"3D printing" A boat? Layer by layer?

[John Smithson]

So, had another unconventional thought... what is the practicality of using a CNC machine to print a boat layer by layer from sheets of plywood?

Taking the CAD design and cut it into 3/4th inch layers. Then cut those layers into pieces that would fit into 4x8 sheets of plywood.

Then just pack them efficiently into sheets and let the CNC machine produce what you need to assemble together.

If you offset the cuts from one layer to another, am I wrong in thinking it would be rather robust?

The other advantage is this could be used to assemble the interior at the same time.

The one issue is that proper clamping would be needed, so likely assembling the boat in larger chunks, and then putting those together. Then sanding and glassing everything in.

Seems feasible to me?

 

[bajansailor]

It sounds to me like it would be a terrible waste of plywood - you would need to have so many sheets.
And if I understand your proposal correctly, you would then end up with something like a plywood deck of cards that is boat shaped - and you would then have to glue all of the cards to each other with epoxy. And spend a fortune in epoxy by doing so.........

 

[gonzo]

CNC does not print but remove material. Additive construction methods are completely different. Also, the fibers on the plywood would be running the wrong way. Read some books on boat building and design. There are very good reasons why methods evolve to what they are now.

 

[John Smithson]

And if I understand your proposal correctly, you would then end up with something like a plywood deck of cards that is boat shaped - and you would then have to glue all of the cards to each other with epoxy. And spend a fortune in epoxy by doing so.........

That's the general idea.

Though instead of epoxy, just use resorcinol (the glue in marine plywood) seems like it would work. Seems like it would be cheaper (just another layer of plywood for all the glue cares).

Epoxy for the fiberglass, but no more than any other method would use I would think?

It sounds to me like it would be a terrible waste of plywood - you would need to have so many sheets.

Ideally you would pack the pieces together to make as little waste as possible. Each layer doesn't need to be a single solid piece. As long as the pieces of each layer don't line up with each other as new layers are added, it seems like it wouldn't be an issue. Especially since your strength is coming from the fiberglass anyway. So essentially it's just a plywood core GRP construction.... just getting that core shape a little differently.

Also, the fibers on the plywood would be running the wrong way

For GRP is that a problem? Cores are often balsa....or foam... a less than ideal direction on plywood doesn't seem like it would cause issue?

Read some books on boat building and design. There are very good reasons why methods evolve to what they are now.

Was just reading about the world of 3d printed boats. It's very interesting to me.

Though I'm not rich enough to have a full thing 3d printed with continuous glass or anything. But getting a hull shape faster to glass over? Seems doable... unless that plywood directionality is really a deal breaker? (Still not sure, do you have any links to reading on the issue in GRP construction?)

 

[wet feet]

Theoretically possible,but slow and wasteful.First you have to slice your design into 3/4 inch layers and arrange each layer so that the pieces don't exceed the length of a sheet.Staggering the joints so that the joints don't all fall in a vertical line is another thing to keep in mind.The vast majority of CNC machines are 3 axis and not that many operators will be eager to cut the sloping surface on one side.It will only be one side because the other side would require an undercut and would likely be left square.The technical challenge will be to prevent the individual segments from being flung around the workshop as each piece will hardly have sufficient area for the normal vacuum hold down to secure it.Yes,a decent CAM program will allow the addition of tabs to prevent breakaway,but its one extra step and the nesting process might optimise sheet usage but there will still be a huge proportion of wastage. The really tedious part is converting the stepped surfaces to a smooth surface both inside and out.while ensuring that the overall shape is maintained.

I won't even mention the miserable process of fairing the glass overlay to an acceptable standard.Or the likelihood of ever recouping the money that went into building something so far outside the accepted norms.If you would like a good practical course in the process,just build a small rowing boat to see how it goes and how much each square foot of hull surface costs in both time and materials.

 

[kapnD]

Your end product will be a fiberglass boat with a heavy, weak plywood core.
To achieve this, you’d be just as well to conventionally layup the outer skin in a female mold and glue blocks or strips of plywood on it.
Perhaps you could find a source for plywood scraps for cheap?
I’ve cored boats with foam scraps from nearby surfboard manufacturers. It worked, and was high quality foam for free, but was labor intensive, sheet materials make so much more sense!

 

[Skyak]

Gonzo is right, and fiber orientation is critical for a wooden boat.
Creating "plywood" in the shape of a boat has been done. It is called "cold molding". It is a terrible misleading name because it takes high temperature and pressure to make strong glue bonds, but the result is a wonderful hull skin. Today some people do it with epoxy to make custom boats but if the world is serious about making everything "sustainable" cold molding production boats should be resurrected.
Boats have been 3d printed in plastic but it isn't competitive for large parts. It will take specialized design and time to prove performance.

 

[Kayakmarathon]

I can see a huge advantage in using a CNC router to cut cross sections for woodstrip construction. It took me about an hour per cross section to loft those sections by hand. I've been researching slide rails and stepper motors to build an XYZ gantry with a laser. The position accuracy is +/-0.2 mm. That's far better than pencil and jigsaw. For less than $3000 I should be able to build a DIY system with control software and firmware that allows me to cut a midship up to 90 cm x 90 cm.

 

[wet feet]

I can see a huge advantage in using a CNC router to cut cross sections for woodstrip construction. It took me about an hour per cross section to loft those sections by hand. I've been researching slide rails and stepper motors to build an XYZ gantry with a laser. The position accuracy is +/-0.2 mm. That's far better than pencil and jigsaw. For less than $3000 I should be able to build a DIY system with control software and firmware that allows me to cut a midship up to 90 cm x 90 cm.

That is slightly different and a very practical solution if you plan to do the job more than once.There are limitations, if you ever need a good cosmetic surface on the cut edges as a laser cut edge on wood based materials will have a scorched edge.For a less expensive way to get a one off set of building moulds you can seek out a local place with a CNC router and get the cutting done there.One tip is to make sure the elements that make up the outline are joined so that the programmer doesn't have to do that for you while the meter is running.A commercial grade machine will do the job in a fraction of the time a hobby machine will take.

 

[Skyak]

I can see a huge advantage in using a CNC router to cut cross sections for woodstrip construction. It took me about an hour per cross section to loft those sections by hand. I've been researching slide rails and stepper motors to build an XYZ gantry with a laser. The position accuracy is +/-0.2 mm. That's far better than pencil and jigsaw. For less than $3000 I should be able to build a DIY system with control software and firmware that allows me to cut a midship up to 90 cm x 90 cm.

Laser cutting doesn't have the forces of CNC so you should be able to do it for less than $1k. Laser cut depth is not optimal for woodstrip molds. $3k would support an actual wood CNC cutter. You could certainly hire the job out and save money and space. If you do set up your own CNC you could pay it off doing jobs for others or better yet, rent the molds out for a hundred bucks a season.

 

[Tops]

3Dirigo: The World's Largest 3D Printed Boat - Advanced Structures & Composites Center - University of Maine https://composites.umaine.edu/3dirigo-the-worlds-largest-3d-printed-boat/

The machine used to print the boat is huge (!) and the boat is relatively heavy IMHO. Still very impressive.

In my world, I have used hobby CNC to make parts and hardboard templates for boats.. Anything bigger than 22" (56 cm) on the hobby CNC and I am joining pieces together. I have made templates up to 12.5 feet (3.8 meters). First picture included is the 8 foot (2.44 meter) side template for a PDR (Puddle Duck Racer). The boat uses 4 of them and the template allowed me to to rough cut and router them all to match.

I have also done many templates with designs printed on regular 'A' sized [A4] paper, taped together into longer pieces, cut by a scissors, and then transferred to wood or foam via careful work with a soft artist's pencil. In all these cases you are still lofting, but in the computer first rather than in space. If the project is more than say 10 sheets I will opt to print them joined/whole on a big printer at work.

I have also made a project, a Tom Blake style surfboard 13'-9" (4.19 meters) that was a combination of stationed CNC sections and a solid nose and tail of stacked CNC pieces sections derived from a 3D model, second two pictures. The 3/4" (19mm) solid wood was recycled from a home remodel. The plans were from a 1930's Popular Mechanics article found online.

The ideas of having a cabinet shop with CNC cut the molds and making mold available for rent are good. The comment about making sure the data is clean and connected is important if you are paying someone else by the hour.

 

[gonzo]

For cutting plywood you simply loft the lines directly on it. Cutting to a pencil line is easy and fast. The multiple steps of CNC small pieces, then rough cut and then use a template for routing is unnecessary unless it is a proof of concept.

 

[Tedd McHenry]

So, had another unconventional thought... what is the practicality of using a CNC machine to print a boat layer by layer from sheets of plywood?

I think you'd want to use individual plies rather than plywood, so that you could control the direction of grain layer by layer. It would greatly increase the cutting time and gluing labour, but you'd end up with a better boat. Is that what you meant by "offset the cuts from one layer to another?"

As for clamping, you might be able to vacuum bag in stages, instead of using mechanical clamps.

Overall, though, I think it would be hard to beat the more conventional method of CNC-cutting plywood and making a hull that's all developable surfaces, like a stitch and glue kayak.

The other advantage is this could be used to assemble the interior at the same time.

I see that as a huge advantage. You might also be able to mix materials, with non-structural interior components made from lower-density materials. For example, interior components might be printed from "digital foam" instead of cut from wood plies.

 

[upchurchmr]

Fantasies where you never build anything are rairly useful.
Lets see any one of these ideas built.

There are too many messy details which ruin simple Popular Science type ideas.
Show us an actual development, instead of wet dreams.

If you are going to do computer aided design, you might as well not force your self into limited shape developable surfaces - make fully contoured surfaces, cnc forms, and strip planked hulls. Why waste all the time on something not optimum.

 

[Tedd McHenry]

If you are going to do computer aided design, you might as well not force your self into limited shape developable surfaces - make fully contoured surfaces, cnc forms, and strip planked hulls. Why waste all the time on something not optimum.

That's true if you have the facilities to build compound-surface designs or are building to production volumes that justify molds. But, for many applications, non-compound-surface designs make a lot of sense. I'll wager there are more stitch-and-glue kayak kits being sold than wood-strip, for example.

"Optimum" has to include how the boat will be made, not just how it will perform in theory.