leveraging large light format 3d printing boat parts

Discussion in 'Boat Design' started by 3DPY, Jun 4, 2024.

Tags:
  1. 3DPY
    Joined: Mar 2022
    Posts: 4
    Likes: 3, Points: 13
    Location: Palermo

    3DPY fbelvisi

    Hello everyone.
    I would like to share my expertise regarding how to make prototype boats that are 3D printed and have composite materials integrated into them.
    As a yacht designer, I began by developing a light system for 3D printing functional parts, which significantly reduces the time and cost involved. To provide an overview of our technology , we published a recent research in the Journal of Marine Science.


    I think the primary boundaries of today are:
    x the understanding of novel 3DP-integrated processes
    x the design techniques that made the part light and stiff.

    I've attached a copy of the article and a photo of a seat we created as a Wallyacht study there.

    wating for your opinions
     

    Attached Files:

    CocoonCruisers likes this.
  2. laukejas
    Joined: Feb 2012
    Posts: 774
    Likes: 19, Points: 18, Legacy Rep: 128
    Location: Lithuania

    laukejas Senior Member

    Hi, Francesco, and thank you for sharing this study. I read through it with great interest, as I'm exploring this technology (printed core with composite reinforcement) for my upcoming boat design as well. The paper was put together really well, and it's great that you made comparisons of test results with FEM studies, especially considering that FEM definition of 3D printed parts is quite difficult. I have several remarks / questions, and I hope we can discuss them.

    1) From my own lab testing, as well as from the results of other similar studies, it appears that due to vastly different mechanical properties of typical 3D printing materials and fiberglass/carbon composites, in most loading scenarios the composite skin takes up the majority of the load, never giving the printed core a chance to be loaded, and should fail before the core does. My lab testing confirmed this, but then again, I was using carbon fiber and epoxy, rather than fiberglass and polyester. Your study mentions that the failure mode was shear of the core with balsa, and as I understand it was similar with the printed samples. Am I correct to assume that the samples with epoxy failed in the core first as well?
    2) Continuing on the previous point, section 3.2.4 mentions that the resultant weight of the product was 11kg, of which printed core was 8kg - a ratio of 3:8. In my testing, because of the vast difference in strength and stiffness between the core and the skin materials, I aimed to reduce the core weight as much as possible, to maximize the usage of the stiffer and stronger material, and typically had a weight ratio of 1:1, which always resulted in skin tensile or compressive failure, never a core failure. For reference, I was using regular PLA with 0.4mm nozzle, single wall, gyroid 7.5% infill. Reference image of one of the test parts: imgur.com https://i.imgur.com/HrR5Zyf.png. I am wondering if you had used less printed core material and more composite skin material to arrive at the same weight, perhaps the failure mode would have been skin failure as well, and significantly increase ultimate strength.
    3) Some of your test samples, such as Nugae structural composite panel in section 3.1.3, uses ribbed structure, rather than being a flat panel with composite reinforcement on both sides and internal printed infill. This is one of the major considerations for me right now as well: whether it is more optimal in terms of strength/weight ratio to design thin panels with structural ribs, or to have a uniformly thick panel with printed infill. The second case is clearly far easier to manufacture in terms of composite application, since it is much easier to work with relatively flat panels, there are no access issues. Of course, thicker sections can be designed in critical locations as well, but overall the panel can remain relatively uniform in thickness, such as in this sample: imgur.com https://i.imgur.com/eRrYrBW.png. Of course, this is more complicated and time-consuming to print, as it does not allow for constant speed and no retractions. Still, even with this method, it remains questionable whether it is more optimal for a uniformly loaded panel to be uniformly thick, or ribbed. I hope someone can comment on this very important consideration.

    I should also mention that I am an amateur with far simpler manufacturing and testing capabilities, so whatever results I obtained, were obtained in a far less scientific methodology and far less controlled environment than yours. Still, I wonder what you think on these points I mentioned, and if perhaps optimizing the design to allow for a lighter core and more composite material would result in superior strength. Perhaps that would also allow for the usage of less expensive 3D print materials, since the core wouldn't be in risk of failure anymore.

    I based a lot of my assumptions, design and testing on the information I gathered from this post by Jan Herich, as well as this study.

    Let me know what you think.
     
  3. 3DPY
    Joined: Mar 2022
    Posts: 4
    Likes: 3, Points: 13
    Location: Palermo

    3DPY fbelvisi

    Hi, your observation is correct. Until we can 3D print using continuous fiber, long fiber composite always has a magnitude of superior performance. Therefore, we need to make 3D printing work as a core or support for complex geometries/tools. Our starting philosophy is to use as little material as possible to create self-supporting geometries based on the isogrid hollow philosophy and reinforce externally or internally within the cavities with composites.

    In failure mechanisms, fracture often occurs due to crushing at support points and detachment of the skin bonding. The adhesion between these is not always consistent in thermoplastic/thermosetting matrices and is a critical factor for which we have developed some solutions. In general, it is a completely different approach but can allow for the creation of one-off pieces in a short time and at competitive costs.

    I saw your applications, very interesting, congratulations! We started with the flat sandwich core approach, creating a core for a Mini 650( Redirect Notice https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.ansa.it%2Fvela%2Fnotizie%2F2018%2F10%2F31%2Fintanto-io-il-mini-650-me-lo-stampo-in-3d_b25f2822-d44d-42ec-800b-351affbd4a2c.html&psig=AOvVaw3gY-nn72gujBTzv89Sp7Nx&ust=1717667931674000&source=images&cd=vfe&opi=89978449&ved=0CBIQjRxqFwoTCJjk_uiZxIYDFQAAAAAdAAAAABAE) . However, given the long times (about 3 times longer), the increase in material, and the poor predictability of fracture mechanisms, we preferred to switch to the hollow grid approach, which is proving more effective in our expereince.
     
  4. Rumars
    Joined: Mar 2013
    Posts: 1,840
    Likes: 1,145, Points: 113, Legacy Rep: 39
    Location: Germany

    Rumars Senior Member

    Thanks for the article, it is a well written piece. Nevertheless it makes me ask questions.
    At 27.5kg/sqm we are into metal boatbuilding territory, even steel. What's the use case for a 47mm balsa cored skin with 2mm glass skins that you are trying to replace? Realistically speaking why would a designer go there?
    Why is this panel B the only sample to compare against?

    There are other questions wich go beyond the scope of the article but wich are very much relevant to the success of 3D printing.
    The most important one is: how does it compare financially against thermoforming the foam core on an adaptive mold? This is going to be the chief competitor as long as 3D printing is confined to core use.

    When it comes to non core uses, is the surface quality good enough to create interiors (for example furniture) without needing further machining work (sanding/filling)? If it is, can it be done cheap enough to compete with other methods? In other words is printer time cheap enough to be able to create a surface the customer can accept as is, or that can be veneered using a commercial iron on veneer.
     
  5. 3DPY
    Joined: Mar 2022
    Posts: 4
    Likes: 3, Points: 13
    Location: Palermo

    3DPY fbelvisi

    Hi Rumars,
    Thank you for your observation,
    The comparison was based on the customization of medium/large composite yachts of 60/70 feet that require customizations, and the competitiveness has been validated up to a number of about 15 identical pieces, among which it is more effective to create industrial molds. For that size, the weights are very common for sandwich structures, also in compliance with regulations that force you to a minimum outer skin thikness. Certainly, even if it can be compared with 3 mm of steel, it must be considered that a 3 mm steel panel to be stiff enough requires many more structures, leading to a much heavier structure. Such thin thicknesses have very limited application areas in boat substructures with rather dense reinforcements. The other structures treated are literally very expensive to conceive in a traditional manner and have proven to be very competitive in manufacturing. In general, today's additive manufacturing allows for the creation of large structures with much greater thicknesses of 10/20 mm, while we work with thicknesses of 1.5/3 mm, which allows us to keep the weights down compared to other applications.
     
    CocoonCruisers likes this.

  6. laukejas
    Joined: Feb 2012
    Posts: 774
    Likes: 19, Points: 18, Legacy Rep: 128
    Location: Lithuania

    laukejas Senior Member

    Francesco makes a valid point. Printed structure/core has several major advantages:

    1) No need for molds, which is great for one-off or low-volume production designs;
    2) Very easy to vary structure/sandwich core thickness, even make gradual transitions between different thicknesses, depending on expected loads, distributing material in a very optimized way;
    3) In complex geometries, far cheaper to manufacture compared to steel;
    4) In case of sandwich core approach - unlike foam, printed core can be made with open cell infill (truss or gyroid), which means that in case of water intrusion, water can flow freely through the core to a collection point for extraction, making it very easy to dry it out, whereas foam core could soak it in and make it very difficult to completely dry out.
     
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.