## 4 thoughts on “Will The Hyperloop Work?”

1. George Turner

I did a cursory check of using concrete pipe, and an 800-mile length (LA to SF and back) of 3 meter diameter concrete pipe (top grade), would cost about \$6 billion. 2 meter diameter comes in at about half that.

My thought was that it might be cheaper to essentially build a road type surface and have the pipe lay on top, on shock mounts, so Earthquakes merely move the roadway under the pipe.

Then I did another check for accelerations due to path deflections, which I learned to do when I was looking at elevated-tower space-launch schemes, where wind loads would slowly sway the launch tube, yet the passengers are having to track those sways at mach 15 or so, potentially inflicting fatal (neck snapping) sideways G-loads. But those sways were pretty large.

So assuming the hyperloop vehicle is traveling at about 700 mph and the individual tubes are 100 feet long, a 1″ deviation (probably expected from thermal expansions, frost heaving, etc) would produce 0.5 G’s, and that’s after smoothing (I’m ignoring jerk and just going with s=1/2at^2). The pulse rate would be about 10 hertz. So the vehicles will definitely need a shock absorbing suspension system.

As an aside, when you take the velocity up from 140 mph (high speed rail) to 700 mph (a five fold increase), the transit time when you cross a shifted segment drops by 5, and since a=2s/t^2, the side accelerations go up 25-fold. Compared to normal rail traffic traveling at 70 mph, the effect is magnified by a hundred. I bring this up because the temptation would be to argue that since such shifts don’t seem to affect normal rail, they wouldn’t affect the hyperloop, either.

Anyway, the length of travel of the vehicle suspension would largely determine the maximum allowed pipe deviation during and after an earthquake. It’s not that the shock mountings just need to dissipate vibration and force on a static member, they have to keep the total deflections within tight limits because a near-supersonic vehicle has to track along the shaft without suffering structural damage or subjecting passengers to dangerous jerk or G-loads. Although, perhaps during an Earthquake some amount of roughing people up would be permitted.

However, considering the extreme weight of the pipe sections, if real-time dynamic control is required to reduce deviations, and if this is required on the joints throughout the entire route, then the system might be unaffordable in the extreme. It would take high power, fast, dynamic motion control with laser positioning references (can’t trust the local terrain, only the the measured relative locations of the pipe segments), G-sensors (which are thankfully cheap), and backup emergency power, because the system isn’t used until a massive Earthquake knocks out the grid.

If more natural position damping can work (such as horizontal bearings and counterweights) then maybe the whole issue goes away as long as resonant frequency swaying doesn’t occur.

If someone feels up to it, modeling some paths through a series of deviations might be in order. I’m inclined to think the vehicle would require lifting pads on all sides, though, because if you hit an up-bounce, the vehicle can’t be allowed to then make uncontrolled or undamped contact with the walls or ceiling.

Added thought: Another approach might be to make the pads into lightweight rings with spokes, one in front of the vehicle and one behind, and then have the vehicle shock mounted to the pads’ central hubs. The pads could then maintain close wall tolerances without passing significant jerk to the vehicle itself, and the vehicle would use it’s smaller diameter both to let air pass on by and use that distance to absorb any major irregularities or shifts in the pipe.

2. Herp McDerp

As an aside, when you take the velocity up from 140 mph (high speed rail) to 700 mph (a five fold increase), the transit time when you cross a shifted segment drops by 5, and since a=2s/t^2, the side accelerations go up 25-fold.

Except for the coolness factor, I’m not sure I understand why Musk keeps talking about pushing the vehicles to near-sonic speeds. The engineering challenges would seem to be a lot more manageable if he reduced the velocity to “merely” 400 mph, and that’s almost three times the speed claimed for the “high-speed rail” project. (I suspect that 140 mph will turn out to be too high an estimate, if they actually build the damned thing.)

A psychological factor may deter people from using a Hyperloop system: there are no windows. On the other hand, you can’t see very much from an aisle seat in a passenger jet aircraft. But then, that jet aircraft isn’t only a couple of meters in diameter, either.

… Cruickshank says: “7.4 million people per year is fine. But the HSR [high-speed rail] system will carry as many as 117 million people per year.

And we all believe that HSR ridership projection, don’t we?