Catapults. - Ancient technology for new purposes.

Echus_Chasma · 2003-01-06 13:22:02

How big will it have to be, approximatly?

Palomar · 2003-01-07 11:15:15

*Interesting. How do you manage to insert illustrations...even moving ones??

Catapults were featured in Heinlein's "The Moon is a Harsh Mistress" (items from Luna were catapulted to Earth, calculated to crash into ocean). I thought it was a cool idea to include in the story.

--Cindy

Austin Stanley · 2003-01-08 03:37:45

I expect Catapults would suffer some of the same problems that rail-guns do. Namely that it would be VERY hard to design a payload that could survive the stress of being accelerate to escape velocity in such a short period of time.

Palomar · 2003-01-08 08:53:07

I'll have to read that book. Robert Heinlein was the first scifi writer I ever read. I moved on to Pournelle and others before completing all his novels. Was it a gravity lever like a Trebuchet or one of those magnetic rail guns? The moon's one sixth gravity would certainly make the job a lot easier.

*If memory serves me right, I believe it was a gravity lever.

My favorite Heinlein tale is "Have Spacesuit, Will Travel."

--Cindy

dicktice · 2004-03-22 09:35:20

Mine is the story; "The Roads must Roll," one of the first to tackle the problems posed by multi-speed moving sidewalks.

SBird · 2004-03-22 10:56:57

I love this idea even though it's got some major flaws. The big flaw is the strength of the arm. Dropping a 100 metric ton weight on a steel beam is going to bend it like spagetti. You're much better off with some sort of hydraulic actuator that can put a load on the lever arm more gradually. You'd have to get a mechanical engineer to give you some sort of estimate of how string a steel arm can get but I'm guessing that it's not practical to build an arm that can get more than a few dozen kg into LEO. However, if you have orbital manufacturing capabilities, the ability to lob raw metal to orbit for cheap is great.

The next biggest challenge will be the aerodynamic forces. Your'e basically accelerating your arm end to something like mach 23 at sea level. It's going to need some sort of heavy ablative shielding (shuttle reusable tiles would get turned to dust by the aerodynamic forces.) It might even be that the air drag would be even more of a problem than the forces generated by gravity and the counterweight. However, if we're willing to regularly replace the ablative tiles, there's no reason that we couldn't get this to work. It might even be advantageous to put some sort of scramset or something on the arm end to help boost the arm speed and to reduce the flexural load on the arm.

Finally, there's the areodynamic load on the layload which won't be insignificant. It's unlikely that any load that's small enough to be lobbed by this launcher would be able to survive the reverse reentry imposed on it.

On the other hand, something like this might work on the moon where your escape velocity is much easier and you have no atmosphere to worry about. Escape velocity is only about 2.28 km/s and orbital velocity can be almost arbitrarily lower than that. This could take the place of an orbital railgun. Railgun proponents always seem to forget that no one has ever made an orbital velocity railgun that performs reliably.

Ian Flint · 2004-03-23 10:46:02

I like the idea for using a catapult to launch from the moon. Earth's atmosphere just seems like too big of a barrier.

I would propose using the Earth catapult as a mini first stage. It would be much easier on the catapult to only launch its payload at 300 meters per second -- subsonic. This would avoid bending, superheating, and sonic booms as well.

I suppose a catapult could be designed to launch a rocket at mach 3, 1 km/s, without causing extreme stress on the equipment.

What do you think?

SBird · 2004-03-23 16:17:16

I did a quick calculation for a lunar catapult. You need to use a long arm for the catapult so that the rotational G-forces are minimized. Assuming a 1 km long arm that flings the mass at a final velocity of 2.38km/s, the end mass suffers a centrifugal G-force of about 578 G's. I'm afraid that you'd have to make your arm out of space elevator building materals. On the other hand, you only have to make it 1 km long instead of 36,000+ km long and the overall stress are probably lower. Unlike the space elevator, this catapult is probably practical to build with real materials we'll have in the next 30 years.

Also, the catapult idea down't really work as is. Using a traditional catapult design, you have to get that 2km/s of acceleration in about 1/2 of a rotation of the arm. That requires a 0.8 g acceleration of the arm end the whole way. This is too much. Any sort of km-long structure that can hold a significant payload and its own weight at 578 G's must be some sort of carbon fiber rope. It can't withstand large accelerations - it'll just roll itself around the central pivot like a rope getting rolled up on a winch.

Instead, envision this alternate strategy - you dig a 1000m deep trench in the ground and mount the rope pivot on the ground at the center of the trench. Now, start by holding the rope parallel to the ground at the top of the trench and attach your payload to the end. Give the payload a big mechanical kick so that the payload and rope have enough momentum to make a complete circle - this has to be just enough rotational momentum to >0.17 G's. By my calculations, the initial kick would have to be at least 62 m/s which is easy. In fact the next step is much faster if this first step has more energy so let's assume a 150m/s intial kick which is nothing too complicated.

Now, the central pivot is hooked to an electric motor. Over the course of several minutes, you just keep feeding power into the pivot and get that rope spinning faster and faster until the tip is going 2.38 km/s. At that point, you just release the payload and it shoots up into a lunar escape trajectory. A slight increase in tip velocity lets you achieve lunar escape velocity. Now, you just use the electric motor as a brake and regain much of your intial energy, catch the end of the rope and get ready to repeat the process.

The nice thing about this technique is that it lends itself to parallelization nicely. With a single central shaft, you could have numerous ropes that each launch a mass per cycle. The total energy expended is very similar to what you'd spend on a space elevator (probably less considering the mechanical coupling and gravitational losses of climbing slowly up a rope out of a gravitational well.)

Assuming that the rope ends up weighing 2.5 g/cm^3, the rope's own weight (ignoring the payload docking mechanism at the end and no taper) puts a stress of 7.08 GPa on it where it attaches to the central shaft. Making a conseravtive assumption about nanotube rope strength of 25 GPa and a 20 kg payload docking mechanism, you can launch a 85.5 kg payload per cycle and maintain a 3-fold safety margin. Further assuming that you have 10 rope launchers, and a total cycle time of 1 hour, you can put 855 kg into lunar orbit per hour or over 20.5 metric tons per day.

This assumes that the payload is just a dumb lump of metal. Some sort of tiny reaction mass thruster could be attacked to make course corrections but otherwise, you've got everything you need to throw mass to any arbitrary location you want. If you're just lobbing mass to a lunar orbit where you've got a big space tug that's collecting this stuff, you can assume that the launch speed can be as low as 1.8 km/s and then the mass per launch cycle goes up to 195.5 kg per cycle per rope for a total of 47 metric tons of mass per day for a 10 rope launcher.

On Earth, it simply doesn't work. As I've pointed out with regards to maglev launch boosters over in the Earth to LEO thread, getting a mach 3 launch boost doesn't help that much. You stil have to have a large rocket to get the necessary orbital velocity. Even is one assumes a mach 3 launch and a 1 km arm, you're still dealing with a 100 G launch stress. I can't imagine any sort of practical rocket that can withstand that sort of force. Plus, the weight of that rocket would be so much that you'd end up with an impractical catapult.

If you tried to reach escape velocity with just the catapult on Earth, you're looking at 7 km/s. The forces on the rope alone would be over 60 GPa at the base. These are the sorts of stresses you see a space elevator but now you're looking at a tip velocity of mach 23. Either you need an impractical booster rocket or or an impractical catapult.

dicktice · 2004-03-23 16:48:06

No wonder those old Romans never got off the planet!

SBird · 2004-03-23 17:03:04

I know, they took one look at what a 7 km/s piece of tether does to an aqueduct when it hit it and decided that they had better things to do with their time.

Bill White · 2004-03-23 17:14:10

Instead, envision this alternate strategy - you dig a 1000m deep trench in the ground and mount the rope pivot on the ground at the center of the trench. Now, start by holding the rope parallel to the ground at the top of the trench and attach your payload to the end. Give the payload a big mechanical kick so that the payload and rope have enough momentum to make a complete circle - this has to be just enough rotational momentum to >0.17 G's. By my calculations, the initial kick would have to be at least 62 m/s which is easy. In fact the next step is much faster if this first step has more energy so let's assume a 150m/s intial kick which is nothing too complicated.

WHy do you need a vertical trench or to rotate at 90 degrees? Orbit is velocity, correct?

Spin your rope maybe 15 degrees off horizontal from a tower on top of a hill. If you spin fast enough, you still attain orbit, correct?

Maybe spin up the system while horizontal - - centered on a tower built ojn a hill top then tilt the tower top 15 degrees with a hinge, then let go. If the hinge were multi-directional you could tilt whichever way allowed the desired trajectory.

Or do I misread your idea? What about doing this on Phobos?

G-force means this is for cargo only, right?

= = =

SBird, I just got an urge to buy an electic motor and some gear and some kevlar sailing line and see how far I can throw stuff. Uh Oh.

I better find a big field.

= = =

Further thought.

Attach 2, 4 or 6 aluminium trusses to the spinning tower.

Attach the payload to the truss end along with your tether, coiled so the payload is secure against the truss. Begin spinning the tower with the payload perhaps resting on a falt surface at the end of the truss.

As the spin rate accelerates, start releasing line as needed to attain the desired radius and keeping the payload from striking the ground. Release at the appropriate speed.

All you need is electric power - - no propulsion systems, no rockets.

SBird · 2004-03-23 18:34:41

Ah, you're correct - that's much easier than diggin a km deep trench in the ground. You really don't even need to have the launcher at an angle if you're going to escape velocity, just have the whole thing rotate parallel to the ground and the mass will sail off into the wild black yonder just fine as long as there aren't any tall hills in the way. (and if there are, they won't be there for long) You can basically shoot your mass in any angle you want by properly timing when you release it. Aside from the central cable, your greatest points of stress are at the central column and at the payload interface. The central pillar's got a total load of about 133 metric tons on it at full speed. This is nothing that's outside good old steel manufacturing. I'm sure thre's cranes and powerplants that have to deal with greater stresses than that. The payload interface stress is well within the sort of stresses that standard steel bolts can withstand.

I'm imagining the cable would be a big loop so that it can just run around the central pillar and a loop on the payload interface. This allows the stress from the cable to be distributed so that it doesn't build up in one area. Some rough calculations show that there's nothing in this system aside from the nanotube cable that can't be easily handled by modern engineering materials.

Your idea of playing out the cable until the system has enough rotational velocity would work well. Get the boom spinning at a low speed, reel out the line and then start ramping up the velocity.

The problem is that you can't really do this in parallel like before. There's too great a chance that the cable will get tangled with teach other. Nonetheless, you're still looking at 2.5 metric tons to escape velocity per day. And that assumes that you only launch one mass per hour. I'm guessing that you're turnaround time will actually be much higher than that - perhaps on the order of 10 minutes. A few of these things on the Lunar equator could move really huge amounts of raw material to orbit.

I went back and looked at the 1.8 km/s lunar orbit but I don't think that's too feasible since there's no provision for a thruster to circularize the orbit. Let's just stick with the 2.38 km/s escape velocity.

Of course, the prime competitor would be a railgun. A railgun would probably be capable of moving more mass to orbit but AFAIK, noone's demonstrated a railgun design that works reliably for large masses.

And yes, the stresses involved would be for anything other than metal or perhaps compressed O2 tanks. Too bad there's little H2 on the moon, it would make an interesting way to start moving H2 to Mars for in situ fuel production.

Bill White · 2004-03-23 21:43:12

What would the needed rpm be?

Use the turbine technology used to generate electric power, just run it in reverse, or what about the electric motor half of diesel electric engines? (No diesel, of course, solar or nuclear electricty, but the electric engine half would seem darn close to off theshelf)

Can such turbines maintain sufficient rpm to achieve escape velocity?

= = =

I really like the idea of building on on Phobos or Deimos to throw food (from Marsian farms); water; bulk fuel; clean clothes and stuff like that to prospectors in the asteroid belt.

The Marsian economy is sustained supporting Earth ships in the Main Belt. Spin your launch turbines with a combination of nuclear and solar power.

Bill White · 2004-03-23 22:58:36

Warning - math challenged poster attempting math. Please double check.

Assume a 100 meter tether spun around a central hub. (This is the final length, perhaps paid out as revolutions increase - start in tighter, perhaps resting on a solid arm.)

Circumference is 6.28 x 100 = 628 meter circumference.

1 rps or 60 rpm equals a 0.628 kps, correct?

4 rps or 240 rpm equals 2.412 kps, correct?

2.38 kps is lunar escape velocity, correct?

Thus we need an electric powered turbine capable of 240 rpm perhaps with a 1000 kg (2000 kg? or 5000 kg?) payload projectile at the end of a 100 meter tether. I suspect using 2 payloads at opposite ends would help help this contraption balanced. (Simultaneous release in 180 degree opposite directions? Turbine spin down recharges electric batteries. . .)

My liberals arts trained intuition suggests General Electric sells stuff off the shelf to electric power utilities that can handle the job.

Thoughts (or is my math all wrong?)

SBird · 2004-03-23 23:53:47

The problem with using a shorter arm is that the relative centrifugal force upon the end weight is much greater to achieve the same speed. Centrifugal force = mv^2/r where m is the mass, v equals the radial velocity and r is the length of the rotation arm. Therefore, to get a certain radial velocity, reducing the arm length by 10 causes the centrifugal force to go up by a factor of 10 as well.

On the 1km arm, the G force at the tip is 578 G. With a 100 m arm, your G force goes up to 5780 G which is just too much. The total force on the center pivot goes up to something like 1300 MT which is going to be hard to engineer for. At those forces, you have to be careful of the nanotube rope just slicing through the steel components like butter. In fact, ignoring things like the strength of the rope vs length and other concerns, your total payload capacity goes up linearly with increasing arm length. Therefore a 2km long tether would be able to lob about twice as much payload as a 1km one. However, you quickly start running into all sorts of other issues with the long tethers so I'd say that a 1-2km tether is a good compromise.

The rotation rate also goes down with increasing arm length, the 100m needs to go 227 rpm to hit escape velocity and the 1 km arm needs 22.7 rpm.

Bill White · 2004-03-24 00:37:23

The problem with using a shorter arm is that the relative centrifugal force upon the end weight is much greater to achieve the same speed. Centrifugal force = mv^2/r where m is the mass, v equals the radial velocity and r is the length of the rotation arm. Therefore, to get a certain radial velocity, reducing the arm length by 10 causes the centrifugal force to go up by a factor of 10 as well.
On the 1km arm, the G force at the tip is 578 G. With a 100 m arm, your G force goes up to 5780 G which is just too much. The total force on the center pivot goes up to something like 1300 MT which is going to be hard to engineer for. At those forces, you have to be careful of the nanotube rope just slicing through the steel components like butter. In fact, ignoring things like the strength of the rope vs length and other concerns, your total payload capacity goes up linearly with increasing arm length. Therefore a 2km long tether would be able to lob about twice as much payload as a 1km one. However, you quickly start running into all sorts of other issues with the long tethers so I'd say that a 1-2km tether is a good compromise.
The rotation rate also goes down with increasing arm length, the 100m needs to go 227 rpm to hit escape velocity and the 1 km arm needs 22.7 rpm.

As you can tell, my degree is in American History.

google helps, by the way. . .

Anyway, with a 1-2 km tether, I just dont see a 1 or 2 hour operational tempo. We will need to start with the paylaod held in tight to keep the payload off the ground and pay out tether as speed increases.

That could take some time to accelerate gradually to escape velocity.

But with no air resistance and a well designed hub with minimal resistance (liner induction drive with a small radius mag-lev circle track?) total overall efficiency might be pretty good.

The advantage over a straight line rail gun would seem to be the ability to accelerate gradually, until escape velocity is achieved.

How much shock loading do aircraft carrier arrestor cables experience? 5700 G may be far too much but 2000 G for things like water or methane or tofu paste might be acceptable.

SBird · 2004-03-24 03:23:59

The payload, as long as it's just a hunk of metal or a sturdy tank can withstand huge G forces. I've got a centrifuge in the lab that can hit 30,000G. Ultracentrifuges these days have rotors that can take samples to about 1,000,000 G.

The problem is that your launcher can't take that stress. At 1 km and 578 G, the cable alone without any load already is carrying about 20% of its maximum load capacity. By going to a 100 m tether, the cable is under 10 times as much stress. This is counteracted by the cable being 10 times shorter so it evens out. However, the payload now weighs 10 times as much and therefore has to be 10 times smaller or you'll snap the cable.

As for letting the calbe out, you actually want to do that as quikly as possible. Assuming that the arm is 100 m off the ground and a 1 km arm at full extension, you can let the cable out once the arms is rotating at > 1.8 RPM and a total G force at full extension of 5 G's. This minimizes the forces on teh winch that has to let the cable out and on the cable since it will be under greater stress when wrapped around the release mechanism. After full extension, you can rapidly speed up. I actually think that one could get a spin-up cycle to be fairly fast. Once the cable is moving with any velocity, you can increase the rpm quite rapiely without problems.

Think of a yoyo. Tie the string to a pencil. Try to get the yoyo flying in a circle by rotating the pencil - doesn't work, does it? Now, give the yoyo a starting kick and start spinning the pencil - you'd be suprised at how quickly you can get it up to a high speed. I'm guessing that the cycle time will be limited by the central motor more than anything else.

stratoghost · 2004-03-24 23:11:53

In order to get over inertia, why not use a maglev launcher to accelerate a spaceship before launch. Using India and Tibet as a launch platform could bring down the costs, after the initial installation of the 1,000 km or so long launch platform.

SBird · 2004-03-25 01:09:21

Check out the Earth to LEO thread for some discussion about maglev launchers. It turns out thatunless you get some very high launch speeds, a railgun or maglev launchers just isn't too practical. On the Moon, it makes more sense because of the lower gravity and the lack of air. On Earth, you're subjecting your rocket to incredible aerodynamic stress for what turns out to be a pretty small gain.

Mad Grad Student · 2004-03-25 20:52:10

The stress due to g-forces and air resistence would be so huge with one of these it would be impractical to try to go to orbit with it. However, what if you were only trying to get up to about mach 3, for an opportunity to lose some propellant? Supposing you could launch somewhere in the Hymalias, 20,000 feet above sea level, the equivalent of mach 2 forces would be up in the mach 4 true airspeed range. Rockets are already built to withstand that kind of aerodynamic stress, and max-Q would occur seconds after launch, so from there on out it would be downhill idicated airspeed for the booster. As long as the rocket could withstand the g-forces, this just might work, and could save a few dollars on less propellant.

Let's say you had a pretty efficent rocket that weighed 70,000 pounds and had a 2,000 pound payload. Unfortunately, I'm in trig right now, so I can't do the calculations yet. Can anyone tell me how big a trebuchet would have to be to get this up to mach 4, and what the g-forces would be before release?

ruski_canuk · 2004-03-27 22:04:33

David vs Goliath

If getting to space, at low cost, is as difficult as it is for a young small boy to defeat a battlehardened giant warrior why don't we glimpse into history/legend (depends who you ask) and learn a thing or two from a boy called David.

I have been reading threads on a catapult idea for launching into space. I like the idea though I believe that the paradox lies in the beam of the catapult - the only material that displays qualities which come close to having the strength per m^3 to withstand the insane initial acceleration of taking a sizable payload from 0 to 12 km/s in a rotation is the budding carbon-fiber technology.

Instead of a one time "catapult" into space why not "sling" into space. Build a massive, vertical spinning, sling. You could mount it in a large pit in the ground for better stability and instead of subjecting your "beam" to that crazy initial stress of the catapult, just gradually increase the rotational speed up to the point required to reach the necessesary departure velocity's and then like David let it go! and whola - A giant is slain.

I want to hear from you more educated folks now... :-) (I am only a humble student of engineering)

Justin

Rxke · 2004-03-28 07:00:38

Slingshot launch... Why not horizontal? You can get away with a lower required added momentum, just add an upper-stage to your payload to do the rest... (height) Build it somewhere equatorial, and you have a sling that is even more efficient that way... you could build the payload aerodynamical, so that it gets airborne after a few rounds... so it lifts off the ground (circular) track, going faster and faster, and then release it in the right direction... Then start your engines to speed up even more and climb... That way you could launch a SSTO vehicule that needs less propellant.

(Probably impossible, but who knows)

EDIT... Um... Okay, has already been discussed, in essence, never mind...
But SBird, do you think this kind (your idea) of approach would be economical, compared to purely vertical launches? Once you build the installation *and* it is fairly simple to maintain, it should be, no?

New Mars Forums

Announcement

#1 2003-01-06 13:22:02

Re: Catapults. - Ancient technology for new purposes.

#2 2003-01-07 11:15:15

Re: Catapults. - Ancient technology for new purposes.

#3 2003-01-08 03:37:45

Re: Catapults. - Ancient technology for new purposes.

#4 2003-01-08 08:53:07

Re: Catapults. - Ancient technology for new purposes.

#5 2004-03-22 09:35:20

Re: Catapults. - Ancient technology for new purposes.

#6 2004-03-22 10:56:57

Re: Catapults. - Ancient technology for new purposes.

#7 2004-03-23 10:46:02

Re: Catapults. - Ancient technology for new purposes.

#8 2004-03-23 16:17:16

Re: Catapults. - Ancient technology for new purposes.

#9 2004-03-23 16:48:06

Re: Catapults. - Ancient technology for new purposes.

#10 2004-03-23 17:03:04

Re: Catapults. - Ancient technology for new purposes.

#11 2004-03-23 17:14:10

Re: Catapults. - Ancient technology for new purposes.

#12 2004-03-23 18:34:41

Re: Catapults. - Ancient technology for new purposes.

#13 2004-03-23 21:43:12

Re: Catapults. - Ancient technology for new purposes.

#14 2004-03-23 22:58:36

Re: Catapults. - Ancient technology for new purposes.

#15 2004-03-23 23:53:47

Re: Catapults. - Ancient technology for new purposes.

#16 2004-03-24 00:37:23

Re: Catapults. - Ancient technology for new purposes.

#17 2004-03-24 03:23:59

Re: Catapults. - Ancient technology for new purposes.

#18 2004-03-24 23:11:53

Re: Catapults. - Ancient technology for new purposes.

#19 2004-03-25 01:09:21

Re: Catapults. - Ancient technology for new purposes.

#20 2004-03-25 20:52:10

Re: Catapults. - Ancient technology for new purposes.

#21 2004-03-27 22:04:33

Re: Catapults. - Ancient technology for new purposes.

#22 2004-03-28 07:00:38

Re: Catapults. - Ancient technology for new purposes.

Board footer