Practise the mathematics behind this
The practice portal carries the algebra, logarithms, calculus and vector work this article uses, with worked solutions that explain the reasoning rather than just the answer.
Open the practice portal →A student asked me why a rocket is almost entirely fuel. The honest answer is four lines of algebra, and it uses exactly one idea from the school syllabus: the logarithm. Here is the whole argument, with every number checkable on a calculator, and the reason it decides the shape of every launch vehicle ever built.
Every so often a student asks a question that is worth more than the whole worksheet in front of them. Mine was fourteen and had been watching launch footage. She said: why is a rocket almost entirely fuel? It looks ridiculous. And she was right, it does look ridiculous. A vehicle the height of a building lifts off, and the part that actually goes somewhere is a cone the size of a car. This article answers her question properly, using nothing harder than the logarithms she was already being taught that term.
I want to be clear about what this article is. It is not a popular science explanation where you are asked to accept a conclusion. Every number below is one you can check on a calculator in about a minute. That is the whole point. The rocket equation is one of the few places where a piece of school algebra directly and unarguably decides what humanity can and cannot do, and a student who has followed the argument once tends not to forget what a logarithm is for.
A rocket has no road to push against and no air to bite into. It moves by throwing mass backwards. That is the entire mechanism, and it is Newton's third law with nothing added.
Suppose the rocket has mass and throws propellant out of the nozzle at a speed relative to itself. In a very short slice of time it throws away a small amount of mass and gains a small amount of speed. Conservation of momentum for that slice gives:
The minus sign is doing honest work: the mass is going down while the speed is going up. This is the only step in the article that needs calculus, and even here the calculus is only being used to say "add up all the tiny slices". Adding them up from the starting mass to the final mass gives the result Konstantin Tsiolkovsky published in 1903:
That is it. That is the whole equation. is the total change in speed the vehicle can produce, is the exhaust speed, and the ratio inside the logarithm is how much heavier the rocket is at the start than at the end.
Students often meet as a button on a calculator with no obvious purpose. Here its purpose is unmistakable, and it is worth stopping on.
The reason a logarithm appears is that a rocket gets lighter as it burns. The first tonne of propellant has to accelerate the whole enormous vehicle. The last tonne only has to accelerate the nearly empty shell. So each tonne of propellant buys more speed than the tonne before it. That is a compounding process, and every compounding process ends up with an exponential or a logarithm in its description. Interest on a bank balance, the decay of a radioactive sample and a rocket burning propellant are, mathematically, the same shape of problem.
Read the other way round, the equation says something harsher. Rearranged for the mass ratio:
Speed goes up like a logarithm, but mass goes up like an exponential. Wanting a bit more speed costs you a lot more rocket. There is no engineering cleverness that gets around this, because it is not an engineering fact. It is arithmetic.
Reaching a low orbit around the Earth requires roughly 9,400 metres per second of velocity change once you have paid for atmospheric drag and for fighting gravity during the climb. A good engine burning kerosene and liquid oxygen produces an exhaust speed near 3,000 metres per second. Put those in:
The vehicle has to start at almost twenty three times the mass it ends at. Turn that into a percentage:
Ninety five and a half percent of what stands on the pad has to be propellant, before you have accounted for a single bolt, tank wall, engine or passenger. This is the answer to the question a fourteen year old asked while watching a launch, and it took four lines.
Here is where the mathematics stops being a curiosity and starts being a design constraint that shaped the twentieth century.
Real vehicles need structure: tanks, engines, plumbing, avionics. Call the structural fraction the proportion of a stage that is not propellant. For a well built launch stage that number is around eight percent, and getting it much below that is extremely hard.
Now do the bookkeeping for a single stage going all the way to orbit. Propellant is 95.6 percent of the vehicle, so the structure needed to hold that propellant is about 8.3 percent of the vehicle. But everything that is not propellant, structure included, has to fit inside the 4.4 percent left over. It does not fit. The structure alone is nearly twice the entire budget.
You can work out exactly how good the structure would have to be. Setting the structure mass equal to the whole remaining mass gives a break even structural fraction of 4.36 percent. Below that, a single stage can reach orbit with some payload. Above it, a single stage cannot reach orbit at all, no matter how large you build it, because scaling the vehicle up scales the problem up with it.
Staging solves this by an idea a ten year old can grasp: once a tank is empty, stop carrying it. Split the same 9,400 metres per second across two stages and drop the first one when it is done, and the numbers change completely.
Same engines, same structural quality, same total velocity change. The only difference is that the vehicle stops carrying dead mass. That single change turns an impossible mission into one that delivers about two percent of the launch mass to orbit. Every orbital rocket ever flown, without exception, is staged, and this is why.
Look once more at where each quantity sits in the equation. The mass ratio is inside a logarithm. The exhaust speed is a straight multiplier out front. That asymmetry decides careers.
Doubling the mass ratio from 20 to 40 multiplies your velocity change by (40) divided by (20), which is about 1.16. A sixteen percent improvement for a vehicle twice the size. Doubling the exhaust speed doubles the velocity change outright. This is why propulsion engineers chase exhaust speed relentlessly and why nobody wins a career by shaving grams off brackets.
You will often see exhaust speed written in a different currency, specific impulse:
where is quoted in seconds and is 9.81 metres per second squared. It carries no extra physics. It is the same number in different units, and it exists mainly because it comes out the same whether you work in kilograms or pounds. A kerosene and oxygen engine sits near 300 seconds at sea level, hydrogen and oxygen near 450, and an ion thruster can exceed 3,000, which is precisely why ion engines are used where there is time but no mass to spare.
One more consequence worth seeing, because students almost always guess it wrong. If a mission needs two separate burns, the velocity changes add but the mass ratios multiply:
Two burns of 3,000 metres per second each do not cost twice what one costs. They cost the square of it. Mission planners live inside that fact.
A student meets logarithms somewhere around fifteen or sixteen, usually as a set of three laws to be memorised and applied to questions about pH and compound interest. The laws are correct and the questions are fine, but almost nothing in that treatment tells a capable student why anybody invented the idea.
The rocket equation does. It uses one logarithm, and in return it explains the shape of every launch vehicle ever built, why a Mars mission is harder than a Moon mission by more than the distance suggests, and why the Saturn V had three stages rather than one. A student who has worked through it has done something that is genuinely rare in school mathematics: they have used a function to predict a fact about the physical world that they did not already know.
It also produces the right kind of discomfort. The equation does not care how much you want the answer to be different. That is a useful thing for a bright fifteen year old to meet, and a worksheet on compound interest never quite delivers it.
It says the top speed a rocket can reach equals its exhaust speed multiplied by the natural logarithm of how much heavier it is at the start than at the end. In practice this means speed grows only logarithmically with how much propellant you carry, so each extra bit of speed costs disproportionately more rocket.
Reaching low Earth orbit needs roughly 9,400 metres per second of velocity change. With a typical kerosene and oxygen exhaust speed of 3,000 metres per second, the rocket equation gives a mass ratio of about 22.9, which means the final mass is about one twenty third of the starting mass. Everything else, about 95.6 percent, has to be propellant.
Because carrying empty tanks costs velocity. With a realistic structural fraction of about eight percent, a single stage cannot reach orbit at all: the structure needed would be nearly twice the mass budget left after propellant. Dropping each stage when it is empty removes that dead mass, and turns an impossible mission into one that delivers around two percent of launch mass to orbit.
Specific impulse is exhaust speed divided by 9.81 metres per second squared, so it is quoted in seconds. It carries no extra physics. It is popular because the number comes out the same whether you work in metric or imperial units. Kerosene and oxygen sits near 300 seconds, hydrogen and oxygen near 450, and ion thrusters above 3,000.
Logarithms and exponentials, which are usually taught around ages fifteen to sixteen, are enough for everything except the single line where the equation is derived, and that line can be taken on trust the first time through. The arithmetic in the article can be checked on any calculator.
Yes. Reusability changes the economics, not the physics. A reusable first stage has to save propellant for the return, which means it delivers less velocity change to the upper stage, so the payload fraction falls. The equation is what tells you exactly how much it falls by.
The practice portal carries the algebra, logarithms, calculus and vector work this article uses, with worked solutions that explain the reasoning rather than just the answer.
Open the practice portal →