briancp

As the article mentions, densified wood already exists as a product [1]. But it's basically never used for building materials. As far as I can tell it's main use-case is for electrical transformer support pads and anchorage, as you need a strong, non-conductive material for that. So right now it's a fairly niche product.

I'm not an expert, but as far as I know existing densified wood is basically stronger by virtue of cramming more material closer together - the wood fibers themselves don't change mechanical properties at all, there's just more of them in a given area. So your strength increases are paired with an increase mass as well. If this actually increased the strength of the wood fibers, that would be interesting.

It's an interesting technique, but we already have processed wood products that offer significant strength gains over natural lumber [2], and they haven't displaced regular wood all that much. Normal sawn lumber is often strong enough, and it's just incredibly inexpensive. So it's unclear how much difference another stronger, processed wood product would make to the building market (I can't speak to other uses like body armor or vehicles).

I think long term, the most potential for wood improvement lies with improved trees that grow stronger, faster, and with fewer natural defects. Most agricultural products have been deliberately bred for desirable traits for thousands of years, but we've only just begun to do this with trees. (I wrote more about this here: https://constructionphysics.substack.com/p/stronger-faster-c... )

(Source: structural engineer, formerly worked at a mass-timber focused construction startup).

[1] https://www.roechling-industrial.com/us/products/composites/... [2] - https://www.mjbwood.com/lsl-lumber/#:~:text=LSL%20Lumber%20(....

Posted this in another thread, but there’s some neat photos from the 60s of the telescope under construction here: https://www.naic.edu/history_gal/historicgal.html

Beyond the scientific value, it is (was) just a neat feat of engineering getting this built in the middle of the jungle.

I was in St. Simons a few months ago, and was able to see this ship first hand. It’s enormous, sitting on its side in what looks like just a few feet of water. It’s close to the shore, maybe a thousand or 1500 feet out. A temporary information placard had been put on the fishing pier explaining what it was.

At the time it was all in one piece, but you could see them attaching the lifting equipment along the length, and it was clear it was going to be cut into pieces.

It must take an enormous amount of work to build an internal framework that can support the cut section being lifted from the side. I’m not surprised it’s taking so long so salvage.

A shame. Beyond the loss to science, the equipment itself was impressive. A 900-ton observation platform is like a multistory building suspended a few hundred feet in the air, in the middle of the jungle.

An interesting gallery of the telescope being constructed is here: https://www.naic.edu/history_gal/historicgal.html

This is really interesting, like a computer algebra system[1] but for geometry. Seems like it could be extended to cover the non-shape based documentation (material properties, design loads, etc) as well.

One thing I've never seen is optimization techniques paired up with any sort of in-depth cost data. It's usually based on naïve extrapolations based on total material weight or volume, and doesn't take into account things like crane costs (a function of max component weight and building extents) local material availability, component fixed costs, number of connections, logistics costs, etc. There's a ton of low-hanging fruit in design optimization that you could pick before you even had to reach for AI-based methods, though the AI methods are obviously the long term future of the space.

[1] http://www.math.wpi.edu/IQP/BVCalcHist/calc5.html#:~:text=A%....

The article doesn’t go into much detail, but it hints at the use of a technique that I’ve suspected might have some promise for building design - finding a way to convert a 2D set of drawings or a 3D model to a text-based representation, then using text-based AI tools like GPT-3 to generate new buildings based on input parameters.

Currently the building design process is time consuming and labor intensive, I think there’s lots of potential for automating large swaths of it.

I work as a structural engineer and write a newsletter about, among other things, technology for lowering construction costs.

A tiny house will be cheaper than conventional housing by virtue of being a lot smaller (building costs are usually expressed per square foot or square meter). But as the article points out, these are usually built with fairly high end finishes (wood paneling, granite counters, etc) - their cost per unit area is actually somewhat high. Actual low-cost construction, like the sort used on mobile homes, eliminates most of these expensive finishes, because they’re a major cost driver.

And in the US at least, a house not permanently attached to land has to be financed as property, like a car, rather than real estate, which means the tiny houses attached to trailers will be financed at a much higher interest rate than a traditional mortgage.

Generally I think these are more of a niche lifestyle product, a mobile home for people who don’t want to buy a mobile home, more than any sort of affordable housing solution.

Interesting, it's sort of an inverse desire path[1] - instead of locating where foot traffic is the highest, you locate where microbe traffic is the lowest.

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

Since it’s made of unreinforced concrete, getting it permitted in almost any jurisdiction in the world would be exceedingly difficult. Which is somewhat ironic, as it’s the lack of corrodable reinforcing that’s enabled it’s extremely long lifespan.

This is super interesting, I'd love to hear more about the systems you were using if you're up for discussing. briancpotter@gmail.com

Background: structural engineer, have worked with prefab construction in various capacities

The short answer to this question is "because we haven't figured out how to make it less expensive than regular site built construction".

The longer answer: prefab, or modular, construction has made inroads in various construction niches. Parking garages are nearly all prefabricated concrete for instance. Light framed buildings (either wood or cold formed steel) frequently have their structure panelized off site. Hotels (especially in places where the cold weather means a short construction season) are sometimes built using prefab modules.

It's also super common to have prefabricated lower-level components - wood trusses, bar joists, hollowcore panels, etc.

But any higher degree of prefabrication tends to be more expensive than conventional, not less. They mostly get used when their perceived benefits outweigh their costs. (If you look at a panelizer like Entekra's benefits page, they don't list "cheaper" anywhere). Two major exceptions to this are parking garages and HUD homes, which (among other reasons) are heavily optimized for a particular form factor.

Regarding Bone Structure, it's an interesting system, and I love seeing people tackle this problem space. But I rate the chances of a custom fabricated steel structure ever being less expensive than conventional wood framing as "very near zero". This seems limited to the high end custom home market, and their marketing material seems to indicate that as well.