Why did the Titanic tour submarine implode?

Discussion in 'Boat Design' started by sun, Jun 22, 2023.

  1. DogCavalry
    Joined: Sep 2019
    Posts: 3,209
    Likes: 1,619, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    This forum is a unending source of education.

    I'm curious how something as stiff as CF could be maintained in tension in a matrix of basically syrup.

    And what is the yield strength of epoxy that it can sustain external pressure, and the stress from the prestressed CF...

    Ah, I see. As high as 1690MPa. Well then...
     
  2. jehardiman
    Joined: Aug 2004
    Posts: 3,838
    Likes: 1,211, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    Yeah, epoxy in compression is 15,000-20,000 psi, epoxy in tension is 5,000-6,000 psi. Carbon fiber in tension is 600,000 psi...and effectively zero (0) in compression. You need to design so that the matrix is always in compression.
     
  3. fallguy
    Joined: Dec 2016
    Posts: 7,796
    Likes: 1,735, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    I'm still struggling with intuition on the failure. It seems a no brainer to me this vessel would fail at the middle; like a pipe burst along its length; not circumference. I'd love to see if they did a stress analysis that shows more stress midship.

    Someone who went on a shallower dive is saying the sub was crackling on the way down and the CEO told them ot was normal. I expect it was making micro fissures at all those lamination and glue flaws/variations.

    I'd be willing to speculate that ship would not have made it down to Titanic 100 cycles. How anyone can underappreciate 5000 psi is even more perplexing.

    Am I wrong on the crackling?
     
  4. fallguy
    Joined: Dec 2016
    Posts: 7,796
    Likes: 1,735, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    ?
     
  5. DogCavalry
    Joined: Sep 2019
    Posts: 3,209
    Likes: 1,619, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    I made a scale error in visualization. In simple structures CF is used in compression, but as @jehardiman corrected me, in structures like the titan, it does not have useful strength in compression. Individual fibers have no lateral stiffness to prevent deflection out of the load path, and so can not support any compression load beyond what the matrix can manage, in which scenario, there's no reason to include the fiber. It just displaces matrix material, which is what is doing the work.
     
  6. baeckmo
    Joined: Jun 2009
    Posts: 1,698
    Likes: 706, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    In addition to the questionable production quality, as noticed previously, I am concerned about the possible issues with epoxy aging, and the resulting variations in physical properties, including specific volume changes when it is repeadly cycled through the glass transfer temp range.

    A couple of years ago, we looked into the possibility to use epoxy for a turbine housing prototype. But after some reading we realized that we could not be certain to validate any test results regarding strength and deformation; the variations due to aging would hide what we were looking for. The following summary made it clear that we, as "metal guys", did not have the engineering competence needed to litterally dive into the world of polymers:


    Physical aging of epoxy polymers and their composites

    G. M. Odegard, A. Bandyopadhyay

    First published: 28 October 2011

    https://doi.org/10.1002/polb.22384


    SUMMARY

    This review has discussed the background and fundamental concepts associated with the physical aging of epoxy resins and epoxy composites. From the numerous studies that have been reviewed herein, it is clear that physical aging involves the simultaneous reduction of free volume and conformational changes of the crosslinked molecular structure when exposed to sub-Tg temperatures for extended periods of time. The changes in the molecular structure result in bulk-level responses of epoxies. Specifically; mechanical, thermodynamical, and physical properties of epoxies are influenced in a manner that significantly alters the overall response of these materials.
     
    philSweet likes this.
  7. jehardiman
    Joined: Aug 2004
    Posts: 3,838
    Likes: 1,211, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    Last edited: Jun 26, 2023
    fallguy likes this.
  8. fallguy
    Joined: Dec 2016
    Posts: 7,796
    Likes: 1,735, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member

    Well, that paper is indeed interesting. A bit hard to comprehend and roll into the Titan at its size and 40 MPa pressures, unknown winding angle, etc.

    Noone has ventured to answer my query. Would the expectation of crackling noises be normal?

    I don't think so.

    And I'm wondering one other thing. Were the actions of the operator criminal?

    When an amateur like me sees a dirty winding and hand smeared resin in an unclean room and wonders why anyone did not expect a failure; it is a fair progression to ask..
     
  9. seasquirt
    Joined: Dec 2015
    Posts: 144
    Likes: 61, Points: 28, Legacy Rep: 10
    Location: South Australia

    seasquirt Senior Member

    I'm wondering if this compression of composites problem could be put to use making a better pressure vessel. eg construct a composite pressure vessel with slow curing, cold curing resins, and whatever reinforcement methods, and don't seal it, but lower it to its working depth to 'normalise' the structure as it finishes curing. Like vacuum bagging but equal pressure on all surfaces, and high pressures compressing micro bubbles. Once cured (could take months), then it can be surfaced, and will be optimally 'pre stressed'. Then get it fitted out for its use. 200m deep gives ~ 300 PSI, - so build a composite boat hull soft, seal it, sink it to 200m and let even pressure squeeze and stiffen it with 300 PSI as it cures, and when time is up you have a compacted pre stressed hull. You may need a frame both sides in key locations to hold the desired shapes. Could be the answer for composite turbine blades, pressure tubing, and myriad other high stressed composite components.
    I agree, those lumpy windings photos look disturbing, and I wonder if the Captain had seen them. The first vessel should have been cycled to destruction with cameras and sensors all over to observe.
     
  10. Milehog
    Joined: Aug 2006
    Posts: 601
    Likes: 133, Points: 43, Legacy Rep: 215
    Location: NW

    Milehog Clever Quip

    These stood out and confirm previous posts.

    "Several researchers (Schilloet al. 2015, Castro et al. 2015, Blachut 2016) have shown experimentally and numerically that manufacturing imperfections influence the buckling loads of composite cylinders. The buckling strength of composite cones and domes, which are important structural components of the pressure hull, has been investigated in association with initial geometric imperfections by Castro et al. (2015) and Blachut (2016)."

    "It has been thought that these local deflections were attributed to the initial material and geometrical imperfections (Moon et al. 2010)."
     
  11. jehardiman
    Joined: Aug 2004
    Posts: 3,838
    Likes: 1,211, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    Actually all composites make "noises" under loading. The micro cracking of the fibers debonding from the matrix causes a very high (inaudible) frequency impulse in the material under load that can be monitored. Generally these will occur the first time the structure is loaded and then drop significantly as long as that peak load is not exceeded. However during repeated flexural cycles; the micro cracks will continue to grow, and as they grow, the frequency will get lower. A decade or more ago the US military began using acoustic emission and sense wire methods to "qualify" (as opposed to quantify) the "degradation" of composites under cyclic load, specifically for aircraft composite wings. Note that this is not a hard GO/NO GO, measurement or indication; but rather an indication of significant change in the structure.

    Acoustic emission - Wikipedia https://en.wikipedia.org/wiki/Acoustic_emission

    Structural health monitoring - Wikipedia https://en.wikipedia.org/wiki/Structural_health_monitoring
     
  12. jehardiman
    Joined: Aug 2004
    Posts: 3,838
    Likes: 1,211, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    I know that many of the early composites (free flooding skins generally) used in deep submergence used heat curing and an autoclave, just to prevent the formation of large voids. This was to prevent loss of buoyancy bulk compression at depth. Remember an autoclave is both heat and pressure, so in fabrication you could densify the material. Today, it is more common to use cold cure and vacuum infusion.
     
    rxcomposite likes this.
  13. Eric Lundy
    Joined: May 2017
    Posts: 37
    Likes: 7, Points: 8
    Location: Vermont

    Eric Lundy Junior Member

    I watched a video up on the cnn website: "What it was like inside the the lost Titanic-touring submersible".

    You can see grab rails and the monitors are mounted directly into carbon. Is this an inner sacrificial skin I hope?
    The original video is longer and has a better shot of the monitor mounts. CNN cuts it off. I cannot find the original.
     
  14. fallguy
    Joined: Dec 2016
    Posts: 7,796
    Likes: 1,735, Points: 123, Legacy Rep: 10
    Location: usa

    fallguy Senior Member


    I'm sorry, but I can't help but reflect on crack propogation...

    The entire concept seems in error to me.

    Do you have any comment on criminality, or am I mistaken, or do you want to stay away from the issue?
     

  15. jehardiman
    Joined: Aug 2004
    Posts: 3,838
    Likes: 1,211, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    I'm not a cleared official witness, not my monkey, not my circus.
     
    DogCavalry and fallguy like this.
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.