High-thrust asteroid propulsion

[Sharpy]

How to make a propulsion for an asteroid that doesn't take ages to change its orbit?

Yesterday I've been trying to move a Class E, 30,000 ton, and I barely got some 100ms delta-V into it over the course of some 2-3 hours, using two nukes. I tried Rhinos but the Klaw is not durable enough and once clamped to the asteroid, all stabilizing "enhancements" just fail, phasing through the surface. I managed to push it with triple Rhino, attached to flat MK3 tanks, as I undocked and just pushed against it - but getting it to push at the center of mass in any reasonable direction was about impossible. I tried landing legs around the Klaw and they just sunk into the surface.

Are there any smart techniques to improve the Klaw attachment to the asteroid - make the joint firmer? Or some other way to provide more thrust than what one gets off a couple flimsy nukes?

Edited August 4, 2015 by Sharpy

[FancyMouse]

30k or 3k ton? I thought it was 3k

drilling/ISRU is your best bet. No need to worry about fuel/dV etc, unless fuel is your mission.

Having said that, I've done stupid things like this back in 1.0.2. It took me two nights to bring it back to non-escaping orbit and I swear I will not do that again.

[Sharpy]

No, ISRU and fuel are not my problem. I have the ISRU, and I know how to mine not to deplete it rapidly. It's the time spent doing nothing but decelerating that is the problem. I'd have no problems with making the orbit through apoapsis refilling the tanks at 500x timewarp, then burning it all near periapsis for a minute or two using up the fuel, then repeat the cycle. But currently yesterday's 2h burn brought the apoapsis from halfway to Minmus down to Mun orbit level. I want to *play* the game for goodness sake. Instead I made and ate my supper, I took a bath, I cleaned the shower, I read a story, I washed the dishes, and if it goes on I'll get to vacuuming instead of playing KSP while the burn lasts!

[Laie]

A claw can well handle a Poodle and maybe two, but a Skipper is already too much. KR-2L at full thrust is totally out of the question.

First off, "there's a mod for that": see here for how it has been done using KAS

Personally, I recently tried it with Kerbal Joint Reinforcement installed and found that this was already enough to make the job doable.

Then... I haven't tried this myself, but how about attaching several engines, each with it's own claw? You can't place them perfectly, but tweaking the thrust limits should solve that.

Finally, if you don't need the fuel, you can just mine the asteroid and jettison the ore. The remaining rock will still be the same size class for purposes of contracts, but it will weigh a lot less.

[nigelvn]

try pulling the asteroid

[Sharpy]

Without an obscenely long "tether" this has the side effect of heating the asteroid with exhaust gas. Unless I set up the engines into a (not very efficient "umbrella" pattern.

But this may be the way. I've done some pretty strong long bars in the past, so it should be doable... if expensive, getting that kind of construction to orbit. Plus maneuvering a flexible "snake" 200 meters long is no fun. Or fun, depending how you look at it. Been launching a "snake" of some 150 engine nacelles suspended on 4 turbojets. The things it was doing in the air...

[Geher]

Without an obscenely long "tether" this has the side effect of heating the asteroid with exhaust gas. Unless I set up the engines into a (not very efficient "umbrella" pattern.

Or you don't pull or push the asteroid, but instead you BECOME the asteroid.

No seriously: Add a probe, some rcs and of course the claw to EVERY booster. Seperate every booster form your main craft and attach them with the claw to the asteroid.

I did that once and it worked surpringly well. Just try to get the boosters mostly parallel to each other or you will start spinning.

[Sharpy]

What about fuel routing?

[Geher]

What about fuel routing?

The last time I did this (probably around 0.25) I could pump fuel through the asteroid. I don't know if it still works in 1.0.

[Sharpy]

I can transfer fuel across asteroid through rightclick, but for the engines not to sway the thrusters should be short and stubby, not leaving much room for tanks etc. I'll need to think of a way to achieve - how to place the engines in front of the tanks and roughly flush with the Klaw.

[Geher]

I can transfer fuel across asteroid through rightclick, but for the engines not to sway the thrusters should be short and stubby, not leaving much room for tanks etc. I'll need to think of a way to achieve - how to place the engines in front of the tanks and roughly flush with the Klaw.

I don't think they have to be short and stubby. If they sway a littlebit to the outside, who cares? I think today after work I might try that out with a few Skipper engines.

[Laie]

The problem is that the swaying thruster acts as a gimbal on the rock. And if the engine's gimbal tries to counteract, the swaying will become even worse. That's why puller designs are so popular: pulling will automatically straighten the vessel and thus bypass many of these issues.

If you don't want to pull, It's totally sensible to move everything that creates forces (engines, reaction wheels) as close to the claw as possible. This will at least mitigate the issues.

Not quite on point, but this should serve to illustrate the problem:

[Claw]

Yeah, it's sort of the classic backing up with a trailer issue. If you have ever tried to back up a car or truch with a trailer attached to the back, you know it can quickly become a mess if not handled appropriately.

[Geher]

Okay my attempt didn't work so well. The craft always wanted to tip over in one direction. I could get it more stable by adjusting the thrust of one engine, but it's still far from optimal. The last time I did this it wasn't with an class E asteroid though.

[ajburges]

I am waiting on the 1.25 m core to unlock, but my plan for an asteroid capture system is such:

> 1 mine head: Drills, ISRU, solar, battery, claw, and a docking port sr. Nothing else.

> 4 control nodes: claw, large reaction, wheel, probe core, battery, solar, RCS, normal docking port.

> Power plant. 4 Nukes, 1 cabin (for engineer), 1 LF tank (limited to one to work around ISRU bug), port to attach to mine head, and ports to store confirm nodes. Absolutely no control authority.

The control nodes are solely responsible for attitude, the tug is just a "dumb" thruster limited only by alignment. As a bonus, on delivery the tug detaches to be replaced with the remaining modules of a base! The rest of the modules are needed regardless.

[Laie]

Don't forget to bring an ore bunker. It doesn't need to be large, but you need a non-zero ore capacity or the drills won't work.

The control nodes can have three to five large reaction wheels apiece; depending on the weight of your rock, that may still feel quite sluggish.

Leave the engineer at home, you don't need him on an asteroid. An umanned drill will yield 0.5 ore/second regardless of the ore concentration, which is just as much as the converter can take (or slightly more when creating LF only).

Be aware that mining faster than the converter can handle it will lead to overflowing tanks (ore being dumped into space).

[Geher]

Be aware that mining faster than the converter can handle it will lead to overflowing tanks (ore being dumped into space).

Really? No. I tried this and when the ore tank was full and the drills were still mining, the ore inside the asteroid also stayed the same. --> No ore was dumped into space.

So the only reason to turn off the drills would be to save electricity.

[Teilnehmer]

Really?

That was a bug in 1.0.2. The ore sometimes disappeared very quickly when ore tanks were full.

I don’t know if the bug was fixed in 1.0.3/1.0.4.

[Claw]

I don't know if it was fixed either. I believe the problem was if the setup was such that a small amount of ore was converted to fuel, then a larger amount was mined, the overage on that "game cycle" went to waste. So what would happen was rapid draining with small fractions of excess being thrown away very quickly.

Cheers,

~Claw

[Sharpy]

Yes, the idea AFAIR was that if there's room for 0.1 unit of ore in the ore tank, then the drill picks 1 unit, loads up 0.1 into the tank and discards the rest.

The proper cycle is either to fill tanks, then stop or to manufacture faster than you're mining and burn faster than you're manufacturing.

[Laie]

Yes, the idea AFAIR was that if there's room for 0.1 unit of ore in the ore tank, then the drill picks 1 unit, loads up 0.1 into the tank and discards the rest.

Exactly. This can only bite you if a) you run drill and converter at the same time, and you mine faster than the converter can handle it:

http://forum.kerbalspaceprogram.com/threads/128707

Don't know about you, but I tend to run the equipment all at once; and on asteroids, it's very easy for drills to outpace the converter.

[Sharpy]

Exactly. This can only bite you if a) you run drill and converter at the same time, and you mine faster than the converter can handle it:

http://forum.kerbalspaceprogram.com/threads/128707

Don't know about you, but I tend to run the equipment all at once; and on asteroids, it's very easy for drills to outpace the converter.

a) you drill, convert and burn at the same time. The moment ISRU fills the tanks, conversion stops and as result mining stops.

Unfortunately a) is the right way to go about capturing asteroids and considering their richness is very common.

Still, as long as you're aware of this, just bring some big ore tanks and drill only when needed. Not a big problem, just one to remember.

Also, if you bring enough gigantors, remember ISRU can simultaneously produce [Lf+Ox], [LqdFuel] and [Oxidizer] for nearly tripling its output. And power usage unfortunately.

(even if you have nukes, [Lf+Ox] + [LqdFuel] produces at 1.45 the speed of [LqdFuel] alone. Although in my case Vernors keeping the asteroid from spinning seem to consume just as much as the nukes).

Edited August 4, 2015 by Sharpy

[Mister Dilsby]

All of the above. Here's a practical example, my Class E hauler the 'Kranefly' pulling a ~2000t rock.

Other than an excuse to post a glamour shot, here's a few lessons I learned which I applied to this vessel:

--Be prepared for low thrust if you want to move a big rock. I mean, IT IS A BIG ROCK. Kranefly has six LV-Ns which I think gives it an acceptable ~0.03 TWR when pulling something this size. It takes several minutes to do a big burn. That's why you have to...

--Plan ahead to make your burns as small as possible. Adjust the asteroid's path when it is still months away from its destination. Do inclination burns at the high-altitude node. Make liberal use of gravity assists, etc.

--Always pull, don't push. But this means you have to:

(1) Make sure your engines are far enough away from the rock to operate. Otherwise they will not thrust, AND heat may blow up the asteroid!

(2) Get your ship lined up right. Step one, set the maneuver node and orient the ship in the right direction. Step two, target asteroid center of mass and WITHOUT CHANGING SHIP HEADING get docked to the asteroid. Yes you will have to fly around the rock on RCS to do this. Now your ship/asteroid rig is aimed in the right direction and you will have minimum perturbations as you tow. Ready to tow? Not yet! You must

(3) SET KLAW FOR FREE PIVOT and if you controlled from the klaw to grab, RESET CONTROL TO FORWARD PART OF SHIP. Note that any time you leave the ship and come back to it, the klaw may lock again and you have to make sure it's on free pivot all over again.

--Now you can tow wherever you want--note that every time you change direction on a rock this big, you have to undock, orient to the next maneuver, and re-acquire the asteroid. I remember sharing my v0.90 Kranefly with someone a while ago, and that lovely person said the ship was no good because he couldn't steer the asteroid to a new heading...using reaction wheels. :facepalm: IT IS A VERY BIG ROCK!

Hope this helps!

[Sharpy]

That was a bug in 1.0.2. The ore sometimes disappeared very quickly when ore tanks were full.

I don’t know if the bug was fixed in 1.0.3/1.0.4.

The ore isn't disappearing "very quickly" only at the standard mining speed (which is pretty fast for asteroids) - as if the tanks didn't have the upper limit. It's still present - I forgot to switch the drills off (two days ago) for a couple of minutes and 200 tons of the asteroid vanished into the void. I don't know if it was faster in 1.0.2 but this effect is definitely present in .4