New Armstrong Speculation and Discussion - Thread 2

[meekGee]

I've made my points, and so I'm not going to rehash the same point-by-point arguments. However, I will generally challenge others to:

- Think about New Armstrong in the context of the actual addressable market, not the total market. What fraction of 85% of payload mass to orbit is payload mass that New Armstrong could actually tap into? What fraction of the addressable market today are single launches to single orbits versus mass-constrained constellation deployments?

- Pay attention to the difference between constellation operator profits and launch profits. Starlink generates a lot of profit for SpaceX, because SpaceX and Starlink are one and the same. If SpaceX IPOed Starlink tomorrow, what would the resulting SpaceX profit profile - i.e. launch profits only - look like?

- Think about ROI in terms of margin. A constellation may generate a lot of revenue, but as a tightly-negotiated and tightly-competed block buy, the margins are thin. A launch company would be well-served to consider what the actual profit margins are on any particular mission.

It's a good rule of business - get customers who are themselves profitable.

Also, get customers that buy a lot of product.

Of course they'll negotiate you down, but the alternative of having a small number of customers that you can drive the margin up with - that's a losing proposition.

Most big businesses are high volume, low margin.  You don't even have to have the customers negotiate you down - often your competitors do it for them.

SpaceX has the additional benefit of being its own customer, but that's just because they've been smart and practically started that market.

That's just even more reason for BO to get going on NA.  Amazon/Kuiper is not fully under JB's control, and Amazon is currently looking at a very large money sink in AI that's actually an existential problem. Which means its discretionary spending powers are diminished.

Kuiper was seen as existential "Since Elon has one" but objectively Amazon could use Starlink exactly like Google and Microsoft can.  SpaceX has every reason to treat them exactly the same.

Which means either Kuiper gets real-world competitive or it gets cancelled. So Kuiper won't be a sugar daddy, and BO needs that business.

JB got a lot richer in recent years. If BO is able to deliver, he can fund development. That's their only way forward.

[RedLineTrain]

@sstli2, I have reread your posts and I agree with much of what you say.  But low margin businesses are Amazon's and therefore Bezos's stock-in-trade.  They can be great businesses and indeed low margins can create a competitive impediment against those who are unwilling to dabble in such activities.

New Armstrong's business case is dependent on Kuiper, maybe more than it should be.  I am a bit lukewarm on Kuiper.  I observe that Amazon has many demands on its capital investment dollars and that this could pinch Kuiper.  Kuiper in any ambitious form is a multi-ten billion dollar endeavor.  Competing against Starlink with its current ambitions might require $10 billion per year in capital investments.  I wonder whether Amazon is capable of sustaining that rate for something that while strategically good, is not existential to the business and that results in an inherently high-cost product.

Bezos could choose to expand the markets for which New Armstrong is suitable.  He is an O'Neill disciple, so he may wish to create space stations.  That would provide a good second market for New Armstrong.

Lastly, we should keep in mind that Musk has oscillated between Starship and reusable upper stage for Falcon 9.  It is not so easy to decide which is a better use of time and money.  Bezos will encounter this dilemma.

If Starship is cheaper than New Armstrong (and all others) then Amazon will likely see a another shareholder lawsuit if Bezos tries to force Amazon to avoid Starship. This already happened with Falcon 9.

IMO Kuiper will not be successful. It was a long shot when it was supposed to launch in late 2022, and it is losing potential customers to Starlink for every month of delay since then. It does have Amazon itself as a large captive customer (including every Amazon delivery truck) but I doubt it's enough. By the time New Armstrong launches, Kuiper will no longer need it, one way or another, so New Armstrong would need a different mega-constellation, and by then, the market will be saturated by Starlink and possibly a Kuiper struggling to survive, so another successful constellation is unlikely.

As stated in other threads, the substance of the issue is not the narrow question of shareholder lawsuits.  As you allude to in the second paragraph, the success of Kuiper depends on it lofting satellites in numbers as soon as possible.  Time is of the essence.  Falcon 9 and Starship are wonderful resources that Amazon should not be shy to utilize, if the timing of its preferred launch providers does not line up perfectly.  It has surprised me to watch Bezos, a world-class businessman, fail to grasp this fully.  It's a measure of how strongly Amazon is committed to Kuiper.

[mordroberon]

It seems likely to me that Blue is looking into full re-usability. It seems like their design philosophy is to sink money into building the best version of the system that they can, which can then be amortized across dozens if not hundreds of launches. Whereas spacex is content with a rapid iteration, low cost, mass-produced strategy. Blue's gamble is that while spacex was able to do it first, they will be able to do it better.

As for Armstrong, I don't think it's coming any time soon, i.e. in this decade or next. If it does come I think it would be for launching larger amounts of mass to the moon, basically a replacement for SLS, which I don't expect to outlive the Artemis program. I suspect full reusability will not be a priority.

[Coastal Ron]

...- Think about New Armstrong in the context of the actual addressable market, not the total market. What fraction of 85% of payload mass to orbit is payload mass that New Armstrong could actually tap into? What fraction of the addressable market today are single launches to single orbits versus mass-constrained constellation deployments?

Using TAM/SAM/SOM analysis requires that you imagine what future markets are opened up due to the significantly lower launch costs, so you really need to figure that part out first.

- Pay attention to the difference between constellation operator profits and launch profits. Starlink generates a lot of profit for SpaceX, because SpaceX and Starlink are one and the same. If SpaceX IPOed Starlink tomorrow, what would the resulting SpaceX profit profile - i.e. launch profits only - look like?

SpaceX is a private company, so we really don't know if SpaceX is making a profit on Starlink yet. All we have is speculation, especially since Musk has said that Starship was needed to get most of the Starlink constellation into orbit, and that hasn't happened.

And playing "What-if" games is nonsensical at this point, since it has no bearing on whether Blue Origin should build a supposed "New Armstrong", which we have no idea what its value proposition would be.

- Think about ROI in terms of margin. A constellation may generate a lot of revenue, but as a tightly-negotiated and tightly-competed block buy, the margins are thin. A launch company would be well-served to consider what the actual profit margins are on any particular mission.

Companies that worry about margin tend to be in mature markets with lots of competition, and that is not the situation today.

Today Starship is the only fully reusable launcher that is in development, and SpaceX will be able to charge whatever they want. Historically though, SpaceX has focused on lowering customer prices to inspire more demand, and I think they will do that for Starship too. That doesn't mean they will barely make a profit, just that their customers will see great value in the prices that SpaceX offers.

That will change when a second provider comes online, but we don't know yet how that will affect the market.

[spacenut]

About 20 years ago, there was discussion on NASAspaceflight.com about the need for a 40-60 tons to LEO rocket.  This was initially thought to be Atlas V phase 2 or upgraded Delta IV heavy.  Atlas V phase II would have been a 5-5.5m diameter Atlas with two RD-180's which with ACES upper stage and some solids could get about 40 tons to LEO and with a 3 core heavy version could get about 70-75 tons.  Delta IV heavy would have introduced crossfeed for their heavy version with an upgraded engine and some solids to get 40-50 tons to LEO.  Neither happened because neither NASA nor the Air Force needed it done.  The 40-60 ton range was for a robust cis-lunar program. 

Fast forward today.  We have FH with 40-60 ton capability to LEO and can launch components for the Artemis program without the need for Starship.  Also, New Glenn is about to come on line and it can deliver about 40 tons to LEO and can deliver a lunar lander to the Artemis lunar orbit.  This is also without the need for SLS.  Both rockets with in orbit minimum refueling doesn't need SLS. 

Now, SpaceX is soon to bring on line Starship and one lunar lander will be a version of Starship.  Question is now, does Blue develop New Armstrong and when?  If they develop a New Armstrong, it could very well compete with Starship for a very robust cis-lunar program.  NASA would have dual bidders for redundancy with a New Armstrong.  They will get it with New Glenn vs FH.  So why not bigger.  IF Blue develops a reusable New Armstrong, then with Starship, there will be no need for SLS and it's huge money drain goes away.  Boeing will be left in the dust.  ULA also will be extremely hurt by the two also.  SpaceX and Blue will become the new Lockheed vs Boeing companies. 

[DanClemmensen]

@sstli2, I have reread your posts and I agree with much of what you say.  But low margin businesses are Amazon's and therefore Bezos's stock-in-trade.  They can be great businesses and indeed low margins can create a competitive impediment against those who are unwilling to dabble in such activities.

New Armstrong's business case is dependent on Kuiper, maybe more than it should be.  I am a bit lukewarm on Kuiper.  I observe that Amazon has many demands on its capital investment dollars and that this could pinch Kuiper.  Kuiper in any ambitious form is a multi-ten billion dollar endeavor.  Competing against Starlink with its current ambitions might require $10 billion per year in capital investments.  I wonder whether Amazon is capable of sustaining that rate for something that while strategically good, is not existential to the business and that results in an inherently high-cost product.

Bezos could choose to expand the markets for which New Armstrong is suitable.  He is an O'Neill disciple, so he may wish to create space stations.  That would provide a good second market for New Armstrong.

Lastly, we should keep in mind that Musk has oscillated between Starship and reusable upper stage for Falcon 9.  It is not so easy to decide which is a better use of time and money.  Bezos will encounter this dilemma.

If Starship is cheaper than New Armstrong (and all others) then Amazon will likely see a another shareholder lawsuit if Bezos tries to force Amazon to avoid Starship. This already happened with Falcon 9.

IMO Kuiper will not be successful. It was a long shot when it was supposed to launch in late 2022, and it is losing potential customers to Starlink for every month of delay since then. It does have Amazon itself as a large captive customer (including every Amazon delivery truck) but I doubt it's enough. By the time New Armstrong launches, Kuiper will no longer need it, one way or another, so New Armstrong would need a different mega-constellation, and by then, the market will be saturated by Starlink and possibly a Kuiper struggling to survive, so another successful constellation is unlikely.

As stated in other threads, the substance of the issue is not the narrow question of shareholder lawsuits.  As you allude to in the second paragraph, the success of Kuiper depends on it lofting satellites in numbers as soon as possible.  Time is of the essence.  Falcon 9 and Starship are wonderful resources that Amazon should not be shy to utilize, if the timing of its preferred launch providers does not line up perfectly.  It has surprised me to watch Bezos, a world-class businessman, fail to grasp this fully.  It's a measure of how strongly Amazon is committed to Kuiper.

Kuiper added three F9 launches in December 2023 to fill in for delayed introductions by their primary LVs. Those three appeared to allow them to make their FCC deadline of 1618 satellites by July 2026, at least on paper in an alternate universe where all the LVs achieved the launch schedules projected at that time, and the Satellites were ready. None of that seems to have happened, so they will miss the date. I'm not sure how that affects New Armstrong, but it's at least possible that Kuiper will not be a relevant customer when New Armstrong is introduced.

[deltaV]

Today Starship is the only fully reusable launcher that is in development

There are two other fully reusable launchers in development: Stoke Space's Nova and a future New Glenn upgrade (https://forum.nasaspaceflight.com/index.php?topic=61343.msg2616302#msg2616302).

[RedLineTrain]

@sstli2, I have reread your posts and I agree with much of what you say.  But low margin businesses are Amazon's and therefore Bezos's stock-in-trade.  They can be great businesses and indeed low margins can create a competitive impediment against those who are unwilling to dabble in such activities.

New Armstrong's business case is dependent on Kuiper, maybe more than it should be.  I am a bit lukewarm on Kuiper.  I observe that Amazon has many demands on its capital investment dollars and that this could pinch Kuiper.  Kuiper in any ambitious form is a multi-ten billion dollar endeavor.  Competing against Starlink with its current ambitions might require $10 billion per year in capital investments.  I wonder whether Amazon is capable of sustaining that rate for something that while strategically good, is not existential to the business and that results in an inherently high-cost product.

Bezos could choose to expand the markets for which New Armstrong is suitable.  He is an O'Neill disciple, so he may wish to create space stations.  That would provide a good second market for New Armstrong.

Lastly, we should keep in mind that Musk has oscillated between Starship and reusable upper stage for Falcon 9.  It is not so easy to decide which is a better use of time and money.  Bezos will encounter this dilemma.

If Starship is cheaper than New Armstrong (and all others) then Amazon will likely see a another shareholder lawsuit if Bezos tries to force Amazon to avoid Starship. This already happened with Falcon 9.

IMO Kuiper will not be successful. It was a long shot when it was supposed to launch in late 2022, and it is losing potential customers to Starlink for every month of delay since then. It does have Amazon itself as a large captive customer (including every Amazon delivery truck) but I doubt it's enough. By the time New Armstrong launches, Kuiper will no longer need it, one way or another, so New Armstrong would need a different mega-constellation, and by then, the market will be saturated by Starlink and possibly a Kuiper struggling to survive, so another successful constellation is unlikely.

As stated in other threads, the substance of the issue is not the narrow question of shareholder lawsuits.  As you allude to in the second paragraph, the success of Kuiper depends on it lofting satellites in numbers as soon as possible.  Time is of the essence.  Falcon 9 and Starship are wonderful resources that Amazon should not be shy to utilize, if the timing of its preferred launch providers does not line up perfectly.  It has surprised me to watch Bezos, a world-class businessman, fail to grasp this fully.  It's a measure of how strongly Amazon is committed to Kuiper.

Kuiper added three F9 launches in December 2023 to fill in for delayed introductions by their primary LVs. Those three appeared to allow them to make their FCC deadline of 1618 satellites by July 2026, at least on paper in an alternate universe where all the LVs achieved the launch schedules projected at that time, and the Satellites were ready. None of that seems to have happened, so they will miss the date. I'm not sure how that affects New Armstrong, but it's at least possible that Kuiper will not be a relevant customer when New Armstrong is introduced.

Yes, I hate to sow FUD on Kuiper, but Amazon's actions don't seem to show full "all-in" commitment and execution.

Today, a Vietnamese business magazine stated that SpaceX executives are quoting $11 billion invested in Starlink to date.  My guess is that Amazon needs to spend 2x - 5x SpaceX to get to the same place.  Perhaps $10 billion or more a year in perpetuity, which would seem to amply justify New Armstrong if it were introduced in the medium term.  This is no joke table stakes, so I wouldn't blame Amazon if it did an about-face on the whole business citing changing business realities.

https://theinvestor.vn/spacex-seeks-to-provide-starlink-satellite-internet-services-in-vietnam-d12146.html

[wannamoonbase]

I know that the only acknowledgement of New Armstrong is one comment by Bezo from years ago.  However, I'm going to speculate a bit.

Seeing NG's first booster it made me think that New Armstrong could just be an upsized NG.  Widen the diameter to 10 meters or so and add another ring of 10 or 12 BE-4's for a total of 15-17 booster engines. 

With downrange booster recovery and hydrogen upper stage(s) a NA vehicle could be very formidable.

Then the rest of the vehicle is just upsized, using the same basic design materials, chines, strakes, flight and recovery systems.

NG has been a very long wait but NA could be an evolution with many things proven out on NG.

[meekGee]

I know that the only acknowledgement of New Armstrong is one comment by Bezo from years ago.  However, I'm going to speculate a bit.

Seeing NG's first booster it made me think that New Armstrong could just be an upsized NG.  Widen the diameter to 10 meters or so and add another ring of 10 or 12 BE-4's for a total of 15-17 booster engines. 

With downrange booster recovery and hydrogen upper stage(s) a NA vehicle could be very formidable.

Then the rest of the vehicle is just upsized, using the same basic design materials, chines, strakes, flight and recovery systems.

NG has been a very long wait but NA could be an evolution with many things proven out on NG.

So a 15 m bottom skirt?

I think before they go there they need to develop a more compact BE-4

[wannamoonbase]

I know that the only acknowledgement of New Armstrong is one comment by Bezo from years ago.  However, I'm going to speculate a bit.

Seeing NG's first booster it made me think that New Armstrong could just be an upsized NG.  Widen the diameter to 10 meters or so and add another ring of 10 or 12 BE-4's for a total of 15-17 booster engines. 

With downrange booster recovery and hydrogen upper stage(s) a NA vehicle could be very formidable.

Then the rest of the vehicle is just upsized, using the same basic design materials, chines, strakes, flight and recovery systems.

NG has been a very long wait but NA could be an evolution with many things proven out on NG.

So a 15 m bottom skirt?

I think before they go there they need to develop a more compact BE-4

Probably 10 meter-ish core and a 12 meter ish bottom skirt. 

A stronger smaller BE-4 would help naturally but how many years would that take Blue to develop?

[DrTadd]

Probably 10 meter-ish core and a 12 meter ish bottom skirt. 

A stronger smaller BE-4 would help naturally but how many years would that take Blue to develop?

Does the BE4 need to get smaller? During the EA visit, it was said (assuming I am remembering correctly) that the BE4 was designed 'under stressed' for lots of flights. I would therefore surmise that the engine has extra potential hiding in there untapped. With more flights and units getting inspected after flight, I would bet the motor gets uprated.

Question is how much margin did BO design in to get their 'lots of flights'.

This is kinda the flip side where SX is with SH... except their R2 is maxed out and they need R3 to get their designed payload.

[meekGee]

I know that the only acknowledgement of New Armstrong is one comment by Bezo from years ago.  However, I'm going to speculate a bit.

Seeing NG's first booster it made me think that New Armstrong could just be an upsized NG.  Widen the diameter to 10 meters or so and add another ring of 10 or 12 BE-4's for a total of 15-17 booster engines. 

With downrange booster recovery and hydrogen upper stage(s) a NA vehicle could be very formidable.

Then the rest of the vehicle is just upsized, using the same basic design materials, chines, strakes, flight and recovery systems.

NG has been a very long wait but NA could be an evolution with many things proven out on NG.

So a 15 m bottom skirt?

I think before they go there they need to develop a more compact BE-4

Probably 10 meter-ish core and a 12 meter ish bottom skirt. 

A stronger smaller BE-4 would help naturally but how many years would that take Blue to develop?

Well tight now a 7 circle (3 across) was large enough that they have a 9 m skirt on a 7 m core.

A 19 circle (5 across) would be about 9/3*5=15 m across.  Maybe a touch less since the OD structure remains the same. So maybe 14.

Heavy lift is about thrust density, and BE-4 is terrible at that.  That's why even with 7 engines they have a footprint almost identical to Starship. (And 1/5 the engine count)

A skirt gets progressively worse with diameter since you're making up for an aerial efficiency using a linear structure.

What are the basic design choices that dictate thrust density?  (I'm assuming the nozzle is sized to fit in SL engines)

[JEF_300]

Comparing  the base size of New Glenn and SuperHeavy isn't really fair. Raptor has the highest chamber pressure of any rocket engine, ever. BE-4 flew for the first time a few months ago. It's like comparing Michael Phelps to someone on the high school swim team.

Give Blue a couple years to fly their engine and learn it's ins and outs. The base of New Glenn won't get any bigger or smaller, but I'm pretty confident it's thrust (and therefore thrust density) will go way up.

[meekGee]

Comparing  the base size of New Glenn and SuperHeavy isn't really fair. Raptor has the highest chamber pressure of any rocket engine, ever. BE-4 flew for the first time a few months ago. It's like comparing Michael Phelps to someone on the high school swim team.

Give Blue a couple years to fly their engine and learn it's ins and outs. The base of New Glenn won't get any bigger or smaller, but I'm pretty confident it's thrust (and therefore thrust density) will go way up.

Do you think BE-4 will come close to matching Raptor in Thrust density?  I know it'll improve, but it's at a huge deficit, and Raptor's not done either.

[JEF_300]

Comparing  the base size of New Glenn and SuperHeavy isn't really fair. Raptor has the highest chamber pressure of any rocket engine, ever. BE-4 flew for the first time a few months ago. It's like comparing Michael Phelps to someone on the high school swim team.

Give Blue a couple years to fly their engine and learn it's ins and outs. The base of New Glenn won't get any bigger or smaller, but I'm pretty confident it's thrust (and therefore thrust density) will go way up.

Do you think BE-4 will come close to matching Raptor in Thrust density?  I know it'll improve, but it's at a huge deficit, and Raptor's not done either.

Probably not, but I think it could eventually get close enough that the comparison does make some sense.

[Steven Pietrobon]

A 19 circle (5 across) would be about 9/3*5=15 m across.

I believe that should be 7*5/3 = 11.7 m for a 19 engine version as New Glenn is 7 m wide, not 9 m.

[meekGee]

Comparing  the base size of New Glenn and SuperHeavy isn't really fair. Raptor has the highest chamber pressure of any rocket engine, ever. BE-4 flew for the first time a few months ago. It's like comparing Michael Phelps to someone on the high school swim team.

Give Blue a couple years to fly their engine and learn it's ins and outs. The base of New Glenn won't get any bigger or smaller, but I'm pretty confident it's thrust (and therefore thrust density) will go way up.

Do you think BE-4 will come close to matching Raptor in Thrust density?  I know it'll improve, but it's at a huge deficit, and Raptor's not done either.

Probably not, but I think it could eventually get close enough that the comparison does make some sense.

Maybe.  But compare BE-4 with Raptor v1, and draw a similar conclusion. Since then, Raptor went up by just over 50%. Assume BE-4 will do the same.

Still the same problem.

The question is whether the difference stems from fundamental things (e.g. cycle type) or just from things that can be improved.

If it's the latter, then thrust density can improve by any arbitrary amount. If not, then, well, it can't.

But you won't get a New Armstrong with the current numbers.

[meekGee]

A 19 circle (5 across) would be about 9/3*5=15 m across.

I believe that should be 7*5/3 = 11.7 m for a 19 engine version as New Glenn is 7 m wide, not 9 m.

But the skirt (engine circle) is about 9.  I'm calculating what the skirt of a 19 BE-4 hypothetical rocket would look like.

BTW a 19 BE-4 rocket is still a lot smaller than Starship, just above 1/2 of it.

[Steven Pietrobon]

But the skirt (engine circle) is about 9.  I'm calculating what the skirt of a 19 BE-4 hypothetical rocket would look like.

Pixel counting from image in link below, I get diameter for the base as 7*457/374 = 8.55 m.

https://www.blueorigin.com/new-glenn

The image from the link below gives a BE-4 nozzle exit diameter of 1.9 m.

https://www.reddit.com/r/spacex/comments/amdyi4/raptor_engine_size_comparison_13m_nozzle_scaled/

Pixel counting this base image gives a space of 1.9*19/90 = 0.4 m between the engines.

https://www.nasaspaceflight.com/2024/10/blue-first-stage-testing/

Thus total estimated base diameter for a 19 engine New Armstrong is 8.55 + 2*(1.9+0.4) = 13.15 m. The core diameter could be 13.15*7/8.55 = 10.75 m.