As a retired Aerospace/Aeronautical Engineer with 40+ years of experience, I am 99% certain that the Titanium Flange to Titanium Carbon-Fiber hull joint was the Achilles heel of the design. Primarily due to thermal and pressure expansion/contraction rate differences between the two materials. Joint failure was inevitable.

@JoediyLab

11 ай бұрын

Thanks for the valid insight.

@rael5469

11 ай бұрын

Amen brother. I had some composite training and even at the technician level I can see where the flexing of the glued joint is going to cause a disbond. Repeated flexing under pressure will cause it to disbond. I don't think I've seen a structural joint between composite and metal that didn't include fasteners. @terrymichael5821 could they have held the two end caps to the tube by using rods between the end caps? A minimum of four I would think. Would it have made a difference or kept the tube from collapsing? I think the tube would have collapsed eventually anyway. Tube bad. Sphere good.

@rcampbell4967

11 ай бұрын

I certainly do not have your experience or credentials, but I've been saying the same thing. There are 3 materials; carbon fiber, titanium, and the adhesive. You don't need to be an engineer to understand every material reacts differently to moisture, temperature, and pressure. Stress due to deflection or the relative lack of was put on this joint with every dive and then again at every surfacing. This stress can cause tiny defects to the adhesion starting with the first dive. These defects will likely grow exponentially with each cycle until catastrophic failure is experienced. Or, the carbon fiber delaminated and shettered with such force it harmonically broke completely free of the bonding agent. Either scinario could make sense from my understanding of the 'transcript' leaked.

@gailmcn

11 ай бұрын

A lot of professionals are diagnosing the cause of this catastrophe prior to any formal investigation of the physical evidence. Which I believe is a cardinal sin in any engineering field. Furthermore, these professionals also diagnose according to what they are most familiar with in their own specialty of the broader field. So when they hear hoofbeats outside their bedroom window, they think 'donkey', instead of 'horse', just because they know a bit more about donkeys. But there are several options here for failure, so until the investigation documents are published......

@AORD72

11 ай бұрын

Do you think there could have been a slow leak initially?

The bonding method itself is dubious but the way they were doing it in a warehouse and not a cleanroom is wow.

@JoediyLab

11 ай бұрын

Yes, also the guy in the video wiping the ring/flange did not have gloves on.

@bigjared8946

11 ай бұрын

@@JoediyLab Stockton: It's not like Elmer's glue, it's like peanut butter. I'm convinced this guy's family paid for his engineering degree, which is a thing that needs more scrutiny because "engineers" like that will kill people.

@gaylecheung3087

11 ай бұрын

Leaving oils from your finger is your hands and not in a clean environment with suits and mask gloves

@SteffiReitsch

11 ай бұрын

@@gaylecheung3087 "Leaving oils from your finger is your hands.." ?

@toomanyaccounts

11 ай бұрын

@@JohnSmith-bw8dn they showed the difference between backyard hobbyists and professional sub builders. Triton Submarines is professional sub builders

A couple of additions to your analysis - there is a video of them applying the adhesive to the titanium ring, which shows them using flexible spreaders similar to those used by auto body technicians. While these work well to fair auto body repairs, they are inappropriate for applying adhesive because it's impossible to get an even thickness of adhesive. The spreaders used did not appear to have serrations on the edge. Serrations, in combination with a stiffer metal blade tool, would have made it possible to achieve an even thickness of adhesive. There is also a video of them lowering the Ti ring onto the CF after the adhesive was applied to both parts . There appears to be no adhesive "squeeze-out" after the Ti ring is fully seated, making me think that there was not enough adhesive used. It looks like they put blue painter's tape around the outside bottom edge of the Ti ring to make it easier to clean off squeeze-out, but there wasn't any squeeze-out shown in the video. There does not appear to be any blue tape on the CF. Maybe they had tape on the CF when they applied the adhesive to it, then pulled the tape before lowering the Ti ring, but why would they do that?

@JoediyLab

11 ай бұрын

Yes I was thinking the same thing. The epoxy specs I showed from 3M in the video recommends a 7 mil thickness. They appeared to spread the epoxy too thick on the titanium. Thanks for the comment

@Andy-df5fj

11 ай бұрын

I recall an Oceangate video where Rush was speaking of the glue joint where he stressed they have to use enough, but not too much glue which suggests to me that they didn't rely on excess glue squeezing out, but rather preferred to use a calculated amount to avoid excess. I have no idea why.

@tarantellalarouge7632

11 ай бұрын

exactly the same tool to apply cream on a cake !

@TechWizMaster

10 ай бұрын

you know once under water they wouldn't even need to have the two pieces glued together...the pressure of the water alone will make it impossible to separate the caps even if you'd try...so the glue as NOTHING to do with all this.

@Andy-df5fj

10 ай бұрын

@@TechWizMaster While it's true that the pressure would squeeze on the caps, the glue itself is still subject to pressures and could extrude through the gap. Still, I'm more concerned about the thin titanium flange that interfaces with the inner diameters on the ends of the carbon fiber tube. Under pressure, the walls of a cylinder are the weakest point and as it compresses under pressure, that thin flange is to weak to resist collapse.

The Coast Guard wants 2 years to investigate this accident, if they would just release pictures of all the pieces and photos of where each piece was found, the KZread community would have this thing figured out in a few days

@POVwithRC

11 ай бұрын

Bureaucracy only exists to extend its own life by pretending it is the only qualified party. The USCG and Canada's TSB are prime examples of this. We could certainly do better.

@londonoverground

11 ай бұрын

​@@POVwithRCbollocks

@craigsowers8456

11 ай бұрын

LOL ... that is NOT how "Root Cause and Corrective Action" boards work (and I've attended thousands) ... I would agree that just from the video evidence that there were MANY mistakes made ... but to prevent future occurrences from happening, the entire RCCA process must be followed; keep in mind their recommendations will include multinational "Governmental Oversight" to close the loophole that allowed this tragedy to occur in the first place. Cool your jets ... but DO keep making observations ... would doubt NTSB is reading all these comments ... they ain't perfect and need all the help they can get.

@martinwhite418

11 ай бұрын

The "community" can't do interviews or get court ordered paperwork.

@alihenderson5910

10 ай бұрын

​@@londonovergroundNope, perfectly true. Keep taking the boosters.

Glad to see you mentioned Pagani Automobili. I visited the factory this year and their cars are an engineering marvel. Horacio Pagani developed carbotanium to use it in the highly stressed parts of the frame, it is designed to improve carbon fiber strength under compression as well as mitigating its tendency to shatter at impact.

@JoediyLab

11 ай бұрын

Very cool, would love to see their facility and their processes. Thanks for the information.

@NoNORADon911

11 ай бұрын

@@JoediyLab Mic is good but volume is low.

@bestopinion9257

10 ай бұрын

Pagani was what this Rush guy believed about himself. He was a young employee at Lamborghini and he made its own brand because Lamborghini refused his ideas and remained more traditionalist. Now Pagani is a successful brand and Lamborghini lives thanks to Volkswagen.

What is also interesting is that the "implosion" had enough force to not only take the titanium ring off the carbon fiber, but also sheared the 17 bolts holding the titanium dome onto the ring; since they were found separated.

@danieldupuys2002

10 ай бұрын

Yes it was done by the enormous explosion of the compressed air inside the hull when the carbon fiber tube imploded: similar of a giant "diesel effect"whith a compression ration around 350/1, in an diesel engine, its around 20 to25/1. Note the window also is pushed in front with it's retaining ring ans bolts ( photos wrecked parts)

@arthurdent5357

10 ай бұрын

It's possible the crew unscrewed them so they could attach it to a crane.

@ermias75ermis2

10 ай бұрын

Natures forces laugh at such mere obstacles.. It only took a millisecond to destroy this toy sub.

@78a67h

10 ай бұрын

The CF / Titanium ring bond was way weaker than the bolts holding the hemispheres so the rings must have detached before sufficient force met the bolts to make them shear.

@user-jb5pp5rw9u

10 ай бұрын

Correct on the separation of the hemisphere and ring by failure of the bolts but the bolts were not sheared. Their threads were stripped. The body of the bolt played little part in holding the flange and hemi together beyond its ability to hang on to the threads. Such a powerful force needed to pop both hemi and hull from the flanges. Add to that the speed at which the forces are applied. Such an enormous amount of force applied suddenly is a 'jerk' phenomena studied by the US Navy some years ago. If you apply a steady force up to and past the, in this case, tensile stringth of your load carrying material, you get a textbook fail. But if you apply that much force suddenly, the material will often break because the jerk effect will overpower the material's ability to cohesively stretch to the required level of strain.

With each dive, microfractures would have been forming in the adhesion joint, combined with hard to detect delamination. Once the pressure worked through on a single path through the joint, the pressure would have ripped it apart from the inside out, followed by the compression of the implode.

Sounds much better thanks. Could be louder, otherwise it's a big improvement.

@JoediyLab

11 ай бұрын

Yes, will do

@creakinator

11 ай бұрын

Yes, louder.

Titanium expands about 6 times faster than carbon fiber with increasing temperature. If the interface material between the Titanium and the Carbon fiber was expected to be rigid, it would have to allow for flexing based on temperature cycles of about 25C difference maybe about 1 mm (based on hull size). As a side note, an alleged transcript of text messages came out that they may have been having trouble with a system "A" and switched over to their backup system "B", and their buoyancy was lower than expected, which would likely indicate sea water penetration, and they kept hearing cracking sounds aft.

@JoediyLab

11 ай бұрын

I will have to look for the transcript, have not read it. thanks

@jameskwon7617

10 ай бұрын

If that transcript is real (I've seen a video where it goes through the alleged transcript), then it shows that the sub and mission was FUBAR very early on. The descent rate based on the message chatter put them around 1000M deeper than they should have been for the time elapsed. The reference to A vs B was regarding the battery. Apparently, the primary bank of batteries were drained or low. So they switched to B. There was mention of cracking noises in the transcript. The transcript also indicated that they had dropped their ballast and were terminating the dive. However, their ascent rate was very slow. They also mentioned that they tried to drop the "frame" which might have been the metal skids that was attached to the submersible, which was attached to the hull via a metal strap of some kind. Now, I don't buy the water leaking aft argument. There is nothing aft but the other half of the carbon fiber tube and the Ti endcap (hemisphere). The flaring you see at the end of the submersible is not the pressure vessel. It's an empty void that is not supposed to keep water out at those pressures. Not sure why they needed a faring...it's not like this thing is going 30 knots underwater, but it's not like water was supposed to stay out of there and it didn't. And any leak causing water to ingress the pressure vessel would have come in at extreme pressure and force. The messages wouldn't have been so nonchalant. If the transcript is legit (and there has been no verification yet of this), then it doesn't lead defintively to the cause of the failure, but lends more circumstantial evidence that the cracking noise was a sign of immiment failure of the pressure vessel. Oh, the transcript also shows that some kind of "warning" indicators were all lit up. Was this the revolutionary hull noise detection system? I don't know. As for the very slow ascent rate, I'm thinking perhaps the ballast did not properly separate from the Titan as well as the frame. The system all did seem jury-rigged.

@sapienspace8814

10 ай бұрын

@@JoediyLab I'm quite interested in how the Titan titanium to the carbon fiber "glue" got cured, found some interesting material: "It is essential to pay attention to the thermal expansion behavior when bonding different materials, otherwise stresses will arise in the material that can lead to fractures or weakening. The curing temperature should be kept low, otherwise the outer shell could be subject to thermal damage." I'm pulling signals from my neurons and from one video I'm recalling Rush said something like he put this "peanut butter" glue system in an oven for "curing", this seems to be the polar opposite of what is recommended above...

@JoediyLab

10 ай бұрын

@@sapienspace8814 Thanks for the information, Do you have a link on the curing, I would love to read it.

@starkysorcha2310

10 ай бұрын

@@jameskwon7617 The warning indicators were from the accoustic system (again assuming the transcript is real) they used an accronym which I think was RTW or real time warning

I think they’re onto something there. I’ve been expressing great concern with the glue up process all along. Normally you don’t get a glue bond between two materials without some type of pressure and/or heat. And of course surface preparation is also very important. I just thought that was the most ridiculous thing I’ve seen on this thing since the very beginning. Gluing carbon fiber to a titanium ring. - Sort of like gluing an ice cube to a butter patty in the summertime.

@JoediyLab

11 ай бұрын

Thanks for the comment

@n9zmn

11 ай бұрын

I love your analogy of the ice cube and butter. I wonder, however, how critical the carbon fiber to titanium bonding is in the undersea world it was designed for. Wouldnt all that incredible hydraulic pressure be forcing the end caps IN to the carbon fibre tube? In that case, the flange relationship would be more important than the adhesive bonding. Does that make sense?

@kinte1870

11 ай бұрын

​@n9zmn No it wouldn't push the end caps into the tube. The pressure on the pressure vessel is greater because it's larger .

@vertihvost7675

11 ай бұрын

Do you know what kind of battery they used? 12V lithium?

@craigsowers8456

11 ай бұрын

@@kinte1870 not exactly ... hull imploded first IMHO ... and we both know "Water" cannot be "compressed" so it shot the end caps off like Champagne Corks. And given the videos, it would appear Rush didn't bother to even add the requisite "Primer" between faying surfaced ... otherwise, there'd be some small fibers left on the Titanium flange surfaces. Again, IMHO.

Great video. I’ve been wondering ever since this happened, “How in the heck did they think gluing a submarine together, was going to work?!”. You’ve throughly answered that question. Good job.

Be lucky if that cabin fibre edge goes 2 inches into the titanium ring. Fancy risking your life on 2 inches of a material that wasn't up to the job.

I read a study on the bonding strength of different materials with different prep. I have always found sandblasting metals is preferred and the study confirmed this with titanium. Also a serious and thorough cleaning process so no contaminating residues at all. They were using rags to wipe titanium that might have chemical residues on them and also was there any abrasive prep on the titanium. I am confident the bond was the failure point.

@JoediyLab

11 ай бұрын

Yes it appears so, but we do not know if the carbon failed first or the bonding failed first. We will have to wait for the report, Thanks for the comment.

@StagArmslower

11 ай бұрын

the CF hull's ends were machined for straightness; I wonder if the presents of dust would weaken the joint between it and the flange.

@upperleft9502

11 ай бұрын

They also weren't wearing gloves when cleaning the mating surfaces. He's even leaning his bare forearm on it. Remember kids not to Rush things. 🤠

@gailmcn

11 ай бұрын

The compression force at that depth was also acting on the two end caps, providing some resistance to separating from the hull in at least the lateral plane. It is impossible to determine if the bond was the first point of failure, until examination of the internal side of the flanges for evidence of the type of tear-away, and remaining epoxy, if any. The interior of the flange is not visible in the photograph from the wreckage. Also, no one knows the composition or properties of the epoxy used, but we do know the composition of carbon fiber, and its properties in high compression environments, and what happened with the first CF hull Rush used. So at this early stage, being confident that the failure point is the epoxy bond is a bit like hearing hoofbeats outside your window and thinking zebra, instead of horse. A slight possibility, but far more likely it's a horse.

@Slav4o911

10 ай бұрын

@@gailmcn The glued bond just looks too shady to not be a concern.... but then again.... everything in that sub except for the titan hemispheres looks shady.

I love epoxy, it is my go to product when I am looking for a disastrous outcome

@JoediyLab

11 ай бұрын

Actually the commercial aircraft you ride in has 20 - 35 Percent of carbon fiber parts.

I just noticed in a photo of Stockton Rush (CEO of OceanGate) sitting in the Titan sub. What caught my eye is that one of the bracket for mounting the digital display unit looks to be screwed into the expired carbon fibber hull. If I was 6000 lbs per square inch that’s where I’d go to test the integrity of that sub. R.I.P. to all those who lost their lives.

@AORD72

11 ай бұрын

Yea, surely that is a crazy thing to do, unless that was a internal liner?. Although I think the carbon tube was a thicker than the mathematics said. To support a monitor like that you probably want at least 10mm deep bolts, I think the wall was about 127mm thick.

@andrewcheatle4691

10 ай бұрын

I read ( sorry forget where but it was a legit source I believe ) that there was a glass fibre based inner liner to protect the internal electronics & minimise condensation issues - this was possibly the base on which were mounted internal components.

On the recovered flange (starting at 1:11), what you’re pointing out is it’s outer circumference, which is not the bonding edge.

@iamtomkills

11 ай бұрын

Correct.

@4epa1012

11 ай бұрын

Yep

@stitchem7

11 ай бұрын

On the titanium ring are found two flanges, an inner flange and an outer flange, also a flat surface that is perpendicular to both inner and outer flanges. I call this entire configuration the bonding edge of the titanium ring, as the other edge of this ring is bolted to the forward dome. The bonding surfaces are the I.D. of the outer flange, the flat perpendicular surface, and the O.D. of the inner flange. So, to be clear, I did not refer to the titanium ring as being a flange, I was pointing out the orientation and location of the flanges that are part of the overall titanium ring.

@Slav4o911

10 ай бұрын

@@stitchem7 I wouldn't use glue at all.... I'll just use titanium cylinder between the two titanium hemispheres.... either it'll be bolted or welded. I would not put any glue between me and 360 atmospheres.... even then I would test it a 100 times underwater at twice the depth. If everything is OK then I might get down.... one thing I know from life, things will always get wrong and when they get wrong 2 times overprovision might not be enough.

@stitchem7

10 ай бұрын

@@Slav4o911 Exactly my thoughts, if they would have used a titanium hull/tube, the passengers would probably still be alive. Yes, bolted with proper seals, or welded.

Hi Joe, Sounds great! A tad more volume would be great. Great analysis. The sounds described in previous dives, and if that transcript that is everywhere is legitimate, would suggest microfractures from air spaces in the carbon fiber hull, which seemed to be wrapped like thread on a spool with no cross-fibers. A collapse of the carbon fiber, structural hull would seemingly be a likely cause. In the transcript, they also mentioned RTM hull monitors were "all red" shortly before the implosion, suggesting a hull failure was eminent. The bonding between the titanium rings and the carbon fiber hull was certainly a point of contention, yet there were many. FYI, The USCG has the remains of the Titan vessel in their possession and will seemingly be conducting the analysis. The lack of any pieces of the carbon fiber hull suggests that during failure, it perhaps turned into chicken downy fluff-like pieces. The plexiglass view port was also only rated to 1,300m and is also a potential primary point of failure. The winding of the carbon fiber seemed to NOT be vacuumed or autoclaved, which, if the case, would've left small air bubbles in the wrap, which under pressure were likely failing, as microfractures would have been the sounds the prior passengers and crew were describing, yet certainly not the carbon fiber "settling". That should NOT have been ignored. Despite the additional implosive stresses and heat placed on the structural parts, the titanium rings and caps were still intact. The Locktite adhesive is amazing, yet if not properly applied, such as being off-center, (thin on 1 side, thick on the other, as there was no centering of the tube with centering of the rings that I saw in OceanGate's assembly videos.) Lack of abrasive, lack of chemical bonding, and even improper mixing are also additional concerns. Hardware had also been swapped out, and there are many iterations of this submersible. If the coms are legitimate, the Titan was heavily negatively buoyant upon the beginning of the dive, as their rate of descent was already faster than was planned and accelerated past what their sensors were reading, in an uncontrolled descent. The reported coms say they were at 2960m at 9:17am, then at 9:28am were at 3433m! They had started off at 28m/minute, next depth report they were at 39.46m/minute, then at 43m/minute on their last report before ascending slightly before the implosion. Their rate was reported at 9:30am to be 35m/minute, yet they had already exceeded that rate. Their planned rate of descent was 25.33m/minute if calculated at 3,800m taking 2.5 hours to get there. They were in an uncontrolled descent, which would've caused an increase in stresses on the hull, above and beyond the atmospheric pressure, which is calculated at 1ATM (atmosphere) of air pressure at sea level (14.7psi) + 1 ATM for every 33 feet of seawater (or 10.0584m). At 2.4 miles or 3862.426 meters (12672 feet) they would've been at 384ATMs or 385ATAs (Atmospheres Absolute, as you add 1 atmosphere for air pressure.) At 14.7psi per atmosphere, at the top of the Titanic as reported at 2.4 miles, the ambient static (still with no current, no descent or other stresses) pressure would've been 5,659.5psi. Their reported deepest depth was 3,476m (11,405 feet rounded up to nearest feet) or 346.6ATAs, or just over 5,095psi of static pressure, not including other stresses on the hull. I hope you find this helpful! Please feel free to share. My hope is that young people considering a career in underwater research, engineering, etc. will learn from this accident, and will take the physics of diving very seriously. Physics doesn't care if you don't go home at the end of the day. Thank you for the explanations of the adhesives. As for the "carbon fiber", there are many types. It's been reported that the carbon fiber used was expired aircraft carbon fiber manufactured by Mitsubishi. I haven't heard what type of resin was used. Again, no autoclave or vacuum seemed to be used.😮 I didn't see any crossthreads/mesh used in the manufacturing promotional videos. It appeared as though it was wound semi-uni-directional, like thread on a spool. Prospective and new engineers PLEASE take your projects seriously! People's lives are depending on you doing an excellent job. Never be a "Yes-man"! If somethings wrong or doesn't seem right, speak up! That will limit the lawsuits as well.

@JoediyLab

10 ай бұрын

Thanks for the comment, Yes I will increase the volume.

@vibratingstring

10 ай бұрын

In a small consulting firm with leaders with integrity who are hired by thoughtful owners who trust the intentions of the consultant, you can press to make sure things really are correct in the best technical rather than the narrow definition from applicable regulations. But if anything deviates from that--large corporation, dodgy bosses or dodgy customer, you will simply get fired. Fortunately I have worked most of my career for small firms with integrity. On more than one occasion we have steered the design to being more thorough than the letter of 33 CFR and 46 CFR would require. In the case of ferry design, we were literally ahead of the USCG re-evaluation of average passenger weight. The old and at the time applicable was 140 lbs for protected routes. We used 180 lbs based on our own analysis of population statistics. The new USCG rule is very close to that! Why did this matter? Because ferry stability is what makes the difference between passengers arriving safely in port, versus drowning like the ones in Korea a few years ago. Engineering matters. But in the total scheme of things, most of the time, engineers end up constrained by the rules game they are playing inside of. This is why in passenger vessel safety, we ended up with needless disasters such as the ESTONIA and the SCANDINAVIAN STAR as well as to some extent the HERALD of FREE ENTERPRISE, and of course the overnight dive boat CONCEPTION in California.

Why did the end caps separate from the rings, they were bolted to each other

@JoediyLab

11 ай бұрын

That is a good question, I do not have an answer for you.

@backfromcuba

11 ай бұрын

Big shockwave, tiny bolts.

@danieldupuys2002

10 ай бұрын

Diesel gate! boom!

@JoediyLab

10 ай бұрын

Massive force at depth

My motto is "better living thru 3m". I use 3m products probably 300 days a year. Maybe more. But, there is no "glue" that anyone has designed that will withstand heat at 1 atmosphere, cold at Titanic depth, sea water proof, able to "glue" titanium to much of anything, able to not get pressure washed out at that depth, and be able to recover undamaged from repeated compression/decompression cycles. How rich would I be if I could invent that? And yet goofball here thought this was a great idea.

I feel that another couple of important things people aren't understanding, are the effects of very cold temperatures that the Titan experienced at the depth of Titanic. In my experience, any, if not all, adhesives become brittle at extreme cold. When you combine this with 5546.98 psi (being in salt water as opposed to 5403.09 psi in fresh water), plus a 300-400 mile towing of the Titan, put further fatal damage on this extremely fragile vessel.

@JoediyLab

11 ай бұрын

yup, my mistake I did not check the correct box, Yes valid point about the cold temp.. Thanks for the comment

@johanea

11 ай бұрын

It is above freezing no matter how deep you go? What extreme freezing temperatures are you mumbling about? However, it is entertaining to see all “experts” here. Good laughs indeed.

@Roger_and_the_Goose

11 ай бұрын

@@johanea I've been using composites in the industry for fourty years, I'd put my experience up against yours any day of the week, Einstein

@johanea

11 ай бұрын

@@Roger_and_the_Goose There is no need for such. Calling water that is 2-3C extreme freezing temperatures is enough to get an idea about your experience.

@Roger_and_the_Goose

10 ай бұрын

@@johanea I get the feeling you really must be off your meds, roll over and go back to sleep mate.

Again, well done Joe. Very clear, informative. My takeaway: Examination of the bonding materials and processes used by the hull manufacturer could yield a relatively quick determination of probable cause. Thanks

@JoediyLab

11 ай бұрын

Thanks for the feedback

Joe, have you seen "Forensic Engineering & Failure Anaylsis (sic)" channel's coverage of the Oceangate Titan? He'd done some unbelievable detective work combing through all video's that have been done on the Submersible especially it's construction. He has a strong theory that the Carbon Fiber hull failed where it connects with the Aft Titanium Ring. His theory is based on the frame construction in it's tail section that housed batteries and other components puts undo pressure on the aft section of the carbon fiber hull. He's gone through pictures of that section over time and points out 3 major design changes they made to that weight bearing metal frame over the past 3 years. Forensic Engineering & Failure Anaylsis (sic) channel has been combing though the evidence like Matlock he's even called the manufacturers of these different carbon fibers to figure out exactly which was used and how it was improperly used to create the ships Hull. If the "leaked" final text messages are genuine they support his theory as Stockton Rush last few messages were about loud noises from the Aft section of the Hull. Also he texted it took multiple attempts to drop the metal frame (only part of it actually dropped). In the leaked messages Stockton Rush is having tons of trouble with buoyancy. Even after jettisoning the metal landing gear the Titan is only able to rise 19 Meters in 5 minutes. Before this years dives Stockton Rush covered the entire hull in Rhino Liner. Rhino Liner has never been tested under extreme pressure and may have delaminated the Rhino Liner and allowed tons of water in between the Rhino Liner and Carbon Fiber. This additional weight would have killed the Titans Bouancy making it impossible to rise back to surface even if it didn't have a catastrophic implosion Again these are not my theories they are brilliant thought from the channel "Forensic Engineering& Failure Analysis" (sic). Does the bonding agent really make much of a difference with all the pressure pushing the hull together I'd imagine you could bond that thing with Vaseline and it would stay together with the water pressure keeping everything sealed. Cheers!

@DaHitch

11 ай бұрын

His videos might be interesting, but keep in mind he also explicitly states that "he doesn't believe in gravity", before going on to demonstrate how he completely misunderstands the concept of Newton's Apple. So I'd take any conclusions he draws with more than a few grains of salt. Having said that, water between the Rhino liner and carbon fibre would add as much weight as the water between the fairing and the hull. Nothing, while the sub is under water.

@JoediyLab

11 ай бұрын

Thanks for the comments. I will have to watch the channel you mentioned.

@chunkybuttz844

11 ай бұрын

I would hesitate to accept that channel’s information. I too began watching them but the man is tough to follow as he is very erratic and applies an elementary understanding.

@SoulDelSol

11 ай бұрын

Not if the cylinder buckled in and thwop pops out lid

@SoulDelSol

11 ай бұрын

That may explain why they sunk so quickly too. Rhino liner wth tho

Ocean gate seemed very price sensitive, so that might have been an epoxy selection criteria… although better epoxy would likely not changed the outcome. Also, the CEO seemed to resist people telling him flaws in his plans… so he likely would have discounted research that conflicted with his decisions

@JoediyLab

11 ай бұрын

It appears so, it might have even clouded his judgement . Thanks for the comment

@yachtcaptain7320

11 ай бұрын

I might be wrong ... but the glue they employed looks awfully like JB Weld.

Audio sounds great maybe turn up the gain just a touch. Thanks for another great video!

Wow, great information on the bonding process between the carbon fiber and titanium elements. I had flashbacks of diving on my submarine and hearing the hull creak as we went below certain depths. I am not an engineer, but these were the forces that made me concerned for the bonding process. With each dive/ascend cycle there had to be a case hardening type of effect that would eventually weaken the bond. I'm sure it's a case of no balls, no blue chips. The lesson has been learned. RIP my fellow submariners.

@JoediyLab

11 ай бұрын

Thanks for the feedback

@TechWizMaster

10 ай бұрын

you know once under water they wouldn't even need to have the two pieces glued together...the pressure of the water alone will make it impossible to separate the caps even if you'd try...so the glue as NOTHING to do with all this.

@user-jb5pp5rw9u

10 ай бұрын

@@TechWizMaster It's likely that reasoning was taken as 'innovation'. It's not a bug it's a feature! Why make a big deal of holding the flange to the tube when we all know the only way that glue will fail is if there's a hole in the hull at depth. Kind of like the acoustic sensors in the hull that tell you the hull is failing. 'Everything's in the red,' is not a useful diagnostic alert. Or that it won't matter if you're on the surface and you can't open the hull without unbolting the hemisphere 'because you're already dead.' Dead by design. Squashed like a roach in a corner nobody could reach without pointy shoes. This is some of the most bizarre thought paths for an engineer. It's such a blind dependency on the numbers and the strategy that the alarm is just there to let you know you failed at a critical point. What you gonna do about it.? You're already dead.

the more clips of this the more I cannot imagine this not being some kind of joke or psyop. I have good working knowledge of composites/epoxies/carbon fiber...anyone who does any amount of study would find this design to be a widowmaker in about a day or 2. The fact that it made it 3000+ meters is nothing short of a miracle. The design might work if several changes are made.

@JoediyLab

10 ай бұрын

Thanks for the feedback

@cragar71260

10 ай бұрын

From a retired/disabled mechanical designers view I just CANNOT understand why stockton did not see that this adhesive bond alone was TOTALLY inadequate. Carbon fiber, yes is a great material but not without the what... 8' span of the hull being supported.i would have drawn alternative drawings showing titanium rings spaced evenly apart along with horizontal titanium supports running horizontaly evenly spaced about the dia of hill that are connected mechanically

Well done, Joe. A significant improvement with the new microphone. +1 sub. 🎉

@JoediyLab

11 ай бұрын

Thanks

Sounds much better Joe thanks. I’m enjoying your analysis and videos 😊

@JoediyLab

11 ай бұрын

Thanks

Very informative! Thank you for sharing your knowledge. You are an excellent teacher!

@JoediyLab

11 ай бұрын

Thanks

(Mistake on bonding edge.) The flange you speak of is on the 'bottom' side in the photo at 1:40 and can be seen that it was partially ripped away by the implosion. The edge your cursor is pointing at is the bolted flange. You can tell the boning flange is the one that has the large radius leading to the body of the ring as in the photo of the ring hanging by the rigging before bonding. It appears that both of the 2 inch overhanging flanges of the bonding edge have been ripped away and in one of the recovery photos you can see a deformed section of the bonding edge flange still attached to the ring.

@JoediyLab

11 ай бұрын

Thanks I will double check that.

@johndraper3210

11 ай бұрын

@@JoediyLab *EXACTLY... I was going to point that out to... Not really... I have no idea what a flange is🤷‍♂️🤷‍♂️I'll see myself out*

@oldmech619

11 ай бұрын

Thanks for pointing that out.

@holdernewtshesrearin5471

11 ай бұрын

The bonding surfaces cannot be seen in the picture. The titanium sealing ring slipped over the O.D. of the CF tube and bonded with about a 1" wide related edge at the O.D. and I.D. and about a 3.5" to 4" (the 5" wall thickness minus whatever was machined away to true and rebate the surfaces) wide surface at the end face of the CF tube.

@johndraper3210

11 ай бұрын

@@this_number 👊

Yes Joe the sound is spot on and thanks for the articles. Really appreciate your work.

@JoediyLab

10 ай бұрын

Thanks

This sub's design is only good enough for one trip and that's it in my opinion. Who in their bright mind, honestly thought that this design was sound enough to be used over and over and over again? Was there even a "Life span" estimate made to Titan?

@gailmcn

10 ай бұрын

@ftffighter given Rush's multiple statements about the "virtues" of breaking rules and ignoring safety concerns, it would not have made any difference. The companies involved in production of the hull probably did warn him of the dangers.... he would have ignored them, if he even told them the truth about how deep he planned to go.

@johnarcarese172

10 ай бұрын

Disagree - the sub was designed to go to 4,000 meters. Anything greater than 2,000 was dangerous, as we have seen. If it had been designed to go 8,000 meters, a trip to Titanic would have been far safer.

Interesting Joe. No way I would have got into that sub to go down 40M let alone 4KM. With a background in heavy industrial / petrochemical refrigeration I was surprised that if they built it in this way there was no through bolts holding the titanium flanges across the longitudinal length like a standard flanged component and perhaps these could have supported the CF tube by incorporating CF saddles or similar, I’m sure there’s a whole range of other issues like compression of the CF tube and you still need bonds or gaskets between, personally I’d prefer steel, stainless or titanium. If you haven’t seen it, the hydraulic press channel has done a pretty good model simulation of a carbon fibre tube with bonded steel endcaps, the failure mode seems pretty clear in that case that the CF reaches a pressure at which it completely and in an instant implodes and seperated completely from the bonded ends. Obviously, there’s variables and he doesn’t have the same materials, epoxy or can’t reach the full pressure but it’s interesting and they also attempt to demonstrate the effects of the pressure cycling. Check it out for yourself kzread.info/dash/bejne/dIV7ptl-Y9yzkpM.html

@JoediyLab

11 ай бұрын

Bolting through carbon fiber created isolated stress points around the holes. Bonding in the aerospace industry is very common, but you have to use the correct epoxy and the correct bonding process. Thanks for the comment

@gailmcn

11 ай бұрын

@@JoediyLab I think that was a key issue here, Rush's undergrad degree was in aerospace engineering. (Masters in business). I believe he used what he "knew" from aerospace, which has basically opposite and much smaller pressure differentials, from deep sea engineering concerns.

11:08 I've worked with Scotch weld it has HORRIBLE adhesion properties with metals such as aluminum and titanium. We used it as a bedding compound for damaged fastener countersinks in carbon fiber panels on F/A-18 Hornets. I'm trying to figure out why they did not have mechanical fasteners going through the flanges and carbon fiber. I understand that it is under compression while underwater. At least on the front hatch there's a fair amount of shear and tensile force on the flange when the hatch is open.

@JoediyLab

11 ай бұрын

Good question. Thanks for the feedback

@craigsowers8456

11 ай бұрын

Firstly, the idea is to NEVER put mechanical fasteners in CF structures if possible; by cutting the CF Tow, the strength of the individual Tow (where one drills thru) is diminished to the now length of that between fasteners; not the original length/strength intent by Design Engineering. But it really wouldn't have mattered as the Hull imploded first IMHO ... that 6,000 psi would have kept the dome/flange securely in-place ... until the Hull imploded and pushed them right off ... no amount of fasteners or adhesive would have helped ... "Water" does not "compress" and when it gave way, the Water had to go somewhere (out the 2 ends). Bad design and execution all the way around. Rush even stated in one video that the "Adhesive" was actually NOT for the stated purpose ... the intent was to act as a "Buffer" between the different expansion/contraction properties of Titanium and CF. And free advice on your F/A-18 cured panels you're countersinking ... put a piece of 3M approved masking tape over the area you intend to countersink and then, with a sharpened CS bit, drill. You'll never have another CS with edge "Tear Out". Works as well on "thru holes" where tear out happens on the back side. Trust me on this Bro.

@patrickradcliffe3837

10 ай бұрын

@@craigsowers8456 No, No, No these were repairs to the damaged countersinks after the removal of striped, stuck fasteners. We had a few younger Airframers that would go a little wild with a drill trying get stuck fasteners out. As far as not having mechanical fasteners we had mechanical fasteners going through structural carbon fiber panels attached to aluminum. I agree it does not help when being compressed at 6,000 psi, but it is not always at the pressure and my question still stands with regards to tensile, and shear forces applied to flange that has the door on it. When open that door acts like a lever on the adhesive joint twisting and pulling on it.

@craigsowers8456

10 ай бұрын

@@patrickradcliffe3837 Copy that ... thought you were talking "new build"; still good hip pocket info on use of tape. As the Door can only be opened on the surface (for ingress/egress of passengers), I'm sure there are forces exerted ... it's a damn heavy door ... so hinge line may take its' toll one adhesion of the flange to hull bond line ... but since Rush elected to not do ANY Ultrasonic (and X-Ray if it looked really bad on the scope), wouldn't know that answer; if you don't inspect, everything is Peachy ... until it's not.

I can only see the tube delaminating first, from all the radial pressure around the tube along with the endcaps pressing inward on it axially. It's the polar opposite type of application I ever would've expected carbon fiber composite to be used for, and the tube giving way would've allowed anything else to happen - the titanium ring(s) separating, etc...

@JoediyLab

10 ай бұрын

Thanks for the comments

This is a very accurate analysis based on evidence and testing results. This is exactly how we should analyze this unfortunate disaster. One thing, I noticed when Stockton Rush was talking about the Bonding Process, he stated..."The thickness of the Bond must not be too thick or too thin". The only way to determine what this thickness should be is through Bond Testing. QUESTION1: Where is this Test Data QUESTION2: How was the Bonding process controlled to ensure that the EXACT THICKNESS was achieved. It looked to me as though this was just "eye balled".

@JoediyLab

10 ай бұрын

There is no published test data that I can find, Oceangate was using bondo spreaders to spread the adhesive I did not see any thickness gauges being used. Thanks for the feedback

the titanium ring needs to be etched in hydrofloric acid and then anodized before bonding.

@martinwhite418

11 ай бұрын

No, the way to have prevented this would have been to hydrochloric acid etch Stockton Rush.

@wshyangify

11 ай бұрын

@@martinwhite418 Give that man a break, he's currently under tremendous pressure now.

@martinwhite418

11 ай бұрын

@@wshyangify Baahahahhahh.

@kevinkelly2513

11 ай бұрын

I thought the acid etch was to passivate the titanium, and prevent galvanic corrosion with the carbon.

@huntandrew4409

11 ай бұрын

@kevinkelly2513 The acid etch will remove pasivation and make the surface rough, if anodized will make a thick oxide moleculer layer,easier to bond to. This is very common with aluminum with painting And the chemical cleaning will make surface clear from oils,well until someone puts hands on it. That part of Shiney titanium

Looking at the epoxy joint depth and sizes of the viewport/end cap bolts, it looks like oceangate was depending on the external hydrostatic pressure to hold the joints together.

@JoediyLab

11 ай бұрын

That would be very likely.

@martinwhite418

11 ай бұрын

Yes, but what pressure do you start with? Anyone who has used wood glue. Too much pressure, and you squeeze the glue out, and you have no bond. Too little pressure, and your joint looks awful. Car mechanics, valve cover gaskets, and are designed to be uniformly tightened to seal the valve cover. But you don't go diving with your car.

Very informative scientific review of possible cause. Do more please increase the volume.

@JoediyLab

11 ай бұрын

Will do

New mic is a great improvement 👏. Ive enjoyed your set of videos on this sad affair, engineering is always important, no exception. Its certainly possible that the end caps to cylinder components bonding failed during build, or first handling during build; the end caps being spigoted, would stay in place without bonding? So often, a precision assembly is lifted, moved, handled by the lowest paid. Many thanks for sharing your knowledge and insight. Best regards from the UK. John.

@JoediyLab

10 ай бұрын

Thanks John for the feedback

Pretty familiar with a adhesive like that and it is used in aerospace applications. I used it for avionics racks many different type of throttle quadrants, And many boeing parts even f22 raptor stuff I think its a very popular aerospace adhesive.

@JoediyLab

11 ай бұрын

Thanks for the comment, Yes Epoxy adhesive is very common in the aerospace industry

@vertihvost7675

11 ай бұрын

Loctite 9394?

@craigsowers8456

11 ай бұрын

... and I'd have made sure you used a "Primer" before applying the Adhesive right? (I was on the F-22 Program as well and the Primer is clearly listed in the job cards). Evidently, Rush skipped that step ... but really wouldn't have matter ... the Hull went first IMHO and no amount of Primer or Adhesive would have saved this vessel.

@stompinmcallister1312

11 ай бұрын

@@craigsowers8456 Agree

@stompinmcallister1312

11 ай бұрын

@@vertihvost7675 No it is a two part epoxy almost like JB weld . Mixed if my memory serves was a 2-1 mixture by weight. The place I worked at got it from McMaster car

Agree, but further when the stress is applied (in fact very high stress) there will be differential strain between the Ti and composite. Cracking of the joint will occur as a result of each dive with water ingress. Failure becomes inevitable.

Very clear audio just run it through a compressor audacity has one. Great info

A nicely presented analysis of this problem with the adhesion. I think the lack of any remains of any carbon fibre anywhere is a real red flag. The known fact that the submersible survived previous trips to the Titanic indicates that the bond was sufficient, that the carbon fibre was sufficient for a given number of trips or service cycles. Fatigue was the craft's downfall and gradual delamination of the carbon fibre layers meant that the hull failed as shown in tests here on KZread, as it collapsed it twisted very fast as it contracted for a very tiny amount of time, probably shearing the glue right off the end flange with such force that it almost looks polished. In short, the end caps couldn't keep up with the rest of the hull as it twisted during disintegration. The overall design was terrible, that was to blame rather the bond itself.

@JoediyLab

10 ай бұрын

Thanks for your feedback

Yes, Composite fiber has relatively little compressive strength, so is a fundamentally flawed sole material that leaked, causing dieseling within the cabin.

@JoediyLab

11 ай бұрын

Thanks for the comment, more than likely.

@MarkShinnick

11 ай бұрын

@@JoediyLab Max frustration, apparently, for all attempting to get through to this guy. There's only so much reasoning power remaining once someone sinks so much bias invested like he apparently had. Thanks for your patient style of explanation that helps people visualize the physics.

@MarkShinnick

10 ай бұрын

@@JoediyLab Some clarification: The fibers of the Unidirectional tow windings had no apparent fiber to bind them; all that existed is epoxy. Progressive splitting with water intrusion, and operational friction rubs, between the unreinforced wound carbon tows would produce the reported sounds. The tube itself could thus have essentially split.

I think the portal assembly blew out at the onset of implosion but the pressure was still increasing and the end caps probably ejected from the carbon fiber as it was crumbling

@JoediyLab

11 ай бұрын

It appears so. we will know for sure after the investigation is finished. Thanks for the comment

@iuhsdihdslifuvholuidfh

21 күн бұрын

That makes the most sense, There had to be internal pressure to do that.

Your audio is a vast improvement! Thanks.

@JoediyLab

11 ай бұрын

Thanks

New mic sounds great, Joe.... love your vids on this topic!! Makes me miss school, You could be a professor

@JoediyLab

10 ай бұрын

Thanks for the complement.

So… How did the domed end caps get separated from the flanges? They were held together by 17 bolts… If those got separated by the implosion, then the adhesive pieces never had a chance to begin with…

@luciano2166

11 ай бұрын

good question. i want to know the answer too...

@MissX905

11 ай бұрын

likely needed a 100 bolts but that would have only saved the end cap not the little viewing window. Plus what were the bolts' strength that he used? How old were they? How tight were they torqued onto the rings and other part? What glue was being used, was it ok to be used where a vessel was always in water? Was the glue new or had it been sitting around for a bit like in some overheated area, etc etc?

@luciano2166

11 ай бұрын

@@MissX905 The CEO probably didn't care about the torques and provenance of the adhesives or the quality of the screws. Probably everything came from scrap metal.

@timjackson3954

11 ай бұрын

If the adhesive was just serving as an o-ring seal, always in compression, then that's not a problem. That the shock-wave of implosion sheared it off is unsurprising but irrelevant. There are a couple of possible failure modes though. One is that the CF tube flexed enough to part the seal and the sealant was too rigid to follow it. The other is that the sealant flowed under pressure and blew through. Neither of those modes would seem likely to implode the CF tube though, and would be relatively slow. See the Hydraulic Press channel's example where his sample tube seal failed: it just quietly flooded.

@JoediyLab

11 ай бұрын

That is a good question, I do not have an answer for you .

If you watch the video of the "fire piston" (link below), it provides a possible explanation for what might have happened to the combustible materials inside the submersible. In my opinion, it appears that the hull and everything within it vaporized. To confirm or challenge this theory, it would be helpful to examine the interface where the hull once met the titanium ring. Specifically, we should look for signs of heat scoring or shearing, as well as any indications of air bubbles where the failure occurred. In a typical diesel engine, air is compressed roughly 14 times before fuel is added, and it ignites instantly under those conditions. Now, imagine if the compression were 400 times! When the implosion occurred, it generated extremely high temperatures, potentially surpassing those found on the surface of the sun. While I calculated the temperature to be around 120,000 degrees Kelvin, I find it hard to believe that it reached such a level. Nevertheless, it must have been exceedingly hot inside the submersible. In a diesel engine with a 14:1 compression ratio, all of the fuel burns. Gasoline engines typically have compression ratios ranging from 8:1 to 12:1. Although I'm not an expert in automotive fuel combustion, as a well-informed layman, I understand that higher compression ratios result in more efficient fuel consumption. With a compression ratio of 400:1, it is highly likely that everything within the hull was incinerated within an extremely short duration of approximately 0.0038667 seconds. This calculation is based on the following: the length of the tube was 5.8 meters, and the speed of sound in water is 1500 m/s. Regarding the delta air volume (pressure constant), we can consider 400 times 300 degrees K, resulting in 120,000 degrees K. By compressing a fixed volume of air into 1/400th of its original volume (pressure of ocean constant). We can apply the Universal Gas Law (PV = nRT) in this scenario. I hope that someone will point out any mistakes or provide additional insights. I have made some assumptions to simplify the calculations, so I acknowledge that the actual situation is likely more complex than what I have presented here. I look forward to hearing your thoughts. Thank you.kzread.info/dash/bejne/domguceheJvJeJM.html

@JoediyLab

11 ай бұрын

Great info thanks

@gailmcn

11 ай бұрын

@SciHeartJourney. I also thought as you did initially, based on a typical gas compression scenario in an implosion sequence. Then when I saw the photos of the wreckage, I was surprised that none of it showed signs of scorching or even heat deformation, so I commented on that. A couple people with some experience pointed out these factors: 1. the total volume of gas was small, and compressed by the extremes at that depth, it was extremely small. 2. the gas compression occurred in a very large volume of extremely cold water, and at that millisecond or so, in what was no longer an enclosed hull (global fracturing of the CF hull). They both said that any compression/heat/ignition phase, if it occurred, would have negligible effects in this situation. (The exchange is in the comments somewhere on the BBC News episode on video of the transfer of the wreckage at St. John's).

@cecilbrisley5185

11 ай бұрын

Ever waved your finger through a candle flame? Duration of that heat is important. It was surrounded by and instantly quenched by a huge mass and volume of sub freezing material. Similar to how a 2000 degree spark from a sparkler does nothing to you or your clothing. Difference between heat and temperature. Alot of temperature, not very much heat in contrast to the surrounding materials.

@SciHeartJourney

10 ай бұрын

@cecilbrisley5185 Thank you for so clearly demonstrating the difference between heat and temperature. 👍

@Slav4o911

10 ай бұрын

I think you're right, the temperature was so high everything inside burned in an instant, along with the people inside.... that's why no bodies and no parts of the carbon fiber cylinder can be found. The more I look at what happened, the more I think it was a suicide. If you want to kill yourself and burn to dust yourself at the same time in the fastest way possible, that's one of the ways. It looks like some ritual suicide.... but why take innocent people with him.... He knew exactly what he was building, an underwater self combusting incinerator... just looking like a submersible. He was probably hoping it would implode the first time he goes down... but the thing magically held up... that's why there was no need for any safety.... because it was a death trap. People think these things happen only in the movies.... that's why everybody was so careless. That's why nobody paid any attention to the gluing process. It wasn't build to held up.... look at all the other submersibles and how they are build, especially the pressure chamber. But this.... this was build with little glue hoping the water will held it together... nah.... it was constructed as a death trap from the very beginning.

0:18 WAY WAY better audio quality!! Thank you!!! 👍🙂👍

@JoediyLab

11 ай бұрын

Thanks

@PoorMansChemist

11 ай бұрын

@@JoediyLab Anytime! Great video as always. I found the mention of van der Waals forces as the bonding force for some of these adhesives a bit odd. vdW forces arent noted for their strength. If it were me Id rather trust my life to some sigma and pi covalent bonds between the adhesive and the substrate. Surely there is something that will react with titanium that can also he made to bond to some other polymer matrix.

Thankyou, very interesting and the audio is perfect..

@JoediyLab

10 ай бұрын

Thanks

I can't imagine using adhesives when you are dealing with 6000 psi. Maybe an outer titanium tube welded to the titanium end caps, with an inner tube of carbon fiber reinforcement that's replaceable. But it sounds like carbon fiber has micro damage every time it is cycled and there is no way to determine when it will go catastrophic.

When the sub was crushed, you had an energy release in the neighborhood of 47lb of TNT detonating (if I’m recalling the correct numbers). That released energy would focus itself at the boundaries between materials - the titanium would have been ringing like a bell struck by the worlds largest firecracker, using that energy to free itself of all bonded material. Not a materials engineer (I do software, not hardware), but I’d expect the rings to be free of adhesives after that, even if the bonding was sound.

@gailmcn

10 ай бұрын

Also the 5 day submersion in salt water may have had an effect

Scotch DP420NS-BLACK is my favorite epoxy. I specified it for potting portions of connector bodies on some audio equipment years ago. That non-sag property is amazing.

@JoediyLab

11 ай бұрын

Very cool, thanks for the comment

@martinwhite418

11 ай бұрын

Yes, maybe they should have put the people in the sub and potted to whole sub. Abaahahahahahaahha.

Good post, and it answered many questions I had about carbon fiber vs. titanium, esp. the bonding aspect. BTW, your sound was a bit low; I had to turn up the volume.

@JoediyLab

11 ай бұрын

Thanks will address the volume in the next video.

I totally agree that maybe the biggest issue was that bond and not the hull itself...maybe carbon fiber was not a bad idea...but the bond together definitely couldn't withstand that preasure.

@anthonytimpson4975

11 ай бұрын

nah the carbon fiber was a bad idea

@JoediyLab

11 ай бұрын

Yes not sure what failed first, the flange or the carbon fiber, when then blew the flange off. In the recovery video that appears to be no carbon fiber visible.

@MrNota500

11 ай бұрын

The list of what was done right is shorter than what was done wrong. Carbon fiber was definitely not the right choice. That's right from the Navy, rush said the Navy did experiments with composites and concluded they were not fit for pressure vessels. So he knew but thought he knew something the Navy didn't. I think he brought his aeronautics engineering to the Ocean Gate projects and the two are completely different environments. What works in one does not apply to the other.

@luissoto5942

11 ай бұрын

@@MrNota500 that's all good...but in order to move on and perfect something we need to see what was done wrong and how can we make it better...yea he fucked up...but it's to late for that know...got to admire what he tried to do.

@gailmcn

11 ай бұрын

@@MrNota500 Thank you! Almost nobody acknowledges the profound difference between the basic issues in aeronautics and deep sea aquanautics. You can see in.the design of the Titan the influence of Rush's aerospace training, and see it in his use of carbon fiber. In fact, he only had an undergrad degree in aerospace engineering, with only 2-3 years total of professional experience as a test pilot engineer. His Masters was in Business Admin. This was not a guy who should have been running the production of a submarine vehicle, especially for such extreme depth.

Thank you for the clear video and great explanation of adhesive types. There would have been millions of pounds of collective force holding that bond against the carbon fibre cylinder in addition to the millions upon millions of pounds of compressive force radially on the cylinder. A possibility I have considered is that due to the extreme pressures the epoxy adhesive may have extruded itself into the end grain of the carbon layers and possibly played a role in initiating the obvious delamination process that would have led to the failure.

@MarkShinnick

10 ай бұрын

The fibers of the Unidirectional tow windings had no fiber to bind them. Progressive splitting with water intrusion, and operational friction rubs, between the unreinforced wound carbon tows would produce the reported sounds. The tube itself could thus have essentially split.

@JoediyLab

10 ай бұрын

Thanks for the comments

Great Video. Thank you, Titan should have been purposed for a coastal, low depth sub for tourists to see fish, reefs, and an eel living in a old pirate boot. Ideal for a low cost sub. Deeper you go, the deeper you need to go in your pocket.

@JoediyLab

11 ай бұрын

Thanks for the feedback

Excellent analysis and video. Thank you.

Watching the video of them gluing this up in a warehouse, it’s obvious that hair, dirt, air bubble, or moisture will contaminate the adhesive and there’s basically nothing controlling the thickness or consistency of the adhesive application.

I was senior chemist at British Aerospace in the 80s, where we built satellites. We had occasions to bond cfrp to titanium. Before bonding, we always grit blasted the titanium bonding surface. I remember having an Easter weekend ruined due to be called in to attend a senior level meeting following a failure of a Ti - cfrp component in acoustic testing [ placed in a test chamber & bombarded with some VERY loud pink noise, to simulate launch conditions]. When I looked at the [very clean] failure surface, there was no evidence of blasting, just very mild abrasion.

@JoediyLab

10 ай бұрын

Very cool information, thanks

Great snooping around glues!!! Thanks!

@JoediyLab

10 ай бұрын

Thanks

Perhaps a silly question but was the carbon fiber hull wrapped around an inner titanium hull? Or was the titanium only located on the end caps/rings?

@Steve-du2nz

11 ай бұрын

they wrapped it around a stainless steel cylinder from what ive seen.

@brinnpope3050

11 ай бұрын

@@Steve-du2nz Thank you for the reply!

@JoediyLab

11 ай бұрын

That is a mandrel for holding shape. Once the carbon fiber is cured they pull the carbon fiber off of the mandrel. Thanks for the comment

@christinecollins6648

11 ай бұрын

Wound around a metal form that was removed- so the carbon fiber stood on its own

@brinnpope3050

11 ай бұрын

@@JoediyLab - Thank you for clarifying. Wow, that’s truly frightening. I’ve seen a lot of crazy things happen with carbon fiber bike frames (cycling bikes, not motorcycles). Stress cracks that go unaddressed and then completely fail. Different situation and use, obviously, but still…yikes!

Keep up the great work, Joe.

@JoediyLab

11 ай бұрын

Thanks

Thank you for the great info and the pop filter. If you could also adjust your audio level it would be tip top. I have my volume maxed to hear ya. Thanks again!

@JoediyLab

10 ай бұрын

Yes , next video major changes we applied, Mic position and volume levels were adjusted.

Since the pressure is applied to the outside of the 'capsule', aren't the forces present pressing the end caps onto the composite tube as well as compressing the tube?

@backfromcuba

11 ай бұрын

They are. The problem is that the two materials don't compress equally.

@JoediyLab

11 ай бұрын

Correct, as we discussed in the first video carbon fiber and titanium compress and expand at different rates.

@stanley917

11 ай бұрын

@@JoediyLab not the point. The end cap is captured on the end of the tube by the flange on the titanium cap, the deeper it goes, the more force pressing the two surfaces together. The external pressure would keep it in place without bonding material (assuming perfect mating surfaces no leakage - bonding material in a sense compensating for surface imperfections, acting as a gasket) Latest released communications indicate they knew they were in trouble, the real-time monitoring system (RMS) warnings were active, they jettisoned ballast and the landing gear in an attempt to surface before the breech and collapse. Of course, it is the internet, all information is always suspect.

Great video. Sound was good but I did have to turn up the volume a wee bit.

@JoediyLab

10 ай бұрын

Yes working on mic position and volume level, thanks

9:54 When entering the depth figure the calculator was set for fresh water, as opposed to sea. Does that make any discernable difference in this exercise?

@JoediyLab

10 ай бұрын

yes a bit, my mistake, I usually set that, when recording the video I forgot

EE myself, half-century career in aerospace, defense, commercial air transport. We had a saying, modified slightly by me: “you can always tell a structural engineer. You just can’t tell ‘em much.”

@JoediyLab

10 ай бұрын

Very cool, Yes I have always heard that this is no such thing a a civil engineer. Thanks for the comment.

Im a Dr. Quantum Hyperdimensional Widgetology and the Chief Chaos Engineer of Rubber Duck Debugging. So, worry not, for my opinion carries weight not only due to my amusing title but also due to the depth of knowledge and experience that lies beneath it. I am here to share my expertise, engage in thoughtful discourse, and offer valuable perspectives on any topic at hand. During an implosion, the hull of the submarine is the component most likely to experience failure. The hull is designed to withstand the external pressure of the water, but when subjected to an intense implosion, the immense forces acting upon it can exceed its structural limits. The intense inward pressure can cause the hull to buckle, rupture, or collapse, leading to a catastrophic failure of the submarine's integrity.

the glue is used to keep pieces together at surface level. At a depth of 10m, the water pressure will push the dome, and the ring in between, against the cylinder with a 3000kg force. At the Titanic depth the force will reach 100000t and more. Under a similar load, the titanium ring act like a rubber seal under its own elastic deformation phase. The Titan submersible was designed to take structural advantage by the outside pressure, in keeping pieces tight together. The most likely cause of the disaster is still the inward failure of the pressure chamber, by collapsing of the carbon fiber hull.

@JoediyLab

10 ай бұрын

Thanks

@alessandromaffei4769

10 ай бұрын

10000t, not 100000t.

Sounds much better! Nice work!

@JoediyLab

11 ай бұрын

Thanks!

if everything was mounted to the titanium rings, what's the extra force the descent at high speed 30 instead of 2 or 3 (according to the texts) would have at the glue section where the engineering/life support equipment in back was attached on it. Being that there seams to only be a couple inch glue section where all the torque would be applied. And what is the top rated speed of the Titan?

@JoediyLab

11 ай бұрын

Good question thanks

Thank you for your contribution here. It's nice to see people digging up some references and old as well as newer papers on bonding etc and getting interested in the topic. However I do wonder what you are doing with your stress calcs. They don't make sense to me. The water pressure is not the pressure on the adhesive. It's more interesting than that. There are actually multiple aspects to the problem, and you've accidentally addressed one that isn't what you were getting at. 1. pressure differential across (radial direction) laminate: that is water pressure, the delta p that is trying to drive water through the laminate. 2. The hoop stress (circumferential) stress on the cylinder, due to the overall hydrostatic situation. 3. The longitudinal stress on the cylinder, due to overall hydrostatics , the end caps "pushing" on the cylinder 4. differential rates of contraction of cylinder and titanium end caps gives a shear stress along the bonded edge Note that 4 is potentially complicated for a variety of reasons. Your comparison of stresses is however incorrect, and you quote values without noting the differences between shear strengths and compressive strengths and tensile strengths. Typically if only one "strength" is even given it is tensile. But the other problem is whether you are looking at the strengths of the epoxy resin (neat) or epoxy compound (thickened) or a strength of a bond to a representative material. Properly reading bonding agent values is not simple! Assuming a uniform (metallic or other ordinary noncomposite) Hookean material, the hoop stress is simply p*r/t, and the compressive longitudinal stress is simply p*r/(2*t) assuming thin skins. In this case we are not quite thin--but for the compressive longitudinal stress it is derived very simply as pi*r^2*p/(2*pi*r*t) which simplifies to p*r/(2*t) Take all this and assume 5 thick and an assumed radius of 30" and you get: 6000 psi * 30" / 5" = 36 ksi hoop stress (if metal) and 18 ksi longitudinal compressive stress. Now, to calculate the bondline shear stress is beyond the scope of a youtube comment! And we don't know enough about the titanium geometry, or thickness. So we cannot figure out its behavior. We could make a rough estimate of the carbon contraction but it is all guesses. Note that the laminate schedule for the cylinder is unknown--the Composites World article doesn't match the appearance of the build in the video. Also there is tremendous confusion out there about how many carbon subs were built. My understanding is only one--cyclops 2--which was *renamed* Titan. There is some level of doubt about whether a second cylinder was built at some point. My gut sense is no on that. The laminate schedule affects the compliance--and therefore the deflections as well as actual stress state. The apparent schedule from videos has no longitudinal material. The deflection of the tube under hydro tatic pressure subjects it to potential surface buckling of the inside skin--due to the combination of the radial pressure of 6 ksi combined with the incipient euler buckling developing from end cap pressure.

@JoediyLab

10 ай бұрын

Thanks so much for the corrective feedback. Yes I was made aware of that and will be doing a followup.

Also great work on the microphone. 100x better!

Sounds Much Better. Great Content.

Boat builder here. I couldn’t believe this epoxy type adhesive application to the titanium. Anytime we laminate aluminum for strength, it is always roughed up with a grinder for better tooth. I didn’t see any of that on the flange of the titan. Also in the video they are laying up wet fiber in an overlap which is just not even close to the quality of composite you would get with pre preg carbon and vacuum bagging. Really surprised that death tube lasted several dives.

@JoediyLab

10 ай бұрын

Thanks for the feedback

Would a longer rim on thee flange overlapping the outside of the carbon fiber cylinder have improved the stability of the bond?

Stockton Cheap skated this vessel to almost no end,,it was like having a bumper car on a NASCAR race track !!! & He knew it but he just didn't care bcus he probably thought it would hold up enough to give him time to construct a better submersible.

@JoediyLab

11 ай бұрын

Thanks for the comment

Your mic sounds great!

@JoediyLab

11 ай бұрын

Thanks

Very interesting information...thank-you.

@JoediyLab

10 ай бұрын

Thanks for the comment.

Great video joe

@JoediyLab

10 ай бұрын

Thanks 👍

Wow. excellent solution, it sounds great!

@JoediyLab

11 ай бұрын

Thanks

Did the Titan have ballast or trim tanks, and if so were they on the inside or out?

@JoediyLab

10 ай бұрын

They used drop weights

reminds me of the ORing which brought down the Challenger…path if least resistance will always be at a seam…two different materials glued together which expand and contract at different rates causes friction…and friction begets wear

@JoediyLab

10 ай бұрын

Thanks for the comment

The mic and pop filter are fantastic! A clear improvement from what you were using last week. Your plosives are way down.

Great analysis. I would like to add that very small imperfections in the glue layer can become bigger to the point we’re the water on the outside at almost 400 atmosphere want to go in. That would cause further failure of the glue at very high speed, almost like water jet cutting of metal, let alone glue. This would be happening in a very short time (10 - 100 ms) and trigger total failure of the glue connection.

@JoediyLab

10 ай бұрын

Thanks for the feedback

Interesting info! Just a tip for future videos, increase the zoom size of pages you're showing. You're only streaming at 720p so the small text is very hard to read.

@JoediyLab

10 ай бұрын

Thanks for the feedback yes I have zoomed the browser in future videos

It's actually interesting to learn that this was the first time a vessel like that failed by just imploding in all of history. And deep sea exploration is already actually almost about 100 years old.

@JoediyLab

10 ай бұрын

Thanks for your comment

@gailmcn

10 ай бұрын

@rosomak8244 Yes, and in all that deep sea submersible history, traditional engineering rules of safety/risk management were carefully observed....it wasn't until "break the rules and dump all the 50 year old white guys" ethos that there was a catastrophe.

Could the The viewing port transparent material have failed first and then pressure blew the end caps off the retaining rings as the carbon fiber material disintegrated.

@JoediyLab

11 ай бұрын

That is a very good question, The viewing port was rated for only 1300m so yes that is a possibility.

The Titan was glued together. Who could have guessed it wouldn’t hold?

The geometry of the glueing area, and the difference in expansion between the materials is equally important as the bond quality. Loads can skyrocket with such huge glue diameters.

@JoediyLab

10 ай бұрын

Yes thanks

The cylinder of oceangate was ten feet not 12 feet. There is a picture of the cylinder standing upright next to a twelve foot ladder. Just an observation, keep up the commentary it is very informative

@LetsGo__111

10 ай бұрын

Nope! It was 8.3ft long thank me later!

Wouldn't the compressive pressure on the bonding joint be way higher than the pressure due to depth, simply due to the reduction in surface area (compared to the dome's cross section)?

@JoediyLab

10 ай бұрын

Yes that is correct, I had an engineer give me some calculations.

@davedavis6020

10 ай бұрын

@@JoediyLab There was a confusing comparison of the hydrostatic sea pressure (in psi) to the strength of the adhesive (also in psi). The actual compressive stress on the glue joint is much greater than the sea pressure. It's elementary engineering, and the video's narrator apparently doesn't understand this.

@JoediyLab

10 ай бұрын

@@davedavis6020 You are correct, I was corrected by another viewer. Check the comments, there are calculations given

Your audio is great 👍

@JoediyLab

10 ай бұрын

Thanks

Mic and popshield sound good. Nice and warn. There's a touch of sibilance but nothing dangerous. Hope that helps.

@JoediyLab

10 ай бұрын

Thanks for the feedback

Good job very informative

@JoediyLab

10 ай бұрын

Thanks

Thanks for the informative piece

@JoediyLab

11 ай бұрын

Thanks

How was the outer white cover attached? Also, it looks like he had screwed lights and a monitor into the interior walls of the sub

@JoediyLab

10 ай бұрын

It appears that the external structure tied into the titanium flanges and the framework that made up the skids. The drop weights also were attached to the skids. It would have been foolish to drill holes into the carbon fiber, as some folks have indicated. Thanks for the great question..