Submarine boat missing - on the way to the sunken "Titanic"

[RocksInMyHead]

I agree the fibers will make the matrix weaker in compression.

But it's not just the compression they should worry about but also against deformation, even just tiny imperceptible amounts of deformation can lead to rapid progression of failure of a cylinder under high external pressure.

On the other hand, CF composites have very high stiffness to weight ratios and resists deformation very well. I suppose it's a bit of a compromise to sacrifice a little bit of compression strength to have good stiffness against deformation.

Compression forces lead to deformation. Carbon resists deformation very well, right up to the point where it doesn't - and because it is weaker in compression than in tension, that threshold is much lower under compression.

I also agree with more tests. Particularly with cyclic load testing. I did a search on it and found nothing. I can only assume OceanGate didn't do any cycle load tests on the pressure hull.

That would be a massive oversight on their part as it couldn't be that super expensive to do cycle load testing on a pressure vessel. You only need to enclose it in a slightly larger pressure vessel built of cheap but strong material like steel and opened and closed with many bolts. Nothing too elaborate nor prohibitively expensive. Put the test article inside. Sealed and bolted shut. Fill with sea water, increase pressure to max design depth using compressed air. Cycle pressure to simulate cycle loads of repeated dives. They can also cycle temperature in addition to simulate changes in temperature at different depths.

I don't know how many cycles it needs to survive to be deemed safe and still remain economical but the test itself is entirely economically feasible and OceanGate could have easily done the cycle load testing themselves.

Cyclic load testing would be a very expensive process for this, actually, because it would require a whole second submersible to be built - you can't do load testing properly unless it's done to failure. It would definitely be the safest option though (and probably should be done for a commercial vessel). The compromise is non-destructive ultrasound imaging. After construction, they can scan the structure with ultrasound to determine if there are any flaws. This would then be repeated at certain intervals to check for damage so that the vessel could be retired before it undergoes catastrophic failure. Rather than that, they were apparently relying on real-time acoustic monitoring of the hull to identify any issues in the hope that they would be able to surface before it failed.

They could also test in addition the resilience of the pressure hull against the implosion of equipment like the external cameras or any external equipment that can implode under high external pressure. They can see if the human occupied pressure vessel can survive the shockwaves from equipment implosions.

That would also be a good idea.

 

[timewerx]

Compression forces lead to deformation. Carbon resists deformation very well, right up to the point where it doesn't - and because it is weaker in compression than in tension, that threshold is much lower under compression.

Cyclic load testing would be a very expensive process for this, actually, because it would require a whole second submersible to be built - you can't do load testing properly unless it's done to failure. It would definitely be the safest option though (and probably should be done for a commercial vessel). The compromise is non-destructive ultrasound imaging. After construction, they can scan the structure with ultrasound to determine if there are any flaws. This would then be repeated at certain intervals to check for damage so that the vessel could be retired before it undergoes catastrophic failure. Rather than that, they were apparently relying on real-time acoustic monitoring of the hull to identify any issues in the hope that they would be able to surface before it failed.

That would also be a good idea.

Even steel and titanium is much weaker in compression than tension. Many structural materials are.

Although steel and titanium is much less brittle compared to carbon fiber which can make either metal more durable against rough handling. To make carbon fiber endure as much cycles, you just have to make it thicker and they did.

The problem with carbon fiber is any minor incidents that would not be a problem with metal parts like a tool allowed to fall and hit the carbon fiber parts may cause material to delaminate under the point of impact even if there are no external signs of damage.

You'll have to be extremely careful around carbon fiber which maybe a little challenging to do out in the open sea with the waves so most definitely, the carbon is a poor material of choice for the situation. But I get why they would choose carbon fiber. It's about as capable as titanium but cheaper.

The real time integrity sensor is a joke. Anything that might snap at great depths may lead to rapidly progressing failures. If the sensor detects anything, it might be too late. They might as well use an alarm sound for the sensor that says "you're doomed! see you in the next life!"

 

[timewerx]

Seems like there's an important junction where the titanium cone is fastened to the CF cylinder. I suppose the high pressure could make that joint somewhat self sealing with the right design.

The ends are glued watertight to the cylinder. I don't think the problem would have originated there unless it got damaged from handling or from manufacturing defect.

 

[RocksInMyHead]

Even steel and titanium is much weaker in compression than tension. Many structural materials are.

They actually aren't. Homogenous materials have a defined yield strength that is more or less the same under both tension and compression. The reason why CFRP has a lower compressive strength than tensile strength (by about 40-50%) comes down to its composite nature. Wood is similar (in behavior, not strength) since it also acts as a layered composite.

Although steel and titanium is much less brittle compared to carbon fiber which can make either metal more durable against rough handling. To make carbon fiber endure as much cycles, you just have to make it thicker and they did.

The problem with carbon fiber is any minor incidents that would not be a problem with metal parts like a tool allowed to fall and hit the carbon fiber parts may cause material to delaminate under the point of impact even if there are no external signs of damage. You'll have to be extremely careful around carbon fiber which maybe a little challenging to do out in the open sea with the waves so most definitely, the carbon is a poor material of choice for the situation.

Cyclic compressive/expansive forces due to pressure loading will be far more likely to create delamination issues given the thickness of the material. Delamination is caused by the flexing of layers against each other. If the force isn't sufficient to cause deflection within the composite structure, it can't result in delamination. Broadly speaking, I wouldn't really be concerned about rough handling unless an impact actually caused visible damage to the surface.

But I get why they would choose carbon fiber. It's about as capable as titanium but cheaper.

In some ways it's as capable, in others it's more capable, and in still others, it's less capable. They're very different materials with very different properties.

The real time integrity sensor is a joke. Anything that might snap at great depths may lead to rapidly progressing failures. If the sensor detects anything, it might be too late. They might as well use an alarm sound for the sensor that says "you're doomed! see you in the next life!"

Yup.

 

[DaisyDay]

The epitome of a human interest story. Like the miners, how long till it’s made into a movie.

Kinda a bummer of a movie.

 

[timewerx]

They actually aren't. Homogenous materials have a defined yield strength that is more or less the same under both tension and compression. The reason why CFRP has a lower compressive strength than tensile strength (by about 40-50%) comes down to its composite nature. Wood is similar (in behavior, not strength) since it also acts as a layered composite.

True if we're just talking of material yield strength.

But not true if we're talking of submersible pressure vessels. Because the way these pressure vessels are designed, they will fail or implode under increasing external pressures (compression) through "buckling" far below the material's yield strength even for metals.

With less compressive strength of the material, you simply give it thicker skin and OceanGate did do it in their Titan sub which had much thicker design thickness compared to other metal-PV submersibles.

Broadly speaking, I wouldn't really be concerned about rough handling unless an impact actually caused visible damage to the surface.

I used to work in airport ground operations and ground crews are instructed to be more careful around composite aircraft or around composite parts of aircraft.

My cycling friends who owned expensive carbon fiber bikes have attributed frame failure of their own bikes where even minor impacts have occurred with no visible damage. Many with such experiences have since switched to metal frames. It's common knowledge among enthusiasts if rough handling is unavoidable, one must avoid using structural components made of carbon fiber.

 

[RocksInMyHead]

True if we're just talking of material yield strength.

But not true if we're talking of submersible pressure vessels. Because the way these pressure vessels are designed, they will fail or implode under increasing external pressures (compression) through "buckling" far below the material's yield strength even for metals.

Sure, but we were just talking about yield strength (or at least I was). The yield strength of CFRP is about 40-50% lower under compression than it is under tension. This makes any pressure vessel built out of CFRP to withstand exterior pressure (rather than interior) inherently weaker.

With less compressive strength of the material, you simply give it thicker skin and OceanGate did do it in their Titan sub which had much thicker design thickness compared to other metal-PV submersibles.

This is generally true, but it's not quite so simple with a composite material like CFRP. As I said in my previous post, the fact that it's layered works against it under compression since it has a very limited capacity to contract elastically before it begins to delaminate. In contrast, a homogenous material like steel can expand and contract significantly without compromising structural integrity. So, in order to make it completely safe, the material needs to be thick enough that it cannot deform at all. Depending on size and design of the vessel, that may well push CFRP outside the bounds of viability as a material.

I used to work in airport ground operations and ground crews are instructed to be more careful around composite aircraft or around composite parts of aircraft.

My cycling friends who owned expensive carbon fiber bikes have attributed frame failure of their own bikes where even minor impacts have occurred with no visible damage. Many with such experiences have since switched to metal frames. It's common knowledge among enthusiasts if rough handling is unavoidable, one must avoid using structural components made of carbon fiber.

Typical carbon bicycles (I've also owned a few) have tubes only a couple mm thick at most. And while I'm sure aircraft use thicker material, it's likely not 5" thick. It's much easier to flex (and therefore potentially cause delamination) a thinner structure. Again, with a structure of that thickness, I would generally only be concerned about visible surface damage (cutting, gouging, anything that compromised the integrity of the fibers). It would take a very large amount of force to create a large enough amount of deflection within the internal structure to cause delamination. Obviously, if it got smashed into the side of a ship or dropped from a significant height, that could be an issue, but I wouldn't classify that under standard "rough handling". I'm just saying that it wouldn't need to be handled with kid gloves

 

[timewerx]

Typical carbon bicycles (I've also owned a few) have tubes only a couple mm thick at most. And while I'm sure aircraft use thicker material, it's likely not 5" thick. It's much easier to flex (and therefore potentially cause delamination) a thinner structure. Again, with a structure of that thickness, I would generally only be concerned about visible surface damage (cutting, gouging, anything that compromised the integrity of the fibers). It would take a very large amount of force to create a large enough amount of deflection within the internal structure to cause delamination. Obviously, if it got smashed into the side of a ship or dropped from a significant height, that could be an issue, but I wouldn't classify that under standard "rough handling". I'm just saying that it wouldn't need to be handled with kid gloves

The two journals below discuss the matter at hand.

Low energy impacts such as from rough handling or clumsiness or encountering debris from normal use can cause subsurface damage in composites even in the absence of visible surface damage.

2nd article shows laminate thickness only changes subsurface failure mode in low energy impacts. This means even thick laminates are susceptible to subsurface damage even from low energy impacts even in the absence of visible surface damage. Obviously a thick laminate will have smaller % of damaged fibers/layers against the total number of layers. Not a very big problem unless it's on a submersible's pressure vessel.

The damage is not the result of bending of the whole test article following an impact but from the highly concentrated force at the impact location causing localized flexing of layers right under the impact point even if the whole composite part did not flex at all.

Impact damage characterisation of carbon fibre/epoxy composites with multi-layer reinforcement

Low-velocity impact tests were performed to investigate the impact behaviour of carbon fibre/epoxy composite laminates reinforced by short fibres and …

www.sciencedirect.com

Thickness-dependent energy dissipation characteristics of laminated composites subjected to low velocity impact

This study investigates the low velocity impact properties such as damage thresholds, critical energy thresholds and damage process of laminated compo…

www.sciencedirect.com

Bonus research: thick composite laminates are complicated to make and prone to defects.

https://www.iccm-central.org/Proceedings/ICCM17proceedings/papers/IC4.1%20Gower.pdf

 

[RocksInMyHead]

2nd article shows laminate thickness only changes subsurface failure mode in low energy impacts. This means even thick laminates are susceptible to subsurface damage even from low energy impacts even in the absence of visible surface damage.

The second article cites multiple studies that support exactly what I said: a thicker laminate will be significantly more resistant to impacts and delamination is much less likely to occur due to impact as laminate thickness increases.

The damage is not the result of bending of the whole test article following an impact but from the highly concentrated force at the impact location causing localized flexing of layers right under the impact point even if the whole composite part did not flex at all.

Yes, and localized flexing of the layers requires significantly more energy if the material is very thick.

Bonus research: thick composite laminates are complicated to make and prone to defects.

https://www.iccm-central.org/Proceedings/ICCM17proceedings/papers/IC4.1%20Gower.pdf

Yep. I'm not trying to say that building the sub out of carbon fiber was a good idea. I'm just contesting your theory that damage from low-energy impacts could have been a contributing factor. The far more likely failure mode is delamination due to multiple pressure loading/unloading cycles.

 

[timewerx]

The second article cites multiple studies that support exactly what I said: a thicker laminate will be significantly more resistant to impacts and delamination is much less likely to occur due to impact as laminate thickness increases.

Yes, and localized flexing of the layers requires significantly more energy if the material is very thick.

I think you missed out some of the information as I might expect. The journal is a bit long and boring read, sorry about that.

Damage still occurred with low energy impacts on thick laminates. Just the nature of damage is different. Instead of delamination, it experienced fiber fractures. Although fiber fractures are not as big problem as delamination if only a handful of fibers fractured near the surface, and could be ignored in many cases. But when the CF is used as a PV in a deep submersible, the consequences can be far more serious.

Yep. I'm not trying to say that building the sub out of carbon fiber was a good idea. I'm just contesting your theory that damage from low-energy impacts could have been a contributing factor. The far more likely failure mode is delamination due to multiple pressure loading/unloading cycles.

I know, you made it very clear from the start that Carbon Fiber is a bad idea and we both agree on that.

Personally, I would not decide to use carbon fiber over titanium to save cost and weight. I'd go for 3D printed maraging stainless steel PV. 3D printing will permit the use of organically-designed hollow reinforcing structures consisting of a large number of very small and stubby truss-like members throughout the PV. Such reinforcing will allow the compression strength of the PV get much closer to the material's yield strength in compression resulting to significant savings in weight by reducing the total amount of metal used in the PV's construction.

 

[durangodawood]

This reminds me of a bike frame materials discussion.

Personally I'm a little wary of CF forks.

 

[timewerx]

This reminds me of a bike frame materials discussion.

Personally I'm a little wary of CF forks.

While CF parts are really strong, they just don't tolerate clumsiness and rough handling for long term use. They can get damaged from seemingly small impacts even if there are no visible damage. If you can be completely careful around a CF bike and extremely rarely gets into accidents, it should last you a very long time!

I did find it annoying that many steel bikes sold today comes with CF forks! Just when you think you can afford to be clumsy around the steel bike, but not near the fork!

 

[RocksInMyHead]

Damage still occurred with low energy impacts on thick laminates. Just the nature of damage is different. Instead of delamination, it experienced fiber fractures.

It stated:

Fiber fracture was the dominating failure mode for thick laminates while delaminations were more influential in the failure of thin laminates. They also stated that the threshold load for the major damage was independent of the incipient impact energy (the same conclusion as it was reached by Belingardi and Vadori [19]), but it was highly dependent on the laminate thickness.

In other words, delamination is less common in thicker materials compared to fiber fracture AND as thickness is increased, the threshold load for major damage (i.e. fiber fracture) is also increased. At least, that's how I read that part.

Unfortunately, I don't have access to the full Shyr and Pan paper, so I wasn't able to look into their results in detail, but it should be noted that the article you cited was looking at the properties of pretty thin composites (maximum of 8mm) relative to what was used in the Titan.

Also of note, fiber fracture is much more likely to be visible than delamination.

 

[RocksInMyHead]

This reminds me of a bike frame materials discussion.

Personally I'm a little wary of CF forks.

Meh. I've owned 5 carbon bikes over the years and 4 carbon forks. Never broken a fork (including the cheap ebay mountain bike fork), and the only frame breakage was due to poor packaging on the part of the guy I bought it from (it arrived broken). Had that bike repaired professionally and still ride it today.

While CF parts are really strong, they just don't tolerate clumsiness and rough handling for long term use. They can get damaged from seemingly small impacts even if there are no visible damage. If you can be completely careful around a CF bike and extremely rarely gets into accidents, it should last you a very long time!

I did find it annoying that many steel bikes sold today comes with CF forks! Just when you think you can afford to be clumsy around the steel bike, but not near the fork!

I don't treat my carbon bikes any differently than my steel and aluminum bikes. Could an impact that wouldn't bother a steel bike break a carbon bike? Sure. But I'm not going around dropping my steel bikes on rocks or hitting them with hammers any more than I am my carbon bikes. And honestly, I've seen just as many failures (not on my personal bikes, but among friends and acquaintances) of steel and aluminum frames as carbon.

 

[durangodawood]

While CF parts are really strong, they just don't tolerate clumsiness and rough handling for long term use. They can get damaged from seemingly small impacts even if there are no visible damage. If you can be completely careful around a CF bike and extremely rarely gets into accidents, it should last you a very long time!

I did find it annoying that many steel bikes sold today comes with CF forks! Just when you think you can afford to be clumsy around the steel bike, but not near the fork!

Thats exactly the problem. If its not essential to maximize performance (ie youre not paid to race), then why take the chance on invisible damage - especially if youre not made of money and expect to keep a bike around for a long time.

This is compounded by the issues of rack mounts, etc. But Im a weirdo who likes to bike for travel and exploration and just long fun rides.

 

[durangodawood]

Meh. I've owned 5 carbon bikes over the years and 4 carbon forks. Never broken a fork (including the cheap ebay mountain bike fork), and the only frame breakage was due to poor packaging on the part of the guy I bought it from (it arrived broken). Had that bike repaired professionally and still ride it today......

4 data points is not exactly confidence inspiring for such a high stakes issue.

(With CF in the discussion I just knew you 2 would be cyclists!)

 

[RocksInMyHead]

4 data points is not exactly confidence inspiring for such a high stakes issue.

Fair enough lol, but I'm hardly the only person to ever own a carbon fork.

 

[timewerx]

Meh. I've owned 5 carbon bikes over the years and 4 carbon forks. Never broken a fork (including the cheap ebay mountain bike fork), and the only frame breakage was due to poor packaging on the part of the guy I bought it from (it arrived broken). Had that bike repaired professionally and still ride it today.

I don't treat my carbon bikes any differently than my steel and aluminum bikes. Could an impact that wouldn't bother a steel bike break a carbon bike? Sure. But I'm not going around dropping my steel bikes on rocks or hitting them with hammers any more than I am my carbon bikes. And honestly, I've seen just as many failures (not on my personal bikes, but among friends and acquaintances) of steel and aluminum frames as carbon.

That makes you very careful around your bikes and safe to own CF bikes!

My aluminum bike has been felled over sharp objects a few times, NOT by me but by other family members. One crash by me over rocky terrain bending the rear axle and rear dropout, dropped heavy tools over it by me like a big adjustable wrench and countless minor collisions with cars, trucks, and motorcycles. Only one collision with a motorcycle resulted to damage via the wheel spokes. A few of these collision had me falling over the handlebar but unhurt and my bike undamaged.

My riding condition is very crowded, very chaotic, and simply quite unforgiving to bicycles why incidents are very common.

I kept riding the same bike for many thousands of miles now despite the damages. The damage hasn't progressed a bit and the bike kept riding as smoothly and quietly as when it was bought new.

One cyclist I talked to who promotes Giant bikes, owns and rides several sponsored carbon bikes, has a daughter who race for Giant locally, tells me I shouldn't own a carbon bike! He told me himself, he would never ride his carbon bikes in the city where the chance of collisions skyrockets to the heavens.

 

[timewerx]

Thats exactly the problem. If its not essential to maximize performance (ie youre not paid to race), then why take the chance on invisible damage - especially if youre not made of money and expect to keep a bike around for a long time.

This is compounded by the issues of rack mounts, etc. But Im a weirdo who likes to bike for travel and exploration and just long fun rides.

I've been riding a bike damaged by several collisions and one major crash for many thousands of miles now. It's an aluminum bike with aluminum wheels with steel fork. Not a single part is made of carbon fiber.

I would have replaced the whole bike for peace of mind if I only have free cash laying around but never did. The bike kept on running and shifting smoothly and quietly as the day I bought it new.

I swear by steel. Would be my material of choice for the most rugged operating conditions.