# Why is SpaceX not also working on a smaller version of Starship?

SpaceX is planning to retire Falcon 9, which would leave it with only Starship. While a Starship launch is expected to be cheaper than a Falcon 9 one, a downscaled Starship launch would be cheaper still.

Moreover, a downscaled Starship would be easier to develop and iterate upon, paving the way for the full scale Starship. It could be as similar as possible to the full scale Starship (but of course allometrically scaled). So it might even lower the total development cost and risk.

It could conceivably use seven times less Falcon engines, ending up with five on the first stage, and one on the second. That smaller Starship could be used for almost all the satellite launches, as it would still be a respectable size.

• They just trimmed the top off of it? twitter.com/NASASpaceflight/status/1197265917589303296 – ta.speot.is Nov 21 '19 at 2:49
• "a downscaled Starship launch would be cheaper still" why would you think so? Things don't scale effortlessly, up or down. Even the mere fact of having two launchers instead of one introduces extra expenses. And that's completely ignoring R&D, which is a significant part of the costs of dealing with rockets in general. – Luaan Nov 21 '19 at 12:11
• The original design a few years ago was based around a 12m diameter. Musk has also said an 18m design might come at some point. Maybe the current 9m design is the smaller version... – James Thorpe Nov 22 '19 at 9:18
• The Falcon 9 uses two types of Merlin engines: sea-level optimized, and vacuum-optimized variants. The Merlin is a LOX / RP1 gas-generator cycle engine with thrust around 900 kn. The Starship and Superheavy booster are designed to use two variants of the new new Raptor engine. The Raptor is a LOX / Liquid Methane full-flow staged combustion engine and has a thrust of around 2400 kn. – Dragongeek Nov 24 '19 at 13:48

There are several compelling engineering and design reasons why a bigger spaceship makes sense and several reasons why making a mini-starship does not make sense for SpaceX specifically (and their vision).

First and foremost, Elon Musk has made it clear that his goal for the company and the future isn't to provide cheap satellite launch capabilities, it's to put people on Mars. Building an inferior version of the Starship isn't on the "critical path" to putting people on Mars for Elon. SpaceX is very much Elon Musk's company and it follows his vision.

That said, here are a couple other reasons why SpaceX specifically might not want to make a Mini-Starship (although a Mini-Starship might make sense for some manufacturers):

• Bigger rockets can be more efficient, fuel-consumption wise, as dry mass can be a smaller portion of the overall rocket mass. For example, if your flight computer and avionics system masses 100 kg, it would have a large effect on a rocket which can only lift 1000kg to LEO. If your rocket is able to put up 100,000kg at a time, the 100kg flight computer suddenly becomes much less of the overall dry mass and thus your rocket is more efficient.

• Building big vs building small. One of the things that is "revolutionary" in Starship's current design/construction progress is that it's being welded together in a field, outdoors. SpaceX believes that, when your rocket is big, you can get away with looser tolerances, which in turn equates to money saved. For example, the Sea Dragon proposal was designed to be assembled in a shipyard and had an enormous lift capability for a very low cost. Starship is similar. If a Mini-Starship were constructed, it would probably require tighter tolerances and smaller, more exact parts. Really, this boils down to my first point again. If you're building a small rocket, every weld, bolt, and wire has a more significant impact on a lighter, smaller rocket's efficiency.

• SpaceX believes that a design similar to the two Starships currently being prototyped will work. If they instead scrapped what they have developed so far and built a Mini-Starship, they'd have to essentially start from zero again. The design for a rocket is so complex, that you can't just resize it and have a functional design afterwards. SpaceX learned this lesson the hard way when developing Falcon Heavy. Initially, Elon had thought that it would just be "strapping three boosters together" but in the end, SpaceX had to develop the center core almost entirely from scratch.

• SpaceX has a limited amount of employees and money. If they decided to work on a Starship and Mini-Starship concurrently, the pace on both projects would be cut in half if not more. Elon has stated that development speed is critically important to him and diluting his engineers with extra projects means that Starship takes longer to build.

• The smallsat launch market is heating up. Several companies, notably Rocket Labs, are focused on sending small satellites to orbit. SpaceX is betting that there will be no shortage of demand for mass sent to space and then it won't matter how big the rocket is that goes up, as they'll be able to fill every rocket with payloads. Even then, the 2 million launch cost is so low that even if Starship flies mostly empty, it would still be making a profit. Addendum: In the comments several concerns have been raised: • Building a Mini-Starship would be cheaper than a full sized Starship • On this, I don't disagree. If you were starting from scratch, building a 1/8 scale starship would probably be cheaper than a full sized one. • SpaceX has already sunk a lot of money and development time into ITS/BFR/Starship over the years. Simply shelving these efforts to work on an inferior rocket would be seen as a waste of money. • While it would be cheaper, I don't think it would be significantly cheaper. The big costs are the Raptor engines as they're highly complex and are a revolutionary design. Still, all Starship prototypes we've seen so far (Starhopper and the late MK1) haven't been equipped with the full compliment of engines. Starhopper only had one and the MK1 was only briefly test-fit with three engines. • Starship and the Superheavy booster are two different craft. Starship is designed to be equipped with three Sea-level raptor engines and three vacuum-optimized engines for redundancy. On a Mini-Starship, only a single Raptor engine would fit which wouldn't be efficient and wouldn't provide the redundancy that multiple engines allow. An entirely new Merlin-sized Methalox engine would need to be developed which just isn't going to happen when SpaceX has the working Raptor in production. • A Mini-Starship could be iterated more rapidly • Maybe, but neither the Mini-Starship or Starship are small desk-top or garage-shop construction projects. While working with smaller components would be easier, cranes and heavy equipment would still be required for both. • The labor that's connected to the actual size of the rocket doesn't take up much of the overall assembly time. Yes, welding together a bigger rocket takes longer than a small one, but installing avionics, sensors, engines, wires, and writing software wouldn't suddenly become quicker if the rocket were physically smaller. • Mini-Starship could provide cheap launches while Starship is developed • Starship (or Mini-Starship) are not SSTO vehicles. They both require the additional development of the Superheavy booster. Also, there is no reason why SpaceX needs to stop operating the Falcon 9 series. They have plenty of rockets with plenty of reuse cycles left and despite the non-reusable second stage, Falcon 9 can easily undercut the price of any other launch provider. Until the Vulcan or Ariane 6 rocket become operational, SpaceX won't even have to lower their prices (and profit margins) for Falcon 9 to be the cheapest provider. • Beliefs that upscaling the Mini-Starship to a full sized one would be easy or lots could be shared during development • I'm not quite sure where this belief is coming from. In the engineering world, you can't just scale something up and expect it to work. For example, take the Cessna 172, which is the most built aircraft of all time. Why don't larger planes just look like the Cessna but bigger? Because, simply scaling up a design, no matter how fantastic just doesn't always make sense. • The parts of development that could be shared don't justify building an entirely new rocket. The engines are already built and building fuel tanks and the actual structure isn't simple per se but it's rather straightforward compared to developing a FFSC engine. Aerodynamics and in-flight control would also be explored in a Mini-Starship and maybe heat-shield tech, and would be transferable, but again, SpaceX clearly doesn't believe this would be cost-effective • SpaceX could reap profits in the smallsat market with Mini-Starship • Just because the smallsat market is heating up doesn't mean that bigger satellites are going anywhere. Especially with Starlink, SpaceX isn't hurting for cash or launch contracts. Dealing with smallsat launches where each satellite only pays like 50k isn't where SpaceX sees profits. • "When your rocket is big, you can get away with looser tolerances, " this is yet to be proven. The sloppy-looking welds on the vehicle in Boca Chica scare me. – Organic Marble Nov 20 '19 at 18:38 • @OrganicMarble Fixed – Dragongeek Nov 20 '19 at 18:40 • @OrganicMarble Especially on the windward side, during the hypersonic flight regime. How will those bumps and welds affect the hypersonic flow? Terrifying to think about. And how do you model that in a wind tunnel? – geoffc Nov 20 '19 at 18:53 • @OrganicMarble Prescient comment. It blew up a few hours after you made that comment. – ceejayoz Nov 21 '19 at 2:18 • "Blew up" -- they were conducting an over pressure test and reached the point it couldn't take the over pressure. It looks dramatic, but they knew going in it was a possibility. – Rob Crawford Nov 21 '19 at 22:01 There was some talk of modifying Falcon 9's second stage to give it full reusability in November of 2018. The idea was that a reusable second stage would be used to test out Starship technologies. This effort was scrapped 10 days later in favor of accelerating development on the current stainless steel design of starship. The superheavy/starship architecture is designed to facilitate exploration and exploitation of space beyond simple Low Earth Orbit, which necessitates a larger booster and spacecraft. Since development costs tend to outpace materials, manufacturing, and fuel costs; it makes more sense to build a single large vehicle that can do many jobs than two vehicles for separate missions. • This answer assumes that developing a smaller Starship in parallel would significantly increase the total development costs, which is unclear to me. As I wrote in my question, developing a smaller version could even decrease some of the total costs. Falcon 9 is a very different rocket, so it being scrapped doesn't seem like strong evidence. – Stephane Bersier Nov 23 '19 at 23:55 Similar to how a small plane can't really carry any people around the world, but a larger plane can, it turns out there are challenges to making really small rockets. These are magnified when you take in to account full reusability. It turns out that Starship is about the smallest spacecraft that makes sense for a fully reusable spacecraft. A few things to consider. The pressure of the tanks has to remain about the same. The strength of a tank is pretty much dependent on the thickness of the tank, thus you get a lot more stored fuel with a larger rocket for the weight. Rocket engines have a maximum thrust to weight ratio when they are larger. Starship is intended to support missions to orbit even when some rockets fail. The size pretty much again comes to Starship sized. Heat shielding is a bit trickier, but I believe there is a similar minimum mass that is required to truly be effective. Essentially it has to absorb the energy of the spacecraft and dissipate it out. With a larger spacecraft, the density per area ratio is lower, allowing for more effective slowing down, and also less heat shielding required. Lastly, there was a similar thought process with Falcon 1 vs Falcon 9. It turns out that many of the costs of a launch are fixed, the flight analysis, coupled loads analysis, etc all have to be done regardless if you have a single small satellite or a huge one. Concentrating on a larger load allows those costs to be minimized. Bottom line is, it is far more efficient for a spacecraft of Starship's goals to focus purely on a spacecraft roughly the size of Starship. Robert Zubrin even mentioned this at a recent Mars Society meeting. • In particular, the surface area of fuel tanks/heat shields goes up as the square of the ship scale, but the volume goes up as the cube, which is what provides the favorable economies of larger ships. Of course, this doesn't continue indefinitely, as you become limited by materials strength. Thus, there is an optimal size. – Lawnmower Man Nov 21 '19 at 3:40 • Cube-square law does not apply to pressurized tanks. The structural mass of a tank is proportional to pressure and volume. See Pressure Vessel#Scaling – Rainer P. Nov 21 '19 at 8:49 • @RainerP. Right, but that doesn't apply very much to rockets. Unless you're using pressure-fed engines, the internal pressure in the tanks is relatively low, and the resulting loads are much less important than the axial compression due to thrust and dynamic pressure or aerodynamic bending loads. Also, with liquid propellants in the tanks, only hydrostatic pressure matters for the tanks, and that only scales with length. – TooTea Nov 21 '19 at 10:06 • Most of the arguments of this answer are flawed. – Stephane Bersier Nov 23 '19 at 23:37 • For paragraph one: in the atmosphere, smaller planes have a smaller range because of the square–cube law. This doesn't apply in space. While rockets launched from Earth do have to contend with the atmosphere initially, the energy lost to drag is already very minor for a Falcon 9. – Stephane Bersier Nov 23 '19 at 23:38 The simple answer, as already stated by others, is that small rockets do not align with Musk's goals of putting people on Mars. Additionally, Spacex already have a small (OK, medium sized) launcher in the form of Falcon 9. They claim that Starship + Falcon Superheavy will be able to undercut it. One possiblity I wouldn't disregard (if Spacex chose do go down that route in future) is to use the upper stage of Starship without a booster to launch payloads to suborbital speeds. A payload would then need its own kicker stage to get into orbit. Spacex have already revised their Earth-to-Earth passenger scheme to one without a booster so using the upper stage of Starship without a booster is not an entirely new idea. The kicker stage could be a simple solid rocket, or a reusable stage - whatever is decided in the future. I would note that when Falcon 9 is retired, Spacex will have several hundred surplus Merlin engines which could be given one last use in an expendable kicker stage. Spacex has gone for the upper size end of the market, with very few competitors, all of them expendable: SLS (projected to be ready soon) and Long March 9 and Yenisei (projected to be ready late 2020's.) They have avoided the crowded lower size end of the market. If one of the current players in that market becomes big enough to compete with Spacex, they may regret it in 20 years. But there is no sign of that happening yet. I wouldn't discount spacex developing a smaller launch system later - but it would be done in an opportunistic way on the back of the Starship program. Spacex does have one Spinoff project not directly aligned to the goal of getting to Mars, and in typical Musk style, it is unique: the Starlink Satellite constellation project. Spinoffs like this are obviously a necessary way to raise funds. • I feel like your last paragraph contradicts your first one... – Stephane Bersier Nov 23 '19 at 23:28 • @StephaneBersier How so? Musk's main goal with Spacex is to build a large rocket to take people to Mars. For that he needs both money and a way to gain experience with his rocket. One way to make money is with the starlink constellation, which is something nobody else is doing (or at least, not on anything like the same scale.) Another way would be to build a small rocket, which is something a lot of others are competing to do, therefore SpaceX have avoided this. – Level River St Nov 25 '19 at 2:22 • SpaceX is actually already planning to compete in the smallsat market with their Smallsat Program. Given the expected specs, a mini-Starship would have no trouble competing in the smallsat market while still maintaining large profit margins, serving the same role as the Falcon 9 in their current Smallsat Program plans. – Stephane Bersier Nov 25 '19 at 18:45 • @StephaneBersier Spacex's Smallsat program is a rideshare program, not a small rocket program. It's a great option if you don't care too much what orbit you're going into or when you launch, since all satellites on the launch go into virtually the same orbit on the same launch date. Small rocket operators like Rocketlab will give you a bespoke orbit and launch when you want. That's the advantage they have over rideshare, and in theory they can charge more for clients who need it. The trouble is there's a lot of other small rocket operators either up and running or coming soon. – Level River St Nov 25 '19 at 23:54 • Exactly. I didn't say it's a small rocket program. My arguments still hold: SpaceX could make money for its ultimate goal with a mini-Starship, just as it's doing with Starlink. If SpaceX can make money with its current Smallsat program using a Falcon 9, then a fortiori it could do so with a mini-Starship. – Stephane Bersier Nov 27 '19 at 0:14 Right now Starship and it Superheavy booster are both open design. Current plan is for 35 engines on 1st stage, 6 on 2st stage, 150ton to LEO ascend payload, dry mass 85t, return to Earth payload 50t. All of this could change during rocket development, because of technological challenges, but also money which will have SpaceX available during development process. They already change Starship design many times ( number of fins, TPS types, design of legs ) and they could keep changing it constantly after first test flights. At the end they can finish with much smaller, medium version of Starship with less engines, less payload to LEO, smaller gross mass. For example 20,30t to LEO Starship/Superheavy with just total of 15 Raptors, but maybe 20,30t to LEO payload, but still with 30-40 Raptors total. It depends, on what would be optimal for lowest possible cost per kilogram to LEO, GTO, which will be main goal. Figure 2 million per flight is just < aspiring goal > and shouldn't be taken as fact. For example ESA Arianespace want with expendable Ariane 6 ( payload 20t to LEO, 10t to GTO ) achieve same cost per kilogram as F9R, which is about 5000 per kilogram to LEO ( more than twice to GTO ) and with Ariane next program, which could use potentially reusable LOX/Methane engine Prometheus, improve Ariane 6 cost per kilogram further by factor of two. But that of course is not yet given.

Same as is not given that Starship/Superheavy cost per kilogram to LEO or GTO will be better than F9R, since second stage reuse is much harder than 1st stage ( both technically and economically ), and no rocket with massive size of Starship/Superheavy ( it will have see level trust 3-5 times of Saturn 5 ) was ever build, let alone try to be reused.

• The Starship design is indeed not final. However, it is to be at least a super heavy-lift launch vehicle, which excludes the possibility of it being as small as the mini-Starship described in the question. – Stephane Bersier Nov 30 '19 at 0:44