h a l f b a k e r yNo serviceable parts inside.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
Spinning around the longitudinal axis of the ship are two donuts, one in front of the other. Each donut is a fuel tank: on the spokes connecting the front donut to the ship are rocket nozzles; the rear donut's spokes are adjustable vanes (picture a jet turbine impeller).
Inside the ship is a nuclear
reactor which generates electricity for the craft, but it's primary job is to spin up the donuts counterrotationally to each other (while the body of the ship remains stationary).
(Now to help demonstrate the main principle, let's take a look at ducted fan propulsion. As well as the front fan which is moving the air backwards there is often a rear stationary fan: it's job is to turn the rotational component of the air coming out of the front fan into straight thrust, thus making the fan more powerful than it would be otherwise. Anyways...)
So you have two counter-rotating bigass flywheels: each filled with fuel (and even if they aren't, they're still quite heavy).
The front donut provides downwards thrust from the nozzles with a roll-plane component provided by centrifugal force (which splays the thrust out, neither adding nor subtracting to the downwards thrust). The rear donut comes along and the vanes, backed by centrifugal force, redirects the sideways component downwards (and adds to it). Blade angles adjust as the donuts' rotation decreases, until they're straight up when it's stopped.
To spin up the flywheels in the first place, we use a nuclear reactor: it produces far less power than a rocket engine, *but* rockets aren't run for extended periods of time, whereas a reactor is running all the time... so you could impart quite an amount of potential energy to the flywheels while waiting to use the rockets again.
Of course you'd want to make sure you were pointed in the right direction before spinning up even though turning the ship is doable with the flywheels on gimbals: the ship shifts from ---|-|---> to ---/-/---> to make limited turns while avoiding precession. Major attitude adjustments are best done while the flywheel system is at rest, of course.
Let's look at an example flight plan (a simple one: burn to get going, then again to stop at the destination):
After positioning the ship to the right orientation, have the space station spin up the flywheels for you (or do it yourself by firing the rockets sideways then refuel after you've spun up).
Blast off assisted by the flywheels: burn time and initial flywheel speed are calculated such that all the energy from the flywheels is dumped into the thrust: when the burn's finished the flywheels are motionless and the entire ship can be turned with attitude jets without worrying about precession.
Reposition ship in the eventual docking orientation. Start spinning up the flywheels using the nuclear reactor. Go get a coffee; it's gonna be a long trip.
When you get to your destination, again use the flywheel to assist decelerating to rest.
Inspiration
Rotational_20effect_20ion_20engine
which probably isn't the same idea. Note that the Hybrid might work with ion engines as well, but the flywheels would have to be spinning much faster to get an equivalent ratio of speed increase. The flywheel could be much lighter, of course. [FlyingToaster, Nov 08 2009]
[link]
|
| |
oh come on, why bone this: it's a good theoretical exercise if nothing else. (whining edited out) |
|
| |
Mine's the second fishbone.
Rocket gas velocities are higher than jet engine gases. At the size of a jet engine, it's a real challenge to operate the engine at the necessary speeds without it coming apart.
Have you ever noticed a rocket engine with post-combustion turbine wheels? Ever wondered why not?
OK, your rear turbine flywheel is going to be loaded with fuel, and spinning around the outside of something big enough to be habitable. Any guesses as to what will happen? |
|
| |
It doesn't *have* to be a fuel-filled donut but I was hungry when I posted it and still am. |
|
| |
But why use dead weight when you can use live ?: you want something heavy enough such that it will keep rotating through an entire burn and rocket fuel seemed the obvious choice :) |
|
| |
The idea is that you get to store the constantly produced energy of a nuclear reactor, then use it to assist with thrust, thus saving fuel; not entirely unlike using a constantly running motor/gen to charge a battery for an acceleration-assist. And unlike using up more rocket fuel for a longer burn, you still have a functional nuclear reactor at the end of the trip. |
|
| |
And the cool thing is... using a flywheel for motive power ... in a vacuum. And it's "free" in the sense of you don't have to toss more irreplaceable stuff out the back to get higher acceleration. |
|
| |
F'rinstance, continuously spin 25 meter radius donuts at 600 rpm each and *double* the effective thrust of nozzles spaced along the rim (based on hydrogen-peroxide propellant which is the only rocket fuel I could quickly find a detonation speed for). |
|
| |
//continuously spin 25foot radius donuts at 600 rpm each// |
|
| |
Sounds like you need to draw yourself a picture. And do some math. |
|
| |
If you're pointing the rockets radially outward, then you get to add the rotational speed to the exhaust velocity. However, it's the direction that's wrong: the exhaust would be exiting on a line 45 degrees outward from a tangent line to the edge of the donut, but in its own plane. All around. Result: 0 thrust. |
|
| |
If you're pointing the nozzles "down" to give a thrust vector, then the rotation of the donut is perpendicular to your thrust and you get no added thrust velocity. However, the outward vector would cause the exhaust to form a cone shape, and since it starts from the outer edge of the first donut, it completely misses the second donut. Which is therefore going to have to be far larger if it's going to do that redirect & accellerate thingy. |
|
| |
Another item you might take note of is that since the exhaust is not constrained to the inside of a tube, there will be no rotational component. The gas just flies outward. A vane built to catch and redirect the gases is going to have to be a ring, not a radial vane. |
|
| |
//why use dead weight when you can use live ?// - the problem isn't whether it's live weight or dead weight; the problem is that it's weight. With the numbers you quoted, you're going to be looking at 3000 g's of centripetal acceleration. Take your rest fuel mass, add the structure weight, divide by the number of support vanes, multiply by 3000, find out how strong those vanes need to be. Calculate safety factors based on it being a very long walk home. |
|
| |
//math//... hmm. okay 25 *metre* then (or 1800rpm) |
|
| |
//pointing nozzles down... misses second donut//so one of the donuts in my math-example-anno is 24metres wide, the other 26, and they're a foot apart, or any other combination of whateverage. |
|
| |
//ring vane// nope: oh a ring *would* work (and be simpler), but you'd only get a partial deflection. As you pointed out, the stuff flies off tangtially, not radially, so it hits the vanes pretty much dead on. Picture the two donuts as being very close together. |
|
| |
Post was writ' to show how you could turn rotational energy into thrust in space. Materials science is just gonna have to catch up, is all. |
|
| |
//sp. cetrifugal.// ... really? |
|
| |
I never knew the difference and went all the way over to Wikipedia to figure out once and for all which was which. Sadly I will probably forget by the next time I need to know. |
|
| |
Meanwhile, materials science aside, this will work. Originally I had the two donuts, both with nozzles and vanes, with the nozzles pointed at the vanes on the other donut, pulsed firing of course. This is (slightly) more elegant, but far from perfect yet. |
|
| |
//sp. cetrifugal.// ... really? I would think "centrifugal" more likely. |
|
| |