Tourist submersible visiting the Titanic is missing Part 2 69

[dik]

Apparently it can. In addition, it can debond and lose compressive capacity.

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

-Dik

 

[LittleInch]

So apart from being apparently towed behind the vessel on the latest trip I also saw somewhere that it was regularly transported overland from the Atlantic coast to their base in Everett, Washington state.

Plenty of room there for the odd knock or point damage which wouldn't show up.

Didn't realise this is basically a suburb of Seattle - no wonder they did some sort of work with Boeing.

Debonding under compression didn't take long to find things like this

https://www.degruyter.com/document/doi/10.1515/secm-2015-0057/html?lang=en

Doesn't look good. basically gradually fails until "it completely disbonds" once you've reached the critical limit.

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

 

[3DDave]

AFAIK they did no work with Boeing - at most they bought material that was scrapped from Boeing without informing Boeing of their plans.

Considering that this effort parallels Theranos - the leader wants to break out of the mold, fires or ignores everyone with experience in the field, opting instead to surround themselves with ignorant (not stupid, but uninformed) Yes people it is clear there were steps taken to keep the end use from anyone with relevant knowledge.

The difference is the leader at Theranos was smart enough not to participate in the use of the product. This time it was self-correcting.

 

[SnTMan]

^Agree.

Recall this post from Part 1: FacEngrPE (Mechanical)1 Jul 23 19:34

Refers to an article from Design News about successful submersibles made from CF composites.

OG's problem was apparently not so much use of CF composites as such, but rather with design and manufacture of the same. Very different corporate cultures.

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

 

[SwinnyGG]

Does CF "bruise" n the same way GRP does?

Yes.

It's a common failure mode for composite tanks. There's a safety measure to catch them - in internal delam will result in pressure moving between plies, and will 'blister' the tank wall before failure, so they're easy to catch by visually inspecting the tanks.

For a tank wall in compression, I would expect any delaminations within the wall layup to result in visible blisters inside the tank. In theory you'd be able to catch them via visual inspection, but for the OceanGate design, you would need to disassemble the pressure vessel and pull out their internal tank liner to inspect the tank. I haven't heard conclusively either way but given the general safety culture that was on display in other areas of the business, I have a strong doubt they were doing that after every dive or with any regular frequency really.

2. What resin was used and how was it cured? My experience with composites is that thick laminates are susceptible to thermal runaway with a risk of spontaneous combustion. Maybe they used room temperature curing resins, which in itself has inherent dangers.

I don't know that we have specific information on the resin, but there was some info about the manufacturing process in the previous thread. They used a very long, low temperature cure. It was 7 days at 125[sup]0[/sup]C, or something to that effect.

 

[SwinnyGG]

So, was the failure so violent as to rip an end ring free from the titanium hemisphere or did the recovery crew remove the bolts while onboard the recovery vessel to make subsequent handling safer and easier?

I have a hard time seeing any failure mode where the end bells stay on - either the CF hull buckles and the end bells are blown off, or the acrylic window failed and the end bells were blown off. There would have been a huge pressure spike due to water hammer, an order of magnitude beyond the pressure the sub was design for and also of the opposite sign.

 

[LittleInch]

For normal thickness material I can see that but when it was what 5 inches / 125mm thick maybe those normal bruises didn't show, plus the internal and external "cladding" would have reduced the visibility of this.

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

 

[SwinnyGG]

For normal thickness material I can see that but when it was what 5 inches / 125mm thick maybe those normal bruises didn't show, plus the internal and external "cladding" would have reduced the visibility of this.

Certainly possible. As I said I doubt they were pulling out the internal liner to inspect the tank wall, and even if they were - I agree it's possible that the telltale blisters may have been difficult or impossible to detect visually.

I would expect that even at this wall thickness they would be very easy to detect with ultrasound. But again, continuous safety inspections weren't a part of their process as far as we know.

 

[MechinNC]

I've read a fair amount about CF testing in compression. Some good stuff linked here..

But what about fatigue response in compression? Seems like a vital bit in this failure.

I could look it up, but I'm being a bit lazy!

 

[1503-44]

I calculated the 29,000 psi compressive load using area of 11ft2, of the 5 inch wall. If it's 50mm thick at the joint, that's going to be about 2.5x higher stress on the face of the joint. 75ksi +/-
I hope the titanium was acting as a retainer ring.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[SWComposites]

But what about fatigue response in compression? if there are no delaminations, then fatigue response is typically good in compression. However, if there are delaminations, then they can grow and eventually buckle under compressive loading.

 

[SwinnyGG]

The fatigue point is not something we should dismiss IMO, but I'm not sure if less than 10 full cycles to rated depth really would qualify as a fatigue failure in my mind.. Reads to me like gross overload.

 

[dik]

I don't think it's really fatigue, but rather damage to the material on loading.

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

-Dik

 

[IRstuff]

Based on the videos posted, it looks like they also made a rather huge design change between the first and second cylinders, since the first had a rather conventional looking "spool" upon which the CF was laid, but the second one was flangeless, presumably to make it easier to lay the longitudinal plies, but that possibly weakened the connections between the cylinder and the endcap mounts.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[GregLocock]

Pretty colors for management

https://www.youtube.com/watch?v=TT-Z_3xfFeE

Cheers

Greg Locock


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

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

 

[SnTMan]

The layup has to be done on some kind of mandrel. It either becomes part of the finished cylinder i.e. a thin shell, or the mandrel has to be easily removed from the cured layup. Segmented or something.

An integral mandrel would prevent any sort of inspection of the inside surface of the finished cylinder. It might hinder of any NDE of the same.

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

 

[rodface]

@snTMan

what would be typical material/thickness for a mandrel that is intended for leaving-in-place? A removable mandrel seems very interesting, what is it coated with to prevent sticking, is the removal destructive?

 

[SnTMan]

rodface, I can't really say, but maybe a 1/4" thk shell, backed up by a more substantial mandrel of some sort would likely do. One of the layup videos in the Part 1 thread seems to show some bolted end plates, no clue as to further details.

Mold releases are routine, but imagine trying to remove one cylinder in intimate contact with another with no draft, etc. One being a kind of fragile composite. Ir could prove impossible.


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

 

[Brian Malone]

I have no experience with fabrication of a composite structure of this size but a segmented shell would not be difficult to make or use. The zero draft becomes a non-issue. Is a fully cured 5 -inch thick wall carbon-fiber tube really a 'fragile composite' for handling and working with for the conditions of manufacture? I cannot imagine this tube could not have cores removed without issue. At one atmosphere and room temp, the tube is highly capable for any load cases, no?

 

[rodface]

@SnTMan

Would be fascinating to learn the exact procedures that were carried out, but unfortunately I am sure that the parties involved will do their best to take this information to the grave.

I have been thinking about this disaster a lot; it seems that a promising avenue of submarine engineering development has been chopped off at the knees because of the sloppy work done here. Given enough funding, R&D, experimentation, the carbon fiber submersible could very well be realized. But thanks to this incident, it may be another 50 years before these concepts come to fruition through some means.

It's almost like witnessing a disaster that stalled the onset of powered flight by several decades. Imagine how different the world would be if Kitty Hawk was delayed until the 1940s.