Hacker News Reader

Comment by CalChris | Hacker News Reader

> a repeat of the BART mistake.

Probably a different mistake but BART uses 5'6" rail gauge. Standard US rail gauge is 4'8.5". This means that every railcar, every everything, has to be custom made with no economies of scale.

No one can remember why they did this.

View on HN

> BART uses 5'6" rail gauge. Standard US rail gauge is 4'8.5".

Interesting. When I saw this, I wondered if it was to match another standard gauge that was judged to be superior. But, looking at the list of rail gauges on wikipedia[0], 5'6" is the gauge for railways in India ("Indian broad gauge" to be specific). The initial BART rolling stock seems to have been build by Rohr, Inc.[1], which is based in California. So it seems unlikely that they were benefiting from any economies of scale from Indian rail construction. Rohr also seems to have built rolling stock for the Washington, D.C. metro, which uses a near-standard 4'8.25" gauge[2].

Perplexing.

[0] https://en.wikipedia.org/wiki/Track_gauge

[1] https://en.wikipedia.org/wiki/Rohr,_Inc. (you'll need manually add the period to the end of the URL, because of how HN converts this into a link)

[2] https://en.wikipedia.org/wiki/Washington_Metro#System

if, 150+ years ago, when we were building all the rail infrastructure in India, someone said - that there would be no fast line from Delhi to London in 100 years - people would think you were nuts - right now it's tricky to get from London to Berlin, India is near impossible

It most likely would have happened, if the airplane hadn't been invented. Now you can fly from Delhi to London in 8-9 hours.

There is the theory that Rohr was an aerospace company and didn't know about standards; I find this hard to believe. Some say there was a study about car dynamics. But the article I read said that no one actually remembers. The design docs are in a transportation library in Monterey and they apparently don't say. Here's a collection of theories, all plausible with no confirming evidence:

http://myprogressiverailroading.com/rail_forum/maintenance_o...

BTW, Pittsburg uses 5'6" as well.

DC Metro is even more ridiculous. It's only 1/4" different from standard gauge.

>https://en.wikipedia.org/wiki/Rohr,_Inc. (you'll need manually add the period to the end of the URL, because of how HN converts this into a link)

FYI, the URL encoding for "." is "%2E", so replacing period with %2E works: https://en.wikipedia.org/wiki/Rohr,_Inc%2E

It is pretty common knowledge why they did this. Original plans for the BART had it traveling over the Golden Gate Bridge into Marin County. Due to high winds across the bridge, they wanted a wider gauge to prevent trains from tipping.

Yes, this is a popular theory but there's no evidence in the fossil record.

Is there evidence in any direction? This really sounds like the most plausible hypothesis, to me at least.

With strange, custom-construction-requiring stuff like this, I tend to wonder how much of it would be explained by "help self/a buddy get rich". It's practically guaranteeing that there's going to be a single company supporting it for all time, which is a potentially-extremely-lucrative business deal.

If that's the case, obfuscating the reasons for these decisions is a rational move.

It's a bit cynical to say this but running HSR on an incompatible network protects it against future politicians who may seek to sabotage it by merging networks.

This sounds similar to the reason why the Glasgow Subway runs on a 4' (1219mm) gauge: to stop it from being taken over by the numerous (at the time privately owned) railways around the city which were all standard gauge.

Marin was not a participating county, and that was established long before gauge would have been chosen.

Given standard gauge trains transverse plenty of bridges which frequently have high-winds, that seems… dubious.

Wtf is happening?

I want to believe that the different gauge standards is a bit like the proliferation of programming languages in software - they solve their slightly different problem space well. Is this true?

Reply on news.ycombinator.com