Op-Ed: The Load Bearing Rocket, A Falcon 9 Analysis

While NASA and its international partners look to the Moon and beyond under the Artemis Program, the mid to late 2020s will see a dramatic shift to the landscape of Low Earth Orbit (LEO). The International Space Station, which has served as the hub of human space exploration for 25 years, is going to be retired. The precise timing of the ISS’ retirement has become somewhat dubious recently; different involved parties have differing opinions on how long the station can be made to last, but one thing remains clear: the future of LEO stations will be commercial in nature. Under the Commercial LEO Destinations program at least two privately operated space stations will be in place by the time the ISS is deorbited, assuming all goes to plan. Privately developed, launched, and assembled space stations will be maintained by privately operated crew and cargo spacecraft, a dramatic shift from the landscape the ISS was born from. The emergence of private space stations remains a part of years yet to come, but the transition from government to commercial spaceflight has been ongoing for many years now.

Throughout that transition, one company has stood out from the rest, established itself with a massive lead, and absorbed a significant fraction of the entire launch market. Space Exploration Technologies Corp., better known as SpaceX, needs no introduction. The imagery of Falcon 9 blasting off from Launch Pad 39A, only for the booster to soar back to a landing zone and touchdown safely has become perhaps the most iconic symbol of spaceflight’s current age. SpaceX has been recovering and re-flying Falcon 9 boosters since 2016, and seven years later the achievement remained unmatched. The only other orbital-class stages recovered have been from Rocket Lab’s Electron rocket, which land in the ocean via parachutes, and while this year the company re-flew a recovered engine, a full first stage has yet to fly a second time. SpaceX and Falcon 9 are also no stranger to the ISS. From Cargo Dragon’s first mission in 2012, to Crew Dragon’s upcoming 8th ISS crew rotation, SpaceX has been well ahead of its competition in regards to ISS services. With Boeing’s Starliner Crew Flight Test increasingly delayed, and Northrop Grumman’s Cygnus cargo spacecraft now set to fly on Falcon 9 in the wake of the early retirement of the Antares 230+ rocket, Falcon 9 is now the sole pillar of American access to the International Space Station. One may argue that, if the vehicle carrying the entire responsibility of maintaining American presence on the ISS flies routinely, reliably, and affordably, there is little cause for alarm. However, the present state of affairs does present risk, and is a scenario NASA specifically organised commercial programs to avoid.

Redundancy Maintains Capabilities

The value of redundancy often goes understated in discussions of spaceflight capabilities. There is an impression among many that if one good system exists, there is no value in an additional system if it cannot match or exceed what is offered by the first. There are, however, very real modern examples of the consequences of relying solely on one launch system. Ariane 5 was the flagship rocket of European space access for 27 years, and sole-flying member of its family for the last two decades. A true workhorse, Ariane 5 delivered a variety of cargo into space, including commercial satellites, ISS supply missions, international payloads and interplanetary probes, all with a near-spotless launch record. Yet, European access to space now finds itself in a critical state. Only one part of the blame falls on Ariane 5: the simple fact that it retired in July of this year. Ariane 5 was always going to retire, and the previous three generations of Ariane rockets had their retirement met by the inaugural flights of their successors. Until 2023, an operational member of the Ariane Family had been actively flying since 1979. The risk was always there that a successor might not be ready to fill the gap left by its precursor; spaceflight is an industry prone to delays and setbacks after all. After 44 years of flights that inevitable possibility finally occurred, and Ariane 6 has not yet flown.

The reasons why Ariane 6 has not yet flown could be grounds for another article entirely, but regardless of those reasons, Europe now finds itself in a launch gap. No other domestic system falls within Ariane 5’s payload class; furthermore, sanctions against Russia have removed Europe’s access to Russian launch vehicles, and recent mishaps with the smaller Vega-C rocket place its own future in jeopardy. Europe has entered 2024 with nearly no domestic access to space at all, let alone reliable or flexible access. While writing this piece, it was announced ESA has begun considering moving a European weather satellite, set to launch on Vega-C, to Falcon 9. Japan finds itself in a similar situation with the transition from the H-II rocket to the newer H3, which failed on its first launch. While H-II’s final flight is only months away and H3’s second launch attempt is planned to precede it in February, Japan is still cutting it close. Uncertainty in H3’s availability has already led valuable domestic payloads set to ride on the flagship rocket, such as JAXA’s Martian Moons Explorer, to stand down from launch. 

It is the existence of redundant systems in the modern American launch market which prevents a launch gap like the one facing Europe from occurring. Multiple systems often exist within the same size class, and multi-launch contracts are often distributed between two systems of similar capacity to ensure operations do not fully rely on one system, which would be one point of failure. In the wake of the Shuttle Program, selecting two different systems to maintain resupply cadence to the International Space Station was seen as a vital decision. By selecting both Northrop Grumman’s Cygnus spacecraft and SpaceX’s Dragon spacecraft, both with their own launch vehicles, a failure or unavailability in one system would allow the slack to be picked up by its counterpart. This greatly reduced risk and increased the likelihood of a constant resupply stream existing for the ISS in the post-Shuttle landscape. This decision led to the Commercial Resupply Services (CRS) Program ending up in the best case scenario: two systems simultaneously ensuring a steady stream of supply missions to the ISS. 

NASA policy even allows for payloads to be re-manifested on alternative systems should the original be deemed unsuited for launch. A somewhat recent example can be found in NASA’s TROPICS constellation. TROPICS is a weather-observing constellation originally planned to consist of six CubeSats, and the three-launch contract for TROPICS was initially awarded to Astra’s Rocket 3.3. In June of last year, Rocket 3.3 lifted off carrying the first two TROPICS satellites, but failed to reach orbit after consuming fuel at a higher-than-expected rate. The failure and resulting loss of the first two TROPICS satellites resulted in Rocket 3.3 losing the second and third of its contracted launches. TROPICS Missions 2 and 3 were instead handed to RocketLab’s Electron rocket, and these launches were completed successfully in May of 2023. It is through this process that Falcon 9 is now set to fly Cygnus, but in this case it presents a potential problem.

Increasing Reliance and Demand

When Russia launched its invasion of Ukraine, the Antares 230+ rocket which previously launched Cygnus became unavailable due to the destruction of the Ukrainian plant which produced its tankage and sanctions against the first stage’s Russian-built engines. Because the first stage of Antares 230+ could no longer be manufactured, maintaining the expected flight rate of the Cygnus cargo spacecraft required moving it to a new launch vehicle. Luckily, Cygnus is built to be agnostic, meaning it can launch on any system that has the payload capacity to send it to space. However, at the time almost all the launch vehicles in the size class to do so were either not yet operational, such as Vulcan-Centaur, or had their launch manifest spoken for, such as Altas V. In order to maintain the rate of Cygnus flights to the space station, Cygnus’ next three launches fell on the shoulders of the only system available–Falcon 9 Block 5. Meanwhile, Northrop Grumman is working with Firefly Aerospace on a new first stage for the Antares rocket. This so-called “Antares 330” configuration is intended to launch in the latter half of 2024.

While Falcon 9 is once again a provably reliable system, for the time being it now represents the sole pillar of the Commercial Resupply Services Program, until the Antares 330 rocket and/or Dreamchaser space plane get their wings. Thankfully, in the time since the NG-19 mission and upcoming NG-20 mission, Vulcan-Centaur has entered operation. Less fortunately, Vulcan-Centaur is currently booked for 2024 and 2025, so procuring a launch for emergency Cygnus launches would require purchasing flights from customers. NASA does have the authority to do this; it would then be up to United Launch Alliance and the launch customer to organise a new launch schedule, or for the customer to seek out an alternate launch provider.

SpaceX’s Cargo Dragon Spacecraft, however, is not an agnostic design, and can only be flown on Falcon 9. The crew variant of the spacecraft, the aptly named Crew Dragon, is the sole operational crew system in the United States, and has seen its own demand increasing since its first flights. Axiom Space recently flew its third mission on Crew Dragon, and plans to fly twice this year, its own cadence steadily increasing. Additionally, SpaceX is planning to conduct the first flight of its own private Crew Dragon flights, Polaris Dawn, this year. Increasing Dragon’s flight rate by nature of its design means a slight additional increase to Falcon 9’s flight rate. With Starliner still waiting on its Crewed Flight Test, any other near-term desires for a commercial flight to Low Earth Orbit will fall to Crew Dragon.

Delays in alternative launch vehicles further shovel flights into Falcon 9’s launch manifest. Amazon subsidiary Project Kuiper aims to deploy a large internet mega-constellation, and initially booked rides on upcoming launch vehicles, primarily RS1, Vulcan-Centaur, New Glenn, and Ariane 6. Delays in all of these systems led to changing manifests. While these launch vehicles remain part of Project Kuiper, many launches shifted to the operational Atlas V rocket, including the inaugural launch of Kuiper satellites. In December of 2023, Amazon announced another new rocket entering the program–Falcon 9 Block 5–which will launch Kuiper satellites three times to further increase the expediency of the constellation’s early launches. Additionally, while Vulcan-Centaur recently launched the first flight of NASA’s CLPS program, the rest of the currently contracted CLPS missions with stated launch vehicles are set to launch on Falcon 9 and Falcon Heavy. This pattern repeats for many programs, and as more and more space startups desire rides to space, more fall on Falcon 9’s back. According to Next Spaceflight, Falcon 9 Block 5, not counting Falcon Heavy, has 47 government and commercial launches scheduled for 2024.

Falcon 9 Block 5 is, indisputably, one of the most reliable rockets ever flown. At time of writing, Falcon 9 has had 266 consecutive successful missions, its last failures were in 2015 and 2016, and the modern Block 5 variant has never failed once. With this in mind, even discussing the idea that Falcon 9 could experience a launch failure, despite the ever-increasing responsibility of the system, seems misguided; few people would be surprised if Falcon 9 never experienced a launch failure again. Even the aforementioned Ariane 5 rocket had a mostly successful career after it entered full operation, but therein lies a significant distinction of Falcon 9. Falcon 9 has not yet hit the limits of its operational capacity, and if there is anything the last year of flights demonstrates, it’s SpaceX’s commitment to finding that limit.

Falcon 9’s flight rate has increased dramatically over the last few years, and there is no indication that it will slow down. In 2023 SpaceX set itself the ambitious goal of achieving 100 flights across all its operated launch vehicles. While the total launch count was slightly bolstered by Falcon Heavy and the two flight tests of Starship, the bulk of SpaceX flights in 2023 were of course Falcon 9. Falcon 9 flew a whopping 91 times and only four of those flights used previously unflown core stages. Adding on the flights of Falcon Heavy and Starship flight tests, the total is brought to 98 orbital launch attempts in 2023. While SpaceX did not quite hit a triple-digit launch count this year, the achievement remains impressive. However, there is an asterisk to this flight rate: the main driver of this explosive cadence is not the demands of the market, but SpaceX itself.

Starlink

For all Falcon 9’s launches in 2023, only around a third were customer payloads. Of Falcon 9’s 91 launches in 2023, 63 were flown for SpaceX’s Starlink mega-constellation. SpaceX currently operates over 5,300 functional Starlink satellites according to the webpage of astronomer and astrophysicist Jonathan McDowell. Starlink currently represents the majority of all satellites in orbit, and even now the constellation’s current population represents only half of about 12,000 satellites planned to be simultaneously operational. In February of 2022, SpaceX claimed the capacity to produce 45 Starlink satellites every week, and in the time since has demonstrated the ability to deploy those satellites just as quickly. However, Starlink itself has changed significantly since its inception.

Starlink satellites, much like SpaceX’s launch vehicles, have evolved over time. When operational Starlink launches began back in 2019, each Falcon 9 carried approximately sixty satellites per flight. These satellites represented the initial “1.0” version of the Starlink satellite bus. Eventually, Starlink’s design was iterated on, and upgrades brought with Starlink v1.5 also brought increased mass, meaning Starlink could now only be launched in batches averaging around 50 individual units. Based on SpaceX’s claim of 45 satellites produced per week, a weekly launch cadence remained sufficient to keep up with production. This iteration of the Starlink bus was not its final evolution, however, as SpaceX intended for Starlink deployment to be handed over to their forthcoming Starship/Superheavy launch vehicle. With the increased payload capacity of Starship, Starlink would inaugurate its v2.0 configuration, a much larger version of Starlink built to take full advantage of Starship’s increased fairing volume and its ability to deliver high payload masses to LEO. Due to this increased scale, SpaceX emphasised that Starlink v2.0 requires Starship, but slips in the launch vehicle’s development necessitated a change of plans.

On February 27th of 2023, SpaceX launched its first set of Starlink v2.0 “mini” satellites. This latest iteration of Starlink is a downscaled version of v2.0 that improves significantly upon the performance of v1.5, while being small enough to fly in Falcon 9’s fairing. This version of Starlink is intended to bridge the gap while Starship remains in development, but comes at a cost. While v2.0 mini is small enough to fit within Falcon 9’s fairing, it remains significantly heavier. Starlink v2.0 mini can only be launched in sets as high as 23 individual satellites, roughly half the number of units per launch seen with v1.5. While each satellite has more capacity than their predecessors individually, maintaining global coverage still requires the same amount of individual satellites. In order to maintain the same rate of deployment, the previous weekly cadence of Starlink launches has become biweekly, as we’ve seen from Falcon 9 this year. Delays to Starship’s development have placed a considerable burden on the Falcon 9 vehicle, forcing it to double its flight rate out of necessity. Furthermore, the stated on-orbit lifetime of earlier Starlink generations is around five years. Therefore, over the next two years we should see the first launched Starlink groups be retired and deorbited. Without Starship, in order to maintain the population of the Starlink mega-constellation while continuing its growth, Falcon 9’s flight rate will increase even further. According to SpaceX flight software engineer Catherin “Erin” Ishimoticha, SpaceX’s flight goal for 2024 is 144 launches.

Thankfully, as previously stated, Starlink is an internal payload produced by SpaceX themselves. If the demanding flight rate does prove strenuous for Falcon 9 over time, it is within their own capability to scale back, though that does not seem to be their current intention. It is also not entirely clear what development milestones are required before Starship will be permitted to launch Starlink missions, and the precise timeline for Starship’s development can be hard to predict. Ship 25, which flew as part of Starship’s first Integrated Flight Test, was observed to have a hatch speculated to be designed for the deployment of Starlink satellites, but by the time of launch this hatch had been welded shut. Ship 28, which is expected to fly as part of Integrated Flight Test 3, also has this hatch, and will test it on-orbit according to statements made in the January 9th Artemis Program teleconference. Depending on SpaceX’s willingness to fly Starlink satellites on Starship test flights, it’s not out of the question that the beginnings of Starlink’s handover to the new launch system could begin in 2024, but that remains deep in the realm of speculation. For the time being, if Starlink remains a major priority and Starship continues to stumble on its path to operational flights, SpaceX cannot meaningfully decrease Falcon 9’s flight rate while still maintaining their mega-constellation.

Potential Strain?

Despite Falcon 9’s increasing flight rate, payloads continue to be delivered safely, but what adjustments and compromises are being made to accommodate and maintain Falcon 9’s high-cadence? SLC-40 has been the primary site of Falcon 9’s increasing cadence since its reopening in 2017. While SpaceX began launches from Pad 39A that same year, its primary purpose is supporting launches of Crew Dragon and Falcon Heavy. As such, the rate of launches from 39A has not dramatically increased. SLC-40, however, has seen an almost exponential increase in launches since flights from the pad restarted following repairs and adjustments made in the wake of an on-pad explosion in 2016. As described in a Smithsonian Magazine article, In December of 2017 John Muratore, who headed SpaceX operations at SLC-40 at the time, explained to reporters that SLC-40 would be able to handle one launch every week. Falcon 9 went on to launch from SLC-40 12 times that year, a far cry from the pad’s supposed maximum cadence. 2023, however, was the year that SLC-40 met Muratore’s claimed cadence, with Falcon 9 launching 55 times from SLC-40 alone. Falcon 9’s third available launch pad, SLC-4E, sees overall fewer launches, but its cadence has increased at a similar rate. According to a report by Space Explored, SpaceX’s Vandenberg launch manager Nate Janzen claimed 2024 will see 50 launches from SLC-4E alone. If current trends continue, 2024 could see these launch facilities operating far beyond their originally targeted operational limits, while a new Falcon family launch site, SLC-6, is not expected to be operational until sometime in 2025.

Launch sites are the most visible aspect of Falcon’s flight pipeline. How SpaceX is adjusting to increasing the rates of vehicle processing and integration is harder to speculate on, and even assertions regarding the limitations of Falcon’s launch facilities remains largely conjecture. Facilities are additionally not the only limitation to the rate at which spaceflight operations can be conducted. The human element, the safety and health of SpaceX’s workforce, is often even less visible to outside observers. A recent investigative piece published by Reuters, while more so focused on SpaceX’s Starship Program than on Falcon 9, does provide insight into the consequences of SpaceX’s rapid pace. Ultimately, only SpaceX management can truly know the relationship between their responsibilities and operational capabilities, or what required and possible flight rates look like. It is SpaceX’s obligation to ensure that reaching their own operational targets and fulfilling their commercial launch orders does not compromise the safety of their system, particularly in this period where the US component of ISS relies on Falcon 9 alone.

A historical parallel could potentially be drawn to the culture seen in the early years of NASA’s Shuttle Program. In the early 1980s, the ambition was for all US payloads to take advantage of the Space Transportation System’s nature as a partially reusable launch vehicle, and its potential for a high flight rate. At the time, the goal of reaching 25 flights a year, every year, was as lucrative as it was ambitious. 1985 was the Space Shuttle’s most daring year yet. That year, the Space Shuttle fleet flew a gauntlet of nine challenging missions. STS missions successfully deployed a slew of commercial, scientific, and national security payloads. The system delivered astronauts from around the world to space, from nations with no prior crewed spaceflight experience, and even flew sitting US politicians. 1985 also saw the inaugural launch of Space Shuttle Atlantis and the first fully operational Spacelab mission. 

Outwardly, 1985 seemed to be the best year for STS yet. An Abort to Orbit on STS-51-F or the impromptu repair operation on STS-51-D were not seen as indications of a system straining to meet programmatic requirements, rather as acceptable growing pains. Going into 1986, STS seemed ready to face another challenging year of flights. The schedule laid out 15 ambitious flights; the first saw the flight of another US politician, future NASA administrator Bill Nelson, and was a successful start to a new year.

Engineers famously warned NASA executives of the risks of pushing Challenger to fly its next mission, STS-51-L, knowing that the low temperatures on launch day could result in the O-ring joints which join the segments of STS’ solid rocket boosters together to become brittle and fail. The decision to proceed with launch was the result of many factors, programmatic pressure to maintain the Shuttle’s flight rate was among them. Additionally, degradation and erosion in the O-ring joints had been observed on boosters recovered from previous, successful, Space Shuttle flights. The risk of failure was diluted and improperly communicated, to the point where failure was determined to be an acceptable remote possibility, especially stacked against the risk of setting program goals back.

STS-51-L is known today as one of the most infamous tragedies in human spaceflight. The loss of Challenger’s seven crew members, including a school teacher, was a harrowing lesson in the costs of pushing sensitive operations to their limits, and an incredibly public one. The signs that the STS program was already operating at its current limits were already there, simply ignored. The disaster sent shockwaves through the industry that are still felt to this day. The Department of Defense expanded its Titan IV program from 10 to 39 launches, and the Atlas V rocket developed to replace Titan still flies to this day, and has flights booked as far out as 2029.

This is not to accuse SpaceX of being on the brink of an event on the scale of the Challenger Disaster and disruption to the launch ecosystem. Falcon 9 and Crew/Cargo Dragon represent a very different system; for one, crew and cargo flights are separated, and Crew Dragon has a demonstrated launch escape system with incredibly high NASA oversight. Furthermore, previous stand-downs of Falcon 9 have not lasted longer than a year. Statistically, if Falcon 9 is to experience a launch failure in the future, it would most likely occur during a Starlink launch; no crew would be at risk. The number of scheduled Falcon 9 launches in a given timeframe is far higher than it was in 2016 however, so it’s likely a failure would be more disruptive now than it was then.

The purpose of this analogy is to draw comparisons between two very different systems, highlighting ways in which these complex machines can approach their limit when supported by workplace culture. Could SpaceX make the decision to slow down Starlink flight rates, or at least stop increasing the launch rate until SLC-6 comes online, or until existing launch facilities can have their operational limits assessed? Potentially. Will SpaceX choose to slow down a program stressed by their CEO as a necessary revenue stream, especially when the alternative launcher’s operational timeline is unclear? Not very likely. It is possible that SpaceX has already determined these limits, and has a set internal target that they haven’t reached yet. It could be possible that Falcon 9 will continue to bear Starlink’s increasing launch demands until Starship is online. SpaceX being a company famous for its mentality of learning from flying and testing to failure gives cause for some alarm, but again all remains very speculative.

Closing Thoughts

Even when taken as a whole, either at face value or with some further speculation involved, it can be difficult to assess the overall risk of a launch failure, for any system, before one actually happens. In fact, if Falcon 9’s flight rate was not continuing to increase and instead levelled at the cadence seen in 2023, there would be little basis for alarm. Furthermore, when the previously mentioned SLC-6 site begins flying Falcon 9, it may serve to alleviate the launch load from SLC-40 and 4E, allowing both flight rate and pad down-time to increase. Alternatively, SLC-40 and 4E may maintain their cadence growth, and SLC-6 will simply work to raise the overall launch rate alongside them—only time will tell.

Regardless, it remains the case that a market in which all payloads are funnelled into one launch system is, no matter how effective said system is, inferior to one that includes alternatives in regards to mission assurance. As far as NASA and the International Space Station program are concerned, Falcon 9’s time as the load bearing rocket will soon be coming to a close. Both Boeing’s Starliner and Sierra Nevada’s Dreamchaser are scheduled to fly vital test flights to the ISS this year, no sooner than mid-April. The ISS program could enter 2025 with operational flights of new, fully parallel crew and cargo spacecraft under its belt. Meanwhile, for the wider commercial market, it may take some time for new launch systems to not only come online but reach their full cadence as Falcon 9 did many years ago now. However, 2024 has already seen the inaugural launch of Vulcan-Centaur, and Blue Origin has indicated it will hopefully see the first launch of New Glenn later this year. It is safe to say that regardless of if or when Falcon 9 experiences a major setback, it will not be much longer until other systems exist to fall back on. The launch landscape may enter a bold new era soon enough.