Explosions Are Great, Actually.

The history of spaceflight is the story of weirdos, explosions, and stacked odds.

Yesterday, some very lucky people got to play Kerbal Space Program in real-life. And luckily, we got to watch.

Their prototype was a 394’ (120m), 11 million lbs (~5,000t) behemoth that generated twice the thrust of the Saturn V at launch. Nominally, it’s going to consume somewhere between 44 thousand lbs (~19.7t ±0.4) to 46 thousand lbs (21.45t) of propellant (methane and liquid oxygen) per second.¹

For comparison, the Saturn V consumed around 28 thousand lbs per second (or, more precisely, 12,885kg/s). Using the international unit of African bush elephants, Starship will consume an extra ~51 bush elephant equivalents of propellant per minute, give or take some elephants (but what are a few elephants between friends?).

Their project is so massive that when they lit the candle, its infrared signature could be observed from Geostationary Orbit,

A compilation of IR data from NOAA’s & NASA’s GOES-16 satellite. The satellite is in geostationary orbit and provides continuous high spatial and temporal resolution imagery of the Americas.

The satellite that captured the above IR images (GOES-16) also saw the rocket’s plume (1), its shadow (2), and its destruction(3). The satellite’s sensors capture images in strips that are stitched together. Accordingly, different parts of the image below are likely to be from different time points.

The test ended in the most Kerbal way possible. 172s into the flight, they lost control. After 67 seconds of donuts, as their flight control system worked furiously to regain control, whoever was in charge of Range Safety triggered the vehicle’s Flight Termination System. The prior sentence is a fancy way of saying that they triggered explosive charges on the fuel tanks, rupturing them to reduce the size of the eventual explosion.

It was spectacular.

Pre-launch, I tried to start a betting pool around the launch’s outcomes. No one took me up on the offer.

People were surprised that I was so cheerful about the rocket meeting a fiery end. I tried to explain that it's the first time in history that someone has lit 33 rocket engines simultaneously. The record was previously held by the Korolev’s N-1. All four of its prototypes met a fiery end.²

SpaceX is trying to outdo them. They’re building something 60% more powerful than the N-1. Something that pumps many, many elephants of propellant per second. And they’re trying to build it in a way that can withstand the violence of flight and launch with minimal refurbishment a dozen-plus times.

The odds have always been stacked against them.

Post-launch, the reaction was split. Almost everyone in the space community loved every second of it. But outside the community, most of the world saw it as a failure and a waste. This sharp split isn’t something new. It has existed for as long as we’ve been building rockets.

Explosions and rocketry go together like bread and butter. SpaceX didn’t invent this golden combination. It has always been this way.

Back when Robert Goddard was a newly minted undergraduate trying to make his way in science, he was kicked out of Worcester Polytechnic Institute for causing one explosion too many. Before JPL was JPL, it was the Suicide Club (Or, Suicide Squad), so named because its members were clearly suicidal. They were kicked out of Caltech after an explosion that either set a dorm on fire or embedded a steel plate in a wall (depending on whom you ask). But either way, kicked out they were. As John Clark put it in ‘Ignition!’, “Pioneers are seldom appreciated.”

Out in the desert, the Suicide Club blew up many an experiment on their test stand. But work continued, culminating in JPL’s first rocket experiment,

Four times that day they tried to test fire their small rocket motor. On the last attempt, they accidentally set fire to their oxygen line, which whipped around shooting fire! These were the first rocket experiments in the history of JPL. They tried again on Nov. 15, 1936, and their experiment finally worked.

— JPL’s Early History. Accessible here.

Their work eventually became the first rocket startup, Aerojet, that developed and deployed the first commercial rocket motors in the US. One-by-one, starting with the dropout Jack Parsons, the founders were kicked out and lost control. As von Kármán put it,

Unhappily for us, no bank would lend us money; bankers hadn’t yet come to think of rocketry as a stable business.

— von Kármán, Theodore & Lee Edson.”The Wind and Beyond.” 1967.

Even as rocketry grew up, it remained true to its explosive roots. Much of the safety data used today was derived in experiments with a high-eyebrow singe potential,³

Came the day of the first trial. The propellants were hydrazine and WFNA. We were all gathered around waiting for the balloon to go up, when Uncle Milty warned, "Hold it—the acid valve is leaking!"

"Go ahead —fire anyway!" Paul ordered.

I looked around and signaled to my own gang, and we started backing gently away, like so many cats with wet feet. Howard Streim opened his mouth to protest, but as he said later, "I saw that dog- eating grin on Doc's face and shut it again," and somebody pushed the button. There was a little flicker of yellow flame, and then a brilliant blue-white flash and an ear-splitting crack. The lid to the chamber went through the ceiling (we found it in the attic some weeks later), the viewports vanished, and some forty pounds of high-grade optical glass was reduced to a fine powder before I could blink.

I clasped both hands over my mouth and staggered out of the lab, to collapse on the lawn and laugh myself sick, and Paul stalked out in a huff. When I tottered weakly back into the lab some hours later I found that my gang had sawed out, carried away, and carefully lost, some four feet from the middle of the table on which the gadget had rested, so that Paul's STIDA could never, never, never be reassembled, in our lab.

— Clark, John Drury. Ignition!: An informal history of liquid rocket propellants. Rutgers University Press, 1972.

You cannot do the impossible without a few explosions along the way.

The journalist Jay Barbree⁴ once called Alan Shepard “a cool cat” and an “educated daredevil” during an interview. There is no better description of spaceflight.

You have to be a little crazy and a bit of a daredevil to do it. The field was started by misfits and hermits who dreamed of worlds beyond their own, and it is continued by successive generations of weirdos seeking to do the impossible in some shape or form. Be it landing a car on Mars or building skyscrapers that can fly.

But getting the golden mixture of sanity, knowledge, and craziness right is critically important. It is the same as rocket fuel, the wrong mixture mixed in the wrong proportions will blow up in your face. It is critically important that craziness is never mixed with stupidity, or you end up in the desert running away from carelessly handled UDMH and NTO.⁵

We are splitting a fine line here. Most people are unable to distinguish between those capable of pulling off impossible feats and those who are just unhinged. Careful, calculated discernment is necessary.

Calculated discernment is where the educated part of ‘educated daredevil’ comes in. Every single grand space project that has succeeded was successful because of people who understood every part of the system they were trying to create and could discern the difference between acceptable risk and stupidity. That’s what separates the Alan Shepards from the Lawnchair Larrys.

It is at this point some people will remark, “Why not make it as safe as possible then? And make sure that the system works no matter what?”

The problem is that we’re primitives. We’re dealing with immense energies and forces with materials and understanding that’s still quite rudimentary. For example, the first stage of Saturn V’s raw thermal energy output was around 46,788.5 MJ/s per engine. Or 233,942.5 MJ/s for all five engines. For a 168s burn duration, the total thermal energy that was channeled by the machine is roughly equivalent to 2/3rds the yield of the weapon used at Hiroshima. At lift off, the output power of the Saturn V exceeded the output of all nuclear power stations operational in the US at the time.⁶

Starship’s first stage is in excess of that.

The history of spaceflight shows that holding on to the idea that these forces can be dealt with without any risk is a dangerous notion. It’s also an expensive one. And you end up falling into the trap that post-Apollo NASA fell into,

Risk is a combination of the probability of an undesirable event and its consequences. There are different flavors of consequences: technical, cost, schedule, safety, political. As a project becomes more expensive, the financial consequences of failure become more dire. That risk can be mitigated, but those procedures are expensive, increasing the cost, thus increasing the risk…ad infinitum.

Because of spiraling costs and schedules, NASA did not launch a single planetary mission between 1978 and 1989.

— Frank, Elizabeth. Faster, Better, Cheaper: A maligned era of NASA's history. 2019. Accessible here.

During the ‘90s, President Clinton appointed Daniel Goldin to be NASA’s administrator, he is the one who diagnosed and first defined the above paradox of how chasing safety led to greater and greater risks,

There’s a paradox at work here that creates a downward spiral. Launching fewer spacecraft means scientists want to pile every instrument they can onto whatever’s going to fly. That increases the weight, which increases the cost of the spacecraft and the launcher. Fewer spacecraft also means we can’t take any risk with the ones we launch, so we have to have redundancy, which increases weight and cost, and we can’t risk flying new technology, so we don’t end up producing cutting edge technology.

— Daniel Goldin, 1992-3(?)

Administrator Goldin’s answer was simple. They were going to eliminate the systems management technique used during Apollo, and they were going to leverage commercial-off-the-shelf advances in miniaturization to put new technology in space to perform new missions faster, better, cheaper.

Tacitly, though not explicitly, he endorsed the idea that risk was okay. And the result was a mini-golden age.

During Administrator Goldin’s Faster, Better, Cheaper (FBC) era, NASA frequently had projects that were delivered before time and under budget,

The Viking mission to Mars in 1976 cost $1.06 billion in real-year dollars and took 6 years to develop. The Pathfinder team was instructed to send a lander AND rover (Sojourner) to Mars in half the time and 1/14 the budget. They succeeded.

— Frank, Elizabeth. Faster, Better, Cheaper: A maligned era of NASA's history. 2019. Accessible here.

Sojourner was the first successful rover on Mars, and the entire team’s budget was less than that of a single instrument on Viking.

Of course, with acceptance of risk comes risk. In 1999, four out of five of the FBC missions failed spectacularly. It didn’t matter that the total cost of all 16 FBC missions was still less than the Viking missions and Cassini. It didn’t matter that NASA’s Inspector General was asked to investigate FBC and recommended “fully incorporating” it into NASA’s approach. What mattered to the politicians, the press, and the public was that NASA failed. The affair was a “national embarrassment.”

Administrator Goldin was hauled in front of Congress, and that was the end of NASA’s SpaceX era.

There is a sharp difference between those who play status games, and those who are interested in creation and exploration. That difference can be measured in explosions.

1. Derived from an appendix submitted by SpaceX to the FCC. Engines were ignited around at T-3s, MECO was supposed to occur at 169s. Assuming depletion of all 3,400t of propellant, this gives 19.76t/s. The estimate can also be derived bottom up, each Raptor 2 engine consumes 1,400 lb/s, or 650 kg/s. For 33 engines, 21.45t/s. I’ve given both of these estimates for the sake of accuracy. It’s likely that the truth is either somewhere in between or around these numbers.

2. You don’t have to be particularly clever to work out the odds.

3. I am forever grateful that they risked their eyebrows so that I don’t have to risk mine.

4. He covered every crewed launch of the US space program between 1961 and 2011. He also wrote moonshot with the Mercury 7’s Deke Slayton and knew the Mercury 7 astronauts personally.

5. Rule of thumb, if you don’t immediately know what these acronyms mean, you definitely shouldn’t be messing around with hypergolics.

6. Thanks Dennis Silin for double checking my calcs and the interesting comparisons!