Tourist submersible visting the Titanic is missing 101

[waross]

They wouldn't have time to say;
"Oh sh#t"
RIP

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[boofi]

From James Cameron:

"“This OceanGate sub had sensors on the inside of a hull to give them a warning when it was starting to crack,” he told ABC News. “And I think if that’s your idea of safety, then you’re doing it wrong. They probably had warning that their hull was starting to delaminate, starting to crack. ... [W]e understand from inside the community that they had dropped their ascent weights and they were coming up, trying to manage an emergency.”"

It sounds like they had SOME kind of monitoring system for the hull. Maybe too little, too late.

That is my thought as well. They said it broke into 5 major pieces. That would be 2 end caps, the tail section, the landing struts, and the cylinder. For all of these things to separate I believe the cylinder they're all attached to would have to fail.

Did they mention seeing the carbon cylinder? I thought it was just a debris field with the fairing, the two end pieces (end bells), and various debris.

Shells under external pressure are often buckling critical long before they reach the material compression limit.

https://youtu.be/CJSaAjJMK5w

ASME code may be considered overly conservative, but you really really do not want to operate near the structural buckling limit in any design.

I wonder what the design process really looked like for this vehicle. Did these younger engineers just bust open Roark's and call it a day? (I say this as a younger engineer )

 

[Tomfh]

If the viewport failed first, wouldn't the sudden in-rush of water at 400 atmospheres over-pressure lead to some dramatic "water hammer" effects

Yes it would have been beyond violent. The pressure spike once it filled would have torn it to pieces.

 

[waross]

I saw a remark that the viewport would fail optically long before it failed structurally.
There was also a comment that the viewport was laminated.
Is it reasonable that the laminations could deflect quite a bit before structural failure, but even a small deflection may affect the optics of the inter-lamination adhesive?

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[GregLocock]

I think direct crush of the composite cylinder is the most likely failure to result in the spontaneous ejection of the two ends. Viewport failure would just fill the thing very rapidly.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[dik]

Young engineer... What's Roark's?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[davidbeach]

Wasn't there a time in the late '80s when NASA thought that all the 50 year old white guys were being too stodgy as well? Nah, those O-rings are fine...

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[boofi]

Roarks formulas for stress and strain. It's actually a really popular reference for all kinds of stress calculations. You'll see it referenced all the time in engineering calc packages. I was being a bit tongue in cheek, it's a legitimate source of information. An excellent reference but not a full substitute for sound engineering judgement; e.g. not using a material that could have issues in compression, bonded with questionable methods to a titanium end cap.

That reminds me, does anybody know what kind of adhesive they used to bond the carbon fiber to the titanium? Was the titanium also heat shrunk onto the hull?

 

[enginesrus]

Why haven't any pictures of said parts been released?
I posted a video above and it shows the titanium ring being glued to the main body, above poster asked how it was attached.

Around the 4 minute area.

https://www.youtube.com/watch?v=4dka29FSZac

 

[SWComposites]

Some kind of paste adhesive. The process in the video looked sketchy. Rings were not shrunk on.

 

[boofi]

I wonder if that was a promotional video or if that really was the final assembly. I would think something like this would require a much more controlled environment than what is shown on there.

 

[SWComposites]

Well, it should have been a controlled facility. But apparently they knew better.
This whole thing reeks of hubris and start-up mentality and having/wanting to do things on the cheap.

 

[SWComposites]

Anyone interested in the art and science of shell buckling should check out the wealth of info here:

https://shellbuckling.com/index.php

Buckling of unstiffened cylinders are particularly sensitive to imperfections.

 

[TugboatEng]

I was curious about the acrylic window so I googled some things about acrylic and fatigue. My experience with acrylic is that it is sensitive to damage from creep. As a result I will not use interference fit screws in the material.

I did find this interesting document that explains the failures of Plexiglass 55 (cast acrylic).

https://apps.dtic.mil/sti/citations/AD0407801

It is a long ways beyond my own understanding.

With the evidence available today, despite the concerns about the acrylic window, there was a hull failure.

 

[LittleInch]

I can't find it now but I'm sure I read from Cameron that the deep sea "community" thinks that some of the weights were dropped and the sub was trying to rise to the surface.

Probably difficult to know now and not sure how they knew but as probable as anything. Horrible to think that they knew it was failing just before it finally went. Ugh.

The forces when you think about them are just huge. About 2.5 tonnes per square inch, on every square inch and this thing wasn't small. I get about 7,500 tonnes axial compressive force from the ends alone. Have I got that right?




Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[waross]

Daily Mail

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[SnTMan]

"[W]e understand from inside the community that they had dropped their ascent weights and they were coming up, trying to manage an emergency."

Maybe they did have time to say oh sh*t...

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[KevinK2]

Just read this vessel carried 46 passengers in 2021 & 2022, min of 9 trips with 5 passengers, 11 trips if there were 4 passengers.
The sandbag theory could be based on the location of the bags, vs the wreckage.

The buckling link was informative for thin walled shells of revolution.

Kevin Kelleher, P.E. (retired)
Internal Mechanical Eng'g Consultant
DuPont ESD Specialists

 

[EdStainless]

Other than the Titanic this section of the ocean floor is very flat and clean.
And this was about 1mi from the Titanic.
Yes, the US Navy hears everything. Their report was from very shortly after communications was lost on the decent.
That would be before the vessel was reported lost.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[boofi]

From another discussion on reddit:

https://www.reddit.com/r/AskEnginee...nd_yield_on_past_occasions/jp81ykl/?context=3

Bristol Composites Institute | Realising the Potential of Carbon Fibre Composites in Compression (2017)

"It appears to be due to the stress-strain curve of carbon fiber being very different from the stress-strain curve of the matrix of epoxy that holds it together. For that reason, "yield" is misleading, because carbon fiber has a non-linear stress-strain curve. At high loads the divergence between the stress-strain curve of carbon fiber and that of epoxy is quite a lot, and this causes the two to unbond from each other, at a microscopic level, which might not even be detectable if it is all held together by other fibers around it.

When the two stress strain curves begin to diverge, it causes even more delamination and concentration of force on the remaining bonded parts, worsening the mismatch and accelerating further delamination, resulting in a feedback cycle that ends in failure.

For this reason, carbon fiber composite materials fatigue and fail with repeated cycling to extremely high loads that take it into the part of the stress-strain curve where this divergence happens, so "stressed beyond yield" should be "stressed beyond onset of microscopic delamination"."

@SWcomposites