Category: Spaceflight

  • In the late hours of June 18, SpaceX experienced a catastrophic failure of Starship 36 during propellant loading for a routine static fire test — yet another in a string of setbacks for the program this year. Rapid iteration can be a valuable tool, but celebrating failure as inevitable — without accountability or thorough investigation and remediation — poses a dangerous precedent for spaceflight and needlessly damages public perception.

    Ship 36 exploded during what should have been a routine operation for a rocket coming up on its tenth flight. The next morning, SpaceX released a statement that — while emphasizing that data reviews were ongoing — stated it appears that a “potential failure of…a COPV containing gaseous nitrogen in Starship’s nosecone area” caused the anomaly. Sure, sounds plausible. COPVs are notoriously fragile and have caused failures of SpaceX rockets in the past, but the failure of Ship 36 isn’t happening in a vacuum.

    This is the fourth major mishap for the Starship program just since the beginning of the year, following the in-flight failures of the Ship during the past three IFT missions. In the aftermath of each of these anomalies both SpaceX and its fans say the same thing: “It’s okay, we got good data! Fail fast, learn faster!” You don’t seem to be learning much if you fall flat on your face four times in a row.

    Both SpaceX and its fans seem to look at failures and shrug, labeling it as progress and not risk, regardless of actual outcomes. Constantly looking at a spaceflight program through this lens only serves to make real safety lapses invisible until a catastrophe unfolds. Refusing to come out and admit failure, instead opting to shroud it behind terms like “learning opportunity,” “data gathering,” and casual refrains diminish the gravity of rockets exploding. This vehicle was only partially filled and still caused significant damage to infrastructure. Imagine how much worse a full Ship or integrated launch vehicle would be.

    While nobody was injured this time, frequent large scale public failures of Starship over the past five years has eroded public trust in not only SpaceX, but the regulators that have oversight over the company. Are agencies like the FAA or NASA — who SpaceX has a lucrative contract with to provide Starship as a moon lander — losing leverage if failures become routine? Without the deterrence of regulatory scrutiny and strict oversight across the entire lifecycle of a rocket, complacency may spread.

    The future of U.S. lunar exploration hinges on Starship working. NASA needs the Starship HLS variant ready to land humans on the Moon by 2027 in preparation for the Artemis III mission, but repeated failures are slowing progress and sowing doubt that SpaceX will be able to deliver. Not only does Starship need to demonstrate that it can successfully get to orbit and back, but it also needs to demonstrate rapid reusability, large scale on-orbit cryogenic refueling, and be crew rated — all in the next two years. That goal seems more far-fetched every time Starship experiences a failure.

    Logging what went wrong and calling it a learning opportunity isn’t enough. Organizations must transparently fix root causes and findings as has been the norm for decades. In the wake of the Space Shuttle Challenger disaster, the Rogers Commission mandated a total redesign of the Space Shuttle Solid Rocket Boosters — adding capture features, heater circuits, extra O‑rings — and instituted independent oversight and a formal Office of Safety, Reliability, and Quality Assurance. The fleet was grounded for nearly three years, and NASA allowed only a cautious, certified return to flight.

    There is recent precedent for commercial companies as well: After the in-flight failure of Falcon 9 during the CRS-7 mission in 2015, NASA’s Independent Review Team concluded the failure was due to a design error — SpaceX used an industrial‑grade strut instead of an aerospace‑grade one, and lacked proper safety margins. Following the report’s release, SpaceX overhauled its design processes, upgraded strut materials, and incorporated stricter part screening before returning to flight in.

    No such independent reports have been issued after any of the latest string of Starship failures as they are not required because no NASA payloads have flown. This only serves to foster an environment where indifferent risk-taking is prioritized over slower methodical failure analysis and disposition. The development of Starship is funded in part by public tax dollars. It should be scrutinized as such.

    Spaceflight is inherently risky. There are infinite ways a launch can go wrong and only one way it can go right, but that does not mean we should settle for routine explosions. Regulators, relevant government agencies, and launch providers should develop industry standards to determine when continued failure should trigger serious review or programatic changes, and agencies funding these programs in part — like NASA and the DoD — should be empowered to commission independent reviews of failures.

    None of this is to say we shouldn’t celebrate data-driven learning, but we should weigh it against the expense of public trust, crew safety, and long-term sustainability. Normalizing failure or hiding it behind euphemisms only serves to dull curiosity and critical evaluation — spaceflight deserves better.

    This latest in a string of failures coming from Starbase is not just another data point showing an uncomfortable trend. It is a symptom of an industry-wide culture that has grown far too forgiving of repeated missteps. We thrill at ambition — but without responsible accountability, “fail fast” becomes “fail forever.”

  • I was elated when I saw hardware I built go to space for the first time in January. I had worked for my current employer for a bit over a month and had never imagined it would happen so quickly. However, in that mix of emotions was a deep-seated discomfort: this was a rocket headed to orbit — it could have very easily been headed to a war zone. Nearly everything in aerospace is considered “dual-use,” meaning it has direct applications for civilian and military use. It’s the reason — aside from math being hard and me hating school — that I dropped out of my aerospace engineering degree two and a half years ago. I couldn’t stomach the fact that the technology I would be developing could be used both to land on the Moon and take the life of another person.

    And yet, here I am — in the industry anyway. Crawling back not as an engineer, but as a technician. Not building missiles, but orbital rockets. The distinction matters to me, even if it’s a blurry one. I didn’t return to aerospace because I made peace with the dual-use problem. I came back because I want to further our reach into the cosmos, and because this job lets me put my hands on the machines that leave Earth. That’s magic. It still feels like magic.

    But magic doesn’t erase morality.

    The rockets I build are peaceful, at least that’s what I tell myself. They launch satellites that improve life on Earth, Moon landers, and probes studying the cosmos. That distinction matters to me, but it comes with a single, very large, caveat: the Department of Defense is probably our largest customer. Lol, perhaps lmao. Whether it be an early-warning missile defense satellite for the Space Force or a [REDACTED] for the National Reconnaissance Office, these are all tools for war that my beloved “peaceful” rocket launches.

    With the same rocket — just a different payload, a different customer — a technology of wonder becomes a technology of war. The lines between NASA, commercial space, and the DoD aren’t lines at all — they’re gradients. And if I’m being honest, there are days I’m not sure where my values fall along that gradient.

    This isn’t an abstract problem for me. We have contracts with the DoD, I have friends and coworkers that work on things they can’t talk about, and we launch our rockets from a Space Force base. Hell, when someone asks me what I do for work and I want to keep it vague I tell them I work in defense. It’s the truth, but sometimes the truth isn’t what we want it to be. I’m all for strengthening the might of the U.S. military when it’s resources are helping Ukraine target Russian assets on occupied land, but when kids in Gaza are being turned to pink mist by airstrikes, it’s hard not to feel complicit in something monstrous.

    There’s no easy resolution here. I don’t have a tidy moral calculus that tells me it’s okay to keep doing what I do. I just know that this industry, this work, still calls to me. I believe in space — not in some capitalist, “space economy” kind of way, but in the old-school dreamer sense. I want us to explore. To learn. To connect people and maybe one day stand on another planet not as conquerors, but as caretakers, explorers, and innovators.

    But wanting that future means wrestling with the present. It means acknowledging that space isn’t neutral — it is political, strategic, violent. The same boosters that launch telescopes launch targeting systems. The same companies bidding for NASA science missions are bidding for Pentagon contracts. I’m part of that ecosystem, whether I like it or not.

    Maybe that’s why I’m writing this. Not to absolve myself, and definitely not to condemn others in the industry — many of whom are just as conflicted as I am. But to name the discomfort out loud. To refuse the luxury of ignorance and to remind myself that even if the lines are blurry, my values still matter in how I show up to this work.

  • Below is an edited version of a blog post I wrote for work. Only proprietary and/or export controlled information has been omitted.

    In the [Big Blue building], down the way past mountains of crates, inside the CWA,1 and inside a storage cabinet lives a handful of tubes of Krytox 240AC.2 Now, Krytox is one of those materials you didn’t know you always needed until you first use it. We use it liberally here at [Certain Blue Aerospace Company] for everything from o-ring lubrication to preventing fitting galling. You can find more than a few tubes in just about any work center. However, a startling trend has struck the Krytox supply in my local CWA: not a single tube has its cap.

    I can already hear you saying “What the fuck is this rando yapping about?” I get it, this seems like a stupid thing to not only be upset about but also write a blog post about. My issue lies in the fact that this shows a blatant disregard for 5S3 procedures, lack of work ethic, and a dangerous level of complacency.

    First of all, let us go over why tubes of Krytox — or anything else for that matter — have caps. Caps serve two purposes: to keep the stuff in the tube inside and keep stuff out of the tube outside. Believe it or not, we work on hardware inside of CWAs for a hopefully very obvious reason: keeping components clean. Sure, we use Krytox like it’s aerospace-grade mayonnaise on an aluminium sandwich, but PFPE contamination is a genuine concern for many of the components we work on within the confines of the big bubble cube.

    It isn’t uncommon for tubes of consumables to be covered in small amounts of Mystery Goo — often the material they’re holding. But what about the times the Mystery Goo isn’t what you think?

    That smear on the outside of your Cor-Ban? It might be Krytox from a capless tube. It might also be DOW 730, Loctite, Molykote, or one of the many other word-salad chemicals all jumbled in a cabinet with varying levels of sealing and cleanliness. Without a cap — and without caring — you’ve got no way of knowing. You may not even think twice when you grab a particularly slimy container. In an environment where small amounts of residues can result in bond-compromising contamination or broken down seals that “whatever” mentality suddenly becomes a work work stoppage, a nonconformance, or worse: an anomaly report.

    A missing cap is a symptom. It’s what happens when people treat their tools and workspaces like everybody else’s problem. In a high-stakes environment like aerospace, your discipline bleeds into everything: torques, inspections, even how you log your work. If you can’t take five seconds to replace a cap, wipe off a container, or swap your gloves, what else are you letting slide?

    5S isn’t just a buzzword filled “lean thing” managers like to slap on laminated posters. It’s the foundation of creating a consistent, reliable workflow not mired in chaos. “Sort, Set, Shine” starts with small stuff: replace a cap, wipe a container, swap your gloves. “Standardize, Sustain” is about building a culture where that becomes second nature. If that can’t be done, what are we even doing here?

    Just two weeks ago I removed four tubes from the CWA with no caps, cleaned up the pool of Krytox oil4 that was contaminating everything in the cabinet, and replaced them with fresh materiel. Tonight there was no oil puddle, but there were six capless Krytox tubes. I’ve taped a note on the bench top above the storage cabinet that reminds those using the supplies inside to remember 5S principles.

    I’m not here to be a Krytox cop. In fact, I hate getting needlessly pissed off over problems that shouldn’t exist. But I am here to give a damn about the work we do. We should all take pride in the work we do every step of the way. We’re building hardware that’s going to space for [fuck’s] sake! if something as tiny as a cap can tell you if someone cared — or didn’t — then it’s not just a tube of grease anymore. It’s a marker of pride, ownership, and embracing Team Blue.

    1. A controlled work area. In this case, a clean room. ↩︎
    2. Krytox 240AC is a perfluoropolyether (PFPE) lubricant grease commonly used in aerospace due to its stability in extreme environments. ↩︎
    3. 5S is a workplace organization method that stands for Sort, Set in order, Shine, Standardize, and Sustain, aimed at improving efficiency, safety, and cleanliness. It helps eliminate waste, streamline processes, and maintain a productive, organized work environment. ↩︎
    4. Krytox also available in oil form. ↩︎