Main Engine Cut Off - T+256: The Kuiper Logjam

Episode Date: August 25, 2023

Amazon moved their Project Kuiper prototypes from Vulcan to Atlas V. Between that and some recent conversations I’ve had, I thought it would be a good time to check in on Kuiper and to see how they���re progressing towards deployment. I do some math, and it’s not good.This episode of Main Engine Cut Off is brought to you by 34 executive producers—Chris, Donald, Dawn Aerospace, Will, Tim Dodd (the Everyday Astronaut), Benjamin, Steve, Theo and Violet, Pat, Russell, The Astrogators at SEE, Harrison, Tyler, Jan, Bob, Kris, Lars from Agile Space, Pat from KC, Ryan, Lee, Stealth Julian, Warren, Craig from SpaceHappyHour.com, Fred, Matt, Frank, SmallSpark Space Systems, Joel, David, Joonas, and four anonymous—and 830 other supporters.TopicsAmazon moves Project Kuiper prototypes from Vulcan to Atlas 5 - SpaceNewsAmazon signs multibillion-dollar Project Kuiper launch contracts - SpaceNewsArianespace signs unprecedented contract with Amazon for 18 Ariane 6 launches to deploy Project Kuiper constellation - ArianespaceBlue Origin opens rocket engine factory - SpaceNewsBarry Jenakuns on X: “Fun fact, because Bezos own's Blue Origin, Amazon have to declare how much money they spend(/receive) on them. Assuming that this is for the 27 launch option, the $2.7 billion spent would give a per launch cost for New Glenn of $100 million.”Episode 120 - Big Dumb Satellites - Off-NominalThe ShowLike the show? Support the show!Email your thoughts, comments, and questions to [e-mail] @WeHaveMECOFollow @[e-mail] on MastodonListen to MECO HeadlinesListen to Off-NominalJoin the Off-Nominal DiscordSubscribe on Apple Podcasts, Overcast, Pocket Casts, Spotify, Google Play, Stitcher, TuneIn or elsewhereSubscribe to the Main Engine Cut Off NewsletterArtwork photo by ArianespaceWork with me and my design and development agency: Pine Works

Hello and welcome to Main Engine Cutoff, I am Anthony Colangelo, and I've got a ton on my mind that I want to talk about on the show. I've got some stuff on financial stories I want to talk about, some lunar landers that have been going on lately. I'm going to save those two topics for in the near future. I'm trying to schedule a couple of guests to come on and talk with me about those topics. I think it'll make a better show. And in the meantime, I got somewhat down a rabbit hole on Amazon's Project Kuiper. And I haven't been able to stop thinking about this and doing math on this. So

just wanted to break this down with you and see where Amazon's at with this project. It's been a while since I talked about it on the show. So I thought it'd be good to check in, especially given the critical moment that they find themselves in, not only on the launch side, but on the satellite side as well. And really, this is sprung from me having two space lunches this week here. Tim Dodd happened to be in Philadelphia for a day. So we were able to grab some coffee and lunch, got down a bunch of rabbit holes, and part of that was New Glenn-related stuff. The next day, I'd already scheduled lunch with Caleb Henry,

our friend of Quilty Analytics, who's been on the show many times to talk about the satellite industry, finance industry, and we got down this rabbit hole as well, so I kind of just got stuck on it since earlier in the week and has been kicking a lot around my head. So there also has been some recent news. So the recent news is that Amazon decided to move their first two prototype satellites off of the first launch of United Launch Alliance's Vulcan onto the Atlas 5 vehicle that they bought a couple launches on as well. So this is going to be the

first launch that Amazon does for Project Kuiper next month or about one month away from this earliest launch date of an Atlas V carrying two prototype satellites for Project Kuiper. To set the stage a little further, Project Kuiper is Amazon's Starlink competitor, a 3,236 satellite constellation to provide internet globally. Um, they formerly had bought a ton of launch vehicles to launch these satellites. Um, so they bought nine Atlas fives. They bought, uh, now I'm going to forget the exact count, but they bought almost a hundred launch vehicles between Atlas five, New Glenn, uh, Vulcan and Ariane six from Ariane space. They bought almost a 100 launches to get these

satellites up into orbit because there is a really pressing timeline on this whole project. Amazon got approval back in 2020 for 3,236 satellites, and the FCC guidelines for that approval impose a particular timeline. So half of the constellation has to be deployed by July 2026. So that means by three years from right now, they have to have 1,618 satellites up, or they get their numbers cut back in terms of what they're approved to fly. So three years from now, they got to get 1,600 up. And then three years from then, they have to get the other 1,600 up., that that last phase is not the really hard gate. You know, if you're able to get half your constellation up in that first six year time slot that the FCC gives you, you're fairly likely

going to be able to make that second time slot because, you know, the indication of getting up to speed on those launches in the first six years is that you should be able to cover it for the back three, which is why it's structured that way. But given where Amazon's at, this seems like a really tall order. And that's half of what I want to talk about today. But to stick on this recent news for a second, Amazon moving prototype launch vehicles from Vulcan over to Atlas. To shed light on that even more, right? Originally, a couple years back, Amazon planned to fly these first two prototype satellites on a launch vehicle called RS-1 that ABL Space is beginning to fly. ABL launched their first RS-1 vehicle at the beginning of the year, only lasted a couple seconds before they had a failure, have yet to make it to the launch pad for their second attempt,

doesn't seem like they've got the quickest pace in the world. So at this point, Amazon is like, that's no longer useful to us. And it was last year when they moved their satellites off of ABL onto ULA's Vulcan vehicle. They were going to fly in the very first launch of Vulcan, right alongside Astrobotics Peregrine Lander going to the moon. Everything was going to be great because Vulcan was always, you know, a couple months away from flying. Here we are in August 2023, and Vulcan is still a couple of months away from flying. ULA says end of the year, Vulcan might be on the pad. All indications I have is that that's incredibly, incredibly hopeful. I expect it to fly early next year at best. So again, we're still, you know, several months away from this thing hitting the launch pad.

And at this point, Amazon needs to get these two satellites up into the sky. They already have nine Atlas fives on order for use under project Kuiper. So they had decided to switch these two prototype satellites over to an Atlas five. Now these satellites are not big. They're not small, but they're not big. They're 500 kilograms a piece is the best estimate that we have for the size of these satellites. The Atlas 5 configuration that they'll be flying here is called the 501 configuration, a five meter fairing, no solid rocket boosters. So it's kind of the lowest capability that they have on the Atlas versions that they bought. And that can do something like eight tons to low Earth orbit. So, you know, significantly over what they actually need to be able to launch

this to low Earth orbit, these two satellites, but not really the worry, right? Amazon just needs to get these things up and flying. So if you want to say it's overkill, that's cool. But for Amazon and the amount of money that they're spending on Project Kuiper overall, really a drop in the bucket if it's a couple tens of millions different than what they could have flown these things to orbit on a different vehicle. They already had this deal signed. Presumably it would be a little easier to adapt whatever they were already doing for Vulcan over Atlas because there is supposed to be a lot of commonality there from the ULA side. And there is a definite ban on flying on SpaceX vehicles from the Amazon side because of historical bad blood, not only between Bezos and Musk, if you want to go that route, but I think

even a lower level, the people involved on Project Kuiper have had some issues with Musk in the past. Clearly don't want to fly on a SpaceX vehicle because, I mean, if you were trying to get something up to orbit super quick, 500 kilograms, I feel like you probably could have found a spot on a transporter mission in the not-too-distant past, or even in the distant past, and flown a little earlier. But that's, you know, that's a really weird thing to get our head around, and it would be major speculation to figure out exactly why that did not occur. But whatever the case is, I'm not necessarily worried about

the idea of flying two satellites on an Atlas 5 from the launch rocket perspective. You know, like I said, Amazon's spending a ton of money. They always say that they're spending about $10 billion on Project Kuiper, and that, by all accounts, is ridiculously under what they're actually going to be spending on this. There was a tweet last month that showed that that because jeff bezos owns blue origin amazon has to declare how much money goes to and from blue origin um and through that there's some assumption here because it says blue origin and an unnamed third party in this uh that we assume

to mean united launch alliance because uh amazon is sending money to United Launch Alliance for launch services. They're buying Blue Origin BE-4 engines for their vehicle, so it eventually makes their way to Blue Origin. So that money is accounted for in here, yada yada. This disclosure said that between now and 2028, Amazon will be paying $7.4 billion for satellite launch and related services to Blue Origin and this third party. Again, that's only part of the launch cost for Amazon, for Project Kuiper. That includes Blue Origin's New Glenn, United Launch Alliance's Vulcan, but that does not include the Atlas V vehicles that they've bought, the nine of those, and the 18 Ariane 6 vehicles that they bought to fly Project Kuiper satellites to orbit. So just the launch services alone, and again, this math tracks with, even if you go back a

couple episodes of this podcast, where I'm talking about the potential sale of United Launch Alliance, that Amazon is probably spending on the order of $5 billion on ULA, because not only are they buying launches, but they're investing in infrastructure for ULA to be able to up their flight rate of Vulcan to support Project Kuiper. So they were probably already spending $5 billion-ish. That kind of all tracks with what we know. But this, again, this number $7.2 billion between now and 2028 doesn't even include, $7.4 billion, sorry, doesn't even include other launch vehicles. They're probably spending $10 billion on launch and infrastructure alone before you talk about the satellite production side of thing the user

terminal production side of things all of this software that needs to go into this all of the other stuff that needs to be done for project hyper to actually hit market um you know the ground network maybe they've got a little bit of boost on since amazon itself has been involved in the ground side for a little while um but whatever the case is, you know, the 10 billion number that Amazon has given out in the past does not seem at all accurate given what we've seen here. So, you know, for all that said, a couple tens of million on a particular launch is not going to break the overall Kuiper budget by any means. What I find more concerning about shifting these two prototypes over to an Atlas V is that the strategy here of continuing to fly

these two prototype satellites, at this point in late 2023, I have some questions about that. What's the intention? If they launch these in 2022 or even earlier, if it was early 2022, I get it because you're still, you got a bunch of years before, you know, the bulk of your satellites need to go up. So maybe it is worth flying a couple of prototypes, seeing if you can catch some easy errors before you go and do a huge run of serial production. But at this point, you've been sitting around for a little while and it's been evident that you're not going to be able to fly these two prototype satellites on any really helpful timeline um you know have you produced a bunch of satellites already or are you still just waiting on these two prototype satellites to go up before you get into mass production of these satellites

if you have produced a bunch of satellites are you going to go back and fix stuff based on what you find on these two prototype satellites or is is that not worth it? If that's not worth it, can you not ship these satellites? Can you not like wait a couple of months and put up a whole plane of satellites on an Atlas 5 vehicle? You're still going to fly just the two? You know, those are the things that I'm really curious about. And we'll probably almost never uncover this kind of stuff from internal to the company. But I don't know, it just feels really weird to be this late in 2023 and say, we're still going to fly the two prototype satellites on a launch vehicle that, you know, has all the confidence in the world that that thing will work. It's not like, uh, it's, it's

the riskiest thing to even be able to test your entire plane of satellites. So we're then really not in a position to spend 40 satellites or whatever it is, if they were able to fly like the full on Atlas 5 with some solid rocket boosters, they're not in the position to send up a whole batch of satellites. I mean, even on this one with no solid rocket boosters, they could fit more than 10. Certainly more than two. I don't know. Is that is that how I guess maybe at that point, it's not helpful.. I don't know. I guess maybe at that point it's not helpful, but I don't know. Just something feels off about that generally. And it makes me concerned that they don't have a huge production lineup and running for these satellites. The launch side

is more of a log jam. And I have some math to prove that. And we can do some exercises to think through exactly how this will all go down. But before we get to that part, I want to say thank you to all of you out there who's supporting Main Engine Cutoff over at mainenginecutoff.com slash support. There are 864 of you. The numbers are a little bit lower. I will mention Patreon had a bunch of issues at the beginning of August. So if you are, if you think you should be getting headlines, but you aren't seeing it in your podcast player, go check your Patreon account because they had a bunch of issues. Just check back in there.

I know a bunch of you dropped off at that moment, and I think we're still missing a couple of you. So if you hear this and you're not seeing headlines in your feed, go check that out. But anyway, there are 864 of you over there who have either no issues or have checked your Patreon account, and I'm so thankful for your support. That includes 34 executive producers who produced this episode of Main Engine Cutoff. Thanks to Chris, Donald, Dawn Aerospace, Will, Tim Dodd, the Everett Astronaut, and sometimes Lunch Haver, Benjamin, Steve, Theo Violet, Pat, Russell,

The Astrogators at SCE, Harrison, Tyler, Jan, Bob, Chris, Lars from Agile Space, Pat from KC, Ryan, Lee, Stealth, Julian, Warren, Craig from SpaceHappyHour.com, Fred, Matt, Frank, SmallSpark Space Systems, Joel, David, Eunice, and four anonymous executive producers. Thank you all so much for the support as always. This is a 100% listener supported show and I could not do what I do without all of you. If you want to help support the show and get some extra podcasts in your life, I just mentioned a couple times, Headlines. It's an entire other show that I do. Every single week, I run through the stories that are going on in space, keep you up to date on the ones that are important to know about, filter out the ones that you don't need to know about because there's a lot

of stuff that happens these days that just doesn't rise above the line of things that you should be particularly concerned about. So let me read all the space news for you and keep you up to date on just the stuff that you need to know. I think it's a great way to stay up on everything to help support the show and to continue to have Miko in your life. So thank you all so much for those of you that are over there. If you are not yet, I recommend checking it out because I really do think you will love it. And if you don't, I'll Venmo you $3 back or something. So you got a money back guarantee if you sign up and you hate it. I will say this as well. I don't delete the episodes of Headlines after a couple of weeks, you know?

If you sign up, you've got like tons of shows, hundreds of shows to go back and listen to if that's your kind of thing. I think those spoil quickly, but some people like going back through the archives and they are all there waiting for you. If you've not checked it out yet, you've got a ton of stuff waiting there. So head over there and help out support the show. Thank you all so much for the support. All right, let's do some launch math and some schedule looking because none of this makes any sense. Uh, like I said, the FCC guideline or the, the, I don't know what guideline is probably not the right word for it. The FCC, uh, what's the word I'm looking for? I guess rules? These are rules? They're not more like

guidelines, unlike Pirates of the Caribbean. Uh, so they've got three years to launch 1,618 satellites, or else their license will get cut down, or they'll have to apply for, this is, honestly, if you want my real take right now, and then you can log off the rest of the show while we do fun math, they're probably going to apply for an extension and say something, something COVID, something, something inflation, space is hard, please forgive us. Like that'll probably happen. They got their license in 2020. So, you know, bada bing, bada boomboom, COVID delays, so we've got another couple years on the lifeline here. I think that's probably what's going to happen, but let's just see if they can get there without doing that, because the less risky way would be actually to ship all these

satellites to space. Well, there's a couple of problems here. Number one is, they got nine Atlas 5s to fly. Now they're going to have eight with this other prototype launch that's going to happen. They're going to have eight Atlas Vs that they can use a full batch of satellites on, right? So if we do some math there, the Atlas V 551, which is the variant with five solid rocket boosters, that can put up almost 19 tons, metric tons, to low Earth orbit. Based on the amount of satellites that we know they're fitting on the other vehicles, they're probably fitting, I don't know, 35 Kuiper satellites on a fully loaded Atlas 5. For reference, they're putting 45 satellites on each Vulcan, 61 satellites on New Glenn,

they're putting 45 satellites on each Vulcan, 61 satellites on Newt Glenn, and we got a range of 35 to 40 satellites on Ariane 6, which there's some more stipulations within there. But that's the general guideline on how many they can fit on each of these. So if we do some math and we say they're able to launch all the other Atlas 5s within the next three years. That's 280 satellites that can make their way to low-Earth orbit. Pretty good start. Not a huge chunk of the 1,618 that they need to launch by three years from now, but a pretty good start. Now, just looking at the schedule for the other three vehicles, none of them have flown yet, right? And they're all still several months off from actually getting to their first flight. Several months at best, I'll say. So that's number one issue to really contend with is a bunch of vehicles that haven't flown yet that we don't exactly know what their ramp up is going to be like either in terms of production,

flight rate, any issues that might crop up in the early days. So to talk through that, Vulcan, you know, I'm thinking early next year, maybe six months from now, they get the first Vulcan flight off. And that one contains the Astrobotic Lander, nothing else really. The second flight they've said for a long time is the cargo mission to the ISS for Sierra Space with the Dream Chaser. And then, I don't know, they've got a bunch of national security flights that are in the backlog for Vulcan. So does the Kuiper mission start to jump the manifest of these national security missions? Probably not. So, you know, the first launch of Vulcan that carries Kuiper satellites, if it happened in 2024, I would be absolutely shocked. Absolutely shocked.

I would be absolutely shocked. Absolutely shocked. Which means they've got a year and a half to fly with Vulcan. New Glenn, on the other hand, I don't know if that's the other hand. I should just say in addition. We don't really know what the timeline is there overall. What we do know is they won the launch of Escapade from NASA, which is supposed to launch mid 2024 out to Mars. We have not yet seen a ton of hardware from New Glenn. There's been more of it in recent months, I will say.

The other day, Tori Bruno tweeted out a picture from the Huntsville engine factory that Blue Origin has set up, which is supporting Vulcan, but also supporting New Glenn. And it was more impressive than I expected. You know, I didn't know exactly what was going on there. I don't feel like I've seen a ton of pictures out of that Huntsville rocket engine factory before. They did say, you know, that they opened it up. Let's see, they opened it in February 2020. And at the time they said, we'll be making engines by the summer. Well, it was February 2020. So I assume they weren't making engines by the summer. Well, it was February 2020. So I assume they weren't making engines by the summer of 2020, given Wave's hand at the general state of the world from mid 2020

onward. So I really haven't seen many pictures inside Huntsville, but they had a bunch of stuff in flow there. And Tori Bruno called out that they had the flight engine for the second Vulcan flight in final assembly at the plant. And there was a bunch of other stuff around, you know, that's in addition to their stuff that they were building out in Washington state. So we have very little insight into how many engines they've been able to make in, in all, uh, we know that ULA has received a couple, they've integrated a few on a vehicle. They've test fired a few on a vehicle they blew one up on a test stand uh which was seems like much less of an issue than it was originally made out to seem like uh or at least

talked about like so that's less concerning but it's just a matter of like we don't know the exact division of the early run of be4 engines that blue origin's doing between vulcan and new glenn how many new glenn engines do they actually have to play with right now? Have they started making any? Have they started making some? I have a suspicion that they've started making some of the New Glenn engines and that they're not just sending all the first engines to ULA for Vulcan. But we'll really only know that once we start to see a flight booster roll out of their Florida factories with engines attached. Will we see that in the next few months? I really hope so. I feel like every year I hear a rumor that by the end of the year,

Blue Origin hopes to have a Pathfinder on the pad, and we've yet to see that. I'm sure there's a rumor floating around somewhere out there that says the same thing this year. Let's just be optimistic in this segment and say that they have at least a Pathfinder out in the next six months. They're able to fly maybe mid-2024. Maybe they do get that Escapade flight off to Mars. And here we go, we've got New Glenn flying this time next year. So that gives them two years with the potential for New Glenn flights. New Glenn, we know a couple launch contracts, but we don't really know any hard commitments on timelines for those launch contracts. We don't even really know that many

hard payloads for those flights. We know Escapade going to Mars, and we've just seen a lot of agreements other than that. So, you know, Telesat, I think, was going to launch some satellites on there. That's a constellation that's still years off. So I don't know anything that's going to eat the early schedule of New Glenn. So Kuiper might be able to demand the early launches from New Glenn's manifest. So in our optimistic scenario, they've got a year and a half with Vulcan available, and they've got two years of New Glenn available. Ariane 6, I have literally no idea what to say about this segment. They're not going to be flying Ariane 6 until early to mid next year, 2024.

flying arian 6 until early to mid next year 2024 they are initially flying arian 6 they being arian space in the two solid rocket booster variant um blue origin is flying with or sorry amazon is flying with the four solid rocket booster variant of arian 6 um and to get even more complicated amazon bought 18 launches on Ariane 6 16 of them are with upgraded solid rocket boosters so Ariane 6 needs to fly not only its initial version with two solid rocket boosters and get through the teething issues with two solid rocket boosters and then get through a couple of missions that they have

for institutional payloads from Europe with that two solid rocket booster version. Then they need to go to the four solid rocket booster version and fly that. And then eventually they can fly two Amazon Project Kuiper launches before they need to go upgrade to the bigger solid rocket boosters for the other 16 launches that Amazon bought. So if you can figure out that log jam, I don't know, between now and three years from now, let's just, I honestly think there will be zero Ariane 6 flights with Project Kuiper satellites on top before they hit that 2026 deadline. Again, we're going to just go optimistic in the scenario and say the two four solid rocket

booster variant launches that Aririanespace agreed to actually flies within the next three years. So where does that leave us? Well, we got to come up with some numbers on how many flights we think the Vulcan and Newt Glenn might fly. Let's just give Arianespace credit for two launches. We got eight Atlas vehicles that are going to fly with a full complement of Project Hyper satellites.

So if you're keeping score, we are at 352 satellites put into orbit. The two prototypes, 35 satellites on eight Atlas Vs and two satellites on Ariane 6. This is the most ridiculous segment I've ever done this show, but we're pushing on through. If you're bailing at this point, I'll see you next week. A year and a half worth of Vulcan flights to deal with. I don't know. They've got dream chaser flights they've got national security flights and they've got project kuiper flights they have to ramp up to an actual flight rate i'm gonna say in 2024 if ula gets three vulcans off i'll be thrilled um none of them are going to be for project kuiper the year and a half from that point forward, can they get six off given Vulcan's ramp up? Given that Amazon can make enough satellites, I'm going to give them credit for

six on the optimistic side. And on the New Glenn front, I'm going to say they have like full control of the New Glenn payload manifest besides escapade. New Glenn's ramp, I expect to be very slow compared to Vulcan. So in the course of two years of a vehicle flying, I don't know. I'm going to give them four launches of Project Kuiper satellites. And that brings us to 866 satellites launched in what I assume is a very optimistic scenario by July of 2026. So they are, I'm going to do some quick math, they are 752 behind where they need to be by the FCC rules. Just for fun, I ran the math on where they would need to be by July of 2026 to be beyond the FCC rules. Within three years, they will have to have flown

nine Atlas Vs, 13 Vulcans, nine New Glens, and six Ariane 6 vehicles. An absolutely astronomical amount. There's no way they are ever reaching anything close to that number. My most optimistic scenario puts them at 866 satellites. I don't think they'll get even close to that. And I'm really, really interested to see what happens here because not only would this be a major bummer for Amazon to struggle this hard at this, you know, admittedly incredibly complicated project, incredibly expensive taxing project. project, incredibly expensive taxing project. SpaceX is clearly the outlier in both their configuration of their mission, and they're just always the outlier in the industry. But they have a huge fleet of vehicles that they can fly. They have not only the ability to pay an internal cost

for the vehicle, rather than the sticker price. They have the ability to manage their entire schedule. They're reusing hardware so they can fly very frequently. There's, you know, the biggest problem that Starlink has right now is that very clearly, by the amount of satellites going up on each Falcon 9 launch right now, they are soon to be hurting from the fact that Starship is not flying yet to let them get enough satellites up to orbit. soon to be hurting from the fact that Starship is not flying yet to let them get enough satellites up to orbit. They're down to flying 15 to 20 some satellites per Falcon 9. You know, they on the original versions, they flew 60. And now they're flying what's called the V2 minibus for Starlink. That's not even the full size V2 satellite. So, you know, how much is that putting into

their future potential with Starlink in terms of capacity and throughput and all that kind of stuff? They need Starship flying, and that's the biggest risk right now for Starlink. On the Amazon side of things, you know, they are not in control of their own launch schedule. They are paying a ton of money for the launches themselves. They are investing heavily in infrastructure from, you know, ULA's infrastructure at the Cape to Arian Space's infrastructure in French Guiana. We don't really know exactly what, but probably a similar deal with Blue Origin to some extent.

They're like $10 billion in on launch alone before they get to building a satellite production line and user terminals and all of the complicated stuff that is a requirement of this kind of system working. all of the complicated stuff that is a requirement of this kind of system working. And the thing that's risky here for the sector overall is if Amazon can't figure this one out, if they can't crack this problem, I'm not sure there's a lot of companies out there that will go, well, I bet we can do it. Amazon, given what they're able to do with AWS and everything else that's under the Amazon brand, they have a really good reputation for these kind of very complex, technologically driven projects

that are heavily focused on infrastructure. And here they are staring down a project that I'm sure is just exploding in cost, exploding in complexity, and the schedule is getting out of hand up against governmental guidelines. And I just feel like that could put the industry in a really weird spot where investment in this kind of project completely dries up and people start looking elsewhere for, okay, well, that methodology didn't work. What are we going to do now?

You know, we've already seen OneWeb go through several different phases of unraveling to land at where they are now, which is, you know, they did get to orbit, but it's in a different format than they thought before, different ownership structure. The size of their constellation was constantly in flux. So again, we're left in another scenario where SpaceX is this huge outlier, and we have one other competitor out there, and one that looks like they're going to struggle to

even reach the same capability level as the single competitor on the market. Two weeks ago on an off nominal, we had Karan of K2 Space, one of the co-founders, on to talk about what they're working on, which are these very large satellites to fly in the post Starship era. And honestly, the post New era, where mass and volume are things that we worry about a lot less. And they're trying to figure out what the industry is to do in that world. And they're working on these very large satellite buses that have very large power generation capabilities, which would allow you to do very high-throughput satellites for, in their ideal, less cost than it is right now to do a high throughput satellite all the way out at GEO, which is like billions of dollars. They're looking to take a couple zeros

off that. And their theory is that medium Earth orbit is going to have a bit of a rebound when they're able to very cheaply put up very power intensive satellites for not a lot of money. And you could buy tens of these satellites and put them in medium-earth orbit and provide a lot of throughput with a lot of power generation, a lot of capability. And I just wonder, looking at this Amazon situation,

if their theory is not entirely correct, that there's going to be a swing back to not needing to build thousands of satellites, not needing them to launch thousands and hundreds of times for thousands of satellites. And if the latency that you find in a couple thousand kilometers versus a couple hundred kilometers, if that's good enough for a lot of use cases and the next generation of satellite interconstellations come back around to, you know, the medium Earth orbit where we kind of skipped

over for a little bit. We went from GEO, there's a hot minute where Mio was in play, Leo took over all the shine, and all the money was going there. Maybe it was a little too much for everyone that is not called SpaceX. And, you know, maybe K2 is right, and they're going to bounce back with more capability for less costs. If Mio is the right sweet spot, it's more palatable to investment types, it's more, it's more, you know, uh, it's more palatable to investment types. It's more, it's more, you know, the ability to budget for that project is a little bit more in sync with, uh, reality, I think. And, um, I don't know, I don't know where that leaves us. I'm just like between this storyline and what Karam was talking to us about on off nominal. I'm like, Hmm, maybe there's something there. Maybe they got that right. And, um, you know, I, I hope I'm wrong that the Amazon project does not unravel to the same extent that I'm envisioning here. But I am a little bit worried

that if it does go that way, that it will have major repercussions in the kinds of projects that the industry at large takes on. And it will have some, you know, downstream effects from Amazon's experience in that regard. So didn't mean to leave it on a downer note, but I've just been flummoxed by this math. You let me know how many satellites you think Amazon will launch by July of 2026. My optimistic take was 866, and I think that is way too high, way too optimistic, but we shall see. So that's all I've got for you. Thank you all so much for listening to this completely bizarre episode of Main Engine Cutoff. If you've got any

questions or thoughts, which I'm sure there will be many, hit me up on email, anthony at mainenginecutoff.com I would plug some of the other social networks, but there's nine Twitters now and I don't feel like talking about it, so just email me if you've got questions or thoughts and I'll bring it up on the next show if you do. Until then, thanks all so much for listening. Thanks for

your support, as always, over at mainenginecutoff.com slash support, and I will talk to you do. Until then, thanks all so much for listening. Thanks for your support as always over at managingcutoff.com slash support, and I will talk to you soon.