Moving Asteroids

guest blogger john hare

I briefly got into another discussion about moving asteroids recently. It involved parking an eleven ton spacecraft next to the asteroid and letting the gravitational attraction between the two shift the asteroids orbit. Then when the spacecraft gets too close, use the thrusters to open the gap again. When I said that a piece of thread would have just as much pull, and that just as much propellant would be used in either case, it was suggested that I learn something about conservation of momentum. Without further explanation from the other guy, I just assume that it is another of those concepts over the intellectual head of this redneck. If the rock is dangerous though, we need to do something about it.

Thinking about the subject, with the eleven ton vehicle and fifteen year time frame in the article, I think much more efficient use of mass, time and money can be done. They suggest in the article that a half kilogram or so of force would be applied by this gravitational tractor concept. I’m going to try newtons this time to see if I can get them right. Fortunately there are several people here to straighten me out if I screw it up. 5 newtons force for an hour is 18,000 newton seconds  applied. In a day that is 432,000, and a year is 157,680,000, and 15 years is 2,365,200,000 newton seconds. That’s a real big number, maybe. The F1 Saturn engine averaged about 7,000,000 newtons, so in about 338 seconds, one of them could apply as much total force. The problem is propellant of course.

During the discussion the ideas of painting part of the NEO and vaporizing the surface for reaction came up. In other discussions nuclear bombs, electric engines and fancy flyby trajectories are mentioned. I think most people miss the point that most asteroids have enough internal energy to move themselves if properly persuaded. The rotational energy alone is more than enough to shift an orbit to safety if we are clever enough to tap into it for propulsion purposes. The ideas here are not new, just my interpretation.

An asteroid is a natural for a beanstalk. A tiny rock with 10 m/s escape velocity and a four hour day would have a NEOsync orbit at about 23 km. If the composition  contains enough steel, iron, or other tensile useful materials in attainable form, then an in situ material beanstalk is feasible. Here I am assuming that something can be extruded with material properties half as good as terrestrial rebar. A 100 m/s stalk tip velocity would be at 230 km which would be the total length of the system. With a taper ratio of two, a mass ratio of one results for tether to tip payload.


If a thousand tons of NEO material can be extracted for beanstalk use, then thousand ton payloads become feasible from the tether tip at 100m/s. It would seem that throwing a thousand tons at 100m/s would deliver a 100,000,000 newton impulse to the NEO. 24 such payloads would exceed the impulse delivered by the gravity tractor in the other article. One a beanstalk is set up though, sending dozens or thousands of payloads is just a matter of extracting and bagging NEO material.

The neat thing about a beanstalk much longer than NEOsync is that the energy needed to lift and throw the payloads is supplied by the rotational energy of the NEO itself. Once past NEOsync, the payloads fall out to the end point, and they are quite capable of lifting the next payload off the ground with a light spectra tether that masses a fraction of a percent of the material lifted. This spectra tether is separate from the beanstalk and just used for propulsion purposes, returning to the ground after each lift. There is no need for energy delivery to the vehicles, which are basically bags with tether brakes.

One the beanstalk is built and reaction mass collected, the operation waits until the proper orbit to throw the payloads/propulsion sacks to Earth or Lunar orbit. It seems likely that this window will be during perihelion and last a month or so. A thousand ton payload every four hours for a month puts 180,000 tons of material on trans Earth/Luna trajectory. At the same time, it delivers an 18,000,000,000 newton impulse to the NEO, which should move it out of the danger zone during the flyby that happens in a dozen years or so. This impulse, delivered earlier and more concentrated than the gravity tractor idea, should be proportionately more effective, even discounting the nearly eight times more power.

If harvesting is not wanted for some reason, and mission mass is the critical restriction, then one ton payloads could be done with a hundred kg spectra tether every four hours for the fifteen years suggested in the article. The thousand ton units seem more worth chasing and catching to me for a space faring economy.

The following two tabs change content below.


I do construction for a living and aerospace as an occasional hobby. I am an inventor and a bit of an entrepreneur. I've been self employed since the 1980s and working in concrete since the 1970s. When I grow up, I want to work with rockets and spacecraft. I did a stupid rocket trick a few decades back and decided not to try another hot fire without adult supervision. Haven't located much of that as we are all big kids when working with our passions.

Latest posts by johnhare (see all)

This entry was posted in Uncategorized. Bookmark the permalink.

16 Responses to Moving Asteroids

  1. A_M_Swallow says:

    Local gravity around the meteor is very weak so an iron ribbon may not need tapering.

    A rotating tether at 1 Astronomical unit from the sun will be going through a temperature difference of several hundred degrees each rotation, check that the ribbon material can handle this.

    A solid meteor may rotate the tether but a dust cloud will let it go. The effects on a mixture will have to be investigated.

    Does the meteor rotate in a useful direction?

  2. MG says:

    Connecting the six orbital elements of the NEO with the E-M system is not likely to be a common occurence. I am unclear how the hoisting tethers get back to the surface (if that is your intent) for another load. Also, why not go with a much longer tether (say, 20000km), accumulate mass, and then cut the long tether once? That will have a larger “Isp” than the system you propose.

  3. johnhare john hare says:

    Connecting the orbital elements doesn’t have to be a common occurence if you have 15 years head start as in the article I read. Spend a few years setting up the system before the window opens.

    The seperate hoisting tether could either loop back to the surface like a roller coaster chain, or be carried back with very small tether climbers.

    The longer tether would give much higher Isp if feasible. Feasible would include relative difficulties of in situ tether manufacture. If high grade steel strips can be cut out of a vein with little difficulty, then the longer the better. If making the steel is an expensive and time consuming process for a low grade product, then smaller with more distinct payloads is better.

    Using the long tether once negates much of the attraction of the tether. To me, one of the main attractions is the ability to use one manufactured article until it wears out. If it can be built once and used indefinately, then payback becomes very attractive.

    One strong point you make is that a higher Isp system would require less total material to be processed. This to would require intimate knowledge of the target NEO. If good information is not available prior to mission launch, then it might be best to plan for the worst and hope for the best.

  4. MG says:

    I think the mass of the tether would be so small that the mass of the mechanisms required for in-situ tether construction would dwarf one transported from Earth. The mass of the excavation equipment, even if it is a grappler that can push boulders, will likewise likely outmass a suitable tether.

    One other consideration… Once you create a binary object, especially if you can move the center of mass appreciably, you change the trajectory of the system, and likewise can significantly alter the future trajectory during any flyby. Indeed, a flyby is a really good time to jettison mass, because you get a bigger *oomph* from it when it is inside a gravity well.

  5. johnhare john hare says:

    Bringing a tether from Earth would optimize to smaller payloads due to tether mass. Excavation equipment mass would be an interesting excercize to look into. That could easily make or break the whole project. It is possible that an asteroid with a metal core or vein could have the tether cut out in strips like an apple peeler. The cutting equipment could get quite interesting.

    Cutting loose mass during a flyby is one of the ways I see of capturing material into the Earth Lunar system. Sooner or later I might go into detail on how I think we could use the Lunar relative velocity to reduce capture velocity to an acceptable level. Short version is that a mass approaching with an Earth relative velocity of 2 km/sec at infinity (beyond effective gravitational acceleration)could be timed and steered to approach the moon at 1 km/sec at Lunar infinity. Using crude tether manufacturing to turn the whole thousand ton mass into a long metal bar, the center of mass reaches closest Lunar approach at the 2.2 km/sec escape altitude. Center of mass is at 2.5 km/sec at this time. Size depending on the attainable length of the crude tether, an outer section is detached with the lower section staying in eccentric Lunar orbit.

  6. jsuros says:

    I like MGs point that lifting enough mass far enough out from the center of mass converts the angular momentum of the asteroid into a binary object whose individual trajectories are each different than the original system once you sever the connection. Of course, that’s a LOT of mass to lift up past 23km.

    Also, here’s a crazy thought. Why not use a solar kiln and a robot bulldozer to bake regolith into bricks and then build a regolith filled tower up off the surface of the asteroid? This would seem to be simpler than manufacturing extruded metal structures, nor would it rely upon discovering metallic bodies that could be cut up for a similar purpose. Baking bricks, stacking bricks, and pouring fill into a berm are activities we could make robots to do today. Could a brick tower reach 20 odd klicks up to an asteroid’s synchronous orbit? Might be fun to try.

  7. This is a cool idea I had never heard of before – thanks! How about this for a “make it practical” measure:

    The hardest part for me is the manufacture of the tether in situ. I don’t think that is really necessary, though! Instead of shipping a micro manufacturing plant, just ship up a pre-made cable. The first 24 km or so is just cable. The lander uses a small rocket to deploy the first 24 km of cable. Then the lander crushes some rock, puts it in a bag, and attaches the bag to the cable and reals it out some more.

    The basic idea is that the cable’s mass pulls the bags containing rock slowly past 23 km, at which point the bags of rocks help pull as well. Initially, the bags would hold only a few grams. When the end of the cable is reached, multi-ton bags could be used. Once the entire cable is reeled out, you drop it and get a large, instantaneous acceleration.

    This lets you land far less mass on the asteroid, and get a far higher Isp. The equipment is a simpler crusher and bagger. The cable would mass maybe 1/1000th the mass moved.

  8. Doug Jones says:

    Hey, this is a clever idea, John! I agree with David Summers, just send a tether system to the asteroid, it will pay for its own mass quickly once it is in service, and to figure out details of the deployment a practice tether or three ought to be sent to a non-threatening asteroid for practice, first.

    Bear in mind that the direction of thrust is not arbitrary, it is constrained to be within the plane of the object’s rotation. Somewhere along that plane, though, it is very likely that there is at least one direction that will give good results for moving the asteroid’s closest approach away from the Earth. (Typically the best is to thrust along the velocity vector to change the orbit period, this moves the impact point 3x faster than any crossrange thrust can acheive, and can be applied continuously to allow maximum impulse in the shortest time.)

    That the hurled objects might approach Earth is not particularly likely, with a 100 m/s delta-V they’d probably be wildly scattered. A fat tether with large payloads should not be needed, just keep a skinny one really busy.

    If you consider ten tons of xenon reaction mass used in an array of NSTAR thrusters with Isp = 3300 at 92 mN thrust each (, the total impulse would be about 3E8 N.s, requiring 20 thrusters to run for 5.5 years.

    The same impulse in 100 m/s gravel would require 3E6 kg of material mined (3000 tonnes), and to deliver the same impulse over the same 5.5 years would require a throughput of 0.018 kg/s. If you throw one lump on every four hour rotation of the asteroid, that has to be a 265 kg lump (or bag of regolith). If you have ten lumps in transit at any time, they only have to move up the tether at about 230km/40 hr, or a whopping 1.6 m/s.

    Forget a standalone tether plus clothesline. all you need is a circular clothesline-like loop, with a wheel at the bottom rotating it and the bags. A loop of tensioned line hangs into the sky, passing a big flimsy wheel around a hundred meters across turning every 200 seconds or so… every four hours a bag comes down on the cable, gets filled, and swings back up. Ten empty bags are coming down at any moment, ten full ones going up, and the wheel grinds out power for the scraper and bag stuffer.

  9. Eric Collins says:

    You can also improve your ISP and your system robustness by starting up multiple tethers.

    Unfortunately, it is very difficult to plan such activities without an accurate characterization of the actual mass, composition, distribution, and rotation of the targeted asteroids. What is really needed is a better understanding of these objects so that we can better plan to exploit them once we finally get the means and the motive to do something with them.

    To get us started along this path, I would recommend that someone begin thinking about how to design a class of cube-sat that can be mass produced and then sent out on many parallel survey missions. Each probe would have an ion drive so that many bodies could be encountered and assessed. The Dawn mission to Ceres and Vesta should provide us with some useful experience conducting this kind of survey operation.

  10. johnhare john hare says:

    As usual, I had several assumptions that I didn’t include in the post.

    One was that we were going to harvest the asteroid in the process of changing its’ trajectory to a safer orbit. My mental scenerio was that the launch is 15 years before potential impact just as in the gravity tractor article. Then a low delta V window from asteroid to Earth is located in the 3-6 year from mission launch timeframe. That 3-6 years is used to rendezvois, land, set up all the equipment, and mine material to send. During the launch window, when delta V change to intersect Eath orbit is under 100 m/s, a payload is launched every 4 hours as accurately as possible on a Hohman orbit to a Lunar capture system. Under these conditions, the number of possible launches is limited, so payloads should be as large as possible. Large payloads want large tether masses, which possibly leads to in situ tether manufacture. After this major effort, the orbit is checked to see if it now misses the Earth by an acceptable margin. It might be desirable to wait on the next window to send more material to the Lunar orbital factories, or thrust as much as possible to change the trajectory to something safer.

    If the only purpose is to change the orbit to something safe, then a much smaller tether becomes optimal with continous ‘thrust being applied in a constantly changing direction. You probably want the higher Isp available with an Earth manufactured tether, and the higher throughput available from the pure clothesline system. This even allows more than one throw per NEO 4 hour day, as pure propulsion could be scattered across 90 or so degrees since propellant is free and efficiency is less important. Perhaps a launch every 5 minutes for an hour and then a 3 hour break and repeat. This would allow an even lower mass tether per unit of thrust per hour.

    NEO survey is an absolute requirement. It wouldn’t do any good to send a regolith scooping mission to a solid nickle iron body with no dust. Or to depend on collecting iron from a body that has none.

  11. Mike Puckett says:

    Guys, REALLY, and I mean R-E-A-L-L-Y need to read that.

    Via anon at

  12. Mike,
    I’ve already read several of them a few weeks ago (you’ll recognize the name of the author of one of the papers…), but was going to wait to mention them until they were presented (and until I had some breathing time after our first LLC window).


  13. MG says:

    Hey… a Jon Goff is on one of the references. Do you know him, Jonathan?


  14. johnhare john hare says:

    Well I’m certainly not going to start making suggestions on the depot architectures, not being qualified and all. No sireee, not a word, haven’t a clue where to start, no ideas at all,…. but maybe.:-)

  15. David OHara says:

    To slowly move an asteroid, put a big inflated spherical reflector near it with half silvered. It focuses sunlight onto a small spot vaporizing whatever is there, even nickel/iron. This provides thrust over a long period of time. This is nothing more than vacuum evaporation as is done for making optics on earth.
    One also might use this to mine an asteroid by using the vapor pressure of different materials (melting point too) to evaporate different materials onto a large substrate.
    This big reflector would be similar to the old Echo satellite.

  16. johnhare john hare says:

    This has been suggested many times and is the type of complex problem I was trying to avoid. Stationkeeping on the mirror must be precise. Boiled rock/metal Isp really sucks and you are not getting much propellant mass flow, so the thrust takes a double hit. The mirrors are subject to fogging with the vaporized material. We are talking about megaton or better objects. Fractional newton thrusts are going to take a long time to get the job done, maybe more than we have.

    The sharp nudge from throwing kilotons of rock can easily dwarf any movement by the boiling rock thruster.

Leave a Reply

Your email address will not be published. Required fields are marked *