Comments by "Glen McGillivray" (@glenmcgillivray4707) on "Solar Eclipse Timer" channel.

So, for the lens to work it has to be practically touching the sphere. To make it that close, it becomes load bearing, and the sphere will clamp down on it. Personally I'd do something along this idea, but fill the layer in between with a suitable refractive liquid, which can be vented through an internal port to a syringe/piston setup, to let the dome compress and crush inwards and seal properly, while the volume taken up by the crush can be 'vented' What sort of liquid would work, why wouldn't it wet the sealing surface, what sort of maintenance would you need to do to it? I don't know, and experiments would need to be done. it's just the idea that a liquid can take the incoming pressure and volume changes under 4 kilometers of water pressure and move itself out of the way without imploding, or affecting the loading characteristics particularly, and be contained behind a nice flat view port to give you the optical effects we desire.
4
@wackyvorlon can't react chemically, nor 'wet' the sealing surface by accident. It's a close enough match to be useful, but I'd still say experiments are needed. I'm tempted to say maybe just use water and add a suitable salt/dissolved material to prevent both corrosion and tweak the refractive index.
3
When the deflection of the dome was quoted at three quarters of an inch, that's a lot of fluid that needs to move out of the way without loading the internal flat inner surface. Otherwise you are talking about applying four kilometers of water pressure on a dome, which will load the internal flat surface with an equal force thanks to hydraulic loading. you'd need a pressure tap, which would allow the compression to occur while maintaining a suitable pressure in the core void as the outer arch crushes under the stresses.
3
@wackyvorlon I like the enthusiasm. But you'd be better to check what that'd look like as vinegar. Because of course I cannot prevent bacterial or yeast contamination.
2
The fix would be so simple. Mount your damn lifting eyes on the outside of those landing gear. Ideally on the tops, concealed by the cover. But no, they wanted them on top for the sake of looks. Would have been so simple to bolt on some triangle hanging loops but no, weaken the most vulnerable component, that's clearly the wisest choice.
2
So, for the lens to work it has to be practically touching the sphere. To make it that close, it becomes load bearing, and the sphere will clamp down on it. Personally I'd do something along this idea, but fill the layer in between with a suitable refractive liquid, which can be vented through an internal port to a syringe/piston setup, to let the dome compress and crush inwards and seal properly, while the volume taken up by the crush can be 'vented' What sort of liquid would work, why wouldn't it wet the sealing surface, what sort of maintenance would you need to do to it? I don't know, and experiments would need to be done. it's just the idea that a liquid can take the incoming pressure and volume changes under 4 kilometers of water pressure and move itself out of the way without imploding, or affecting the loading characteristics particularly, and be contained behind a nice flat view port to give you the optical effects we desire. (Copied from a similar post)
2
The fact is, the forward mounting ring was sheered flat, it used to have a C profile. My suspicion is the carbon fiber initially crushed inwards, and momentum drove the viewport out of the dome entirely. Importantly when we look at the sensors on the forward barrel, they reveal non-linear deformation on the final few dives. This suggests water intrusions, and or carbon fiber debonding and a general compromised structure at that point. The Dome would probably have failed in time. But it didn't get enough time to get that opportunity.
2
So. We know the outer dome crushes under pressure. Because four kilometers of water is a lot of mass when every meter weighs one ton. The lens, cannot touch the dome or it will effect the load bearing inner surface in an unexpected manner, at any point. At first I considered a lubricated inner surface to minimize stress transfer, but the inner lens will still be crushed by the dome slowly being crushed, which will mess with the loading regardless. So you need to compensate. I can only come up with a larger optically clear void with fluid, intentionally connected to a hydraulic tap off to maintain a suitable pressure inside based on the internal flat windows pressure resistance. A Lens would be crushed like the dome, a Fluid will need to be displaced, and the sealing surface needs to be kept in perfect condition to ensure no leaks under the loading. This still needs a lot of testing, but wouldn't be critically flawed like this solid arc.
2
A fluid would need to be displaced when the dome contracted, and expand again when surfacing. A hydraulic ram with a gas drive would have been enough, or even a spring loaded ram. But you do need to tap off 380 atmospheres of pressure or it'll load up the inner plate and you'll watch an implosion on the inside while the outer dome is (initially) intact.
1
@smedleyx a fluid lens contained behind a flat panel would have worked easily enough. But it would have costed, and he might have envisioned it as a needless expense and complication. The fact that hemispheres aren't as strong as arches under extreme compression isn't immediately obvious, so the 'simple' fix seemed viable. It wasn't. neither was their 1 inch composite composite, there were mechanisms to preload the structure and use void space to balance the loading, but they chose to break the structural integrity of their tube instead. because the solution looked so simple, but every engineer worth his salt hated these compromises. If you build 5 concentric 1 inch tubes rated to 800 m depth, by using valves to keep the intense pressure out, you can preload each ring while diving. You can even monitor pressure between layers to identify when things are failing. But no. Just try monolithic composites in a manner never attempted before and say it'll work despite multiple experimental tests.
1
So, for the lens to work it has to be practically touching the sphere. To make it that close, it becomes load bearing, and the sphere will clamp down on it. Personally I'd do something along this idea, but fill the layer in between with a suitable refractive liquid, which can be vented through an internal port to a syringe/piston setup, to let the dome compress and crush inwards and seal properly, while the volume taken up by the crush can be 'vented' What sort of liquid would work, why wouldn't it wet the sealing surface, what sort of maintenance would you need to do to it? I don't know, and experiments would need to be done. it's just the idea that a liquid can take the incoming pressure and volume changes under 4 kilometers of water pressure and move itself out of the way without imploding, or affecting the loading characteristics particularly, and be contained behind a nice flat view port to give you the optical effects we desire. (Copied from a similar post)
1
@solareclipsetimer thanks for the reply. But my answer was directed at the original poster, and gave a simple explanation for the process. I believe with some curiousity that 5 layers, without glue, as five pressurized components without sanding down the outer layers, pressurized to each 1/5 the total depth, perhaps by isolating the layers at appropriate depths might have shared the loading better, and resisted the load better, each acting as unique pressure vessels, each with seperate imperfections. Unfortunately they tried to build a monolithic structure and built so many imperfections in with their process, it was doomed from the start unfortunately. Regardless it's beyond our understanding why the man in charge didn't comprehend just how dangerous he was acting in his pursuit of offering a unique tourism experience. And we will never understand why exactly he allowed such negligence to occur. You'd think after noticing a variation in the gauges as recorded, you'd take the sub to be inspected, then dissect it to understand the fault, with the presumption you may need to build a Core Mk 3 as a consequence.
1