i am impressed with the effort you put into attempting accurately model the composite tube. as for going down, you can really see why the sphere wins out but i would probably go down in anything thats been certified by group of suitable subject specialists.
Rush aggressively avoided certificates and NDT (Non-destructive Testing). He *knew* his submersible menace was fundamentally flawed, and knew if he went for certification or testing, he'd be forced to stop and rethink, which would have been financially ruinous. So, instead, he gaslighted, obfuscated, and crossed his fingers. We see how THAT worked out...
The joke in academia is that only 2 people will ever read your thesis, the first being you and the second being your supervisor. So 4,205 reads is not bad haha.
@@JustBadly I can agree that a fair part of papers have criminal (it's not an overstatement) mistakes in statistical analysis, but I truly believe we should all still appreciate how powerful this tool is. I get it, it's a joke, but in real life If you think that statistics is a lie then either you do statistics wrong or you don't understand the data to interpret it properly.
Actually, there was a study a few years back that showed that the huge majority of theses are never even quoted . It may not mean that no ones read them, but it’s close
I defended my thesis (something about foundations), no one seemed to care. I discovered my thesis had some major flaws in it afterwards :D It's fair to say no one read through my paper. (Luckily I ended up not working in my field (construction engineering) so everyone can relax.)
Timecodes: 0:00 - Intro 0:35 - Breaking the rules with good engineering LOL 1:21 - Best Submersibles Design according to SCIENCE ! 2:41 - Failure mode of the EGGmersible 3:26 - Failure mode of the otherMersible 3:55 - Model details - definition of Composite Cylinder 4:11 - Optimization of CFRP Layup 4:35 - How to improve the CFRP-Titanium Interface? 5:08 - Model details - how to built the sub? 6:02 - Model details - how to mesh the sub? 6:24 - Why I am an eggspert and you are not 7:00 - my PhD thesis 7:24 - Subscribe, Like and Comment 8:09 - Become a Patreon ! Become a Patreon: www.patreon.com/posts/oceangate-titan-85704969?Link& References: PhD thesis: www.researchgate.net/publication/337465846_ROBUST_DESIGN_OF_BUCKLING_CRITICAL_THIN-WALLED_SHELL_STRUCTURES Paper 1: link.springer.com/article/10.1007/s11804-020-00168-w Paper 2: www.tandfonline.com/doi/abs/10.1080/17445302.2017.1389253 Optimized PlyLayup Paper: www.sciencedirect.com/science/article/pii/S0263822318341710?via%3Dihub
Look up Aluminaut. Exactly the same performance requirements as the Titan (well actually superior in virtually every way). The US Navy DSRV uses multiple spherical pressure compartments. The crew are in the forward one and the rescued submariners are in the aft two (at the pressure in the submarine to avoid decompression sickness).
This is the best RUclips can be: hear a *real* expert in the field about a story of interest. This must be by far and away the most technically profound content on the Titan disaster. Thank you!
Yeah I was thinking especially about generative design and topological optimization, combined with additive manufacturing to create unique and unibody shapes from titanium alone. Your video is a very close vision to that, and I love it!
In practice, Navy submarines appear to have internal and external structures (floor diaphragms and bulkheads as well as structural hoops on the exterior) that reinforce the cylindrical hull. And of course they are metal.
I lived in the Florida Keys for years. I remember an old salt captain who'd sailed around the world many, many times telling me, "the sea is a cruel mistress - and if you're going to step foot in her home, you will follow her rules". Later he said, "when you're in the ocean, you're not at the top of the food chain anymore - in fact, you're at the very bottom; and every single creature there knows this - so remember, when you're out there swimming, something is looking up from the depths at you, and is seriously considering having you for dinner".
The creatures in the ocean are nowhere near as dangerous as the ocean itself. The environment itself is utterly hostile to humans and will kill you before any fish manage to. You can drown in waist deep water.
It's all about tension and compression. Think of the old days. You want to make a lashing between the lower mast and the top mast of a boat. A piece of hemp rope would work wonderfully because it can handle tension. You wouldn't hack out pieces of sand stone for that application. Now you want to build a vault in a cathedral. Sand stone can handle the compession and is what has been used traditionally. A HEMP ROPE CAN'T HOLD UP A VAULT. That is the Titan in a nut shell. Now change from hemp to carbon fibre. Those carbon fibres, the way they were oriented, did NOTHING to the integrity of the vessel. The epoxi matrix had to take all the compression stress. Mysteriously it worked a few times. But it was doomed to fail eventually. The epoxi would squish out like marshmallows to the sides under compression. That is delaminate.
Raising my hat to you! Seeing the ease you handle the complex designs, my trust in the younger generations of engineers is back. At least they have good teachers!
So the connected spheres model does exactly what I thought one should do, if you don’t have a plain round sphere but something elongated, you probably can gain a lot of strength by “reinforcing” the weakest parts which in the case of the connected spheres is seen in those thick rings at the connection points. If you had a cylindrical thing that is metal and not carbon fiber, that could mean an extra ring in the middle of the cylinder or a gradual thickening. Only really useful in homogeneous materials though. At least those were my guesses as someone who didn’t specifically study these type of material sciences…
I’m impressed by the passion and love Dr. Ronald Wagner puts into engineering… outstanding. As of the models, we use them exactly to idealize and screen between concepts… the most promising are the ones we move forward for testing. Testing is not the place to find out there are issues. We keep in mind that also we design with safety factors depending on risk. Materials and manufacturing processes have inherent variability that may not be captured in a in-silico model… therefore material testing, validation and certification is always needed.
@@hnrwagner If anything, engineering concepts can reveal corruption, like on 9/11! Sadly, people still believe the towers collapsed due to fires created by jet fuel!😂😂😂
@hnrwagner Exactly my friend! When I initially encountered that idea, I kind of laughed, because it appeared "self-evident," but I looked into the work of Simon Stack's 'September Clues' and realized the deception was much more multilayered! Sadly, people like you and I are still outliers! Thanks for responding!
Thanks for posting this, So in a nut shell… if it was totally made out of titanium instead of with carbon fibre section it would be 4 times stronger, with multiple sphere joined together it would be 8 times stronger, for an egg shape it would be 12 times stronger, and a complete sphere would be 20 times stronger.
Only slight geometric adjustments would have significantly enhanced the structure’s ability to react the hydraulic pressures at depth. The cylinder should have been designed with an outward curvature (a relatively large, single radius is adequate - say large enough over the cylinder’s mean diameter to create a four inch bulge at the center of its span). This curvature would have given significantly more stability against bottle-necking of a straight cylinder. The four inch barreling would also permit space to potentially include a stiff frame (doughnut or washer shape) at the center span for significantly enhanced crush resistance… while maintaining a constant diametral pass through clearance for passengers. The curved cylinder’s shape resists pressure on the outside by resistance to flattening the curvature to the point where inflection to reverse curvature and crumpling begins to occur. The deformation tends to elongate the cylinder, generating significant longitudinal kick force all around the shell. This force has a secondary and beneficial effect of pressing the ends of the cylinder against the hemispherical end cap and rings… pressing the end bond line rather than prying and shearing it under flexure deformation of the cylinder. There is actually little benefit, by the way, from the outer step of the titanium ring at the interface. That titanium would have been better positioned to increase the inner step’s wall thickness… furthering the stiffness and ability of that flange portion of the ring to react the inward pressure at the overlapping interface and circumferential bond line.
I have seen analysis, that potentially it could not have started as just buckling of cf shell, but maybe leak on joint between two materials as they have so much different mechanical properties under stress. In that case just a pinhole with such water pressure would quickly grow and tear apart epoxy and cause whole thing to collolapse regardless of cf part shape... Altough I don't know if we officially know what was failure mode, was is a leak first or just straight up buckling of weakened tube? Hm 🤔
Und jetzt nochmal ein cfk Zylinder mit optimierter Wandstärke und Ringspanten(eventuell noch Längsspanten). War das Innere Volumen bzw. Das a nutzbare Innere Volumen deiner Testkörper gleich?
Thank you so much for sharing your expertise and knowledge, very much appreciated. Probably the best video on the design flaws of different submersible shapes. I wonder why Stockton Rush never consulted people like your self? Maybe he would have been told something he didn't want to hear, and therefore pay higher production costs, perhaps just sheer arrogance? We'll probably never really know.
Video from the University of Plymouth (UK) pressure testing carbon fiber submarine hull deigns has been online for over a decade along with the paper for anybody to read. Though you don't have to be a engineering genius to figure out some extra reinforcement ribbing couldn't hurt. Lot of military interest back then as lighter hulls mean more speed along with some desirable acoustic properties.
the modern definition of the word "rush" to rush (verb - rushed; rushing; rushes) intransitive verb 1: to get into service a product based on a fundamentally unsafe design, to reject any safety concerns as insults -- "This experimental carbon fiber submarine has been rushed into deep sea passenger service without consideration for safety."
It would have resulted in a fail. Stockton knew that. It would have upset his business model. Just a real tragedy that he put business before human life, even his own.
@@adcashmoWell, the first Titan hull did failed the test as Stockton had a second hull made and replaced it. The one involved in the accident is the second hull. You know this thing call “Google” you should try to use it…
One of my first thoughts upon seeing this cylindrical thing was "if they wanted to elongate it, wouldn't it be better to at least use an ellipsoid?" Yeah, it may be more expensive to make out of metal, but if they were going for wrapping their hull in the carbon fiber ribbon, that's one thing that wouldn't mind custom shapes (not that a spool of fiber is of much use against external pressure). Thank you for demonstrating how about anything is so much better than cylinder. But seriously, how does one even name his sub TITAN and not actually make the pressure hull of TITANium?
It was the resins that contained the sea... The sleeve of your coat does not resist the pressure... Your arm gives resistance to your coat... and carbon fibers need resins... The resistance was given by the resins ... It's a submarine made of epoxy
@@eduardoamaro7867 Yeah, exactly. An "innovative composite material" that actually defeats its own own purpose in this application. And they didn't even weave it properly. I'd rather trust plywood
[38,-38,68,-68,90,90,90,90,38,-38] , the best after l could find it is from: Stacking sequence influence on imperfection sensitivity of cylindrical composite shells under axial compression, Friedrich et al.
@@hnrwagner -thanks- that’s for axial compression so compression normal to the surface may be different-in one of the videos I saw an early tube being “wet wound “ with a fibre orientation of 90deg . How does this work in your FEA ( I know it’s a bit radical having no off axis fibre but this guy was a maverick!)
Interesting, but what about buoyancy? It would be interesting to calculate the crush depth for each model using the wall thickness that results in neutral buoyancy (with a constant internal volume). The buoyancy of course has to be positive so it can be ballasted to neutrality, but this criterion gives the limits of the design and material selection. This was Rush's stated reason for using Carbon Fibre: high strength to buoyancy ratio.
that's probably the biggest reason he went with carbon fiber, cuz if the pressure hull was not buoyant he'd have to have massively complex systems to insure ability to surface. His core "innovation" in his mind, is probably doing this naturally buoyant pressure hull so all it has to do is drop the weights to surface, so super simple super safe.
@@Mike-Bell Synthetic foam would be crushed by the pressure unless protected by its own pressure hull, in which case that hull may as well just be filled with air, getting us back to the original strength to buoyancy problem. You would need some incompressible material less dense than water. The original deep-dive submersible, Trieste, used a giant tank of petroleum (the underwater analog of a balloon), but that does hamper manoeuvrability if you want to explore a wreak!
@@john_hind Syntactic, not synthetic. Syntactic foam is glass balls in a matrix and specially made to be incompressible for extreme pressure. Its v expensive.
@@Mike-Bell Ah, learned something today - thanks! 'Syntactic foam': weird name, sounds like some abstract concept from AI or linguistic philosophy! That is how Cameron's pillow shaped sub worked? Did the Titan have any of that, or was it relying entirely on positive buoyancy of the pressure hull? Still think it would be a useful benchmark of the quality of a material and shape to dimension the hull thickness for neutral buoyancy with a standard internal volume and then compare crush depths. Would be a standard benchmark for comparison, though obviously the actual dive depth would need to be derated (or buoyancy added) for a safe design.
Sad to realize that Sea Gate needed to hire and trust only a few talented people, including an engineer like you, and they could have produced a far superior product.
What Ocean Gate needs to know and keep in mind is that shipbuilding and aerospace manufacturing has changed... Now new technology approved in 2020 provides another tool and a different way of doing things... Submarines are going to weld automatically with laser due to new technologies and aviation. builds stamped monocoques... eliminating rivets... The qualitative leap is given by technology and people who know how to listen... what is said...
I have not watched all of your work. But here are a couple thoughts that seem largely overlooked I do not see that the carbon fiber would add compression strength with the layout that was used. It is like soaking rope in epoxy. Great for a tank to hold pressure, not great for enormous external compression loads A main reason why he chose composite over metal was the buoyancy issue. If you use metal… you have to work very hard to overcome the added weight as compared to a composite
How about using a linked-spheres model, reinforced with some metal rings around the outside, but then enclosed in a cylindrical can filled with a lighter-than-water oil? Alternatively, I wonder how well something vaguely like a dirigible could work with metal frame and layers of flexible carbon fiber skin that was in tension?
Not really work hard - there is a robust solution already with syntactic foam. But this foam is very expensive at the quality needed for a DSV, and Titan was supposed to be cheap. And yeah, when I first heard the sub was made of Titan and composites, I actually expected a design like a COPV.
actually, the amount of energy required is impractical. We can do it, but our society has energy crisis (the thing with climate change is we can't use fossil fuel freely anymore). What I mean is, the energy (fuel cost) per Kg of payload. eg: Titan hull cost USD 1million, but to send it to the moon will cost USD 1 billion.
@@francessimmonds5784 My bad. I were some quite deep ones out there, but it seems Conshelf III was the deepest one so far at 300m. I do think there is a case to be made though, that a deep-sea habitat would be easier to make. So long as it is not required to get up to the surface under its own power, it can be as heavy as it likes, and thus a thick-walled steel structure should suffice. If a submersible can go to this depth, so can a fixed structure. At work we routinely use quick connect fittings for hydraulic systems rated for 200 bars operating pressure, and hydrauliccylinders with operatingpressures of 700 bar. I see no reason you couldn't make a docking port of sorts for a submersible that would work at 400 bars, to get people in and out. Lifesupport is basically the same as for space, except less temperature fluctuations and radiation. Now if anyone would actually do it before we build a base on the moon, that is a different matter entirely.
Wow! Those animations were so cool. Thanks for sharing. Ive been really interested to learn more about this kind of topic, since i heard about the titan submersible. The animations were a pleasant bonus.
How would it affect the strength of the cylindrical shape if you devide it into two sections with a wall? of course that would have to have a hatch in it in order to go from one side to another. Normally I would expect that to increase the structural integrity but I have no idea how that behaves when applied to a carbon fiber structure under pressure. Could that even turn out as a problem, when you have a more rigid section in the middle that kinda creates a counter force that pushes outwards while the rest of the body gets deformed by the outer pressure? I guess the major problem with carbon fiber is, that it simply pulverizes instead of getting deformed when enough force is applied.
From my understanding Karl Stanley's submersible is made basically from merging 2 spheres, it would be interesting as real-life example of non-standard design.
Since forever, simplified models are a great starting point to research different designs and avoid predictable mistakes. +1 for the original designs. Personally I’d trust a sphere over anything else, it is a proven design and likely the most researched, the best understood, and the most widely tested of all types of pressure vessels.
Very nice work but it's not clear what the ground rules are. If we insist on an interior volume which at least encloses the Titan interior and minimum mass of titanium shell, what is the best shape? Will it be a prolate ellipsoidal shell that is slightly thinner at the two ends?
Use the front and back parts from the original sub. And cut out the carbon fibre part. Would be able to seat 2. So make 3 separate pods. But that would be super heavy.
Great video thanks for sharing! The crazy thing is it's possible to make a cylinder-shaped sub from titanium and add buoyancy with syntactic foam, had Oceangate done that it probably would have been fine. A sphere is always better of course but just choosing the right wall thickness and stiffening rings is enough to make it work. It would be interesting to see your evaluation of large acrylic windows. What is the largest window that can be used at 4000m depth and still be safe.
Are these better for the same mass and internal volume? If you have made the pressure hull weight a lot more you cant really compare it to the original.
I was wondering: How would the carbon fiber hull act if you put one or two titanium or steel rings inside to support it? In my head that should put the hull in tension which should be where carbon fiber excels.
@@eduardoamaro7867 Yes. That's what I'm talking about. Put titanium rings inside the sub and the carbon fiber + resin that used to be in compression is now in tension. The geometry of the whole orchestra completely changes. It's basically a tent now. Canvas with no support won't do you much good. Put a stick inside, and you got yourself a tiny house. You'd need to keep the rings away from each other with a frame of some sort, and the fiber would need to be wrapped in all directions but that's about it. The entire carbon fiber structure would be in tension, just as is required to actually have the carbon fiber as a load bearing element and not just a filler for some epoxy. Still not as good as full titanium hull, but better than unsupported bunch of resin.
Ocean gates attitude towards inovation was perfect provided you didn't put people in it. Send it up and down automatically untill it destroyed itself and learn from it . They were kind of using space X method for testing to destruction only putting people on the rocket which is completely nuts.
It's kind of funny knowing that when says "adding value to humanity" he means it in dollar terms as shareholder return. Sadly, I don't think it's going to work out this time.
Look at the mouniting of the thruster brackets , they attached the brackets to the CF wIith lag screws. then they moved some holes and possibly didnt fill the hole(s), the port horizontal hruster. Also the synaptic foam assembly, plastic straps and woodscrews.
Would placing structural support beams inside the cylinder help the strength? I know it would make moving around inside the sub more difficult for passengers, but the trade-off might be worth it.
I was intrigued by the forces on the 4 sphere dome and how they increase with pressure opposite the window. So I assume that the force of the window being pushed in is acting like a wedge, eg, the cone narrows too fast giving it a mechanical advantage. If so, perhaps the angle of the opening could be changed to avoid that wedging action (i.e., make the inner window diameter larger or smaller). I once read about car hitting a dome home, resulting in no structural damage where the car hit, but on the opposite side, , the frame popped apart (as the force of the collision was focused on the opposite side).
Subscribed. I use FEA regularly, but not in this level of detail (SolidWorks Simulation and Ansys Mechanical). Question, though - why does your model of the titan appear to show that the Ti hemispheres are thicker than the CF tube? Is that just an abaqus display issue, or are you neglecting the behavior of the titanium hemispheres (as it is accepted that the CF tube is the weak link)?
at the time I created the model, the info was that the Ti hemispheres where like 160 mm thick, more recent data show a thickness of about 85 mm. Currently I am updating the model and add more details :)
Oceangate wasn’t ‘innovating’, in-fact they were ignoring basic design principles, that isn’t innovation - that’s hubris. And they paid the ultimate price for that hubris.
I would think a honeycomb shape would be best. Then have each area be used for different things. Might be to heavy to add a skeletal structure inside of it
Hi Dr. Wagner. How much can we rely on the results from a failure theory like Hashin - assuming the CFRP manufacture is good? Is this all the FEA you would need to carry out for a structure like OpenGate, or would a real FEA report of a structure like OceanGate's sub be a lot more expansive? On another topic - I used to work with Metal Matrix Composites. They were made by squeezed-casting. The CTE mismatch between the alumina matrix and the aluminium matrix created a large pre-tension of the fibres once the part cooled after casting. Hence the material has an amazing compressive strength. Do you think a MMC would have been a good material for the sub?
Very interesting video. I think I missed the part where you explain how you decided on the material thickness. If you made the titanium the same thickness as the CFRP, then we are not comparing apples to apples. For a fair comparison I think the wight to bouyancy ratio must be kept the same, and internal volume must be kept the same. After all the whole point of the CFRP was to keep the wight down, so it would not need (as much?) Bouyancy foam on the outside.
I only kept thickness constant, the demonstration was mainly about the shape which are better than the Titan, consideration of bouyancy and internal volume would not change the statement that the egg and the merge spheres are better than the cylinder.
After jettisoning the landing gear, if the carbon tube is "squashing" a bit, the buoyancy volume is decreasing. Basically their "air balloon" of the pressure chamber is shrinking, loosing buoyancy with each second. Hense a faster descent, and almost nil ascent after the jettison.
I've been wondering about a hexagonal or octagonal structure made of titanium tubes with CF wound around it, perhaps with additional support structure inside. My thinking is that CF in a shape other than a circular tube could take advantage of the tensile strength of the material. Of course I know that with pressure vessels moving away from circular or spherical is always a bad idea. I'm just... curious how it would behave.
As an alternative to the egg shape you could have a large sphere with a smaller sphere attached at the rear. You could still have some of the space advantage and it could possibly be stronger than the egg.
Got a specific geometry for a 5ft/152.4cm ID × 12.5ft/3.81m of a parallel-sided barrel of titanium staves of wedge-xsec for 10km at yield, to certify 4km. 280 staves 2.625" bottom, 1° top is 2.66" at 4.5" height. For 5-psngrs 30"/seat, footwell, bulkheads between. The faces take the pressure as circumferential compression as a physical geometry what homogeneous shells do internally. This implies the total compression/face is far less than the pressure applied the reason to consider it. Appreciate your modelling for this vs bulbs for cost, ease of mfg & qa of beams/staves it looks bomber/overbuilt at 4.5". 🍺
I don't think buckling strength is the weakest link when you consider that 2 materials are joined by glue. The different stiffnesses of these materials means that at depth, the carbon-fiber composite would deform more than the titanium. This would put increased stress on the joint and have a potential to break the glue holding the parts together. When this happens, even a tiny hole would quickly ingress water in a rapid decompression, imploding the vessel long before anything buckles. I think moving to all titanium would not only improve buckling strength, but the material stiffness would be uniform, allowing the failure mode to be buckling instead of a glue bond failure. Do you think there's a way to model this kind of differential deformation causing additional stress in the glue bond? Another variable to throw in the mix is cyclical loading. Could fatigue cracking occur with multiple cycles? Is there a way to model how many cycles until you expect a fatigue crack? These innovative designs may be more susceptible to fatigue cracking, so it might worth looking at. I expect buckling strength to be a good starting point for submarine design, but there are other failure modes that impact this design too.
Titan would still be in operation if engineering just like Ronald shows had been done. Given that software for this type of analysis has been available for awhile now, Oceangate’s failure to use it is inexcusable. A license to use it for a year would probably cost less than what Oceangate charged each passenger for a trip. But even without finite element analysis software, successful designs of submersibles have been around for quite some time. Good engineering judgment goes a long way. What were they thinking?
I'm sorry if this has been asked and answered, but do these comparisons include imperfections for the buckling pressure? What do you think would be a reasonable size of imperfections in these cases? Love your channel by the way.
Dont want to nitpick too much, great video otherwise. You need to balance better, the interview with Stokton Rush is WAY louder than the rest of the video.
Yes the Russians have a small submarine which has a similar design with several spheres connected for it's pressure chamber - I guess it wasn't innovative enough for Rush.
They had such a submarine. It sank in 2019. The spheres had been fine, but using the wrong type of battery inside a submarine had in hindsight been also a preventable accident. Actually, the Russians officially try to repair the burnt-out submarine. I really hope though, they find a good excuse to simply forget the project.
“You are remembered for the rules you break.” He isn’t wrong, but he probably didn’t mean to be remembered as a cautionary tale.
He'll be remembered for all the wrong reasons.
A new rule of thumb is: "Don't be a Stockton Rush about it "
I don't know what to think of this late individual Stockton Rush , i can't make sense of such foolishness!???
@thabzmad7265, I’d call it a “Rush job”
@@thabzmad7265 Don't (Stockton) rush into that decision....
i am impressed with the effort you put into attempting accurately model the composite tube.
as for going down, you can really see why the sphere wins out but i would probably go down in anything thats been certified by group of suitable subject specialists.
Rush aggressively avoided certificates and NDT (Non-destructive Testing). He *knew* his submersible menace was fundamentally flawed, and knew if he went for certification or testing, he'd be forced to stop and rethink, which would have been financially ruinous.
So, instead, he gaslighted, obfuscated, and crossed his fingers.
We see how THAT worked out...
Let me bring forth the glorious tale of Aluminaut! Nothing wrong with the shape of the pressure vessel.
The joke in academia is that only 2 people will ever read your thesis, the first being you and the second being your supervisor. So 4,205 reads is not bad haha.
not even the supervisor, he only reads like beginning and end
The other joke is that 'there are lies, damn lies and statistics'.
@@JustBadly I can agree that a fair part of papers have criminal (it's not an overstatement) mistakes in statistical analysis, but I truly believe we should all still appreciate how powerful this tool is. I get it, it's a joke, but in real life If you think that statistics is a lie then either you do statistics wrong or you don't understand the data to interpret it properly.
Actually, there was a study a few years back that showed that the huge majority of theses are never even quoted . It may not mean that no ones read them, but it’s close
I defended my thesis (something about foundations), no one seemed to care. I discovered my thesis had some major flaws in it afterwards :D It's fair to say no one read through my paper. (Luckily I ended up not working in my field (construction engineering) so everyone can relax.)
Timecodes:
0:00 - Intro
0:35 - Breaking the rules with good engineering LOL
1:21 - Best Submersibles Design according to SCIENCE !
2:41 - Failure mode of the EGGmersible
3:26 - Failure mode of the otherMersible
3:55 - Model details - definition of Composite Cylinder
4:11 - Optimization of CFRP Layup
4:35 - How to improve the CFRP-Titanium Interface?
5:08 - Model details - how to built the sub?
6:02 - Model details - how to mesh the sub?
6:24 - Why I am an eggspert and you are not
7:00 - my PhD thesis
7:24 - Subscribe, Like and Comment
8:09 - Become a Patreon !
Become a Patreon:
www.patreon.com/posts/oceangate-titan-85704969?Link&
References:
PhD thesis:
www.researchgate.net/publication/337465846_ROBUST_DESIGN_OF_BUCKLING_CRITICAL_THIN-WALLED_SHELL_STRUCTURES
Paper 1:
link.springer.com/article/10.1007/s11804-020-00168-w
Paper 2:
www.tandfonline.com/doi/abs/10.1080/17445302.2017.1389253
Optimized PlyLayup Paper:
www.sciencedirect.com/science/article/pii/S0263822318341710?via%3Dihub
Look up Aluminaut.
Exactly the same performance requirements as the Titan (well actually superior in virtually every way).
The US Navy DSRV uses multiple spherical pressure compartments. The crew are in the forward one and the rescued submariners are in the aft two (at the pressure in the submarine to avoid decompression sickness).
That comment sure didn't age well.
There should be an option somewhere to pin this post to the top of the comment thread :)
This is the best RUclips can be: hear a *real* expert in the field about a story of interest. This must be by far and away the most technically profound content on the Titan disaster. Thank you!
"ROBUST DESIGN OF BUCKLING CRITICAL THIN-WALLED SHELL STRUCTURES"
When that's your PhD thesis, yeah I'd say you're an expert. 👍👏
Or maybe we can say an eggspert!😅
@@Stealth639 eggspert LOL, love it
Lol he should of been fired by ocean gate as well as the other guy that tried to help them
Yeah I was thinking especially about generative design and topological optimization, combined with additive manufacturing to create unique and unibody shapes from titanium alone. Your video is a very close vision to that, and I love it!
In practice, Navy submarines appear to have internal and external structures (floor diaphragms and bulkheads as well as structural hoops on the exterior) that reinforce the cylindrical hull. And of course they are metal.
Fools break the rules until the rules break the fools.
🤣
Thank you for taking the time to create these videos and post them for all of us to watch freely. You are amazing. Thank you!
I lived in the Florida Keys for years. I remember an old salt captain who'd sailed around the world many, many times telling me, "the sea is a cruel mistress - and if you're going to step foot in her home, you will follow her rules". Later he said, "when you're in the ocean, you're not at the top of the food chain anymore - in fact, you're at the very bottom; and every single creature there knows this - so remember, when you're out there swimming, something is looking up from the depths at you, and is seriously considering having you for dinner".
That why I like to swim in a pool, swimming in the ocean is somewhat terryfing
*increases fear of open water *EVEN MORE
The creatures in the ocean are nowhere near as dangerous as the ocean itself. The environment itself is utterly hostile to humans and will kill you before any fish manage to. You can drown in waist deep water.
Would you take a dive in the Eggmersible?
Yeah, as long as it’s not made by oceangate…
Perhaps after a decade of other people testing it.
I’d only consider it if it were certified by independent organizations as safe.
It's all about tension and compression.
Think of the old days. You want to make a lashing between the lower mast and the top mast of a boat. A piece of hemp rope would work wonderfully because it can handle tension. You wouldn't hack out pieces of sand stone for that application.
Now you want to build a vault in a cathedral. Sand stone can handle the compession and is what has been used traditionally. A HEMP ROPE CAN'T HOLD UP A VAULT.
That is the Titan in a nut shell.
Now change from hemp to carbon fibre.
Those carbon fibres, the way they were oriented, did NOTHING to the integrity of the vessel. The epoxi matrix had to take all the compression stress. Mysteriously it worked a few times. But it was doomed to fail eventually. The epoxi would squish out like marshmallows to the sides under compression. That is delaminate.
Yes, it was my first thought too, to make the center section egg-shaped...
Awesome that you picked up my question regarding the lay up, thank you !
With that egg design, you eggceeded my expectations
It was truly eggceptional
Raising my hat to you!
Seeing the ease you handle the complex designs, my trust in the younger generations of engineers is back. At least they have good teachers!
So the connected spheres model does exactly what I thought one should do, if you don’t have a plain round sphere but something elongated, you probably can gain a lot of strength by “reinforcing” the weakest parts which in the case of the connected spheres is seen in those thick rings at the connection points. If you had a cylindrical thing that is metal and not carbon fiber, that could mean an extra ring in the middle of the cylinder or a gradual thickening. Only really useful in homogeneous materials though. At least those were my guesses as someone who didn’t specifically study these type of material sciences…
I’m impressed by the passion and love Dr. Ronald Wagner puts into engineering… outstanding.
As of the models, we use them exactly to idealize and screen between concepts… the most promising are the ones we move forward for testing. Testing is not the place to find out there are issues.
We keep in mind that also we design with safety factors depending on risk.
Materials and manufacturing processes have inherent variability that may not be captured in a in-silico model… therefore material testing, validation and certification is always needed.
engineering is really great, love it, not corrupted like medicine nor can it be corrupted
@@hnrwagner If anything, engineering concepts can reveal corruption, like on 9/11! Sadly, people still believe the towers collapsed due to fires created by jet fuel!😂😂😂
@@edkiely2712 people also still believe that planes hit the towers
@hnrwagner Exactly my friend! When I initially encountered that idea, I kind of laughed, because it appeared "self-evident," but I looked into the work of Simon Stack's 'September Clues' and realized the deception was much more multilayered! Sadly, people like you and I are still outliers! Thanks for responding!
All the bs out since this accident happened this is only Chanel make sense and explain as engineering way. Thank u
Thank you for sharing you hard earned expertise!
eggsperties dont forget it.
Thanks for posting this,
So in a nut shell… if it was totally made out of titanium instead of with carbon fibre section it would be 4 times stronger, with multiple sphere joined together it would be 8 times stronger, for an egg shape it would be 12 times stronger, and a complete sphere would be 20 times stronger.
Thank you so much this is fantastic! I had the idea for multiple spheres and the egg, and so excited to see you made these models. Thank you.
as an eggspert I do what I must ;)
@hnrwagner now do a butt plug shaped one and compare it to how f-d you be with that instead.
This analysis is bloody brilliant.
Only slight geometric adjustments would have significantly enhanced the structure’s ability to react the hydraulic pressures at depth. The cylinder should have been designed with an outward curvature (a relatively large, single radius is adequate - say large enough over the cylinder’s mean diameter to create a four inch bulge at the center of its span). This curvature would have given significantly more stability against bottle-necking of a straight cylinder. The four inch barreling would also permit space to potentially include a stiff frame (doughnut or washer shape) at the center span for significantly enhanced crush resistance… while maintaining a constant diametral pass through clearance for passengers. The curved cylinder’s shape resists pressure on the outside by resistance to flattening the curvature to the point where inflection to reverse curvature and crumpling begins to occur. The deformation tends to elongate the cylinder, generating significant longitudinal kick force all around the shell. This force has a secondary and beneficial effect of pressing the ends of the cylinder against the hemispherical end cap and rings… pressing the end bond line rather than prying and shearing it under flexure deformation of the cylinder. There is actually little benefit, by the way, from the outer step of the titanium ring at the interface. That titanium would have been better positioned to increase the inner step’s wall thickness… furthering the stiffness and ability of that flange portion of the ring to react the inward pressure at the overlapping interface and circumferential bond line.
I have seen analysis, that potentially it could not have started as just buckling of cf shell, but maybe leak on joint between two materials as they have so much different mechanical properties under stress. In that case just a pinhole with such water pressure would quickly grow and tear apart epoxy and cause whole thing to collolapse regardless of cf part shape... Altough I don't know if we officially know what was failure mode, was is a leak first or just straight up buckling of weakened tube? Hm 🤔
Und jetzt nochmal ein cfk Zylinder mit optimierter Wandstärke und Ringspanten(eventuell noch Längsspanten).
War das Innere Volumen bzw. Das a nutzbare Innere Volumen deiner Testkörper gleich?
This is the greatest RUclips channel of all time
that what I think too
Thank you so much for sharing your expertise and knowledge, very much appreciated. Probably the best video on the design flaws of different submersible shapes.
I wonder why Stockton Rush never consulted people like your self? Maybe he would have been told something he didn't want to hear, and therefore pay higher production costs, perhaps just sheer arrogance? We'll probably never really know.
Have you ever tried a Hexagon tube for a pressure test? I would like to see how much stronger a round tube would be compared to a Hexagon
can try, interesting idea
My guess is it would collapse at the flat faces as they buckle inwards leading to the hex profile taking on the shape of a 6 pointed star shape.
Video from the University of Plymouth (UK) pressure testing carbon fiber submarine hull deigns has been online for over a decade along with the paper for anybody to read. Though you don't have to be a engineering genius to figure out some extra reinforcement ribbing couldn't hurt. Lot of military interest back then as lighter hulls mean more speed along with some desirable acoustic properties.
OceanGate should have hired you to peer test their design. Sadly for all concerned that didn't happen.
I am also not that expensive ;)
the modern definition of the word "rush"
to rush (verb - rushed; rushing; rushes)
intransitive verb
1: to get into service a product based on a fundamentally unsafe design, to reject any safety concerns as insults
-- "This experimental carbon fiber submarine has been rushed into deep sea passenger service without consideration for safety."
It would have resulted in a fail. Stockton knew that. It would have upset his business model.
Just a real tragedy that he put business before human life, even his own.
@@adcashmoWell, the first Titan hull did failed the test as Stockton had a second hull made and replaced it.
The one involved in the accident is the second hull.
You know this thing call “Google” you should try to use it…
One of my first thoughts upon seeing this cylindrical thing was "if they wanted to elongate it, wouldn't it be better to at least use an ellipsoid?" Yeah, it may be more expensive to make out of metal, but if they were going for wrapping their hull in the carbon fiber ribbon, that's one thing that wouldn't mind custom shapes (not that a spool of fiber is of much use against external pressure).
Thank you for demonstrating how about anything is so much better than cylinder.
But seriously, how does one even name his sub TITAN and not actually make the pressure hull of TITANium?
It was the resins that contained the sea... The sleeve of your coat does not resist the pressure... Your arm gives resistance to your coat... and carbon fibers need resins... The resistance was given by the resins ... It's a submarine made of epoxy
@@eduardoamaro7867 Yeah, exactly. An "innovative composite material" that actually defeats its own own purpose in this application. And they didn't even weave it properly. I'd rather trust plywood
You gotta have shape in three dimensions for that kind of pressure. I like that series of connected spheres design.
Ronald -another great video-what was the optimal layup for the carbon fibre?
[38,-38,68,-68,90,90,90,90,38,-38] , the best after l could find it is from:
Stacking sequence influence on imperfection sensitivity of cylindrical composite shells under axial compression, Friedrich et al.
@@hnrwagner -thanks- that’s for axial compression so compression normal to the surface may be different-in one of the videos I saw an early tube being “wet wound “ with a fibre orientation of 90deg . How does this work in your FEA ( I know it’s a bit radical having no off axis fibre but this guy was a maverick!)
Volume too low but interview clips are super loud. Going to re-watch with subtitles.
Impressive work there
You have to know the rules very well, before you have any room to consider if, when, how, and how much to break them.
Would titanium rings inside help the instability.
Interesting, but what about buoyancy? It would be interesting to calculate the crush depth for each model using the wall thickness that results in neutral buoyancy (with a constant internal volume). The buoyancy of course has to be positive so it can be ballasted to neutrality, but this criterion gives the limits of the design and material selection. This was Rush's stated reason for using Carbon Fibre: high strength to buoyancy ratio.
that's probably the biggest reason he went with carbon fiber, cuz if the pressure hull was not buoyant he'd have to have massively complex systems to insure ability to surface. His core "innovation" in his mind, is probably doing this naturally buoyant pressure hull so all it has to do is drop the weights to surface, so super simple super safe.
Titanium centre hull in place of carbon fibre composite would add 2.5 tons. To offset the extra weight require 4.9m3 of syntactic foam.
@@Mike-Bell Synthetic foam would be crushed by the pressure unless protected by its own pressure hull, in which case that hull may as well just be filled with air, getting us back to the original strength to buoyancy problem. You would need some incompressible material less dense than water. The original deep-dive submersible, Trieste, used a giant tank of petroleum (the underwater analog of a balloon), but that does hamper manoeuvrability if you want to explore a wreak!
@@john_hind Syntactic, not synthetic. Syntactic foam is glass balls in a matrix and specially made to be incompressible for extreme pressure. Its v expensive.
@@Mike-Bell Ah, learned something today - thanks! 'Syntactic foam': weird name, sounds like some abstract concept from AI or linguistic philosophy! That is how Cameron's pillow shaped sub worked? Did the Titan have any of that, or was it relying entirely on positive buoyancy of the pressure hull? Still think it would be a useful benchmark of the quality of a material and shape to dimension the hull thickness for neutral buoyancy with a standard internal volume and then compare crush depths. Would be a standard benchmark for comparison, though obviously the actual dive depth would need to be derated (or buoyancy added) for a safe design.
Sad to realize that Sea Gate needed to hire and trust only a few talented people, including an engineer like you, and they could have produced a far superior product.
And listen to them and not fire upon raising concerns.
What Ocean Gate needs to know and keep in mind is that shipbuilding and aerospace manufacturing has changed... Now new technology approved in 2020 provides another tool and a different way of doing things... Submarines are going to weld automatically with laser due to new technologies and aviation. builds stamped monocoques... eliminating rivets... The qualitative leap is given by technology and people who know how to listen... what is said...
Yeah, I too hate it when my Seagate hard drive implodes.
They couldn't afford talented engineers, let alone listen to them.
"Sea Gate" made hard drives. Ocean Gate made soft submersibles.
Thats some good work there man!
thanks
I loved your video demonstration. I’ve subscribed to your channel and will watch all of your videos.
The video clip blew up my car speakers, but it was worth the information
0:33 hearing damage alert!
I have not watched all of your work. But here are a couple thoughts that seem largely overlooked
I do not see that the carbon fiber would add compression strength with the layout that was used. It is like soaking rope in epoxy. Great for a tank to hold pressure, not great for enormous external compression loads
A main reason why he chose composite over metal was the buoyancy issue. If you use metal… you have to work very hard to overcome the added weight as compared to a composite
How about using a linked-spheres model, reinforced with some metal rings around the outside, but then enclosed in a cylindrical can filled with a lighter-than-water oil? Alternatively, I wonder how well something vaguely like a dirigible could work with metal frame and layers of flexible carbon fiber skin that was in tension?
Not really work hard - there is a robust solution already with syntactic foam. But this foam is very expensive at the quality needed for a DSV, and Titan was supposed to be cheap. And yeah, when I first heard the sub was made of Titan and composites, I actually expected a design like a COPV.
I’d say a base on the moon is more likely than a base under deep sea, and be easier to build.
actually, the amount of energy required is impractical. We can do it, but our society has energy crisis (the thing with climate change is we can't use fossil fuel freely anymore). What I mean is, the energy (fuel cost) per Kg of payload. eg: Titan hull cost USD 1million, but to send it to the moon will cost USD 1 billion.
I think you need to rethink that statement. Several underwater bases already exist. Look up underwater habitat.
@@2testtest2 I did say “deep sea”. The deepest underwater habitat I’ve found was 67ft , a far cry from the 12,500 feet of titanic.
@@francessimmonds5784 My bad. I were some quite deep ones out there, but it seems Conshelf III was the deepest one so far at 300m.
I do think there is a case to be made though, that a deep-sea habitat would be easier to make. So long as it is not required to get up to the surface under its own power, it can be as heavy as it likes, and thus a thick-walled steel structure should suffice. If a submersible can go to this depth, so can a fixed structure. At work we routinely use quick connect fittings for hydraulic systems rated for 200 bars operating pressure, and hydrauliccylinders with operatingpressures of 700 bar. I see no reason you couldn't make a docking port of sorts for a submersible that would work at 400 bars, to get people in and out. Lifesupport is basically the same as for space, except less temperature fluctuations and radiation.
Now if anyone would actually do it before we build a base on the moon, that is a different matter entirely.
@@2testtest2 "How submarine dock while underwater" -by Scott Manley
Great presentation, great work, thank you. There is no substitute for evieence based engineering, as you show.
Congrats on the success in academia and RUclips!
He will definitely be remembered, it's just that he's no longer here to know about it.
Fascinating. One of the more interesting videos on this subject. Thank you.
Wow! Those animations were so cool. Thanks for sharing. Ive been really interested to learn more about this kind of topic, since i heard about the titan submersible. The animations were a pleasant bonus.
Rush will be remembered for breaking the rules and the fatal consequences resulting in deaths of everyone onboard the titan.
„Breaking the rules LOL“ - best caption 2023 ;-)
Please do a 3D rendering of my imploded ear canals at 0:33. Kidding.
How would it affect the strength of the cylindrical shape if you devide it into two sections with a wall? of course that would have to have a hatch in it in order to go from one side to another.
Normally I would expect that to increase the structural integrity but I have no idea how that behaves when applied to a carbon fiber structure under pressure. Could that even turn out as a problem, when you have a more rigid section in the middle that kinda creates a counter force that pushes outwards while the rest of the body gets deformed by the outer pressure? I guess the major problem with carbon fiber is, that it simply pulverizes instead of getting deformed when enough force is applied.
From my understanding Karl Stanley's submersible is made basically from merging 2 spheres, it would be interesting as real-life example of non-standard design.
I believe the US DSRVs had conjoined spheres.
actually 3 spheres- just like the DSRV
You are wise beyond your years for sure
The egg-shaped submersible resembles the Turtle, built in 1775.
Good analysis and very credible to see your literature publications
Thanks
Since forever, simplified models are a great starting point to research different designs and avoid predictable mistakes. +1 for the original designs. Personally I’d trust a sphere over anything else, it is a proven design and likely the most researched, the best understood, and the most widely tested of all types of pressure vessels.
Very nice work but it's not clear what the ground rules are. If we insist on an interior volume which at least encloses the Titan interior and minimum mass of titanium shell, what is the best shape? Will it be a prolate ellipsoidal shell that is slightly thinner at the two ends?
Damn, a h-index of 18 is hella tight. Huge brother.
Thanks
Love your work, very well done!
He was right about Mars. That’s Elons fever dream that will never happen.
Use the front and back parts from the original sub. And cut out the carbon fibre part.
Would be able to seat 2. So make 3 separate pods. But that would be super heavy.
Great video thanks for sharing! The crazy thing is it's possible to make a cylinder-shaped sub from titanium and add buoyancy with syntactic foam, had Oceangate done that it probably would have been fine. A sphere is always better of course but just choosing the right wall thickness and stiffening rings is enough to make it work.
It would be interesting to see your evaluation of large acrylic windows. What is the largest window that can be used at 4000m depth and still be safe.
Are these better for the same mass and internal volume? If you have made the pressure hull weight a lot more you cant really compare it to the original.
I was wondering: How would the carbon fiber hull act if you put one or two titanium or steel rings inside to support it? In my head that should put the hull in tension which should be where carbon fiber excels.
The hull was made of resins... THE FIBERS TOLERATE THE TENSION IN THE SAME DIRECTION OF THE FILAMENT... The sea was contained with epoxy resins...
@@eduardoamaro7867 Yes. That's what I'm talking about. Put titanium rings inside the sub and the carbon fiber + resin that used to be in compression is now in tension. The geometry of the whole orchestra completely changes. It's basically a tent now. Canvas with no support won't do you much good. Put a stick inside, and you got yourself a tiny house. You'd need to keep the rings away from each other with a frame of some sort, and the fiber would need to be wrapped in all directions but that's about it. The entire carbon fiber structure would be in tension, just as is required to actually have the carbon fiber as a load bearing element and not just a filler for some epoxy. Still not as good as full titanium hull, but better than unsupported bunch of resin.
That's a solid eggspert flex.. Congrats on 1000 citations!
Respekt an deine wissenschaftlichen Leistungen!
Danke,war auch viel viel Arbeit
you dont need a different shape... just make the hull thick enough.
Ocean gates attitude towards inovation was perfect provided you didn't put people in it. Send it up and down automatically untill it destroyed itself and learn from it . They were kind of using space X method for testing to destruction only putting people on the rocket which is completely nuts.
It's kind of funny knowing that when says "adding value to humanity" he means it in dollar terms as shareholder return. Sadly, I don't think it's going to work out this time.
He ranks pretty highly on the list of names eligible for a Darwin award
Look at the mouniting of the thruster brackets , they attached the brackets to the CF wIith lag screws. then they moved some holes and possibly didnt fill the hole(s), the port horizontal hruster.
Also the synaptic foam assembly, plastic straps and woodscrews.
Would placing structural support beams inside the cylinder help the strength? I know it would make moving around inside the sub more difficult for passengers, but the trade-off might be worth it.
What's the manufacturability of any of those designs, and how does it compare to the original?
I was intrigued by the forces on the 4 sphere dome and how they increase with pressure opposite the window. So I assume that the force of the window being pushed in is acting like a wedge, eg, the cone narrows too fast giving it a mechanical advantage. If so, perhaps the angle of the opening could be changed to avoid that wedging action (i.e., make the inner window diameter larger or smaller).
I once read about car hitting a dome home, resulting in no structural damage where the car hit, but on the opposite side, , the frame popped apart (as the force of the collision was focused on the opposite side).
Excellent review. Thank you. 👍
Sehr interessant. Danke dafür.
That interview on 0:37 looks like something Michael Scott would say
the multiple spheres would greatly benefit of more filler around the intersections, to reduce the angles and add rigidity
Subscribed. I use FEA regularly, but not in this level of detail (SolidWorks Simulation and Ansys Mechanical). Question, though - why does your model of the titan appear to show that the Ti hemispheres are thicker than the CF tube? Is that just an abaqus display issue, or are you neglecting the behavior of the titanium hemispheres (as it is accepted that the CF tube is the weak link)?
at the time I created the model, the info was that the Ti hemispheres where like 160 mm thick, more recent data show a thickness of about 85 mm. Currently I am updating the model and add more details :)
Oceangate wasn’t ‘innovating’, in-fact they were ignoring basic design principles, that isn’t innovation - that’s hubris.
And they paid the ultimate price for that hubris.
Why you not use the assembly-module to assemble all parts? It seems you created a new part by merging all parts....
You certainly warrant a subscription 👌
thanks
I would think a honeycomb shape would be best. Then have each area be used for different things. Might be to heavy to add a skeletal structure inside of it
Nice work.
Hi Dr. Wagner. How much can we rely on the results from a failure theory like Hashin - assuming the CFRP manufacture is good? Is this all the FEA you would need to carry out for a structure like OpenGate, or would a real FEA report of a structure like OceanGate's sub be a lot more expansive? On another topic - I used to work with Metal Matrix Composites. They were made by squeezed-casting. The CTE mismatch between the alumina matrix and the aluminium matrix created a large pre-tension of the fibres once the part cooled after casting. Hence the material has an amazing compressive strength. Do you think a MMC would have been a good material for the sub?
Very interesting video. I think I missed the part where you explain how you decided on the material thickness. If you made the titanium the same thickness as the CFRP, then we are not comparing apples to apples. For a fair comparison I think the wight to bouyancy ratio must be kept the same, and internal volume must be kept the same. After all the whole point of the CFRP was to keep the wight down, so it would not need (as much?) Bouyancy foam on the outside.
I only kept thickness constant, the demonstration was mainly about the shape which are better than the Titan, consideration of bouyancy and internal volume would not change the statement that the egg and the merge spheres are better than the cylinder.
After jettisoning the landing gear, if the carbon tube is "squashing" a bit, the buoyancy volume is decreasing. Basically their "air balloon" of the pressure chamber is shrinking, loosing buoyancy with each second. Hense a faster descent, and almost nil ascent after the jettison.
Take care you volumes when you add clips, the 0:35- clip is painfully loud compared to the rest of the video
I've been wondering about a hexagonal or octagonal structure made of titanium tubes with CF wound around it, perhaps with additional support structure inside. My thinking is that CF in a shape other than a circular tube could take advantage of the tensile strength of the material. Of course I know that with pressure vessels moving away from circular or spherical is always a bad idea. I'm just... curious how it would behave.
As an alternative to the egg shape you could have a large sphere with a smaller sphere attached at the rear. You could still have some of the space advantage and it could possibly be stronger than the egg.
Wouldn’t the weak point be where the 2 spheres are attached to each other? And wouldn’t it fail like 3:42 ?
Great video and explanation
Got a specific geometry for a 5ft/152.4cm ID × 12.5ft/3.81m of a parallel-sided barrel of titanium staves of wedge-xsec for 10km at yield, to certify 4km.
280 staves 2.625" bottom, 1° top is 2.66" at 4.5" height. For 5-psngrs 30"/seat, footwell, bulkheads between.
The faces take the pressure as circumferential compression as a physical geometry what homogeneous shells do internally. This implies the total compression/face is far less than the pressure applied the reason to consider it.
Appreciate your modelling for this vs bulbs for cost, ease of mfg & qa of beams/staves it looks bomber/overbuilt at 4.5".
🍺
I think the weight change from CFRP to Titanium for the cylinder is huge.
I don't think buckling strength is the weakest link when you consider that 2 materials are joined by glue. The different stiffnesses of these materials means that at depth, the carbon-fiber composite would deform more than the titanium. This would put increased stress on the joint and have a potential to break the glue holding the parts together. When this happens, even a tiny hole would quickly ingress water in a rapid decompression, imploding the vessel long before anything buckles. I think moving to all titanium would not only improve buckling strength, but the material stiffness would be uniform, allowing the failure mode to be buckling instead of a glue bond failure. Do you think there's a way to model this kind of differential deformation causing additional stress in the glue bond?
Another variable to throw in the mix is cyclical loading. Could fatigue cracking occur with multiple cycles? Is there a way to model how many cycles until you expect a fatigue crack? These innovative designs may be more susceptible to fatigue cracking, so it might worth looking at.
I expect buckling strength to be a good starting point for submarine design, but there are other failure modes that impact this design too.
Titan would still be in operation if engineering just like Ronald shows had been done.
Given that software for this type of analysis has been available for awhile now, Oceangate’s failure to use it is inexcusable. A license to use it for a year would probably cost less than what Oceangate charged each passenger for a trip.
But even without finite element analysis software, successful designs of submersibles have been around for quite some time. Good engineering judgment goes a long way. What were they thinking?
I'm sorry if this has been asked and answered, but do these comparisons include imperfections for the buckling pressure? What do you think would be a reasonable size of imperfections in these cases?
Love your channel by the way.
Do you have any data on how much bigger the egg shape could be if you only wanted 2x
Dont want to nitpick too much, great video otherwise. You need to balance better, the interview with Stokton Rush is WAY louder than the rest of the video.
oh okay, I noticed it only after upload, RUclips changes the volume of the video, it was different before I uploaded.
That's because Stockton always wanted his voice to drown out the views of others
Interesting... 🤔
Yes the Russians have a small submarine which has a similar design with several spheres connected for it's pressure chamber - I guess it wasn't innovative enough for Rush.
They had such a submarine. It sank in 2019. The spheres had been fine, but using the wrong type of battery inside a submarine had in hindsight been also a preventable accident. Actually, the Russians officially try to repair the burnt-out submarine. I really hope though, they find a good excuse to simply forget the project.