Reuse business case

[Jim]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

[meekGee]

The bottom line.

Reuse if implemented correctly is highly economically successful.

 Globally last year the total was 135 successful launches. Up >50% from the numbers in 2016 (83 successful launches [Gunter's Space Page 2016 year of launches list]).

91 of 2021's successes were performed by just three launch vehicle families:  CZ (DF-5 based), Falcon 9, and R-7.  Only 29 of those flights provided successful first stage recoveries (defined as returning a usable first stage).  The truth is that China's all-expendable program has been responsible for much of the growth in launch numbers recently, building whatever "space environmental sensing" network they are building up there.     

 - Ed Kyle

Are you saying that out of three leading launch systems, only one is reusable?  That's hardly news...
Also out of those three, only one is truly commercial.

[tssp_art]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

Agree. To elaborate, it's not that ULA is unable - though they might be. It's that they are likely not allowed. They are owned by Lockheed Martin Space (50%) and Boeing Defense, Space & Security (50%) and, they are chartered to maximize the return to those companies by maximizing profit on the sale of Atlas and Delta launches and, after much negotiation, the new Vulcan launcher. But to do anything else they need permission from both parent companies to invest money instead of providing that money as profit. That permission has not been forthcoming. Not for ACES or SMART Reuse (which both seemed like good and sensible investments) much less for something like a Starlink competitor. Part of that reluctance may be just the need to add profit to the bottom line of the parents. But it could also be that those efforts do not align with the goals of both parent companies - or that they compete with efforts that one or both parents are considering. Bottom line is that ULA has highly and artificially constrained access to market. They are not free to pursue options available to SpaceX, Amazon, or even Rocket Lab,

[trimeta]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

I tend to agree: I raised the question largely rhetorically, to suggest that reuse is economical if and only if you have a high flight rate, and having a high flight rate is only feasible if you've got your own in-space activities.

I suppose there is the question of Project Kuiper buying from third-party launch providers, but that seems like a short-term market, since ultimately they're going to prefer launching with Blue Origin (not exactly a single company running both launch and in-space activities, but quite close). So your market only lasts until New Glenn hits its stride. And any hypothetical new reusable vehicle would arrive at around the same time anyway.

[meekGee]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

But the parents can.

This whole ULA vs. parents is just a corporate naming game.  ULA was the launch services of the parents and now it's merged and incorporated separately, but the overall picture is the same.

Starlink and SpaceX could also have been incorporated separately, just like Kuiper and BO are too.

The only difference is that when it's a single entity, it's impossible to finger point across divisions as easily as it is to finger point across companies, since it's the same CEO...

[trimeta]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

But the parents can.

This whole ULA vs. parents is just a corporate naming game.  ULA was the launch services of the parents and now it's merged and incorporated separately, but the overall picture is the same.

Starlink and SpaceX could also have been incorporated separately, just like Kuiper and BO are too.

The only difference is that when it's a single entity, it's impossible to finger point across divisions as easily as it is to finger point across companies, since it's the same CEO...

That's a valid point -- if Boeing or Lockheed Martin went into the in-space activities business (to a greater extent than they already are, Boeing's existing satellite busses don't seem to demand high launch cadence for example), they could push for ULA to make high-cadence rockets with reusability. Although there's still one wrinkle: since ULA has two owners, you need both of them to approve such a change in direction, but if only one of those owners has an in-space activities business to justify it, this might be a harder sell.

[meekGee]

And sure, Starlink will likely make them a great deal of money in the long run, but do we expect ULA to also diversify into in-space applications which require high launch cadence?

it can't

But the parents can.

This whole ULA vs. parents is just a corporate naming game.  ULA was the launch services of the parents and now it's merged and incorporated separately, but the overall picture is the same.

Starlink and SpaceX could also have been incorporated separately, just like Kuiper and BO are too.

The only difference is that when it's a single entity, it's impossible to finger point across divisions as easily as it is to finger point across companies, since it's the same CEO...

That's a valid point -- if Boeing or Lockheed Martin went into the in-space activities business (to a greater extent than they already are, Boeing's existing satellite busses don't seem to demand high launch cadence for example), they could push for ULA to make high-cadence rockets with reusability. Although there's still one wrinkle: since ULA has two owners, you need both of them to approve such a change in direction, but if only one of those owners has an in-space activities business to justify it, this might be a harder sell.

True but part of the supposed attractiveness of the merge was risk mitigation for each of the parents..

[markbike528cbx]

Well, ULA (more probably the Boeing/Lockmart owners) have seen the light.

https://forum.nasaspaceflight.com/index.php?topic=48584.msg2357028#msg2357028
https://twitter.com/Free_Space/status/1511360115752026116

Irene Klotz
@Free_Space
To most efficiently fulfill its launch commitment to Amazon, ULA is moving ahead with program to recover Vulcan BE-4 engines for reuse, @torybruno tells @AviationWeek on sidelines of #SpaceSymposium

I've sometimes thought that Tony's anti-reuse public pronouncements to be lacking in enthusiasm.

I guess there is a bright side to the "Where's the engines Jeff?" catcalls for BE-4 delays. 
If you reuse them, you don't have to buy new.

[Lars-J]

Re: SMART reuse...
I'm surprised but pleasantly so - But I remain skeptical that this kind of reuse will provide the cost savings they are hoping for. BUT I am excited to see them trying, and I do wonder how many Vulcans will be launched before they start attempting recovery.

[sdsds]

Re: SMART reuse...
I'm surprised but pleasantly so - But I remain skeptical that this kind of reuse will provide the cost savings they are hoping for.

I think Bruno has pretty much said ULA is no longer "carrying" the costs associated with keeping the Vulcan design open to an alternate engine provider. But ULA would certainly prefer to decouple Vulcan's flight rate from the production rate of its engines. Reuse is the other way to do that, right?

[JayWee]

I guess there is a bright side to the "Where's the engines Jeff?" catcalls for BE-4 delays. 
If you reuse them, you don't have to buy new.

Not really, the current iteration of BE-4 is not designed to be reused. ULA hinted in the Europa Clipper RFI that BE-4 Block2 was coming.
Therefore, there will be another round of "Where's the engines Jeff?"

[edzieba]

Re: SMART reuse...
I'm surprised but pleasantly so - But I remain skeptical that this kind of reuse will provide the cost savings they are hoping for. BUT I am excited to see them trying, and I do wonder how many Vulcans will be launched before they start attempting recovery.

Even if it doesn't save any money, as Rocketlab have mentioned reuse allows for an increase in flight rate without an increase in production rate.

[TrevorMonty]

Re: SMART reuse...
I'm surprised but pleasantly so - But I remain skeptical that this kind of reuse will provide the cost savings they are hoping for. BUT I am excited to see them trying, and I do wonder how many Vulcans will be launched before they start attempting recovery.

Even if it doesn't save any money, as Rocketlab have mentioned reuse allows for an increase in flight rate without an increase in production rate.

ULA still need to produce new booster tanks but they are lot cheaper and quicker to produce than complex engine pods.

Sent from my SM-G570Y using Tapatalk

[Robotbeat]

Are they? The Shuttle external tank cost, what, about $150 million in today’s money? Still pretty dang expensive to make rocket tanks! SpaceX makes Merlins and Raptors for about $1 million apiece (I’m part because they make a LOT of them), and certainly engines must be a significant portion of the pod cost.

I think the issue is ULA isn’t a vertically integrated engine producer, so they have to rely on their external provider for engines, at a large markup. SMART reuse gives them bargaining power, if nothing else. Also, Blue may not be able to bring engine production up to SpaceX speed fast enough, so SMART would allow ULA to get a higher flightrate sooner and just using more tank production and refurb, which they could control in spite of not being vertically integrated.

[DanClemmensen]

SpaceX makes Merlins and Raptors for about $1 million apiece (I’m part because they make a LOT of them), and certainly engines must be a significant portion of the pod cost.

Do they? for Merlins, they need one for each expended second stage, or or about 50/yr, plus nine for each expended F9, FH side, or FH core, and a few replacements. so <100 Merlins/yr.

For raptors, we don't know yet. A bunch to get started, and then not many until building  the mythical Mars fleet. After the expended test boosters (maybe 3?) they need about 3 operational SH to handle the entire world's payload demands, so that's 100 + 100 for SH. They also need six or nine for each SS: pick a number, but the SS inventory must be bigger. Let's say 20 to start, so again less than 200. That's 400 raptors to get to maybe 2026.

By contrast, expendable Vulcan needs 65 pairs of engines, and New Glenn needs 7(?) for each reusable launcher, so maybe 200 total by 2026. That's fewer, but not an order of magnitude fewer.

[Redclaws]

SpaceX makes Merlins and Raptors for about $1 million apiece (I’m part because they make a LOT of them), and certainly engines must be a significant portion of the pod cost.

Do they? for Merlins, they need one for each expended second stage, or or about 50/yr, plus nine for each expended F9, FH side, or FH core, and a few replacements. so <100 Merlins/yr.

For raptors, we don't know yet. A bunch to get started, and then not many until building  the mythical Mars fleet. After the expended test boosters (maybe 3?) they need about 3 operational SH to handle the entire world's payload demands, so that's 100 + 100 for SH. They also need six or nine for each SS: pick a number, but the SS inventory must be bigger. Let's say 20 to start, so again less than 200. That's 400 raptors to get to maybe 2026.

By contrast, expendable Vulcan needs 65 pairs of engines, and New Glenn needs 7(?) for each reusable launcher, so maybe 200 total by 2026. That's fewer, but not an order of magnitude fewer.

It would probably be worthwhile comparing how many Merlins SpaceX has built total, which is somewhere in the neighborhood of 1000, since investments in the production line and learning related improvements have been done over that full run.  (Yes, they’ve also upgraded the engine, so that would apply some discount against the investments and learning over that 1000, but still.)

[freddo411]

SpaceX makes Merlins and Raptors for about $1 million apiece (I’m part because they make a LOT of them), and certainly engines must be a significant portion of the pod cost.

Do they? for Merlins, they need one for each expended second stage, or or about 50/yr, plus nine for each expended F9, FH side, or FH core, and a few replacements. so <100 Merlins/yr.

For raptors, we don't know yet. A bunch to get started, and then not many until building  the mythical Mars fleet. After the expended test boosters (maybe 3?) they need about 3 operational SH to handle the entire world's payload demands, so that's 100 + 100 for SH. They also need six or nine for each SS: pick a number, but the SS inventory must be bigger. Let's say 20 to start, so again less than 200. That's 400 raptors to get to maybe 2026.

By contrast, expendable Vulcan needs 65 pairs of engines, and New Glenn needs 7(?) for each reusable launcher, so maybe 200 total by 2026. That's fewer, but not an order of magnitude fewer.

I applaud the attempt at numerical estimates.   However, by my wild guess, informed by past performance, I'd say that:

* SX will need to produce more than you say (they will make more prototypes, and they will lose/expend a significant number)

* Blue won't produce the numbers you estimate by 2026.   Vulcan won't fly that much by then either.

[GWH]

There has been a few times where I get the sense that ULA made their analysis of SMART vs booster reuse using Vulcan vs New Glenn as the comparison - not SpaceX's Falcon family.

And I keep feeling that within the context of their primary customers it is the right decision - then against all odds it still remains correct within the context of a LEO constellation program.

My hat is off to ULA for finding their fit within a changing marketplace.

[DanClemmensen]

SpaceX makes Merlins and Raptors for about $1 million apiece (I’m part because they make a LOT of them), and certainly engines must be a significant portion of the pod cost.

Do they? for Merlins, they need one for each expended second stage, or or about 50/yr, plus nine for each expended F9, FH side, or FH core, and a few replacements. so <100 Merlins/yr.

For raptors, we don't know yet. A bunch to get started, and then not many until building  the mythical Mars fleet. After the expended test boosters (maybe 3?) they need about 3 operational SH to handle the entire world's payload demands, so that's 100 + 100 for SH. They also need six or nine for each SS: pick a number, but the SS inventory must be bigger. Let's say 20 to start, so again less than 200. That's 400 raptors to get to maybe 2026.

By contrast, expendable Vulcan needs 65 pairs of engines, and New Glenn needs 7(?) for each reusable launcher, so maybe 200 total by 2026. That's fewer, but not an order of magnitude fewer.

I applaud the attempt at numerical estimates.   However, by my wild guess, informed by past performance, I'd say that:

* SX will need to produce more than you say (they will make more prototypes, and they will lose/expend a significant number)

* Blue won't produce the numbers you estimate by 2026.   Vulcan won't fly that much by then either.

Your guess about the BE-4 demand effectively says that the Kuiper contracts are not achievable, since the number for my guess come straight from the ULA and Kuiper announcements.

Your guesses may be better than mine for Starship. From an engine usage perspective, the number of expended SHs  is the big driver. I guessed three, you are guessing may six, so another hundred expended engines.

Of course, the biggest deal is the engine unit cost. The comments here and in the trade press seem to accept that Raptor unit cost will be below $1 M, and Elon dreams of $250,000. If this happens, BO may find to hard to match SpaceX costs even at the same volumes.

[Comga]

Re: SMART reuse...
I'm surprised but pleasantly so - But I remain skeptical that this kind of reuse will provide the cost savings they are hoping for. BUT I am excited to see them trying, and I do wonder how many Vulcans will be launched before they start attempting recovery.

How can ULA “start attempting recovery”?

SpaceX added  restart capability to its engines and tried to “land” an expended booster on the ocean.
When it spun up they added cold gas thrusters in the available volume of the interstage to a booster and grid fins.
Then they added legs to the exterior.
No massive changes were required to the Transporter-Erector
A small amount of rocket payload performance was lost.

Adding even a test version of the inflatable decelerator would seem to take a major modification to the booster. 
Could they fit it in an annular, external collar?

After that, there needs to be room for the severable interface.
ULA is going to either stretch the rocket and its GSE and/or take a hit to the payload.
Giving up performance is not something ULA has done historically.
LOFTID is fine, but until ULA builds some dedicated hardware, it is unclear how ULA could insclude SMART in Vulcan.