The tyranny of the rocket equation

A rocket has nothing to push against. It throws mass out of the back as fast as it can, and the reaction drives it forward. Everything it will ever throw has to be carried from the start, and carrying it costs more of it. That loop is the whole problem of spaceflight, and one short equation captures it.

One line, one logarithm

Konstantin Tsiolkovsky wrote it down in 1903. The change in velocity a rocket can make, its delta-v, equals the speed of its exhaust multiplied by the natural logarithm of its mass ratio: the mass it starts with divided by the mass left once the propellant is gone. In symbols, Δv = vₑ × ln(m₀ ÷ m_f).

Engineers quote the exhaust speed as specific impulse, in seconds; multiplied by standard gravity, 9.80665 m/s², it becomes a speed. An engine with a specific impulse of 300 s, typical of kerosene burned low in the atmosphere, throws its exhaust out at 2.94 km/s.

The logarithm is where the trouble starts. Make a rocket 90% propellant and, at 300 s, it can change its speed by 6.77 km/s. Push that to 95% and it reaches 8.81 km/s. Keep squeezing until propellant is 99% of the rocket, leaving almost nothing for tanks, engines or anyone aboard, and the answer is still only 13.5 km/s. Each extra slice of propellant buys less speed than the slice before it. Drag the slider below and watch the curve flatten.

What the trip costs

Low Earth orbit is not really a height; it is a speed. At 300 km up, a circular orbit needs 7.73 km/s, fast enough to go round the planet in about 90 minutes. A real launch has to buy more than that. On the way up the rocket spends minutes holding itself up against gravity and pushing through the air, and those losses usually add about 1.7 km/s, which is why mission planners budget roughly 9.4 km/s to reach low orbit.

Every destination beyond stacks on top. Raising an orbit from 300 km to geostationary height takes another 3.90 km/s, in two burns of 2.43 and 1.47 km/s. Leaving Earth altogether means reaching escape speed, 11.19 km/s at the surface before any losses. The table turns each of these into the mass ratio it demands.

Why it gets worse, not just harder

Because the speed grows with the logarithm of the mass ratio, the mass ratio grows with the exponential of the speed. For low orbit at 300 s the ratio is 24.4: for every tonne that reaches orbit, 24.4 tonnes left the pad, and 95.9% of the launch mass was propellant.

Halve the budget and the ratio does not halve; it shrinks to its square root, 4.94. Turned around, each extra kilometre per second multiplies the size of the rocket instead of adding to it. That is the tyranny in the title: the requirement compounds, because the rocket must carry the propellant needed to carry its propellant.

The trick of throwing things away

Tanks and engines are dead weight once the propellant they held is gone, yet a single-stage rocket hauls all of them the whole way. Suppose the empty structure is 8% of each stage, a lean figure, and the engines manage 350 s. A single stage carrying a payload of 2% of its launch mass reaches 7.96 km/s. Even with no payload at all it would top out at 8.67 km/s, short of orbit.

Now cut the same rocket in two, with four fifths of the propellant and structure in the first stage. The first stage gives 4.38 km/s and then drops its empty tanks; the lighter second stage adds 6.18 km/s. Same propellant, same engines, same payload, and the total is 10.57 km/s, enough for orbit. Each stage obeys the same equation, and their velocities simply add.

That is why every rocket that has reached orbit so far has used stages, and why boosters that fly back and land matter so much: they keep the advantage of shedding hardware without paying for new hardware on every flight.

Better engines, and their price

The other lever is exhaust speed. At 450 s, roughly what liquid hydrogen and oxygen achieve in a vacuum, the ratio for low orbit drops from 24.4 to 8.42, and the propellant share from 95.9% to 88.1%. A solid motor near 270 s would need a ratio of 34.8.

Electric thrusters go much further. At 3,000 s the same 9.4 km/s would need a mass ratio of only 1.38, with 27% of the spacecraft as propellant. The catch is thrust. An ion engine pushes with a force of a fraction of a newton, far less than its own weight, so it can never lift itself off the ground. It shines in space, where it can push gently for months on end. To leave the pad, thrust has to beat weight, which is exactly what the thrust-to-weight calculator checks.

What this model leaves out

The equation assumes a constant exhaust speed and no outside forces during the burn. Real engines are less efficient at sea level than in a vacuum, gravity pulls on the rocket for the whole climb, and the air drags on it at first. The loss allowance above stands in for all of that, and it is an estimate, not a calculation. Treat the numbers here as the clean, idealised skeleton of a launch, which is what the calculators compute.

Every figure on this page comes from running PhDino's own calculators, and the key values are checked against independent working each time the site is built. The calculators are below if you want to try your own numbers.