Tuesday, January 17, 2017

Railroads and the Flat Earth

One of Eric Dubay's "200 Proofs"—well actually four because Dubay played some misleading games with his numbers—is that railway lines are constructed without the Earth's curvature being taken into account. This is obviously devastating proof that the Earth does not curve away. Right?

Let's do a little thought experiment. You can try this in real life if you have the space. Imagine you have an N-gauge model railroad. The tracks are 9mm apart, and the scale is either 148:1 or 160:1. Let's take the latter figure to save some space.

We're going to lay a straight track for one scale mile. As I said, we need a big space, because this track is going to be 33 feet long. Now, let's pretend that this space is not a perfectly flat floor but a scale model of a mile of curved Earth, and flat, that is, the same elevation above mean sea level throughout. How much lower would the track be at the end of the run than the beginning?

Well, the eight-inch-per-mile squared formula, which I wrote about a bit ago, is a fine rule of thumb for short distances, so we'll go with that. Eight divided by 160 is, conveniently (another reason I chose 160:1) .05 inches, or 1/20th of an inch.

Now, be honest with yourself. Is 1/20th of an inch over 33 feet going to make enough difference to your train to compensate for, even if it was actually downhill (which it's not in full scale—more in a moment)? If fact, even if the track dropped 1/20 of an inch for every foot, or rose 1/20th of an inch for every foot, the train would have no issue with it.

And I'd love to have your carpenter if your floor is level to within 1/20th of an inch over 33 feet. I know my level can't detect that.

Ah, says the flat-Earther, but it accumulates. Over miles and miles and miles, the curve is steeper and steeper. Well, no it's not. At any given point on the track, this mythical absolutely flat track (I really think that people who make this argument have never taken a trip on a train, especially along the west coast of the US), the curve ahead would be exactly the same. It's not as if the train traverses the entire track all at once.

And then (I told you I would get to this), at full scale none of this matters. Because "flat" is meaningless for train construction. If you want the bed to be flat as far as the train is concerned, you need to stay level, that is, perpendicular to plumb, to the center of gravity. To the center of the Earth.

Of course, flat-Earthers say that gravity just doesn't exist. But that is another topic.


  1. So on what projects do surveyors take curvature into account?

    1. On projects where is actually matters. Geodetic surveys, by definition. Long-span suspension bridges need to take curvature into account. I'm sure an engineer could provide other examples. But most projects just don't need to be concerned about it.

  2. Google the LIGO gravity wave detector. Curvature of the earth had dramatic effects on its construction.

  3. Also, Railroads use actual survey data as input to plan, the local topography is vastly more variable than curvature anyway. Since the curvature is gravity equipotential it is neither up nor down - so the only affect on the rail is length and that is measured along the actual ground-plan so no 'curvature' adjustment is required since that is already in the data.

    It's a complete boondoggle - as usual for Flat Earth


Note: Only a member of this blog may post a comment.