February 16, 2011
Lasers To Accelerate Rockets?

The problem with chemical rockets is that most of their fuel basically lifts other fuel. So the rocket fuel burns to lift fuel so it can burn to life other fuel so it can burn. Chemical rockets have a pathetically low payload to fuel ratio. Cheap cargo lift into space has to involve avoiding the need to carry all that fuel.

Overcoming gravity is not easy. Conventional rockets are 97 percent fuel and tanks. Even NASA's mighty Saturn 5 moon launchers had just 3 to 5 percent available for payloads.

NASA wants to send the energy up to a rocket using lasers as it ascends.

A new technology under study would use ground-based lasers or microwaves to zap a heat exchanger on the rocket, releasing more energy from the fuel. The heat exchanger works like a hot plate, spiking the temperature of the fuel to more than 3,100 degrees Fahrenheit (1,704 degrees Celsius), which significantly increases the rocket's thrust.

I've read similar proposals for powering vehicles designed to travel up nanotube beanstalks into orbit. These vehicles would be like elevators that would have their own motors for moving them up. They would need power to lift cargo. Either the beanstalk would need a superconducting cable to carry electric power to the elevators or lasers would need to be aimed at the elevator car. One way of using the power in that case would be via photovoltaics to convert it into electricity. Though the PV adds weight.

A third possibility for cheaper space launch would be extreme acceleration of a cargo (said acceleration would only work for non-living cargo) in a tunnel or mountain side accelerator to impart all needed lift before the cargo even enters the atmosphere. Again, the need to carry fuel would be avoided.

The beanstalk approach takes too much time for humans. The ground-based accelerator causes g forces too high for humans to withstand. So the laser approach aimed at rockets might some day turn out to be the cheap way to put humans into orbit. Their cargo can get shipped separately using the other two approaches.

Share |      Randall Parker, 2011 February 16 10:49 PM  Space Launch

Brett Bellmore said at February 17, 2011 4:22 AM:

The fundamental problem with laser powered rockets is the high capital cost required per launch pound. Estimated at something on the order of $1 billion for a system capable of 100kg payloads. This doesn't mean the system can't work, and be cost effective. It does mean that the system can't be cost effective unless you're launching literally millions of pounds of payload into orbit per year, so as to divide that initial capital investment over a lot of launches.

While I think that, given a low enough launch cost, we could eventually find a million pounds of worthwhile payload to launch each year, we don't have that payload now.

So, like all approaches to access to space which reduce the cost of individual launches by substituting a huge amount of infrastructure for onboard systems, laser powered rockets face the need for a kind of 'phase change' in our launch market. They're only justifiable economically by a market of the scale which can't exist without them.

That makes moving to them very difficult to accomplish.

Russ said at February 17, 2011 6:11 AM:

Why not ship your humans in water (noncompressible), and then use accumulations of said water to gradually act as radiation-shielding?
(Yes, I was a liberal-arts major, so if there's something seriously and obviously nonworkable about that, it's ignorance, not trolling).

Peter said at February 17, 2011 6:29 AM:

Apologies for this pet peeve. We use fuel for rockets in combination with an oxidizer, since we aren't building air breathers. The oxidizer is the "heavy" part. The LOX carried is 2 to 6 times heavier than the fuel in most cases. Also, if we aren't using a bi-propellant rocket to generate the energy, as in the laser rocket, we don't have any fuel, we just have propellant. So when we talk about the weight in either case we should say propellant weight.

Eric said at February 17, 2011 6:57 AM:

"The beanstalk approach takes too much time for humans."

Why? How long would it take to go up the beanstalk to get to the station at the top of the elevator?

David A. Young said at February 17, 2011 9:30 AM:

Well, it's 22,000 miles, so figure it from there. For example, if the elevator's average speed were 100 mph, it would still take something like 9 days. If you were willing (and had the power) to do constant acceleration/deceleration, that might make it more practical.

Brett Bellmore said at February 17, 2011 9:38 AM:

"Why not ship your humans in water (noncompressible),"

Because bones are denser than muscle, which is denser than fat. At the relevant accelerations, a water packed human would arrive in orbit as a layer of crushed bones, topped by a layer of crushed muscles, various internal organs, and the fat on top of the water.

And the problem with taking that long to orbit, is that you go through the high radiation Van Allen belt, and die of radiation poisoning by the time you get there, unless you're in a lead casket. Which would really be heavy.

The best option I've seen is the *rotating* sky hook, combined with an air breathing first stage to reach it.

Bruce said at February 17, 2011 9:53 AM:

We should still be building on Gerald Bull's work (despite his messy end) and shooting non-human cargo into space.

"by November 1962 the 150-kilogram Martlets were being fired at over 10,000 ft/s (3,048 m/s) (6,818 mph) and reaching altitudes of 215,000 ft (66,000 m)."


Nick G said at February 17, 2011 11:13 AM:

You'd only need to accelerate at .1G for about 20 minutes to get to 2,000 MPH. Is there any necessary technical limit to the speed of a beanstalk beyond the rate of power transfer necessary to rise through the gravity well?

Bruce said at February 17, 2011 11:34 AM:

Nick, friction. The elevator has to be attached by something like pinch rollers or else it flies away.

Nick G said at February 17, 2011 12:14 PM:

By design the elevator would be an extremely straight line. I would think a little bit of something magnetic would do it, like magnetic high speed rail.

Bruce said at February 17, 2011 3:18 PM:

No. The elevator would have to flex. Or it would break. And it would be made out of carbon nanotubes or whatever replaces those. And it wouldn't be magnetic.

Nick G said at February 17, 2011 6:19 PM:

Why would it have to flex? We're talking about a geo-stationary orbit, after all. I wonder if it would dip down into the lower atmosphere? Not doing so would protect it, and make things simpler.

As for magnetism, 1) we're not assuming that we can only use one material, are we? The nanotubes would provide structural support, but I should think there would be other components. Further, why can't nanotubes be magnetic? They seem to be mighty adaptable, electromagnetically.

PacRim Jim said at February 18, 2011 12:18 AM:

Why not wait a few decades until biotech and nanotech advance to the point that such journeys become feasible?

Engineer-Poet said at February 18, 2011 2:10 PM:

Oh, geez, another spammer tactic... Randall, please nuke the preceding comment (and then this line).

Graphene is both very strong and electrically conductive, so a ribbon-climber on a graphene ribbon could potentially use a linear induction motor for both propulsion and guidance without physical contact.

Woozle said at February 18, 2011 3:36 PM:

Just to reiterate the legitimate points on this subject:

Once you get out of the atmosphere (less than 100 miles), you're looking more at acceleration than linear speed, and I'm not aware of any reason that magnetic traction and suspension couldn't be used. There are technical hurdles, yes, but ultimately space elevators are the least wasteful method of achieving orbit, by a significant margin (probably orders of magnitude).

That said, 9 days (100 MPH) is not an unreasonable amount of time -- if the capsules have sanitary facilities, are stocked with adequate food, and have internet, who cares how long it takes? It may even turn out to be more practical to have a single belt rotating at a slower speed (say 200-500 MPH) to which the capsules gradually accelerate from a standstill, using either magnetic effects of some kind or mechanical traction. If it's 22 kilomiles to orbit, that's a travel time of 2-5 days. It currently takes 1-2 days to get to the US east coast from Japan, depending on layovers -- and you don't get a chance to sleep. It used to take many weeks to travel from Europe to the New World; a week's travel in near-perfect safety and reasonable comfort seems a small price to pay for access to the next frontier.

But I digress. I'll take whatever system we can actually build (technically and politically) soonest. Just because we go for one particular technology now doesn't prevent anything else from being built.

Engineer-Poet said at February 18, 2011 10:24 PM:

The problem with riding 100 MPH climbers is that you'd spend so much time in the inner Van Allen belt that you'd wind up sick or dead.  The weight of shielding you'd need is way beyond what a reasonable climber could carry.  Maybe boosting at 1.5 G would reduce the travel time enough to be safe, but I'd worry about details like tangential acceleration and whatever happens when you reach the speed of sound in the ribbon.

Lasers and microwaves have the virtue of leaving most of the heavy stuff on the ground (laser-launch using ice lenses is the end-stage of stripping the off-ground part down to Payload, Propellant and Photons, Period).  They're not as energy-efficient as climbers but we can afford to spend a little extra for sensitive cargo like squishy human bodies.

Woozle said at February 19, 2011 5:35 AM:

Engineer-Poet: how much weight? Or, more to the point, how thick would the walls need to be? Although weight is an issue, it's not the killer factor that it is with rocketry.

On reflection, though, I think the "moving belt" design is going to turn out to have other problems -- mainly because of the huge momentum involved; if anything jammed, you'd have serious problems. Self-accelerating capsules and a fast trip through the Van Allen belts would appear to be the best option. There's probably some way the cables can conduct the power for the capsules to use (and some way for them to pick it up through magnetic induction), which would be a much better way of using ground-based power to move them (vs. physically moving the cables).

Even given a faster trip through the Van Allen belt, however, I'd think we would want to have enough shielding to allow for repeated trips, so that's probably something that needs to be taken into account. I really don't see weight as that big of a problem; even if the capsule weighed as much as a Saturn V, it would still take far less energy to lift it to geosync via cable than it would to do so by rocketry. The main constraining factor with cables is how much weight the cables can carry -- and you can always run more of them if you need more load-capacity; once you've got the first one in place, the rest are relatively simple, as I understand it.

Now I have to go re-read "The Fountains of Paradise", because I seem to remember Clarke's capsules used physical traction and storage batteries -- and he would definitely have done his math and taken radiation hazards into account.

Brett Bellmore said at February 19, 2011 10:20 AM:

"Although weight is an issue, it's not the killer factor that it is with rocketry."

On the contrary, it's exactly the same factor: The strongest material we have in the lab, carbon nanotubes, is barely strong enough to build a skyhook, and only if you use a tapered cable, the equivalent of "staging" for rockets. This is no accident, it's because the strength of chemical bonds, which dictates the capabilities of chemical rocketry AND mechanical structures, is inadequate compared to the delta-V demanded to get into orbit.

You can't afford to hang significant hardware off the cable, and you can't afford to make the capsule 20 times more heavy for the payload launched. For exactly the same reason you don't put unneeded hardware on a rocket, and the Apollo astronauts went to the moon without radiation shielding, knowing they'd die if there was a solar flare during the mission: Chemical bonds aren't strong enough.

Being obsessive about weight is as important for skyhooks as for rockets, no less.

Woozle said at February 19, 2011 1:30 PM:

Okay, here's my understanding of how this works. I could be wrong; I Am Not A Rocket Scientist, just a long-time fan of space exploration and "hard" SF.

With an elevator you don't have to have a massive vehicle just to get a tiny payload to orbit -- because you don't have to carry your propellant (or even your primary power source, if my guess is right). (Laser-based rockets don't have to carry their power-source either -- I'll get to this in a bit.)

That's the rocketry "killer factor" I was referring to: every kg of payload translates to more fuel, plus fuel to boost that fuel part way, plus superstructure to contain the larger amount of fuel, plus more fuel to get that superstructure up as far as you need it, and so on. The amount of fuel needed consequently goes up on a greater-than-linear curve with the size of the payload.

That's why you need staging for rockets, to chop off that feedback loop so you don't end up needing an infinitely large amount of fuel for any payload over a certain size. When your payload is large to begin with, even if you are using staging you get to a point where every kilo of weight has a disproportionately large effect because of all the additional mass that must be boosted along with it. There's a similar effect for aircraft, but it's much worse in rocketry.

Yes, laser-based rockets solve part of this problem too, by not carrying their power-source -- but they still need to carry reaction mass. They also lose efficiency as they get further from the laser. They would still improve dramatically on the steep curve for traditional rockets (staged or not), but even so I don't think they come anywhere near the linear cost of a space elevator. They might be more feasible in the short-term, however, since vastly less infrastructure (and therefore capital investment) is needed.

Yes, a cable would be tapered for optimum mass distribution, and this is perhaps somewhat analogous to rocket staging -- but once you reach that critical point where the material can support its own weight (plus a safety factor) to GEO, it's just a matter of adding more material until you have sufficient strength to carry whatever payload you need. This isn't trivial, but it's doable (much easier than getting that first cable connected) -- and it's a one-time expense; once the infrastructure is in place, your cost-to-orbit (energy-wise) is some relatively non-steep linear function of the potential energy (from GEO to the Earth-end of your beanstalk) of whatever you want to put there -- versus a steep non-linear curve for rockets (and a somewhat less steep but still non-linear curve for laser-rockets). Much, much cheaper.

That said...

Before you can build a beanstalk, you've got to clear out all lower orbits (barring unforeseen technology) to prevent the cable from being severed. That's going to take some doing -- we'll need to have a significant presence in space to make this even possible. (The Russians reportedly have an idea about how to do this within the next couple of decades, but I don't know the details. I can find the news article if anyone is interested.) We'll also need to be able to serve from GEO, or by some other means, all the functions of satellites currently in LEO/NEO, unless there's some way to occupy those orbits with a vanishingly small chance of hitting the cable over a very long time-span. (There may be a way to reliably detect potential collisions and move the cable aside, but I haven't heard a convincing proposal for this yet.)

Also, you're going to need enough space presence to be able to manufacture the beanstalk in GEO. You'll also need a counterweight (which can be reeled in and out to adjust for current load on the cable), and your GEO station should be massive (built around a small asteroid, for example) so as to damp out small discrepancies between the current load and the counterweight -- all of which implies some serious mass-moving capacity up in GEO, which we currently lack even in LEO.

So we're going to need the laser-rockets -- or some other seriously-cheaper-than-what-we-have-now option -- just to get to the point where we can build the elevator, unless there's some solution I'm overlooking.

Engineer-Poet said at February 19, 2011 8:34 PM:

The problem with people on slow beanstalk climbers is that you need a certain amount of mass to stop the high-energy particles of the Van Allen belt (which are mostly protons).  Unfortunately, once you have enough mass to stop those, you've got this great spallation target for cosmic-ray primaries which multiplies THEIR effect quite a few times.  Once you've added enough shielding to stop THOSE, you've got a huge amount of dead weight (and a very heavy minimal capsule for the trip).

Laser launch has the benefit of very low recurring costs.  Jordin Kare has been working on this problem for a while, and IIRC he thinks he can produce a system to launch lots of small payloads for a very reasonable cost using semiconductor lasers.  This would not launch anything large, but if you want to move bulk commodities in small packages it's just the ticket.

Using lasers or microwaves to increase the performance of a rocket is a great idea.  Think of an RL-10 upper stage with a really big nozzle.  You burn some minimal amount of oxygen to turn everything to vapor and run the turbopumps, seed with some easily-ionized material like potassium, and hit the nozzle with several GW of radiant power.  Voila, a doubling of your Isp.

Woozle said at February 20, 2011 10:15 AM:

Having accepted that beanstalks are not a feasible tool for the next generation of launch vehicles or anything we can start seriously planning yet...

...I have to point out that your objection about the need for radiation shields for travel through the Van Allen belts poses a problem for any travel to GEO or beyond, not just via beanstalk.

Your objection also only applies to living matter; being able to cheaply lift construction materials and supplies into orbit (and return materials and waste without incinerating it) could still be well worth the price of admission.

Also, there is apparently a proposal to drain the Van Allen belts: http://en.wikipedia.org/wiki/HiVOLT -- I'd add that to the list of likely prerequisites for a beanstalk, along with clearing LEO/NEO and building up significant industrial presence in GEO.

Randall Parker said at February 20, 2011 1:26 PM:


Yes, to use a beanstalk to put people into orbit you have to shield the beanstalk elevator and provide food, oxygen, and other support for them. Better to move people into space very rapidly so that less shielding can be used. Most of the weight that needs to be moved into space to support people in space is not people anyway.

Paul D. said at February 21, 2011 5:08 AM:

Another problem with orbital elevators is the elevator capsule still needs rockets.

Why? Because in the event of failure (the cable breaks or the capsule falls off), there are altitude ranges where just falling back into the atmosphere will kill the passengers. The capsule, even with a heat shield, will hit the atmosphere fast enough, and at a sufficiently steep angle, that the deceleration will not be survivable.

The way to survive is either a longer deceleration burn so its hits the atmosphere slowly enough, or a sideways burn so it goes into orbit or reenters at a shallow angle. IIRC, the delta-V in the worst case is at least 4 km/s.

Engineer-Poet said at February 22, 2011 6:12 AM:

There are non-survivable accidents in most forms of transport; I doubt that it's worth designing for all of them.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright