Spacex BFR to be lower cost than Falcon 1 at $7 million per launch

by

In Elon Musk’s video announcement of the Spacex BFR he indicated that it would be lower cost to launch than the Spacex Falcon 1.

Above was the graphic which showed the Spacex BFR at lower cost than the Falcon 1.

In 2005 Falcon 1 was advertised as costing $5.9 million ($7.3 million when adjusted for inflation in 2015). In 2006 until 2007 the quoted price of the rocket when operational was $6.7 million. In late 2009 SpaceX announced new prices for the Falcon 1 and 1e at $7 million and $8.5 million respectively, with small discounts available for multi-launch contracts.

This would mean at $7 million the Spacex BFR launch 150 tons would have less than a $50 per kilogram launch cost ($23 per pound).

The Spacex BFR only has two large pieces. It has a first stage and the top ship. Both parts will be re-used. Spacex has shown it can land the first stage. Landing the the large top section of the BFR safely is in the design with two redundant rockets to ensure highly reliable landing.

I believe Spacex will succeed with full reusability and with low maintenance and recovery costs with the BFR.

I had extended some cost analysis assuming complete reusability and reuse life of 100 times or 1000 times and per unit costs of $200 or $500 million.

Spacex should be able to profitably get into the cost per flight of $20-40 million.

If Spacex is able to get the volume of customers and the flight operational ability to fly 50 times per year for each BFR then Spacex could have a rocket payback within one year while charging less than $10 million. There are many unknowns about costs to test and refurbish. But a rocket designed for full reuse could conceivably have lower recovery, refurbishment, testing and flight operations.

Spacex could lower the margin on each launch if there is elasticity in launch where they create a larger launch market with lower prices. Spacex will be able to adjust their prices based upon the demand they see at lower prices.

Even with somewhat lower reusability and more costs on the maintenance, Spacex should be able to comfortably get launch costs for a BFR to $30-40 million which would be $200-300 per pound.

Elon also showed a graphic extrapolating this years Spacex launches to have 20 launches in 2017 and 30 launches in 2018. Most of those launches would be Falcon 9 launches and some would reuse the first stage and some would be Falcon Heavy launches.

This would mean that Spacex would make about 300 rocket engines in 2018.

If Spacex has about the same production rate for the comparably Merlin engine sized Raptor engine, then Spacex would be able to make about one third to one quarter the number of Spacex BFR’s as Spacex makes Falcon 9s.

Spacex engine production rate and some other projections

2017  200 engines
2018  300 engines
2019  450 engines
2020  450 engines (pause in production increase with major raptor shift), Elon launches part of internet satellite network 
2021  450 engines (pause in production increase with major raptor shift), 12 BFRs
2022  600 engines (more production increases again), 15 BFR, most of internet satellites launched
2023  800 engines, 20 BFR
2024  1000 engines, 25 BFR
2025  1300 engines, 30 BFR

By 2025, there could be a fleet of 100 BFR. Each could be flying 10-50 times per year if there the market for launches can be grown with $40-200 per pound launch costs.

Major moon and orbital bases in the 2020-2022 timeframe would be trivial with this kind of space capability.

32 thoughts on “Spacex BFR to be lower cost than Falcon 1 at $7 million per launch”

  1. “Maybe space-based manufacturing like perfect ball bearings.”

    I’d be shocked if they could make significantly better ball bearings in zero G. Most people have no idea just how close to perfect ball bearings already are.

    • @ Brett: “I’d be shocked if they could make significantly better ball bearings in zero G. Most people have no idea just how close to perfect ball bearings already are.”

      Not just that but also it would not be worth the extra cost just for a .1% improvement on the bearing’s smoothness.

  2. I think it will be lower than 7million per launch, at least the cost to spacex will be far lower. The market will appear when launch cost are that low. There will be dozens of companies from huge to smallest trying to do stuff in space. Also, Spacex will totally corner the space tourism market. Instead of doing suborbital flights like Virgin Galactic for 200K a ticket for 5 passengers Spacex will be able to take 400 passenger to orbit the planet for 20K a ticket or even take them around the moon (one week trip) for a 100K a ticket food and drink included!

  3. The question is what kind of business could potentially use the capacity. Maybe space-based manufacturing like perfect ball bearings.

  4. The $7M price for a BFR sounds low. SpaceX has 6000 employees right now, according to Google. That’s probably $1B in annual labor cost, including overhead. At $7M per flight, the revenue from 150 flights a year would just cover labor cost, assuming they don’t hire more people. To get $2B in revenue, they’d have to raise prices to $14M per flight, or fly 300 times a year. Is the market big enough for those launch rates?

    • “The $7M price for a BFR sounds low. SpaceX has 6000 employees right now, according to Google. That’s probably $1B in annual labor cost, including overhead. At $7M per flight, the revenue from 150 flights a year would just cover labor cost, assuming they don’t hire more people.”

      Great comment Randy. This should have been on the article!
      Now the question becomes is there a market for 20,000 tons in space?

      I think no!
      My guess is, provided Musk manages to build the BFR, they won’t launch more than 12 per year because there wont be a need for more launches. Who will be the customer? Satellites have been getting smaller, not bigger. One BFR will be enough to launch a constellation of satellites for any purpose.
      However even 12 launches per year at $100M per launch is better than the competition (provided they don’t start offering something similar).

  9. So they finally figured out what went wrong with the Shuttle, it’s not trying to man rate something that doesn’t need to be manned, and it’s not the lack of understanding of technical limitations, the real reason it failed was because they didn’t build 300 of home

    Think about it, if they built 300 Shuttles instead of 5, they could have flown them 5-10 times a day, launch thousands of times a year, and in the process lowering launch cost to that of flying a passenger jet.

    Sure, you may ask how do they pay for 300 shuttles, or where do you find thousands of payloads a year, or what do you do if something like an O-ring fails, but that would only show your lack of vision: if there are 300 shuttles and thousands of launches a year, spaceflight will be so cheap people will choose to live in space in stations like 2001 A Space Odyssey, commute to work in the city on the moon, and make regular visits to Mars

    And that vision is what will pay for the 300 shuttles.

    • Ah, no, what was wrong with the shuttle was that it had unrealistic specifications, and involved a number of compromises which were known at the time to be bad choices.

      For example, Nasa wanted to use titanium in the airframe. This was vetoed on the basis of, IIRC, national security, or some such, and they went with aluminum. But aluminum anneals at such a low temperature relative to titanium, that the requirements for the thermal protection system became much more extreme. Thus the use of those tiles, which had amazing thermal properties, but which were also terrifyingly fragile, leading to the loss of one of the shuttles. (They actually had to pump cold air through the airframe immediately after landing, before the heat of landing soaked through the tiles, or the airframe would have gotten too hot and been ruined.)

      Or those segmented boosters, with the seal that failed killing another shuttle. Built that way only so that they could be manufactured in a critical Congressman’s district; They were originally supposed to be made one piece onsite, but as one piece boosters they were too big to ship across the country.

      Think of the BFR as the Shuttle without the politically imposed mistakes. And, admittedly, more mature technology. But mainly, without the imposed mistakes.

      • “They were originally supposed to be made one piece onsite, but as one piece boosters they were too big to ship across the country.”

        AeroJet, who lost the contract to Thiokol because Senator Jake Garn was owed a political debt by Nixon, had no intention of shipping an 800 ton rocket engine across the country. They demonstrated that you could pit cast a single grain 800 ton motor, store it in a slowly rotating slanted chamber to prevent debonding from the casing, and ship it from their planned casting pit at Homestead AFB in Florida straight up the coast to Port Canaveral. I talked with an AeroJet exec flying cross country once about this. He had been an engineer on the project.

    • Shuttle was the prototype. X33 was supposed to be the more refined version with airline-like operations. You wouldn’t build 300 Falcon9’s instead of 300 BFRs.

    • Even if they had built 300 Shuttles, which could easily have cost half a trillion dollars or more, even allowing for economies of scale, and this at a time when NASA’s entire budget was ~$15 billion/year, that still would not have been enough to achieve the kind of cost reduction we were looking for. The problem was that partially re-usable is little better than non-reusable in cost terms. The kind of effort that had to go into making a new Shuttle Ready to fly would still have been a killer. Each flight would have needed a new External Tank, no way around that. Add to this that after a fairly hard landing in the ocean followed by a soak in salt water, refurbishing the SRBs was almost as expensive as buying a new one. On top of this, the kind of intricate maintenance of the tiles and the fact that you need to essentially do a complete overhaul of the SSMEs every few uses means that airline-like economics would always be a pipe dream for the Shuttle.

      The reason airline economics work the way they do is that it is largely possible to “gas-n-go”, refill the fuel tanks and load a new set of passengers and gt the plane flying in a matter of minutes after it lands. For the BFR, the cost analysis is critically dependent on what seems to me fairly aggressive estimates for low refurbishment costs of the BFR. The system has to be designed with low refurbishment costs in mind to have even a ghost of a chance of achieving that. Musk has a track record of over-promising AND he also has a track record of, eventually, delivering the goods. Based on that, my expectations of BFR is that the initial cost numbers are likely to be higher than projected by a factor of several which, considering the capacity of a BFR, would still be an impressive cut in the per kg cost to orbit. Following that, incremental improvements may well bring the per flight costs down close to the projected costs, which would be a total game-changer.

      • The STS as it ended up was a design by committee compromise to cut the budget. All the earlier proposals were completely reusable with a manned winged Booster along with an orbiter. No solid boosters, no external tank. The unfortunate thing is the system got stuck with that patched together monstrosity for decades without the budget to do anything else.

  11. Even 100 reuses is in the limit of catastrophic failure rate. 1000 reuses seems unrealistic to me.
    They need around 20 times to make the rocket in the same market than obsoleted rockets like F9.
    Very hard but they could perhaps reach that values and getting better from there. If they don’t reach 20 flights reusing per rocket, BFR would be a economic failure.
    But assuming that I’m worng and they really can reach 1000 reuses, that means a lot of tons on Moon, for example. Every payload means a lot of money of development. It seems too much money to put into a space program without a short term wealth return (and probably indirect for another people not directly linked to investors).
    Even 20 journeys per year to the Moon seems too much for me. Only though a repetition way (plain development costs, only copies and copies from the same robots) we could send so much infrastructure really well developed for the Moon (or Mars).

    • I’m with you Zanstel. Seems a lot of dudes here underestimate failure rates and think that hundreds of terajoules of stored energy is going to play nicely with 31 first stage parallel engines (even detuned ones). I fear that our dear Mr. PayPal, whom I respect greatly, is overestimating what he can deliver. Even when his Raptor engine eventually exceeds the SSME reliability benchmark, he has $%#&@ 31 of them in parallel.

      Here is a pretty cool NASA paper that discusses failure rates for a modern lunar mission:

      https://www.nasa.gov/pdf/140639main_ESAS_08.pdf

      Included is a nice figure out showing 5% probability for loss of mission and 1.5% probability for loss of crew.

      • Seems a lot of dudes here underestimate failure rates and think that hundreds of terajoules of stored energy is going to play nicely with 31 first stage parallel engines (even detuned ones).

        I… concur.

        Let’s remember Musk is a great salesman and he tends to over-sell but eventually delivers at least the minimal required parts of his promises; the later is what makes him successful, given people is more willing to believe him when he talks (or boasts).

        In this case, even if reusability doesn’t pan out more than a few cycles of launch, what he is actually getting is attention and investor’s money to make the ships a reality. If people believe in SpaceX and keep buying launchers, they will be able to build this. The more trust and customers, the faster it will be.

        And as the saying says: if you build it, they will come. Once such system exists, even with less favorable costs and reliability than planned, it will be used nonetheless.

        The bane of the space programs is to be so full of brilliant ideas that never go beyond PowerPoint.

  12. Major moon bases and orbital bases wouldn’t be trivial, but they would be much more likely with this kind of capability.
    The shift of focus would then be on the next set of enabling technologies – the habs, the ISRU, power generation, off-world construction, mining and tunneling technologies, etc.
    Who’s going to figure out all of those things? Musk in his Q&A chat said he would leave this to others. But who’s going to do it?

    • Musk in his Q&A chat said he would leave this to others. But who’s going to do it?

      Indeed. Who will take the lead on this? SpaceX can only do so much.

      The problem now seems to be the incredulity of the potential partners developing the hardware, not expecting to have someone ready to deliver such hardware so soon.

      Probably some of them will come forward right after seeing BFR in action.

      • Musk/SpaceX seems to have chosen ISRU as one of the priority challenges. In the Reddit AMA, he mentioned it:
        <>

      • broken quote:
        Q
        Who will design and build the ISRU system for the propellant depot, and how far along is it?
        A
        SpaceX. Design is pretty far along. It’s a key part of the whole system.

    • If it were actually $50 a pound, plenty of people would step forward to do it. You could have kickstarters paying for some of it. Universities would get together on 1km space telescopes. Garage inventors would be testing out concepts.

      That big a change in costs would change more than just costs.

    • There needs to be a focus on profit, not just “neat”. Mining material and energy for Earth use, then specialized manufacturing for Earth use, perhaps tourism. Support science to Mars, etc, but don’t count on it for big profit-the cost is too high!

  13. “two redundant rockets to ensure highly reliable landing” is wrong, Elon said yesterday that 2 landing engines have now been increased to 3, to increase safety and return mass even more.

Comments are closed.