Inside Rocket Lab: How Sir Peter Beck Is Building Neutron’s Engines | Tour 02

Summary

• Rocket Lab’s distressed-asset playbook is real and ongoing: Beck bought the former Virgin Orbit facility — “over $100 million worth of assets in that building” — for $16 million while it was in bankruptcy, machines included. He confirms keeping a tracking list of potentially bankrupt space companies and their facilities (“Oh, yeah”; “Yep”) but won’t name the top target: “That’ll get me in trouble.” He signals there will be more.
• Beck’s sector critique cuts at both zombie companies and their backers: “space companies are really hard to kill,” lingering on serial funding rounds because claims are “really, really hard to corroborate unless you go five layers deep into the rocket equation.” His sharpest line on diligence: “sometimes the really big firms that you think are really good at it actually are the worst.”
• Archimedes is deliberately anti-heroic engineering: it is a staged-combustion engine, unlike Rutherford’s electric pump cycle, and Beck’s goal was “the most boring engine possible.” Where Merlin and Raptor are “really strung out” high-performance engines, Archimedes is built for reuse. Qualification is one hour of running and 40 starts, versus a ~5-minute acceptance test for the expendable Rutherford, which only needs to run for 190 seconds in its life.
• The whole Neutron architecture flows from an “absurd” design requirement — turn the vehicle around in 24 hours — which helped drive methalox over kerosene (no soot in the regen channels) and “a whole lot of really good decisions.” The tradeoffs bite both ways: low chamber pressure aids longevity but makes the engine “much harder to light.”
• Additive manufacturing is the cost weapon: most of both engines by mass is 3D-printed, ~10 people run the automated shop, and Beck calls Archimedes “an incredibly cheap engine… just ridiculously so.” Next year Rocket Lab becomes the world’s first customer of a machine that prints a whole Archimedes — “tall as me” — in one piece.
• Production and test cadence is industrial, not artisanal: one Archimedes every eight days, one Rutherford per day, 930+ engines have been to space, two Stennis test cells operating 20 hours a day, seven days a week; Electron has flown 93 times, second only to Falcon 9. Staged combustion can’t be tested piecewise, so the program is hardware-rich by design: “you literally eat engines.”
• The strategic template: future space giants all look alike — own rocket, own satellite production, own applications — and the Iridium deal (as discussed) is “just the start of our applications layer” because “all the cool kids have got a comms layer.” His closing hot take: “the biggest thing to be done in space hasn’t even been thought about… it’s like we’ve sent our first email at the beginning of the internet.”

Deep dive

1. The $16M Virgin Orbit steal — and the hunting list behind it

• Beck calls the engine development center “arguably one of the best deals of my life”: the old Virgin Orbit facility held “over $100 million worth of assets,” and Rocket Lab bought the whole thing — all the machines included — for $16 million while it was in bankruptcy. They piled into golf carts when the opportunity came up. The payoff: “we never had to worry about a factory for building engines ever again.”
• He keeps a watchlist of potentially bankrupt space companies and the facilities he wants, says there will be more, but refuses to name the top of the list: “That’ll get me in trouble.”
• His diagnosis of the sector’s zombie-company problem — “space companies are really hard to kill… the number of space companies that should be dead and aren’t… is quite incredible” — points partly to space enthusiasts and to the promotional difficulty of verifying claims “unless you go five layers deep into the rocket equation.” Asked if investors do that work: “Not really… sometimes the really big firms that you think are really good at it actually are the worst.”

2. Additive manufacturing is the cost-and-speed engine

• Most of both Rutherford and Archimedes by mass is 3D-printed, but the point is design consolidation, not gimmickry — “we’re not stupid. We don’t 3D-print bolts” — printing multiple parts as one complex architecture, because “at the end of the day, it’s all about speed and cost.” Roughly 10 people run the entire automated shop, printing Inconel, copper, titanium, and a proprietary alloy developed in-house for Archimedes’ environment.
• The hardware economics as told: a 12-laser powder machine is less than $10 million, while some machines can cost more; the longest print runs about three and a half days; the Archimedes thrust chamber is copper clad with a superalloy. Next year a machine arrives that prints a whole Archimedes engine — “tall as me” — in one print: “we’re the first customer of it in the world.”
• The output claim worth underlining: “an Archimedes engine is an incredibly cheap engine. Same with a Rutherford engine. It’s just ridiculously so.”

3. Archimedes: engineered to be “the most boring engine possible”

• Reusability inverts the design problem. Archimedes is a staged-combustion engine, while Rutherford uses an electric pump cycle. Expendable Rutherford only needs to run for 190 seconds in its life, so a roughly five-minute acceptance test suffices; Archimedes must do 40 starts and run for one hour in qualification. Beck wants it to “just go and go and go and go” without concern. His analogy: looking at the engine on an airplane wing, “you wanna know that engine is boring, right? It’s not gonna blow up.”
• The fuel choice follows from reusability and a deliberately extreme requirement — turn the vehicle around in 24 hours, “an absurd design requirement” that “drove a whole lot of really good decisions.” Kerosene leaves soot in the regenerative channels and requires purging; after a methane run the engine is “still shiny stainless steel… just no residue whatsoever.”
• The tradeoff chain, kept intact: relatively low chamber pressure for a staged-combustion engine lowers internal temperatures — good for longevity but “much harder to light.” “You create these things that are ultimately good, but you create a whole lot of problems along the way.”
• On rivals’ engines — Merlin and Raptor — Beck is candid that everyone studies everyone. Those are “really high-performance engines… really strung out,” while Archimedes is designed to be benign and durable; their lessons “probably don’t apply to this so much” because Rocket Lab has its own problems.

4. Hardware-rich development and a brutal test cadence

• The teams have worked on the engine for about three or four years. The initial design was quick, but staged combustion can’t be decomposed for testing: the first hot fire needs the turbopump and everything integrated. The program therefore has to be hardware-rich from the start because it consumes a lot of hardware — “you literally eat engines.”
• His favorite war story: an early hot fire leaned out and “consumed the whole injector… there was literally just a big hole left,” yet the propellants kept mixing, producing thrust, and the engine kept running. The conclusion: “it’s very, very difficult to analyze. You’ve just got to test.”
• The numbers: the Archimedes line produces one engine every eight days; one Rutherford engine is made and shipped out every day; 930+ engines have been to space; two test cells in Stennis, Mississippi, operate 20 hours a day, seven days a week; and Electron has flown 93 times, second-most frequently launched behind Falcon 9. While describing the Archimedes components, Beck also cites a “plus or minus 15,000 horsepower” figure without clearly assigning it to one component.
• Neutron carries 10 Archimedes engines — nine on the first stage and one on the upper stage — because the upper-stage thrust “coincidentally is the ideal amount of thrust to land a rocket.” “It’s just physics.”

5. Vertical integration climbs into applications — and the “first email” thesis

• Every piece of Electron — hardware, software, tanks, engines, and more — is built by Rocket Lab, and Beck names it directly: “that’s one of the key successes of the company, is that vertical integration.” The same philosophy applies to satellites: the flight computer, reaction wheels, and solar panels are all Rocket Lab’s.
• Responding to O’Shea’s mention of the Iridium acquisition, Beck calls Iridium “incredibly strategic” and “just the start of our applications layer” — jokingly, “all the cool kids have got a comms layer.” His template for the industry: the big space companies of the future all look alike, with “their own rocket… their own ability to build as many satellites as they need, and they’re all going to have applications.”
• Tourism is possible, but Beck says he would not be a very good tour operator, is not sure it is exciting for him or the company, and worries about the CEO’s responsibility if something goes wrong: “if anything ever goes bad, guess who’s knocking on the door?”
• The hot take that closes the episode: after 20 years, the industry is well past the government-to-commercial-launch democratization milestone, and “the biggest thing to be done in space hasn’t even been thought about, let alone talked about… it’s like we’ve sent our first email at the beginning of the internet.”