steel is good for like 30ksi (30000psi). The water pressure is only say 6ksi which is significantly less. That means that the steel is much stronger on its own, in every direction, than the water pressure pushing on it. The epoxy in most epoxy composites isn't rated for 30ksi, it's generally less than 10ksi and perhaps even less than 5ksi. Epoxy on its own is capable of standing up to less pressure than the composite will see at 13000ft deep. So the concern is that it's not if but when water manages to wiggle it's way past some of the epoxy between the carbon strands. Sure the bulk composite material can stand up to 6ksi, but if the glue that holds it together is weaker, well, the water can push it out of the way. However deep the water gets it has reduced the effective thickness and this strength of your composite at that point, if it goes deep enough it shoots through and that's the start of a very quickly evolving collapse.

ubermonkey 3y ago
I'm a cyclist. People DO worry about the failure mode of carbon fiber when riding, and the stakes in bicycling are FAR LOWER -- I mean, you'll almost certainly live. From the factory it's gonna be fine, but the worry people have is getting a nick or dent in it in such a way that it compromises the strength in an important dimension. Carbon fork failures used to be a thing, and that can end REALLY REALLY BADLY for a rider on a road bike. So going deep enough that a vessel failure means effectively instant disintegration in a carbon sub? F*ck no, like, on the face of it.
everforward 3y ago
The other part is that carbon fiber is good at handling expansive pressure (pressure coming from the inside of the vessel) and bad at handling compressive pressure (pressure outside the vessel). It kind of intuitively makes sense. Carbon fiber is roughly like a fabric. If you make a tube out of a piece of fabric, there's almost no resistance when you try to compress it between your hands. There's far more resistance if you try to move your hands apart inside it, up until the fabric rips. It was a terrible material for this to begin with between the risks of delamination and carbon fiber being specifically bad at the kind of pressure they were dealing with.