SpaceX Evolution (Appendix H: Secret Sauce)
A serialized version of this book for SubStack
Link to First Installment: https://chrisprophet.substack.com/p/spacex-evolution
As we’ve already discovered, Elon Musk companies have essentially created the perfect medium for tech development. To that end they use a unique technique to actually produce these new technologies, which could be described as their ‘secret sauce.’ They call this technique “the first principles approach to technology” which consists of a five step process to progress from ideation through to mass production, as follows: -
1. Make the requirements less dumb
When planning to build a new piece of technology, whoever sets the initial requirements is often wrong. New realms of technology are uncharted, so misinformed opinion drives the direction of development, often leading to sub-optimal solutions. Therefore the first step is to refine what’s actually needed then determine the best way to physically achieve that goal.
Example: SpaceX planned to recover their Falcon 9 booster via parachute, which subsequently proved impractical because the booster velocity was too high when it hit the atmosphere. Therefore they refined their terms by deleting the parachute requirement, allowing them to progress onto the more practical recovery technique of propulsive descent and landing.
2. Delete the part or process step
“If you’re not adding things back in at least 10% of the time, you’re clearly not deleting enough.”1 ~ Elon Musk
Systems tend to complexity to account for every possible application and ensure safety. Unfortunately complexity increases the number of components which can go wrong and overall cost. Hence the second step after developing a prototype is to cut the number of components and processes it performs to the bare minimum.
Example: the Raptor 1 engine used bolted flanges to seal the propellant manifolds. These flange seals had to withstand partially combusted propellant at over 12,000 pounds per square inch pressure, hence often leaked or failed outright.
“SpaceX metallurgy team developed SX500 superalloy for 12000 psi, hot oxygen-rich gas. It was hard. Almost any metal turns into a flare in those conditions.”2 ~ Elon Musk
For the Raptor 2 engine SpaceX opted to weld these flanges together, removing any need for fixings, which reduced mass and improved the flange seal.
3. Optimize
Once you’ve reduced the number of parts and processes to the bare minimum, redesign each to improve their efficiency and performance. Often this involves a holistic approach to identify what is most in need of improvement and how things could work better together, instead of individually.
Example: the full thrust revision of the Merlin engine dispensed with the majority of control valves to increase propellant flow rate, resulting in a 17% improvement in engine thrust. A subsequent refinement of the engine produced another 16% thrust and improved the overall reliability and serviceability, allowing it to be flown 10 or more times without major refurbishment.
4. Accelerate
Increase the iteration rate as you close on the final product. The last 10% of development is usually the hardest so needs to be attacked with gusto.
Example: In November 2012 SpaceX announced they would develop a methalox engine, that would later be called Raptor 1. In December 2021 Elon announced the Raptor 2 version had entered production – then in May 2023 he announced Raptor 3…
Elon stresses the three prior steps must be thoroughly completed before attempting to accelerate, otherwise you’re probably developing the wrong thing: -
“If you are digging your grave, don’t dig it faster, stop digging your grave!”3
5. Automate
Once all prior steps have been performed and you have iterated as close as possible to the final product – then automate the production process.
Example: Elon spent considerable time trying to automate assembly of battery mats for the Tesla Model 3 vehicle; until he discovered the mats served no purpose and could be deleted…
Five steps of engineering wisdom gleaned from a lifetime developing cutting edge technology, from ideation through to a practical production. Of course this first principles approach to development works best in the correct medium, because without culture there’s chaos.
2 https://x.com/elonmusk/status/1076684059827302400
Many do not seem to appreciate the power of first principles reasoning and the power of deletion. Musk's approach to everything, not just the products, but the companies, and now the government...is to delete things until you have to add them back.
Once you need to add things back, you know where the constraints are. Just like a rocket exploding, many perceive this as a bug in the process, but it's not. It's a feature designed to push the envelope harder than anyone else.