> You stretched 500m to 5km but kept only one valve. Why?
This is the worst case scenario by assuming only one valve is functioning. Theoretically, even that could break, but I'll assume there are enough redundant valves that at least one will always work.
> is it unrealistic for onboard pressurized tanks to keep up with escaping air while outside is getting pressurized?
It depends on the size of the hole. A bullet hole for an average train car size would take hours to become deadly and could easily be corrected by onboard air (depending on how much air is onboard), but a gun isn't going to cause a bullet sized hole. It is quite violent. Something like a catastrophic door failure, or derailment, would deplete the oxygen in less than a second. Basically, the inverse of oceangate; instead of everyone imploding, everyone would explode. Since I also suspect there will be valves on the vessel to handle releasing small amounts of gas enroute (to allow adjusting internal pressures to match destination atmospheric pressures), this could also get stuck open.
I suspect, if anyone were to actually do this, they would go for low pressure (like high altitude) instead of a vacuum. The speed of sound is so high, they could easily reach it in the tunnel. Further, people just need oxygen masks instead of dying a horrible death.
Nobody has mentioned this while following along with all the US hyperloop failures, so it is clear nobody has really tried engineering this thing, IMHO, and why I said my original comment about it. If someone were actually engineering the system, these are all pretty obvious things. As described in the original 1800's systems and by Elon, it is an impossible system. I used to think about this thing all the time in the '90s, so maybe I've thought too much about it.
I'm also curious about other issues, like maintaining low atmosphere or a vacuum (these were the key failures in older attempts in the late 1800's) in the tunnel in an energy efficient way. If it can't be kept, things will deteriorate at an accelerated rate, introducing catastrophic failures early in the system lifetime. There is also maintenance and inspections to consider. Not to mention that underground is already dealing with increased pressure from the earth, it also has to support it while maintaining a vacuum. I suspect above-ground tubes are probably far cheaper to build and maintain, but at that point, you might as well build a train.
Since moving to Europe, I can go pretty much anywhere in Europe in a day. Heck, I can get on a train this evening, sleep in a bed on the train, and wake up on the other side of Europe tomorrow morning for breakfast, for a little more than the cost of an average hotel room. Trains are great, well understood, and pretty fast. The problem the US has (as seen with the California high speed rail), is that they 'want it to be all US based' instead of hiring experts from across the ocean who work on these things every day. The US has no experience building high speed networks, which is part of the reason the hyperloop even has a chance at getting money. It's a collaborative Dunning-Kruger effect.
I think if the US can get to the point where they can develop high speed networks, in general, then stepping up to something like the hyperloop is a good idea. Other nations are still working on the hyperloop and they are making good progress, but I'm not as familiar with their details.