...other than budget, obviously.

Our computers nowadays are good enough that we should be capable of sending a probe to a neighbor star system and back, and get some pretty good analysis reports. We do have batteries that would survive a hundred years before self-discharge renders them useless, and media to keep the collected data for long after that. The current roadblock seems to be speed of the probes — it would take much longer to reach that far than any of our equipment could survive in working order. We need better propulsion to receive the results in any reasonable time frame.

Let me ask the question first in a very short and non-constructive way: Why don't we have any better propulsion?

Now for something less subjective: What roadblocks do the scientists currently struggle against that keep the propulsion of our probes too weak to think practically of missions outside the Solar System? Are there any reasonable predictions or projects of propulsion systems that would considerably improve on what we have? Or is it just budget? Give it a fuel tank big enough and it will fly as fast as we desire? Or are there other considerations like safety in the case of nuclear?

  • 3
    $\begingroup$ In addition to the problem of propulsion, keep in mind that the farther you go out, the more path loss for the communication channel. For the return channel (i.e. probe to Earth), this means either the probe has to transmit with more power or the receiver on Earth has to be bigger and better. In simplistic terms, the amount of energy required per bit of communication will become a bottleneck (I believe, it grows by roughly by r^2). $\endgroup$ Commented Jul 25, 2013 at 8:40
  • 1
    $\begingroup$ @robguinness: There are two solutions to that: 1. The probe comes back, 2. we place "relays" on the way. They can either be segments left after the probe or we can cyclically launch one every few years to keep up with the probe, so that the whole chain travels towards the destination "unrolling" from Earth. $\endgroup$
    – SF.
    Commented Jul 25, 2013 at 8:47
  • $\begingroup$ This is true. But, of course, both options have a cost, so a detailed trade analysis would have to be conducted between adding simply adding more transmit power and adding either return capability or a relay infrastructure. My point was mainly that propulsion is not the only bottleneck for deep space missions. Voyager 1 is going farther and farther out into the edges of the Solar System, but eventually it won't have enough power to transmit meaningful information back to Earth. $\endgroup$ Commented Jul 25, 2013 at 8:54
  • 2
    $\begingroup$ I propose we change the title to 'what is the limits of acceleration of space probes currently' because the speed limitation is technically the laws of physics, the OP's issue is actually one of acceleration to a speed. $\endgroup$
    – user106
    Commented Jul 25, 2013 at 9:22
  • 2
    $\begingroup$ @RhysW: I wouldn't entirely agree. We have extremely powerful acceleration systems but they work for minutes at a time, so the speed gain isn't all that great. We need to make the probes move fast, give them high speed. Obviously the [strength of acceleration * time of acceleration] limits that speed. $\endgroup$
    – SF.
    Commented Jul 25, 2013 at 9:36

6 Answers 6


It's partly the same issues as the launch problem. If you put more fuel in the fuel tanks of the rockets, you then increase the mass. Then to lift that fuel you need to add a bit more fuel to lift that fuel, and so on and so forth.

A similar problem exists with the current propulsion system on probes but before I go into that I'm going to (very briefly) explain travel in space so we can understand the problem.

Travel In Space (assuming traveling in a straight line)

Travel in space is not the same as travel on land. Travel on land requires the constant burning of fuel to be able to replace the speed lost to friction, air resistance, etc.

Travel in space doesn't work the same way, it doesn't require a constant burning, it requires you burn enough fuel to propel the mass to that initial speed, then enough to burn in reverse to slow itself down at its destination.

(Which isn't half and half, it requires more burned fuel to speed up than to slow down, as the slowing down portion has less mass because we can discount the lost fuel burned to accelerate in the first place.)

Back to the problem

Ok so we could just add more fuel to the probes, but then we run into the same issue as the launch issue, adding extra fuel adds extra mass which requires extra fuel of its own to burn in order to propel the increased mass to the same speeds that we wanted.

So really what we want is a method of propulsion other than the existing liquid and solid process.

As you can see here, NASA already has some alternative ideas they want to try out for propulsion, I will briefly cover them below in case the link dies.

Nuclear Thermal Propulsion

Nuclear thermal propulsion – heats a fluid, usually hydrogen, in a high temperature nuclear reactor that creates thrust to move the rocket in space

NASA expects this type of propulsion system to be much lighter and a more efficient method of propelling ships in space.

However, every silver lining has a cloud and this is no exception. What currently stands in our way of using this system is the extreme difficulty to keep hydrogen in it's liquid form.

As you can see here, hydrogen needs to be kept at 20 Kelvin to stay in a liquid form. This proposes many technical challenges, first to reduce the temperature to such a level, then again when trying to stop the liquid fuel from heating up from the high temperatures of the exhaust!

And don't be mistaken, the technical problems with using liquid hydrogen isn't for lack of trying. in fact the idea to use liquid hydrogen as a fuel has been around since at least the 1950's!

Plasma based propulsion

NASA is also investigating a plasma based propulsion system called project VASIMR.

The idea is to use a nuclear reactor (again) and hydrogen (again) to ionize the hydrogen and blast it through a magnetic nozzle.

Obviously this is very technically challenging but there is also the issue that the plasma has to be magnetically shielded form the hardware of the ship or it causes electrode erosion in the engines themselves.

(Excuse my lack of knowledge of the physics around how this bit actually works.)

Not to mention you would also need energy to power the nuclear reactors in each design.


So really, we use chemical based propulsion systems because the alternatives are technologically expensive and difficult. We will struggle to make the chemical propulsion systems propel more because of the multiplying fuel issue (unless we find more efficient fuels). But really the biggest issue isn't so much the propulsion, its the distance!

For example the space station currently orbits us at about 18,000 miles per hour, orbiting the earth once every 90 minutes.

The Apollo spacecraft which flew to the moon traveled faster than that, at about 24,000 miles an hour. These types of speed are inconceivable for travel on earth, being hundreds of times faster than any jet can go.

So really my argument rests on these points, fuel problem, lack of easy alternatives, cost of fuel, sheer distance.

  • $\begingroup$ No mention of solar-based propulsion systems? Of course, they start to become less effective as the probe moves away from the Sun, but they in turn gain more efficacy as they approach other stars... $\endgroup$ Commented Jul 25, 2013 at 8:57
  • $\begingroup$ @robguinness good point, i hadn't considered that, i will find some more information and edit it in $\endgroup$
    – user106
    Commented Jul 25, 2013 at 8:59
  • $\begingroup$ @robguinness you could argue we are technically already doing it, Juno is going to jupiter only through solar power i hear $\endgroup$
    – user106
    Commented Jul 25, 2013 at 9:04
  • $\begingroup$ Yes, solar-powered plasma thrusters have been used for many years, especially in HEO communication satellites. Another form is solar sail technology, which uses the momentum of photons directly to propel a spacecraft. This is much more experimental, but some demonstrator missions have been flown as proofs of concept. $\endgroup$ Commented Jul 25, 2013 at 10:13
  • $\begingroup$ The hydrogen boiling point problem is hard, but not unsolved. Hydrogen has been in use as a fuel for a long time, usually burning with liquid oxygen. The Space Shuttles used it too. $\endgroup$
    – Linuxios
    Commented Nov 21, 2014 at 19:48

Right now, the primary limit is that we're stuck using reaction drives, meaning you have to expend propellant mass to accelerate the spacecraft. So, your total ΔV (change in velocity) is limited by the amount of propellant you can carry and the efficiency of your engines as specified in the Tsiolkovsky rocket equation

$$\Delta V = 9.8 * I_{SP} * ln ( MR ) $$

where MR is the mass ratio

$$MR = {M_{spacecraft} + M_{propellant} \over M_{spacecraft}}$$

The Dawn unmanned spacecraft is using an ion engine with a specific impulse (Isp) of 3100 seconds, which is currently the most efficient engine in use that I know of. If the spacecraft carries its own mass in propellant ($M_{spacecraft} = M_{propellant}$, for a MR of 2) that means we can get a total ΔV of ~21057 m/s. Fast, but not interstellar travel fast. If the spacecraft carries 9 times its mass in propellant (MR = 10), we can get up to ~69953 m/s. Better, but still not good enough for interstellar flight. For a Dawn-like spacecraft to reach 0.01 c (~ 3,000,000 m/s) we'd need a mass ratio on the order of $5.0 * 10^{41}$.1

There's a practical upper limit to how much mass we can launch from the surface of the Earth, which constrains the amount of propellant we can send with the spacecraft.

There are two ways around the problem — one is to accelerate the spacecraft using EM radiation against a sail. There's a project called Breakthrough Starshot that's looking to use a ground-based bank of terawatt lasers to accelerate a gram-scale spacecraft to 0.2 c (59958491 m/s) in the space of about 10 minutes. There have also been numerous ideas of using a sail with the solar wind from the Sun.

The other is to create a true reactionless drive (such as the Alcubierre drive or the EmDrive), which has a number of issues (not least of which is how you get around conservation of momentum and things like that).

  1. The fact that ΔV goes up with the logarithm of the mass ratio is what makes the rocket equation so tyrannical. Every kg of propellant you add has to be accelerated along with the spacecraft, leading to diminishing returns. That's why we didn't use rockets to slow down the Apollo spacecraft or Shuttle orbiters before re-entry; it just translated into too much mass to launch.


Simply put: Fuel and drive efficiency.

There is a maximum amount of vector change possible based upon the fuel load aboard, and the efficiency of the drive in converting that fuel into a change of vector.

To accelerate (which is a synonym for change of vector) for a given time requires a set amount of fuel. All that fuel needs to be aboard, and there's no way to refuel effectively in flight.

So, the fuel and the drive efficiency combine to create a maximum total vector.

And total vector is a synonym for speed.


It is limited by how powerful and dense we can make lasers. A laser sail concept skirts the issues of propellant and the rocket equation, and so offer the promise of the highest speeds attainable. Of course, it wouldn't be easy.

  • 1
    $\begingroup$ Not quite--neither power nor density actually matter because nothing says you can only have one launch laser. You can pile up as many as you need, the limit is beam coherence. $\endgroup$ Commented Aug 25, 2016 at 22:51

The delta-V a space vehicle can achieve depends on the rocket equation. It comes down to the proportion of total mass available to be expelled as propellant, and the velocity of that expelled mass. The velocity of the expelled mass depends on the amount of stored energy which can be converted into kinetic energy.

Chemical rockets have only so much chemical energy stored in the reactants (which are also the propellant mass), which determines the limiting velocity of expelled propellant, which imposes an upper limit on the achievable delta-V of the vehicle.

Switching to a different type of propulsion in which the propellant mass is expelled at much higher velocity can theoretically permit a vehicle to achieve much higher velocities. Of course, it presupposes the availability of an energy source which can store much more useable energy per unit of fuel mass - such as nuclear, and raises the question of just how to accelerate the propellant.

VASIMIR is one example of a thrust technology which expells its exhaust gas at far higher velocity than achievable with chemical propellants. It may not be the technology to propel an interstellar probe (several factors appear to limit the delta-V it could apply to a spacecraft), but some other technology aiming to do the same thing (very high velocity exhaust) just might.

To some extent it does come down to money - to fund research into and development of new propulsion technologies. But it also comes down to basic physics.


Very generally speaking, a reaction engine works by throwing energy out the back to get motion towards the front. Now, according to high-school physics, the energy we get for throwing a chunk of mass (Propellant, in our case) is

$$ Ke = {MV^2 \over 2} $$

So, the faster we throw out the mass, the more energy per-unit-mass we get. In rocketry, this is called Exhaust Velocity($Ve$), with a derived quantity called Specific Impulse ($Isp$). Higher velocities equate to better efficiency, up to the absolute maximum velocity of $c$, at which point high-school maths break down and the equation starts looking more like $$ Ke = {MC^2 \over sqrt(1-(V/C)^2) } - MC^2 $$ Notice how I said Energy instead of Mass. Photons move at the absolute maximum velocity the universe will allow, and are therefore the perfect propellant for a rocket. Technically speaking, a simple flashlight is the absolute apex of rocket technology.... Except for one tiny issue.

While photons pack the maximum possible kinetic energy per unit mass (How photons have kinetic energy without having mass is beyond me, but they do. Let's just call it mass for simplicity), their mass is immeasurably tiny. Your flashlight might be the absolute apex of Efficiency, but its actual thrust is practically nothing. It would take years for you to notice your flashlight had moved at all, which brings us to another issue.

While it would take years for your flashlight rocket to start moving, it would take mere hours for the batteries to die. To give our photon rocket the lifespan and power to do anything at all, we'd have to use a small nuclear powerplant to fuel it. With all that extra mass, our already tiny acceleration is squashed by hundreds of tons of reactor.

Technologies like Ion-drives and VASMIR face similar limitations. In order to deposit enough energy into their reaction mass to be effecient, they have to cut down the propellant flow to a tiny trickle of what it could be, meaning they have very, very low thrust. In addition to that, they also require large amounts of electrical current, meaning they face the same issue as our nuclear-powered flashlight rocket.

Taking all that into consideration, the holy-grail of rocketry would be a high thrust, high efficiency engine. There's only a few currently-theoretical contenders for the title, such as the Zurbin NSWR or Project Orion. Most if not all of them have some fairly serious drawbacks, and because one of them involves Using nuclear weapons for propulsion it's unlikely to receive funding any time soon.


Not the answer you're looking for? Browse other questions tagged or ask your own question.