5 ms·

What I think is interesting here is how in the larger rocket industry, specially small launch all the competitors talk about how they need to make rocket produc

What I think is interesting here is how in the larger rocket industry, specially small launch all the competitors talk about how they need to make rocket production like car production. Lowering the price or production by mass production. SpaceX the most successful company doing by far the most launches, is producing about 1 new Falcon 9 Upper Stage about 2x a week at most. A successful car is produced 5000x a week, maybe even 10000x. That gives you real cost savings, but 1-2 is really just serial production. That does give some savings but nothing like automotive. So any small rocket company that talks about improving launch cost by mass production and talking about 'using methods from the automotive industry' is just delusional. Even if the were spectacularly successful and the was actual demand (there isn't) and launching as often as SpaceX. The cost saving from mass production would still be really quite small. Going from the price of your first rocket to say 100 every year saves money, but nowhere near as much as reuse does. RocketLab the only company that actually launches small sat regularly was absolutely convinced mass production was the way to go and then the CEO literally eat a hat and pivoted to making the rocket reusable instead. But somehow the we gone make small launch cheap with mass production story still convinces some investors. Edit: Prat of why I mention this is just recently I was listening to interview with the CEO of Augsburg Rocket Factory from Germany and they made that argument again.

Gareth321 3y ago
I'm not sure I agree. Mass production scales exponentially, sure, but that doesn't mean one producing two Falcon 9 Upper Stages per week isn't experiencing exponentially more efficiencies than one producing a competing rocket at a rate of one per month. No matter how you cut it, more production is typically better for cost efficiencies.
- panick21_ 3y ago
  Yes its better, but my point is your not gone have cheap rocket launches because you produce 2 rockets a week. Your not gone beat a company that has rocket reuse. And you sure as shit wont compete with a larger reusable rocket.
  - bryanlarsen 3y ago
    Reusability isn't a panacea. Refurbishment is hand bespoke labor and its costs can add up quickly. The obvious example is the space shuttle, which cost $1B to refurbish. Elon has said that it costs 10% of new cost to refurbish a first stage. It's believed that a first stage costs $20M to make, so that puts refurb at about $2M. It's plausible that the first stage of a medium-class rocket can be built for less than $2M. Just barely plausible, but plausible. You're not going to do it if you pay $16M for a pair of BE-4 engines, but if you can get two Raptors for $250M apiece... The elephant in the room is Starship. Designed for reusability from the beginning, it's quite possible it pushes that $2M figure down besides reusing the second stage as well as the first.
    - panick21_ 3y ago
      You can potentially build a small vehicle for the cost of a reused Falcon 9, but you don't compete on the rocket market, you compete on the payload to orbit market. > but if you can get two Raptors for $250M apiece... But if your competitor builds the same rocket with 3 Raptors that is reusable your still gone lose.
      - hirundo 3y ago
        They're aiming for $250K per Raptor, not M. The goal is less than $1K/ton of thrust.
        panick21_ 3y ago
        I know but not sure what that has to do with the discussion on reusability? Sure if you have a $250K per Raptor that is mass produced then you can beat a reusable RS-25 that cost 140M.
      - bryanlarsen 3y ago
        $250K per Raptor. Nasty typo that ruins my argument!
- bumby 3y ago
  It’s rarely discussed but a disproportionate amount of aerospace cost is quality control. High level QC doesn’t tend to scale as well as mass production.
  - bhouston 3y ago
    I think as you do more pieces you can invest in more automated ways of doing QC as well. As the increased fixed cost for the QC infrastructure is amortized across the larger number of produced units.
    - TeMPOraL 3y ago
      It all adds up. Scaling up production requires simplifying both product design and the entire process, minimizing the amount of complex manual work and bespoke components. That reduces QA/QC workload as well.
      - bumby 3y ago
        A lot of the QA cost is with pedigree and traceability and that doesn’t scale nearly as well. A $1k bolt isn’t expensive because it’s a bespoke design; it’s expensive because you have to maintain tight control from everything from material coupons to storage. This is one area that makes a lot of manufacturers balk at the selling to aerospace companies; the amount of QC work just doesn't make sense for the amount of product they're making. Another area is supplier control, which SpaceX has struggled with in the past. That doesn't scale well either, but maybe they can bring that cost down by vertical integration of manufacturing (although that doesn't really help as much for the pedigree cost).
        TeMPOraL 3y ago
        Yes, but I believe there are still synergies here. A $1k bolt may be expensive because of QA, but if you're making/buying enough of them, it could make sense to change the QA process to one that has higher false positive rate, but is cheaper to perform. Even if the change makes you throw every other bolt out, most of them perfectly good by the old process, if the new process means dropping per-item-cost from $1k to $250, it's very much worth it. In a similar way, increased scale also makes it reasonable to invest up front to redesign the final product so it doesn't need any of those $1k bolts in the first place.
- rbanffy 3y ago
  There is also a point where you simply run out of either payloads or launchpads and processing facilities. While it's arguable that running out of payloads isn't at all bad, as it opens the market for more competitive pricing, running out of launchpads and facilities to use requires sizeable investment that, in turn, requires higher cadence to be profitable.
onethought 3y ago
Mass production techniques can still apply to low production volumes. Controlling production, vertical integration, driving down cost of manufacturing. Sure if you look at the whole rocket you can say "only 2 per week" but zoom in a little bit, Falcon 9s have 9 engines, Starship has what 33? It's still not car volume levels of course, but SpaceX are dreaming of a day where there are 1000 starships making a mars trip at once. 33,000+ raptor engines for a single transfer window. I know this is pie in the sky at this point, but it kind of makes sense.
- panick21_ 3y ago
  I think you misunderstand your point. I'm not saying it bad to have mass production. Its clearly a benefit and increasingly so as you scale up. I'm saying that if you are a small rocket maker who argues to investors 'our launches are gone be cheap because we will have mass production' is not really credible. And many current small launch vehicle startups still argue this regularly.
  - whoknowswhat11 3y ago
    The other issue is these companies aren’t really beating spacex with mass production- rocket engines being a good example - spacex has manufactured well over 700 Merlin engines. So when these companies claim they are going to undercut spacex dramatically with ‘mass production’ (10x spacex) they never get close to that and many struggle with basic reliability
  - LeifCarrotson 3y ago
    As a controls engineer who does a lot of work for the automotive industry, I think you might be missing the long tail of production here. Sure, a few current flagships have production rates of 5,000 per week, with takt times of 45 seconds or less, and really impressive rolling production lines that won't stop for years, which feature a screaming-fast robot at each corner of the car that get featured in documentaries. But while those lines turn out an impressive amount of product, they're a small part of the whole industry. For every billion-dollar line at Ford building one black Model T (or F-150) a minute, there's a few dozen machines at tier 2 and tier 3 suppliers that might be building, say, a brown leather center console trim panel for the special color option for a less popular car. They might run 3 shifts per week while it's in production, and one a month for the next 10 years building service parts for mechanics. But they're still built with mass-production manufacturing techniques: The parts go in a nest sized just for them cut by CNCs, the right amount of glue is applied evenly by automated sprayers, the adhesive is wet out by an automated press that records the force and position to a production database, the completed assembly undergoes 100% inspection in an automated poka yoke station. In contrast, that brown leather trim panel assembly machine is a one-off, hand-built by craftsmen. The nest might be cut by a CNC, but it goes on a rectangular sub-base with a hole pattern cut on a Bridgeport and held in a generic machinist's vise. The weldment that forms the table was cut, welded, and powder-coated by hand. The dowels that locate the nest were pressed using a manual hydraulic press, or maybe using a hammer, into holes reamed by hand. It underwent weeks of debug and run-off, with a handful of last-minute changes, until eventually the manufacturer told the integrator they were happy. Small launch vehicle startups that argue they'll save money by using mass production are not suggesting they'll have TAKT times of 20 seconds per rocket, 3 shifts per day 7 days a week. They're merely saying they won't have individual master craftsmen fitting parts together by hand. There's a lot of room between the two extremes.

Gareth321 3y ago

I'm not sure I agree. Mass production scales exponentially, sure, but that doesn't mean one producing two Falcon 9 Upper Stages per week isn't experiencing exponentially more efficiencies than one producing a competing rocket at a rate of one per month. No matter how you cut it, more production is typically better for cost efficiencies.

panick21_ 3y ago
Yes its better, but my point is your not gone have cheap rocket launches because you produce 2 rockets a week. Your not gone beat a company that has rocket reuse. And you sure as shit wont compete with a larger reusable rocket.
- bryanlarsen 3y ago
  Reusability isn't a panacea. Refurbishment is hand bespoke labor and its costs can add up quickly. The obvious example is the space shuttle, which cost $1B to refurbish. Elon has said that it costs 10% of new cost to refurbish a first stage. It's believed that a first stage costs $20M to make, so that puts refurb at about $2M. It's plausible that the first stage of a medium-class rocket can be built for less than $2M. Just barely plausible, but plausible. You're not going to do it if you pay $16M for a pair of BE-4 engines, but if you can get two Raptors for $250M apiece... The elephant in the room is Starship. Designed for reusability from the beginning, it's quite possible it pushes that $2M figure down besides reusing the second stage as well as the first.
  - panick21_ 3y ago
    You can potentially build a small vehicle for the cost of a reused Falcon 9, but you don't compete on the rocket market, you compete on the payload to orbit market. > but if you can get two Raptors for $250M apiece... But if your competitor builds the same rocket with 3 Raptors that is reusable your still gone lose.
    - hirundo 3y ago
      They're aiming for $250K per Raptor, not M. The goal is less than $1K/ton of thrust.
      - panick21_ 3y ago
        I know but not sure what that has to do with the discussion on reusability? Sure if you have a $250K per Raptor that is mass produced then you can beat a reusable RS-25 that cost 140M.
    - bryanlarsen 3y ago
      $250K per Raptor. Nasty typo that ruins my argument!
bumby 3y ago
It’s rarely discussed but a disproportionate amount of aerospace cost is quality control. High level QC doesn’t tend to scale as well as mass production.
- bhouston 3y ago
  I think as you do more pieces you can invest in more automated ways of doing QC as well. As the increased fixed cost for the QC infrastructure is amortized across the larger number of produced units.
  - TeMPOraL 3y ago
    It all adds up. Scaling up production requires simplifying both product design and the entire process, minimizing the amount of complex manual work and bespoke components. That reduces QA/QC workload as well.
    - bumby 3y ago
      A lot of the QA cost is with pedigree and traceability and that doesn’t scale nearly as well. A $1k bolt isn’t expensive because it’s a bespoke design; it’s expensive because you have to maintain tight control from everything from material coupons to storage. This is one area that makes a lot of manufacturers balk at the selling to aerospace companies; the amount of QC work just doesn't make sense for the amount of product they're making. Another area is supplier control, which SpaceX has struggled with in the past. That doesn't scale well either, but maybe they can bring that cost down by vertical integration of manufacturing (although that doesn't really help as much for the pedigree cost).
      - TeMPOraL 3y ago
        Yes, but I believe there are still synergies here. A $1k bolt may be expensive because of QA, but if you're making/buying enough of them, it could make sense to change the QA process to one that has higher false positive rate, but is cheaper to perform. Even if the change makes you throw every other bolt out, most of them perfectly good by the old process, if the new process means dropping per-item-cost from $1k to $250, it's very much worth it. In a similar way, increased scale also makes it reasonable to invest up front to redesign the final product so it doesn't need any of those $1k bolts in the first place.
rbanffy 3y ago
There is also a point where you simply run out of either payloads or launchpads and processing facilities. While it's arguable that running out of payloads isn't at all bad, as it opens the market for more competitive pricing, running out of launchpads and facilities to use requires sizeable investment that, in turn, requires higher cadence to be profitable.

onethought 3y ago

Mass production techniques can still apply to low production volumes. Controlling production, vertical integration, driving down cost of manufacturing. Sure if you look at the whole rocket you can say "only 2 per week" but zoom in a little bit, Falcon 9s have 9 engines, Starship has what 33? It's still not car volume levels of course, but SpaceX are dreaming of a day where there are 1000 starships making a mars trip at once. 33,000+ raptor engines for a single transfer window. I know this is pie in the sky at this point, but it kind of makes sense.

panick21_ 3y ago
I think you misunderstand your point. I'm not saying it bad to have mass production. Its clearly a benefit and increasingly so as you scale up. I'm saying that if you are a small rocket maker who argues to investors 'our launches are gone be cheap because we will have mass production' is not really credible. And many current small launch vehicle startups still argue this regularly.
- whoknowswhat11 3y ago
  The other issue is these companies aren’t really beating spacex with mass production- rocket engines being a good example - spacex has manufactured well over 700 Merlin engines. So when these companies claim they are going to undercut spacex dramatically with ‘mass production’ (10x spacex) they never get close to that and many struggle with basic reliability
- LeifCarrotson 3y ago
  As a controls engineer who does a lot of work for the automotive industry, I think you might be missing the long tail of production here. Sure, a few current flagships have production rates of 5,000 per week, with takt times of 45 seconds or less, and really impressive rolling production lines that won't stop for years, which feature a screaming-fast robot at each corner of the car that get featured in documentaries. But while those lines turn out an impressive amount of product, they're a small part of the whole industry. For every billion-dollar line at Ford building one black Model T (or F-150) a minute, there's a few dozen machines at tier 2 and tier 3 suppliers that might be building, say, a brown leather center console trim panel for the special color option for a less popular car. They might run 3 shifts per week while it's in production, and one a month for the next 10 years building service parts for mechanics. But they're still built with mass-production manufacturing techniques: The parts go in a nest sized just for them cut by CNCs, the right amount of glue is applied evenly by automated sprayers, the adhesive is wet out by an automated press that records the force and position to a production database, the completed assembly undergoes 100% inspection in an automated poka yoke station. In contrast, that brown leather trim panel assembly machine is a one-off, hand-built by craftsmen. The nest might be cut by a CNC, but it goes on a rectangular sub-base with a hole pattern cut on a Bridgeport and held in a generic machinist's vise. The weldment that forms the table was cut, welded, and powder-coated by hand. The dowels that locate the nest were pressed using a manual hydraulic press, or maybe using a hammer, into holes reamed by hand. It underwent weeks of debug and run-off, with a handful of last-minute changes, until eventually the manufacturer told the integrator they were happy. Small launch vehicle startups that argue they'll save money by using mass production are not suggesting they'll have TAKT times of 20 seconds per rocket, 3 shifts per day 7 days a week. They're merely saying they won't have individual master craftsmen fitting parts together by hand. There's a lot of room between the two extremes.