Reading List 12/14/24
Japanese semiconductor manufacturing, EV reliability, indoor air quality, mall to residential conversions, and more.
Welcome to the Reading List, a weekly roundup of news and links about buildings, infrastructure, and industrial technology. This week we look at Japanese semiconductor manufacturing, EV reliability, indoor air quality, mall to residential conversions, and more. Roughly 2/3rds of the content is paywalled, so for full access become a paid subscriber.
Other items of note:
I’m interested in talking to some architects and developers about how they decide on the design of the buildings they work on. If you’re an architect or a developer and would be interested in chatting, please send me an email at briancpotter@gmail.com
If you’re a paid subscriber and want to suggest a topic for a future newsletter, feel free to do so in the comments. No guarantees I’ll select it, but lots of previous posts have been written due to suggestions from folks.
How many full-time writers can Substack support?
As a writer on Substack, it’s of obvious interest to me what the economics of the platform are, and how successful someone can expect to be on it. One measure of this is: How many people earn enough writing on Substack that it could be their full-time job?
One way of estimating this is to use Substack’s leaderboards, which rank the most popular Substacks across a range of different categories, and count how many publications have at least a thousand paid subscribers. Many Substacks charge on the order of $5-6 a month, which at 1,000 paid subscribers is $60-70,000 a year. This is higher than the median individual income in the US (~$37,000), but not that much higher, once you take into account the higher taxes from being self-employed and Stripe/Substack’s cut. (Because of the way Substack’s leaderboards work, it will also include folks with fewer than 1,000 subscribers who charge substantially more per month.)
Counting up the publications yields about 700 publications total:
This might underestimate how many people Substack can support full-time. For one, the most popular Substacks make millions of dollars a year, enough to support several full-time writers. For another, many people can support themselves making less than $60,000 a year, and I suspect the tails of these distributions might be quite fat, with lots of folks making not that much less than that. On the other hand, it might also overestimate them, if writers live in very high-cost areas, or have lucrative careers, and would only pursue writing full-time at higher levels of income. But it’s probably in the ballpark of the true number.
Japanese semiconductor manufacturing
The international competition at the bleeding edge of semiconductor manufacturing tends to focus on the US, Taiwan, Korea, and China, but Japan is apparently eager to stay in the race as well. It’s pouring billions of dollars into a company called Rapidus, which aims to escape the brutal capital costs of a modern fab using what seems to be a more Lean-inspired, one-piece continuous flow manufacturing method that minimizes inventory and cycle time. From the Financial Times:
At the heart of the Rapidus project is an attempt to prove that bespoke chips can be efficiently and profitably produced in small quantities rather than large batches, an idea that overturns the received wisdom in advanced semiconductor manufacturing…
In the three years since that final Zoom call, Rapidus has leapt from concept to reality. A massively expensive plant is rising from the forests of Hokkaido, 900km north of Rapidus’s headquarters in Tokyo…
Today, Rapidus and Japan’s hope of success rests on two highly contested propositions. The first is that the surging AI market means that there will be sufficient demand from smaller customers for customisable special-use chips — bespoke designs that prioritise efficiency and can outperform more generic chips, such as those produced by Nvidia, in specific tasks…
The second, more controversial, bet is that it can reject the core industry premise of large-scale batch manufacturing — printing hundreds of wafers at the same time — in favour of a much quicker single-wafer process…
Rapidus will, says Koike, boast the “world’s shortest total cycle-time,” meaning the total amount of time it takes to process a wafer in a fabrication plant. And he thinks he can hit yield rates of up to 90 per cent within a year.
Bloomberg also has an article on Japan’s broader semiconductor push, which includes a new TSMC plant:
These nascent chip cities are leading a revival in a country that once dominated the industry — and aims to do so again. In 1989, the late Shintaro Ishihara, author and future Tokyo governor, urged Japan to “move at least five years ahead of other nations.” Instead, it stumbled due to US trade pressure and missteps during an industry transition. As South Korea and Taiwan took over, Japan’s share fell to less than 10%.
Drilling technology books
This week’s newsletter was all about the evolution of PDC bits for oil and gas (and now geothermal) drilling. If you’re interested in reading more about oil and gas technology, there are a few good books on the topic. One is Groundbreakers, which is sort of an encyclopedia of oil and gas technological advances, structured as a chronological narrative. It’s a great high-level survey of the development of oil and gas technology, though no individual entry is particularly deep (the PDC bits entry, for instance, is only about 3 pages long).
For a deeper treatment, check out Roughness, Rock Bits and Rigs, which chronicles the evolution of oil drilling technology in Alberta from 1884 to 1970. This is a much deeper, more thorough academic book that seems to give a lot of detail on the advances (because it ends right before the PDC era, I only skimmed this for information on pre-PDC roller-cone bits).
And of course, no mention of oil technology books would be complete without mentioning The Prize, Daniel Yergin’s magisterial history of the oil industry.
Other advances in drilling technology
Every revolutionary technology ultimately risks being superseded by an even better technology, and PDC bits are no different. Even as they continue to improve, there’s a variety of investigations into alternative drilling technologies, including energy-based methods that don’t use drill bits at all. Via this BBC article on enhanced geothermal:
…Quaise Energy, a spin-off from the Massachusetts Institute of Technology (MIT), for example, are aiming to drill holes as deep as 12 miles (20km) to access temperatures of 500C (932F) or more. To do so, they are turning to a tool that draws on years of research into nuclear fusion power. "While others are putting shovels in the ground, we're putting microwaves in the ground for the first time," says the company's co-founder Matt Houde.
He and his colleagues are experimenting with millimetre-wave directed energy beams that vaporise even the hardest rock. It focuses [a] high-powered beam of radiation similar to microwaves but at a higher frequency onto a segment of rock, heating it up to 3,000C (5,432F) so that it melts and vaporises. By directing the beam so it bores through the rock, holes can be created without the debris and friction created by traditional drilling techniques…
Slovakia-based GA Drilling, meanwhile, is exploring a different high-energy drill technology to bore into the Earth's crust. It is using a pulse plasma drill, based on very short high energy electric discharges that disintegrate rock without causing it to melt. This avoids creating any viscous molten rock, which can be difficult to remove and can stop drill bits penetrating further. "Since the process is very swift with short shocks crumbling the rock, there isn't time for melt to form – so the need to pull up and replace the bit is greatly reduced," says Igor Kocis, chief executive and chairman of GA Drilling. "Five to eight kilometres (3-5 miles) is a target for our current development programme – and later 10km-plus," he adds. "These depths will allow nearly universal access to geothermal power."
This report from the Cascade Institute gives some more details about energy-based drilling technologies, and other potential technologies that could close the gaps and make enhanced geothermal practical.
EV reliability
One minor mystery of EVs is that while theoretically you might expect them to be much lower maintenance than ICE cars (due to vastly fewer moving parts), in practice this hasn’t been the case. Earlier this year, for instance, Hertz ditched 20,000 EVs due to their high repair costs. This is only a minor mystery because you can probably chalk it up to the relative maturity of the two technologies, but it’s nevertheless interesting.
But the maintenance gap, while still substantial, seems to be closing. From E&E News:
The group — which surveyed owners of cars from the past three model years — found that EVs have 42 percent more problems than gas cars. That's a sharp decrease from last year's survey, which found 79 percent more problems in EVs.
“What we see with every type of technology and vehicle and manufacturer is that it just takes some time to get the bugs out and work out the new processes,” said Jake Fisher, senior director of auto testing at Consumer Reports. “The more time and history the manufacturer has with [electric vehicles], the more the reliability will improve.”
…Problems are more common in EVs because they “are very often packaged with tons of other electronics and other new technologies," Fisher said. "Some of these EVs are from very new manufacturers that have growing pains."