quote:

Not defending this guy, eh? And really, why wouldn't you?

Probably the same reason you wouldn't.

quote:

If all the proper testing and certification was done as you say

What I have said is that the carbon fiber tube was tested and certified. I have gone on to correct blatant inaccuracies some members have stated as fact when it came to subsequent testing.

quote:

why not defend the CEO then for putting paying passengers inside his submersible?

Because I'm not defending him or his submersible. In fact, I continue to point out the report of the viewport being rated for 1/4th the pressure required in contrast to the tube which was rated at twice the pressure.

And just like safes that are certified, the manufacturer has no control over what a third party does with it once it leaves their possession. Any modifications or wear and tear is the responsibility of that third party.

The proof is in the pudding. In this case the pudding killed 5 people. That's enough to know that this design was not suitable for the mission.

It doesn't matter who certified it to do whatever, it still killed 5 people.

Might carbon fiber be good to go with a different design? Maybe, idk because I'm not a cf engineer.

So he bought a tube from a company that designed it in a computer and tested scale models to show it could withstand the static pressure requirement. I see no info in any of the posted quotes for cyclic load testing to full design pressure until failure to know what the cyclic limit is. And I wouldn't trust anything less than multiple full size samples tested because of the potential for variation in the winding process. Say one fails at 20 cycles and one at 40 cycles due to variation in manufacturing. Hard to know with a sample size of one.

When you cannot afford to build and test full scale prototypes to failure, you have to use time proven and validated design principles based on the whole of the historical experience with actual examples that have been successful in the field.

And that's for everything in this case - the tube, the bonding method of the endcaps, etc.

What keeps getting missed is the entire PROCESS:

Design, design validation testing to failure under static and cyclic loading, manufacturing, manufacturing validation testing to failure of multiple samples, slow ramp up in service at progressive depths until full service depth is reached - all UNMANNED. Slow ramp up of manned missions at progressive depths until service depth is reached. All the while inspecting for damage/cracks after every dive (probably with X-rays and/or ultrasonics), base on KNOWN cyclic failure data from previous testing.

This is how we Engineers (M.E. U of IL 1994) do things. A similar process is followed for most complex engineered things. Cars, engines, aircraft, spacecraft, etc. When things are so big, costly, and/or complex you can only make one and can't test to failure, you have to be VERY conservative in the design and construction to make for damn sure it never fails.

But all this was too costly and time consuming for Rush. And he knew better than all those old white experienced submariners, until he didn't know better.

quote:

Originally posted by Fly-Sig:
The proof is in the pudding. In this case the pudding killed 5 people. That's enough to know that this design was not suitable for the mission.

The proof is that 5 people were turned into vaporized pudding in a millisecond.

Fiber and carbon fiber wrapped cylinders have been used for a long time now to contained pressurized gas. They are quite successful but have limits.

The dynamics of keeping pressure in, is a lot different than keeping pressure out.

If he truly felt he found some ground breaking technology that would allow this to happen he should have built several. Send them to depth multiple times packed with sensors. Every so many dives you take one of the test models and break it down, cut it apart, inspect everything for any evidence of failure. If you complete 150 dives with the last one and 125 dives with the predecessor and perhaps 100 dives with the one before that and have no evidence of structural failure then I'd say your design is probably good for 75 dives to 80% of that test depth.

Guess what... that is expensive but then again true development is NEVER cheap if it's done right.

quote:

Originally posted by 229DAK:

quote:

All three of those points were engineered and certified

Other than Rush, certified by whom? He appeared to have dodged any legitimate certification process, in the name of innovation, etc.

At the very least, you can get certification from any number of country's Board of Shipping. The point of such is so there's an accepted standard, where insurance coverage is available (Lloyd's), which would help indemnify the owners, operators and any passengers. Secondarily, if the owners wish to sell the craft, they've got evidence that it was capable to function as specified by a third-party; negligence notwithstanding.

Late last year, ultra-deep diving submersible DSV Limiting Factor, its support ship and accompanying gear was recently sold to a private ocean research company. Without the certification, the transaction probably doesn't happen or, is chock full of escape clauses. Plenty of video (I posted one on pg.9) of how it was built and the efforts that went into its operation, which included plenty of 50-something aged white guys.

Rush was a fool, a willful ignoramus, who intentionally tossed aside all previous research and development, in an effort to 'disrupt the system'. He wanted to shove his innovation into the face of the 'deep-diving industry' which kept telling him he could't do 'this or, that' because of engineering, or, physics or, material science or...experience . This is the mindset I found around a lot of the tech industry as the dot-com era died and the tech-boom grew from its ashes. All these guys with big bank accounts, wanting to get into their passion projects outside of tech, wanting to 'DISRUPT the industry'. Well, he did it, this may in fact push a new chapter of standards into international regulatory orgs for submersibles

I’ve only been able to follow a bit of this as I’m traveling, but there’s been quite a few incorrect and partially correct statements about CFRPs (carbon fiber reinforced polymers).

My background is Aerospace Engineering and my MS and PhD were on composites related topics. My MS was on a portion of a technology to try and detect damage in composites and my PhD was related to stopping cracks from growing and coalescing. My advisor was a preeminent composites manufacturing expert and I ran his composites manufacturing lab while I was a student. I also had a similar role for the Army Corps of Engineers when I was a student. I tell you this to give you an idea of the depth of my knowledge and have an informed opinion on how to value what I say. I have worked in the oil and gas industry for 15+ years, so not a composites related job and I would no longer call myself an expert but I am knowledgeable. I am NOT a marine engineer or know a ton about it, but I understand the very basics.

As others state, FRP (fiber reinforced polymers) are strong in tension and less so in compression. The thing I haven’t seen is that composites can be tuned to have different layers in directions to best make use of the material. This is difficult to achieve for “collapse” loading, but it can be improved from every fiber in one direction. I’d like to see their lay-up. The info someone else found did surprise me as they had a relatively low temperature cure cycle and void content was reasonable, but not super low.

One of biggest challenges with using composites in high collapse pressure/deep sea applications is the manufacturing of thick composites. There are technical challenges with the resin chemistry, thermodynamics, and curing temperature profile. It’s also difficult to fully consolidate the composite to minimize voids. My advisor was working on this in the early ‘90s as the Navy was very interested. It’s very challenging and I’m not aware of anyone who’s been successful over the last 30 years.

I won’t get into the mechanics, but larger diameter vessels require thicker walls for a given pressure and material. So carbon fiber might work great for ROVs, but is much more concerning for larger vessels with humans inside. Others have already mentioned fatigue of composites. This is a complex subject due to the anisotropic (directional dependence of mechanical properties) nature of composites. Generally you need tension to grow cracks, but complex shapes and materials can create some interesting stress patterns. Someone else mentioned FEA which is on target. It’s complex to do with composites due to this anisotropy, but most commercial packages have some capability to handle this. Now that someone tracked down at least some dimensions I may take a crack at it and ballpark things.

One other aspect that has been discussed is damage tolerance and fatigue resistance. Again, “it depends” is the answer but generally is worse than metals and much more difficult to detect and manage. A robust NDE (non-destructive evaluation) program is required when composites are used, especially in mission critical applications where loss of life is a possibility. It’s also not advisable to add any holes, slots, screws etc. in composites.

I’ve limited my comments to mostly technical aspects, but I’d be remiss if I didn’t comment on the engineering process, though it appears controversial. The general engineering process that appears to have been in place (I want more details before finalizing my thoughts) lacked rigor. Some other engineers in the group have made some great points (window rated to ~1/3 of the target depth??). The only thing I’ll add is that the lack of reviews when so much risk is present is scary. An engineer’s ego should never be bigger than their expertise. This was the case here. We should never repeat mistakes made with blood.

My apologies as I’m typing on a phone and probably made several mistakes. I also couldn’t easily reference which member said what. I’ll try and get back to this on Tuesday when I’m back in front of a computer.

This message has been edited. Last edited by: Kampy,

^^^^ What he said!

I thought this summed it up pretty well

quote:

my MS and PhD were on composites related topics.

In your professional opinion, do you think it is possible to construct a viable/safe deep sea submersible using carbon fiber composites?

I've read this entire thread through, parts of it twice, and I still cannot find the section that proves that the hull was the failure point. Was it really the hull? Did a pin-hole leak occur, or a crack? Was it caused by a temperature-related failure vice pressure-related? Instead of the hull, was it the mating site of fiber to metal? Instead of the hull, was the failure in the view port? Perhaps the hull was guiltless.

An aircraft wreck can go years without having an agreed-upon reason for the failure. Think TWA 800. That's after a large percentage of the wreck is recovered.

With this wreck, how much of it can ever be recovered? How much can studies of that material reveal?

I am not a structural engineer. I am a cynic. I'm of the opinion that IF we ever understand how this craft failed, it will be after many years, not a few days. If ever.

There isn't any definitive proof yet that hull collapse was the failure mode. Everyone's discussing it because implosion is most likely what happened. This is also corroborated by the Navy actually hearing what could have been an implosion. There are quite a number of ways this could have happened, some of which you mentioned.

As a matter of fact, one theory is that the viewport breached, as it was not actually rated for that depth. I'm wondering if that would cause such an implosion. Water rushing (no pun intended) in would equalize the internal pressure, taking stress off the hull. Of course, the crew would still be dead and the craft lost, so no overall difference.

I'm wondering if debris will be recovered for further analysis.

I'm just thinking they should ask Celine to sing for any services but if she's unable, backup plans would be Kenny Loggins singing "Highway to the Danger Zone" or Garth singing, "I've Got Friends In Low Places.'

Viewport, endcaps, tube in the middle, were all part of the "hull". My limited understanding is that a failure of any part of the hull would result in an equalization of pressure; a sudden and instantaneous one. With the greater pressure on the outside, that sudden, instantaneous, equalization of pressure is called an implosion. This happens faster than the brain can think; no time for even an "O" in "Oh, shit!". The air inside compresses and heats up, a lot. So much so that it ignites and incinerated everything inside. Then it explodes.

Thanks for the informed comment, Kampy.

quote:

Originally posted by whanson_wi:
With this wreck, how much of it can ever be recovered? How much can studies of that material reveal?

It has been posited that, given the extreme violence of the implosion, the carbon fiber main body of the vessel probably disintegrated. One imagines the same of the view port material. It seems likely the only things that would have survived more-or-less "intact" would have been the metal bits.

quote:

Originally posted by kkina:
There isn't any definitive proof yet that hull collapse was the failure mode. Everyone's discussing it because implosion is most likely what happened. This is also corroborated by the Navy actually hearing what could have been an implosion. There are quite a number of ways this could have happened, some of which you mentioned.

As a matter of fact, one theory is that the viewport breached, as it was not actually rated for that depth. I'm wondering if that would cause such an implosion. Water rushing (no pun intended) in would equalize the internal pressure, taking stress off the hull. Of course, the crew would still be dead and the craft lost, so no overall difference.

I'm wondering if debris will be recovered for further analysis.

There isn't, and there may never be, with the catastrophic failure that happened, all we really know is that it failed, imploded, and they found some parts on the ocean floor.

Could be the port failed, maybe the ring failed, might be they didn't get a bolt tight to torque, or they didn't replace those bolts they use to close the port and over fatigued one or more and the bolt failed.

The discussions over carbon fiber are interesting, maybe the CF failed, but at this point it's nothing more than conjecture, we have zero proof on what actually failed, and the only way we'll know is if one of the recording devices inside captured the event and was housed in a part of it that didn't vaporize, like a black box in the tail section they supposedly found.

quote:

Originally posted by Flashlightboy:
I'm just thinking they should ask Celine to sing for any services but if she's unable, backup plans would be Kenny Loggins singing "Highway to the Danger Zone" or Garth singing, "I've Got Friends In Low Places.'

Well…I’m sure Sir Elton John could modify “Candle in the Wind” like he has for 1,000 other “world events”. Just sayin’…

quote:

Originally posted by kkina:
I'm wondering if that would cause such an implosion. Water rushing (no pun intended) in would equalize the internal pressure, taking stress off the hull.

No need to wonder that, because that’s not how pressure differential works.

Any breach of the hull/viewport/flange/endcap would have caused an implosion. There’s no “equalizing of pressure” in the way you are thinking about it, a slow and steady filling of the submersible. It’s a rapid and catastrophic equalization, with pressure and heat generation that is absolutely beyond our comprehension.

quote:

and the only way we'll know is if one of the recording devices inside captured the event and was housed in a part of it that didn't vaporize, like a black box in the tail section they supposedly found.

Yeah right. Homeboy was using these to communicate with the surface. Off the shelf. Inspirational.

quote:

Originally posted by reloader-1:

quote:

Originally posted by kkina:
I'm wondering if that would cause such an implosion. Water rushing (no pun intended) in would equalize the internal pressure, taking stress off the hull.

No need to wonder that, because that’s not how pressure differential works.

Any breach of the hull/viewport/flange/endcap would have caused an implosion. There’s no “equalizing of pressure” in the way you are thinking about it, a slow and steady filling of the submersible. It’s a rapid and catastrophic equalization, with pressure and heat generation that is absolutely beyond our comprehension.

Well, I wasn't thinking it would be a gradual filling. It would still be almost instantaneous, but through the viewport, not with the hull flattening.

I guess this is the point where we would do the actual experiment.

quote:

Originally posted by kkina:
I guess this is the point where we would do the actual experiment.

The experiments have been done.

During an implosion of the kind the Titan almost certainly suffered, everything rushes violently inwards, compresses greatly, then "rebounds" violently.

This is why the submersible wasn't just crushed, but, essentially, exploded.

IOW: It's highly unlikely they're going to find her main tube somewhere, crushed like a discarded beer can.