SpaceX Outlines Satellite Internet Plan

SpaceX launched its 12th resupply mission to the International Space Station from NASA’s Kennedy Space Center in Florida at 12:31 p.m. EDT on Monday, Aug. 14, 2017. (Credit: NASA Television)

Statement of
Patricia Cooper
Vice President, Satellite Government Affairs
Space Exploration Technologies Corp. (SpaceX)
Before the
Committee on Commerce, Science and Technology
United States Senate
October 25, 2017

Selected Excerpts

SpaceX’s consumer focus sets it apart from most other proposed NGSO system. SpaceX has designed its system with the primary purpose of providing broadband service directly to end-users, particularly individual households and small businesses. Meeting this distinct direct-to-end-user goal demands far more on-orbit capacity, which in turn drives the larger number of satellites in the design and the focus on spectrum re-use efficiency.

Initially, the SpaceX system will consist of 4,425 satellites operating in 83 orbital planes (at altitudes ranging from 1,110 km to 1,325 km). This system will also require associated ground control facilities, gateway earth stations, and end user earth stations.

Using Ka- and Ku-Band spectrum, the initial system is designed to provide a wide range of broadband and communications services for residential, commercial, institutional, governmental, and professional users worldwide. SpaceX has separately filed for authority to operate in the V-Band, where we have proposed an additional constellation of 7,500 satellites even closer to Earth, our Very Low Earth Orbit, or “VLEO,” system. In the future, these satellites will provide additional broadband capacity to the SpaceX system and further reduce latency where populations are heavily concentrated.

To implement the system, SpaceX will utilize powerful computing and software capabilities, which will enable SpaceX to allocate broadband resources in real time, placing capacity where it is most needed and directing energy away from areas where it might cause interference to other systems, either in space or on the ground. Because the satellites will beam directly to gateways or user terminals, the infrastructure needed on the ground—particularly in rural or remote areas—is substantially reduced, essentially addressing the “last mile” challenge and helping to close the digital divide. In other words, the common challenges associated with siting, digging trenches, laying fiber, and dealing with property rights are materially alleviated through a space-based broadband network.

SpaceX intends to continually iterate and improve the technology in the system, something that our satellite manufacturing cost profile and in-house launch capability uniquely enables. The ability to modify service as necessary, as well as refresh the technology of the satellite system through iterative spacecraft design changes and phased, continuous deployment, is critical to meet rapidly changing customer demands and responsibly utilize spectrum. This approach will ensure that the system remains adaptable to existing and future customer demands.

For the end consumer, SpaceX user terminals—essentially, a small flat panel, roughly the size of a laptop—will use similar phased array technologies to allow for highly directive, steered antenna beams that track the system’s low-Earth orbit satellites. In space, the satellites will communicate with each other using optical inter-satellite links, in effect creating a “mesh network” flying overhead that will enable seamless network management and continuity of service. The inter-satellite links will further help SpaceX comply with national and international rules associated with spectrum sharing, which distinguishes our system from some of the other proposed NGSO constellations.

Overall, SpaceX has designed our system to achieve the following key objectives:

(1) Capacity. By combining the umbrella coverage of the LEO Constellation with the more intensive coverage from the VLEO Constellation, the SpaceX System will be able to provide high volume broadband capacity over a wide area. SpaceX will periodically improve the satellites over the course of the multi-year deployment of the system, which may further increase capacity.

(2) Adaptability. The system leverages phased array technology to steer dynamically a large pool of beams to focus capacity where it is needed. As noted, optical inter-satellite links will permit flexible routing of traffic on-orbit. Further, the constellation ensures that a variety of frequencies can be reused effectively across different satellites to enhance the flexibility, capacity and robustness of the overall system.

(3) Broadband Services. The system will be able to provide broadband service at fiber-like speeds, the system’s use of low-Earth orbits will allow it to target latencies comparable to terrestrial alternatives. SpaceX intends to market different packages of data at different price points, accommodating a variety of consumer demands.

(4) Efficiency. SpaceX is designing the system from the ground up with cost-effectiveness and reliability in mind, from the design and manufacturing of the space and ground-based elements, to the launch and deployment of the system using SpaceX launch services, development of the user terminals, and end-user subscription rates.

SpaceX soon will begin the process of testing the satellites themselves, with the first two prototypes launching within the next several months. Following the successful demonstration of our space and ground technology, SpaceX intends to begin the operational satellite launch campaign in 2019. The remaining satellites in the constellation will be launched in phases through 2024, when the system will reach full capacity with the Ka- and Ku-Band satellites.

The constellation will be operational well in advance of full deployment, and we expect to begin offering services commercially as early as deployment of 800 satellites. SpaceX is highly experienced with cutting-edge debris mitigation practices and has deep ties with the domestic and international institutions tasked with ensuring the continued safety of space operations.

This year, SpaceX proved out this concept with the successful launch and landing of three flight-proven Falcon 9 boosters, placing high-value telecommunications satellites into orbit for commercial satellite operators. Each Falcon 9 first stage will soon be capable of at least 10 flights with no refurbishment and many more flights after minimal refurbishment, resulting in significant cost reductions.

Dramatically lower launch costs and the demonstrated capability to launch nearly every two weeks (or less) allows SpaceX affordably to deploy larger numbers of satellites for its own NGSO constellation at a pace not previously possible. Moreover, affordable access to space also allows SpaceX to refresh the constellation technology over time, driving down the cost of producing each satellite and making it easier to add capability to meet consumer demand and dynamically react to an evolving market.

  • Vladislaw

    How many launches will be needed to put up the first 800 sats? Anyone have a guess?

  • Michael Vaicaitis

    “Each satellite will weigh approximately 850 pounds and will be the size of a small car. SpaceX plans to launch these satellites aboard its Falcon 9 and upcoming Falcon Heavy rockets, with up to 50 satellites per rocket. This means that the initial deployment of 800 satellites would require a minimum of 16 launches to complete, and all 4,425 satellites would take at least 90 launches.”
    Somebody has been doing mass per launch calcs rather than volume per launch.

    “Each satellite SpaceX proposes to put into orbit, without its solar panels extended, is the size of an average car, measuring 4m by 1.8 by 1.2m and weighing 386kg.”
    Not sure if these are the dimensions of the satellites or this journalists idea of the “size of an average car”.

    The “size of a car” comparison seems common, and the mass of 386kg rings true too – though will that include propellents? – either way the mass and the dimensions strike me as out of proportion. Best guess for now is 10-20 sats per launch, so I’ll take the average of fifteen, which is over 50 launches. Another guess of around $15 million per launch and $500,000 per satellite gives $400 million in hardware and $800 million in launch costs for the first 800 satellites before they are able to start getting paid for the service. But then you did ask for a guess.

  • Sorry, I don’t know. But I’d like to point out that the question could be refined to, “How many reusable boosters and how many upper stages would be needed to launch the first 800 sats?” If, for example, the satellites were small enough that 11 of them could be launched at a time like they did with ORBCOMM and if they have, say 18 first stages that have been recovered, then it would require only four relaunches of each booster but 73 new upper stages to put up 800 satellites.

  • Andrew Tubbiolo

    Would love to watch the maint-and-regen crew learn their ropes with this oversupply of boosters. Man, they’re just falling out of the sky! The balance between production, maint/regen, and sales is a real frontier of engineering, operations, and salesmanship. There’s going to be some great books written about what’s going on right now.

  • windbourne

    I have to wonder if they will create a larger fairing with more volume?
    Otherwise, it would be fewer sats /FH launch.

  • duheagle

    That may well depend upon whatever the deliverables of that USAF Raptor contract turn out to be.

  • Michael Vaicaitis

    Why would they ever use FH to lift 15-20 386 kg satellites? – so less than 10 tonnes. They’ll be using a Block 5 F9 (over and over again) with the first stage having plenty of fuel to return to a land touchdown. Perhaps the majority of the constellation will end up going by BFR, if they’re lucky.

    Also, the larger fairing is called BFR.

  • Michael Vaicaitis

    see my reply to windbourne – the deliverables for that usaf contract will be Raptor – full speed ahead to BFR.

  • Michael Halpern

    What oversupply, once they stockpile enough block 5s production will shift to BFR

  • Andrew Tubbiolo

    I wonder if that will really happen, and if it does, I question how wise such a move is. Look how much longer it has taken Falcon 9 to mature to this stage, and how late Falcon Heavy is. Space X’s record on meeting schedule would not make me feel safe cutting production of Falcon 9 and running off inventory. Given past history, I’m ready for BFR to take a decade to be made operational, I think it will be 5 or 6, but I’m ready for a decade. And at those rates I still think they’d be doing a great job.

  • Michael Halpern

    I doubt it will be a total cut, until they are certain they don’t need any more 1st stages, most think BFR is late 2020s earliest, basically by SLS mission 3 – 6 if it gets to 6

  • Andrew Tubbiolo

    Well that’s the crux, will it be a total cut? Elon said they’d build BFR at the current factory and as such it would be a total cut. Shotwell said the next week that there would be a new factory at the waterfront. That would allow a trickle of production to continue which is all that’s needed to bridge the extended gap (I think will happen) between Falcon and BFR.

  • Lee

    Given some assumptions, it’s pretty easy to figure out the approximate number of launches:

    If you want the most cost-effective sat constellation, you’re going to want to launch into the plane the sat needs to be in. You’re not going to add a bunch of thrusters to every sat, driving up the cost of the entire constellation, just so you can do a one-time plane change. This is why the constellation won’t be launched en-masse by BFR (especially since BFR won’t be ready by the time SX wants to start launching).

    4425 sats / 83 planes gives about 53.3 sats/plane. For argument’s sake, let’s round that to 54 sats/plane.

    Divide 54 by 3 and you get 18 sats/launch, a number we feel could fit into the current payload shroud.

    Thus, there should be 3 launches per plane (roughly) for a total of 3 x 83 or 249 launches. Assuming 2 launches/week, this comes out to around 2.5 years to launch the initial constellation.

    How they get partial service with 800 sats is a little more speculative, but assuming they launch 18 sats into around half the planes, you get close to the “800” number that was mentioned.

  • duheagle

    Given that BFR production will be stood up in a new facility, there’s no major conflict between Falcon production and BFR production. New Falcons can be cranked out for as long as they’re needed.

  • duheagle

    Going from Falcon 1 to Falcon 9 was as big a jump as going from F9 to BFR will be. SpaceX did that in four years – 2006 to 2010. To hit Elon’s 2022 unmanned Mars departure date, BFR has to fly by 2021 – four years hence. Personally, I think 2020 is eminently possible given that SpaceX is now both much larger and more savvy an organization than it was in 2006.

    Too many attach too much freight to the tardy arrival of Falcon Heavy. But Falcon Heavy pretty quickly proved to be something off to the side of SpaceX’s actual critical path shortly after its announcement. Falcon 1, Falcon 9 and, now, BFR are SpaceX’s critical path. This “it’ll take a decade” nonsense about BFR is, frankly, incomprehensible to me.

  • duheagle

    “Most?” It would be interesting to see an actual show of hands on BFR’s projected debut. I sure don’t think it’s going to be “late 2020’s earliest.” And I also think I’ve got company on that. I think the odds are no worse than even money that BFR will fly before even SLS Block 1.

  • duheagle

    The BFR plan is evolving very quickly. That’s maybe the biggest reason I think all this talk about BFR taking a decade to get up and running is just bonkers. The new factory plan introduces new items into the critical path, but not new kinds of items. It solves some potential problems as well – no potentially tricky timing in trade of Falcon for BFR production being the main one. Overall, a solid net plus in my view. The F1-to-F9 transition took four years and also involved the new vehicle being built in a new factory. I think, as usual, SpaceX will improve on its past performance.

  • Andrew Tubbiolo

    My bet is still for a decade. Call me bonkers. I’ll call myself wrong if they start hopping a large rocket structure made of wound carbon composite within the next three years. Yes they are bigger and better than 11 years ago, but they’ve been sitting on the shoulders 20 year old aluminum-lithium technology perfected in the 1990’s, and engines rooted in the same decade. Look I hope I’m wrong, and I was wrong about them losing another booster due to pushing the launch rate this year (so far). So I hope I’m wrong about this one too. But for now I’ll stick to my guns, fully ready to recognize my wrongness should history show me to be so.

  • duheagle

    I still suspect USAF wants some kind of Raptor-powered larger 2nd stage for at least FH. SpaceX will be happy to do this even if it modestly impacts BFR because SpaceX wants to keep USAF happy. USAF is the best friend SpaceX has in the U.S. gov’t. at the moment. USAF wants to do space stuff quickly and cheaply to fend off Rep. Rogers’s Space Corps idea. SpaceX is the available and willing means to that end. USAF really has no conflicts anymore where SpaceX is concerned. The X-37B mission pretty well demonstrated that. And USAF has that new “advanced EELV” program ginning up that SpaceX can probably get a nice slice of cash from to put toward remaining BFR development.

    Contrast that with SpaceX’s still problematical relationship with NASA. Half of NASA loves SpaceX and the other half is trying, at every opportunity, to kick the company in the nads. The SpaceX-USAF relationship started out more than a bit rockily back in 2014, but now it’s all sweetness and light. What USAF wants, USAF will get.

  • Michael Vaicaitis

    For the most part, i like your thinking. With regards to BFR, how is it that you alone are privy to the precise date that it will not be available by?. Of course, I understand what you mean, but then BFR “could” (aspirationally) be ready to go to Mars in 2022, so it could be in limited service in 2020, in time to help with the lifting some of those 4425 M/LEO sats. Lifting such a relatively trivial mass of these sats could leave BFR with sufficient delta-v to deliver to more than one plane/altitude in one trip. Also, using BFR could mean 1 or 2 trips per plane as opposed to your suggested 3. In any case, using BFR would be cheaper than using F9. And finally, any work for BFR prior to Mars would be good operational experience/testing.

  • Michael Vaicaitis

    I agree with your analysis of SpaceX’s blossoming relationship with USAF, but I am curious…
    “I still suspect USAF wants some kind of Raptor-powered larger 2nd stage for at least FH.” …what is it that has fuelled your suspicion in this regard. The initial contract was reported as working toward a new Falcon second stage, but that was back in the good old days of ITS. The imminent arrival of BFR as a be all do all commercial workhorse changes the landscape considerably. USAF’s “heavy” launch plans are well catered for over the next few years by their already contracted for commitments to Delta-IV-H and the welcome arrival of FH. BFR on the other hand provides the dual benefits of vastly increased lift capability, and vastly reduced costs. Why, do you suspect, would the USAF go out of its way to delay BFR for the sake of possibly one or two missions on a new Falcon second stage?. Since their stated policy is to move towards cheaper more agile systems, why would they even need FH with a Raptor upper stage?. What payloads are likely to be thought of, lobbied for, funded and developed, just in time to launch on a FH+Raptor upper stage, but not be able to wait an extra few months for BFR?. And whose to say that a meaningful increase in upper stage+payload mass wouldn’t then mean even more changes to the already heavily modified FH centre core.
    All in all, I think it far more likely that SpaceX have sold USAF on the benefits to be had from BFR, and both organisations are keen to expedite its development into service asap.

  • Kenneth_Brown

    I wouldn’t bank on this venture happening in the near future. While Tesla is a separate company, SX has loaned them a bunch of money and that could create problems if Tesla’s fortunes are depressed for any length of time.

    I’d really love to see the business plan on this. I get gobs of speed from a cable internet connection that costs me less than $50/month with no stated limits (that they’ll publish in bigger than 6pt type). Can SpaceX’s beat what I pay now? How many subscribers will they need to break even on operations? What are the estimates on per account bandwidth? How they put the satellites up and what downlink bands are less important that if the business model for the service is realistic or not.

  • Mr Snarky Answer

    “Elon said they’d build BFR at the current factory”

    Context matter. Engines are yes built at the current factory, as are avionics. The tanks will not be, and that’s what counts on crowding out F9.

  • Michael Halpern

    Even in best case scenario for BFR, it won’t be before SLS unless SLS is cancelled, they have to make their money for BFR off of F9 block 5 first, block 5 isn’t coming out until early next year, then they want to get the stockpile of boosters they have at that point to near time to refurbish so they can just not refurbish them as they transition to BFR, they probably can push it out in6 years time, the question is if they would really want to

  • ThomasLMatula

    You are thinking of this from the old access model. This is as much a gamer changer as Falcon and Tesla.

    The key is to make the access free and make money on the targeted advertising, content subscriptions and services provided.

    Remember this is a global service so you will be able to reach billions of devices, not just laptops, but anything that needs net access, including Teslas, your smart phones, etc. There are a number of ways to generate revenues when you are networking billions of devices.

    Remember, a couple billion plus are online globally. Say 6 billion devices. If it costs you $12 billion to deploy the system and operate it for 2 years you only need to generate a $1.00 a device a year to break even. There are many ways to do so.

    For example, charge smart phone makers, TV makers, etc. a simple $5-10 for a app that allows their devices to connect to this network for free, forever…

    Or those with existing devices $5 for each device for life time access.

    Good by mobile phone providers, cable companies, Internet service providers.

  • Kenneth_Brown

    No thank you in big capital letters. When you are getting something for “free”, YOU are the product. What you describe means that Elon would by trying to become the next Google although he would be capturing everything about somebody on a continuous basis. When you are home, when you aren’t. Where and when you work. Where you shop. How aggressively you drive and what places you visit on a regular basis. Do you get your teeth cleaned every 6 months? How often do you visit your doctor? Do you go to a shooting range?

    It’s already pretty scary what can be inferred with very little data from just one Rewards Card about you. Given access to larger amount of information would be even worse. When you apply for a job there are certain things that you cannot be asked. These days it’s already pretty easy to get a dossier on somebody for a few bucks that would contain all of the information that an employer is forbidden to ask. But, it could get far more personal and employers might just decide that according to your destination records they purchased from Elon, you spend too much time at establishments that serve alcohol or cater to a group with an alternative lifestyle. Or, you attend the “wrong” church or that you don’t attend church.

    Facebook is finally getting called out for a disturbing amount of targeted advertising that could only have come out of voice calls that people didn’t realize would be harvested or on a device with a FB app even though the information to trigger the ads wasn’t from something typed in on FB. I’m not too hopeful, but many people are soft of coming to understand that even if they don’t have secrets, there are things that they would rather have kept private. Having different suppliers for different things isn’t a perfect way to keep private matters private, but it’s a step in the right direction.

  • Kenneth_Brown

    Tesla isn’t a game changer until they can turn a profit. Anybody can develop and sell a product at a loss. Loan me a few billion and I’ll prove it.

  • We briefly discussed this elsewhere, but a quick back-of-napkin makes a compelling case for manufacturing Raptors in Hawthorne starting in late 2019, and rolling to a Raptor upper stage for F9/FH in the same timeframe.


    1) Chances of BFR full deployment by 2025 approach 100%.

    2) 30 F9/FH launches in 2018, with 30% CAGR after that = 720 F9/FH launches in the period 2018-2025.

    3) Expendable core launches drop to nearly zero once FH is up and running.

    4) Block V is good for at least 50 launches.

    5) At least 3 Block V cores will be complete by end of 2017.

    6) At the end of 2014, SpaceX claimed to be producing 4 Merlins a week. Let’s assume that that’s stayed pretty much the same.

    7) F9/FH second stage recovery never happens, at least operationally. (I could see using some S2’s to prove out BFR technology, but that’s an engineering curiosity at best.)

    That would imply that you really only need a fleet of 15 Block V’s. Figure 5 extra FH center cores and 5 spare cores for accidents or maintenance unpleasantness, for a total fleet of 25 cores.

    Current core production rate must be at least 16 cores/year, given the launch manifest for 2017. That means that all block V cores could be complete by mid-2019.

    Now let’s look at Merlin production. 4/week is about 200/year. Figure the 25-core fleet needs 2 spares per core and your total requirement for the core fleet of engines is 275–which also comes out to mid-2019.

    But now we need an extra 720 engines for second stages. That would imply that you’d have to keep the Merlin line open through 2022.

    But if you go to a Raptor upper stage, that doesn’t have to happen. If you do pilot production and testing of Raptors and a RUS in late 2018 or early 2019, by mid-to-late 2019, you could cut the Merlin line over to full production of nothing but Raptors.

    By mid-2019, launch rates will be approaching 50/year. Even if you only produce 2 Raptors a week, you’ve got the F9 cash cow demand handled, and at least 50 extra Raptors to begin the testing campaign.

    This all seems to fit with the aspirational schedule pretty well. I’d guess that, by 2020, Hawthorne will be a Merlin-free zone, except for maintenance and refurbishment ops.

  • Just up-thread, I did a back-of-napkin assuming that SpaceX had to be able to operate the F9/FH architecture through 2025, which is a 3-4 year pad on the aspirational BFR schedule. It’s a decent case for cutting F9/FH over to a Raptor upper stage in late 2019 and then just crankin’ out Raptors instead of Merlins.

  • Shotwell said that they’d committed to building BFR cores in a new facility in Long Beach, so they could go straight onto the ship. But I would guess that Raptors will be built in Hawthorne.

  • My guess is that they’re fairing-limited, not weight-limited. Figure two layers of 6 sats each in the existing fairing, which is 67 launches.

    But! If they cut over to a Raptor upper stage, they might manage a fairing with an interior diameter of 5 m instead of 4.5, and something slightly longer. Now maybe you’re at 3 layers of 8 sats each: 33 launches.

    Split the difference: 50 launches.

  • ThomasLMatula

    True, but they allow for a few folks to slip through the cracks. The vast majority will buy in.

    Look at how many folks use Google as their search engine, have gmail, and Chrome for their browser. And they are also looking to provide Internet access as well 🙂

    And if you own a Tesla Elon Musk already knows where you go, how fast you drive, any stops you run… If you doubt it look at this story.

    Tesla Remotely Expanded Car Batteries Near Irma’s Path, And Questions Linger

  • duheagle

    That might be how it goes. A lot depends upon just how much floor space is devoted to the Merlin final assembly line now. My guess is that, compared to how much is needed for producing stages, not very much. A new, separate line for Raptor final assembly might not take up much room at all. The machine tools to make Raptor parts are probably already all or mostly in place making Merlin parts. Those could do double duty as Merlin production ramps down and Raptor production ramps up.

    I’d like to see the Merlin stay in production for awhile as I have been noodling a military application for it post-Falcon. So long as all the tooling and CAD/CAE/CNC stuff is kept, though, that may not matter too much.

  • duheagle

    F9 may continue to use the current S2, but a new Raptor-based upper stage for FH seems increasingly likely. USAF seems to want it for something. On another thread, Bob Oler suggested that something may be the X-37C. I think he may well be right. If so, the new FH S2 will be bigger and fully reusable.

    With F9 still putting up expendable S2 stages, the Merlin line would have to stay open, but the only way Raptor competes with Merlin for shop floor space is on a final assembly line – which doesn’t have to be very large, really. The same machine tools currently cranking out Merlin parts can make parts for Raptors too. So there’s little or no net floor space increase needed for Raptor in that respect.

    Based on pictures and videos I’ve seen of the inside of SpaceX’s Hawthorne works, I think there are four actual F9 S1/S2 production bays. If SpaceX really badly needs some extra space for BFR component production, it could shut down one of these bays and still have plenty of capacity to make any additional F9 S1’s and S2’s needed once Block 5 has replaced all prior versions of the F9 S1 in service. That will most likely happen well before the end of 2018.

    There’s every reason to expect the cutover from Falcon to BFR to be very smooth.

  • duheagle

    You’re the campus lefty and I’m the cranky conservative. But you seem to think the world is largely unchangeable and governed largely by the narrow interests of stick-in-the-mud incumbents. Whereas I see us heading pell-mell into a future of constant change at an accelerating rate and find that a Good Thing.

  • duheagle

    A large part – possibly even most – of the BFR development engineering has already been done on dimes SpaceX has already made on its own. What the Block 5 F9’s will do is pay for the new infrastructure needed to build the thing. Should this prove insufficient – and I don’t think that will be the case – there’s always the USAF waiting in the wings with advanced EELV money on offer – which Elon will take in any event. USAF likes SpaceX a lot these days. The feeling seems to be mutual. Each learned a lot from the other – and about the other – during that EELV certification process for F9.

    SLS Block 1 is not going to fly until 2020. And maybe not even then. BFR will fly by 2021 at the latest. Personally, I expect it to do so in 2H 2020. I think SLS vs. BFR is going to be a squeaker.

  • duheagle

    The only thing that makes sense as an answer to your question I have to credit to Bob Oler who thinks it’s the X-37C. That hadn’t occurred to me before, but it makes all kinds of sense. X-37C would be big enough to carry a half-dozen crew, could launch either manned or unmanned and was planned for a “naked” launch (no fairing) like the original Dream Chaser atop a high-end Atlas V. Falcon Heavy would be a much cheaper launch vehicle. With a Raptor-powered reusable upper stage it would also considerably expand X-37C’s mission envelope relative to the Atlas V or an unimproved FH. Being a crew-optional vehicle, one can even see how it would appeal to the Fighter Jock Mafia that runs USAF these days. That might explain how the USAF space weenies got money to proceed with it. That and the fact that USAF is under a lot of pressure from folks like Rep. Rogers about its being the Rip Van Winkle of space these last two decades or so. All the pieces just seem to fit together.

  • duheagle

    SpaceX may have a hard time beating $50/month. My service costs over three times that. So if SpaceX can beat my cable company when they get 800 sats up, it’ll be hasta la vista, Baby. I think I’ve got a lot more company than you do.

  • Andrew Tubbiolo

    Actually I was much more a cranky conservative in the past. I was cranky because I believed in concepts of conservatism as being closer to natural law. Those ideas were formed growing up in the Cold War. After watching Iraq and Afghanistan and the neo-cons, and now the descent into Trump, I’m a ‘atheist’ of political systems, (relative to you, yes, I’ve moved left.) and inclined to look at results only. I still in many ways consider myself to be an Eisenhowerist. I don’t discount your thoughts about where we’re going. I do recognize that there are limits to the amount of change that systems of power can undergo in any unit of time. ULA is facing death, and they know it. 2017 saw Falcon operate like Shuttle was promised in the 70’s schedule wise, next year it will be reusability. In 2019 the price will fall thru the floor. They will act, and I think they are acting. Hence my little conspiracy theory I posted, and I thank you for engaging it. I enjoy our teat-a-te’s.

  • redneck

    I tried to work with a group that called itself conservative until it became clear that they were actually statists that wanted themselves in charge. The more I look, the more I see various ideologies as tribal rather than idealistic. If the other side is for it, it must be bad, no need for further discussion. At any given time, there will be a few people that are hard core for one of the ideologies and a majority following something that sounds good to their particular life experience.
    Libertarian comes closest to an ideology I can support, though there are a few things I have a problem with, open borders being the most flagrant.
    To me, Left, right, liberal, conservative, Republican, and Democrat have become blurred almost out of usefulness. Many of our problems are caused by the attempts of one to dominate the other rather than efforts to do what is right.

  • Michael Vaicaitis

    A plausible enough theory perhaps, except what is the time scale for this mythical beast?. Do USAF have funding in place?, has development started?, or do you believe this is a top secret project almost ready to go?.

  • Michael Halpern

    I guess I am too used to thinking in terms of how long spaceflight development usually takes, we will see. the industry is prone to setbacks after all.

  • Andrew Tubbiolo

    Tribalism makes sense, we’re a tribal species. It’s where we all started, and where we go when we don’t have anything bigger. Gangs are a perfect example of what I think is ad-hoc tribalism. Most political theories are really schemes to get other people to do the dirty work for you dressed in one form or another.

  • You’re not the target customer. Some customers that are:

    1) People outside of wireline broadband service range.
    2) Developing world customers (overlaps with #1).
    3) Corporate/commercial customers.
    4) IoT / instrumentation / global tracking customers.
    5) Mobile backhaul (relaying stuff from cell towers into the core network).
    6) Internet transit customers, especially ones who need transit away from established fiber plants.
    7) Content distribution network customers (overlaps with #6).

  • I’d be very worried if I were using the tooling for both engines at the same time. Sure, you can invest in a separate set, but that’s a pretty big expense.

    Also, you want to pull a lot of skilled labor from the Merlin program onto the Raptor. Maybe you can do that and simply lower Merlin throughput, but I wouldn’t discount the possibility that there are key personnel that have to be doing one or the other. Until you develop the expertise for both lines, you may have to choose.

    I agree that it’s not your big consumer of floor space, but I doubt that that’s the gating factor.

  • publiusr

    I wonder if you could put a Falcon core with legs atop BFR….
    Land one on the Moon. Fire off another–have that land on the moon a mile from the first one.


  • Hey, Di­ck, this is obviously out-of-the-blue, but I remember we had this X-37C conversation as the driving force behind actually deploying a RUS, and it recently occurred to me that there’s a much more obvious, straightforward reason: EELV2’s requirement for short-cycle to GEO.

    FH has enough delta-v to get 11.5 t to GEO in full expendable mode, but the F9 S2 probably can’t handle the 5.25-hour coast to GTO apogee. I suspect that that’s a pretty strong case for working the kinks out of autogenous pressurization on a RUS–as well as some kind of sustainable power system. They can probably deal with the boil-off over that amount of time, but helium and power will kill ’em dead on a kerolox S2.

    So now we’ve got:

    1) Higher energy for X-37C.
    2) Long coast for GEO insertion.
    3) A nice way to ramp Raptor production early.
    4) And maybe support for a bigger fairing.

    Beyond that, I recently started fooling with a 2-launch crewed mission to LLO using FH (lunar lander sold separately!), and the FH S2 isn’t even close to being able to coast and then provide enough delta-v for TLI. I wound up tossing in a chopped-down S9 S2 as a sorta-3rd-stage TLI module, but it would also require a major upgrade to mission duration. The RUS, or possibly a chopped-down version of it, would solve a lot of problems there, as well.

    I still don’t think that NASA is going to touch BFR/BFS for a crewed mission–not without launch escape. My guess is that SpaceX doesn’t care that much, but having a RUS in their bag of tricks might make them a little more interested in providing crewed capability to LLO some time in the next 10 years. But the EELV2 GEO thing seems extremely plausible.