Comment by jcranmer | Hacker News Reader

Comment by jcranmer | Hacker News Reader

Parent Full thread

> No other civilized country in the world has such lousy transportation infrastructure. Especially trains. The U.S. should have bullet trains going everywhere, but no.

It depends on what metric you use. The US freight rail system is the envy of the world, for example.

The reality is that HSR isn't cost-effective if the trip takes more than about 5 hours. When you superimpose that time limit on the US, there's rather few viable HSR routes. A Chicago/NYC run is at that very margin of viability--and compared to most European countries, that's a longer trip than travelling from one border to another. It's quite easy to set up routes in UK or in Germany that hit all of the major cities in fairly short order, but that task is impossible in the US. Especially grating is the lack of collinearity between major cities; the Midwest in particular ends up having to be a star topology, since Chicago-X doesn't include multiple cities, and there's no route useful enough to be worth bypassing Chicago.

> The reality is that HSR isn't cost-effective if the trip takes more than about 5 hours.

Can you expand on that? It's not obvious to me.

It’s certainly not cost-effective on a relative basis when compared to air travel. The Delta Shuttle runs Chicago-NY-DC every hour on the half hour for a song. Hard to compete with that on rail.

With rail you do get the benefit of landing in the middle of the city rather than an airport 30 mins from city center, but it’s still far less expensive in both time and money to fly when the trip exceeds about 4 or 5 hours.

And, indeed, high-speed rail isn't very price competitive with budget air a lot of the time either. Doesn't mean it's a bad idea; transporting a lot of people who are willing to pay a premium off the roads and out of airports may well be a net win. But understand that it's premium travel.

Several reasons:

* Airplanes travel about twice as fast as top-speed HSR. By 5 hours, even assuming generous airplane penalties, there's no speed advantage to HSR.

* For business travel, you've already lost the entire working day at that timepoint, so downtown-to-downtown isn't as much of an advantage anymore.

* You start to have to have nasty scheduling times for intermediate stops. Having to disembark at 3:00 AM is not going to be conducive to ridership.

* Sleeping cars and full-service dining start to become necessary around that point in time, and that's really going to eat into your operating costs.

I drove the number of 5 hours largely on the meal concern, but it does seem that most planning exercises tend to profitability wall closer to the 3 hour mark.

The US is 2680 miles across, and the minimum speed for new high speed rail is 160MPH. So, it would take less than 16.8 hours for a non stop high speed to go coast to coast.

In other words, the 5 hour rule of thumb predicts that a modern system where the coastal cities were densely connected and there were a few money-losing lines that crossed unpopulated regions would be wildly profitable.

State-of-the-art high speed rail (like in China) would be even faster, and make even more sense.

It's unlikely that you'd be able to push trainsets above 220mph or 350 km/h in revenue service. Even then, the actual average speed for a Chicago-NYC run would be closer to 160mph, since you'd have to make some stops along the way, and there's no way you're maintaining 220 in the Appalachians. The shortest plausible route would be essentially the turnpike route (parallel the NJ, PA, OH, and IN turnpikes in a NYC-Philly-Pittsburgh-Cleveland-Chicago route), which is going to run about 800 miles or so--or about 5 hours. More importantly, you can't shrink the project to a smaller viable project--not connecting Chicago and NYC is going to give you a losing line, and not hitting that top speed is going to be too slow.

And Chicago/NYC is the "easy" connection. Atlanta, Texas Triangle, and Florida are simply too far away, let alone trying to reach across the Plains to hit Denver, Phoenix, and Las Vegas en route to LA/SF.

Well in other places, the HSR just tunnels through the mountains. Rome to Bologna, which is a comparablyish distance to Harrisburg to Pittsburgh is a good example. Highish speed rail already exists between NYC and Harrisburg and Pittsburgh-Cleveland-Toledo-Chicago is comparably easy (especially if you can use the already-electrified SSL tracks to get in Chicago).

By my count, a NYC-EWR-PHL-HAR-PIT-CLE-TOL-SBN-CHI train would directly serve metros with ~50 million people (and indirectly serve much of the rest of the northeast and midwest).

Realistically, your problems start between roughly Youngstown and Harrisburg. Between Harrisburg and Altoona, the terrain is ridge-and-valley, which means the tradeoff is twisting to follow a water gap or punching a small tunnel through the ridge. West of Altoona, the dissected Allegheny Plateau begins, but you have to first make an 800-1000 foot climb to reach the plateau, which means in practice a ~7mi, 2% grade tunnel. From thence to around Youngstown, you're on a dissected plateau, which means every water course (for whom the idea of a straight line is anathema) has carved a gorge. It's a rather different situation from just tunneling through a massif.

Sure, you can fix issues with tunnels and bridges. But that doesn't come cheap. And what is by far the most expensive portion of the line is the part that's necessary for any sort of traction, which greatly increases the chances of the entire project being a white elephant (keeping in mind, again, that even in the best-case scenario, you're still at the very margin of being able to run a profit). The economics look worse than the Turin-Lyon HSR, which is already criticized as an expensive, unnecessary boondoggle.

Reply on news.ycombinator.com