[cockerpunk]

Originally Posted by Mahler9th

Perhaps of interest:

https://www.cbsnews.com/news/titanic-submersible-interview-transcript-with-oceangate-ceo-stockton-rush/

"The other thing with carbon fiber is, you can't cut holes in it. It doesn't like that. And so these [cables] go into the titanium, and that's how we get our data in. So that's how you get things in and out of the sub, is through the little port."

"So the key on that one is, we have an acoustic monitoring system. Carbon fiber makes noise. There're millions of fibers there. There are 667 layers of very thin carbon fiber in this five-inch piece.

It makes noise, and it crackles. When the first time you pressurize it, if you think about it, of those million fibers, a couple of 'em are sorta weak. They shouldn't have made the team.

And when it gets pressurized, they snap, and they make a noise. The first time you get to, say, 1,000 meters, it will make a whole bunch of noise. And then you back off, and it won't make any noise until you exceed the last maximum.

And so when, the first time we took it to full pressure, it made a bunch of noise. The second time, it made very little noise.

We have eight acoustic sensors in there, and they're listening for this. So when we get to 1,000 meters, if all of a sudden we hear this thing crackling, it's, like, "Wait, did somebody run a forklift into it? You know, has it had cyclic fatigue? Is there something wrong?"

And you get a huge amount of warning. We've destroyed several structures [in testing], and you get a lotta warning. I mean, 1,500 meters of warning.

It'll start, you'll go, "Oh, this isn't happy." (LAUGH) And then you'll keep doin' it, and then it explodes or implodes. We do it at the University of Washington. It shakes the whole building when you destroy the thing.

So that's our backup for the hull. And we're the only people I know that use continuous monitoring of the hull."

"And as I said, the warning is about 1,500 meters. It's a huge amount of pressure from the point where we'd say, "Oh, the hull's not happy" to when it implodes. And so you got a lotta time to drop your weights, to go back to the surface, and then say, "Okay, let's find out what's wrong."

"Like, over here, we blew this one up. So this is all a one-third scale [model of the Titan]. We were able to blow this up intentionally, to hear what it's like with our acoustic monitoring system. What we wanted to verify was, we can detect the carbon fiber failing way before it happens, so that you can stop your descent and go to the surface.

And that's what we found out here. So we now know what this shape sounds like when it's uncomfortable and right before death.

It's the loudest thing I've ever heard in my life.

When you go beyond 6,000 PSI in the test chamber that we were using at the University of Washington, they have to empty the building. Only essential personnel can be there."

"Well, obviously, the hull is critical. And if the hull being carbon fiber—and being a unique material, and having the potential for things like water intrusion—but if that hull goes, that could kill people.
You wanna make sure it's infrequent and also detectable. And one of the ways to do detection is this acoustic monitoring system.

So we bench-checked it against the system that Boeing uses. We made our own system that was much smaller and more sensitive, actually.

And then we also have strain gauges. So we look at the hull every step of the way.

And so we get to 1,000 meters and the hull's acting differently, the strain gauge goes off, or the acoustic monitoring system is showing a lot of noise, you call off the dive. You come to the surface. You find out what's wrong.

And so that's what we tested with all of these components, taking it to destruction, and also testing at the deep-ocean test facility, the entire sub, and being able to see: What does it sound like on its first dive, and what's it sound like now? And we can keep comparing it and making sure it's still putting out that beautiful sound—of silence."

acoustic emissions is a lot of things ... but a fail safe it is not. AE is a chaotic mess, and you dont actually know for certain what is happening.

also they are measuring the "warning buffer" in additional feet of pressure, but if its cracking, than its already failed and is weaker that it was before. ie, a better spec for that buffer would be a combination of time, and cycle count. it might not be 1500 feet of additional pressure (in this case the term would actually be head, but whatever), but if you left it at whatever pressure it was at when it cracked, over time, it would continue to crack likely. and if you removed that pressure, and added the same pressure again, you would see additional damage.

ie, the "warning buffer" he talks about here is a lot more complicated than he says it is. every single AE detection is a weakening of the structure.

EDIT: every interview with rush i read sounds like when elon billionaire-splains second year engineering school concepts like its ground breaking. rush sounds the exact same.