Titan sub noise heard on board support ship. 2

[LittleInch]

Just watched a new documentary by the BBC and others like Discovery so if you're interested look out for it. Implosion: The titanic sub disaster.

Key new part for me is that they released video of the moment a loud "crump" was actually heard on board the support vessel apparently through the hull with Rushs wife Wendy doing the monitoring of the sub and her saying "what was that?". They then got a message from the sub which must have been somehow delayed saying dropped two weights which confused them. It is pretty chilling to listen to alright. That and the fact that they were thought not to be at the ocean floor implies they were getting some warnings.

This is a clip but there are others.

Not much we didn't already know, but they have noted dive 81 (out of 88) where there was a very large bang heard which they reckon was the point at which the hull was on its last legs after a major internal failure.

The Discovery documentary maker back in 2022 was very unimpressed with the whole thing and persuaded the company to can their documentary as he was convinced it would end in failure and didn't want to be seen to promote the operation.

Looks like it's on Discovery pretty soon. If you're interested, it's pretty good and not that big into the human interest side, but does have input from the wife of the man and his son that died which is quite moving.

How to watch 'Implosion: The Titanic Sub Disaster' - Stream the thought-provoking documentary online

Scientists explore how the Titan submersible implosion was more than just a tragic accident

www.tomsguide.com

 

[Tomfh]

even a partial set of the test prescribed by the ABS - just one of many certifying bodies, the tragic outcome probably could have been avoided.

Agree.

Cyclic testing is what it needed in my view. Cycle the pressure thousands of times and see what the cracks do. They’d actually already done that indirectly with the first hull; but turned a blind eye by the looks of things to all the horrible cracking that was occurring.

 

[IRstuff]

Ultimately yes, SR was responsible for the culture and safety but the employees turned a blind eye, enabling it as long as possible.

Not strictly true; they knew quite early on that any dissent was going to result in immediate dismissal. Their hope was that they would be able to sway Rush to a safer conclusion, but that was not going to happen. While the Netflix documentary is heavily biased by the former employees, the facts are that multiple top management personnel were canned when they dissented, and SR did not do much to protect the submersible, even to the point of leaving it in Canada, exposed to the elements in winter, because that was cheaper than doing the right thing.

 

[TugboatEng]

Carbon fiber and titanium are not inherently affected by weather. The effect of moisture on the bonds to metals that passivate is a relative new and unknown subject. The article I quoted above is from 2017. Maybe it wasn't so much arrogance but ignorance.

 

[RolMec]

just a proposal to get some lessons learned from all this: if anyone here plans to do something new&groundbreaking, with passengers on board, then to do it from first principles with the balls to take in a / some devils advocate and to never rename a new and unknown subject like Ti-CF-bonding to "glueing" or so in order to dodge a snag

 

[TugboatEng]

RolMec, I'm annoyed that everyone is hyper focused on CF tube and ignoring the CF to titanium connection that is apparently critical to the hoop strength or the tube. I'm using terms like gluing for attention when I should be using hydration to describe the problem. Hydration isn't interesting enough to attract attention.

 

[Tomfh]

I'm annoyed that everyone is hyper focused on CF tube

Yes, the joint is a question mark and might’ve contributed. But we’ve got clear evidence the carbon fibre hull was cracking badly, and those cracks were getting worse with every dive. That’s not speculation, that’s documented. The previous hull was cracking to pieces, far from the joints. So while the joint might have played a role, the main failure path still points to the hull itself.

 

[GregLocock]

I'm withTugboat on this, designing composite to metal interfaces is an order of magnitude more difficult than designing composite tubes. The cracking sounds could be composite failures or joint failure or both. The bell that is being struck by these impulses is the same in either case.

 

[XR250]

Why is that joint so critical? Seems like the end caps would simply be held in by hydrostatic pressure? What purpose do they perform besides preventing leaks and keeping get ends from failing off out of the water?

 

[LittleInch]

It's the fact you have two different materials joined together subject to all sorts of differential forces including bending, thermal contraction / expansion, water pressure trying to get into the joint, end cap force creating both direct axial load and some bending which may change as the depth increases, some flexing of the titanium ring, etc etc.

Any joint is a weak spot and when you're joining two radically different materials such as titanium and Carbon fibre over quite a short distance, there could be very high stress concentrations in the CF shell and onto the glued joint itself.

We might only be talking 1 or 2mm but the bending moments and stresses in something that big going to that depth would have been huge.

I don't know directly which of those two elements would or could have been the "stronger" or the more flexible of the two, but I do recall seeing FEA type drawings showing the CF tube being squashed in the centre more than the end caps thus creating bending moments and probable localised stress concentrations at the join.

In any event, the clear information that was being given by the submersible during its dives showed that the fibres were constantly being broken and the test failures showed this. The complete lack of type testing for multiple dives, FOS testing and pure ignorance or hubris about people raising objections and barriers is something to consider the next time a new way of doing things is mooted.

No one really likes giving the client / the boss bad news, but sometimes you just need to triple check your data and then deliver it, having taken suitable copies first and another job to go to... I've had to do it a few times and its not good for business for sure... One time it was a new way of making pipes and one of the suppliers pointed out in a meeting with some investors that you could / they had made a few test metres ok, but when you wanted to creates hundreds of metres, there was a physical mismatch issue inherent in the design which caused it to lock up when you went more than that, which had been ignored / not seen by the inventor. That was a meeting I'll remember for a long time.

 

[RVAmeche]

I think one of the core issues with classing is they don't have any rules for submarine carbon fiber construction so it wouldn't have been able to pass? That seemed to be one of OceanGate's primary claims, outside of saying they're useless and just a paperwork exercise.

 

[RVAmeche]

Also the whistleblower protection or lack thereof was comical. What exactly is the protection if they can still just sue you 6 ways from Sunday to make you backdown and withdraw the complaint?

 

[IRstuff]

No one really likes giving the client / the boss bad news, but sometimes you just need to triple check your data and then deliver it, having taken suitable copies first and another job to go to...

That only works if your client is rational and not full of themselves. The Netflix doc. points out numerous instances, notably, one where OG paid to get outside analysis of test results and got the bad news and Rush simply stopped working with them. Ditto Lochridge, who was immediately fired after submitting his safety assessment. Likewise, despite his selling of the acoustic and strain sensors as a safety measure, Rush ignored all the clear indications that fibers were breaking and claimed it was "seasoning" of the CF hull. He was also warned that leaving the submersible in Canada for the winter was likely to result freeze/thaw damage and ignored that warning as well.

It's pretty clear that Rush was someone who would ignore any bad news and would stop doing things that would result in bad news for his pet theory; there's simply no way to get through to someone like that. This is where too much of the bullheadedness needed to succeed is a terrible thing.

Regardless of what exact mechanism resulted in the hull failure, the likelihood of the disaster was absurdly high.

 

[IRstuff]

I think one of the core issues with classing is they don't have any rules for submarine carbon fiber construction so it wouldn't have been able to pass? That seemed to be one of OceanGate's primary claims, outside of saying they're useless and just a paperwork exercise.

That was Rush's dodge to avoid classing, but it's pretty simple; Rush had zero proof or analysis that his submersible was robust enough for the dive depth, so he knew it couldn't get classed at all.

 

[RVAmeche]

Agreed. But my understanding was the classing agencies don't even have a set of standards they could've applied to the vessel to class it if they wanted to. Not sure if OceanGate had done proper engineering/testing to show that it was good for a certain number of dives if they could've gotten certified or if the CF construction alone would prevent it from the start.

 

[LittleInch]

Agreed. But my understanding was the classing agencies don't even have a set of standards they could've applied to the vessel to class it if they wanted to. Not sure if OceanGate had done proper engineering/testing to show that it was good for a certain number of dives if they could've gotten certified or if the CF construction alone would prevent it from the start.

I don't think that is correct.

If you read post #77, they could easily have applied Appendix D which apparently

"..but a designer can apply for creating an ASME PVHO Case for the use of non-standard materials and the testing and results would be submitted for review and approval."

So like a lot of codes there are approved materials and ways of doing things, but there is a way to use undocumented materials or designs, but it take a LOT more effort ( and time and money). OG and SR just wanted to bypass / short circuit / ignore all established norms about how to deal with innovative designs and materials.

 

[RVAmeche]

Ahh that makes a lot of sense

 

[Brian Malone]

The US Coast Guard document is here:

https://www.news.uscg.mil/News-by-Region/Headquarters/Titan-Submersible/smdsort32560/publicationdate/

Doc.CG-104 ABS

 

[TugboatEng]

IR, the FEA diagrams showing the tube bending in the center are assuming the glued joint to the titanium dome has not failed. The titanium dome has an inner lip for location in the CF tube. The lip was sheared off in the implosion, not bent. I read that as the failure at the joint being caused by shear, not moment.

XR, he hoop strength of the tube is insufficient without the titanium dome. Any failures of the connection between the CF hull and the titanium ring will cause the CF hull to collapse.

 

[IRstuff]

FEA also assumes homogenous and uniform hull, which we already know is not the case. All those pops and bangs represent damage to an already nonuniform structure, so it's quite likely that the hull implosion collapsed and pulled the the hull and the adhesive from poorly prepped surfaces in the domes

As I said recently, the actual failure mode is irrelevant to what Rush did, or did not, do in his insanely poor execution of what might have been a step forward in undersea exploration.

 

[btrueblood]

"It's the fact you have two different materials joined together subject to all sorts of differential forces including bending, thermal contraction / expansion, water pressure trying to get into the joint, end cap force creating both direct axial load and some bending which may change as the depth increases, some flexing of the titanium ring, etc etc."

Even with more-or-less homogenous and predicatable materials (steel, titanium), there is a large difference in deflection vs. applied pressure at the joint between end cap and cylindrical shell; hence, making that joint predictable and safe is not simple.