We might be able to do very slightly better with a few different chemical fuels, but only slightly. And those slightly better fuels would have significant drawbacks. Using liquid oxygen as your oxidizer is one thing, using fluorine is quite another.
If we want more energetic rockets we need rockets that aren’t powered by chemical reactions.
Note that the ion engines mentioned before have a very high specific impulse, which is good. Specific impulse measures how much thrust is produced for a given amount of reaction mass, so it would be the equivalent of miles per gallon in a car. High specific impulse means you get lots of bang for your buck. The problem is that specific impulse, contrary to what was mentioned above, is not the only consideration. A very efficient engine that puts out a tiny amount of thrust has a very high specific impulse, but it wouldn’t be able to lift itself out of Earth’s gravity. Think of the high MPG cars we have–they get good MPG by sacrificing performance. Muscle cars are very wasteful of gas, but they have very good performance.
So if you know the specific impulse of your rocket engine, and you know how much reaction mass you have, then you know exactly how much change in velocity you have. And it turns out that to lift yourself into Earth orbit requires a ludicrous amount of delta-v. The only way to get more delta-v is to add more fuel, but fuel has mass, and so when you burn the extra fuel most of that energy is used to move the extra fuel. And so you have the Saturn V rocketswith gigantic amounts of fuel all to lift a relatively tiny service module and lunar module to high orbit.
I’m only partly through Rocket Propulsion Elements, so I can’t give a full answer yet :).
That said, let’s do a slightly more precise theoretical calculation. The SSMEs use a mixture of ratio of 6:1, not the stochiometric ratio of 8:1. As I mentioned, this has practical benefits but does reduce the theoretical limit.
So the combustion process looks something like this:
3 O[sub]2[/sub] + 8 H[sub]2[/sub] = 6 H[sub]2[/sub]O + 2 H[sub]2[/sub]
Only 6 of those H[sub]2[/sub]s get burned, so our fuel to combustion product ratio is 12/112 = 10.7%. H[sub]2[/sub] releases 143 MJ/kg of energy when burned, so we get 15.32 MJ/kg in the combustion products.
Doing a KE calculation, we get that 15.32 MJ = 0.51 kgv^2, so v = 5535 m/s. Divide by 9.81 m/s[sup]2[/sup] for a 564 s Isp limit.
Therefore, the SSME is actually achieving 80% of its Isp limit given its mixture ratio.
One thing I realized about the Isp chart you linked to earlier is that it assumes you have a working fluid available. Jets (turbofans especially) get huge efficiency gains by virtue of them using large amounts of air as reaction mass (and not just oxidizer). For a given thrust, increasing the mass decreases the necessary exit velocity, but since the energy requirement goes down with the square of velocity, you gain efficiency.
Incidentally, one particular nozzle inefficiency is in cosine losses–the fact that the thrust is not exactly in line with the direction of the rocket. However, this is generally a fairly small amount; on the order of 1%.
As far as chemical rockets go, hydrogen-oxygen may be the limit of what could ever be practical. There are only so many elements and so many ways of combining those elements to yield energy. Simply because hydrogen is so light, reactions involving hydrogen are going to be among the most energetic on a per gram basis. There’s hydrogen-fluorine of course and some serious research was done on it; but in addition to being hideously dangerous, fluorine is rarer and more expensive than oxygen and the exhaust product would be hydrogen fluoride- a deadly corrosive gas in its own right. Wikipedia claims that the oxidation of beryllium to beryllium oxide is the most energetic chemical reaction by weight, but there’s no practical way you could ever use that as rocket fuel.
Beyond that there’s there have been proposed exotic metastable states of different elements. For example ozone (O3) would be a more powerful oxidizer than oxygen (O2). There’s been interest in atomic hydrogen (as opposed to common molecular hydrogen, H2). The problem with all of these is that word “metastable”, as in “prone to spontaneous explosive decomposition”. The more energy a metastable material can release, the inherently harder it is to keep stable. Most of the proposed exotics would explode if not kept refrigerated to within a fraction of a degree of absolute zero; whereas hydrogen and oxygen are perfectly stable so long as they’re kept separate.
It’s just barely possible that by some clever trick a highly-energetic metastable exotic could be safely stored and used as a rocket fuel, but I’d say the odds are against it.
And other (also barely possible) metastability candidate is metallic hydrogen. Hydrogen enters a metallic phase under extreme pressures. It has not been completely ruled out that the phase might persist upon release of the pressure. If so, it presents a possibility for very dense hydrogen storage (not to mention the extra energy from the pressure release).
There are straightforward ways to get orders of magnitude more performance, right now, with technology we already have (in terms of, we have all the pieces, they just aren’t assembled and billions of dollars in engineering work has not been done).
It’s actually really simple. You separate the tasks of supplying mass to throw out the back of your rocket and supplying the energy to propel that mass.
For leaving earth, you bolt an inert block of metal onto the bottom of the rocket. You then use a very large array of lasers to ablate material off the metal, giving you thrust and 2000 ISP or better (hydrogen is only ~400 ISP). This is possible because the lasers that provide the energy sit on the ground and do not have to be carried by the rocket, and if you concentrate a lot of them onto a single target plate, the temperature of the material leaving the rocket can be incredibly hot.
For travel in space, you do the following :
For leaving somewhere, you do a trick similar to the lasers. You fling little iron bb sized pellets at incredibly high velocities (tens of kilometers per second) with an electromagnetic gun mounted onto something in vacuum like the Moon. Your ship doesn’t carry rocket fuel at all, just a line of superconducting magnets to slow the pellets down, transferring momentum to your craft.
For slowing down, you can either :
a. Use a planet’s atmosphere to slow down for free (not possible if you are going too fast, though)
b. Haul some kind of nuclear electric engine on your craft. You already have a line of superconducting magnets and have been catching pellets, so you could hang on to the pellets, and when you get close to the destination, fire up a nuclear reactor on your craft and use the energy to power the magnets to fling the pellets.
c. Use antimatter. For interstellar journeys, this may be the only way.
d. Have a pellet launching gun at the destination.
Let’s do a bit of math to keep things in perspective. A Falcon 9 launches about 13 tons into LEO.
At 2000 s Isp, that implies a mass ratio of about 1.6:1, or 20,800 kg. Suppose we want to accelerate at 15 m/s^2 (about 1.5 G). A bit slow, but usable. That’s a force of 312,000 N.
If our exit velocity is 20 km/s, then we need a mass flow of about 15.6 kg/s. Energy for one second of thrust is thus 0.5*15.6 kg * (20,000 m/s)^2, or 3.12 GJ. Therefore we need 3.12 gigawatts worth of laser.
That’s without taking any inefficiencies into account. I’d be surprised if you could get more than perhaps 10% efficiency out of a real-world system; lasers are fairly inefficient to start with, and the energy transfer to the propellant mass is going to be far from perfect given all the plasma messiness that’s happening there.
All this is sustained over 10 or 15 minutes, so you can’t just use batteries or a capacitor bank. You pretty much need a good sized nuclear plant to drive the system.
I’m not seeing any objection in your post that is more than a mild incovenience. What I was assuming when I wrote the above post was :
You have a civilization that actually cares about mass access to space, and sending large amounts of people and cargo between planets. This means the civilization spends 5-10% of it’s entire GDP, between all wealthy nations, on the effort.
10-20 years of development time from the day you raise NASAs (or, ideally, several competing groups with funds awarded based on results) budget to a trillion or so a year.
I consider that to be “possible within known technology” : nothing in my post is impossible or even all that difficult compared to known methods and techniques.
Not a problem. You obviously would build the launch site somewhere with access to all 3 legs of the U.S. power grid. Here’s a good spot. The launch trajectory is straight up - and, conveniently, the Tres Amiga super station is not in a bad spot, surrounded by mostly empty wasteland on all sides.
You don’t build a dedicated nuclear plant, you just use excess U.S. power grid capacity during launches. This might be at nights, or possibly in the afternoon if solar power becomes a major thing. You’d have to be doing multiple launches a day to make use of all the capital investment - probably several hours of back to back launches every 15 minutes or so, daily. Rest of the time would be downtime and preparing the next lineup of payloads for launch.
I think ablative laser propulsion would be a very safe method of accessing space. A vertical steam catapult or something would fling the payload + fuel block at a low velocity into the air. In the event of a failure during launch, you would turn off all the lasers (just cut the power with redundant computer controlled switches), detach the payload from the fuel block with explosive bolts and deploy a parachute. There would not be a massive tower of fuel tanks and oxidizer to detonate. (you would still need RCS, so the spacecraft wouldn’t be totally safe, but it would be better)
Oh, a quick bit of figuring : if you do 13 tons a launch, 16 launches a day (every 30 minutes for 8 hours of excess electric grid capacity.) you’d have a capacity of 208 tons a day. You could launch an ISS worth of payload in 2 days.
Like I said, just putting things in perspective. It could be done, but in terms of scale it would be one of the top few things that humans have ever accomplished.
We’re just barely at the point where satellites can communicate with each other via laser. Such precision would be a prerequisite for a launcher, except that you need a few billion times the wattage.
As a point of comparison, the Manhattan Project used 1/7 of the total US electrical power in its heyday. So, scaled to today, that 30 GW doesn’t seem so bad. But it’s in the same ballpark, so I posit that the project would need the same kind of existential threat that prompted the Manhattan Project for it to gain public (or government) support.
At least a laser launch system could, in theory, get a payload off the ground and into orbit.
These super-efficient ion drives that were mentioned upthread have a thrust-to-weight ratio of less than one, so could never lift themselves off the ground, no matter how big you make them.
It’s not that bad. I’ve seen various rough estimates of the scale of the effort. The lasing modules needed are commercially available on the open market today - it’s nothing special, just a few billion worth of diode lasers. Each module would be combined to a single coherent beam and focused onto a mirror about 1 meter across (to get a small enough spot size during the launch). The total module would be about the size of a shipping container, with a window for the main focusing mirror, and there would be about 1000 or so of these identical subunits.
You don’t get your beam power out of 1 big laser, you get it out of 1000 or so smaller lasers, each of which is itself composed of thousands of commercially available modules that actually generate the light.
Anyways, there’s no commerce or government need to launch the equivalent of the entire ISS every 2 days - but if there were a *need *for such a capability, it could be developed quickly. There are much better methods for getting to space than the ones currently in use, which were originally developed for weapons.
You wouldn’t try to launch 13 tons at once, either - the proposal I saw was for a 1 ton launcher.
One ton ain’t much. It might be adequate for bulk cargo, where one could develop very cheap modules that shuttle fuel or whatever to some orbiting craft, but it’s barely enough even for microsatellites.
The laser modules alone will be in the ballpark of $10/watt, and I think you’d be hard-pressed to get the whole thing (from aiming systems to power distribution) for under $100/watt. So even a small-scale system would be something like $250B.
If we go with your guess of 16 launches a day, we have 5,800 tons/yr. That’s 450 F9 launches–quite a bit. But at that scale, with reusability, they should be down to $30M/launch at the least. So the $250B is only earning $13.5B a year–about a 5.4% interest rate. Just barely breaks even given typical business rates. And that’s without any extra overhead.
Given the other downsides it’s hard to see how it would be worth it, at least unless the costs can be dropped by an order of magnitude. Maybe my guesses are too high but I don’t think I was particularly pessimistic.
I suppose that if you really optimized the hell out of your system, you could run it 24 hrs a day and get it down to maybe 20 min per launch. That’s a factor of 4.5x on the numbers I used, and could make it practical. Short of building a generation ship, though, I’m not sure what we’d use that capacity for.
Fair enough. So laser launch only really becomes an effective idea at a certain level of scale, since, obviously, if you were building big enough, those laser modules would be produced on such a scale that they wouldn’t cost so much per watt. Your $100 a watt estimate probably wouldn’t be accurate once you are buying hundreds of billions worth - you can subdivide the cost into the engineering and R&D costs (several billion for developing the system, tens of billions for building large scale automated factories to produce the final design) and the materials cost (probably pennies per watt).
Similarly, at a big enough scale of launches - back to back 24/7, I guess, at the cost estimate you gave, it would probably be cheaper. At $100/watt, it would be worth it to buy generators and have dedicated power plants to feed this thing. A natural gas burning combined cycle power plant costs only about $0.91 a watt electric for the actual generators and turbines and building to put them in, or $9.10 if you are assuming only 10% efficiency from power grid to lasers.
It’s been pointed out for a long time that space is the ultimate giant economy-sized package: the more you’re willing to spend in total, the lower the unit costs become. This graphic shows the vast potential for reducing the size and weight of a round trip to the moon if a lunar fuel factory and orbital depots were established. The problem is that it makes 100 trips to the moon cheaper, not one; and at least until now the demand simply hasn’t been there.
Yeah, I think to make it work, you have to assume that you can drive the costs down somehow. As you point out, it seems reasonable–the modules are all the same, so it should be possible to build some big factories to churn them out. It’s not obvious that this has to be the case, though.
As Lumpy said, space is the ultimate economy of scale industry. The issue with this proposal and similar ones is that any kind of launch requires a really high peak power, and you have to build your machine for that peak (making it expensive). Expensive enough that you have to run it continuously to make it worth it… but you then need the demand for it.
Anyway, it’s not to say that it’s a bad idea or anything, just that it requires a particular set of economic circumstances which don’t seem to apply today. Could be that things are different in 20 years–if SpaceX’s resusability efforts take off, they might grow the industry enough to reach the threshold we need for a ground launch system of some kind. Laser launch does seem to be one of the more practical ones.
Hey, maybe we can sell it to the military as an anti-ballistic-missile defense. It would do a hell of a job at that in its spare time…
Another subtle factor that would probably make a gigantic difference : the lasing modules are going to sit on the ground. It doesn’t matter if a percentage of them fail prematurely, because you would have excess laser capacity to make up for this.
This is a radically different point on the quality : cost curve than the prospect of building rocket cores. You can make things that work-ish for orders of magnitude less money than things that have to always work or bad things happen.
Essentially, with the laser part of laser launch, you don’t care if redundancy requires you to buy a lot more heavy equipment, because the rocket does not carry this equipment.
Of course, the heavy use of laser launch would also change the economics for building spacecraft and satellites. If you can launch a replacement cheaply and conveniently, you don’t have to spend nearly as much money on R&D and construction.
A rocket works on newton’s third law, For every action, there is an equal and opposite reaction. So rocket to go up, mass has to be toss out the back. The more mass being toss out and really fast the more thrust.
So for rocket to have high thrust you have two option for high thrust.
toss that propellant mass out the back with a lot of velocity.
Or
2 toss a propellant mass that as lot of mass.
**Where does energy come from? Well the mass of propellant being accelerated needs energy to be accelerated and we get this by fuel.
The more mass being toss out the back and really fast the more thrust more fuel needed . The problem with rockets is the fuel is used up really fast.
But we want that more mass being toss out the back and really fast for high thrust.
Is that possible? It sounds like it might be subject to some of the same optical constraints as have been discussed in relaton to whether it is possible to focus the sun’s rays into a point hotter than the sun.
By increasing the number of lasers you are increasing the number of light sources, and also the effective aperture of the laser. Yes, there is a limit to the temperature you can reach, but that increases every time you add a laser.
