Governments, for disaster relief, per the article, for starters. Me, since I'll have the printer anyway.
My impression of a 3d printer for houses is that it would have to be bigger than the house. Getting it to and from the site to the next job would be rather daunting.
The gantry from the picture is taler than a house, and the rrails are longer than a house, but there could eaily fit in the back of truck. And a truck would work fine, no?
3D printing is optimal for thermoplastic.
".. at this point in time". This is only because that's where everybody is focused. Concrete pumper trucks are quite common around here, its only a matte of time before some braino decides use one with a printer (Ferb, I know what we're going to do today...)
There just seems to be an economic limit to the size of thermoplastic items.
Governments providing free housing for disaster relief? Didn't FEMA buy trailers for victims of Katrina and then hold back on them? And I don't see much in the way of governments providing housing for Haiti since their earthquake. In general, everyone talks about providing housing for disaster relief, but follow through is mostly forgotten.
We did have a village of 600 here in Taiwan that was destroyed by mudslide and they were given a whole new village with new housing, but not by the government. It was by the Tsu Chi Buddhist Disaster Relief Fund. The only stipulations were no one can live there if they smoke tobacco or drink alcohol. Some people actually refused due to these terms.
I am sorry, but the reality is that most houses are provided for paying home buyers. And housing - both new and the repair of old - are rather important engines of the economy. They provide jobs for a wide variety of craftspeople. This is why the USA is having a 'jobless recovery' - nobody can afford to buy housing.
These days everything is downsized - restaurants no longer have waiters and dishwashers, you serve yourself; banks no longer have tellers or even branch officers; you visit the ATM and have to deal with people in authority over the phone or internet; gas stations are self-service in many states; and on and on.
The simple reality is that everyone that doesn't have a job to depend on is depending on someone else to provide for them. The whole economic model of off-shore manufacturer and a consumer economy is turning rather grim.
Needing a crane and rails to put the 3D printer in place? I built homes in Oregon and the most we needed was a backhoe and a dump truck (unless you needed to drill a well). Generally, the materials are two or three semi-truck loads and the foundation concrete is 3-4 truck loads.
I realize this guy is an engineering professor in Southern California, but he sound more like an 'ivory tower' professor than a nuts and bolts engineer.
If anything could be done like this, I suspect an automated foundation excavation and installation would be optimal, not the whole building. But the simple fact is soil varies from mile to mile and each foundation take a bit of a different approach.
Once the foundation is in, construction is pretty standard. It is nice to have a small crane to lift the roof trusses into place, but they can be lifted by 3 guys if no crane is available. Or an all-terrain fork lift can do the job.
Yes, you can use concrete pump trucks, but don't expect the concrete to have adequate strength to retain shape immediately. It isn't cooling, it is setting.
As I mentioned above 'shot-crete' - a form of sprayed concrete plaster is about the only fast setting concrete available and it is rather nasty and messy to use. People have even sprayed domes out of it that are water tight (and boats). But it all starts with having the steel mesh in place to make the shape -- lots of steel.
Then there is the problem of when the printer goes haywire and prints ten copies when you only wanted six.
No one uses those automated road-building machines for a half a block of pavement. They're meant to construct tens and even hundreds of miles of interstate. That's where their cost-savings come in. You won't find automated home construction building the one-off, or even a small sub-division. It only really works if you're building hundreds, thousands, or tens of thousands of units.
(Speaking of automated and semi-automated road construction, isn't it curious that Caterpillar -- who's business is building trucks and other machinery for making roads -- is contributing to the 3D home concept? After all, a graded road with concrete median and curb is but the same concept, just a little flatter...)
Where these homes are most needed, standard concrete is not the best choice of material (in the video he talks about the example wall being made of a composite material with about 3X the PSI rating of reinforced concrete). Water, not sand, is the more precious commodity.
You guys talking about the traditional methods of construction are still thinking in the Western or developed world mindset. Experienced laborers for safely building houses are hard to come by in most undeveloped countries. Like these places are full of skilled roofers!
I mostly agree with Loopy, obviously, although there is one caveat; with modern accelerants it's possible to lay down a paste style concrete that achieves a good fraction of its strength in hours. This is what is used to create the shells for Monolithic Domes. However, this material still needs reinforcement and part of the Monolithic Dome construction process is manually placing rebar reinforcement before the shotcrete is sprayed. (MD's begin with a shell of polyurethane foam, which is strong enough to hold up the shotcrete while it cures.) There are schemes for embedding steel fibres in shotcrete as it is sprayed, but they are nowhere near as strong as actual dimensional rebar. It's also important to deal with rebar rusting; most accelerants ruin the alkalinity of concrete, which protects the reinforcing steel from rust. Rusting reinforcement can actually blow the concrete apart because iron oxide is bigger than iron. And all concrete is porous.
I seem to recall reading about experiments where either Fiberglass strands or carbon fibers were mixed in with the concrete when it was poured. Something like that may end up being an option for a large 3-D printer like this.
Even if you take it to places that need the housing, like Pakistan (after the recent flooding) or the Phillipenes (more recent flooding), the economics to do it is just not in place. Saudi Arabia and Kuwait used to be the ideal places for large scale housing projects, but no longer.
Given that you have a huge tract of land, you have a rail spur with a complete concrete batch plant, a fleet of concrete trucks, and another fleet of concrete pumps, and adequate supplies of rebar, windows, paving materials, and so on - you could use a 3D printing concept to build.
On the other hand, kits of pre-fabricated steel geodisc domes that can be assembled by local residents in areas requiring immediate disaster relief due to on-coming winter or typhoons would have immediate shelter that could be later relocated or enhanced where it stands.
Conceptually it is a neat idea. But the will to make it a reality is rather dubious.
On the other hand, kits of pre-fabricated steel geodisc domes that can be assembled by local residents in areas requiring immediate disaster relief due to on-coming winter or typhoons would have immediate shelter that could be later relocated or enhanced where it stands.
They already do something similar for emergency housing, domes or otherwise.
You mentioned trucking costs and transportation complexity for 3D houses, but you make it sound like all the steel for these domes just fly there on their own. And where are they pre-built to begin with? You won't find steel plants in most of the countries we're talking about, so that probably means shipping, which for the cubic feet of material talked about here is astronomically expensive. A non-starter. At least with earthen construction the raw materials are more or less nearby.
Fibers other than steel flex too much, and don't assert themselves until after the matrix material cracks. This makes such fiber-reinforced concrete significantly inferior to steel reinforced, even steel fiber reinforced.
Gordon, the limiting factor in shotcrete is portland cement. Portland cement is expensive and standard concrete is mixed to minimize its use, by carefully filling as much of the final space with carefully graduated aggregates. Most ready-mix has aggregates up to a couple of centimeters in size, and a typical ready-mix recipe is less than 1/10 portland cement by weight. For shotcrete and mixes that have to be pumped through nozzles like the one in the video, you can't use large aggregates. This means you have to use more portland cement, typically around 1/3 by weight, which makes the mix stronger whether you care about that or not. It also makes the mix more expensive no matter what other additives you use because portland cement is 10 times as expensive as aggregates (much less water to cure it). Additives are a significant additional expense for shotcrete and, I would assume, for the video's extrusion mixes.
Unfortunately, no matter what you do to it the tensile strength of unreinforced concrete remains Smile, and unless carefully built to leverage their mass in compression (like the buildings in the video from Iran) concrete structures need a lot of tensile strength simply to safely hold up their own roofs.
I suppose that the real demand would be for a rammed earth robot. You can use rammed earth from local soil and enhance it with a variety of local available materials. China has structures that have lasted thousands of years from rammed earth made with manual labor. In fact, major portions of the Great Wall of China are rammed earth, some covered with brick veneer.
Steel too heavy for shipping, aluminum is half the weight of steel for the same volume. And concrete is actually a bit lighter than aluminum per cubic foot (about 5 pounds). But all of these are sourced from rich industrialized countries and there are even less aluminum industries on a global basis than there are steel.
Steel mills can be constructed as local mini-mills with local steel ore deposits and local coal. The quality may be a bit poor, and getting it actually rolled into structural shapes is another issue. But, it can be done if a country has the will and the capital. At one time, I helped write a proposal for Nigeria to do so and it was actually considered. But, the government was overthrown a few months later and the project dismissed.
I think you begin to see, that any realistic project will have source material for a robot from local material and not depend on a high energy cost material provided from an energy rich country. So items, such as roofing membranes may be worthwhile to import, but the floor and walls and roof trusses are best provided locally.
In sum, architecture has always had to rely on the local wealth to be able to select the best techonology for a project.
Has anyone ever seen a Volkswagen driven lumber mill. That was listed in the Whole Earth Catalog of the 1960s and can be moved from site to site to mill all aspects of a home's construction. Not much need for a robot though.
The problem with materials like metal is that they are too strong. You have to use far more of them than you should need for secondary reasons.
For example, while most species of wood test to 3,000 psi or so in compression, for safety reasons and to allow for flaws it's usually allowed 500 to 1000 psi in construction. This means a 2x4 stud in a load bearing wall which is constrained from buckling can support several thousand pounds.
Steel and wood cost about the same per pound, but steel is 10 times as heavy as wood (400 lb/cuft instead of 40 lb/cuft) and also about 10 times stronger (40,000 psi) This means the column of steel you need to replace a 2x4 stud should have a cross-sectional area of about 1/10 the cross sectional area of a 2x4, or about 1/2 square inch.
In practice there is no way to keep such a shape from buckling unless it is externally constrained. A wood house that has been framed but not skinned will support itself; cold-formed steel studs are very wobbly and unstable until the sheetrock or siding is put in place to stabilize them.
If you use hot-formed shapes like I beams a structure the size of a residential house becomes massively overbuilt and, therefore, hyper-expensive. It's similar to the problem with concrete for only slightly different reasons. Wood really hits a sweet spot between cost, strength, and stiffness which nothing else comes close to matching. And plastic materials, including anything bonded together with epoxy or other resins, are fantastically expensive compared to the materials normally used for construction.
Dirt, of course, is cheap but dirt hits you in the labor and material staging pocket. Because its tensile strength is so low you must use truly massive amounts of earth to build a practical wall that doesn't buckle, which means that even if the dirt is free the cost of making it into bricks and moving it around becomes prohibitive. This is why there are very few outfits that sell adobe bricks; shipping costs eat you alive if they have to be shipped more than a few miles.
Thinking out of the box is a good thing of course, but there are a lot of sound reasons why modern construction looks the way it does. I have my own issues with building codes requiring excess material use to pad the pockets of various industries, but a lot of the basics are no different than they were 300 years ago.
The bottom line is that no matter what material you use, if it has nearly perfect properties you will use a minimum of ten thousand pounds or so of it to safely enclose a typical residential house-shaped space against typical weather, animals, and burglars. Then, if your material is at all imperfect in any way, you start adding to that. A modest adobe house can weigh a substantial fraction of a million pounds (which also gets you into foundation issues). And the list of materials cheap enough to deploy in such quantities at all is rather short. There are no plastics on it.
Adobe houses are workable in the SW and in poor areas. And there has been research done to make them more resilient and so it doesn't take a trade school grad to build but locals using low tech equipment. The only draw back it's low tech and not sexy like throwing a $20 million dollar robot at it with a small army of logistical support vehicles and equipment to support it.
The only draw back it's low tech and not sexy like throwing a $20 million dollar robot at it
As I pointed out in the very previous post, a major drawback to adobe is labor and transportation costs. Now if you have an army of unpaid and available labor sitting around with nothing else to do adobe can make a lot of sense. But in every accounting I've ever seen a first-world contract built adobe house is significantly more expensive than a conventional frame house, and it's because of the labor and transportation. People who aren't doing communal sweat equity don't build adobe unless they want it for the thermal or other architectural properties.
Here's the rub, you don't have all that nice 1st world materials and a gold plated logistics system in place(no Lowes or HD). You work with what you got. Labor, well there's plenty in India or Pakistan where the that robotic joke and it's army of expensive logistic support vehicles.
And BTW if you ever bothered to look at what constituted housing in these regions or south of the border, mud bricks made up a lot of it, made from on-site local material. The same in pre - WWII Eastern European country side.
Of course these people didn't have the luxury of modern Western accounting methods so they have some housing instead of none. They didn't have to pay some trucker $250 to haul a single load 2 miles to the work site and people worked together(try getting a bunch of Americans to work together in a communal setting and they'll sue each other). They didn't have to hire so-called professionals(that charge $200 just to walk in the door and charge $2000 and take 3 days to dig a small ditch) so costs for them were and are minimal.
The other thing is that these people don't need the massive living space that Americans need today to get by so the dynamics change again.
Bottom line: Home building in the 3rd world isn't rocket science and a lot of practical people have given it serious thought and worked out alternatives that take into account local resources and skill sets.
Robots created by ivory tower egg heads aren't needed.
Of course these people didn't have the luxury of modern Western accounting methods so they have some housing instead of none. They didn't have to pay some trucker $250 to haul a single load 2 miles to the work site and people worked together(try getting a bunch of Americans to work together in a communal setting and they'll sue each other). They didn't have to hire so-called professionals(that charge $200 just to walk in the door and charge $2000 and take 3 days to dig a small ditch) so costs for them were and are minimal.
As a contractor myself, I'd like to clarify: you're paying a reasonable rate based on market demand and contractor overhead. Have you looked to see how much you cost whatever company you work for? It's not just salary, it's benefits: healthcare, accounting, legal protection, physical resources, discounts at places, employee education opportunities, and job security. A contractor has to cover all of that themselves, and so have to charge more than the "standard employee $/hr".
And anyway, even if in 3rd world countries "they didn't have to pay" and "people worked together", that doesn't reduce the cost: the homeowner must reciprocate and give up his own time to help his neighbors (==cost). It's merely a different form of exchange.
My spending is your income and your spending is my income. That's the way economic growth works and housing that is localized and labor intensive is important as an engine of economic prosperity. Of course, it is sad to witness many people not wanting the jobs and allowing illegal aliens to do the work and remit their gains to their home countries with taxes unpaid while they enjoy the benefits of their host country.
But at the core of all this is a very real desire to own your own paid for home as security.
When you import a robot and technically complex materials to do the task faster and 'better', you are spending your money elsewhere. It goes to a corporation that hopes to achieve substantial profits by diverting money that would have been spent locally. Where they divert it to is rather awkward - maybe for good purposes, maybe for the bad.
In most instances, where corporations and governments can help is not particularly in providing housing technologies that are more efficent, but in terms of providing safe and sane locations for housing. And also providing education in how to DIY shelter that is safe and useful over the long-haul.
The issues here are far more complex than one neat techno=solution. The realities are that we are NOT building on the Moon or Mars at this point. (And in both places, tunneling might be the first phase of habitat construction). And we are not providing anywhere near enough housing for disaster relief in this disaster prone world. In the other marketplaces for housing, people are pretty much accepting the product available as being an improvement over past efforts and progressing towards more comfort, even if they might be duped (Modern wood homes in America are just not as durable or repairable as the homes built before WWII).
So, this requires a multiple solution approach and some real motivation to provide better housing with better infrastructure to those that need it. There is a corruption that favors providing housing for those that will pay enough for middlemen to make a profit. All too often, the people that are willing to DIY a home are not empowered by local construction codes and zoning. And all too often we see urban renewal removing a so-called blighted neighborhood by removing the people that cannot afford better housing and providing a new environment for more affluent people. This seems to have happened again with the London Olympics redevelopment of London's East End.
In other word, there are a lot of 'grand schemes' that tend to favor the promoters, not the person in need of decent shelter.
In the 1960s and 1970s Japan invested heavily in industrial robotics to assist in automobile production. While the specifics of the challenges were different, the general criticism is not much changed -- it's cheaper with manual labor, the machines are more expensive than traditional methods, the machines are not capable of working with traditional materials and assembly line methods, whatever.
History has proven those naysayers wrong. The Japanese introduced robotics for quality consistency, and along the way developed manufacturing techniques that better meshed with the needs of robotic assembly. These days you don't even want to buy a mass produced car that's been hand-made. Too much chance for human error. The cost reductions due to robotic manufacturing, while now considerable because it's become it's own science, is secondary.
Amazon recently purchased Kiva Systems for three-quarter of a billion dollars so they could be on the forefront of automated warehousing. More than 80% of all warehouses in the country are not suitable for a single robot, let alone an army of them. Does this mean that because there are technical challenges to overcome that robots have no place in warehouse automation? If you think so, better write a letter to Jeff Bezos and tell him he blew $750+ million on a harebrained idea.
Yes, building a whole house automatically is a far-fetched idea. Who says it wasn't. It's a "grand challenge." It's made to be difficult, and far-reaching. Like other grand challenges, it's also made to act as a springboard for smaller ideas along the way. You don't have to be a genius to think of the simpler alternative uses for this. It's pretty clear Caterpillar isn't interested in getting into the home building market. So it's quite possible they view the same machine for making roads, curbs, pilings, road dividers, just about anything you can think of, in any dimensions, any shapes, using many types of flowable and air-cured materials.
I'll remind folks that desktop 3D printed objects are basically useless in comparison to injection molded plastics. A 3D part is much more expensive on a materials basis, and far weaker. I don't see these limitations stopping people from trying new things and pushing the end of the envelope.
using many types of flowable and air-cured materials.
There is exactly one material which can be mixed into a flowable form and permanently chemically set into a solid, and which is cheap enough to use to print a structure the size of a house. That material is portland cement. There is nothing else. I spent the better part of a year investigating alternatives, and no matter how frugal you are with aggregates and meshes and eggshell arrays every single alternative either has even more terrible mechanical properties or is a hundred times as expensive.
Even portland cement isn't so simple; if you want to use it as shotcrete or ferrocement you must use more because you can't use large aggregate, and it's likely that the admixtures (at a minimum you need accelerant, air entrainer, and plasticizer), while a relatively small physical part of the final mix, will cost nearly as much as the cement.
The only remotely interesting alternative to concrete and shotcrete/ferrocement is papercrete, cast paper fixed by portland cement. Papercrete is a lot lighter than concrete and is much more balanced between tensile and compressive strength; while not very strong, a lot can be used cheaply and the resulting wall inherits some nice properties such as being an excellent thermal insulator. It does not really need reinforcement, since if you use enough of it to get the compressive strength you need and you bond it properly the cellulose fibers contribute a comparable tensile strength. A typical papercrete wall would be 30 to 50 cm thick, but would weigh less than a quarter what a similar looking adobe wall would weigh.
I invented two different forms of papercrete which were far superior structurally to anything that has ever been used to build a house. One system requires elaborate compression forms (but they can be made by amateurs with hand tools); the other requires a couple of extra ingredients beyond the paper, cement, and sand. Both are stronger, stiffer, and have better water rejection than anything in regular use by the people building real houses. The uncompressed formula can even be pumped like concrete (if you try this with normal papercrete the water squeezes out and you are left with a nonfluid brick of fiber).
Unfortunately, paper is mostly empty space and cellulose is madly attractive to water, and no matter what I did to my samples I could not get them to resist absorbing two or three times their dry weight in water when presented with an infinite supply, and it floods in this part of the country. Papercrete is a great material for the US desert southwest, where most experimental papercrete structures are built, but ironically it requires a lot of water and it takes a long time for the water used in the creation process to evaporate out of the bricks. (My techniques are both far superior on this score, but not good enough to use in Mississippi.) A papercrete structure which becomes saturated will mold before it can dry out and becomes uninhabitable. This has happened to quite a few of the real experimental structures which were built, even in drier climates.
There are a few other combinatorial techniques; there are two commercial adobe-like processes which use portland cement on one hand and plaster of paris on the other as binders, but they still have problems with environmental water and involve considerable logistical expense because of the low tensile strenght and mass. For practical reason most modern adobe and adobe-like structures also have conventional truss roofs, which reduces the advantage of using earth construction considerably. (By contrast a lot of papercrete houses are domes and barrel vaults, as this is much easier to do with the more lightweight and tensile material.)
Anyway, what I'm circling around to here is this: In order for 3D printing to become a practical method for building sized structures, the necessary innovation isn't the printer; the necessary innovation is the building material. And if there is a significant breakthrough there, it will affect all existing construction techniques in positive ways as well as possibly making 3D printing possible. It's unlikely that 3D printing will ever be competitive with conventional mass manufacture, though, for the same reason that 3D printed small parts will never be competitive with injection molded equivalents when there is enough volume demand to justify designing the mold.
@Gordon
Automated automobile manufacturing is not the same as automated housing. It is a smaller, more technically intensive product that is shipped globally.
Automated warehouses have the warehouse itself as the shell for the robot.
Housing robots have to be deployed to a building site and redeployed again and again and that mobility is costly. Also the exposure to weather and sun will take their toll.
Frankly, arguments against automation that claim labor is cheaper have all been doomed to failure. But we are now in a dire situation where the economics of automation and "supply side economics" are NOT working and the world is seeing a shift back to "demand side economics". People require employment to have disposable income to support a consumer economy. When you automate and downsize all the more costly jobs, people are left with a bare bones existence and cannot afford much beyond basic food, shelter, and clothing. Creative financing failed to keep these people able to purchase housing, so we had the recent economic crisis.
More automation and more cutting the cost of things by taking the human labor out of production is not going to serve to make a brighter future. Labor unions were likely right to resist automation as an economy is foremost about jobs and not just about creation of a high GDP that has little or no relationship to individual prosperity.
If you don't have buyers with cash, you find it difficult to sell anything.
3D printing is pretty much a hobbyist approach and a fad as it now stands. But when one looking at modern CNC machine shops and such, very few people are able to produce huge amounts of material. The same has happened in automated farming, in the automotive industry, in steel making, and so on. Job destruction across the board and just the suggestion that everyone retrain. There are no more typesetter jobs for book publication, printers (as people) are replaced by DIY copy centers.
@localroger
Rammed earth that has additives, such as a water-resistant binder (think liquified asphalt emulsion) has been demonstrated as a very good alternative to concrete in many venues. It certainly isn't as strong as concrete, but most of the world is happy to build one story structures with walls being dependent on mass and compression forces for stability.
Papercrete sounds interesting - as long as termites and fungal decay are not issues. There has been a lot of new concrete materials recently, including thin fiberglass and portland cement panels that are very durable.
What parts is Boeing printing? Critical structural elements or arm rests for seating? Maybe toilet seats?
The simple fact is this issue is foremost about Material Science. We have a vast corpus of knowledge about the strength of materials and the newest addition to high strength is carbon fiber. But the whole 3D approach seems to be divided into assembly by additive thermoplastic processes or by printing a liquid polymer in a tray and adding some sort of dust until you build up an item with polymer and the dust as filler.
I guess using a plaster or goo might evolve as a third process, but none have the tensile strength for floor and roof spans without building massive arches. It is a return to the era before wood and steel were engineered into longer spans of lighter material. Reinforced concrete spans are very sophisticated engineering and often done in a factory setting where conditions and specifications can be held to tight limits. You go into the field with a robot and quality is never going to be easy to achieve.
Nonetheless, 3D printing certainly has a place in the contruction of models for research and development. In aeronautics and in nautical engineering, to scale models can vastly save costs as testing a smaller model in a wind tunnel or a tank do provide good data about the designs performance. Nothing wrong with using 3D printing to get the model built rapidly and at a low cost.
What parts is Boeing printing? Critical structural elements or arm rests for seating?
The simple fact is this issue is foremost about Material Science. We have a vast corpus of knowledge about the strength of materials and the newest addition to high strength is carbon fiber. But the whole 3D approach seems to be divided into assembly by additive thermoplastic processes or by printing a liquid polymer in a tray and adding some sort of dust until you build up an item with polymer and the dust as filler.
I guess using a plaster or goo might evolve as a third process, but none have the tensile strength for floor and roof spans without building massive arches. It is a return to the era before wood and steel were engineered into longer spans of lighter material. Reinforced concrete spans are very sophisticated engineering and often done in a factory setting where conditions and specifications can be held to tight limits. You go into the field with a robot and quality is never going to be easy to achieve.
Nonetheless, 3D printing certainly has a place in the contruction of models for research and development. In aeronautics and in nautical engineering, to scale models can vastly save costs as testing a smaller model in a wind tunnel or a tank do provide good data about the designs performance. Nothing wrong with using 3D printing to get the model built rapidly and at a low cost.
I would like to clarify: what you see on TV and in Make Magazine is not all there is to 3D printing. In particular, companies like Boeing and Airbus are 3D printing critical parts out of metals.
Airbus wing brackets: the organic foreground part is 3d printed. In the background is the existing machined part. EADS Innovation Works
“At the EADS labs in Filton, U.K., researchers demonstrated that they can print out several different metal parts for airplanes with a technology that uses a laser to heat metal powders until they form solid metal shapes. Using this technique, EADS has printed metal hinges for engine covers: the hinges allow the covers to swing open for engine maintenance. The parts have intricate shapes that maintain strength while cutting the weight of the part in half. The new hinge has been put through the tests used for conventional parts and shown to meet performance requirements. ”
I have held in my hands 3d printed parts that Boeing has put onto 777 test aircraft. They were magnetic (hall effect) encoders used for the flaps. They save a bunch of money by being able to print out a durable complex plastic part. I believe it was made of nylon.
Unfortunately, ...could not get them to resist absorbing two or three times their dry weight in water .... A papercrete structure which becomes saturated will mold before it can dry out and becomes uninhabitable.
...In order for 3D printing to become a practical method for building sized structures, the necessary innovation isn't the printer; the necessary innovation is the building material.
So, if the papercrete could be dried such that it did not get moldy, and not reabsorb water, this would be a viable alternative? Hmmm.
Could this be done by applying heat? That is, the water baked out without destrying the paper? Or if the paper is destroyed, turn the rest to ceramic like when bricks are fired?
How about sealing the material after its dried to keep water out? Seems this should be solvable.
prof_braino, what I discovered is that papercrete cannot be sealed. I did a series of absorption and evaporation tests on prepared samples. You can slow the rate at which it absorbs water, sometimes considerably, but you cannot stop it from becoming completely saturated if it has enough time exposed to a water source.
Papercrete actually starts out saturated and must be allowed to evaporate dry before the bricks can be used. This is usually done by leaving them out in the sun for a few weeks (where hopefully it won't rain). Once assembled into a structure, should they become saturated again it is almost impossible to get them dried out before mold forms.
Papercrete derives all of its tensile and compressive strength from the cellulose fibers, and its stiffness from the portland cement particles locking the fibers together wherever they intersect. I did some burn tests which essentially removed the cellulose (a "2-bag mix" will not burn, and I had to sit the blocks on a charcoal bed). The end result was a very friable porous matrix of cement particles.
prof_braino, what I discovered is that papercrete cannot be sealed.
Can't even coat it with epoxy? Maybe some mono-molecular car paint? I'm not saying these are viable or cheap, I just want to know why nothing can work.
I was thinking like building a papercrete "smokehouse". Have a fie inside not hot enough to bun the papercrete, but hot enough to dry it. Then seal the outside. I read that straw bale houses last over a hundred years, because the plaster on the outside hermetically seals the straw, and after the initial oxygen is consumed its completely stable until torn down.
Seems like there's GOT to be a way to do this. Finding this is a first step; Making it simple and easy is a separate step.
Then again, maybe I should just make a bot that builds structures out of straw bales. Actually, that might not be a bad idea....
There is exactly one material which can be mixed into a flowable form and permanently chemically set into a solid, and which is cheap enough to use to print a structure the size of a house. That material is portland cement.
The Romans built hydraulic-setting concrete structures still standing that have no portland cement in them. The Chinese are big in building concretes that are likewise not portland based. Co-portland mixes (varying ratios) constitute some 20-30% of new construction in the UK and EU, and some buildings, roads, and bridges are made with as little as 10%or 20% portland mixes in them. These other cements are used much more heavily in the UK, EU, and Asia. Not in the US, so yes you're going to have trouble in finding a suitable alternative here.
Whether a non-portland mix is "cheaper" depends on where in the world it's milled. In the US most mills produce portland cement, because that's what the industry demands. Other types might be more expensive in the US because they are not currently produced in the same quantities, and unlike some countries we have little or no legal pressure to reduce the use of portand cement.
So, the rest of the world is aware of your difficulties in finding a good alternative to portland cement. Doesn't mean no one is working on the problem. Ceasar felt your pain!
Concrete and cement techologies in the USA seem to center around Purdue University being the leader. The trends have pretty much been driven by concretes with much greater compressive strengths for taller structures and the use of slip forms for reductions in labor and the need to provide complex forms.
All this has been most recognizable in bridge building as concrete protects the rebar from corrosion by seawater. Steel bridges have fallen out of favor.
Mega-structures appeal to engineer's ambitions and reduction of maintence is always an issue with new construction techniques - just because the replacement of old has its rather annoying maintenace history in the mix.
I am quite enthused by seeing metalurgical processes involved in 3D printing and even the fact that removing costly mold development from the thermoplastic production. These are really useful and needed developments.
I am not quite sure about the Portland Cement versus other alternatives. But when one begins to consider less compressive strength as being adequate due to structures being less ambitous in scale, all sorts of alternative open up.
Cellulose is problematic by its role in the environment. Termites and fungus love the stuffl and in pure form it is not very stable. I suppose you could 'chemically' treat or enclose it, but why bother with something like epoxy when you can better use the epoxy with glass or carbon fiber, which are more inert?
Steel pre and post tensioning have not been mentioned. But even rammed earth could beneift from the addition of steel to stabilized shocks to the building system.
I've yet to fly in Boeings all plastic Dreamliner -- maybe never will. But it certainly does seem that is where we are headed.
The Romans built hydraulic-setting concrete structures still standing that have no portland cement in them...
Well yeah, I was aware of non-Portland hydraulic cements, but as you observe it's almost impossible to get them in the US and Portland is so ubiquitous because its properties are so well understood, a wide range of well-behaved admixtures are available, and it's more waterproof than for example Roman mortar. I have also made papercrete stabilized by plaster of paris instead of PC, which is an interesting material.
The thing is, at the end of the day, all hydraulic cements are very similar to one another in ways that other self-setting materials aren't. All hydraulic cements take significant time to cure, have poor tensile strength, and create a porous result. There are significant differences; calcium-accelerated PC and Roman cement don't get nearly as high-PH basic as PC does, which is actually bad if you want to use reinforcing steel that can rust. Roman cement takes a *long* time to cure, which is why you have those mixes with a little PC to fix the structure in place while the alternate cement gets its act together.
The thing is, when you look at other induced-setting printing materials, they are very different and far more expensive than any hydraulic cement.
Comments
Governments, for disaster relief, per the article, for starters. Me, since I'll have the printer anyway.
The gantry from the picture is taler than a house, and the rrails are longer than a house, but there could eaily fit in the back of truck. And a truck would work fine, no?
".. at this point in time". This is only because that's where everybody is focused. Concrete pumper trucks are quite common around here, its only a matte of time before some braino decides use one with a printer (Ferb, I know what we're going to do today...)
maybe not
http://www.designboom.com/weblog/cat/8/view/17984/dirk-vander-kooij-endless-saloon-table-rocking-chair.html
We did have a village of 600 here in Taiwan that was destroyed by mudslide and they were given a whole new village with new housing, but not by the government. It was by the Tsu Chi Buddhist Disaster Relief Fund. The only stipulations were no one can live there if they smoke tobacco or drink alcohol. Some people actually refused due to these terms.
I am sorry, but the reality is that most houses are provided for paying home buyers. And housing - both new and the repair of old - are rather important engines of the economy. They provide jobs for a wide variety of craftspeople. This is why the USA is having a 'jobless recovery' - nobody can afford to buy housing.
These days everything is downsized - restaurants no longer have waiters and dishwashers, you serve yourself; banks no longer have tellers or even branch officers; you visit the ATM and have to deal with people in authority over the phone or internet; gas stations are self-service in many states; and on and on.
The simple reality is that everyone that doesn't have a job to depend on is depending on someone else to provide for them. The whole economic model of off-shore manufacturer and a consumer economy is turning rather grim.
Needing a crane and rails to put the 3D printer in place? I built homes in Oregon and the most we needed was a backhoe and a dump truck (unless you needed to drill a well). Generally, the materials are two or three semi-truck loads and the foundation concrete is 3-4 truck loads.
I realize this guy is an engineering professor in Southern California, but he sound more like an 'ivory tower' professor than a nuts and bolts engineer.
If anything could be done like this, I suspect an automated foundation excavation and installation would be optimal, not the whole building. But the simple fact is soil varies from mile to mile and each foundation take a bit of a different approach.
Once the foundation is in, construction is pretty standard. It is nice to have a small crane to lift the roof trusses into place, but they can be lifted by 3 guys if no crane is available. Or an all-terrain fork lift can do the job.
Yes, you can use concrete pump trucks, but don't expect the concrete to have adequate strength to retain shape immediately. It isn't cooling, it is setting.
As I mentioned above 'shot-crete' - a form of sprayed concrete plaster is about the only fast setting concrete available and it is rather nasty and messy to use. People have even sprayed domes out of it that are water tight (and boats). But it all starts with having the steel mesh in place to make the shape -- lots of steel.
Then there is the problem of when the printer goes haywire and prints ten copies when you only wanted six.
(Speaking of automated and semi-automated road construction, isn't it curious that Caterpillar -- who's business is building trucks and other machinery for making roads -- is contributing to the 3D home concept? After all, a graded road with concrete median and curb is but the same concept, just a little flatter...)
Where these homes are most needed, standard concrete is not the best choice of material (in the video he talks about the example wall being made of a composite material with about 3X the PSI rating of reinforced concrete). Water, not sand, is the more precious commodity.
You guys talking about the traditional methods of construction are still thinking in the Western or developed world mindset. Experienced laborers for safely building houses are hard to come by in most undeveloped countries. Like these places are full of skilled roofers!
-- Gordon
I seem to recall reading about experiments where either Fiberglass strands or carbon fibers were mixed in with the concrete when it was poured. Something like that may end up being an option for a large 3-D printer like this.
Given that you have a huge tract of land, you have a rail spur with a complete concrete batch plant, a fleet of concrete trucks, and another fleet of concrete pumps, and adequate supplies of rebar, windows, paving materials, and so on - you could use a 3D printing concept to build.
On the other hand, kits of pre-fabricated steel geodisc domes that can be assembled by local residents in areas requiring immediate disaster relief due to on-coming winter or typhoons would have immediate shelter that could be later relocated or enhanced where it stands.
Conceptually it is a neat idea. But the will to make it a reality is rather dubious.
And 20 hours is still completely absurd.
They already do something similar for emergency housing, domes or otherwise.
You mentioned trucking costs and transportation complexity for 3D houses, but you make it sound like all the steel for these domes just fly there on their own. And where are they pre-built to begin with? You won't find steel plants in most of the countries we're talking about, so that probably means shipping, which for the cubic feet of material talked about here is astronomically expensive. A non-starter. At least with earthen construction the raw materials are more or less nearby.
-- Gordon
Gordon, the limiting factor in shotcrete is portland cement. Portland cement is expensive and standard concrete is mixed to minimize its use, by carefully filling as much of the final space with carefully graduated aggregates. Most ready-mix has aggregates up to a couple of centimeters in size, and a typical ready-mix recipe is less than 1/10 portland cement by weight. For shotcrete and mixes that have to be pumped through nozzles like the one in the video, you can't use large aggregates. This means you have to use more portland cement, typically around 1/3 by weight, which makes the mix stronger whether you care about that or not. It also makes the mix more expensive no matter what other additives you use because portland cement is 10 times as expensive as aggregates (much less water to cure it). Additives are a significant additional expense for shotcrete and, I would assume, for the video's extrusion mixes.
Unfortunately, no matter what you do to it the tensile strength of unreinforced concrete remains Smile, and unless carefully built to leverage their mass in compression (like the buildings in the video from Iran) concrete structures need a lot of tensile strength simply to safely hold up their own roofs.
Steel too heavy for shipping, aluminum is half the weight of steel for the same volume. And concrete is actually a bit lighter than aluminum per cubic foot (about 5 pounds). But all of these are sourced from rich industrialized countries and there are even less aluminum industries on a global basis than there are steel.
Steel mills can be constructed as local mini-mills with local steel ore deposits and local coal. The quality may be a bit poor, and getting it actually rolled into structural shapes is another issue. But, it can be done if a country has the will and the capital. At one time, I helped write a proposal for Nigeria to do so and it was actually considered. But, the government was overthrown a few months later and the project dismissed.
I think you begin to see, that any realistic project will have source material for a robot from local material and not depend on a high energy cost material provided from an energy rich country. So items, such as roofing membranes may be worthwhile to import, but the floor and walls and roof trusses are best provided locally.
In sum, architecture has always had to rely on the local wealth to be able to select the best techonology for a project.
Has anyone ever seen a Volkswagen driven lumber mill. That was listed in the Whole Earth Catalog of the 1960s and can be moved from site to site to mill all aspects of a home's construction. Not much need for a robot though.
For example, while most species of wood test to 3,000 psi or so in compression, for safety reasons and to allow for flaws it's usually allowed 500 to 1000 psi in construction. This means a 2x4 stud in a load bearing wall which is constrained from buckling can support several thousand pounds.
Steel and wood cost about the same per pound, but steel is 10 times as heavy as wood (400 lb/cuft instead of 40 lb/cuft) and also about 10 times stronger (40,000 psi) This means the column of steel you need to replace a 2x4 stud should have a cross-sectional area of about 1/10 the cross sectional area of a 2x4, or about 1/2 square inch.
In practice there is no way to keep such a shape from buckling unless it is externally constrained. A wood house that has been framed but not skinned will support itself; cold-formed steel studs are very wobbly and unstable until the sheetrock or siding is put in place to stabilize them.
If you use hot-formed shapes like I beams a structure the size of a residential house becomes massively overbuilt and, therefore, hyper-expensive. It's similar to the problem with concrete for only slightly different reasons. Wood really hits a sweet spot between cost, strength, and stiffness which nothing else comes close to matching. And plastic materials, including anything bonded together with epoxy or other resins, are fantastically expensive compared to the materials normally used for construction.
Dirt, of course, is cheap but dirt hits you in the labor and material staging pocket. Because its tensile strength is so low you must use truly massive amounts of earth to build a practical wall that doesn't buckle, which means that even if the dirt is free the cost of making it into bricks and moving it around becomes prohibitive. This is why there are very few outfits that sell adobe bricks; shipping costs eat you alive if they have to be shipped more than a few miles.
Thinking out of the box is a good thing of course, but there are a lot of sound reasons why modern construction looks the way it does. I have my own issues with building codes requiring excess material use to pad the pockets of various industries, but a lot of the basics are no different than they were 300 years ago.
The bottom line is that no matter what material you use, if it has nearly perfect properties you will use a minimum of ten thousand pounds or so of it to safely enclose a typical residential house-shaped space against typical weather, animals, and burglars. Then, if your material is at all imperfect in any way, you start adding to that. A modest adobe house can weigh a substantial fraction of a million pounds (which also gets you into foundation issues). And the list of materials cheap enough to deploy in such quantities at all is rather short. There are no plastics on it.
Check out
http://calearth.org/
As I pointed out in the very previous post, a major drawback to adobe is labor and transportation costs. Now if you have an army of unpaid and available labor sitting around with nothing else to do adobe can make a lot of sense. But in every accounting I've ever seen a first-world contract built adobe house is significantly more expensive than a conventional frame house, and it's because of the labor and transportation. People who aren't doing communal sweat equity don't build adobe unless they want it for the thermal or other architectural properties.
And BTW if you ever bothered to look at what constituted housing in these regions or south of the border, mud bricks made up a lot of it, made from on-site local material. The same in pre - WWII Eastern European country side.
Of course these people didn't have the luxury of modern Western accounting methods so they have some housing instead of none. They didn't have to pay some trucker $250 to haul a single load 2 miles to the work site and people worked together(try getting a bunch of Americans to work together in a communal setting and they'll sue each other). They didn't have to hire so-called professionals(that charge $200 just to walk in the door and charge $2000 and take 3 days to dig a small ditch) so costs for them were and are minimal.
The other thing is that these people don't need the massive living space that Americans need today to get by so the dynamics change again.
Bottom line: Home building in the 3rd world isn't rocket science and a lot of practical people have given it serious thought and worked out alternatives that take into account local resources and skill sets.
Robots created by ivory tower egg heads aren't needed.
As a contractor myself, I'd like to clarify: you're paying a reasonable rate based on market demand and contractor overhead. Have you looked to see how much you cost whatever company you work for? It's not just salary, it's benefits: healthcare, accounting, legal protection, physical resources, discounts at places, employee education opportunities, and job security. A contractor has to cover all of that themselves, and so have to charge more than the "standard employee $/hr".
And anyway, even if in 3rd world countries "they didn't have to pay" and "people worked together", that doesn't reduce the cost: the homeowner must reciprocate and give up his own time to help his neighbors (==cost). It's merely a different form of exchange.
Americans do work together on community projects: http://en.wikipedia.org/wiki/Open_source
But at the core of all this is a very real desire to own your own paid for home as security.
When you import a robot and technically complex materials to do the task faster and 'better', you are spending your money elsewhere. It goes to a corporation that hopes to achieve substantial profits by diverting money that would have been spent locally. Where they divert it to is rather awkward - maybe for good purposes, maybe for the bad.
In most instances, where corporations and governments can help is not particularly in providing housing technologies that are more efficent, but in terms of providing safe and sane locations for housing. And also providing education in how to DIY shelter that is safe and useful over the long-haul.
The issues here are far more complex than one neat techno=solution. The realities are that we are NOT building on the Moon or Mars at this point. (And in both places, tunneling might be the first phase of habitat construction). And we are not providing anywhere near enough housing for disaster relief in this disaster prone world. In the other marketplaces for housing, people are pretty much accepting the product available as being an improvement over past efforts and progressing towards more comfort, even if they might be duped (Modern wood homes in America are just not as durable or repairable as the homes built before WWII).
So, this requires a multiple solution approach and some real motivation to provide better housing with better infrastructure to those that need it. There is a corruption that favors providing housing for those that will pay enough for middlemen to make a profit. All too often, the people that are willing to DIY a home are not empowered by local construction codes and zoning. And all too often we see urban renewal removing a so-called blighted neighborhood by removing the people that cannot afford better housing and providing a new environment for more affluent people. This seems to have happened again with the London Olympics redevelopment of London's East End.
In other word, there are a lot of 'grand schemes' that tend to favor the promoters, not the person in need of decent shelter.
History has proven those naysayers wrong. The Japanese introduced robotics for quality consistency, and along the way developed manufacturing techniques that better meshed with the needs of robotic assembly. These days you don't even want to buy a mass produced car that's been hand-made. Too much chance for human error. The cost reductions due to robotic manufacturing, while now considerable because it's become it's own science, is secondary.
Amazon recently purchased Kiva Systems for three-quarter of a billion dollars so they could be on the forefront of automated warehousing. More than 80% of all warehouses in the country are not suitable for a single robot, let alone an army of them. Does this mean that because there are technical challenges to overcome that robots have no place in warehouse automation? If you think so, better write a letter to Jeff Bezos and tell him he blew $750+ million on a harebrained idea.
Yes, building a whole house automatically is a far-fetched idea. Who says it wasn't. It's a "grand challenge." It's made to be difficult, and far-reaching. Like other grand challenges, it's also made to act as a springboard for smaller ideas along the way. You don't have to be a genius to think of the simpler alternative uses for this. It's pretty clear Caterpillar isn't interested in getting into the home building market. So it's quite possible they view the same machine for making roads, curbs, pilings, road dividers, just about anything you can think of, in any dimensions, any shapes, using many types of flowable and air-cured materials.
I'll remind folks that desktop 3D printed objects are basically useless in comparison to injection molded plastics. A 3D part is much more expensive on a materials basis, and far weaker. I don't see these limitations stopping people from trying new things and pushing the end of the envelope.
-- Gordon
There is exactly one material which can be mixed into a flowable form and permanently chemically set into a solid, and which is cheap enough to use to print a structure the size of a house. That material is portland cement. There is nothing else. I spent the better part of a year investigating alternatives, and no matter how frugal you are with aggregates and meshes and eggshell arrays every single alternative either has even more terrible mechanical properties or is a hundred times as expensive.
Even portland cement isn't so simple; if you want to use it as shotcrete or ferrocement you must use more because you can't use large aggregate, and it's likely that the admixtures (at a minimum you need accelerant, air entrainer, and plasticizer), while a relatively small physical part of the final mix, will cost nearly as much as the cement.
The only remotely interesting alternative to concrete and shotcrete/ferrocement is papercrete, cast paper fixed by portland cement. Papercrete is a lot lighter than concrete and is much more balanced between tensile and compressive strength; while not very strong, a lot can be used cheaply and the resulting wall inherits some nice properties such as being an excellent thermal insulator. It does not really need reinforcement, since if you use enough of it to get the compressive strength you need and you bond it properly the cellulose fibers contribute a comparable tensile strength. A typical papercrete wall would be 30 to 50 cm thick, but would weigh less than a quarter what a similar looking adobe wall would weigh.
I invented two different forms of papercrete which were far superior structurally to anything that has ever been used to build a house. One system requires elaborate compression forms (but they can be made by amateurs with hand tools); the other requires a couple of extra ingredients beyond the paper, cement, and sand. Both are stronger, stiffer, and have better water rejection than anything in regular use by the people building real houses. The uncompressed formula can even be pumped like concrete (if you try this with normal papercrete the water squeezes out and you are left with a nonfluid brick of fiber).
Unfortunately, paper is mostly empty space and cellulose is madly attractive to water, and no matter what I did to my samples I could not get them to resist absorbing two or three times their dry weight in water when presented with an infinite supply, and it floods in this part of the country. Papercrete is a great material for the US desert southwest, where most experimental papercrete structures are built, but ironically it requires a lot of water and it takes a long time for the water used in the creation process to evaporate out of the bricks. (My techniques are both far superior on this score, but not good enough to use in Mississippi.) A papercrete structure which becomes saturated will mold before it can dry out and becomes uninhabitable. This has happened to quite a few of the real experimental structures which were built, even in drier climates.
There are a few other combinatorial techniques; there are two commercial adobe-like processes which use portland cement on one hand and plaster of paris on the other as binders, but they still have problems with environmental water and involve considerable logistical expense because of the low tensile strenght and mass. For practical reason most modern adobe and adobe-like structures also have conventional truss roofs, which reduces the advantage of using earth construction considerably. (By contrast a lot of papercrete houses are domes and barrel vaults, as this is much easier to do with the more lightweight and tensile material.)
Anyway, what I'm circling around to here is this: In order for 3D printing to become a practical method for building sized structures, the necessary innovation isn't the printer; the necessary innovation is the building material. And if there is a significant breakthrough there, it will affect all existing construction techniques in positive ways as well as possibly making 3D printing possible. It's unlikely that 3D printing will ever be competitive with conventional mass manufacture, though, for the same reason that 3D printed small parts will never be competitive with injection molded equivalents when there is enough volume demand to justify designing the mold.
Automated automobile manufacturing is not the same as automated housing. It is a smaller, more technically intensive product that is shipped globally.
Automated warehouses have the warehouse itself as the shell for the robot.
Housing robots have to be deployed to a building site and redeployed again and again and that mobility is costly. Also the exposure to weather and sun will take their toll.
Frankly, arguments against automation that claim labor is cheaper have all been doomed to failure. But we are now in a dire situation where the economics of automation and "supply side economics" are NOT working and the world is seeing a shift back to "demand side economics". People require employment to have disposable income to support a consumer economy. When you automate and downsize all the more costly jobs, people are left with a bare bones existence and cannot afford much beyond basic food, shelter, and clothing. Creative financing failed to keep these people able to purchase housing, so we had the recent economic crisis.
More automation and more cutting the cost of things by taking the human labor out of production is not going to serve to make a brighter future. Labor unions were likely right to resist automation as an economy is foremost about jobs and not just about creation of a high GDP that has little or no relationship to individual prosperity.
If you don't have buyers with cash, you find it difficult to sell anything.
3D printing is pretty much a hobbyist approach and a fad as it now stands. But when one looking at modern CNC machine shops and such, very few people are able to produce huge amounts of material. The same has happened in automated farming, in the automotive industry, in steel making, and so on. Job destruction across the board and just the suggestion that everyone retrain. There are no more typesetter jobs for book publication, printers (as people) are replaced by DIY copy centers.
@localroger
Rammed earth that has additives, such as a water-resistant binder (think liquified asphalt emulsion) has been demonstrated as a very good alternative to concrete in many venues. It certainly isn't as strong as concrete, but most of the world is happy to build one story structures with walls being dependent on mass and compression forces for stability.
Papercrete sounds interesting - as long as termites and fungal decay are not issues. There has been a lot of new concrete materials recently, including thin fiberglass and portland cement panels that are very durable.
I will participate in this war of facts:
http://www.businessweek.com/technology/3d-printing-coming-to-the-manufacturing-spaceand-outer-space-01092012.html
As for quantity of 3d printing machines, hobbyist machines will soon overtake commercial if they haven't already.
The simple fact is this issue is foremost about Material Science. We have a vast corpus of knowledge about the strength of materials and the newest addition to high strength is carbon fiber. But the whole 3D approach seems to be divided into assembly by additive thermoplastic processes or by printing a liquid polymer in a tray and adding some sort of dust until you build up an item with polymer and the dust as filler.
I guess using a plaster or goo might evolve as a third process, but none have the tensile strength for floor and roof spans without building massive arches. It is a return to the era before wood and steel were engineered into longer spans of lighter material. Reinforced concrete spans are very sophisticated engineering and often done in a factory setting where conditions and specifications can be held to tight limits. You go into the field with a robot and quality is never going to be easy to achieve.
Nonetheless, 3D printing certainly has a place in the contruction of models for research and development. In aeronautics and in nautical engineering, to scale models can vastly save costs as testing a smaller model in a wind tunnel or a tank do provide good data about the designs performance. Nothing wrong with using 3D printing to get the model built rapidly and at a low cost.
I would like to clarify: what you see on TV and in Make Magazine is not all there is to 3D printing. In particular, companies like Boeing and Airbus are 3D printing critical parts out of metals.
http://blog.ponoko.com/2011/05/14/aerospace-industry-adopting-3d-print-technology/
Apparently, Boeing is making 3D printed turbine engine parts. I can't find the source for that, though.
http://www.technologyreview.com/featured-story/426391/layer-by-layer/
So, if the papercrete could be dried such that it did not get moldy, and not reabsorb water, this would be a viable alternative? Hmmm.
Could this be done by applying heat? That is, the water baked out without destrying the paper? Or if the paper is destroyed, turn the rest to ceramic like when bricks are fired?
How about sealing the material after its dried to keep water out? Seems this should be solvable.
I really want to print my next house.
Papercrete actually starts out saturated and must be allowed to evaporate dry before the bricks can be used. This is usually done by leaving them out in the sun for a few weeks (where hopefully it won't rain). Once assembled into a structure, should they become saturated again it is almost impossible to get them dried out before mold forms.
Papercrete derives all of its tensile and compressive strength from the cellulose fibers, and its stiffness from the portland cement particles locking the fibers together wherever they intersect. I did some burn tests which essentially removed the cellulose (a "2-bag mix" will not burn, and I had to sit the blocks on a charcoal bed). The end result was a very friable porous matrix of cement particles.
Can't even coat it with epoxy? Maybe some mono-molecular car paint? I'm not saying these are viable or cheap, I just want to know why nothing can work.
I was thinking like building a papercrete "smokehouse". Have a fie inside not hot enough to bun the papercrete, but hot enough to dry it. Then seal the outside. I read that straw bale houses last over a hundred years, because the plaster on the outside hermetically seals the straw, and after the initial oxygen is consumed its completely stable until torn down.
Seems like there's GOT to be a way to do this. Finding this is a first step; Making it simple and easy is a separate step.
Then again, maybe I should just make a bot that builds structures out of straw bales. Actually, that might not be a bad idea....
The Big Bad Wolf seconds the idea.
The Romans built hydraulic-setting concrete structures still standing that have no portland cement in them. The Chinese are big in building concretes that are likewise not portland based. Co-portland mixes (varying ratios) constitute some 20-30% of new construction in the UK and EU, and some buildings, roads, and bridges are made with as little as 10%or 20% portland mixes in them. These other cements are used much more heavily in the UK, EU, and Asia. Not in the US, so yes you're going to have trouble in finding a suitable alternative here.
Whether a non-portland mix is "cheaper" depends on where in the world it's milled. In the US most mills produce portland cement, because that's what the industry demands. Other types might be more expensive in the US because they are not currently produced in the same quantities, and unlike some countries we have little or no legal pressure to reduce the use of portand cement.
So, the rest of the world is aware of your difficulties in finding a good alternative to portland cement. Doesn't mean no one is working on the problem. Ceasar felt your pain!
-- Gordon
All this has been most recognizable in bridge building as concrete protects the rebar from corrosion by seawater. Steel bridges have fallen out of favor.
Mega-structures appeal to engineer's ambitions and reduction of maintence is always an issue with new construction techniques - just because the replacement of old has its rather annoying maintenace history in the mix.
I am quite enthused by seeing metalurgical processes involved in 3D printing and even the fact that removing costly mold development from the thermoplastic production. These are really useful and needed developments.
I am not quite sure about the Portland Cement versus other alternatives. But when one begins to consider less compressive strength as being adequate due to structures being less ambitous in scale, all sorts of alternative open up.
Cellulose is problematic by its role in the environment. Termites and fungus love the stuffl and in pure form it is not very stable. I suppose you could 'chemically' treat or enclose it, but why bother with something like epoxy when you can better use the epoxy with glass or carbon fiber, which are more inert?
Steel pre and post tensioning have not been mentioned. But even rammed earth could beneift from the addition of steel to stabilized shocks to the building system.
I've yet to fly in Boeings all plastic Dreamliner -- maybe never will. But it certainly does seem that is where we are headed.
Well yeah, I was aware of non-Portland hydraulic cements, but as you observe it's almost impossible to get them in the US and Portland is so ubiquitous because its properties are so well understood, a wide range of well-behaved admixtures are available, and it's more waterproof than for example Roman mortar. I have also made papercrete stabilized by plaster of paris instead of PC, which is an interesting material.
The thing is, at the end of the day, all hydraulic cements are very similar to one another in ways that other self-setting materials aren't. All hydraulic cements take significant time to cure, have poor tensile strength, and create a porous result. There are significant differences; calcium-accelerated PC and Roman cement don't get nearly as high-PH basic as PC does, which is actually bad if you want to use reinforcing steel that can rust. Roman cement takes a *long* time to cure, which is why you have those mixes with a little PC to fix the structure in place while the alternate cement gets its act together.
The thing is, when you look at other induced-setting printing materials, they are very different and far more expensive than any hydraulic cement.