3 ms·

> The big one is designing the sub to travel to 95% of the expected maximum depth, leaving a 5% margin of safety. Sorry, but this does not follow. I believe yo

> The big one is designing the sub to travel to 95% of the expected maximum depth, leaving a 5% margin of safety. Sorry, but this does not follow. I believe you are confused about the fact that the Titan was designed to go down to 4,000 meters, and the Titanic lays at 3,840 meters. And because 3,840 is so close to 4,000 you believe they had no safety margin. Do I understand you well? But the thing is when you design a submarine which can go down to 4,000m you don’t design a hull which instantly implodes at that precise depth. You add a safety margin on top of your design requirement and that is what you design for. And then using calculations, simulations and tests you convince yourself that your submersible is still safe even at that extra safety padded depth. I’m not saying that this is how they designed it. After all it seems their engineering culture was deficient in several ways. But you simply can’t conclude that they had no safety margin because they dove close to their design limit.

bagels 3y ago
Safety factors are in terms of the yield stress. So, if they had a safety of factor of 3, this means that they should be able to withstand 3x the expected stress. Pressure is linearly related to depth, so it should have been designed for 12,000 meters with a reasonable safety factor like that. Of course, this doesn't take in to account sharp impacts, which, especially for carbon fiber, dramatically change the strength requirements. Some ballpark typical numbers for safety factor: spacecraft ~1.2, airplanes ~1.5, automobiles ~3
beowulfey 3y ago
Yes but you also don’t want to operate so close to the safety margin. If you need to travel to 3800m, you design a sub to safely withstand 1.5x that. And then agreed that yes it should not implode right if it were to reach that depth. At least this is how I remember learning it, which I admit was a long time ago.
- krisoft 3y ago
  > don’t want to operate so close to the safety margin. How do you know they were operating close to their safety margin?
  - beowulfey 3y ago
    I think I’m misinterpreting what “rated for 4000m” means. My impression is this is the maximum safe depth the sub is capable of reaching without damage. And that they are bring it regularly to 95% of that depth. My understanding of safety margins, as I was taught, is that one should design such that you have a safety margin of 1.5x the intended use case. In other words, for 3800m expected frequent usage, it should be able to reach a depth of 5700m. Alvin, for example, can reach a depth of 6500m. But you’re right, we don’t know what that 4000m means, really. I’m just assuming it means safe depth without damage.

bagels 3y ago

Safety factors are in terms of the yield stress. So, if they had a safety of factor of 3, this means that they should be able to withstand 3x the expected stress. Pressure is linearly related to depth, so it should have been designed for 12,000 meters with a reasonable safety factor like that. Of course, this doesn't take in to account sharp impacts, which, especially for carbon fiber, dramatically change the strength requirements. Some ballpark typical numbers for safety factor: spacecraft ~1.2, airplanes ~1.5, automobiles ~3

beowulfey 3y ago

Yes but you also don’t want to operate so close to the safety margin. If you need to travel to 3800m, you design a sub to safely withstand 1.5x that. And then agreed that yes it should not implode right if it were to reach that depth. At least this is how I remember learning it, which I admit was a long time ago.

krisoft 3y ago
> don’t want to operate so close to the safety margin. How do you know they were operating close to their safety margin?
- beowulfey 3y ago
  I think I’m misinterpreting what “rated for 4000m” means. My impression is this is the maximum safe depth the sub is capable of reaching without damage. And that they are bring it regularly to 95% of that depth. My understanding of safety margins, as I was taught, is that one should design such that you have a safety margin of 1.5x the intended use case. In other words, for 3800m expected frequent usage, it should be able to reach a depth of 5700m. Alvin, for example, can reach a depth of 6500m. But you’re right, we don’t know what that 4000m means, really. I’m just assuming it means safe depth without damage.