MachinePix Weekly #55: Brian Potter, author of Construction Physics
MachinePix Weekly #55: Brian Potter, author of Construction Physics
Brian talks about the complexities of construction, his favorite buildings, and the scariest thing he's had to deal with on a construction site. This week's most popular post was a giant floodgate.
This week I sat down with Brian Potter, the author of Construction Physics. I first learned about Construction Physics when it was featured by Substack as in its weekly “What To Read” column.
Brian’s newsletter was particularly timely during lockdown with posts like “Lumber FAQ: Why wood has gotten so dang expensive”, but it as the progress studies pieces like “Why did agriculture mechanize and not construction?” that got me hooked.
Anyways, Brian was kind enough to help me demystify some questions I’ve always had about construction—and as part of this construction-themed week, he’s stepped in as a guest-editor on the Week in Review of @machinepix posts below.
The most popular post this week was a unique floodgate:
“Flood control technology always makes me think of Holland, where they've spent hundreds of years figuring out how to prevent their country from flooding, and have become experts at it. This floodgate is apparently based on the the Hagestein weir and lock complex, which is also a visor-type floodgate.”
—BP
I’m always looking for interesting people to interview, have anyone in mind?
—Kane
Interview with Brian Potter
I originally found your newsletter because it was featured in Substack’s What to Read—what made you decide to start writing about your experiences in construction?
So I used to work at a company called Katerra, which is mostly known now because of its massive bankruptcy [laughs]. But they were a construction startup with a really massive amount of funding from SoftBank, and they were coming in and saying “we're going to change the whole construction industry. It's incredibly inefficient, blah, blah, blah. But we're going to make buildings in factories the way we make everything else and not do them one at a time manually, which is a crazy 200 year old process.” And when I joined, I was thinking these guys have the proper theory of how to change the industry, and they have the money to do it.
But obviously it didn't work out. So I was like, okay, but why didn't it work out? And then as I started research, it's like, oh, there's been a lot of other efforts for a very similar idea: you know, really large industrialized buildings in factories, and none of those worked out either. What is it about this industry that is so resistant to these attempts to do things in a more mass produced, more efficient way. So I just started writing about trying to understand the industry kind of one little bit at a time and understand kind of how the parts fit together and how all the mechanisms and incentives work—and that became Construction Physics.
Construction, especially large commercial construction, sometimes feels like a blackbox with dozens of different parties. All these players—construction companies, contractors, subcontractors, engineers, etc—get thrown around. Can you help us understand the main players, and what your role was as a structural engineer?
How a building comes together is basically this: a real estate developer has an idea for a building, based on some belief in the market—home ownership is doing this, offices are doing this, demographics are doing this—and bet if they put a building here, it will sell. They get financing for it and then put a team together to build it.
There are two main players and a lot of sub players. One of the main players is the design team, which is led by the architect. The other is the construction team, which is led by the general contractor. The architect typically hires out the various specific designer roles, which are a bunch of different engineering disciplines: the structural engineer who designs the load bearing portion of the building, the mechanical engineer who designs the HVAC systems. There's a plumbing engineer, there's the electrical engineer, all that stuff—under the design team led by the architect. They put together the design documents for the building which are basically a list of what the building needs to be and instructions on how it goes together.
All those design documents are then given to the general contractor, who then goes and also hires their own little team of subs that are each going to do a specific task. So they'll hire somebody to just do the concrete or just run the wiring and kind of all that stuff. Sometimes there's some back and forth between the two teams. Sometimes it’s “you didn't show this” or “we’ve encountered these environmental conditions and thus we need to change this”. That's kind of how it roughly goes together.
Got it. When someone says construction company, I should think “general contractor”.
“Construction company” is almost always going to be somebody in the general contracting role, the leader of the team that does the building.
Architects almost always say that they're architects and engineers almost always say that they're engineers. Sometimes it gets a little mixed up because design-build companies that do both. And then sometimes developers are part of that and sometimes there's not. So there’s always a few exceptions and gray areas depending on the specific players.
I read a great piece in the New Yorker about Arup’s exceptional structural work. Which engineering or architectural firms are doing really exciting new work these days?
Arup’s great, but to be honest I don't follow the really high-end engineers and architects super closely. What I'm really most interested in is firms working on making the whole process more efficient.
The guys at the really, really high end are kind of the opposite of that, basically really pushing the envelope of stuff, which is in some ways orthogonal. Like Ferrari and Toyota, right? They're both car companies, the teams at Ferrari are really pushing certain envelopes in some ways, but not in the same ways that Toyota is with mass production.
Are there any companies you can think of pushing the envelope on processes and efficiency?
Yea—This company called Equilibrium that is focused on mass timber. Mass timber hasn't been used to its full potential with modern construction, and it’s really interesting to see the work they’re doing with it.
For the audience, which is generally technical, but not necessarily in construction: what is mass timber?
Sure. So mass timber is basically exactly what it sounds like. It's timber elements that are very large. In the US we do a lot of wood construction, but a lot of it is what's known as light framed timber, which is just like two by fours, two by sixes combined in various ways.
Mass timber has different properties. The biggest difference from a code perspective is that mass timber has very different combustion properties. Basically in a fire it'll form a char layer on the outside, and that will protect the inside of the wood from being burned. And so it can actually last a lot longer in a fire. Light frame timber is just too small, you really can't really rely on its fire resistance too much. And because of that, you can build higher with mass timber—much bigger, more densely occupied structures than you can with light framed wood. You can build a skyscraper with mass timber in a way that the code would not allow you to, and you would not want to anyway, with light frame wood just based on the fire risk.
Possibly naive question, but you said skyscraper. “Skyscraper” makes me think of steel; what is the calculus between choosing mass timber and steel?
Mass timber has environmental properties that are really nice: it’s a carbon sink. The relative energy it takes to produce steel versus how much went into growing the wood is also very favorable towards mass timber.
Also, people really like being surrounded by wood. There’s a reason furniture in your house is mostly wood. Even cheap stuff, perhaps particle board or actual cardboard, has a wood veneer, right? Not a fake metal finish. People just really liked the look of it and then the feel of it. Architects are really enthusiastic about mass timber for that reason.
There are trade-offs to consider in engineering with wood versus steel, but it’s not obvious that steel is strictly better even for large structures, and we’re just starting to explore mass timber.
Are there canonical mass timber buildings that we can look at and appreciate?
There's not like a ton in the US, it's a little bit more popular in Europe. It's actually interesting because if you go to like an old building, 1920s, 1930s, a lot of times you can find these like really big, beautiful timber members. It got superseded by cheaper materials and ways of building. As for new ones, there’s the Catalyst Building, a mass timber university building in Spokane, and it's really neat.
There's this one called T3, in Minneapolis and then I think they built one in Atlanta as well. A lot of tech companies are building mass timber offices as well. It’s starting to become much more popular.
How about a personal favorite building or structure?
This is one I've mentioned before, and I've actually written a whole essay about it. There's this temple complex in Japan called Ise Jingu, and it qualifies as a mass timber building because it's made from these really big, enormous logs.
But what's really interesting is this tradition where every 20 years they tear down and build an exact copy of the buildings. And they've been doing it for like over a thousand years. So you have these buildings that look very new, a brand new version of a 1,300 year old building that has existed continuously since the end of the Roman Empire, which is very neat to think about.
What are some common misconceptions about construction you often run into?
Technical people, but not necessarily in the construction world, often oversimplify the diagnosis of the problems in the construction industry—often I hear that we need to do this one thing, and that would solve the problem; but there’s this complex web of institutions and risk profiles and like logistics and even the physical size of things that all sort of interact in a way that makes it very difficult to improve on a process.
That's another thing: people think people are not actively trying to improve the way things are done; that construction is an industry that is content to just do things how they’ve always done it. But it's such a competitive industry that if you can come up with a reliable way of doing something more efficiently or in a way that saves money, people will be very very interested in that. It just turns out that it's very hard to do and the way to do it is not necessarily obvious.
What’s the craziest story you’re allowed to share from your time in industry?
This one is from the first job I had, where they were building a parking garage for a hospital and they had started construction. But then it came to light that the surveyor had sited the building wrong and it literally didn't fit. We had frantically redesign the rest of the building around this part that had been put up because it had been surveyed wrong.
There was this one project where I was working for a company that made these concrete slabs. As part of the way that they work, they were slightly cambered up when they had been set. That's just kind of part of the way that this product worked. It’s explained in our drawings and everybody who specifies things knows that this happens. But this architect didn't know that this happened and the developer was very upset. They insisted that we try to remove this upward bow in these slabs.
What we ended up having to do was drop these 10,000 pounds weights of grout on top of the slabs to try to get them to come down a little bit. These slabs had more steel reinforcement in them than strictly necessary for production efficiency reasons. What we had to do was, while this was weighted down, we had to go under the slabs and cut the reinforcing steel. And so we had guys going under the slab, they were cutting it, and the whole slab would kind of thump down. I had to tell them in the field exactly what reinforcements that they could cut and where they had to cut it. Like the bomb defusal scene in a movie.
But it worked. It brought the upward bow down a little bit. I think the client ended up still not super thrilled, but it wasn't because of anything that we had done, we did our best.
You mention a lot of fine details like camber in concrete slabs. Do you have fun tips for newbies on looking at buildings with a professional, critical eye? What do professionals look for that we may miss?
I like to see how much I can tell about the history of the building just based on what you can see from the outside. Structural systems or building systems kind of go in and out of style.A lot of times you can kind of get an idea of when it was built, just based on what they used to build it.
A good example is concrete. Now the things they use for parking garages are called double tees, which is a concrete slab and then two vertical webs. And that makes the floor of the parking garage. For a ten to fifteen year window in the sixties, it was very popular for everything. And so a lot of times you can see those things being used for things like factory walls. If you see a manufacturing building or something like that with concrete walls, but they have these weird little wings on the outside, it's because it was built with this precast system and you can basically pinpoint the date as to when it was built just on that.
There's a bunch of things like that. Nowadays brick veneer, especially for a single family home, is pretty much in a few spare spots because the labor to do it is so expensive. You typically see it on the front and not really on the sides, but as you go back, you can see how expensive labor was by how much brick is in a building and how ornate it is. The farther back you go, the more brick it is. Really old stuff from the twenties and thirties can have like this really beautiful, ornate masonry work that you never see today just because it's so expensive to kind of get that stuff.
I see a ton of startups trying to tackle construction now; are there any particular opportunities that you’re personally excited about?
Construction is really, really constrained by how far you can ship something efficiently in a single day. A lot of times you're limited to basically like a 300 to 500 mile radius and that has a lot of downstream effects. If you can only ship 500 miles away, the size of the market you can serve is limited, which limits how big your factory can be, which limits how efficient it can produce things.
This isn’t necessarily directly related to construction, I'm really curious to see if breakthroughs in logistics, things like self-driving cars or like electric cars, can change the economics of logistics and expand the shipping radius. If that limit gets stretched out, I suspect that we'll have some downstream effects on the construction industry.
What was your favorite piece to write for Construction Physics?
The one that I wrote about Katerra was nice because there really just wasn't that much information available about what happened outside of the company. I wrote my perspective from having worked there. I ended up getting just like a ton of messages and feedback from people who had also worked there that basically said, “I'm so glad that somebody wrote down what it was like there, because so much of the reporting that you heard about it from outside just didn't reflect what it was like there at all.”
And now that it’s not a growing concern, it’s all ephemeral. All the information could have disappeared outside of people’s memories of it. So I'm glad there's something that describes what it was like there and what we were trying to do.
Any side projects you’re working on right now?
The Construction Physics newsletter takes most of my hobby time!
Any favorite books or books you’re reading now?
I read a lot of books just for the newsletter. So I recently read this interesting book, Neighborhood Defenders, which is basically the people that go to these zoning board meetings to protest against any sort of new housing or any sort of construction development. And it's a relatively small, specific group of people, but they have an outsized effect on what gets built in the country because these zoning boards, if you go and make a complaint to them, they're going to consider what you say even if it’s not sensical.
Just by going to these things and showing up and saying, “I’m worried about parking, I’m worried about the environment it's going to affect my views of the lake” or whatever, they really can end up shaping a lot of what gets built and really make it a lot harder to build like say a multi-family apartment development, which, ironically, is often much more environmentally efficient. The book is an interesting study of how this all works and who are the ones that show up to these meetings and what do we know?
What’s your favorite simple (or not so simple) tool or hack that you think is under-appreciated?
I have a lot of issues with my joints. I've found that if you need a brace for a finger or thumb, by far the best thing to use is a roll of blue painter's tape. Just wrap it around your finger because it's the exact right thickness to brace it in a way that will allow you certain freedom of movement that works by far better than anything else.
The Week in Review
“I didn't know about this bridge, but it'll be the longest suspension bridge in the world when it's complete. Its main span is 2,400 feet longer than the Golden Gate Bridge's.”
—BP
“Most pools are made out of concrete. Myrtha's are interesting in that they make all theirs (even the permanent ones) out of stainless steel. I wonder how often these portable competition ones get re-used.”
—BP
“Bricklaying robots are super interesting! Bricklaying is maybe the simplest task that has resisted being mechanized (almost all brick and masonry is still set by hand). I wrote a whole article trying to work out why.”
—BP
Postscript
I learned that airline pilots are very opinionated about different commercial aircraft:
I am always looking to connect with interesting people and learn about interesting machines—reach out.
—Kane