This article is based on the assumption that finding opportunities to create tangible, monetary value in space is the best way to propel humanity to the stars, and something that must be pursued. With that understanding in place- how is profit created by going to space right now?
A simple fact to start with- profit is created when someone is provided value, and they are willing to pay for it. The only reason to perform an activity is if more value can be created than it costs to perform that activity. This could be something like ‘a satellite was launched to relay wireless communications, and the company can sell the relay service for more than the cost of manufacturing, launching, and operating the satellite’. It could also be ‘the United States launched a manned mission to the moon because all the people in the US felt a little bit prouder to have their nation put a man on the moon, and were willing to pay higher taxes to cover the expenses of the endeavor’.
This isn’t a complex topic- activities are profitable and worth doing when the value they create is greater than the cost to perform. Another modifier to investigate is how much of that profit creation can be captured, but for the sake of limiting complexity we won’t dive into that topic.
With that said, we need to know the cost of activities as a threshold to overcome, and that threshold must always be considered when discussing profitable activities. The cost to get to space is exceptionally high, as physics itself makes it challenging. The Tsiolkovsky rocket equation and the high gravity of Earth combine together in unfortunate ways. With typical (chemical) engines, the design margins required to put payload into space is very low, leading to complex vehicles. The historical ratio of a vehicle to put anything into orbit has been 85 - 95% of the rocket launch mass if fuel and only about 1% is payload. Aggressive engineering might shift that to 2% payload. In any event, the vehicle must be highly capable and large to put anything useful into orbit.
Unfortunately, given the tight engineering margins a positive feedback loop is created to make the barrier to entry for space even higher. The effective loop is that a rocket to take things to space is very expensive. Because the rocket is expensive, the people building payloads are conservative with their designs because they need it to work right the first time to not waste their large investment. Because the payload is very expensive, they want a very high reliability rocket to make sure their expensive payload reaches orbit. This drives them to want a more conservative, and more expensive rocket. Then the loop just keeps repeating.
Besides the positive feedback loop that comes with the rocket cost drive, space is just essentially a horrifying environment to operate in. There are quite a few reasons why, and almost all of them drive more expensive spacecraft design.
First of all, spacecraft must bring their own electrical power production. Usually this is in the form of solar arrays, but nuclear is also feasible- typically this is in the form of radioisotope thermoelectric generators (RTGs), which is a fancy way of saying radioactive fuel that can be used to generate power. RTGs are not widespread, and closely controlled due to nuclear proliferation efforts- so we’ll mainly focus on solar arrays.
The spacecraft must also control its own temperature- which is much easier said than done. Space is not “cold” as many people think, but effectively just a vacuum, like the channel between the walls in a thermos. Heat is wicked away from a spacecraft as it’s radiated away into the universe, which is at a background temperature of 3 deg kelvin. Essentially the same way you can feel the heat from a fire as you get close (radiative heat transfer) is the same thing that a “warm” spacecraft is doing into the vast emptiness of the universe. But on the flip side, the sun puts out a lot of heat, and so just by sitting near the sun a spacecraft absorbs enough heat from the sun to keep it around the right temperature. The end result is a basketball painted blue placed in the same orbit as the Earth ends up being about the same temperature as our average planetary temperature. But for a spacecraft, with power generation, large flat solar arrays, and weird angles- temperature can get very wonky very easily. Typically temperature can be kept in range by the right coatings, but even then different parts of a spacecraft can vary widely in temperature, and thermal management is a critical task, either actively or passively.
Van Allen radiation Belts: Wikipedia
In addition to temperature, the radiation environment of space is particularly nasty. The first thing to talk about is the things that stop radiation. First of all is Earth (or any planet’s) atmosphere, which slows radiation down significantly. More importantly, the significant magnetic field of the Earth blocks radiation as well. The flip side is the magnetic field also traps energetic particles in a band called the Van Allen Belts. A fun note is that the Van Allen Belts cause the Aurora Borealis when they dip down to the atmosphere near the magnetic poles where the magnetic field is weakest. The Van Allen belts go through the Medium Earth Orbit region, so any spacecraft in that region have to be especially robust to high energy particles. This is one of the big drivers of the GPS constellation cost, as GPS operates in Medium Earth Orbit for geometric reasons, but most other spacecraft avoid it when possible.
Low Earth Orbit has the magnetic fields to protect it from radiation, but not atmosphere, so spacecraft in that region must still deal with some radiation, but less so than in geostationary or deep space. The big fear is cosmic radiation, which is an odd type of radiation. Cosmic radiation is effectively atomic nuclei accelerated to near light speed. These heavy particles have significantly more energy than photons, and astronauts in the space station have noted “flashes of light” in their eyes caused by random cosmic radiation. In addition, the solar radiation environment is more significant outside the atmosphere and there’s just generally more stuff going on up there. The elevated radiation levels cause a significant amount of cancer in astronauts (every trip up is a bit of a danger), but more importantly it messes with electronics. The typical impact is “bit flips”- memory in electronics systems randomly flip when they are hit by radiation. Programs randomly stop working, instructions fail, and in general electronics don’t work well. There are many techniques to mitigate this impact- such as triple modular redundancy (TMR), where 3 sets of circuits are run in parallel and the outcome is ‘voted’ of the three circuits. There are a few different methods, but the end result is spacecraft electronics are challenging to make and operate, and thus more expensive.
Another constant risk is micrometeoroids and debris. Things go fast in space, and the minimum orbital velocity is 7.8 km/s. Small rocks are constantly whizzing around at even faster speeds (meteors). We don’t deal with any down on earth of this because they burn up in the atmosphere, but spacecraft are constantly being shot at- but most of the shots miss so it’s rarely an issue. Nevertheless, traditionally spacecraft had to be designed around having holes punched in them every once in a while.
Another reason spacecraft are expensive is the only way to communicate with them is with very long distance communications. Due to the tyranny of the communications link budget equation, there is always a tradeoff between communications rate, directionality (and tight pointing requirements), antenna size, distance, and power. Effectively you’re always balancing one for the other, and the power requirements scale exponentially with the distance between the two points communicating between each other. Needless to say, things in space are very far away, so communications can be quite challenging. The real challenge is that a failure to point correctly to achieve communications can lead to mission failure. Communications are critical not only to producing value, but the pure existence of the spacecraft.
In addition, the method of putting things into space is to put them into a small, hot, vigorously shaking box on top of a tube of explosives (also known as a rocket). Satellites need to be designed to work in space, but also to collapse into a small space and survive a high vibration environment. These two vastly different design regimes both need to be accounted for in the spacecraft design. This may change with the advent of on-orbit assembly and manufacturing, but typical applications have everything built on Earth.
Finally, one of the worst parts of spacecraft is the historical inability to fix anything after deployment. Once a spacecraft is up and in place, it is very difficult and very challenging to even physically interact with it. Preventative or responsive maintenance and refueling has, historically, been almost impossible. NASA has done a few repair missions of the Hubble telescope, and there is an emerging commercial capability for in-orbit servicing (led by Northrop Grumman), but it’s not expected to be a regular, low cost service. Spacecraft need to work perfectly over their entire lifetime- every time. That drives significant engineering and quality assurance costs, and conservatism in capabilities.
None of these factors are- in isolation- huge hurdles. And other industries and markets face similar challenges. But few other engineering and business opportunities involve such an overlap of challenges, and they all build on each other and escalate to be a unique and exceptional challenge.
Launch Cost Reduction - futuretimeline.net
The takeaway here is space is very expensive, and even with new trends in smallsats, improved electronics, and lower launch costs this fact will not completely change. Even if SpaceX’s Starship achieves its objective, you can’t expect launch costs to be less than $100/kg to $500/kg, which will still be more than ten times more expensive than to get anywhere on earth. Historically, it has been more than 1000x times more expensive than getting anywhere on Earth. This is definitely on a downward trend, but it will continue to be massively expensive to travel to and operate in space.
Now that we’ve talked about the general theme that space is expensive, how do we quantify that? I’ve tried to build up some composite cost estimates using publicly available data. I use $5000 / kg for current launch costs, based on SpaceX’s $1M / 200 kg rideshare pricing. I guess $500 / kg for starship launch costs in 5 years, based off the assumption of $10m / launch operational costs to SpaceX (based on $2m propellant costs), a 100k kg payload, and an 80% gross margin by SpaceX. Given the vast capital recuperation activities required by Spacex, I feel like this is a reasonable but conservative estimate.
For manufacture of on-orbit vehicles, each application typically needs its own unique analysis. That said, a Starlink satellite costs about $1m. They are, admittedly, mass manufactured, but it’s a starting point to work with. A starlink satellite is about 260 kg, and I typically use about $4,000 / kg for mass produced, complex spacecraft, and $40,000 / kg for development or low quantity satellites in the near term with current space design margins (< 15%), based on a 1u (1.3 kg) manufacturing cost. In the future, when launch costs have dropped significantly, I expect margins to open up to 50%, such as with US aircraft design requirements. I expect spacecraft to weigh approximately 30% more, but costs to reduce to $10,000 / kg for low quantity builds. I expect no change for mass produced quantities, as terrestrial aircraft costs come out to be about $2000 / kg (by price / mass), so few additional savings can be gained. This results in an effective $13000 / kg cost at current, low margin spacecraft mass designs.These are very rough estimates, but approximate operational (not including development) costs.
With these two large cost drivers combined, I utilized the below, very rough, estimation for cost of putting things in space by mass.
My estimates for the cost of space equipment placed into LEO by mass
With this huge cost threshold to overcome, the question is- what activities can be performed in space that create so much value they can overcome this hurdle?