Rockets of Today

— stats and specs explained —

For each rocket, I will summarize the rocket’s key specifications and statistics at the bottom, in an abbreviated format. This summary includes:

  • Mass: the total weight of the rocket stack with fuel but without payload, in metric tons (t).
  • Diam: the diameter of the primary cylindrical cross section of the rocket’s main stage, not counting protruberances and strap-ons, in meters (m). When side boosters are not optional, I will give a second figure in parentheses for the width with those included, again ignoring minor protruberances. Sometimes the second figure is inexact.
  • Thrust: the initial upward force exerted at launch, in kilonewtons (kN). On Earth it takes ten kilonewtons to lift one metric ton — to be exact, Earth’s gravity gives one metric ton a weight of 9.81 kN at sea level. (A kilonewton in English units is 224.8 pounds of force.) The narrower the ratio between the sea level thrust and the weight, the more fuel is wasted in the first minute of a launch. On the other hand, a wider ratio may mean that the payload is subjected to harsher G forces during flight. These numbers may be imprecise, and in some cases where data is scant, it’s possible that vacuum thrust may have been conflated with sea level thrust. When solid boosters are involved, initial thrust can also be confused with peak thrust or average thrust... in these cases particularly, published figures are often incomplete or contradictory, and the figure I give may be guesswork.
  • Imp: the specific impulse of the main first stage engine(s) in vacuum, in kilometers per second (km/s), because I consider the habit of expressing it in seconds to be misleading and bogus. This is the amount of momentum produced per unit of fuel, or the thrust produced per rate of fuel consumption. Its real units are meganewtons per ton per second, but that cancels down to a speed. For pure rockets it really is a speed, namely the average backward velocity of the exhaust gases leaving the craft. For air-breathing engines, or any other situation where the mass being moved backward is more than what the vehicle carries with it, it’ll be a much higher value than the observable exhaust speed. At sea level the values are typically at least ten percent lower than what the engine can do in vacuum, and the higher the figure an engine achieves, the more it loses in thick air.
  • type: next (with no preceding label) comes a phrase which designates the rocket’s power cycle as one from the following list, which is loosely sorted from the simplest design to the most sophisticated. These cycles are different solutions to the problem of how to feed fuel into a running engine, which is a challenge because the pumping power needed can be enormous. Hover over any type (or on mobile, touch it) to see an explanation of how it works:
    • solid fuel The oldest and simplest type of rocket motor consists of a large chamber pre-packed with a solidified blend of fuel and oxidizer. The fuel tank and the combustion chamber are one and the same. Once ignited, it cannot be put out. Thrust may vary considerably over the burn time. This is a feature, not a bug, but keep in mind that nominal thrust specifications for solid motors can’t be taken too literally; the value may be an average, a peak, or neither.

      In orbital craft, most common mixes use ammonium perchlorate as the oxidizer, blended into urethane — to be more exact, hydroxyl-terminated polybutadiene, known as HTPB — as the fuel and binder. Often there is also powdered aluminum or magnesium in the mix as additional fuel, and a few adventurous builders have been known to mix in high explosives for extra spiciness. The oxidizer makes up the largest part of the mix — around 70%. Some older solid rockets, such as those on the Space Shuttle, used polybutadiene acrylonitrile (PBAN) instead of HTPB; it has slightly better performance, but it costs more and takes a long time to cure, so its use has become rare even though its exhaust is less toxic. Perchlorate has toxicity issues too... a green alternative is ammonium dinitramide, but it hasn’t yet come into much use. We could use such an alternative, as current solid fuels produce far worse air pollution than any liquid fuel. (Fun fact: on Mars, there are perchlorates in the soil. This would make it difficult to grow potatoes there.) The exact formulations of solid fuels are often proprietary, involving many secondary ingredients. They are working on higher energy solid fuels based on boride compounds.
    • hybrid This type of motor uses solid fuel but fluid oxidizer. The fuel is typically some kind of polymer, such as urethane or nitrile or nylon. Typically the oxidizer is either forced into the fuel/combustion chamber by use of a pressurizing gas, or is itself a gas before compression, such as nitrous oxide. There may be significant wastage of oxidizer, and so far this has only been used suborbitally.
    • pressure-fed This is the simplest type of rocket to use fluid propellants. In this design, the fuel and oxidizer are forced into the combustion chamber from their tanks by the use of a separate pressurizing gas, as in a spray can. Helium is the most commonly used pressurant, because of its lightness. This is usually used for small engines with low thrust, because combustion has to occur at a pressure well under that of the fuel tanks. This is especially used in cases where engine startup has to be very quick and reliable, such as in reaction-control thrusters or propulsive landing rockets, which for the same reasons normally use hypergolic fuel. The disadvantage is that since the tanks are large pressure vessels, they have to be heavy.
    • electric pump Some new rockets of small size have fuel and oxidizer pumps driven by electric motors. This gives a lot of performance with a simple and robust design, but has a weight penalty because the rocket has to carry a pack of batteries. Apparently it does not scale well for large engines.
    • independent turbine Some of the oldest liquid-fueled rockets have fuel and oxidizer pumps which run off of a gas turbine which is not powered by the rocket fuel. It can use a pressurized gas, but more commonly it uses a separate gas-producing chemical, such as hydrogen peroxide. In these cases, such engines are sometimes officially described as using a gas generator cycle, but I count it in a separate category.
    • tap-off This is the simplest arrangement for having a liquid fuel rocket engine supply its own power to run the fuel pumps. A small opening in the combustion chamber pipes some of the hot exhaust to a turbine which spins the fuel and oxidizer pumps. The turbine exhaust is dumped out at low pressure, making this an “open cycle”. Like all turbopump designs, starting the engine requires spinning the turbine by other means before ignition.
    • expander Many liquid-fueled rockets use the fuel or the oxidizer as a coolant outside the combustion chamber and nozzle, in order to keep them from melting. In this engine cycle, that cooling step does double duty, as it also boils the cryogenic liquid used as coolant. This boiling produces enough gas pressure to power the turbopump.

      In the “expander bleed” version, unburned fuel or oxidizer is then dumped overboard, making it an open cycle. In the “closed expander” version, the exhaust of the turbine goes directly into the combustion chamber to be burned. This means it is limited to moderate combustion chamber pressures. A compromise version injects the expanded gases into the bell instead of the combustion chamber, making it semi-closed. The “dual expander” version has separate turbines and pumps for the fuel and the oxidizer, usually with both closed. One turbine may be downstream from the other, running at lower pressure. As yet, the dual version pretty much only exists on paper.

      The expander cycle cannot easily scale up for large engines or high levels of power, because the rate of heat transfer through the walls is limited by the square-cube law. The power limit is much tighter for a closed expander than for an open one. But where it’s applicable, the expander does a very reliable job, and the mild temperature in the turbines makes them durable. It works best with liquid hydrogen. For all of these reasons, it has mostly been used in upper stages, but recently it has started appearing on main boosters.
    • gas generator This is a very popular mainstream type of rocket engine — the default for large liquid fueled engines, so widely used that it powers the majority of orbital launches. It diverts a small portion of the fuel and oxidizer into a separate burner, which powers the turbine(s) that power the pumps. It is capable of high power output and high pressure, but subjects the turbines to tremendous heat, and loses some specific impulse because the turbine exhaust has to go out a tailpipe at low pressure. The heat level in the turbine may be mitigated by using a fuel-rich mix there, which worsens the tailpipe losses.

      A variant is to inject that exhaust back into the bell, making it a semi-closed cycle. The injection happens about halfway down, where the pressure is low enough. This helps protect the lower part of the bell from heat by creating a layer of comparatively cool nonburning gas along the outside of the expansion zone, and contributes a bit of additional thrust. The most famous engine to employ this injection design was the F-1 at the bottom of the Saturn V. The cool exhaust gas would create a visible curtain of dark soot around the flame, for a short distance below the bell. Sometimes the vacuum version of an engine will use bell injection though the sea-level version has a tailpipe; the Merlin used in the Falcon 9 in an example of this.
    • staged combustion The objective of this design is to make sure that all of the fuel and all of the oxidizer go into the combustion chamber at full pressure, thereby maximizing specific impulse. To do this, not all of it arrives unburned. What you do is pick one component, most often the oxidizer, and inject a small amount of the other — just enough to get into the stoichiometric zone where it’s capable of burning. This combusts in a preburner and then goes through the turbine, powering the pumps, and then the exhaust goes into the combustion chamber... but the exhaust is mostly unburned oxidizer. There it mixes with the remaining fuel. The pumps can run at extreme pressure, so that even the pressure on the drain side of the turbine is still very high — often higher than gas-generator engines ever achieve.

      The oxidizer-rich variant is a Russian specialty, which they use with both cryogenic and hypergolic fuels, and made a great success of at a time when American engineers thought it couldn’t be done. The other versions can only be used with fuels that form no soot, such as hydrogen or methane. With such fuels, a fuel-rich design is the norm. The most famous example is the hydrogen-fueled Space Shuttle Main Engine, or RS-25, which may have been the most advanced engine of the twentieth century.

      The “full-flow staged combustion” variant is the ultimate liquid fuel power cycle: it uses two separate turbines, one pumping oxygen with a bit of fuel and the other pumping fuel with a bit of oxygen, each powering its own side’s pump. Of all liquid fuel rocket engine types, this is the most complex and difficult, and has the highest performance. SpaceX’s Raptor is the first one of these to ever leave the ground.
    For liquid fueled engines, the engine type will be followed by a fuel type in parentheses, usually from the following list. If it’s combined with an unusual oxidizer, that will also be mentioned; if none is stated, then cryogenic liquid oxygen (lox) is the default for non-hypergolic fuels — for UDMH or MMH the default is N2O4. Hover or touch to see notes:
    • UDMH or MMH Unsymmetrical dimethylhidrazine (H2NN(CH3)2) is the most widely used of a family of fuels based on hydrazine (N2H4). They are called “hypergolic” fuels. This means means they need no igniter, as they immediately combust on contact with their matching oxidizer, which usually consists of one or more nitrogen oxides, the most common being N2O4. When more than one is used, the blend is often referred to as MON, for “Mixed Oxides of Nitrogen”. A common alternative to UDMH is MMH, or monomethylhydrazine (CH3(NH)NH2) — it is hypergolic with the same oxidizers. Sometimes different hydrazine-based propellants are blended. Even pure hydrazine — the most dangerous of this family — can be mixed in, as long as it is moderated by the others.

      These fuels are mainly used in top stages and small thrusters, because they make it easy to reliably start and stop an engine many times. These hydrazine fuels are less suited for heavy use in the atmosphere because unfortunately they are viciously toxic — like, don’t even let a drop of it touch your skin. (The exhaust is far more benign than the fuel, but still contains a significant amount of nitrogen oxide smog.) Crews who handle the fueling process practically have to wear space suits on Earth. Another key advantage is that they don’t have to be refrigerated. This makes them easy to store for months or years. Many military missiles have used the stuff because a rocket fueled with it can be kept ready and waiting for long periods, and it has better performance than solid fuel. And this convenience has in the past attracted some engineers to use it on orbital boosters, for example the Titan, the Proton, and the older models of the Ariane and Long March lines, but this practice is largely coming to an end.
    • kerosene Kerosene is similar to jet fuel or diesel oil, in that all three are mixtures of chain hydrocarbons of medium length. The kerosene used for rocketry is very highly refined, to reduce the presence of longer-chain molecules which can produce solid gunk. Most of the molecules are in the range from C10H22 to C16H34. It forms a liquid which combines relatively low volatility at room temperature with low viscosity. It is usually combined with liquid oxygen, meaning that the rocket’s fuel tank has to be kept at a much higher temperature than the lox tank next to it. This can be avoided with room-temperature oxidizers such as nitric acid or peroxide, at some cost in performance — an approach sometimes used in military missiles. Kerosene is convenient in the atmosphere but spells trouble in deep space missions because the fuel can freeze, so it is generally not used for that role. It’s difficult to ignite, which is a good thing in that it’s not prone to explosions if something leaks. Because of the low volatility, it may need to be pressurized with helium.

      Occasionally, rockets use turpentine, which is a bit lighter and more volatile than kerosene. Most of it is made of various isomers of C10H16, but it may also include a wide variety of more complex organic molecules. The hydrocarbons generally include benzene rings, unlike the linear molecules that predominate in kerosene; this makes them more reactive at room temperature, behaving as solvents rather than oils.
    • propane or propylene Here we have hydrocarbons with a few carbon atoms apiece, the most typical being propane (C3H8). This makes them light enough so that they evaporate rapidly at room temperature, so to be stored at room temperature they require compression to remain liquid, like the propane or butane sold in retail stores. When chilled they gain a lot of density and no longer require a pressure vessel. They are easy to ignite and not prone to forming soot or gunk. Propylene (CH3CH=CH2) burns hotter than propane, but is not as easy to obtain in bulk.
    • methane Methane is a hydrocarbon with just one carbon atom (CH4). It has to be compressed to very high pressure to make it liquid at room temperature, so for rocketry it is liquified by chilling it to very cold temperatures, similar to that of the liquid oxygen it combines with. Otherwise it cannot be stored in a light or compact tank. Because no two carbon atoms are bonded to each other, it burns very cleanly and cannot form soot. Methane is the principal constituent of the natural gas that’s piped to your house, making it very inexpensive. The density is not as low as you might think — about 80% that of room-temperature kerosene. (But then, kerosene can also be made more dense by chilling it.)
    • hydrogen Liquid hydrogen (H2) has to be chilled to even colder temperatures than liquid oxygen does — 20 kelvin rather than 80. Even at those temperatures it has a low density and takes up a lot of room, and it also costs more than other common fuels. Also, it is difficult to get high thrust from a hydrogen engine, so it’s rarely used at sea level unless accompanied by side boosters. But it gives a higher exhaust velocity when combined with oxygen than any other fuel. About the only chemical combination which could yield higher velocity would be to combine lithium with fluorine, and that would be ferociously toxic, as well as vastly more expensive.

      Extra hydrogen can be mixed into any other kind of fuel as a “tripropellant” mix, just because having unburnt molecular hydrogen in there lowers the average molecular weight of the exhaust and hence raises its average velocity. But this is cumbersome and complex to do in any case other than where hydrogen is already the primary fuel; in this case you just make the mixture fuel-rich and you get that velocity boost, plus an engine that runs cooler and has less stress.
    If the rocket requires side boosters for liftoff, and they differ in type from the core stage, I will add their type with “and”, for example “gas generator (hydrogen) and solid fuel”.
  • Payload: the maximum mass, in metric tons (t), that the rocket can lift into a stable low orbit. This is what determines a rocket’s size classification. This is followed, in parentheses, by the percentage ratio between the payload mass and the initial mass. This ratio tends to be somewhat proportional to specific impulse, and may tend to be better for large rockets than for small ones.
  • Cost: a rough estimated price for lifting a full payload to low orbit, in millions of US dollars per metric ton ($M/t). This may not be an easy stat to obtain, especially outside the USA, where the numbers may reflect currency exchange rates more than engineering costs. When a rocket has optional strap-on boosters, their use may improve this figure, as with their help the capacity increases faster than the cost.
  • Record: three numbers separated by slashes, the first being the number of successful launches with real payloads, the second being test flights that didn’t quickly fail, and the third being failures. (We define failure much more stringently if a real payload is involved.) This record applies to the listed model and closely related predecessors and siblings, but not major revisions; for instance, the record for the H-IIB includes the H-IIA and H-II but not the original H, and the Long March 3B record includes the 3C and the original 3, but not the 2. These numbers may sometimes fall behind reality as launches continue, so we append the date at which it was last updated.

Here’s an example for an actual rocket. This is a modern rocket which hasn’t varied much over its history, and has a nice simple set of stats: the Electron by Rocket Lab.

Electron: mass 12.5 t (early ones were lighter), diam 1.2 m, thrust 224 kN, imp 3.4 km/s, electric pump (kerosene), payload 0.3 t (2.4%), cost $20M/t, record 45/1/3 through June 2024.

And here’s a more complicated rocket, with more options: the Atlas V by ULA. Before the stats, we name the specific configuration of the rocket to which these stats apply, which will usually be the lightest and simplest option. In this case a note is added on the payload capacity to show the gains when optional strap-ons are used. When it comes to the record, this rocket’s history is so long that we add a popup which can be viewed by hovering the mouse over the record text (or touching the spot on mobile devices) to view the records of historical versions. Only a few old rockets have a popup like this. Some others have a parenthetical note of the record for their legacy versions.

Atlas V 401 (no extra boosters): mass 334 t, diam 3.8 m, thrust 3827 kN, imp 3.3 km/s, staged combustion (kerosene), payload 9.8 t (2.9%) [18.5 t (3.2%) with 5 boosters], cost $11M/t, record for “V” (through June 2024) 100/0/0! (1 crewed) — for legacy versions about 268/19/47 (4 crewed)
Records of Atlas orbital versions:
early ad-hoc  ’58–’95  45/10/8
Atlas-Able ’59–’60 0/0/3
Mercury-Atlas  ’59–’63 4/4/2 (4 crewed)
Atlas-Agena ’60–’78 90/0/20
Atlas-Centaur  ’63–’97 60/5/14
Atlas II ’91–’04 63/0/0
Atlas III ’00–’05 6/0/0
— so far: —
Atlas V ’02– 100/0/0 (1 crewed)

(The exclamation point is just because the Atlas V’s record of 100 launches without ever losing a single payload is a unique accomplishment unmatched by any other rocket. The second most used rocket with a perfect record is the Delta IV, also from ULA, with 44 flights if you include the Heavy version.)

For historical comparison, a space shuttle could theoretically lift 27.5 metric tons to low orbit, at a cost of around $18 million per ton in today’s money. The Saturn V could lift 135 tons — a number that has never been attempted again since, but probably will be in the next decade. The cost might have been about $14 million per ton incrementally in today’s dollars, but if you prorate the development budget over the small number of launches that were made, each one cost many times that.

For some of the more interesting rockets, this summary of stats will be followed by a link labeled “[show stages]”. If you click this, it will reveal a table of more detailed facts about each stage of the rocket, including optional ones. Depending on how many stages there are and how wide your browser window is, you may have to scroll the table horizontally to see all of them. Once shown, the link changes to “[hide stages]”. For some types which have more than one rocket in the family, there may be a link for each; in other cases there may be a single combined one, if some stages are shared between more than one model. Combined ones describe what they cover with a note in parentheses after the link.

The stats shown include many of those above: mass, diameter, thrust, specific impulse, fuel type, and engine type, but broken down for the individual stages. Several other figures are also included. To clarify everything, it may be best to look at a real example. We will use the Atlas V for that purpose:

Stage name AJ-60A (early) GEM-63 (2020+) Atlas CCB Centaur III
Role (pos) count booster (S) ×0-5 booster (S) ×0-5 core (1) upper (2)
Diameter (m)   1.58   1.61   3.81   3.05
Liftoff mass (t) 46.7 49.3 305.1  23.1
Empty mass (t)  2.2  5.2 21.1    2.2 *
Fuel mass (t) ~13    ~13.5  ~76.3  ~3.0
Oxidizer mass (t) ~30    ~30.5  ~208     ~17.7 
Fuel type HTPB HTPB+Al kerosene hydrogen
Engine Aerojet-Rocketdyne
RL-10C ×1-2
Power cycle solid solid staged expander
Chamber pres. (bar) 267    24  
Ox./fuel ratio   2.3?   2.3?   2.72   5.88
Thrust, vac max (kN) 1690     1650     4152      106.3 *
Thrust, SL initial (kN) ~1110      3827    
Spec. imp, vac (km/s)   2.74   2.74   3.31   4.41
Total imp, vac (t·km/s) 117    120    943    ~91   

Starting from the top, the first row gives each stage’s name. For instance, the second stage is named “Centaur” in this case. Sometimes no definite name is known, but most commonly, it’s just some jumble of letters and numbers. There may be parenthetical notes here to indicate stages which are used in only some versions of the rocket.

The second row gives the stage’s role and position. In the main stack, the position is given as a number in parentheses, counting from the bottom, so the main booster is designated as “(1)”. Auxiliary boosters stuck onto the side are designated as “(S)”. Before this is a role description word, which uses terms such as “core” to designate the main booster that supports the rest of the stack, “upper” for a stage above that which is needed to reach orbit, or “kick” for a small stage used only after orbit has already been reached. After the position number there may be a multiplier: in this example, the side booster is followed by “×0-5”, which means that a given launch may use no such boosters, or any number up to five. Or the word “opt” may appear here if the stage is optional, which generally applies to topmost stages.

The next few rows are similar to the values already listed in the short summary, with additional detail. The diameter may be the same for upper stages as for the core, or may vary; if side boosters are used, they are generally narrower. The liftoff mass gives the total heft when fully fueled and ready. The following row gives the empty or “dry” mass — the weight of the hardware alone with no consumables included — which is an important figure in calculating the overall delta-V capability of the stage. In some cases this may not be known. The next two rows hopefully give the mass of propellant used, with the fuel and oxidizer listed separately. Often these figures are not given out, but if we know the fuel/oxidizer ratio and the dry mass, they can be calculated. Values that are calculated or estimated are designated, as in this example, with a “~” character before the number, to indicate that it may be approximate. (The ratio itself is given several rows further down, expressed as how many tons of oxidizer are consumed per ton of fuel.)

The next several rows shift from describing the stage as a whole to describing the motor which drives it. First we name what kind of fuel it burns, which may be an abbreviation in cases like the “HTPB” in the booster column (which stands for hydroxyl-terminated polybutadiene, a type of urethane similar to the artificial rubber in car tires). Then we give the name of the engine, which often includes the name of the company that made it, for engines not made in-house by the rocket builder — “Energomash”, for instance, is the Russian company which builds the RD-180. For solid fuel there isn’t usually a distinction between the name of the motor and the name of the stage. The engine name may be followed by a multiplier, as in this case we see “×1-2” after the RL-10C, meaning that the Centaur stage can be equipped with either a single or dual engine. Next comes power cycle, given briefly without distinguishing subtypes such as expander bleed vs closed expander.

Next comes the internal pressure level which the engine produces in its combustion chamber when running at full power, in bar (one bar is 100 kilonewtons per square meter, which is pretty close to sea level atmospheric pressure). Generally speaking, a higher number here means a higher performance engine. No figure is available for the solid booster because the number is both difficult to measure and rather variable over the duration of the burn. Even a weak rocket engine like the RL-10C produces astonishing pressure, when you consider that one whole side of the chamber is practically wide open, with only a mild restriction at the nozzle. This helps give one an appreciation for the colossal amounts of power and energy which get pushed through such a comparatively small device, and why the hard part of building a rocket engine is not the chamber or nozzle or bell, but the fuel pump that pushes the propellant into it. The only way to keep the pressure so high in such an open chamber is to shove in the fuel and oxidizer at an almost impossibly rapid rate, so that the flame can barely squeeze through the width of the nozzle.

The final engine stat is the oxidizer to fuel ratio, as already mentioned above. Sometimes these numbers are calculated, estimated, or outright guessed — for instance, in the case of the solid booster, the exact mix of propellants is a trade secret involving many secondary ingredients, so our ballpark estimate of 2.3 is followed by a question mark. For liquid engines the ratio may also be inexact; sometimes different versions of an engine tweak the mix ratio, and some engines even change it in midflight, in order to balance tradeoffs such as peak power vs manageable temperature.

After this, the last four rows describe the stage’s overall performance. First comes the maximum thrust under vacuum conditions, in kilonewtons. Then comes the initial thrust at sea level, which is lower. This is the type of thrust which was given in the brief summary above the table. The figure there may combine the thrust values of the core and the side boosters, but in this case it does not because the side boosters are entirely optional. Note that the sea level thrust for the solid booster is marked as estimated or approximate. This is common with solid motors as the thrust may vary considerably over the duration of the burn. For upper stages, no figure is given for sea level thrust, as it is not applicable.

The final two rows give specific impulse and total impulse. Specific impulse is as described for the brief summary: the amount of momentum produced per ton of propellant, in vacuum. Total impulse is essentially the specific impulse multiplied by the supply of propellant — the total amount of momentum which a rocket stage is able to produce before running out. This measures the rocket’s overall motive capability. This stat is the best gauge of how much rocket you’re getting for your money, and for solid motors especially, it’s the one essential stat that cuts through the approximations and equivocations to give you the real picture.

If a value in this table has some nuance which needs a footnote to explain it, you’ll see a “*” after the contents of the field. This indicates that there’s a text note which can be seen by hovering the mouse over the cell. For example, the vacuum thrust figure for the Centaur stage has an asterisk, and if you hover the mouse you see “Doubled if two engines are used.” Likewise, the Empty Mass value has an asterisk, and a note estimating that the value might be about 2.5 tons with dual engines. Sorry, as yet there is no way to view these supplementary notes in most mobile browsers.