Fixing the Railroads
cause nothing lasts forever, even cold November trains
Leading Off: New York City is (slowly) banning natural gas hookups in new construction. This type of restriction may be necessary, but is not obviously necessary1. Heat pumps are significantly more than 100% effective2 at moderate temperatures, but require substantial electricity at low temperatures. They also don’t burn natural gas directly, and thus can use carbon-neutral energy sources easier.
During some recent furnace maintenance at my house, I asked about installing a heat pump, and was told I would need to hire an electrical contractor to increase the maximum electricity draw first — and also that my furnace had 10 years of depreciation left before it made sense to replace.
One of the ideas I’ve been pondering for 2022 is performatively going truly carbon-neutral. That will not happen: I will still be burning natural gas in my house, and I will not be buying fake carbon credits. It would be nice if somebody else was attempting to be performatively carbon-neutral — if someone is doing this and has a blog, let me know and I will promote it.
Some Good Ideas: I assume most of you know about Astral Codex Ten (formerly Slate Star Codex), and some of you enjoy its normal fare more than others. Yesterday, a very interesting collection of funding grants was announced. If you haven’t done so already, you should take a look.
Tweet of the Week:
Was it really a year ago that the Bean Dad tweet captured Twitter mindshare for a few days in early January?
High Speed Rail - why doesn’t America have it?
The most important thing about high-speed rail (which is often forgotten) is that you can't run slow trains and fast trains on the same track. The second most important thing (which is also forgotten) is that railroads are expensive.
❓ A freight train leaves Omaha at 3PM traveling west towards Denver at 59 MPH. At 3:45PM, a passenger train leaves Omaha towards Denver on the same track at 125 MPH. When do the trains crash into each other?
Most of the American rail system is used for freight, and that freight is more economically moved at 59 MPH than 125 MPH. Without at least three tracks3, it will simply be impossible to move trains that quickly. Most of the US rail system has either two tracks or one track — if you have sidings to allow trains to pass you only need one rail for service which is slow and infrequent enough.
So, we need to build out a substantial capital expense for additional rail lines. We also will probably need to pay to electrify the rail lines — in general HSR lines do not use diesel locomotives, and we want to go carbon-neutral anyway. And, most likely, rather than trying to make due with a three-track system, the railroad company would go for a four-track system: two sets of train lines, two separate rights-of-way, with different curve banking and track quality.
And creating a new right-of-way is somewhere between “slow and expensive” and “impossible”. The original rail right-of-ways were created when nobody lived there (or rather, nobody that the railroads and the US government cared about lived there). Today, it would involve dealing with hundreds of individual land owners, many of whom would be financially or politically motivated to delay or prevent the project.
In short: as far as the work needed to have a high-speed line between Omaha and Denver, approximately none of it is done.
What would it cost?
The corridor between Omaha and Denver certainly will be one of the least expensive corridors - the terrain is easy and the intermediate cities minimal. Other countries manage to build HSR for $25-40 million per mile. I will assume $75 million4 per mile. That makes this corridor cost roughly $40 billion in initial construction costs. To keep the numbers simple, I will assume a 40 year depreciation schedule. In other words, it costs (roughly) $1 billion per year to operate high speed rail here.
In the 12 months from September 2018 to August 2019, there were roughly 300,000 air passengers between OMA and DEN. As a best-case scenario, I will assume this rail route would have twice that traffic - 600,000 passengers per year. And that number is uni-directional: there are 600,000 passengers going the other way as well.
This leaves us with a high-speed rail cost of $833 per passenger (one-way), not accounting for the expenses associated with the rolling stock. This must be compared to an air route where tickets are around $100 (one-way) for a trip that is faster.
You can imagine more traffic if this line is part of a network — although unless airplanes are completely prohibited I doubt there will be substantial demand for a 14 hour train trip between Chicago and Los Angeles. You can imagine cheaper costs. You can imagine population growth along the corridor.
In the realm of reality, this route simply doesn’t work. Other routes may be more feasible. Milwaukee-Chicago-Indianapolis, for example, might work. But the dreams of a trans-continental electric high-speed rail system are just dreams.
Batteries will never solve the problem of seasonal variation in power usage and production. Chemical batteries certainly will not, and I am skeptical that batteries powered by moving rocks up and down will be a solution. Storing chemical fuels at least has the possibility of doing so. And there are many reason to prefer to convert surplus electricity to methane rather than hydrogen.
An electric heater is considered 100% efficient - 100% of the electricity is turned into heat. A heat pump (which is basically operating an air-conditioner in reverse) is more than 100% efficient - you get substantially more heat (indoors) for the electricity used.
I don’t know what computer game will let you manage trains of varying speeds, but there probably is one. There are a lot of train simulators that will let you create “signal block” systems - if you don’t want to play a “train game” try Factorio. In short - you can coordinate passing on a bi-directional track. It’s not trivial, but it can be done safely.
Initially, this said $100 million - but I did the math with DEN-OMA being 400 miles, not 540. Regardless, the cost will be higher than in other countries — but not as expensive as upgrading parts of the Northeast Corridor would be.