Take a set of linear aerospike engines, say 5-11 opposed pairs (they'll drop off in pairs as you go up)
Mount them on a large piston that can slide up into the tank, engines and all
Put a common bulkhead in the tank, also able to slide like a piston
Take a pair of plasma cutting torches and mount them to opposite sides of a ring that's slowly rotating around the central propellant manifolds, and mount another pair on another ring that counter-rotates
Finally, arrange for cables or something to tension the ends of the rocket together
The result is a rocket that drops its engines (and associated turbomachinery) as soon as their thrust is not needed, and literally cuts off and drops tankage as it goes
(specifically, the counter-rotating torches cut intersecting spirals in the tank, dropping diamond-shaped chunks at intervals)
Plasma torches can, with sufficient amperage, cut thin plating fast enough that I think this would work OK
but there are obviously some ... unsolved problems ;)
actually cables are needed even with thrust, I believe, oddly enough: total thrust is unlikely to be high enough even at launch, and of course it drops over time
I ran some of the numbers to get a general idea
and something like the Falcon 9 would end up with not much more than a bar of tank pressure from thrust, at best, which is pretty bad
strictly speaking you could try tank friction drive instead of cables, I suppose, but I doubt that would work well
inner our outer surface of shell could be threaded, or a central rod, so the compression is done with some kind of screw-type drive, which could simultaneously provide the extra compression push
Not really; I don't have a lot of knowledge of good seal designs, but I assume that's reasonably doable, especially given the low speed of movement
There are hydraulic cylinders that are large, but usually not that large; on the other hand, those operate often at far higher pressures, moving rather faster, and with tighter tolerances for wiggling
Rather than an o-ring type thing, perhaps an inchworm type mechanism, where each seal alternately moves in/out radially, slowly "walking" up the tube. Unfortunately that leaks unless the intermediate volume between seals is differentially pumped.
Interesting to contemplate the unusual team a startup would need to pull this off, probably including a polymer materials scientist, a cutting torch designer, and of course some serious aerospike engineering talent
An easier business plan to implement (but not insure) would be to build and equip a big "Rocket Hackerspace" with CAD and numerical simulation design tools, machine shop, testing, metrology equipment as shared resource, the way incubators share coffee machines, laser printers and conference rooms.
That only occurred to me because every new rocket start-up having to recreate the team and the tools is a big barrier to entry. Just a side-thought.
I wonder how these guys rocketlabusa.com/electron put together a team. It seems step 1 was to be in Southern California where a lot of aerospace people are concentrated.
Then again, the hyperloop went from elon's pdf to a few companies, one of which (I don't remember which) was an virtual collective of engineers not centrally located, wasn't it?
"In 2015, a handful of redditors decided to heed the call of Elon Musk as he announced his Hyperloop Competition. Today, rLoop consists of over 1,300 people from more than 59 countries who have rallied behind the rLoop concept and collaborated on the development of our Hyperloop technology."
@uhoh Fair enough, I really liked the way he explained the potential shape/quantity of the Oort cloud based on observed data-points. Even if it isn't a "perfect sphere" or "exactly X" it's still a cool concept to keep in mind.
Well... keep in mind until it's proven or disproven at least. If we're being honest that lecture exceeded my knowledge by quite a bit, which is why I thought you'd like it.