Comments by "" (@JrgenMonkerud-go5lg) on "Solar Eclipse Timer"
channel.
-
The biggest issue is that the analysts of logitudinal slices as analogous to a bridge is not right unless you have transverse circular symmetry. And with defects you certainly do not have that, this means the hull deforms asymetrically, not only if it buckles but as it bends linearly, which makes the 2d slice logitudinaly, bend like a squeezed circle in some orientation, which puts strain on the outside and compression on the inside, but in the orthogonal direction to the logitudinal analysis, and the kicker is that this loading is not symmetric, one pole puts comoressive stresses on the outside and the other on the inside, like compression of a circle with thickness. Needless to say, the logitudinal plies provide no strength other than to resist in part the deformation caused by the local curvature of the deformation as a function of logitudinal pisitions of transvers deformation, and so the resistance to compressive stresses in certain areas is almost non existent, particularly on some areas up next to the bulkheads, because of another phenomenon of the distribution of pressure in the joint, when the logitudinal slice bending like a bridge, turns the interface outward, so the load is concentrated on the inside of the joint with the titanium. Then when you add the assymetrically deformation in bith the logitudinal slices of different orientations and the transverse slices of different logitudinal positions, the conoressive stress on the interface gets concentrated to just a very small area, 2 points if cpmoressive forces cause no deformation, which is not true, so only 2 very small areas around the joint with the titanium on each side take the entire pressure on the two titanium bulkheads. The titanium bends a little and the carbon compresses elastically a little, butnthe stress distributions on the end of the cyclinder on each side is highly concentrated, and this results in those areas betting broken up first under cycling. This is counterintuitive because it is a result of the specific assymetry in deformation, even if the deformation is very small, very high stress concentration happens on the interface. This is likely wgere the infamous sounds of cracking came from, the softening of the overloaded ends of the carbon hull in two small areas in the inside of the hull. What would happen, is exactly consistent with what rush described for all his testing dives. At first there was a lot ofncracking sound, because this area was softening by cracking, resulting in inelastic deformation there, after a few cycles the area stresses are concentrated into gets a lot larger, and so the cracking noises calm down. Then you go cycle after cycle of minor cracking noises, until the deformation has grown enough that there is a discontinous failure. This happened before the last dive. Their system that measured deformation detected a non linear deformation a dive before they died. That they went for another dive is oure madness. They actually got lucky enough that a serious and detectable faliure, that lead to a rapid change in the deformation innterms of hull displacement showed up onnthe plot between depth and deformation, telling him something cracked and now the faliure point is basically at the dive depth. But they didn't analyse it, or care, either if the two maybe, amaturish nonsense that got a bunch of people killed. They were suoer lucky that the sub didn't fail once the first major faliure took place. You can see on the graphs of the deformation ploted against dive depth, and the plots of carbon fiber faliure, that they were dangerously close to the faliure point because of the rapid change in deformation due to depth. They should instantly have cincluded the hull was finished and they had to retire the hull. But when a large bang happens and a previously almost straight deformation plot wirh depth suddenly turns at the dive depth, you then should know you have been extremely lucky nit to implode. But to go down again one cycle later with the same hull is absolute madness.
5
-
@solareclipsetimer absolutely. What i tried to write as well, is just a more complicated version of the same kind of analysis. Depending on how much asymetric deformation there was, what i described could be less severe ofc. I think the testemoni was good. I'm pretty sure the damage started at the interface between the hull and the bulkheads, but u know, if that is true, it is still going to be the case that damage where the stresses are highest, is also going to cause more displacement everywhere else pretty much, thats just how it goes. The data that exists where you can see the change in deformation per depth dive by dive, with a discontinuous change is crazy tho. I didn't mean ti sound critical, the bridge analog is always super important anyway. The video was super nice. I'm not an engineer but a physicists btw. Would be super silly to take on such responsibility without experience. I think rush and the other guys were in way over their heads.
Take for instance the shape of the plastuc in the window. It was shaped with a flat back, which actually makes it weaker than if there is a cutout spherical surface in the same orientation as the exterior, the reason is pretty simple, and applies to all calculations of stresses, if solid bodies can tug on themselves when deformed, and the flat interior surface, instead of an arch matching the exterior, does that, enging up with a stress concentration on the interior circumference of the interface of the window and titanium. It is so silly that they did this, and with a window rated for 1000 meters. Because they could have made a stringer window with the sane optical effect for idk 50 bucks? You just have the structurally sound shape, plus an acrylic flat plate, or whatever shape you desire, could be convex as well if you want more of a zoom in the window. It is really as simple as an oil with the same refractive index and a cover in any shape you want, and you dont have to compromise at all, and some expansion balloon such that the deformation of the actual windom come and go. But it's basically super cheap and easy to make.
3
-
3
-
2
-
2
-
2
-
2
-
2
-
1
-
1
-
1
-
1