I really don’t see any posts so far as being out of bounds for the topic. Just my opinion.
samclem, moderator
I apologize if I came across as junior moderating, or seemed to claim that Exapno’s post was out of bounds. It was not my intent. That said, “I always have a trump card … that beats any argument you can make” is not a basis for useful discussion, and in this case is both a non sequitur and false.
So far, IMO, only Sam Stone has found a truly serious weak point in the proposal where Musk’s solution simply can’t work. There’s way too much thermal expansion to spread all the way to the end points. There need to be a fair number of joints spread along the way, but their design is non-trivial.
Yeah, that’s a deep post.
I’d been wondering all along if carbon fiber tubes would be better than steel- anyone happen to know the expansion factor?
Also, steel pipes could be temp-controlled to an extent. We supposedly have surplus power, perhaps we can pipe some heated or cooled fluid along sections of the pipes.
I don’t know offhand how well carbon fiber would work.
I had also been thinking of active cooling. I’m pretty sure there’s enough surplus power. But I’m not a huge fan of the structural integrity being reliant on an active system.
It’s a shame that nickel is so expensive. An iron-nickel alloy called invar has an exceptionally low coefficient of expansion and would drastically reduce the need for joints. Way too expensive, though.
I think the solution has to be a passive joint system–maybe even one on each pylon. Each 30 m segment needs about 2.5 cm of expansion room. I think the bearings could be designed to handle a seam like that, but the design is still a bit tricky.
The coefficient of expansion is linear, so if you want to make it a difference of 85 degrees instead of 120, that’s fine. It doesn’t really change anything. Instead of a total change in size of 1386 feet, it’s 981 ft. That doesn’t really change the problem.
The pylons are only 100 ft apart, so each expansion joint would have to allow an expansion of around .9 inches. Pylons for curved sections would have to allow for a slight movement side to side as well. That’s typically the way long structures are built today - any bridge will have expansion joints at the interface between the bridge and the roadway, for example.
The problem is now you have to figure out a way to build in 25,000 expansion joints, each of which is capable of handling the tremendous force of the atmosphere pressing on it while remaining flexible enough to move back and forth and side to side, without leaking, for years on end. I have no idea what it would take to build that and how much it would add to the cost and complexity of the structure, how much strength would be compromised, or whether it’s even feasible.
The tube will also expand in circumference, but since it’s a small tube that’s probably trivial. But it might have an effect on whatever seal is being used for the expansion joints.
This is beginning to remind me of threads on the Civil War, in which people want to talk about nothing but weapons and tactics. That’s what they find fun. But it’s not the whole story. In several important ways, the military aspects are the least determining factors of the outcome of the war. The North, given its superior industrial base, infrastructure, population size, and accrual of immigrants to replace the dead, couldn’t lose the war unless it surrendered. All the military had to do was to be there.
The limitation of discussion of innovation to technology is just as misleading. The technology exists to make flying cars. The technology exists to make moving highways. The technology exists to send rockets full of mail coast to coast. The reasons none of those exist lie elsewhere than the feasibility of their tech. Personally, I’d love to see a space elevator. But if somebody were to say that it can be built from current tech at a cost of $6 billion, I’d point out that they’ve been smoking helium.
That’s not a non sequitur, nor is it false. It’s true and to the point. It’s just a different point than the one you would like made. The history of technological innovation is as meaningful and factual as speculation about what could happen in some future. More, actually.
Carbon fiber is far more expensive than steel. Sometimes that expense is worth it, if you’re making something weight-sensitive, since carbon’s a lot lighter than steel. For a tube that’s just going to be sitting there on a concrete foundation, though, it’d be really hard to justify.
You could, however, alloy the steel to decrease its thermal expansion coefficient. Though I’m sure that would introduce its own set of tradeoffs.
EDIT: And I see that Dr. Strangelove has beaten me to mentioning invar. I didn’t realize it was so much more expensive than steel. There might be other alloys that would provide a reasonable middle ground, though.
Carbon fiber composites can be designed to have a linear coefficient of expansion of zero, or close to it. It depends on the resin you use, etc.
However, carbon fiber has a lot of drawbacks for this type of project. The big one is that carbon fiber is strong only in tension, whereas the tube has to resist bending moments and compression forces. Carbon fiber is also extremely expensive compared to steel. Finally, carbon fiber composites can break down from UV and the resins soften with heat.
There are other metal alloys that can be engineered to have lower coefficients of expansion than steel, but at much higher cost and perhaps by giving up other useful properties.
I could imagine being able to do some kind of hybrid design where you have long straight sections of welded steel, with a large expansion joint at each end, and then perhaps more exotic materials being used in curves. I can picture an expansion joint at each end of a 100 mile run which consists of a larger tube that allows the inner tube to slide back and forth inside it on bushings, perhaps with a baffle system to seal as much air as possible. You wouldn’t need a perfect seal if there are only a handful of these - you’d just design the pumps to account for the losses. Hell, maybe you could even design an impeller around the interface, so as air is sucked into the gap between the two tubes the impeller spins and reclaims some of the energy and is used to help power the pumps.
Forget that. Trying to maintain a near-constant temperature of hundreds of miles of steel tubing would ruin the energy effectiveness of the system. And a failure of the heating system would be catastrophic if the track didn’t have a passive design that could handle expansion anyway. The entire track could be destroyed.
Sure. And when a big passenger jet crashes, hundreds can be killed.
Still, when a train derails you just need to fix the section of track that caused the problem and the whole system is running again. I’m not sure what would happen to the track in the case of a sudden large opening, but I’d think considering the forces involved you might have damage all along its length, or at the least you would require a systematic inspection.
The point I was making was that you just can’t consider the pod near the failed tube section - you’d have to design the system to handle the catastrophic pressurization or you could kill everyone inside the tube.
All of those have pretty obvious reasons why they’re either uneconomical or outside the realm of reasonable speculation. No one has made the carbon nanotubes we need for a space elevator. Rocket mail is just… ridiculous. Moving highways fail even a preliminary cost check.
Look, I agree that there are more factors to consider. But most of these extra factors are utterly unpredictable in advance. Hell, they’re barely possible to analyze after the fact. History moves in fits and starts and sometimes you just have to try something. No one knew for sure, in advance, that Apollo or the Manhattan Project or whatever were actually possible. They looked possible from running the numbers and doing detailed analysis. But who knows–what if the scientists on the Manhattan Project had a change of heart regarding superweapons too early in the project? It could have stalled the whole thing.
Since no one has suggested a plan that doesn’t include an early, low cost, subscale prototype that acts as a proof of concept, I’m not even sure what the argument is about any more.
The non sequitur I was referring to was “it doesn’t exist and no one in the world is actively working on it”. None of this has anything to do with whether the Hyperloop is feasible (whether in a technical, economic, political, social, etc. sense). And it’s false that no one is actively working on it, unless you mean “active” in a sense matching “no one is actively working on writing Wikipedia articles”.
It also treads awfully close to my original statement:
Likewise, “it’s never been done before, therefore it can’t be done” is the stupidest point that can be made against something.
In a nutshell: an argument about technology and an argument about cost are two separate arguments.
But it’s right at least 99% of the time. So how can it be stupid? 
On the whole thermal expansion issue, do a search in the PDF and you’ll see it’s suggested:
“A telescoping tube, similar to the boxy ones used to access airplanes at airports would be needed at the end stations to address the cumulative length change of the tube.”
If Sam Stone’s calculations are right then thats a 700 foot telescoping tube at each end… (and that completely ignores the lateral expansion).
So yeah he is really suggesting no expansion joints the entire length. It seems amazing that of the experts he consulted on this no one actually did the calculations for the total thermal expansion.
I might liken it to a $1 lottery ticket that has a 1% chance at winning a million bucks (also, Musk’s record indicates that he might have a greater than 1% chance at success).
I could accept some sort of huge expansion joint at each end, but that only works if the track is completely straight. The minute you put directional changes into the mix, the design completely falls down. A track that goes 100 miles north, then 10 miles east, then 100 miles north again will have the entire curved section move backwards and forwards by hundreds of feet. And the east-west expansion would push the entire north-south track east or west by dozens of feet. Since the pylons don’t move, that’s not possible.
You need to deal with the expansion in increments using expansion joints. You might be able to get away with putting one before and after each directional change if you can design a sort of sliding adapter than can compensate for hundreds of feet of expansion, but you’d have to have them. The closed, welded tube design does not allow for that.
Also, even in Musk’s design you need 25,000 sliding bearings on the pylons. And if any one of them binds… look out. That’s a lot of maintenance.
The PDF does mention allowance for thermal expansion at the pylons:
“The tube will be supported by pillars which constrain the tube in the vertical direction but allow longitudinal slip for thermal expansion as well as dampened lateral slip to reduce the risk posed by earthquakes.”
Right. The problem is that the proposal directs all of that longitudinal slip to the endpoints–hundreds of feet worth. Friction effects alone make that infeasible. Even without friction, the expansion would ruin the precise shape of the various curves in the design. Something would break or buckle. You could get away with that approach for several pylons worth, but not 25,000 of them.
I was agreeing with you. You can’t pre-stress the tube either, unless you have pylons every meter or so holding it down… pre-stressed continuously welded rail is only possible because the rail is connected so frequently to heavy concrete ballast to contain the forces involved.
So yeah the tube has to be made of unobtainium, a material with no thermal expansion, or else you need to design a leakproof thermal expansion socket that can handle a car going over it at 760 mph.
Actually the car wouldn’t be going over it in the sense of touching it, since the car doesn’t touch the walls, supposedly, but the pressure from the air jets will probably still be quite a bit of stress on the joint.
Ahh, I think I begin to understand. The allowance for thermal expansion in the pillars is only to address forces on the pillars, not the fact that the tube is longer or shorter. Is that correct?
Yes, but the lateral allowance in the pillars of a few inches is completely inadequate. You have the forces of the straight sections before curves expanding pushing and pulling the starts and end points of curves from side to side, potentially the tube could move 100s of feet side to side at the start and end of each curve. Sam Stones response lays it all out here: http://boards.straightdope.com/sdmb/showpost.php?p=16569050&postcount=51