Hacker News Reader

Comment by ak_111 | Hacker News Reader

Why is it more likely that it immediately imploded rather than after a few days?

The latter is more probable to me as it has never been tested under water for more than a few hours at a stretch.

This will also be consistent with your banging hypothesis which was only heard after a day or two when it got lost. And also with the fact that no immediate sound disturbance was registered with the controlling ship.

I of course would very much hope it imploded way before any psychological trauma was inflected on the passengers, but sadly I think the most consistent hypothesis with the little data we have is that it got lost and went astray for around two days then imploded.

View on HN

There's little other explanation for why contact would be abruptly lost. The company said that the sub has multiple forms of ballast, including systems that could be activated manually, and ties that were designed to dissolve after 20+ hours in seawater, allowing the sub to rise even if the crew were unconscious.

> The latter is more probable to me as it has never been tested under water for more than a few hours at a stretch.

That's not true, they've successfully reached the Titanic several times over a dozen or so expeditions. Mike Reiss (a famous writer for The Simpsons) said his trip reached the Titanic but got lost for most of the ~8 hour trip, but that another group on his trip had several hours to explore the Titanic.

https://www.bbc.com/news/world-us-canada-65957709

https://unsungscience.com/news/back-to-titanic-part-2/

> The company said that the sub has multiple forms of ballast, including systems that could be activated manually, and ties that were designed to dissolve after 20+ hours in seawater, allowing the sub to rise even if the crew were unconscious.

is there more information on these dissolvable ties? or their redundant ballast systems?

From David Pogue's report on his trip in 2022:

https://unsungscience.com/news/back-to-titanic-part-1/

Excerpt from the transcript:

But what if the hydraulic system breaks? Well, then they have roll weights.

KYLE: Ah, so, we’ve got these weights here on the side, these are roll weights, we can actually roll the sub and those come off, and that gains us some buoyancy to come back to the surface.

These are pipes that sit on a shelf that juts out from either side of the sub, held in place only by gravity. If everyone inside the sub shifts their weight to one side, the sub tips enough to let these pipes roll off.

If that doesn’t work, there are ballast bags, full of metal shot, hanging below the sub.

KYLE: These bags down below, we drop those off using motors and electric fingers.

OK. But what if the electronics go out, and the hydraulics fail, and everyone inside has passed out unconscious?

KYLE: There’s fusible links within these that actually can dissolve and come back in time if it’s drop off.

Fusible links are self-dissolving bonds. After 16 hours in seawater, those bonds disintegrate, the weight bags drop off automatically, and you go back to the surface.

thank you, couldn't find the right part of the interview

Great, in theory. All the operator has to do now is to show the test reports for all of that. Should be easy, right? After all, those functions can be tested in comparatively safe depths, while being tethered to a surface ship.

> There's little other explanation for why contact would be abruptly lost.

You'd think - but past passengers have said that when they dove they lost contact as well.

Mutiple trip passengers said they lost contact every time.

So multiple anecdotes of loss of contact with the same vehicle without an implosion. The jankiness of this operation continues to be discovered bit by bit.

ETA:

https://abcnews.go.com/US/former-titanic-submersible-passeng...

I listened to Reiss's (the passenger cited in that story) account on his podcast [0], and it's somewhat ambiguous. There are periods of spotty communication and long periods of the sub just getting lost. But not comms system/transponder abruptly going out and not being heard from again.

David Pogue seems to concur with this [1]. He saw them lose track of where the sub was located (wrt to the Titanic wreck), but he said the support ship never lost the ability to communicate with the sub.

[0] https://bleav.com/shows/what-am-i-doing-here-with-mike-reiss...

[1] https://twitter.com/Pogue/status/1671524465736335366

In addition, note that despite losing contact while the sub was on its way down, they did not report it missing until it was overdue after the full mission length.

That indicates a certain blase, routine attitude to communication loss.

This is consistent with this video of a trip. Comms were lost, and the pilot even jettisoned some of the ballast in order to resurface. They continued the descent later though.

https://m.youtube.com/watch?v=RAncVNaw5N0

> That's not true, they've successfully reached the Titanic several times over a dozen or so expeditions.

This is more concerning, given that metal fatigue is an understood phenomenon (1).

(1) https://www.faa.gov/lessons_learned/transport_airplane/accid...

It's a carbon fiber hull, for which I think we know alot less about repeated stress events...

> It's a carbon fiber hull

...With titanium hemispheres at each end, so some sort of poorly understood titanium carbon-fibre interface.

I know first hand how hard it is to design carbon fibre preasure bulk heads, for aircraft with a much lower preasure delta (I wrote my first thesis about how to produce something like that). So, on the sirface, titanium makes sense. Using both, carbon fiber and titatium is just, well, not a good idea. Especially since I have the feeling this whole things wasn't properly calculated in the first place.

Immediate implosion if a far more likely explanation because give that we knew it imploded the most likely time would be immediately after the pressure vessel reached maximum or near maximum stress and it failed. The time actually is not the biggest factor; it is the number of cycles the pressure vessel has endured. An implosion after a relatively low number of cycles is consistent with past incidents with pressurization failures.

As was pointed out, the submersible is pretty large and its a essentially a large cylinder which induces stress in the middle. Usually you make deep sea submersibles out of a sphere or spheroid because you don't have this failure modes.

It isn't surprising it failed. They were strong claims that the company did do enough test, rejected safety advice and even had the CEO claim that at one point the titan wouldn't be able to reach the titanic because it (a model maybe?) had experience a pressurization failure. To say nothing of the materials they were using.

Just comically bizarre that anyone involved in the project thought it would somehow survive let alone be safe.

The most puzzling aspect, from what I understand from the experts, if it imploded at time and depth you are suggesting it would have been very likely to be registered by at least the controlling ship due to the noise. As you highlighted the sub is pretty large, imploding in a millisecond causes huge effects in the surroundings.

So saying it got lost and imploded after a few days solves two problems. Explains why there was no implosion noise originally, and that the banging heard afterwards was the implosion. Also it has a history of losing contact in its previous tours.

They mentioned in the press conference that it was very unlikely that the implosion would have happened after more listening equipment was in use on the scene. I believe there were sonar buoys dropped on Monday, so the implosion likely happened before then.

> it would have been very likely to be registered by at least the controlling ship due to the noise

Why would they be listening? They would be looking at a device that was communicating via SMS or some radar screen at best.

> and that the banging heard afterwards was the implosion

Every 30 min?

Sounds fun. If I was driving a nuclear sub in the Atlantic sea, I'd definitely find an excuse to pop over and play with the kids splashing in my pond.

> Explains why there was no implosion noise originally,

We don't know that. There's a chance a noise was recorded but it wasn't made public in order not to compromise rescue attempts.

It turns out that it’s exactly what happened.

Yes - according to the WSJ and an anonymous US official: https://www.wsj.com/articles/u-s-navy-detected-titan-sub-imp...

There’s a series of episodes of Smarter Everyday where the host goes on a US nuclear submarine and interviews the crew and shows some of the equipment. In one of the episodes, they talk to navigators, who explain that submarines can be invisible to sensors in certain locations because the ocean water is not uniform. Temperature, currents and salinity vary, so a sound could be attenuated in some directions.

"It imploded right away" explains everything.

"It imploded later" requires at least two failures, first of power/comms, and THEN of the structure itself.

Occam's Razor

Your occam is incorrectly structured. Power/comms is irrelevant to the equation in question, imploding, and doesn't add.

It imploded later requires only one failure (the structural failure), just as the relatively short time duration option requires only one failure. Right away is also a later event.

We have no way of knowing what its structural true condition was in terms of whether it was more likely to make it a very short duration or something more like a day.

As others have pointed out, there were multiple, redundant failsafes on the ballasts which would have led an intact sub to surface even if the crew were incapacitated or dead.

You're completely ignoring that the comms went out. All we know is that comms went out early, and debris was found later. What single failure would cause both of those pieces of evidence?

You're right, I think, but we also know that several ships heard sounds after the communication loss. You'd also need to account for that with your theory.

One ping only.

> Occam's Razor

is only good as a heuristic for finding which hypothesis to test first.

It's basically a scientist's "where there's smoke there's fire".

Or in this case, where there's a debris field...

No, it doesn't as it would depend on the root cause of the implosion, like whether or not it crashed hitting the bottom then imploded or imploded partially descended from it's target depth. Slamming into the ocean floor would point to other things than the structure itself being the root cause.

How does it explain why the support boat heard no implosion while listening for the 15 min pings?

Wouldn't an implosion be really loud?

Exactly this, "imploded straightaway" doesn't explain no implosion registered AND probable banging noise heard a few days later, so it doesn't explain everything.

"Lost then imploded" explains both by adding only a small extra assumption so by occam's razor is strictly superior to imploded straightaway as far as I can see.

If you ignore ballast failures and want it to be that way, sure

People are putting a lot of weight on the whole "one vessel heard a rhythmic sound while exploring". From what I've seen of these investigations the ocean is a noisy place and sometimes it gets mistaken for signal. We saw a lot of similar reports from the MH370 investigation.

My money is on simple catastrophic failure of the hull and it not being detected either because the private company is run by jokers who weren't listening for it or because they have been running around like chickens with their heads cut off because the CEO and paying customers just died and they don't want to have to report that to the family, government, media, insurance company, etc... Either explanation is plausible, but I'm slightly more inclined to go with the second simply because they were actively trying to communicate with the sub when it happened and it seems so improbable that they could miss it.

Loud is relative. You wouldn't hear it standing on deck of the support ship. And the hydrophones they were using for communications and pings were possibly (likely?) passed into an FFT, band-pass filtered to look for the expected frequencies of pings, and triggered on a signal spike in that range. I doubt they had somebody just listening to a straight up amplified signal straight from the hydrophones. Even if they did, someone unfamiliar with what they were hearing might not recognize it as an implosion event and attribute it to something else. And given the apparent attitude and methodology of the whole operation, it wouldn't surprise me if they didn't bother making recordings of the raw data. So it's entirely possible they wouldn't hear/notice an implosion event.

As for the banging sounds heard days later, there are a thousand possible sources of those, especially with a dozen or more other vessels on the scene. Anything from a buoy with a loose part clanking away, to some waves breaking and clapping on the surface, to a whale slapping its tail, to extraneous noise from one of the search vessels or its equipment, to other sources of sea life making a racket. It's a bit like trying to listen for the buzzing of a bee in a basketball stadium with a game going on. Any signal you get is far more likely to be something else, but when you have literally no better options, every signal, no matter how unlikely, is worth investigating. Then they report they investigated and found nothing, and in a bit of target fixation, people assume it must have been people in the sub. Understandable, but it's not necessarily the most likely/logical conclusion.

And in this case, the sub had systems for resurfacing even if power was lost (including automatically after a set amount of time). It's highly improbable that it was astray for days before it imploded. The only way this could have happened is if it somehow got stuck on part of the Titanic wreckage and was unable to free itself.

The sub had 7 redundant ways to surface (drop weights / ballast), several of which work without power, and one of which triggers automatically after ~20 hrs of exposure to seawater.

The only way it wouldn't have already surfaced on day 1 is if it got stuck on something (and lost power, unlikely), or it imploded.

What I don’t get is what factors could have caused it to implode now when it didn’t previously?

Material failure after a certain number of stress cycles.

Failure modes for advanced composites are less well understood than for traditional metals as well. The sub's pressure hull was also made out of three disparate materials joined together which adds additional complications. Carbon Fiber in particular is notorious for performing flawlessly until it catastrophically fails in an instant.

I know doing amateur rocketry pressure vessels work until they don't. Motor cases will gladly handle multiple launches and then on the 20th launch, explode. I think it's a matter of the metal fatiguing over time but I'm not sure how you measure the rate or severity.

Fatigue failure:

https://community.sw.siemens.com/s/article/what-is-a-sn-curv...

> Why is it more likely that it immediately imploded rather than after a few days?

That would mean it had two mostly independent failures (which is less likely than one).

It unexpectedly lost communications about 2 hours into its decent. That is a possible/probable point in time where something went wrong.

It’s also the point where they pass the depth that the porthole was rated for.

They got lucky with that until they didn’t.

They even fired a guy for whistle blowing about the porthole.

Are you telling me, that the porthole of a DSV, intended to dive to the Titanic, was EDIT: not: rated for the depth the Titanic is at? This whole operation is getting sketchier by the minute...

If you mean wasn’t rated then yes.

It’s a shitshow from start to finish.

They didn't want to spend the money to build a portal rated to 4000m.

https://techcrunch.com/2023/06/20/a-whistleblower-raised-saf...

  The filing states that OceanGate refused to pay for the manufacturer to build
  a viewport that would meet the Titan’s intended depth of 4,000 meters. The
  Titanic lies about 3,800 meters below the surface.

  The filing also claims that hazardous flammable materials were being used
  within the submersible.

> Why is it more likely that it immediately imploded rather than after a few days?

Weren't there a lot of boats and other resources listening once it was reported missing?

It's one thing for the tender to not hear an implosion when no one else was following this excursion. It seems harder to explain no one hearing an implosion when a growing number of resources were trying to detect any signs of activity from the sub.

If the stories about some parts only being certified to 1.3km instead of 4km are true then it was probably operating closer to the yield point than ideal. My guess is that metal fatigue started to become an issue and it failed too quickly for anyone to react.

If the were too cheap to design and build it properly then they were too cheap to check for wear and tear properly.

Fatigue occurs after oscillating stress levels (far) below the strength of the material. So - stuff will break under low stress, if you apply it often enough. It must oscillate, otherwise it will never break. Some material, like aluminium, have very low minimum thresholds.

A sub is subjected to static stress mostly. Quite a lot actually at these depths. And having it certified for 1.3 km hints to over stressing.

If the sub imploded after the search had begun, would the search vessels have detected the sound?

The USCG admiral coordinating the search effort and doing all the press briefs said that yes, they most likely would have heard it if it happened after they got the listening buoys in the water.

Reply on news.ycombinator.com