Kerbin's Third Moon

[OpenWorldAddict]

This is more of a question then a challenge, but would it be possible to drag a large asteroid into a stable orbit around Kerbin and then land on it, do science on it, and even build a mining base on it. It would become Kerbin's third moon.

There is enough space between Kerbin and the Mun, and the Mun and Minnus to have another object orbit Kerbin.

If you think it is possible, then show me how it would be done?

[HoloYolo]

Yes, also I think this should be moved to Genaral Disscusions. People have done this muiltiple times.

[windows_x_seven]

Redirect Gilly.

[Laie]

This is more of a question then a challenge, but would it be possible to drag a large asteroid into a stable orbit around Kerbin and then land on it, do science on it, and even build a mining base on it.

If you think it is possible, then show me how it would be done?

Dragging it into a stable orbit is entirely possible though a bit tedious: the claw can transmit only so much force before it starts to bend and flutter. If you don't want to use mods to help with this, you can only ever provide very little thrust (like, one or two Poodles' worth) and hence have to deal with very long burn times. Rotating the thing into the right direction for a burn also requires a lot of patience.

It will have no gravity. Any base you want to build needs to be anchored using one or more claws.

The only science you can do is collecting soil samples.

Edited July 13, 2015 by Laie

[Snark]

Dragging it into a stable orbit is entirely possible though a bit tedious: the claw can transmit only so much force before it starts to bend and flutter. If you don't want to use mods to help with this, you can only ever provide very little thrust (like, one or two Poodles' worth) and hence have to deal with very long burn times. Rotating the thing into the right direction for a burn also requires a lot of patience.

One way to mitigate the bending-claw problem: Make it so that the thrusting craft has no torque authority at all, and is thrusting straight at the asteroid's CoM. That is, make sure that the thruster has a locked gimbal on its engine, and has disabled all reaction wheels (including the command module, if it has 'em). This means its thrust axis stays lined up straight through the asteroid's CoM and flutter is minimized. Of course, it also means that the craft is useless for steering the asteroid (i.e. rotating it to the desired orientation), but you can make up for that by attaching other craft that have plenty of reaction wheels but no active engines. They do the torque, the thruster does the thrust.

That said: I personally find the claw to be so horrendously buggy and game-destroying that I've completely sworn off it forever, or at least until Squad has the bandwidth to fix it. But the above technique worked well for me until I got so frustrated with game-destroying bugs that I swore off the claw.

[Grumman]

One way to mitigate the bending-claw problem: Make it so that the thrusting craft has no torque authority at all, and is thrusting straight at the asteroid's CoM. That is, make sure that the thruster has a locked gimbal on its engine, and has disabled all reaction wheels (including the command module, if it has 'em). This means its thrust axis stays lined up straight through the asteroid's CoM and flutter is minimized. Of course, it also means that the craft is useless for steering the asteroid (i.e. rotating it to the desired orientation), but you can make up for that by attaching other craft that have plenty of reaction wheels but no active engines. They do the torque, the thruster does the thrust.

Or you could just turn off the engine, turn on the reaction wheel, and turn it. No need to make them separate craft.

[Snark]

Or you could just turn off the engine, turn on the reaction wheel, and turn it. No need to make them separate craft.

Yes, but how do you steer it while you're thrusting? Unless the CoM line-up is perfect (which in practice it never is), there will be a smidgeon of torque error while thrusting, and you'll need some sort of torque authority to correct it. The problem is, if you put that torque authority on the thing doing the thrusting, it causes it to bend in a way that makes the thrust torque worse. This results in a positive feedback loop that causes the runaway waggle problem. So it's helpful to have a way to torque the asteroid that doesn't include torquing the thruster.

[Landwalker]

Redirect Gilly.

I was actually wondering this morning if redirecting Gilly would even be possible. My assumption is that "established celestial objects" (like Gilly) have different behaviors when it comes to things like that than asteroids do, but I don't know that for sure.

[Grumman]

I was actually wondering this morning if redirecting Gilly would even be possible. My assumption is that "established celestial objects" (like Gilly) have different behaviors when it comes to things like that than asteroids do, but I don't know that for sure.

They do. Even with infinite fuel (and you don't even need to cheat to get it, now that we've got mining), planets and moons effectively have infinite inertia - they're going where they're going whether you like it or not.

[Sharpy]

Yes, but how do you steer it while you're thrusting? Unless the CoM line-up is perfect (which in practice it never is), there will be a smidgeon of torque error while thrusting, and you'll need some sort of torque authority to correct it. The problem is, if you put that torque authority on the thing doing the thrusting, it causes it to bend in a way that makes the thrust torque worse. This results in a positive feedback loop that causes the runaway waggle problem. So it's helpful to have a way to torque the asteroid that doesn't include torquing the thruster.

You don't. Deactivate thrusters, enable reaction wheels, correct, then reverse that and accelerate again.

How viable would be using multiple tugs? By adjusting thrust limiters you should be able get a setup that has a summary CoT through CoM of the asteroid. (actually, on a single craft with widely placed engines you can achieve that too... but it will put a torque on the Klaw.

One more option: use the klaw for stabilization only, placing it on a flexible joint (a beam attached to a stack of Spider engines?) and use several girders as the actual "pushers"/"landing legs". So when you thrust, the klaw only prevents slippage while the girders carry the forces.

I was actually wondering this morning if redirecting Gilly would even be possible. My assumption is that "established celestial objects" (like Gilly) have different behaviors when it comes to things like that than asteroids do, but I don't know that for sure.

Unfortunately truth is KSP doesn't really use two-body model physics. It uses single-body with gravity-less orbiters. E.g. your actions don't influence the planets, just the same as asteroids don't have gravity at all.

By the way, people have been placing asteroids in LKO, making them the refuelling stations. No ferrying fuel around anymore.

Edited July 14, 2015 by Sharpy

[Snark]

You don't. Deactivate thrusters, enable reaction wheels, correct, then reverse that and accelerate again.

Yes, that works-- I've done that, too. I find it to be really tedious, though-- takes forever to get the thrusting done. Having reaction-wheel torque craft separately attached to the asteroid allows firing on continuous thrust and is a lot more convenient. For me, at least. Until the kraken strikes and makes me regret succumbing to temptation and using the klaw because I'd really really like to play with asteroids.

Gosh darn I wish they'd fix it...

[xtoro]

Tip: Pull, don't push. I've managed to tug several roids with success. Use a long girder with claw on the end, and engines on the other end angled slightly outwards radially but towards the claw end. Leave the pivot free and slowly ramp up the engines. The asteroid will swing around a bit but eventually come to a rest with its CoM directly behind you. Keep ramping up the throttle slowly till you're at 100%.

[Invader Jim]

It would become Kerbin's third moon.

That's no moon....it's a space station.

[Laie]

Gosh darn I wish they'd fix it...

I've only run into the "frozen orbit" issue so far. Which seems to be triggered by the claw flexing too much. At least that's what it says in the bug report, even though I can't even begin to imagine what the one may have to do with the other.

Annoying as it may be, I wouldn't call it a kraken. But that's beside the point. What I wanted to say is that Kerbal Joint Reinforcement helped me on both counts: less flexing, no more frozen orbits.

[moronwrocket]

I had seen that suggestion before, and managed to successfully, very very painfully, capture a class E in Kerbin Orbit. Having the ISRU on a separate craft proved a bad idea - that craft and the pusher weren't perfectly aligned, so they constantly wiggled and bounced. And the RCS/torque drones were notoriously kraken-prone due to side-mounted Klaws on SAS modules - they would often explode during physics loading, and constantly jittered. In the end, I decided to not bother because of several problems:

1. The asteroid rotation doesn't stop, even when in timewarp. *Especially* when in time warp. Made it very hard to stay a fixed direction. On top of that, the RCS/torque drones didn't seem to apply torque perfectly and would cause constant over-rotation.

2. Circularizing the orbit on Kerbin's equator would have taken 2/3 of the mass of the astertoid (~1500 tons) and hours of thrusting, which would have made the whole thing not really worth it for fuel.

3. The ISRU unit has a bug where it would keep draining the ore from the asteroid even though the containers should have been full, probably because of using one full Ore unit to top off fractional amounts of monoprop/fuel usage.

4. MY contract was for ejecting the asteroid from the solar system, so staying at Kerbin was only for a short while anyways.

5. The ISRU and pusher modules didn't have RCS thrusters on them (my mistake). The two nuclear outriggers did, so I used them to try and center each on the asteroid CoM.

Right now I edited the save to restore the lost mass, and debating trying to eject it from LKO to solar system escape - needs something like 7km/s dV, which is close to 7 real-time hours of burn - more when burning at 50% to account for off-axis burning. I may just edit the save again, since I feel like I earned the contract completion.

Don't ever do this:

Something about having the Klaws mounted on the side of the SAS modules caused the whole craft to wobble and shake, and sometimes explode.

I did not try a pull configuration, that might work better, but I doubt it. How does the floppy asteroid dampen oscillations? I would think that would make problems worse. Also, this:

https://en.wikipedia.org/wiki/Pendulum_rocket_fallacy

If I was to try again, here's what I would do:

1. Try a class D instead - E is just too big.

2. Use one vessel, ISRU + pusher combined.

3. Only use 1 nuke.

4. Try using Vernors on the RCS/torque drones - they provide more thrust and only need LFO.

5. Turn on drill to fill tanks, then shut off drill. Turn on ISRU to fill tanks, then shut off ISRU. Don't leave either on while thrusting for long time periods, or while defocused.

6. Instead of trying to put the drones in specific spots, make sure the Vernors point specific directions.

7. Make sure all vessels have full RCS controls (vernors or whatever).

Best of luck to anyone else.

- - - Updated - - -

Just had another thought. One thing that might work is to have your thrusters in a radial-4 configuration. And using KER, you can modify the thrust on each engine to get the thrust torque as close to zero as possible. The flexible Klaw might still prove to be a problem though.

Dammit, now I want to try again. Currrrsee you KSP!

[Laie]

The pendulum thing is *not* a fallacy when dealing with asteroids in KSP. Consider this:

Having the engines in a puller configuration bypasses most of the wobbly rocket issues.

As to your contract: mine the asteroid for what it's worth. Every ton of ore you extract will make the asteroid weigh one ton less. Your TWR will go up as you go, the thing will become easier to handle, but it will remain an E class for the purposes of completing your contract. If you're confident that you have enough fuel, you may just mine and discard the ore in order to shed mass.

[xtoro]

Pulling works much better with a free pivot. Changing directions is completely different and presents similar problems as pushing when the asteroid starts spinning, but if done slowly it can be easily done. If my memory serves me, I also disabled engine gimbal since they were at the front and directions were inversed. Not sure if this is still an issue for the latest versions of KSP... But reactions wheels and thrusters work great.

[moronwrocket]

I did mine it dry, accidentally - the slag weighed over 200 tons, and I didn't have enough dV with the asteroid as-is to get it out of Kerbol orbit. So I added back the missing mass, and reduced its slag mass as well. Once I mine it fully, then I'll try launching it out.