Hacker News new | past | comments | ask | show | jobs | submit login
How to design a house to last 1000 years (constructionphysics.substack.com)
594 points by ddubski 4 months ago | hide | past | favorite | 391 comments

This is a very odd design document and it makes me think this author has thought a lot about materials and buildings but not actually built anything.

The steel moment frame, to someone with shallow knowledge, sounds so strong and resilient. But in fact, a rigid steel structure is more vulnerable to seismic (and even wind) loads than wooden framing which can flex and move and dampen those loads naturally.

The stainless spec for the frame is just pure silliness. Looking at my notes now, for steel beams buried in the ground:

  200 microns of rust per year in very aggressive soils, but it rusts on both sides, so make that 400 microns.
... which means that it takes ~25 years to rust through naked 3/8 steel buried in the most aggressive soils.

... which also means that unburied steel, protected from elements, up in the air, is going to last more than 1000 years.

Oh, and also, the SS is more brittle so you've made your seismic issues even worse.


If I had an unlimited budget and was aiming for >1000 years I would pour the piles to bedrock with stainless rebar inside fly-ash concrete and top those pilings with plate connectors into which you could socket large wooden columns (perhaps 8x8) and build the structure with large wooden members connected with steel connectors and column caps, etc.

I would only use steel members if the span called for wood that was too big (like a 24' span needing a 8x14 or whatever).

"rigid steel structure"

Steel is springy. Tall steel-framed buildings and bridges routinely sway in wind, which is usually harmless to the structure but annoying to occupants. Unless you get harmonic oscillation, where the energy stored in the motion builds up, which can be a problem and has destroyed bridges. Much of seismic design involves connections which raise the resonant frequency of the structure so it can't oscillate at a low frequency with high amplitude. That's what those triangular reinforcement beams one sees in San Francisco really do. It's also what all those rectangular trusses under the Golden Gate Bridge do. Those were a retrofit.

Wood's flexibility usually causes problems at joints. Nailed joints are not very strong in tension. Most construction today in areas with earthquakes or high winds involves metal reinforcement of joints. There's a collection of galvanized sheet metal parts for that at any Home Depot.

Tension joints for wood are seen in classic Japanese construction, in boats, and in cabinetry. Not so much in modern houses, partly because they work better in hardwood. I wonder if, in the next installment, the author will discuss those.

I would agree that connections should be steel - I like the thick gauge, heavy duty simpson connectors, plates and column caps, etc., which allow you to lock in beams and columns with 3/4 machine bolts, etc.:


> That's what those triangular reinforcement beams one sees in San Francisco really do.

Interestingly enough that’s not quite right. They were added in response to the Tacoma Narrows collapse which was not, as is popularly misstated, destroyed by harmonic resonance but by aerostatic flutter.

Certainly resonance due to cars or pedestrians can damage a bridge, but that’s a separate issue.

Right. But you have to have a structure capable of long-period oscillation to be vulnerable to that particular problem. That's an inherent problem for long span bridges, but a building has to have a big unsupported span to be vulnerable to that problem. Sports stadium scale, though... All that potential lift.

Rigidity is actually a variable. rigid in the simplest of definition does mean no springy action. Rigid through various fixed connectors implies some spring like behaviors.

This is my favorite topic and I could go on and on about this. Point: it is a matter of precise definition. How rigid are we talking?

I'm fascinated by structural engineering as a sort of hobbyist (who has done some basic home construction/renovation and plans to build a house/houses one day) and would definitely not be opposed if you went on and on about it.

But everything is a spring.

Literally everything. Harmonic oscillators all the way down.

I'm Hooked. Tell me more.

There’s a mass of theory out there. For some, that puts a spring in their step but others shouldn’t let their spirits be dampened.

Check out de Broglie wavelength. https://en.m.wikipedia.org/wiki/Matter_wave

Spring Theory

In January

> Tension joints for wood are seen in classic Japanese construction

Japan also has several wooden buildings that purport to be 1,000+ years old—although these situations inevitably lead to Ship of Theseus debates.

"... And then it burnt down" was probably the funniest thing I heard touring the "thousand" year old castles in Japan. Being completely levelled by a fire seemed to happen about once every 200 years.

Japan basically rebuilds everything on a 20-30 year schedule. The temples are not an exception.

> "Steel is springy. Tall steel-framed buildings and bridges routinely sway in wind, which is usually harmless to the structure but annoying to occupants."

exactly, i recently mentioned my swaying-in-an-earthquake story[0], which was in a class A (steel+concrete) highrise office building. driving steel into the ground doesn't automatically mean it will rot and/or break in the first earthquake/windstorm that hits, even if that's a general possibility, given that engineers do think about that stuff when designing buildings. the gp comment is classic bullshit, plausible sounding but unconcerned with truth.

[0]: https://news.ycombinator.com/item?id=29665129

You misunderstood my comment about steel in the ground ... I was trying to convince the OP that they don't need to worry about the corrosion since even in terrible circumstances the steel still lasts quite a while.

With regard to your skyscraper experience:

I'm not sure this is an apples-to-apples comparison.

Skyscrapers are not made of skyscraper-height columns - they are a stack of elements that are connected every X height that has a well known flex per connection. It's also (hopefully) a uniform flex at every connection.

But a smaller building would, indeed, have unbroken steel members (like a column) and you might "successfully" connect them to one another with an incredible amount of rigidity.

It will either be tremendously strong throughout (good for you) or there will be some tiny piece of the chain that isn't as strong and can fail.

I would be confident attempting this on a very small building.

I would be hesitant to attempt this on a medium, two-story building. I would want wood framing.

For aesthetic reasons, I would want that wood framing to be big timbers. I'd rather spend my money on those than on stainless steel roof framing :)

they are a stack of elements that are connected every X height that has a well known flex per connection. It's also (hopefully) a uniform flex at every connection.

The flex is supposed to be in the beams, not the connections. Stress concentration is bad. Here's an intro.[1] Beams are easy to analyze, and tend to meet their specs, while connections are hard to analyze, and are subject to construction mistakes.

The January 1994 Northridge CA earthquake caused damage at beam-to-column connections in steel moment resisting frames. That got a lot of attention. Few buildings collapsed, but a lot of joints needed to be fixed or reinforced.[2] Welded flanges with bolts turned out to be weaker than expected.

Now, there's a style of construction where all the joints are rotational. That's seen in older truss bridges.[3] In classic designs, all components are in pure compression or pure tension. This shows in the construction; the tension components are flat plates or cables, and the joints are big steel pins. Those are easy to analyze, and if you take a statics class, that's a homework assignment. Popular for railroad bridges.

But building skeletons aren't usually built that way. They usually have rigid connections. You do see some buildings with lots of diagonals and pin joints. Long span roof trusses, which are a lot like bridges, are often built that way. Look at buildings with large atriums and you'll often see pin joints.

[1] https://www.thestructuralmadness.com/2014/04/possible-types-...

[2] https://nvlpubs.nist.gov/nistpubs/Legacy/IR/nistir5625.pdf

[3] https://en.wikipedia.org/wiki/Truss_bridge

Same with long low steel buildings, they usually have a lot of room for expansion built into them because the tops are close enough to the foundation that the angle of the uprights would change a lot if not dealt with in such a way. For higher buildings this isn't as much of a problem because the parts closer to the foundation will likely be closer to ambient.

In one particular case (the steel truss roof of a convention center in Amsterdam) there are many such places where slack was built into it, the structure is many 100's of meters wide and if connected rigidly would put undue stress on the uprights and the foundations.

Rigidity is good for some things and bad for others. In seismic design, inertial forces tend to decrease as structures become more flexible (Good!). But the consequence is that things move more (Bad!) meaning that all the nonstructural things get damaged - drywall, chimney, ceilings, etc. If things move too much, they also are subject to various types of degradation - yielding, fatigue, etc.

The other structural aspect not mentioned in the above discussion is strength. You can trivially get an order of magnitude more strength than required by even the harshest of loads using steel in a small structure like this. The same cannot be said about wood. For instance, a single 3/4" A325 bolt will be able to resist about 40,000lb shear or 70,000lb tension. The entire base shear of this size of structure in a code-design earthquake would be somewhere around 4,000lb.

[clarification: I mixed up my unit conversion. A 3/4" A325 bolt resists ~25kip shear or 31kip tension. One of the joys of working in an industry that primarily uses customary units, while codes are written in SI units]

gotcha, my bad for misconstruing your point. joint strength relative to span strength is definitely a non-obvious issue to the average home owner-builder.

i also vastly prefer wood/mass timber for aesthetic reasons. mass timber has better burn characteristics than steel, and i'd recently read that builders are actually starting to surround steel columns with cross-laminated timber (rather than concrete) for that reason[0], while providing greater strength/flexibility and better aesthetics. that's probably what i'd want if money were no object.

[0]: mentioned in this article, but i'd read more about it elsewhere: https://www.vox.com/energy-and-environment/2020/1/15/2105805...

p.s. - i've also daydreamed about building warehouse style: a separate steel superstructure for the roof integrating solar panels and solar heating, with a simple stick-built house underneath.

"... and i'd recently read that builders are actually starting to surround steel columns with cross-laminated timber ..."

Somewhat relevant - might interest you:


TAMedia office building in Zurich.

> ...mass timber...

Not sure if this is a regional / language difference, but I've more commonly seen it called "engineered wood" [0]. Which apparently now also includes transparent(?!) wood composite [1].

[0] https://en.wikipedia.org/wiki/Engineered_wood

[1] https://en.wikipedia.org/wiki/Transparent_wood_composite

my impression is that 'engineered wood' is a slightly different aggregation than 'mass timber', with engineered wood including things like plywood, while mass timber is focused on the 'timber' aspect (more structurally focused, usually larger).

> Tension joints for wood are seen in classic Japanese construction

If you’re referring to the wood bracket system, that is true, but for posterity, it didn’t originate in Japan. Imported, very likely via Korea (which also uses it in their classic construction), from China.

Steel is springy.

Unless it's not. Steel isn't one homogeneous thing. Depending on what trace elements you add, steel can have wildly different values of strength, rigidity & hardness.

> If I had an unlimited budget and was aiming for >1000 years I would pour the piles to bedrock with stainless rebar inside fly-ash concrete and top those pilings with plate connectors into which you could socket large wooden columns (perhaps 8x8) and build the structure with large wooden members connected with steel connectors and column caps, etc.

Are there any existing buildings constructed that way that have stood for >1000 years?

If not, my approach would be to copy an existing building that has stood for >1000 years in a region that has had several of the same kind of natural disasters that happen at the place I'm going to be building.

York Minster in England is mostly 1000 years old. There are some nice stories about how it handled a serious fire in 1984 here, (starting about 18 minutes in) https://www.bbc.co.uk/sounds/play/m0007pws. eg, the 2000 year old Roman drainage system got used for the first time in 1000 years because lots of water got inside the building from the fire hoses.

When rebuilding it, they had to decide whether to remake it in the same way it was originally, or to use a modern approach. They decided to use oak beams again because of lack of evidence about what happens to steel structures after hundreds of years. But then they couldn't find any oak trees big enough.

There is a story about one of the Ivy League universities. IIRC, in one of the century-or-two old buildings, there was a need to replace some main interior beams in a roof. Among the staff, someone knew someone who remembered something about this. Turns out, there was a special grove of trees, still under control of the university, which had been planted to provide replacement beams -- after 150 years of growth.

An old story about people having thought for the long term.

I've heard that as a story about Westminster, which makes me suspect the whole thing is an urban legend...

When planning to rebuild Notre Dame de Paris, there was a proposal to use titanium beams instead of the flammable oak ones that caused the deflagration in the same place, but they opted to go the original route, which involved finding old oak trees across France to harvest.

I would certainly use wood, and then just design it like the Japanese do for their old buildings, which are continuously repaired and after 1000 years you would still have the same home but it would be a Ship of Theseus type of situation. Wood has been used for building for thousands of years, and I wood imagine that humanity will forever be using wood for building, so there is a small chance you would lose the knowledge of how to build with it.

Unfortunately German only: https://www.rheingau.de/sehenswertes/sehenswuerdigkeiten/gra...

The oldest stone house of germany, ca. 1k years old. The wood is dated to 1035..1075 ad.

But Im sure that house is pretty young for Italian or Egyptian standards…

Could be prone to some survivorship bias.

Sure, but what's the alternative?

Dig out all other buildings that failed over the last 1000 years, figure out how they failed and why, and take action?

You can just copy what is known to work instead.

Or like, look at any of the million dilapidated houses and realize that 1. most houses leak at some point and 2. termites eat wood. Then conclude that actually, maybe .0001% of wood houses lasting for 1000 years is not actually good evidence that wood is a material that easily lasts 1000 years. Then think twice about using wood.

Personally I'd a concrete dome: https://en.wikipedia.org/wiki/Pantheon,_Rome

The claim here is that one would look at _ALL_ 1000+ year ld buildings, and then decide what to build.

You are building the strawman of looking at _ONE_ 1000+ year old building, and then building that because it must work. Nobody is arguing about this, and probably everybody here would agree that doing this is a bad idea.


With that out of the way, there aren't many 1000+ year old buildings made out of wood. There are some, e.g., japanese temples, that IIRC have burned a couple of times along the way, but are 1000+ year old and build out of wood.

But most buildings in the 1000+ year old range are made out of stone and only partially out of wood (cathedrals, etc.).

Concrete with basalt rebar. It doesn't rust like steel.

Steel beams buried in the ground, no matter what they're made of, are going to rot away quite rapidly if they contact any other metal, due to galvanic corrosion. I think you'd be better off with basalt fiber reinforced concrete for anything going in the ground. It's going to be strong, and it won't rot.

There is also basalt rebar. Last I checked, some building codes allow equivalent 1:1 tradeoff with classic rebar - even though the basalt rebar is stronger.

From what I understand, basalt rebar does not have the usual problem of regular rebar where oxidizing (rust) expansion can cause cracking in concrete. Additionally, the temperature expansion rates are much closer (rebar vs concrete).

My main hesitation would be that it is relatively new so we don't have that much data on how well it ages. Overall, it seems better than rebar classic by far (other than cost - which hasn't been scaled) - but I don't build structures for a living.

> Last I checked, some building codes allow equivalent 1:1 tradeoff with classic rebar - even though the basalt rebar is stronger.

Basalt rebar has a much lower elastic modulus, so when it fails, it fails. Steel on the other hand has a much larger elastic region - where the steel stretches while still carrying the same load. Part of the benefit of reinforced concrete is elastic failure - essentially every (properly) designed reinforced concrete structure is designed to fail with the tensile side of concrete cracking while the steel rebar still caries load, giving visual evidence that the concrete has failed. This is called elastic failure. This works because steel has a large ductile range. Basalt rebar doesn't fail the same way, so substituting it 1:1 seems inappropriate, regardless of it being stronger (per unit area).

I did find one random study[1] that substituted basalt rebar 1:1 with steel and compared the failure modes, and found that basalt-reinforced had cracking load essentially the same as a steel-reinforced, but a much larger ultimate load. Interestingly, the higher ultimate strength more than makes up for the much lower elastic modulus.

[1] https://www.icontrolpollution.com/articles/investigation-of-...

Here is a testimonial about a modern cave home: https://dengarden.com/misc/The-Pitfalls-of-an-Underground-Ho...

The dugouts in Coober Pedy are a modern example of this. I'll bet they're still there in 1000 years, albeit with damage to the exterior facades.

With .5" of monthly rainfall and nearly no seismic or insect activity to speak of, you could make any house in Coober Pedy last nearly forever.

I would like to know more reasoning behind choosing a frame without diagonals. I have had to repair many old garages and shacks which had wooden frame but lacked diagonals. Often we had to push them upright and brace the frame with diagonals to avoid problems in the future.

A cube-like building frame without diagonals or shear walls is unstable and does not sound so strong or resilient at all.

in a typical wood-frame wall, it's (usually) plywood nailed to the studs that acts to resist shear forces. you shouldn't need diagonal cross-members, unless there's not enough structural plywood cladding or not enough studding.

>makes me think this author has thought a lot about materials and buildings but not actually built anything

Here's an interview with the author: https://on.substack.com/p/what-to-read-construction-physics and his LinkedIn page: https://www.linkedin.com/in/brian-potter-6a082150

Anything's possible, of course, but he does seem to have the right credentials and experience for the subject matter.

> If I had an unlimited budget

at some point on the price scale you can just build a titanium cast for a house and pour molten rocks in, probably a basalt. what's the yield strength of igneous rocks? maybe the roof need to be arched.

So, a cave in a basalt mountain then.

I like your thinking. ;)

What are the load-baring capabilities of a Basalt Cave during the Apocalypse?

/MontyPython style...

Is that a European basalt cave, or African?

Titanium is quite expensive though; a house made from titanium likely would be broken up and stolen for the scrap value.

Titanium would be the mould of the cast, you remove it once the basalt set

An artist friend of mine back in the `90's made a series of sculptures, which they intentionally created to last for hundreds of years, out of plywood and bondo since it has no scrap resale value :) Each sculpture was also fitted to it's own storage crate, so that it could just be packed up and forgotten.

I was thinking about this when the lava firehose was going in Hawaii. I imagined a big metal shell that you could rotate under it to make a bunch of giant house sized domes really fast.

Is that something that we've actually done at any kind of scale?

Many cold war bunkers are kind of like this, but using rock that's already in-situ: start with solid rock (either in a mountain or in bedrock), remove some of it, in the cavity add linings of concrete, copper and steel to enhance properties as desired (copper for better EM shielding).

Not the same, but the Pyramids of Giza are essentially huge rock houses. I don't think they provide very modern amenities though :)

Well, we've done it with concrete tents [1].

It seems plausable you could begin with that kind of shelter and use that as a basis for further reinforcement.

[1] https://www.concretecanvas.com/cc-shelters/

On a global scale, probably, yes... but I'm either 1 or 0 correct.

Maybe tungsten molds instead?

yes, sorry, got the terms switched around apparently

>> If I had an unlimited budget and was aiming for >1000 years I would pour the piles to bedrock with stainless rebar...

Yeah they reject reinforced concrete, but the reason it doesn't last is the type of rebar used. To then site stainless later seems odd. IMHO our roads need to be built with stainless rebar.

Agreed on the design.

I worked in a post and beam barn built in the 1600s. It will be there in 2200. Basically, if you need the roof maintained and those buildings will last forever.

Personally, I’d do a stone foundation with post and beam. Over 1000 years, luck and location mean more anyway. Chances are the building will be flattened by a war or other calamity over that timeline.

I think you must have skimmed some parts. He isn’t burring steel into the ground. It’s only to be used inside the buildings shell and above the foundation. Thus it’s generally going to stay dry and mostly avoid significant corrosion issues, especially considering the generous safety factor suggested.

The foundation is concrete piles to bedrock supporting a floating concrete foundation. Clearly that limits building locations, but plenty of places have bedrock reasonably close to the surface.

The point you are replying to is that stainless steel is unnecessary, because 'normal' steel will ast more than 1000 years in the conditions described.

Not when one of the assumptions is that the shell will be seriously compromised at some point. Stainless steel is just another durability vs cost tradeoff, but it could be what makes the difference between repair vs rebuilding and it’s not that expensive.

Wood for example could hypothetically last that long, but both fire and leaks are a real concern.

Okay, so your step #1 for building a 1000 year house is to just... have a house that never has any leaks or termites? I can see some issues.

If you want something to last, you plan for when it does go wrong. You should be designing a house that survives ten once-in-a-century floods and storms. Freezing pipes. Hot and cold. Termites, carpenter ants, mice.

> ... which also means that unburied steel, protected from elements, up in the air, is going to last more than 1000 years.

Rust is protective. That's why cars can rust so quickly, because vibration breaks the rust flakes off and exposes steel to more air and water. Steel beams are subject to bending and vibration. Steel buried underground is not. A larger problem is also that rust is expansive. I'm not positive what you mean by plate connectors, but nailed tie plates push their nails out over time. Screws and bolts work fine.

The biggest problem comes with using treated wood and steel together. Treated wood, even the non-arsenic ones, use copper compounds. That causes galvanic corrosion. It'll even eat through zinc-coated steel. Galvanized steel is enough for at least a couple decades, but I have no idea about a millenium.

> author has thought a lot about materials and buildings

> but not actually built anything.

That was my impression. As soon as they started talking about unenforced concrete pilings (drilled? monopile?) to bedrock and stressed steel framing I wasn't certain I would enjoy anymore or that it was a good use of my time.

I think costs at this stage are unrealistically low "probably in the neighborhood of $1000-2000 per square foot ... (8 to 16 times as much as conventional construction)". ~$250-400 sq ft is the cost of modern labour expediated and material optimized building. They're describing stainless steel framing with what would be (what?) 316 SS in S-beam with a custom end plate? A blob of rolled 1" x 6' 316 SS is $200.

There are the examples we could draw from:




Material wise we'd look at clay, solid high-density rocks, non-ferrous metals (lead, aluminum, copper, tin), dense naturally mold resistant woods (cedar, redwood), high density with high oil content woods (https://en.wikipedia.org/wiki/Lignum_vitae), and easily replaceable sacrificial surfaces, dirt with live plant, cobble rock.

Building techniques aren't going to improve with technology, we already have examples that have lasted -- we're looking at building a heap temple with ancient style water and sewers. This is either a rock temple or a log-house with highly resistant woods.

Either we aim for a light-weight footprint or we find solid rock for building on. Solid rock is the most appropriate.

Our fasteners are all based on managing gravity. Rock with concave and convex connections under gravity. Wood pegs in non-load bearing configurations. Lead sheets with crimped folded seams. Solid copper sheet trays with crimped folded seams in rock trays.

The first thought should be "what happens if I drive a truck into the side of this building 10 times?" And the answer should be "not much, you move a few things around, but there is limited stressed coupling, and things rest on top of each other." If you do substantially damage the building, all you should be doing is reassembling the pile.

I called the author out for not having built anything, and in honesty, I haven't built using these ancient approaches, so, perhaps this is a self-destructing prophecy.

No you're right.

Stainless steel is insane, in those conditions it will corrode in several decades, if not faster.

Foundation is just laughable, it won't last 200 years, much less 1000.

Like you say, a structured pile of rocks with everything else easily replaceable is the recipe.

The question actually should be not of a several truck hits, but of several fires, like, complete burnouts. That's what would happen in 1000 years.

"If I had an unlimited budget and was aiming for >1000 years I would pour the piles to bedrock with stainless rebar inside fly-ash concrete and top those pilings..."

What would situation be if you used proper hot-dipped galvanizing instead? (And I mean proper quality hot-dipped galvanizing, not the usual 'toy' stuff that passes for galvanizing these days.)

I've a good reason for asking, as I've an example of galvanizing that is in such good rust-free condition after being exposed to the weather and elements for 65 years that I would not have believed the fact unless I'd actually known its background and seen its condition, which I've done on multiple occasions for reasons I've explained below.

The galvanizing in question involves an extant clothesline that's still rust-free and in good condition, it consists of three galvanized, multistrand (7 strands if I recall correctly) of clothesline wire that span the full length of the backyard and that terminate on wooden poles with horizontal 'T' beams spanning them (it's a typical clothesline that everyone's familiar with).

The clothesline was installed in the backyard of my patents home in the mid 1950s and in 1957 an electrical conduit tee junction of 3/4" diameter (for joining conduits) that was made from folded steel was wrapped freely around the middle clothesline and a dog chain attached. This meant that the dog could pull the tee junction along the line so that he could gain almost full access to the total area of the backyard.

Despite the wear and tear on the clothesline from the steel in the tee constantly sliding over the line—to wear that's even extended to the galvanizing being worn off from the high points of the strands of wire by the steel tree to the extent that it's exposed the underlying bear steel yet nevertheless this clothesline still shows no signs of rust after 65 years.

I've studied chemistry for some years so I understand how galvanizing works but I still find it incredible that this galvanized wire is still rust-free after 65 years. Frankly, I find it so astonishing that every time I visit my parents' old home I cannot but help to cheek up on the status of said clothesline.

Incidentally, little remains of the electrical tee junction, it was made of mild steel and painted black, the paint offering little protection and there would have been no paint that the point of interface between the steel and the wire.

> I would only use steel members if the span called for wood that was too big (like a 24' span needing a 8x14 or whatever).

What about laminated timber to achieve these spans?

I expected something much simpler : bricks, stones, and wood. It's not like we are running low on examples of 1000+ years buildings (https://en.wikipedia.org/wiki/11th_century_in_architecture). These castles, cathedrals, farms, were built to last, so it's appropriate to use them as examples.

We can, however, apply modern technics and materials when they make sense : insulation, windows, waterways... Prefer wood, wool and steel over plastics or composite materials and you're good to go.

On a side note, I'm currently buying a house (old farm) with over 200 years old plain oak carpentry. The thing is absolutely massive and would be unimaginably expensive to build today. With the proper care, it might last another 200 years without issue. Remember, Notre-Dame de Paris used 300 years old trees cut in ~1150 for its roof -before the 2019 fire-. With the proper care it would have still be standing today. I find that to be deeply humbling.

> We can, however, apply modern technics and materials when they make sense : insulation, windows, waterways...

The article raises an important point that could be a problem when trying to mix old and new build techniques. When talking about brick walls it says:

> One tricky thing with this type of assembly is that while it has performed well historically, it doesn’t necessarily play nice with more modern, energy efficient construction. A solid brick wall was traditionally designed to be exposed on the inside, exposing it to interior heat and allowing it to dry. Adding interior insulation makes the house much more comfortable, but also changes the thermal dynamics, potentially causing freeze/thaw damage in the brick, and allowing moisture to accumulate between the brick and the insulation. This is one of the many details that would need to be worked out for the complete design of the home.

You can't simply build something following the examples of a castle or a cathedral, but then add modern insulation, because the insulation won't allow the masonry to breath and dry to both sides, leading to water damage, mold, rot, ...

At the end of the day, a badly insulated building can be made to last for ages passively by just making it breath, so the temperature and humidity vary with the weather but are kept in check by passive external factors (e.g. the sun shining on a external wall dries it from the outside). While a well insulated building absolutely needs constant mechanical HVAC with fine tuned control. You can have a well insulated building or a passive building, you can't have both (by the way, the "Passive House TM" insulation standard is a complete misnomer, being that mechanical ventilation is its second biggest tenet).

Construction seriously depends on location. However, the simplest solution if you want high insulation factors is to have a second air tight structure with an air gap to your brickwork. Just make sure to properly ventilate that air gap. The same approach can then be used on a slate roof.

Essentially you end up with a home inside a shell. There are several advantages to such structures such a potentially great sound insulation and aesthetics, but it’s not cheap. Having an essentially air tight structure requires a hvac system to match. A combination of heat exchanger, filter, humidity control, and temperature control let you have a very comfortable environment while still benefiting from significant insulation.

>Essentially you end up with a home inside a shell.

Which then works against the headline design goal: a house that lasts for 1000 years. How many times would you need to rip down and replace the interior shell? How much more difficult would be demolition if you can't use large equipment and smash the whole lot flat, but instead doing by hand from inside, carefully avoiding damage to the outside shell?

First it’s the system used in the article in part because it’s easy to modify brick structures. Anyway, it depends on your design goals. For a hypothetical 1,000 year home you could have an inner shell of any long lived material including wood which could last that long inside a dry shell as long as you can avoid insect damage.

That said, the real trick to a 1,000 year home is making something that people want to actually use for the next thousand years to avoid it getting replaced rather than the weather or minimal maintenance issues. Because frankly that’s the real threat, which suggests going for something interesting might beat pure utility. An ideal location might be enough but a hook like artistic frescos, intricate stonework, unusual architecture, famous occupant etc could also get you over the hump until the buildings age inherently becomes a draw.

PS: Don’t forget flexibility needs to be part of the design. Electricity, central heating, phone service, AC, and eventually the internet are all relatively recent inventions. I can only assume plenty more systems are going to become normal in the next 1,000 years.

A lesson I've been repeatedly learning since buying an old house, is that many design decisions that seem dumb now were actually optimized very well for what was available at the time. It can be a real challenge to try to retrofit modern efficiency and comfort into a home that was designed for constraints and expectations of another time.

We recently replaced our furnace and found the footprint of the original coal-fired "octopus" gravity furnace, and learning about the operation of the old furnace makes the seemingly inadequate ductwork make more sense. Instead of a furnace blower (which hadn't been introduced yet at the time the house was built) the air moved around the house by convection and relied on a temperature differential between the center of the house and the outside walls. Hot air came up a few ducts in the middle of the house, and cold air came down through return ducts on the outside walls.

Unfortunately the chimney was also acting as a radiant heating element, and one of the upstairs bedrooms has become much colder since switching to the higher-efficiency furnace (which scavenges much more heat from the exhaust, and vents out the side of the house). Ultimately I'm sure the much more efficient furnace will be worth it, but there are trade-offs that will need to be addressed.

My pet peeve is when people claim that heavy old cast iron radiators are "always better" than new flimsy steel ones.

Yes those are better for old houses with old insulation or no insulation because then you wanted radiator to keep warm so you can sit close to it and get yourself warm. Where with new thing ones you want to heat up the air so you don't want radiator to be warm but air in well insulated building.

> here with new thing ones you want to heat up the air so you don't want radiator to be warm but air in well insulated building.

This is more complicated than this, and the construction industry has actually gone back and forth on this topic for the past 40 years. A few consideration against your point:

- When you heat air, you dry it up, which can be pretty uncomfortable for the inhabitants.

- Air stratification is a big concern: when you heat air up, it raises up to the ceiling, warming nobody but the spiders in the corners. It's not uncommon to have a multiple Celsius degree difference between floor and ceiling, even in an air-tight room.

- A house cannot be air-tight because you need to bring new air for breathing, but also to avoid excess moisture in wet rooms, which means your precious warmed-up air is being vented away, and you bring cold air from outside instead. There exists sophisticated systems with heat exchangers to collect heat from the outgoing air and warm the incoming air, but those are really expensive and far from ubiquitous.

- Under standard conditions, thermal comfort comes in roughly equal part from the radiative flux (from the walls) and conduction/convection (from the air) one. So you can achieve a similar level of comfort with lower air temperature when you have a warm radiative surface in a room.

At the end of the day, old cast-iron radiators are pretty good from a heating standpoint. (But they have multiple drawbacks, like being really big, and so heavy you can't move them when you want to refresh the room paint, etc. which means that nowadays, people usually prefer underfloor heating instead)

> There exists sophisticated systems with heat exchangers to collect heat from the outgoing air and warm the incoming air, but those are really expensive and far from ubiquitous.

Energy recovery ventilation units are not that expensive and are required by code for new construction here in Ontario as of 2017. Adds some cost to the HVAC of a new build, but not that much.

> Energy recovery ventilation units are not that expensive […] Adds some cost to the HVAC of a new build, but not that much.

I've recently had my (individual) house built, and it was 3-4 times more expensive than a regular system. It's not a prominent part of the entire budget for a house, but the difference still made me chose the regular one. But maybe a significant part of the price difference was just a premium for having the high-end tech, more than a real cost difference for the builder.

> are required by code for new construction here in Ontario as of 2017

Interesting. Even for individual houses? Maybe that's a good idea to make it mandatory, if the premium hypothesis above turned to be true.

I believe the requirement that came in earlier for large apartments and condos, but 2017 was the date for detached homes (sadly by far the most common form of development here)

I’m not a super expert, but I believe HVAC here in new builds is about 20-30k, split fairly evenly between parts and install. ERVs are around 1-2k for parts and another grand or so for installation (and significantly easier when done at build time). It likely helps we have such extreme climate our HVAC costs are already quite high, so an extra $1000 here or there isn’t a big deal.

Places more moderate it could easily be a much bigger percentage of the budget and come with reduced ROI. Right now over night I’m heating about 30C temperature difference and I live in one of the most mild parts of the country. Recovering energy is a big deal, as is capturing moisture to reduce load on the humidifier (evaporating water is far from free).

Thanks for your answer. That's a great reminder that everything building-related depends a lot on the weather of the region things are being built!

Yes that is what I am describing as well - that it is more complicated. I am not saying that new steel radiators are somehow superior.

Well yes, but the example house doesn't seem to be all that optimized for energy efficiency. Look at the illustration showing a fireplace located at the end of the house. That's going to waste so much heat compared to a centrally located one. So I'm not sure how much attention was paid to energy efficiency here.

I came here looking for a comment like this. A fireplace partially exposed to the exterior (as is common nowadays) is essentially entirely decorative and can actually be a net negative in terms of heating the space. seeing that made me question the practicality of the authors other design choices that I know little about.

Mortar holding stones from even Roman times, is still wet in the middle.

> It's not like we are running low on examples of 1000+ years buildings (https://en.wikipedia.org/wiki/11th_century_in_architecture). These castles, cathedrals, farms, were built to last, so it's appropriate to use them as examples.

This is an easy place to run in to a survivorship bias problem - pick a random 1000 year old farmhouse. How many farmhouses built exactly the same way didn't last for various reasons? Is the fact that one is still standing due to luck, or was it destined to last 1000 years the day it was built?

If you look at that list, almost all of the surviving buildings are churches or religious places. If the goal is to build something that lasts 1000 years, then you need to factor in political and utility questions as well besides fixating on materials.

Your home can have perfect architecture, but if a rich person buys your land after you are gone and decides what you've had sucks, it will just get demolished and replaced with something more modern. That rarely happens with churches.

Ding ding ding. You can make a canvas tent last 1000 years if you have a maintenance team constantly making repairs. This is 100% an issue with human incentives, not construction technology.

> That rarely happens with churches.

I agree. If the church didn't survive, they would just build a new one.

I imagine there lie the remains of hundreds of separate cathedrals under one. But you would never say "the church fell down". Rather, "there was an accident and renovations were required".

I see them as a Ship of Theseus, where the most long-lasting examples were determined through a lot of trial and error.

Isn't this a thing with Notre Dame? It's been a while but I remember the opening of Hunchback mentions the rebuilding right?

Many cathedrals have been rebuilt two or three times, and building a bigger church over the top of a smaller one is normal, but there are also plenty that have survived more or less as they are for at least several centuries. (Hell, there's an 11th century church in my home village, and it still has the door on the wrong side because the village itself was moved due to plague).

Straight masonry without rebar lasts essentially forever - think of all those Roman viaducts that are still in use. Water can wear through it eventually if the shape lets it, especially in places with freeze-thaw cycles (so there's a saying that a church will survive as long as there's someone around to clear the gutters). I guess hurricanes would probably do it if you're in a place that gets those. But there's just not a whole lot to go wrong with what's essentially a big lump of stone. (Of course a bare stone building is not particularly comfortable for living in, and you have to be a bit more careful about how you maintain wood or fabric on the inside - but the building shell itself will last as it is)

It’s not clear how many of the no longer extant old homes and farms fell down due to design issues, due to neglect, or were pulled down for economic or aesthetic reasons. Most architectural styles go through a process where they stop being made, start getting neglected, most of them get pulled down, and the few remaining ones get lovingly restored as they get old and rare enough to become retro rather than just outdated. This tells us very little about how well they were built, even with survivorship bias in mind, because very few buildings will survive an utter lack of maintenance, and less still if many were torn down for not being stylish enough to command the level of rent that the owner is hoping for.

It doesn't matter that the others fell, we now know what definitely didn't fall - which is all we need to get to.

Survivorship bias says to look at all the buildings that failed and why they failed rather than looking at the building that survived. You might imitate build but you will not imitate the environment that allowed this one to survive.

It matters if you want 99% of the buildings that you build last 1000 years, rather than 0.001%.

The fact that there's still a lot of buildings still standing doesn't mean they were built perfect, it just means their construction has the possibility of lasting a 1000 years. For each of those buildings I bet there were 10 more with the exact same building techniques that are no longer standing.

A 1000 years is a lot of time to go without a fire. I think just because of the fire risk wood is simply out of the question. Well unless you can protect the wood against fire like the OP is doing with the steel.

Considering modern times, I think you would have to go one step further and also consider gas explosions and possibly being bombed as well. Just imagine how many 1000+ yr old buildings must have been in Germany before WW2.

There are more old buildings left in Germany than one might first assume. The devastation of WW2 was most concentrated in the major cities, and minor cities and villages often have old centers that survived fully intact. This was a bit surprising at least for me when touring the smaller places.

There's another bias here that available materials were very different 1000 years ago.

Designing with WW2 in mind is preparing for the next conflict based on the last one. Climate change will have implications, which tend to come out as fire and flooding. This will impact siting.


I'm very curious about this kind of stuff. And now especially "activhaus" ideas. Being a software guy, probably from envy.

Ages ago, I started remodeling while my Belgian coworker (working from Belgian) started his new home construction. My house in the USA Pacific Northwest is timber framed with tar shingle roof. Coworker's house, IIRC, was block walls and ceramic roof. My house so temporary, their house built for the centuries. A real home.

In North America, I now find the homesteading and packed earth style homes most compelling. Basically, latest tech Arcosanti. (But with better finances.)

Rammed earth you definitely need to be careful of the humidity and mitigation internal humidity is going to be important. Good technique though for arid/desert climates.

Timber Frame + external insulation with anairtight building envelope is a really good construction method that will be energy efficient and last a long time.

With the right clay mixture, rammed earth can be made more or less waterproof like brick, or you can make actual fired rammed earth bricks. Adding graphene flakes can significantly improve the structural qualities, and added carbon black can increase thermal conductivity (heated flooring) or provide em shielding.

There are lots of materials to play with and mix up for structures.

Timber frame and a lightweight roof is superior to block walls and ceramic in any scenario that includes seismic activity, which includes the entire American west coast.

If wood stays dry it will last for centuries. Of course given the climate here you'll need to maintain and eventually re-shingle your roof (In the same region and my 34 year old tar shingle roof sprung a leak a few days ago) but if you do that your house will likely last just as long as your coworker's.

Ceramic shingles are replaced after 40-50 years as well. At least in Germany.

Arcosanti is a lot of concrete, as I recall.

Ya, no longer ideal. I'm keen to see what replacements people come up with. Like hempcrete, packed earth.

What is behind your thinking that plastic won’t last? Plastic is estimated to take 1000 years to decompose in landfills. Obviously it did not exist millennia ago, but I wouldn’t write it off as not future proof.

I too have a 200 year old house and the sagging mentioned in the article seems inevitable in wood.

Plastics break down in sunlight, becoming brittle and easily broken. They're also basically impossible to repair if they do break. They can also been eaten by rodents.

However plastic is a great material for dark, rodent free places.

“Decompose” being the tail end of the deterioration process, with the usable period being far shorter.

Plastic breaks down in tiny pieces and is almost impossible to repair when broken. It's also very difficult to reuse or recycle when you want to. If your house is still standing after a thousand years, it's also because its habitants were able to remodel it without destroying its materials.

As a matter of fact, the amazing carpentry I was talking about previously is made from "fresh" oak, but a significant part of it was also taken from a previous building (church, farm, stable or monastery). It's doing just fine, apart from a few out-of-place mortises. I tried reusing plastic pipes, once, but that wasn't a great idea. It's fine for a few DIY, though.

Stones, bricks, wood are great mostly because they can be reused, but also because they will continue to look great afterwards.

(Note: You might argue that "your" wood is not easy to reuse as well. This happens because we put nails and screws everywhere and we prefer less-dense wood over heavier ones (pines vs oak))

plastic becomes very brittle very quickly (years), and falls apart into tiny bits. it's these tiny bits that (we expect) take forever to decompose.

even if we understood plastic to retain its original qualities over time, it is impossible to speculate about a thousand-year lifetime. it's unusual to come across any plastic a hundred years old, intact or not.

Wood is flammable, and so it’s kind of out of the question. I’d build out of stone or brick, single story but larger footprint. Then of course ground source heat pump and all electric appliances, and solar nearby..

modern wood construction is more resistant to fire than steel, steel loses its structural strength when it gets hot while wood chars on the outside (becoming fire resistant) and doesn’t lose its strength

I've seen the aftermaths of a few barn fires over the past few years. These were old structures, pre-1950s based on appearance, framed with heavy timbers and clad with wood planks and a steel roof. After the fires, the timber frames were still standing on their stone foundations, charred and black, but there was nothing to be seen of the rest of the structure.

I mean.... Cheap Old Houses had a 1000 year old house for sale not long ago.

There's also plenty of timber buildings that last for millenia.

If I were to design a house with the idea that it should serve future generations, I'd design it to be reconfigured, recycled, or torn down easily.

After reading some of Stewart Brand's writing, I've learned to love ugly buildings.

The parent poster is probably referring to "How Buildings Learn", a book by Stewart Brand with lots of architectural pictures and commentary about how various buildings have been (re)used over the decades or centuries.

Brand posits (a) human needs change faster than buildings age, and thus buildings must adapt to that change over the lifespan of the building

(b) there are two reasons a building lives to be more than 100 years old: either the building is historic / well loved enough that we live with a building's warts even though it doesn't meet our needs perfectly (a Parliament building, a church, a house that cannot be modified) OR it is so flexible or easy (cheap) to modify that it can suit many purposes. (A small commercial building that can hold a dentist office or a restaurant or a law office or a nail salon, a house with an extension, a warehouse that can be converted to a modest factory floor, etc)

Buildings that cannot be adapted are torn down and replaced.

The book is excellent, and beautiful, and I recommend a physical copy to everyone.

It's fascinating how culture has effects on architecture. So many older American homes have small kitchens that are separate from a formal dining room. Historically, a family member or cook would be making food separately, out of the way, and then it was presented in the dining room.

Today, everyone wants a kitchen that's integrated with the dining area. Cooking has been culturally elevated – people don't feel like they need to do it out of the way. But unfortunately, many of these older homes cannot be easily modified. Walls are often load-bearing instead of being reconfigurable.

Not just American homes. In Europe all the new flats and houses are designed with kitchens merged with the dining room. Even the old flats are often redesigned by owners or developers by tearing down the wall(s), separating the kitchen from the other room(s). I personally hate this trend because I don't like getting cooking odors all over the place. I'll probably just build a wall whenever I'd be forced to buy such a place.

The solution to cooking odors is to have adequate ventilation that exits the structure. It blows me away that this is not code in every kitchen and bathroom. I live in rules-heavy CA and my kitchen vent "exhausts" back into the kitchen.

Every time there are people gathering at someone's house, they tend to congregate in the kitchen as that's where the food action is all at. Give me one big open space with kitchen, dining space and a living room all in one. Open concept is highly desirable in modern housing.

With proper ventilation this isn't an issue. My kitchen is centrally located, so I installed a range hood made by a Chinese company (Fotile, approximately $1300 on Amazon). It works exceptionally well. Even the smokiest/smelliest cooking odors don't escape.

This is super funny to me. I just bought a house where the kitchen is separate. My parents, and my mother-in-law were all like... "you can knock down the wall and bring the rooms together"... and I was like: "hell no. As if I want a bunch of kids playing tag while I'm chopping things with sharp knives and running around with burning oil."

...Leave me alone to cook an awesome meal in peace. I'm working in there not messing around.

I'm actually designing a new house and was surprised at the resistance I'm getting from the builders/architects in having a kitchen which is out of the way of the rest of the house. Like you, I'm not interested in dumping hot oil all over a kid/dog/mother who decides to run through the kitchen despite the blockade I set up to keep them out.

That, and kitchens require so much maintenance to keep clean and presentable. I hate how my entire house feels dirty because it's visible from any public spot in the house that I didn't get around to cleaning a few pans the night before.

I'd guess that's because of another phenomenon: A "good" house is one ready to sell to the next guy. Youre never living in your own house, always the next buyer's house. Hence the need to keep up with the latest fads, desired by the next guy who is buying from you for the sake of the next guy.

What I see a lot of in my region (US PNW) is open kitchens integrating with informal dining areas (breakfast nook, or seating at a kitchen island) and living rooms. But most houses are still built with a separate dining room. Heck, my house was even built with a butler's pantry to connect them. That's pretty common.

You have a separate dining room if you’re building a 6 room or larger MANSION. Calling a building with a separate dining room on top of a kitchen plus informal dining room as a Regular home seems to be a stretch.

That's regional, of course. I'm talking bog-standard everyday houses in the 2500-3500sf range, no mansions.

2500sf is 232 square meters. That's three normal family city apartments by European standards. I guess houses just run bigger there but sounds quite excessive nonetheless.

"Excessive" is a moral judgement. City apartments aren't typically 2500sf in the US, either. I was referring to single family detached homes, which are by far the dominant type of dwelling in my region. Land is cheap, wood is cheap. A 2500sf house will be a good bit cheaper than a little flat in most western European cities, I bet.

Didn't mean it as a moral judgement, sorry if that was not the right choice of words.

This seems cultural for sure. I grew up in an apartment built in early 1970s with separate kitchen. To this day most apartments in my country are planned w/o open kitchens.

It is actually an ongoing joke with my partner, who has lived an open-kitchen life until meeting me.

PS: would be interesting to correlate enclosed kitchen with cost and availability of home labourers (slaves, maids, cooks, etc)

Agreed. The idea that homes don't last 1000 years because of their construction quality is conflating correlation and causation. Homes don't last that long because ideas about what (or where) a home should be don't last 1000 years. Heck, in recent times, they hardly last 100 years.

> Agreed. The idea that homes don't last 1000 years because of their construction quality is conflating correlation and causation. Homes don't last that long because ideas about what (or where) a home should be don't last 1000 years. Heck, in recent times, they hardly last 100 years.

And (IIRC) in Japan they often don't last past one owner, since a new owner will typically want to build a new house for themselves on the lot. An old home actually lowers the value of a lot, since you have to factor the demolition cost into the price.

> in Japan they often don't last past one owner

This has been brought up before on HN and I remember a comment mentioning that they thought it was due to the fact that earthquake proofing technology advances quickly in Japan so an old house might not be up to the most current standard. Curious if that's correct, and if not, why is this so common in Japan then.

A thing I've noticed while attending school in Japan was that many old "Machiya" houses get torn down and more modern and western houses get built. At least this was the case in Kyoto. Earthquake-proofing the house is one aspect but I think people in Japan prefer to own a more western home. It's too bad because those old Japanese homes are getting taken down

It's not a case of "more western home".

What you see as a "more western home" is a more energy efficient outer shell. Which just happens to be fairly universal.

I agree on that, although there are residential homes in Japan that implement energy efficient outer shell, while still keeping the Machiya look. There definitely is a preference in Japanese society for western style homes.

As a Japanese, I think modern houses in Japan as just a "house" rather than "western style house". That's the default rather than preference. Traditional Japanese style house isn't an option for most people. Cost and insulation would be a big reason.

Most modern Japanese homes are not "western style" at all. They're usually built by large industrial concerns (e.g., Toyota, Sekisui, etc.) that prefabricate many components at scale. The process of building such a home in Japan begins with a company architect fitting their design system onto your lot. The price is fairly predictable, and relatively cheap—-under US$300K in most places. The home is designed with more or less the following priorities: 1. Safety 2. Ease of maintenance 3. Efficiency 4. Cost

And more recently, considerations for SDGs (sustainability goals) with respect to materials used.

Longevity is not a priority, as homes are expected to depreciate in value over roughly 20 years. There are many reasons for this, but my personal opinion is that industrialization has made it possible to upgrade the technology of the home at a pace and a price that favors rebuilding.

As for machiya and kominka, local governments like Kyoto have tried to intervene to preserve the traditional homes. The "no build" lots do not allow a property owner to build on anything other than the load-bearing structure for the old home. As a result, you have many empty lots and coin parking lots around Kyoto where the old home was unsalvageable or where the owner could not afford a renovation. To be honest, only tourists/foreigners would want to stay in a machiya for the novelty of it. Although they were marvelously engineered for their time, they tend to be rough living compared to the extremely easy and cheap prefab homes. There is also the problem of craftspeople who can maintain these old homes dying out, which adds to the cost of keeping them.

I don't think it's possible to attribute any of this to a general "Japanese" attitude, though. On the one hand, one of the most important and longest-lasting spiritual sites in Japan is Ise Jingu, which is rebuilt from scratch every 20 years as a Shinto ritual of renewal. On the other hand, you have some of the oldest and largest wooden structures in the world still standing in Nara.

As usual, it's complicated.

Earthquake is part of it. However culture is also part. Houses are built to last 20 years: even if you like where you live they still assume you will be rebuilding in 20 years. As such they can cut corners to save money - no problem so long as you rebuild every 20 years.

I think earthquakes, volcanoes and the like may have been the original reason that the culture sees buildings as transitory things. But like many cultural things it's now self-sustaining.

I immediately thought about the schools in smaller towns around here that were built in the 90s, which were built to last, and are now sitting as barely used community or senior centers. Society's needs can change a lot in 30 years, let alone 1000.

The author basically agrees with you, lots of good commentary in the Conclusion section of the article:

> Designing a building for an extremely long lifespan is in some sense a bet on a certain kind of future - one where tomorrow’s physical infrastructure needs aren’t all that different from todays. And because physical infrastructure is hard to change once it’s in place, it’s also an attempt to bring that kind of future into existence. But if you think agglomeration effects should push cities to get larger and denser, or if you think we’re likely to see some cities shrinking as the nature of the economy changes, or if you think building technology is likely to change significantly, an extremely durable, an extremely long-lived house is perhaps less desirable.

Seems like a lot of the article is about just that. A really robust, reliable structure & foundation, with progressively less permanent things attached. A brick facade can easily last hundreds of years, but not a big deal to replace. Interior walls made of non-load-bearing-wood makes it "easy" enough to reconfigure rooms.

He mentions how the building shape is just a rectangle, which makes it reasonable to repurpose for many uses (he mentions office & separate apartments). He takes care to allow for routing of utilities under and inside the building.

It's not quite as reconfigurable as I think you're getting at - optimized for being torn down, but it's much closer than a typical 100yr house would be.

Wood walls are more complext to rearrange, than steel framed walls.(that is why offices have steel frames under that drywall - easy to install and move)

I think the reconfigured / recycled easily makes sense. Tough to build a product in 2000s for life in the 2900s.

That said the cost to make homes reconfigured/recycled easily is probably quite high and who knows if there will be people with knowledge to be able to perform that work in 400 years. Whereas a shelter can always be used by humans...

I personally believe that you need to find a compromise between a framework that lasts very long and party that can be recycled so you can adopt the house depending on your needs without needing to redo everything (for example foundation and some integral supports could probably be designed as a framework that can be adapted while walls, windows, room configuration, doors, etc. need to be changed eventually and therefore should be recyclable).

We already do this with commercial construction. Every floor is just a big box that can be sectioned off as the new owner sees fit.

The key is that the structure is basically self standing with nothing but the outer walls, and if it's a tower, the center core instead.

Add a vertical chase way too to bottom for easy reconfiguring of cables and pipes and you can basically do anything you want until the structure fails.

Modern residential doesn't build like this because it would be expensive and probably pretty ugly. It's a lot cheaper to get your structure in bits and pieces by stacking walls on walls on the foundation then have just an outside structure and then have the floors and roof spanning outside to outside.

But it could easily be done with the materials we have.

Another major thing is the drop ceiling. It's designed to make access easy so that you can run cables and new plumbing anywhere you want.

It's ugly so you'd want to find a way to have a modular ceiling that isn't a drop ceiling, but maybe that works for you.

Last, for residential, you may butt up against height issues as to be modular like that you're want to have a couple feet extra between floors so you have room to move things around without opening things up.

> Another major thing is the drop ceiling. It's designed to make access easy so that you can run cables and new plumbing anywhere you want. > It's ugly so you'd want to find a way to have a modular ceiling that isn't a drop ceiling, but maybe that works for you.

If you build with trusses for floors instead of joists, you can get a lot of the same benefits as a drop ceiling without the ugliness. You can run plumbing/vents/whatever though the trusses without destroying the whole ceiling to get access. You might not need to cut into the ceiling at all depending on what you’re doing and what existing access you have (e.g. an unfinished utility space may allow access to supply new power cables).

Of course patching the ceiling is hardly a big concern if you’re talking about a structure surviving for 1000 years.

The 1000 years is a separate conversation I think than the modular concept. One could lead to another, but depending on how modular you wanted to be, access to the ceiling is a must.

On the other hand, if you make the trusses large enough and give an /attic/crawl access you could go up there and do the thing without needing to renovate. Then all center walls are non structural and you can move them as you please.

Come to think of it, I might build a house like this.

You can also leave the mechanical on the outside and not cover it too. I guess there's a certain beauty to me in the robot parts but I don't think many would like that.

How often you expect to reconfigure is a huge factor. Drop ceilings make complete sense in many commercial buildings, where access to change/move cables and whatnot probably happens every year. For residences, that sort of work happens a lot less frequently.

You can also get away with ugly ceilings more easily when they are higher. A drop ceiling 8 feet or less from the floor is an eyesore. At 12 feet up it’s a lot less noticeable. Of course if you have very high ceilings you can often just leave them uncovered and have a more industrial aesthetic. You have to be more intentional about routing all the utilities then.

The cost may be high, but it doesn't have to be. That's why I mentioned Brand.

An example he cites is MIT's Building 20[1]. It only stood for 50 years, but that's not too bad for a structure that was intended to be temporary. Some amazing stuff came out of that building.

[1]: https://en.wikipedia.org/wiki/Building_20

This is what things are like in many parts of Japan (see https://www.theguardian.com/cities/2017/nov/16/japan-reusabl...).

Oddly enough, the value of houses in Japan depreciate over time (like cars). It's kind of a cultural thing. People want to live in a new house of their own, so many houses are built in a pre-fab way, with the intent that they'll be torn down and recycled in 20-30 years.

Wood is pretty easy to reuse and basically lasts forever if you keep it dry.

> torn down

Hmmmm. I think that doesn't count on ship of Theseus type grounds.

Be wary of making it too ugly though - people will look after it much better if it's beautiful.

Stainless steel rebar is an often overlooked option. In theory, solid SS rebar should outlast the concrete, but it is a difficult thing to accurately study. In favorable conditions, regular rebar reinforced concrete starts to need major repairs after ~40 years due to corrosion.

The Progresso Pier in Mexico was build over 80 years ago with SS rebar, and reportedly has not needed any renovations. A pier built 20 years later using mild steel rebar has been almost completely destroyed by the ocean.

I wish more large infrastructure projects would use it. The up-front costs can be 2x higher, but the lifetime savings win out in many situations.


Stainless steel needs oxygen[1], otherwise it will eventually corrode with pitting and "crevice corrosion"[2]. I wonder what the ingredients are in stainless rebar, and what the oxygen environment is like.

[1] https://www.thermofisher.com/blog/metals/is-stainless-steel-...

[2] https://www.cruisingworld.com/how/beware-stainless-steel-cor...

Hey this is a little late, but now they are looking at non oxidizing basalt based rebar and mesh's as an alternative to steel. It's work a look. it's got a lot going for it.

I came here to mention that. SS rebar is, indeed, a thing and would be an interesting combination with long lasting (fly ash) concrete, etc.

How long lasting is fly ash concrete?

"How long lasting is fly ash concrete?"

We don't know. Existing structures that use "Roman Concrete[1]" are (roughly) 2000 years old and counting ...

[1] https://en.wikipedia.org/wiki/Roman_concrete

How does the Pantheon in Rome keep its structure? Is it just concrete?

Since it’s 2000 years old now.

Here's a good video on the use and tradeoffs of Roman concrete engineering: https://www.youtube.com/watch?v=qL0BB2PRY7k

TL;DW: whereas modern construction uses rebar as a way to keep concrete from fracturing under tensile stress, the Romans made their constructions enormous so that the weight of the structure itself would compress the material and keep it from failing from tensile stress. Their monuments weren't built huge just because it cool, but also because it was practical. But large concrete constructions are both expensive and take years and years to cure, and depending on your concrete chemistry the strongest mixtures can also be much more difficult to work with.

Thanks, so very cool how they were able to make such structures!

Roman concrete is not like modern concrete. The process in modern concrete does not stop and eventually makes it too brittle and falls apart, Roman concrete does not do this and can last a very long time.


Compressive shapes and good concrete.

You start talking about rebar and other complexities when you want a shape that puts concrete in tension (where it's very weak), instead of compression.

Basically - lots of domes and arches.

Wish we could re-discover how to produce Roman Concrete, which has already proven its long-lasting efficacy. https://en.wikipedia.org/wiki/Roman_concrete.

My understanding is that modern concrete would last as long as Roman concrete would, if we built the same types of designs that Romans used concrete for. The big difference is that we want to span gaps without using large unwieldy arches, so we need tensile strength, so we need to use steel reinforcing bar in our concrete, which is the eventual pathway to failure. (Well, that and heavy machinery like semi-tractor trailers, which the Romans also didn't have to design around)

The Romans did not have the quantity of cheap steel necessary for this, so they ensured only compressive loads on their concrete, so it lasted about as long as you'd expect a random rock subject to only compressive loads in a field to last.

> so it lasted about as long as you'd expect a random rock subject to only compressive loads in a field to last.

Roman Concrete is a bit more special than some 'random rock' thanks to the presence of "aluminous tobermorite" which makes the concrete much stronger and more chemically stable


I stayed at a farm in South Tyrol, Italy this fall. (South Tyrol is actually in northern Italy, near the Austrian border.)

The oldest property record for the place dated it back to the year 1200. It's a large, normal looking house, and the walls are made out of irregular stone blocks mortared together.

For what it's worth, buildings like this aren't that uncommon in Italy.

Do you have a picture?

This is the only one I have of the exterior: https://imgur.com/a/q1Y4nlL

They forgot the most important part about making a structure last 1000 years: it cannot be made of materials that people would conceivably want to repurpose in times of duress or those materials should be in such a form that it is extremely difficult to remove them from the structure. This is how many historical buildings were lost, they were mined for stone to use in other structures. It's also how many modern buildings get ruined by people looking to sell the copper wiring or pipe, and that's in a politically stable era.

A better strategy would be to build the house from massive blocks of the most common stone in the area. The blocks should be large enough that they would require significant effort to move or demolish. I wouldn't recommend using any metal in the structure of the house at all. Even wood could be conceivably stripped in times of need.

"We’re also taking something of a risk using something as valuable as stainless steel - a common failure mode for buildings is for valuable material to be ripped out and repurposed. This can range from looters ripping the copper piping out of a house to sell for scrap, to Londoners reusing the stones from ruined Roman buildings, to countries at war melting down building components to make munitions. I don’t see an obvious way of addressing this problem - the risks of corrosion we’re avoiding with stainless steel seems like it’s worth the tradeoff, and covering it with masonry or concrete seems like it would make it less likely on the margin. But this is another reason not to use something as durable as Inconel - the value of the material would likely exceed the value of the building, which is inherently risky for long-term survival."

Missed it, thanks. I still take issue with the design. I think over 1000 years the risk is much higher than they imagine.

All you can really do is minimize it. You can't eliminate it. I mean, you can't even tell me whether humans will exist in 500 years, or whether intelligent non-humans will be running around (doesn't even have to be "aliens", humans will create them), or what. Trying to second guess how people will valuate things in 900 years is a joke.

Judging from the old towns where I grew up in in Europe, the problem isn't building a house that won't collapse for a 1000 years (a classic half-timbered house [0] will get you through most of the earthquakes to expect here for centuries). The problem is getting it through town fires [1], floods [2], and wars [3, 4].

[0] https://upload.wikimedia.org/wikipedia/commons/thumb/8/8b/Ma...

[1] https://en.wikipedia.org/wiki/Great_Fire_of_London

[2] https://en.wikipedia.org/wiki/St._Mary_Magdalene%27s_flood

[3] https://en.wikipedia.org/wiki/Sack_of_Magdeburg

[4] https://upload.wikimedia.org/wikipedia/commons/thumb/e/ec/Lu...

Stainless steel girders? I see this is a no costs spared build.

At one point the author even considered Inconel girders, but practical considerations on builder experience with exotic alloys made that a bridge too far.

Even so he is planning to have builders come in an brick up the entire frame before the rest of the house is built.

I did like that he realized one of the most important aspects of keeping a house around is to make people want to keep it around. Make sure it doesn't age poorly because then even if the structure is sound people will tear it down because "it is an eyesore".

Stainless steel is also much more brittle than construction steel. And this matters a lot. Not only in case of earthquakes but also just in construction where you have large tolerances. Construction steel will bend plastically, while a more brittle steel can just crack. Welding stainless steel is also inviting issues, but possible in principle.

That, and using both steel and stone in the same structure brings its own class of problems due to the different expansion coefficients which needs to be dealt with in a very ingenious way if that is supposed to last for a millenium (or more).

Seems to me like, given the expense and duration goals, you'd be much better off forgoing steel entirely and creating a stone gravity-bound structure, and making the places where you can't go with stone easy to repair or replace, something for which I'm not sure structural steel is ideal.

> Our other option is slate. Slate roofs have extremely long lifespans and are extremely attractive. But, like copper, they’re more expensive upfront, and require more specialized skills to install (since they’re less common). A slate roof is also extremely heavy, putting more weight on our framing and increasing the risk of damage during an earthquake.

OP apparently doesn't know that slate roof have to be repaired all the time. Slates will age and break, especially if they're nailed (because the metal expands and cracks the slate).

I've spent 10 years with a slate roof, and it has to regularly be fully checked, and missing or breaking slates replaced (because they'll leak).

Screw slate, give me terracota tiles any day of the week. Lighter, way more flexible, and easier to replace when they invariably break.

Yes, another way to spell Slate roof = work. The reason is simple: Slate, layers of fossilized leaves, has a rough surface and frost and the weather in general will work on it and split the layers apart, seeds will find enough purchase to germinate (the handy supply of water certainly helps) and lichen and moss just love to grow on slate.

This whole article to me reads: "I'm planning on an overpriced construction for my house and need a plausible excuse'. It's a status thing and a discussion piece, not a serious project. If you want to build for a millenium: copy the Romans. Done. And even then you're going to have to re-do all the trimmings every so many years because they'll all give out with use. Even staircases made out of solid stone will wear over such time spans.

> Slate, layers of fossilized leaves, has a rough surface and frost and the weather in general will work on it and split the layers apart, seeds will find enough purchase to germinate (the handy supply of water certainly helps) and lichen and moss just love to grow on slate.

25 years ago, when I saw some roofers working to replace an old slate roof on a church outside Philadelphia with asphalt (I was horrified), I asked them why they were taking this (to me) horrible step, since my parents live and stay in homes in the UK with slate roofs that are between 300 and 500 years old.

The roofers laughed and said "yeah, that's probably welsh slate. The stuff here in PA is so much worse than that. Freeze-thaw will destroy it in 20-40 years"

So the observations you're making about slate are true but only for specific slate quarries. There are slate sources that can provide slate which could last for centuries.

The UK seems not to have much of an issue with moss & lichens causing problems with slate roofing (it grows but it isn't much of a problem).

That may be more of a typical “newer stuff is garbage”.

I live in upstate NY, arguably a nastier climate, and it’s not atypical to see 19th century buildings with intact slate roofs.

I think the roofers would have said "oh, and the upstate NY stuff is pretty good too". The slate I've looked at in detail in PA really is pretty bad. It just isn't as dense as the welsh stuff in the UK.

Interesting, I never realized that there can be such a huge difference in quality, thank you.

Slate is not made of fossilized leaves; its foliate structure arises from the metamorphic process as flakes of clay (aluminium silicates) align and merge into sheets under transverse pressure. Any organic material present in the original sedimentary deposit will typically result in a graphite inclusion.

Hm, ok! I totally bought this when it was related to me but you are absolutely correct. It always makes me wonder if there is a faster way to cross check everything in your head to fish out the false stuff other than people taking the time to point these things out. Thank you.

I wish there were a service to sort through the ideas in my head and point out the flawed ones too — until I realize that a) I'd have to agree to one implanted product preference for each hundred thoughts scanned, and b) many of my thoughts would be placed behind trigger warning overlays.

I don't think slate roofs are that bad - our house is an exceptionally exposed spot and has a slate roof and we lose maybe one or two slates a year to storms. Our wooden windows and doors are a far bigger maintenance headache than our roof.

When I got to the part about wood windows it dawned on me that the author is less clued in than he lets on. Wood windows are a maintenance nightmare. You can't open them half the year in a humid climate, or all of the year in an old house that has settled. No window is going to last 1000 years so might as well pick one that will make you hate your house less in the interim.

Count yourself lucky :)

And one or two slates per year is indeed manageable, assuming they are in an accessible spot. If you're unlucky they are not and then you have to get to the spot to apply your fix without breaking more slates, which can be quite a bit of work (remove slates to make a path to the spot, fix, then rehang all the others, and hopefully they were uniform).

I've had one storm bad enough in NL that we lost some rooftiles, which were fairly easily replaced. Since in the rest of the country people had lost whole roofs and other houses in the same street were in much worse shape and comparing with the few houses that had slate I'm pretty happy with my good old 'dakpannen', which are almost maintenance free (due to the angle of the roof).

The worst is thatched roofs. Those require pretty much bi-annual upkeep and tend to become rodent infested. They look pretty in the first 10 years, a bit garish in the second and depending on their state of maintenance horror shows in the last 10. I'll never live in a house with one of those, people like them for status but they tend to be people that can afford to pay others to do their work for them.

>Make sure it doesn't age poorly because then even if the structure is sound people will tear it down because "it is an eyesore".

This is an interesting one, lots of interesting buildings were torn down as old eyesores to be replaced with much more efficient modern buildings in the post-WW2 UK. Nowadays these are considered eyesores ripe for demolition and the buildings they replaced are valued, to the point post-War town planners are sometimes cursed to this very day. Taste changes often, I think the best chance of keeping a building around on these grounds are to make it an interesting or particularly elegant example of a style subjectively considered by most to be timeless.

I know it's purely subjective, but I really think the trend in architecture to do away with ornamentation was quite bad from a 'places real people have to live in' perspective, even if it's interesting from an artistic point of view.

"Eyesore" also depends on the community. Lots of places will keep an eyesore place gladly because they can't infringe on the property owners rights. Many places in Asia are like this in general.

> Lots of places will keep an eyesore place gladly because they can't infringe on the property owners rights.

This is some fun new meaning of the word "gladly"

Sure, but over the course of 1000 years presumably the property will be brought and sold many times. So you need a building at least desirable enough that the subsequent owners won't opt to demolish it.

Of course, designing widely beloved buildings is easier said than done.

This might work in a universe with spherical cows but they seem to hand wave away all human elements. My eyes rolled a loop in my head when they advised an urban location. That's a great way to ensure it gets demolished when a marginally better use for the land comes along.

If it were me I'd just build some monstrosity of a palace in somewhere that nobody wants such a thing and I'd build it out of stuff that's highly inefficient to repurpose, not steel beams. The best way to keep something around is to make its continued use better than any other option so that people take care of it and give it the capacity to withstand a couple generations of neglect without falling in on itself. A castle (metaphorical or literal) on some cheap land along the highway in North Dakota should suffice.

I would expect a southwest desert to be better than North Dakota. Fewer freeze/thaw cycles while wet, and milder weather.

But yes.

You should still use reinforcement in the concrete: Use basalt rebar instead of steel; it will not corrode. Use a nylon fiber in the mix to protect against impact and spalling from fire. Use 5000 psi concrete instead of 2500. This is much stronger and will be less likely to break down under any environment. Use a vapor barrier underneath to help prevent moisture and gas from coming up into the house. Put the foundation on 8 to 10 inches of 1" rock. This will help protect against soil expansion and allow water to quickly flow underneath or out of underneath the house. It will also prevent critters from digging into any possible underground utilities, etc. Any cracks after the concrete sets, fill and then seal the entire pad. Be sure to keep the concrete wet and covered for 30 days to aid in maximum strength.

Instead of a steel frame, use insulated concrete forms, again using basalt rebar. This makes a concrete walled house. Use stainless steel trusses (or onsite galvanized steel) for the roof or build a concrete roof with similar construction methods as the insulated concrete forms.

The siding of the house should be concrete board or other non combustible material (brick or stone)..or both where it makes sense. But be careful on the mortar used..seal it at least if you expose any of it to the weather.

Make sure the eaves are at least two feet out and the eaves over doors more than that. This keeps water away from those areas and the house as a whole.

Make sure you have gutters...good ones.

Make sure your land around the house moves water around it - even in flash flood events.

Have real shutters for your windows.

Where it makes sense, especially those areas exposed to weather, do not use wood.

Instead of slate..use aluminum shingles.

Forget the fireplace...too many potential issues with fire, leaking, etc. They are hard to build for 20 years let alone 1000.

Use Fiberglass windows. The best ones will outlast any hardwood.

I think the constraint of expanding the house prevented a build like you're suggesting. I personally think it's a silly constraint when the plan is already 3000+ sqft, which is larger than anything used as non-communal shelter by humans for our species' duration.

I also dislike how the author completely punted on insulation. I think that's a very important part of any new building. It's easy to insulate your proposed design.

Yes, insulation is very important. I actually built this house for myself. Took 2.5 years...the insulated concrete forms I used have built in insulation for the walls...and for the foundation I used special foam around the edges of the foundation to help with that. A commenter mentioned elsewhere that you would not be able to find contractors to build a house like OP designed - that is a true statement even with my build and that is why I had to build 90 percent of myself (my sons and wife helped too).

That's awesome! It's a dream of mine to someday design my own home, though more from an architectural perspective and not a technical one like this. Will likely be unrealized, though, for financial reasons. Unless I win the lottery or something, but it's kind of hard to do that when you don't play ¯\_(ツ)_/¯

Thanks for all the insights!

I have a few questions regarding your suggestions:

> Instead of slate..use aluminum shingles

- Is this purely a question of price?

> Use Fiberglass windows. The best ones will outlast any hardwood.

- Wouldn't long term exposure of the fiberglass windows to the sun weaken the fiberglass (this effect seems to be called "Fiber Blooming")?

Finally, do you have any links to share regarding these topics for people wanting to build a house but without the technical background?

Slate can be easily damaged by hail and your kids baseballs. Aluminum will be resistant to all the sizes of hail except biblical sizes. Aluminum also will reflect pretty 100 percent of the heat from the sun - allowing you to have an insulated/conditioned attic if you want. As to cost - aluminum shingles are pretty expensive too so probably no cost savings there.

The higher quality fiberglass windows have resins and treatments that are pretty much impervious to UV light...so that has been a problem with fiberglass windows but not so much anymore. Another option are steel or aluminum windows. There are some very high quality windows made out of that material as well.

I do have links...will post them later.

Why do you need shudders for your windows? Is that for hurricane/storm protection?

Out flat in Edinburgh New Town had built in shutters on the inside of sash windows - these were actually really effective at helping to keep the place warm - far better and easier to care for than curtains.

Edit: I should point out that the Edinburgh New Town is quite old, but not as old as the Old Town, obviously.

But wouldn't that be an apples to oranges comparison because shutters are fully opaque while curtains are only partially opaque?

I would also expect outer shutters to insulate the window a little bit and thus limit the heat loss no?

"curtains are only partially opaque"

Not any curtain I've ever encountered - people used to have net screens for privacy and curtains for warmth. But that's going back a bit (i.e. my youth).

Indeed that's what I'm thinking about where the curtains are open most of the day but the net screens are always closed.

It is still a decently common practice in a lot of European countries.

Other option would be to use rolling window shutters on the outside.

Yes. And if a window is broken long term (who knows what 1000 years may bring) the opening can be somewhat protected still and easily so.

And build it either on bedrock or on a stable sand layer. The foundation is key.

It's "steel" here, and "shutters"


It still says 'steal'.

My parents house in Lincolnshire England was built in the early 1500s, so about 500 years old. Have no doubt it will still be there in another 500. Obviously many changes have been made over the years but the core of the house is the same.

Solid, 3ft thick, limestone walls. Lime mortar. Probably no foundations, there has been quite a bit of movement previously but none recently. In one room upstairs the floor slopes by nearly a foot from one end to the other.

As far as we know the majority of the roofing/floor timbers are original.

Limestone slate roof, this needs replacing about every 70years.

Stays cool in the summer and relatively easy to keep warm in the winter. Not efficient in the modern sense though.

The way we live changes, trying to build a house for how people will live in the future is impossible. All we can do is build something that’s maintainable, solid and hope for the best.

I think the danger is that if you aim to design something that will last 1000 years you will over engineer it and it will be difficult to maintain and modify.

I hate to be a stuck in the mud, but a concrete pad, with unreinforced piles is not going to last a 1000 years. those piles are going to be impossible to repair without breaking the slab, or undermining, which means it's expensive to maintain. (yes Roman concrete has lasted 1k years, but thats a different type to the cement they use now.)

The other thing that they've not managed to control is moisture. You can't mix and match steel with lime mortar (I mean you can, but its not wise) You can just put a moisture barrier in there, but you need a way to maintain that (its not like a damp proof course, its far more extensive).

Personally if you want to make a house last 1k years, just make a clay lump house. It'll be far cheaper to build, look more realistic and much more well understood how to repair it.

Have they ever figured out how the Romans made concrete and got away with not using rebar ???.

Yes - basically, high quantities of volcanic ash act as a much stronger binder than is currently used/available now, and the chemistry of the cement they used meant that as it aged it got stronger.

Tangentially - it's not that we can't make concrete that way, it's that for many structures we're building of concrete, building to last 100+ years is over kill, and would increase costs drastically for a building or structure that will most likely be torn down before it reaches its life expectancy.

My making the walls incredibly thick, not having large overhangs, and having the foresight to build an empire in a seismically dead area.

Rome is a moderately seismic area suffering from amplification of seismic waves from earthquakes in the more seismic areas in the mountains. Basically all of Italy is seismic. Part of the Colosseum was destroyed by an earthquake in 1349. The google translation of this Italian page is quite good at explaining what happened https://www.focus.it/cultura/storia/perche-il-colosseo-e-cro...

TLDR, the asymmetrical 4-2 storeys shape is the result of that earthquake when the top of the part of the building built on softer ground felt down.

Yes, the research came out a few years ago: https://pubs.geoscienceworld.org/msa/ammin/article/102/7/143...

tl;dr: we took a long time to realize that the Romans didn't use potable water, they used saltwater; the volcanic ash they had access to was also very important

They only used the arch, so everything was in compression. Rebar gives concrete tensile strength.

Why have the chimney outside? It's very inefficient...


Why was that done traditionally? Because if you had a chimney fire, you could hook your mules to it, pull it down, and save the house.

This seems right but I don't know enough about chimneys, mules or even houses to determine if it's true.

c.f. "Little House on the Prairie". Chimney fire is for real.

If we're talking about chimney efficiency then we should talk about including heat exchangers within the chimney walls.

Jamie Hyneman of Mythbusters added one to his house and talks about it here https://www.youtube.com/watch?v=0T3nIk3S8Wc

I was wondering that too. If I wanted to design a wood-burning apparatus that would be efficient and last 1000 years, I'd use a design that was more like a masonry stove (centrally located on the inside of the house). In any future (I think), a backup source of heat in a cold climate is a necessary redundancy? I would use the most efficient tech now for the burn chamber (rocket?), but also design the burn chamber in such a way to allow for it to be replaced with better tech in the future (perhaps the masonry stove outer structure and thermal bank would support itself - steel exoskeleton? - etc.).

It also causes condensation because of rapid smoke cooling. Tar condensate will slowly seep through chimney wall.

Stainless or ceramic chimney liner slows that process down, but neither will last a 100 years let alone a 1000.

Not to mention wood fireplaces in the home and the noxious byproducts they produce is likely to shorten your lifespan considerably.

Bah. Freezing to death in an ice storm can also 'shorten your lifespan considerably'. There's damned good reason to want a simple, off grid means of heating your home if you lose power.

This is not the case when using modern wood burning stoves though right?

Apparently they're not as great as people have been thinking


Or wood burning fireplace inserts that draw combustion air from outside and expel the fumes/combustion byproducts outside as well. Not ideal for local air quality, but essentially eliminates indoor pollution from them.

Well sure, if you are the only person for miles that uses a fireplace.

Additionally, I think people who are outdoors still have to breathe air.

To that point -- last summer, my friends moved from a mountain town under constant threat of wildfire to a pleasant spot by a lake in the Midwest -- their new neighbor keeps a fire burning 24/7, so they still get to enjoy the terrible air quality all winter long.

Unless it does turn wood into a pure gas without a trace of sulfur, silica etc. it must emit such stuff. Or it comes with filters cleaning fine particles, nitric oxide etc. In the end it does emit CO2.

Wood pellets largely solve this.

the smoke goes up the chimney.

Most of the smoke goes up the chimney. You still open it to light, relight and sweep away the ash and in that time you're going to breathe in some smoke or dust particles which you ideally shouldn't

Mine's gone out...

1 meter wide stone walls, mortar, no foundations and wooden beams is what my family house in Spain is made of. It has been there for several hundred years with absolutely 0 structural remodeling, some cosmetic work has been done. Seeing the house will be there for few more years, not sure 1000 though.

If I am correct some of the Romanesque constructions don’t need wood and those have been there for 1000 years.

The seismic activity in Spain is almost null, which I guess matters for this structures.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact