Putting the “rocket” in rocket science
Rocket engines are at the heart of every past, current, and near future spacecraft. The principle they work off is quite simple to understand. A conventional rocket engine combusts the propellant, converting the chemical energy stored in it into mechanical energy. This energy accelerates the combustion products out of the nozzle of the engine at extremely high velocities (measured in several kilometres per second). By Newton’s third law, the exhaust gas pushes back on the spacecraft, accelerating it in the other direction. Sounds simple, right? Well, it really is not. In this section, we’re going to explore the various types of rocket engines, the reason why staging is so important, and various other aspects related to launching to and moving in space.
Solid rockets – simple yet effective
As their name suggests, solid rocket motors use solid propellant usually in the form of grains of metal mixed with an oxidizer to generate thrust. The propellant is packed into a structurally strong cylinder with a hole down the center that serves as a combustion chamber. When the rocket is lit, a flame front is created on the surface of the propellant around this hole which slowly burns through the propellant towards the outside. During this process, the burnt fuel is being forced out the bottom through the nozzle, thus generating thrust.
If you’ve ever seen any rocket not related to launching stuff to space (like fireworks, hobbyist rocket motors and military missiles), it most likely was powered by a solid rocket motor. These are the simplest type of rocket engine as they do not have any moving parts, use a fuel that is very stable and for the most part non-toxic, and can be tuned to generate copious amounts of thrust without too much difficulty, at least when compared to other rocket types. This makes them cheap and easy to manufacture, maintain, and operate.
They do have a long list of drawbacks though, starting with the fact that they have horrendous efficiency compared to pretty much every other type of rocket engine. Moreover, they are extremely heavy, cannot be throttled (you might recognize why this can be a problem based on the last section) and cannot be shut off once ignited, which makes them potentially unsafe and difficult to work with. This means that (bar some exceptions) solid rockets have been mostly relegated to use in supporting a main liquid fuel based engine as strap-on boosters.
Solid rockets have also found a lot of utility outside of the space industry. As mentioned earlier, they are used extensively by the military, in everything from small air to air missiles to long range cruise missiles to intercontinental ballistic missiles. Here their storability and low cost shine, to the point where almost no other type of rocket is widely used. They are also used in civilian applications for pyrotechnic separation and actuation mechanisms, low altitude sounding rockets and even gas generators for various applications.
Hypergolic and monopropellant rockets – precise and reliable
Hypergolic rocket engines use chemicals pairs like N2O4 and N2H4 that react spontaneously on contact with each other as fuel. In most cases, the fuels are injected into the combustion chamber by a pressurized gas like helium or nitrogen being pumped into their tanks, where they react exothermically, releasing energy and being accelerated out of the nozzle of the engine, thus producing thrust. Some heavier engines can use turbopumps (more on those later) instead of pressurized gas to pump the fuel, but the basic principle remains the same.
Monopropellant engines are similar, though not identical, in that they use a fuel that decomposes when passed over a catalyst to drive the engine. They are simpler as compared to hypergolic engines since they only have to deal with a single fuel instead of two, but get lower efficiency and much lower maximum theoretical thrust. This means that whereas hypergolic engines see a variety of uses, monopropellant engines are mostly restricted to low power, low profile applications like reaction control systems for attitude control.
Both types of engines are relatively simple, and use propellants that are storable. This makes them ideal for probes, satellites and upper stages, where their reliability and ability to restart multiple times is essential. However, they are still less efficient than cryogenic rockets, and most of the fuels they use are extremely corrosive and toxic, which makes working with them expensive and difficult. This means that they aren’t used very often in high impulse applications like launch vehicles (with some notable exceptions).
I’ll also briefly mention cold gas thrusters. These are similar to monopropellant engines, except instead of using a fuel and a catalyst to decompose it, they just rely on releasing a pressurized inert gas like nitrogen out at high enough velocity to create thrust (which is very, very inefficient).
Cryogenic and semi-cryogenic rockets – powerful and efficient
Cryogenic and semi-cryogenic rockets use liquid propellants that have been chilled into the negative hundreds of degrees Celsius to drive the engine. These types of rockets can achieve very high impulse and thrust levels. The propellants, usually liquid oxygen along with a fuel like liquid hydrogen or RP-1 (a highly refined derivative of kerosene) are stored in separate tanks that are insulated to keep their contents liquid. Note that RP-1 is actually liquid at room temperature and so doesn’t require cryogenic temperatures like liquid oxygen and liquid hydrogen. This is why rockets using RP-1 and liquid oxygen are designated as semi-cryogenic.
In a very simplified nutshell, an engine using these propellants typically consists of two very high flow rate pumps (called turbopumps) that drive the fuel and oxidizer into the combustion chamber, where they combine and accelerate out of the nozzle at extremely high velocities, generating thrust. How the turbopumps are driven depends on the engine cycle (something that is out of the scope of this document), but in most cases, they are driven by some of the propellant being heated or ignited before reaching the main combustion chamber, and then being passed through a turbine that is connected to either one or both of the turbopumps.
Rocket engines using cryogenic fuels are very efficient, with hydrolox (liquid hydrogen and liquid oxygen) engines such as the RS-25 engines that powered the Space Shuttle achieving as much as 50% greater efficiency compared to the competition. This makes them well suited for high impulse applications like launch vehicles, and especially for use in upper stages which do most of the work in putting the payload into orbit or onto a trajectory towards its target. As with everything however, they do have drawbacks. They are extremely hard to design, build and operate successfully, which makes them some of the most expensive engines out there. Moreover, liquid hydrogen has a ridiculously low density and boiling point, which means that it is very difficult to store. It requires very large tanks, ones with heavy insulation to keep its temperature below the frosty 20 Kelvin mark needed to keep it in the liquid phase. In fact, it is near impossible to keep it chilled that low for long duration missions, which is why you will almost never see a hydrolox engine on anything that needs to last more than a couple of hours in space.
Semi-cryogenic engines like the Merlin engines used on the Falcon 9 and Falcon Heavy offer a sort of middle ground between hypergolic and cryogenic engines. With respectable efficiency and more than adequate thrust, engines using propellants like RP-1 with liquid oxygen offer good bang for your mass, while being significantly easier to operate than their pure cryogenic cousins. RP-1 is several times the density of liquid hydrogen, so it does not require very large tanks. It is also a liquid at room temperature, which makes it significantly easier to store than hydrogen, especially for longer durations. This advantage is slightly dampened by the fact that the oxidizer (liquid oxygen) still needs chilling to store as a liquid, though nowhere to the degree (pardon the pun) that hydrogen does. One problem with RP-1 based engines that has come up recently is with reusability. Unlike hydrolox engines (whose only exhaust product is water), these engines release soot and carbon contaminants, which clogs up the engines, makes them less reliable, requires more maintenance and overall adds to the difficulty of reusing them.
Measuring the performance of a rocket
Now that you know the major types of rocket engines out there, let’s move on to seeing what metrics we can use to quantitatively distinguish rocket engines, and how the variation of each of these factors affects rocket design, engine choice and mission planning.
First though, let’s look at something that isn’t strictly engine-related, but is something that pops up very often in aerospace – delta v, abbreviated as Δv. Imagine that you are in a spacecraft in deep space away from any and all celestial bodies. You place a ball outside the spacecraft stationary relative to it, and then you activate its engines, accelerating away from the ball. When you run out of fuel, the velocity at which you will be moving away from the ball will be the Δv that you have expended. Basically, you can think of it like you would think about the amount of fuel left in the tank of a car. It roughly indicates the range of the car, or how much farther it can go.
Δv is usually used to measure the impulse needed for orbital maneuvers. For example, a burn that puts a spacecraft in low earth orbit onto a lunar intercept trajectory typically requires 3 km/s of Δv. Note that Δv isn’t actually always equal to the change in velocity, however. Even though orbital velocity in LEO is around 7.8 km/s, the actual Δv needed to get to LEO is about 10 km/s. This is because while the engines are firing, gravity is pulling on the spacecraft, reducing its net acceleration. Similarly, even though a Mars bound spacecraft would need a Δv of about 4.5 km/s, its velocity at the end of the burn will actually be a little less than 7.8 + 4.5 km/s.
Thrust and TWR
Let’s move on to thrust, which is usually measured in kilonewtons or meganewtons for the most powerful engines. Depending on the role of the engine, thrust can actually have a really polarizing amount of importance in engine choice. For a first stage booster, thrust is extremely critical as it alone dictates how much mass the vehicle can put into space. Having a really good upper stage doesn’t matter if your rocket can’t get off the pad. More thrust creates room for you to increase the range of your upper stage, increase the maximum possible mass your rocket could lift, and also help reduce gravity losses due to the greater acceleration.
Another factor to consider is that a lower thrust, higher efficiency engine could actually end up using more fuel to put the same amount of mass into orbit due to gravity losses. Basically, every second the rocket is in flight, gravity is deducting about 9.8 m/s of vertical velocity from it. A lower thrust engine would take longer to attain orbital velocity than a higher thrust engine, and so would lose more Δv to gravity than the higher thrust engine. So, even if it started off with more Δv than the higher thrust, less efficient engine, when the two vehicles actually got to orbit, the higher thrust vehicle would actually have more Δv left over as it lost less of it to gravity.
Now let’s come to thrust to weight ratio or TWR. This term is kind of self descriptive – it’s the ratio of a rocket’s thrust to its weight (not mass). Essentially it gives you a relative measure of how fast the rocket is going to accelerate. TWR is not a static quantity – far from it in fact. As rockets burn propellants, their TWR goes up, so much so that some manned rockets have to throttle back towards the end of the first stage burn to ensure the crew doesn’t feel extreme g-forces. Most typical rockets have TWRs in the range of about 1.2 to 1.5 at launch, as that provides a good balance between high dynamic pressure and fuel losses due to gravity.
Efficiency and specific impulse
Specific impulse or Isp is a quantity that is used to compare the efficiency of rocket engines. It is defined as the ratio of the change in momentum imparted by an engine per unit weight of propellant consumed. The actual value of Isp in this form (measured in seconds) is actually mostly meaningless, but it provides a good relative comparison metric between different rocket engines. An engine having an Isp of 400 s versus one having an Isp of 200 s could crudely be called twice as efficient (though in practice the extra range you would get from such an engine won’t be twice that of the less efficient one).
Isp is related to the exhaust velocity of the combusted propellant, which is why different fuels have such differing efficiencies. For example, kerolox engines using RP-1 and liquid oxygen usually cap out at an Isp of about 350 s, whereas hydrolox engines using liquid hydrogen and liquid oxygen can go as high as 450 s and beyond. This is because the low molar mass and high flame temperature of hydrogen leads to the exhaust products exiting the nozzle at higher velocities and thus the engine has a higher Isp. So, hydrolox engines are almost always the most efficient in any class of engines.
Isp is also not a static quantity that changes with several factors. One of the major factors which affect Isp is external pressure. As the high pressure gas inside the combustion chamber travels into the nozzle and expands, its pressure drops rapidly. In an ideal world, you want the gas to leave the nozzle when its pressure reaches that of the surroundings for maximum Isp. Now of course in a vacuum, this is impossible as expanding a gas to a pressure of zero would require an infinitely large nozzle, which is obviously impossible to make, and so rocket engine designers have to compromise with as big a nozzle as will fit inside the interstage fairing of the rocket.
SpaceX Merlin 1 engine variants (small nozzle: first stage; large nozzle: vacuum) – even though both the first stage and upper stage engine are nearly the same in all respects, the upper stage variant has a huge nozzle for maximum exhaust expansion
In atmosphere, however, things are a lot more complicated. Since the external pressure is not zero, the same nozzle which would be the most efficient on an upper stage would barely generate any thrust at sea level. This is because such a large nozzle would mean that as the gas expanded inside the nozzle, the external pressure would eventually force it to detach from the walls of the nozzle, stopping its expansion and thus greatly reducing its thrust. So, a different nozzle, one which is shorter and narrower, needs to be used on first stage engines. This gives them high (though still lower than vacuum) Isp even at sea level.
Isp is also dictated by the chamber pressure, which – as you can probably tell by the name – is the pressure of the gas inside the combustion chamber. And as chamber pressure is essentially a function of the amount of propellant flowing through the combustion chamber, lowering engine throttle settings causes a lowering in the chamber pressure, and by extension Isp. So rather counterintuitively, it is actually more efficient to run a rocket engine at full power than it is to run it at low power. You can learn more about nozzles and expansion ratios by watching this excellent video by Scott Manley going over this topic.
Other points of comparison
Besides performance, there are also other features that differentiate different rocket engines. Perhaps the most prominent of these is engine gimballing. Engine gimballing is a technique where the nozzle of the engine can change its pointing direction, thus changing the direction in which it is generating thrust. This applies a torque on the rocket, giving pitch and yaw control over it. This mechanism is part of the reason why fins aren’t required on newer rockets – engine gimballing produces more than enough torque for flying an optimal trajectory.
Another important feature is restartability. Since most high performance rocket engine designs use a self-sustaining pumping system (i.e. they use some of the fuel flow to pump more fuel), it is actually quite complicated to start them. So, many rocket engines are only designed for a single ignition, which creates problems for spacecraft that need to perform multiple precise maneuvers over long periods of time. Hence, restartability is actually a pretty key feature of a good vacuum engine, which is why newer designs (e.g. CECE) emphasize on it.
I’ve briefly mentioned staging once before, but never actually discussed it. By now, if you have a good grasp of what makes a particular rocket engine, you should already have a pretty good idea of the necessity of staging, but for the sake of brevity let’s just go over them once.
- Staging gives rockets more Δv by letting them shed the useless mass of empty fuel tanks that would otherwise have to be lifted to orbit.
- Staging allows matching the engines to their ideal external operating conditions, making the rocket more efficient and the engines simpler.
- Staging affords a certain modularity by allowing the use of cheaper, lower power components on flights that have smaller payloads.
- More recently, staging helps reuse of boosters by requiring them to accelerate as little as possible before coming in to land at a launch site.
Usually, the various stages of the rocket are built separately, and then assembled near the launch site by either vertical stacking (e.g. in the case of the Saturn V) or by laying them on their sides, attaching them together and then lifting them to vertical on the launch pad (e.g. for the Falcon rockets) or a combination of both. The lower stages have what is called an interstage which houses the engines of the stage above it until separation. When the lower stage burns out, separation mechanisms in the interstage, usually in the form of explosive bolts or mechanical latches are fired, detaching the lower stage. To ensure that the two stages don’t recontact, most rockets also have pneumatic pushers or small solid rocket motors to push either the lower or upper stage away far enough that the latter can safely ignite its engines in time.
A view from inside the interstage of the second stage of the Saturn V during the Apollo 8 second stage separation – the third stage can be seen firing its separation motors to clear the second stage
Despite the fact that staging adds extra complexity and a lot of additional failure points due to the need of reliably separating each stage, its advantages are so ubiquitous that no orbital rocket in history has been launched that did not use staging. It is a technique that has been monumental in allowing us to build bigger, better and cheaper rockets, and will no doubt continue to be of great use for the foreseeable future, at least until the science fiction of spaceplanes becomes a reality.
Assignments on rockets and propulsion
- Most liquid fuelled rocket engines cannot be ignited in zero gravity without special measures being taken. Give the reason for this as well as ways it is overcome.
- Is it possible to make rocket engines with an Isp greater than that of hydrolox rockets? If so, why aren’t they being used frequently in high profile, high impulse applications?
- Write about how rocket engine nozzles can be designed to be altitude-compensating or be efficient across a wide range of external atmospheric pressures.
- You as the director of the IASA (International Aeronautics and Space Administration) want to send a manned mission to Mars. Your engineers have come up with a spacecraft design that will contain all the life support, entertainment and shielding that your astronauts will require, but will weigh 240 tonnes in orbit. However, the heaviest lift vehicle available to you is only capable of launching 60 tonnes into LEO. Assuming development of a new launch vehicle is not possible, how would you solve this problem?
- https://youtu.be/LbH1ZDImaI8 (long, but overviews almost everything related to mainline cryo/semi-cryo engines)