So for a while I've been interested in 3D printing houses and more recently at container homes and tiny houses. Apis Cor is doing some cool things using eco-concrete that you can extrude. Container homes are a nice idea as well. But the problem with all of these for the colder or hotter climates is that you still need to add or fill in insulation and add weather cladding, paint it, make it airtight and avoid cold bridges. So after researching boat building I stumbled across cold molding boats.

Basically you build a shape out of foam, then spray on fiberglass and resin by hand and vacuum infusion. You could do this using a huge cable and winch driven delta printer (or steward using 6 wiches) quickly assembled using lightweight big poles (like 6 meters high) rammed into the build site and made rigid using steel cables like a tent. Then you spray fast curing PU foam in layers, mill it down lightly to make smooth surfaces and spray on chopped fiberglass and resin. So you would need changeable tool heads to switch from foam to (very) light milling and spraying paint.

Because of the adhesion of the quick setting spray foam you could create absolutely crazy architecture, curvilinear shapes and organic details.

You'd have a structurally sound strong wall with a hard and weatherproof surface after applying a top coat. The structure and materials would also be very lightweight meaning you cut down on transportation costs. Of course this has been patented already but afaik only in the US.

This is only a very rough outline and there would be many challenges. E.g. a large cable driven robot probably needs closed loop / 3D positioning control. And the nozzle for mixing polyurethane foam would be tricky. These chemicals are also toxic until they set. Or you might want to use PET (Polyethylene terephthalate) foamed with a insulation gas instead of PU (Polyurethane) because PET is stronger and cheaper to make / recycle and not toxic. But foaming PET is much harder and requires some arcane two stage pressurization / heating / cooling.

I'm not even sure I'll ever attempt this massive project, but I wanted to share and maybe get ideas or feedback or find others interested in something like this.



Edited 2 time(s). Last edit at 05/08/2017 09:09PM by Dejay.

Interesting idea, but from what you wrote, you'd have to mill the sides of walls layer by layer. So a dual head CoreXY gantry would be better suited or you get lots of tool changes. Also spraying the resin/fibre mix is not that easy to do after you've printed complete walls. ( No_go for a delta )
The towers of a delta or CoreXY would have to be perfectly aligned, which is probably the biggest challenge.

Maybe you could build a fixed frame from stage lightning profiles and change z-height only for the head(s) itself ( like a giant CNC mill )



Edited 1 time(s). Last edit at 05/08/2017 06:33AM by o_lampe.

A thanks, those stage lighting profiles (stage girders?) was what I was thinking of but didn't know what they are called! Of course in the spirit of reprap those poles / girders should be 3D printed as a fiberglass / foam sandwich

And I know this is going to be a rather rambling discussion on my part, it's just an idea that I keep thinking about so I want to share / discuss it. I should probably make a mockup of this idea.

Tool changes would be annoying. But a cable driven system would be much much cheaper, lighter to transport and easier to set up on a build site. I think cables are key to having a scalable and cheap system. No linear actuators. So you'd need a way to automatically swap tool heads through a magnetic system. Not sure how much time you'd loose for tool changes. The patent application shows a pic of a tool head that is turned 90° and mentions robot swarms but... not sure about that. The Apis-Cor 3D house printer also shows a 1 axis spatula flattening the print lines of eco-concrete. Maybe something similar could be done with the foam so you don't need to mill at all, or not that much. Or maybe you could also have two robots cooperating or maybe even two cable driven tool heads hanging from the same girders.

But in any case if you'd need to "stop" every layer or every couple of layers to switch to laminating with chopped fiberglass, then waiting for the laminate to cure... that could get problematic for print times. Maybe you'd need heat to speed up cure times. Maybe a big IR lamp to go over the laminate to fast cure and temper it. Or maybe not use traditional resin but spray liquid ABS / PET with fiberglass. Not sure if any of this is feasible, you'd definitely need someone like a chemist who is experienced with polymers for this.

And about perfect alignment of the delta towers - I think you need a system that can simply "magically calibrate that away". I imagine something like a computer vision system (QR codes) or kinect or lighthouse tracking system on each pole so they know where the poles are in relation to each other and how they are aligned and where the tool head is and where the ground level is. Because precision setup on a build site with lightweight poles and cables would be impossible, or at least very impractical. But with closed loop position tracking the algorithms could compensate for that and it wouldn't really matter. You'd also have to compensate for the cable drooping over distance or just irregular cable winding in the winches.

So the real key technology for this would be closed loop control through computer vision (those rectangular patterns) to track the print head and the position of the pole heads. The programming of the firmware would be tricky too, but not overly much so. Of course for development you could ignore that problem at calibrate manually / measure twice / estimate and compensate for things like cable droop. But it would only be practically useful if the system can do that automatically. The upside is that you just need a few poles and some steel cable and some winches and some camera sensors. You'd have a light truck to basically carry the whole equipment and barrels of foam / fiberglass resin / paint.

For material costs of such a house... I've roughly calculated some costs for this type of construction to something like 40-85€ per square meter wall / floor / roof. So too compare to a tiny house with 12x2.5x2.5 meters (40' x 8' shipping container / tiny house on trailer) would be something like 132 m² surface and somewhere between 5k and 9k €. The expensive part is really the laminating polyester resin. Traditional wood framing is probably cheapest in materials, but you would get a better insulation included in that and don't need any labour costs. And don't have any cold bridges in your insulation.

Cheers

To aid your searching ... stage supporting stuff is generally called truss or trussing, as a random example : [www.globaltruss.com]

Also the patterns for optical tracking are often known as fiducials: [en.wikipedia.org]

UV is used to cure many resins, you could have banks of UV leds on the printhead to get the outer skin set quickly.

Triangular Truss Systems might be a little cheaper, it would be good to move to the next level 3D printing uses, for either homes or boats, but maybe the old ways are better, as many times people make things with 3D printers that can be done better with other materials(eg, dont print a coat hook when you can buy one to do it better/cheap) But still good if a real world example can benefit from the 3D computer control of some tools, at least for making molds for parts, or to assist in the making/shaping of the molds.
Stippling/texturing the concrete to remove the slug trails...I think i'd like to keep resins out of the home, but would have to accept it if living on a boat.

Edited 1 time(s). Last edit at 05/08/2017 02:45PM by MechaBits.

Thanks Zed²!

UV could be an excellent option. I wonder if UV curing resin can be stable in sunlight? You could apply a top coat of course that doesn't have to cure before the next layer.

The trusses don't need to be very strong though, since they only carry like 50kg max and can be tightened using cables tethered to the ground like a tent. The moving parts would be rather lightweight. The cheapest option would be something like 6 meter long 45x45mm or 90x90mm aluminium profiles from motedis. Maybe they would have to carry a bit more if you'd put a kind of circus tent over it to create a cleaner / more controllable print environment (temperature / light / humidity / wind).

Well for boats / yacht moulds this might work too. Although this is already done just using CNC on big blocks of higher density PU foam for a stronger mold and more precise surface finish. Since molds are reused lots of time and only in series production, optimizing the building costs of a mould for a yacht isn't that efficient.

Ideally this technique would be use for building one-offs like a house, or a house boat without a mold, a unique shape. Boats is why I was looking at PET foam - it's cheapest of the suitable sandwich foam materials for boat building (strong, elastic / not friable, closed cell). From what I've read PU foam isn't used for boat hulls. And PET is non toxic during spraying I think. While PU foam vents isocyanates when spraying. So the ideal foam would be PET with an inert / insulation gas as the foaming agent.

About concrete, the patent also mentions using the foam as a casting mold for concrete - so instead of building wooden forms to pour concrete in, you simply use foam, then leave it in as insulation. This might be useful even if you only do this for the foundation or a few structural elements / columns / framework.

Edited 2 time(s). Last edit at 05/08/2017 03:07PM by Dejay.

I've attached a demonstration of the idea and my lack of skills in Fusion 360 to the original message. Imagine there would be cameras and QR codes on the towers and the effector platform. It would probably be better to design with 4 towers and 8 cables. That would make the calibration more complex.

Small link collection:

Hangprinter
[reprap.org]
[www.youtube.com]

B&TRap (page has some more links)
[reprap.org]

Path planning for cable-driven robots
[www.iri.upc.edu]

Stadium-Sized 3D Printers using Winch Robots (aka Cable Robots or Rope Robots)
[www.hizook.com]

Spiderbot
[www.bnpeters.com]

There is the Fraunhofer 8 cable simulator / robot.
[hackaday.com]

Also a few papers from this:
Auto-Calibration Method for Overconstrained Cable-Driven Parallel Robots
[ieeexplore.ieee.org]
(This is needed for more than 6 cables)


This is probably a crazy idea: You could also add even more cables to "pull down" 3 of the cables. You'd have 3 cables coming from the bottom and connected to 3 of the top wires with pulleys. By adjusting the tension of the bottom cables you could adjust basically where the top cable starts above the ground. That would reduce the problem of cable slack, limited acceleration and wouldn't limit the print area that much as having some of the cables coming from the bottom. Never mind, this would be a better idea:
You could have half of the winches on moving sliders rolling up and down the towers. That would create a linear actuator and all the problems that entails, but it would allow you to have tension from below to decrease cable droop, increase print speed and not have to deal with slack cables at certain positions, while also not limiting print space like the hang printer.

Edited 3 time(s). Last edit at 05/08/2017 09:35PM by Dejay.

Let's say you'd get the printing part sorted out, how and when do you install water, electricity, waste water installation?
It would be nicer, when all this was done before the print starts, but the printhead(s) would have to work around the installations.
I've seen a TV-docu about building motorhomes and caravans/trailers, they already have a huge amount of automated production based on aluminum/foam sandwich walls.
If you'd preproduce single walls with installation in a production facility and only have to mount them together on-site, that's the way to go IMHO.

Edited 2 time(s). Last edit at 05/09/2017 03:41AM by o_lampe.

For the installation, I guess you'd just have to plan ahead. Electricity would be easier than plumbing. I would simply have a cable channel going around all the walls around the house (on table height!), so that you can always push in new cables and drill new though-holes into the cable channel if you get a new TV or lamp. And if mess up your planning (like you do), you could always cut into a wall or floor and the foam, insert something, then use spray can foam and fiberglass laminate to fix it up. I think it would be easier to modify because you have less materials and techniques (foam + fiberglass + resin).

I don't really know much about plumbing, but I guess you do install the waste pipes into the foundation? Then you could either mark the areas in your CAD program with preexisting objects. Or have some advanced computer vision / SLAM algorithms (there are finally some inroads for VR / AR) but that wouldn't really be needed. You'd still need an overseer on the site to set the system up and make sure everything goes smooth. And you'd definitely want to consult professionals like architects to go over your plans.

And ideally you'd also have algorithms for your architectural program to add smart wiring and plumbing. E.g. collect grey water separately, use filtered rain water for showering and laundry etc. Ventilation is another thing that gets more complex with the heavy insulation codes that make your house basically airtight. For example you'd want ventilation that has a heat exchanger so if it's hot the incoming air is cooled by the outcoming air and vice versa. Or air coming in through your attic that gets preheated and filtered. 3D printing and better computer aided design could make all these issues cheaper to implement.

Yeah the RVs are made from separately premanufactured fiberglass laminate sheets glued onto the foam panels, then use big CNC machines to cut out windows and doors and hatches. I've only seen a video of a luxury RV maker and they had real carpenters and a lot of manual labour to build the interior of the RV. Personally I'd like a more artificial / sci-fi look though. 3D print the furniture too!

But it's all rectangular! I have this idea of having more "ornate" architecture. Maybe it's just me. Something like this for example: https://www.google.com/search?q=beautiful+ornate+houses. Or this guy who makes amazing tiny houses out of wood. Of course you'd have to paint it yourself (although that could be automated too) and you can't "paint on" wood but hopefully you get the idea. I once saw a photo of a beautiful while temple and that was what brought me to the idea of 3D printing houses (this one). You just can't pay for the manual labor anymore for something like that. You could also add procedurally generated or fractal details to your walls. The possibilities are endless if you could add full color painting to the house printing. Of course you'd also want a lot of your furniture 3D printed and fitting into a more unusual architecture. And really more LED panel area lights.

Edited 1 time(s). Last edit at 05/09/2017 04:32AM by Dejay.

There's always the concrete 3d printer. [www.youtube.com] Of course, he didn't print a house, but still I'm sure the application could be adapted.

Quote

You just can't pay for the manual labor anymore for something like that.

I believe, the white temple was build by slaves, like the pyramids ( or volunteers/interns like they are called today ). Otherwise one must be very rich, and their wealth has made other people poor.

Quote
PDBeal
There's always the concrete 3d printer. [www.youtube.com] Of course, he didn't print a house, but still I'm sure the application could be adapted.

Yeah I've seen and love that little castle!
Apis Cor made a 3D concrete printer as well and they use "eco-concrete" which has the advantages of creating less CO2, contains reinforcing fibers (no rebar needed) and most important of all also being better for pumping since to stays fluid under pressure. So they can use a long hose.
The WASP project in italy is also working on 3D printing using mud / cob and concrete using a small reservoir extruder to avoid the pressure problem.

Just to reiterate my thinking: Foam would be better than concrete since it's more lightweight and you can print overhangs and window frames (maybe even straight ceilings). And you kind of need good insulation for a house anyways. PU foam is the best insulating material because it is airtight and fills up all gaps, and the CO2 contained in closed cell structure conducts heal less than air. And then after the insulation you really only need a little bit of strong surface laminate. Transport costs go down.

So why even use concrete for smaller houses then? You need to surface coat concrete anyways for a wall or a roof to make it water / weatherproof. In fact you need multiple layers of different surface materials in traditional construction. Instead just use fiberglass as an established roofing material comparable in cost to other roofing materials.

Of course this is all speculation since I'm not a structural or construction engineer.

So after learning a bit more about PU Foam and fiber reinforced plastics (for a rather simple project) I just want to add a few thoughts and ramble a bit

Yesterday I remembered continuous fiber printing. It might be much better to build up a shell for the walls and then filling this shell with foam. That way you don't need to mill foam away and then spray on fiberglass. It allows more detail and easier fabrication.

Markforge offers a printer and a carbon / fiberglass filament that apparently is simply a roving that is "prepreged" in nylon. This roving filament is layed down on a freshly printed layer and then their proprietary slicers calculates how much less normal (nylon) filament it has to add on top of this. Good bonding with the "matrix" (epoxy, polyester or nylon) is important so this carbon filament is prefabricated.
The bastards are even trying to patent the software for the slicer and the calculation of strength, while their work is obviously inspired or derived from threads on this very forum (Fiberglass thread extruder from 2010, SpoolHead, Fiber Composite 3D Printing (The Bug), Markforge Carbon Fiber 3D printer, Fibre Core). Are there patents on this by Stratasys?

The downside of this is that you need a new fiberglass extruder, you still are limited by layer adhesion / sheer forces and increased print time compared to just spraying foam. This technique is also similar to Filament Winding. Since you would need a larger extruder it's development could be simpler

Just last month someone actually made a working and small filament extruder for cotton reinforced nylon 3D printing by modifying a standard extruder with a special nozzle. The fiber could be cut in advance before it enters the nozzle. Info on instructables and hackaday.

Quote
"Using cotton thread and a modified 3D printer nozzle of my own design I created high-strength composite parts with a flexural modulus (stiffness) over 14 times greater than unreinforced nylon with my low cost Prusa I3 printer."

An alternative to nylon could be a UV resin as used for "The Bug". Fiberglas is extruded with a low viscosity resin that is cured using UV Leds to create "overhanging" structures.

To really solve the problem with layer adhesion, you would need the ability to 3D print vertical lines on top of previously printed horizontal lines. To allow for at least a limited amount of free fiber orientation. So a kind of 6DOF robot arm that ideally can move or drive around and inside the structure you are printing.

My idea of a 3D carbon fibre printer looks more like a 3D-loom, where single strands of carbon rovings are connected at the crossing points by a drop of resin and an UV laser. This mesh would look like the final part, but needs epoxi filling and curing in a vacuum enclave oven the classic way.

Haha I've actually just seen something like that here. But I figure that was made by a removable mold. "The Bug" does exactly what you describe though, but the whole roving is drenched with UV curable resin.

I guess the right way to think about this would be that you want to print with fiber in a plastic matrix, not extrude plastic together with some fiber. You want as much fiberglass as you can get for a strong and lightweight part. Per weight fiberglass is much cheaper than plastic or resin too, something like $2 per kg.

So you'd want really thick rovings that are impregnated with the resin or liquid plastic. The "extruder" could be at the tip of the hotend, pulling the roving through the melt zone, and rolling on top of the existing surface to deposit the filament. In a fiberglass chopper gun a rotating "blade wheel" cuts the fiberglass rovings when they pass between the blade and an elastic rubber wheel. So they can't be too hard to cut.

This "deposit wheel" would have to turn depending on the position. I guess it wouldn't have to be motorized. As long as you know which direction it is currently facing you could move the print head in a way so that the wheel automatically follows after the print head.

But for this to work you'd need a 6DOF robotic arm printer. And a very smart slicer that calculates a solution and the order of operation for depositing first vertical lines to build up something stable and then vertical lines. Well ok you could also have just short strips of fiber that connect a few centimeters of print lines. But the software might actually be far more challenging than the mechanics of this.

So I had another idea for a more practical and simple solution to this:

You could simply print thin shells for the walls out of normal 3D printer plastic, then spray foam into them and add laminate fiberglass cloth on the outside by hand. So every 20 or 30cm you would stop the print, use those 2k PU foam spray can's to fill up the wall shells with foam (the spray cans are pretty cheap compared to rigid insulation foam panels probably because they don't take up that much volume and can be mass produced and shipped). Then continue the print. PU foam gives a lot of rigidity. Every 1m or so you add fiberglass cloth. It comes in rolls, basically you paint the wall with epoxy or polyester resin and put the fiberglass cloth on it, soak it and roll out any air bubbles.

It's more work by hand but you can design a complex intricate building on the computer and have a large scale 3D printer create it. You'd still need a large scale cable robot of course, but no special extruders or new technologies or crazy 6DOF robotics or new slicers that can make them work. An E3D vulcano extruder would be enough. And straight ceilings would need to be added by hand. Maybe they could be 3D printed in another orientation and then inserted into the print. Theoretically you could print the whole house in parts on a smaller printer and put it together like lego. Basically this would be 3D printing SIPs (structural insulated panels)

Of course a wall with 1mm thick walls out of PLA would still require 2kg of plastic per square meter. But compared to traditional construction it could still be quite cheap. And probably you'd want to use a plastic that is tough, chemically bonds well to both polyurethane foam and either to epoxy, polyester or vinylester resin.

EDIT: Plus, this could actually be done in a factory. 3D print all the SIPs that make the fancy house so they all fit perfectly together (already filled and made rigid with foam), then deliver them on a truck to the build side and glue them together, then add the fiberglass to make them really strong.

Edited 2 time(s). Last edit at 08/12/2017 03:00PM by Dejay.

You forgot the expansion rate of 2k-foam. I've seen metal cases popping apart when too much foam was filled in. There is some force behind that expansion that will bulge out your PLA walls.

Yeah I think you are right. I might have wanted to ignore that for now There are other problems as well like how would you 3D print such large structures without warping when the plastic cools. This might just not work, and the approach to freely spray foam and mill it to shape might just be much easier.

Does PU foam create pressure even if the form is open to the top to let the foam expand upwards? Maybe with only thin layers of foam? Maybe you can formulate it in a special way, or preheat it so it expands and reacts more quickly, instead of slowly and creating pressure with half cured foam. I think those professional wall foam sprayers preheat and cure quickly.

Maybe you could even use a different kind of foam, like XPS foam extruder or PET for much higher strength.

Even if you could 3D print with continuous fiberglass rovings, just a bit of pressure could easily push the walls apart. Maybe some kind of fiberglass infill, bridging the two walls would be enough.

just stumbled by this thread

Thought you may be interested in the MIT robot that prints houses with spray foam
https://www.youtube.com/watch?v=Rgp4ncc1wOQ

... here a pretty old related thread

[reprap.org]

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Oh that is interesting VDX. This even predates the MIT patent! (2007 vs 2013)

... my first LOM-type DIY-3D-printer dated from around 1986, so some years after the basic patents from 1981 (didn't hear of 3D-printers then - so all ideas and work around was "self-invented" ).

But could be, that some of my later ideas and IP-files triggered the development of applications and technologies around laser- and micro-tech-applications

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I mean especially the foam printing patent filed by the MIT. Your post is definitely prior art to at least the basic idea of printing houses using foam. I was previously somewhat discouraged this is patented. It's also nice to see that no idea of mine is ever original haha. All the ideas in this thread were basically already there in 2011!

I think 3D printing buildings with foam has a huge advantage: the light weight, sticky nature and fast curing (if you do it properly with heating and pressure) allows overhangs and fast construction. Plus foam insulation "needs" to be an element in houses anyway. I've also researched about about foaming agents and there are new developments there too. You can formulate the PU / PIR polyurethane foam to be denser have have higher compressive strength.

The real challenge (besides a large scale lightweight robot arm and milling the resulting lumpy surface smooth) is how to apply a skin laminate that provides the tensile strength.

I've seen some carbon fiber laying robots for airplane manufacturing that do some awesome things. A giant robot laying carbon fiber bands on curved surfaces using rollers and such. If you would use cheaper fiberglass rovings and have some kind of fast curing resin (UV or heat) then you could move the robot arm around the structure to apply the laminate and create a very strong composite material in any crazy shape you'd want. You could even print 90° overhang for horizontal ceilings or openings.

The MIT robot video linked by DaGameFace mentioned "manual labour is still cheaper" but I think that is kind of a silly thing to say. This technology would need a lot of investment and development but it has the potential to produce very elaborate and ornate houses without any human labour at all. Things like that could be great for creating or rebuilding houses when disaster strikes (e.g. Syria when the civil war finally ends, or in the future migration due to climate change).

I wonder if you could get investment or crowdfunding to develop this. But I mostly just like talking about these cool ideas, personally I'm pursuing more humble projects atm

... there were much more ideas around this - some from me, some from others -- or "created" while brainstorming

Image JIT-made inflatable parts from contour-cut+fused plastic sheets (think about pre-fabbed inner and outer surfaces or colorizing), coated (or 3D-print-coated) with foam or wet concrete -- and inflated to the end shape, before it cures/hardens.

Another very interesting concept is pre-fabbing constructional elements (or add them as ready parts) into the fabbing process ... imagine "online printing Legos" and assembling them to more complex products

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I do think you need some kind of composite material. Concrete with rebar is a composite material, without the rebar providing the tensile strength it wouldn't be any good.

You could either laminate continous fiberglass rovings or spray chopped strand fiberglass with a copper gun. Or maybe even mix chopped fiberglass into the foam mix on the fly. Maybe you could also use spray concrete with basalt fibers mixed in.

And yeah, I think 3D printing puzzle pieces with some continous fiber and filling them with foam could work well too. If you could print cheap plastic like PET and bond it with fiberglass it would be cheap materials and very tough.

But what I've found after some research into boatbuilding, resin is rather expensive and you need quite a thick outer surface for enough impact resistance. E.g. for a foam cored boat you need at least 2mm fiberglass laminate else rocks will pierce it. Foam gives stiffness, but not impact resistance. So up to a certain boat size the most lightweight way to build is still to use wood laminated with much thinner and less fiberglass and epoxy. Wood has amazing nano-fibers. I've also read about some research where they removed the lignin from wood to either turn it into a great lightweight insulation material or compress it to have a super dense super-wood that is stronger than steel. Not really helpful for 3D printing houses though

... you can use "chips", sheets and prefabbed/shaped parts of this "compacted wood" for "puzzling" too - so a form/structure made of foam and after (or while puzzle-fabbing) coated with bullet-proof wood chips should be strong/rigid enough

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

hyy, i search large 3d printer. i have some ideas, but i wants to hear about more options. i will be happy for recommendation.

Quote
VDX
... you can use "chips", sheets and prefabbed/shaped parts of this "compacted wood" for "puzzling" too - so a form/structure made of foam and after (or while puzzle-fabbing) coated with bullet-proof wood chips should be strong/rigid enough

Not sure how you would shape wood chips. And I think wood is only really as good as the lenght of the wood fibers.

You could slice wood veneer (e.g. what plywood is made from) into long thin bendable "wood noodles" and then 3D print those by glue them on top of each other. Kind of how plywood is made but more in the fashion of 3D printing. You'd just need to start stop constantly, load a new wood noodle, have a fast curing (and watertight) glue, and an extruder that can roll and press these wood noodles on top of and next to each other. Material costs would be cheaper and more sustainable than fiberglass rovings in a resin matrix, but this whole idea is crazy complicated

... I was referring to your "super-wood" - while making, it's super tight compressed vertically, but will keep the XY shape -- or you'll press it into a curved form, so it will get the shape of the form.

Think about the medieval leather armour, made of many overlapping pieces, cut out from hardened leather ... or more modern, bulletproof vests, made with many ceramic tiles in a puffer matrix of fabric and foam.

This "super-wood" chips, arranged (overlapping) on the surface, will give a pretty robust shell surface for a boat, made of construction foam

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]