8 ms·

The first image does show the porthole without any glass (acrylic) in it, but I'm sure an implosion would be violent enough to blow it out anyway. I'm going to

The first image does show the porthole without any glass (acrylic) in it, but I'm sure an implosion would be violent enough to blow it out anyway. I'm going to be very interested in what they determine to be the critical failure point. Porthole is a possibility, but there were some other points of failure in the joining of the titanium domes to the carbon composite hull.

EA-3167 3y ago
I'm interested in seeing the skids, and learning about where they were found relative to the rest of the debris. Were the skids totally mangled, implying that they were attached when the implosion occurred? Is the debris field pretty contained, or very spread out? The former implies that the sub was close to the bottom, the latter implies that whatever happened occurred at a depth which allowed for currents and drift to spread the debris. If the skids are mostly intact then maybe there was some warning of trouble and the people in the sub tried to drop ballast. If the skids are destroyed then whatever happened is more likely to have been sudden and without warning.
- KineticLensman 3y ago
  Yes, this technique was critical to understanding the break up of the space shuttle Columbia as it re-entered. The debris track stretched across Texas and Louisiana, and in general objects to the west were those that had broken off first. Although things were confused because differences in ballistic number affected how far fragments travelled eastward, and in some cases modules broke off, travelled east and then disintegrated into sub-components.
causality0 3y ago
From everything I've read, the entire carbon section of the hull could be considered a weak point.
- KineticLensman 3y ago
  > the entire carbon section of the hull could be considered a weak point It is possible that the carbon fibre delaminated catastrophically, possibly as a result of stresses accumulated over multiple descents.
  - sumo89 3y ago
    The only way the carbon tube hasn't turned to sea dust is if the porthole failed first. When carbon fibre fails it catastrophically fails. I'd put money on the joint between the ti and carbon being the weak point and once there's a chip in the carbon under that much pressure it'll disintigrate.
    - KineticLensman 3y ago
      fair points. Hopefully the investigation will nail the actual cause down.
- ortusdux 3y ago
  This is my understanding as well. I'm not surprised to see that the titanium piece survived unscathed. IIRC, the companies goal was to provide more cost-effective trips by transitioning from the standard solid titanium hull to a Ti/carbon fiber composite. This decision was controversial. They fired their head of QC after he strongly opposed the change. One key argument he made was that the design prevented ultrasonic inspection, which would mean that they could not monitor the health of the composite over time.
- ajross 3y ago
  To be clear, way too much of what I've read on this subject has been excessively confident tut-tutting with little to no real engineering expertise. People want to imagine that the simple use of carbon fiber here was the sole mistake, I guess because it's easy to understand and makes us feel smart. But come on, engineering is harder than that. Carbon fiber pressure vessels have a long history. Composite material failure modes are for sure more complicated than metals but they certainly aren't a mystery. We've been using composites for more than half a century, we know how they work. No, my gut says the root cause here (if we ever find one[1]) is going to turn out to be a more routine kind of engineering fuckup. It'll be just some weak point that the analysis failed to turn up. [1] And sure, if we don't then the "carbon fiber death cylinder" theory will just become part of the folk truth, I guess.
  - mechhacker 3y ago
    There are non-trivial differences in designing vessels loaded with compression vs. tension, and the most experience we have is with internal pressure (e.g. tension). The analysis is very nonlinear and unfortunately not as widely understood as internally pressurized vessels. It requires a lot more analysis, troubleshooting, and expertise that is short in supply. Add the additional catastrophic failure modes for composites and there's a reason that industry experts warned against this.
    - ajross 3y ago
      With respect: that's exactly the kind of glib, drive-by analysis I'm talking about. What you're saying simply isn't true. People have been testing composites in pressure chambers and every load imaginable for decades. Yes, clearly they don't act like steel. But saying that it's impossible is just too easy. Just watch. It's going to be a routine design flaw, like they added a joint or a wiring hole in a bad place and didn't re-run the model. (Edit because this was misunderstood: I'm not saying what the failure is. I'm saying that we DON'T KNOW what the failure is, and that given the choice between makes-me-feel-smart pontification about materials and a routine design flaw in a prototype device, I'm going to bet on the routine failure every single time.)
- TechBro8615 3y ago
  Here's a video of hydraulic pressing carbon fiber to its breaking point: https://youtube.com/watch?v=xWTXeGiM8K8 https://youtube.com/watch?v=xWTXeGiM8K8
  - mechhacker 3y ago
    Worth noting that because he's using a hydraulic press, the load isn't constant. It deforms, fibers break, and thus the load decreases as the deflection stays relatively constant in that time frame. With external pressure the load doesn't decrease. Fibers break -> deflection increases -> more fibers break. There isn't load relief as long as the vessel is still holding pressure.
    - sojuz151 3y ago
      It is even worse. The load increases with deflation
  - N19PEDL2 3y ago
    Imagine how frightening it would have been to hear those crackles inside the submarine. (In reality, however, they had no warning of the impending collapse.)
mechhacker 3y ago
I'll speculate that because they are pulling pieces of the Ti domes without signs of carbon attached to it that the joint there wasn't stiff enough, and that carbon separated during failure without plastically deforming the Ti at the joint. But, that's speculation and I'd need to see more pics with more details.
- _yb2s 3y ago
  I would speculate that all of the carbon basically instantaneously liquefied under the pressure after a critical threshold of fibers failed, leaving the Ti untouched. I also want to mention that the large end caps you are seeing weren't bonded to the hull at all- there was a small Ti flange epoxied to the hull, and then the large domes were bolted to that. The front, with the window, was unbolted and rebolted every time for ingress/egress. There is no way that CF epoxied to thick Ti is going to make a strong enough bond that it will deform the Ti when it detaches. In this case, the geometry of that "glued" joint was such that it was forced together by the pressure underwater, rather than pulled apart- so it was not necessary for it to have any real strength other than for holding the end caps on at the surface, and filling voids to make it seal.
  - mechhacker 3y ago
    I see. Thus that large Ti flange that was bonded to the dome didn't carry a load through the bolts significant enough to deform the dome either. Carbon gone in a flash and comes apart at that joint totally, Ti is left hanging in water.
  - fullstop 3y ago
    I was hoping to see if that titanium flange was still bolted to the cap, but it's obscured by the tarp in the photo.
    - mechhacker 3y ago
      I just saw this: https://www.youtube.com/watch?v=uP4SrLhUHY8 https://www.youtube.com/watch?v=uP4SrLhUHY8 Those tarps may be covering what's left of the crew
_yb2s 3y ago
The hull imploded, not the window, but the shockwave probably pushed the window outward or shattered it, as it was only held strongly against moving inwards. People make a big deal about the acrylic glass not being rated for the depth, but I think it's almost impossible that it was the window, and think it must have been the hull: * The properties acrylic are well understood, and the strength of an acrylic disc is trivial to calculate, even if not "rated" you can easily calculate if it's strong enough * The window should fail visually with a lot of warning before rupture * If the window failed with the titanium end cap, there would be no reason for implosion of the carbon fiber hull The hull on the other hand is an irregular/non-homogenous part that is impossible to model accurately. There was no way to know how strong it was without extensive testing on a number of identical hulls- which they never did. The solidworks modeling method they used to give a ~2x margin of safety was nonsense as it ignores the realities of stress cycles, defects, etc.
- TechBro8615 3y ago
  How was the window "connected" to the hull? Would that be a weak point?
  - Wojtkie 3y ago
    Deep sea portholes are usually cones of acrylic that seat right into a port in the hull [0] [0]https://www.researchgate.net/publication/340442621_Prediction_of_the_service_life_of_deep-sea_conical_acrylic_portholes_at_designing https://www.researchgate.net/publication/340442621_Predictio...
  - EA-3167 3y ago
    This sub was a carbon fiber cylinder with two titanium end-caps, and the porthole was fitted into the forward end-cap.
  - _yb2s 3y ago
    They machine a wide ledge on the titanium, so the window is pushed into the groove/sill under pressure, basically a large window that is covering a smaller hole that it can't fit through. For all practical purposes, it justs sits on there and the pressure holds it on. It would also have bolts and/or sealant to keep it in place when not diving.
Pigo 3y ago
It seems likely much more effort will go into evaluating the failure than went into designing this submersible. It would be interesting to know how much more.
cityofdelusion 3y ago
Videos of the thing being built show a very sloppy epoxy job of joining the titanium ring to the carbon fiber hull. There seems to be no dust control, no climate control, and no attempt to make sure the epoxy is free of voids. I've honestly seen more care go into random youtubers making "river tables" than the epoxy job that went into this critical structural junction.

EA-3167 3y ago

I'm interested in seeing the skids, and learning about where they were found relative to the rest of the debris. Were the skids totally mangled, implying that they were attached when the implosion occurred? Is the debris field pretty contained, or very spread out? The former implies that the sub was close to the bottom, the latter implies that whatever happened occurred at a depth which allowed for currents and drift to spread the debris. If the skids are mostly intact then maybe there was some warning of trouble and the people in the sub tried to drop ballast. If the skids are destroyed then whatever happened is more likely to have been sudden and without warning.

KineticLensman 3y ago
Yes, this technique was critical to understanding the break up of the space shuttle Columbia as it re-entered. The debris track stretched across Texas and Louisiana, and in general objects to the west were those that had broken off first. Although things were confused because differences in ballistic number affected how far fragments travelled eastward, and in some cases modules broke off, travelled east and then disintegrated into sub-components.

causality0 3y ago

From everything I've read, the entire carbon section of the hull could be considered a weak point.

KineticLensman 3y ago
> the entire carbon section of the hull could be considered a weak point It is possible that the carbon fibre delaminated catastrophically, possibly as a result of stresses accumulated over multiple descents.
- sumo89 3y ago
  The only way the carbon tube hasn't turned to sea dust is if the porthole failed first. When carbon fibre fails it catastrophically fails. I'd put money on the joint between the ti and carbon being the weak point and once there's a chip in the carbon under that much pressure it'll disintigrate.
  - KineticLensman 3y ago
    fair points. Hopefully the investigation will nail the actual cause down.
ortusdux 3y ago
This is my understanding as well. I'm not surprised to see that the titanium piece survived unscathed. IIRC, the companies goal was to provide more cost-effective trips by transitioning from the standard solid titanium hull to a Ti/carbon fiber composite. This decision was controversial. They fired their head of QC after he strongly opposed the change. One key argument he made was that the design prevented ultrasonic inspection, which would mean that they could not monitor the health of the composite over time.
ajross 3y ago
To be clear, way too much of what I've read on this subject has been excessively confident tut-tutting with little to no real engineering expertise. People want to imagine that the simple use of carbon fiber here was the sole mistake, I guess because it's easy to understand and makes us feel smart. But come on, engineering is harder than that. Carbon fiber pressure vessels have a long history. Composite material failure modes are for sure more complicated than metals but they certainly aren't a mystery. We've been using composites for more than half a century, we know how they work. No, my gut says the root cause here (if we ever find one[1]) is going to turn out to be a more routine kind of engineering fuckup. It'll be just some weak point that the analysis failed to turn up. [1] And sure, if we don't then the "carbon fiber death cylinder" theory will just become part of the folk truth, I guess.
- mechhacker 3y ago
  There are non-trivial differences in designing vessels loaded with compression vs. tension, and the most experience we have is with internal pressure (e.g. tension). The analysis is very nonlinear and unfortunately not as widely understood as internally pressurized vessels. It requires a lot more analysis, troubleshooting, and expertise that is short in supply. Add the additional catastrophic failure modes for composites and there's a reason that industry experts warned against this.
  - ajross 3y ago
    With respect: that's exactly the kind of glib, drive-by analysis I'm talking about. What you're saying simply isn't true. People have been testing composites in pressure chambers and every load imaginable for decades. Yes, clearly they don't act like steel. But saying that it's impossible is just too easy. Just watch. It's going to be a routine design flaw, like they added a joint or a wiring hole in a bad place and didn't re-run the model. (Edit because this was misunderstood: I'm not saying what the failure is. I'm saying that we DON'T KNOW what the failure is, and that given the choice between makes-me-feel-smart pontification about materials and a routine design flaw in a prototype device, I'm going to bet on the routine failure every single time.)
TechBro8615 3y ago
Here's a video of hydraulic pressing carbon fiber to its breaking point: https://youtube.com/watch?v=xWTXeGiM8K8 https://youtube.com/watch?v=xWTXeGiM8K8
- mechhacker 3y ago
  Worth noting that because he's using a hydraulic press, the load isn't constant. It deforms, fibers break, and thus the load decreases as the deflection stays relatively constant in that time frame. With external pressure the load doesn't decrease. Fibers break -> deflection increases -> more fibers break. There isn't load relief as long as the vessel is still holding pressure.
  - sojuz151 3y ago
    It is even worse. The load increases with deflation
- N19PEDL2 3y ago
  Imagine how frightening it would have been to hear those crackles inside the submarine. (In reality, however, they had no warning of the impending collapse.)

mechhacker 3y ago

I'll speculate that because they are pulling pieces of the Ti domes without signs of carbon attached to it that the joint there wasn't stiff enough, and that carbon separated during failure without plastically deforming the Ti at the joint. But, that's speculation and I'd need to see more pics with more details.

_yb2s 3y ago
I would speculate that all of the carbon basically instantaneously liquefied under the pressure after a critical threshold of fibers failed, leaving the Ti untouched. I also want to mention that the large end caps you are seeing weren't bonded to the hull at all- there was a small Ti flange epoxied to the hull, and then the large domes were bolted to that. The front, with the window, was unbolted and rebolted every time for ingress/egress. There is no way that CF epoxied to thick Ti is going to make a strong enough bond that it will deform the Ti when it detaches. In this case, the geometry of that "glued" joint was such that it was forced together by the pressure underwater, rather than pulled apart- so it was not necessary for it to have any real strength other than for holding the end caps on at the surface, and filling voids to make it seal.
- mechhacker 3y ago
  I see. Thus that large Ti flange that was bonded to the dome didn't carry a load through the bolts significant enough to deform the dome either. Carbon gone in a flash and comes apart at that joint totally, Ti is left hanging in water.
- fullstop 3y ago
  I was hoping to see if that titanium flange was still bolted to the cap, but it's obscured by the tarp in the photo.
  - mechhacker 3y ago
    I just saw this: https://www.youtube.com/watch?v=uP4SrLhUHY8 https://www.youtube.com/watch?v=uP4SrLhUHY8 Those tarps may be covering what's left of the crew

_yb2s 3y ago

The hull imploded, not the window, but the shockwave probably pushed the window outward or shattered it, as it was only held strongly against moving inwards. People make a big deal about the acrylic glass not being rated for the depth, but I think it's almost impossible that it was the window, and think it must have been the hull: * The properties acrylic are well understood, and the strength of an acrylic disc is trivial to calculate, even if not "rated" you can easily calculate if it's strong enough * The window should fail visually with a lot of warning before rupture * If the window failed with the titanium end cap, there would be no reason for implosion of the carbon fiber hull The hull on the other hand is an irregular/non-homogenous part that is impossible to model accurately. There was no way to know how strong it was without extensive testing on a number of identical hulls- which they never did. The solidworks modeling method they used to give a ~2x margin of safety was nonsense as it ignores the realities of stress cycles, defects, etc.

TechBro8615 3y ago
How was the window "connected" to the hull? Would that be a weak point?
- Wojtkie 3y ago
  Deep sea portholes are usually cones of acrylic that seat right into a port in the hull [0] [0]https://www.researchgate.net/publication/340442621_Prediction_of_the_service_life_of_deep-sea_conical_acrylic_portholes_at_designing https://www.researchgate.net/publication/340442621_Predictio...
- EA-3167 3y ago
  This sub was a carbon fiber cylinder with two titanium end-caps, and the porthole was fitted into the forward end-cap.
- _yb2s 3y ago
  They machine a wide ledge on the titanium, so the window is pushed into the groove/sill under pressure, basically a large window that is covering a smaller hole that it can't fit through. For all practical purposes, it justs sits on there and the pressure holds it on. It would also have bolts and/or sealant to keep it in place when not diving.

Pigo 3y ago

It seems likely much more effort will go into evaluating the failure than went into designing this submersible. It would be interesting to know how much more.

cityofdelusion 3y ago

Videos of the thing being built show a very sloppy epoxy job of joining the titanium ring to the carbon fiber hull. There seems to be no dust control, no climate control, and no attempt to make sure the epoxy is free of voids. I've honestly seen more care go into random youtubers making "river tables" than the epoxy job that went into this critical structural junction.