Reuse business case

[TrevorMonty]

ULA has been working on SMART for a long time. I would be really surprised if SMART was not considered in the overall design. I think the boattail is bolted to the tank, thus making it as simple as replacing those bolts with frangible ones. They probably left some annular space to put the HIAD and such.

Not only do they need space for the aeroshell and it's hardware, but in order to separate the engine section from the tank section you not only have to sever the structural elements between the engine section and the tanks, but the electrical and all the propellant pipes.

SpaceX doesn't have to worry about separating engines from tanks because they bring back the entire stage, but just for the propellant pipes, ULA will have to squeeze in hardware to allow for disconnects.

So it doesn't sound simple to me.
...

Not saying it's a walk in the park, but the whole engine separation thing was done in the late 50's on the original Atlas - while the center engine was still firing.

Right, early Atlas were informally classified as a "stage-and-a-half" rockets, since the "half" were 2 × LR89 booster engines that were dropped along with the bottom of the Atlas rocket, leaving just the sustainer engine which had been started on the ground - this was because they hadn't perfected air-start engines yet, but did soon after.

And lots of rockets drop boosters, but you are right that the early Atlas dropped more than the booster engines, and those booster engines used the same propellant that the sustainer engine used.

Color me skeptical that ULA has designed the current Vulcan with all the options needed to start testing dropping the engine section. And if they haven't, then that means they have to redesign (and certify) that new engine section - then they have to find customers willing to fly on those test flights.

Because, again, the reusability features of the Falcon 9 were not specifically related to launching a Falcon 9 payload, so SpaceX didn't have to re-certify the Falcon 9 for each reusable test version. ULA might need to if they don't get reusable Vulcan right on the first try.

I'm not an engineer, so just my $0.02...

There wont be problem building and flying certified expendable and SMART version side by side. SMART version should be low risk with Amazon likely to be first customer given extra launch discount.  LV failure isn't going set them back much given how many satellites they will be making.

With RLVs most failure risks are in recovery  which doesn't affect customer's payloads.

Sent from my SM-G570Y using Tapatalk

[darkenfast]

Of course the problem, once again, is that after several years of expensive effort, SMART will be at about the same (or maybe inferior) place that SpaceX's re-flyable Falcon first stages have been at for several years now. And - SpaceX will be moving on to a new, larger vehicle that re-uses BOTH stages and can carry a lot more.

Of course, Starship has to work, but I would suggest that praying for SpaceX to fail is not a great business plan.

The company that succeeds in being real competition for SpaceX will have to be highly innovative and fast.

[Surfdaddy]

I'm curious - where did the idea of SMART come from?

Was originally proposed for Atlas in 2008 but never went anywhere. At that time HIAD was work in progress by NASA with no flight heritage. When SMART was proposed for Vulcan NASA had completed 2 successful suborbital flight demonstrates of HIAD. Later this year ULA and NASA will do orbital reentry demo of HIAD.


Sent from my SM-G570Y using Tapatalk

So also it was proposed even before F9 had ever flown, even expendably. So SMART probably seemed like a state of the art way to recover expensive components. The problem is, evolution has sort of passed that idea by, and it's not a very superior idea any longer.

[DanClemmensen]

I'm curious - where did the idea of SMART come from?

Was originally proposed for Atlas in 2008 but never went anywhere. At that time HIAD was work in progress by NASA with no flight heritage. When SMART was proposed for Vulcan NASA had completed 2 successful suborbital flight demonstrates of HIAD. Later this year ULA and NASA will do orbital reentry demo of HIAD.


Sent from my SM-G570Y using Tapatalk

So also it was proposed even before F9 had ever flown, even expendably. So SMART probably seemed like a state of the art way to recover expensive components. The problem is, evolution has sort of passed that idea by, and it's not a very superior idea any longer.

Among other reasons, the cost of engines relative to the cost of the rest of a booster has reduced, so the relative benefit of just recovering the engines has gone down. This will vary depending on the total cost of the engines. The Vulcan needs 2x BE-4, while the F9 needs 9x Merlin. Does anyone have cost estimates?

[freddo411]

I'm curious - where did the idea of SMART come from?

Was originally proposed for Atlas in 2008 but never went anywhere. At that time HIAD was work in progress by NASA with no flight heritage. When SMART was proposed for Vulcan NASA had completed 2 successful suborbital flight demonstrates of HIAD. Later this year ULA and NASA will do orbital reentry demo of HIAD.

So also it was proposed even before F9 had ever flown, even expendably. So SMART probably seemed like a state of the art way to recover expensive components. The problem is, evolution has sort of passed that idea by, and it's not a very superior idea any longer.

Among other reasons, the cost of engines relative to the cost of the rest of a booster has reduced, so the relative benefit of just recovering the engines has gone down. This will vary depending on the total cost of the engines. The Vulcan needs 2x BE-4, while the F9 needs 9x Merlin. Does anyone have cost estimates?

The "smart" concept seems to have high built in costs for the recovery just as the costs for F9 booster landing at sea are high.    But the benefits of recovery seem to be inferior to recovering a full booster.

Contrast this with the design of Neutron and starship, where the recovered booster is optimized to have no parts falling into the ocean, or landing out at sea.

Hard to compete and win with a flawed concept.

I will say it again.   I don't think ULA will develop SMART, despite all the PR hype.

[DanClemmensen]

Among other reasons, the cost of engines relative to the cost of the rest of a booster has reduced, so the relative benefit of just recovering the engines has gone down. This will vary depending on the total cost of the engines. The Vulcan needs 2x BE-4, while the F9 needs 9x Merlin. Does anyone have cost estimates?

The "smart" concept seems to have high built in costs for the recovery just as the costs for F9 booster landing at sea are high.    But the benefits of recovery seem to be inferior to recovering a full booster.

Contrast this with the design of Neutron and starship, where the recovered booster is optimized to have no parts falling into the ocean, or landing out at sea.

Hard to compete and win with a flawed concept.

I will say it again.   I don't think ULA will develop SMART, despite all the PR hype.

I guess that your qualitative analysis is correct, but engineers and accountants are supposed to work with numbers. Fundamentally, if the fully-loaded cost of refurbished or (SMART re-engined) booster is lower than the fully-loaded cost of a new booster, then reuse makes sense. Otherwise not.  For example, for Falcon, you need to look at the cost of trucking a new booster across the country versus the cost of catching a flown booster on a barge and returning it to port.

[freddo411]

Among other reasons, the cost of engines relative to the cost of the rest of a booster has reduced, so the relative benefit of just recovering the engines has gone down. This will vary depending on the total cost of the engines. The Vulcan needs 2x BE-4, while the F9 needs 9x Merlin. Does anyone have cost estimates?

The "smart" concept seems to have high built in costs for the recovery just as the costs for F9 booster landing at sea are high.    But the benefits of recovery seem to be inferior to recovering a full booster.

Contrast this with the design of Neutron and starship, where the recovered booster is optimized to have no parts falling into the ocean, or landing out at sea.

Hard to compete and win with a flawed concept.

I will say it again.   I don't think ULA will develop SMART, despite all the PR hype.

I guess that your qualitative analysis is correct, but engineers and accountants are supposed to work with numbers. Fundamentally, if the fully-loaded cost of refurbished or (SMART re-engined) booster is lower than the fully-loaded cost of a new booster, then reuse makes sense. Otherwise not.  For example, for Falcon, you need to look at the cost of trucking a new booster across the country versus the cost of catching a flown booster on a barge and returning it to port.

True.   But accountants and CEOs know that going from a company that builds and launches expendables, to a company that develops, tests, flys, recovers and refurbishes/remanufactures boosters requires considerable hiring, retraining, and investment in physical infrastructure.   Elon has tweeted that reuse development cost something like 1 billion dollars.

So as you say, if a recovered set of BE-4 engines costs $X and a new set costs $2X or $3X it might makes sense to recover them.   But I don't see the parents of ULA allowing for a billion dollar investment in "smart" reuse.    A steel man version of their logic might go like this:   

* We don't think we will save 1 billion dollars over the life of Vulcan
* We think we can charge the full cost of the Vulcan to most of our customers.   
* A lower cost Vulcan will end up having a lower price to our customers, our profit per launch won't change
* Our customers prefer/demand first time use rockets
* We estimate we need a fleet of 30 engines to service our yearly launch rate.   Not very many fewer than if we expend
* We can save 100s of millions of dollars by NOT hiring more engineers to work on "smart"

So, disclaimer here, this is my estimation.   I could well be wrong.   The case for not doing "smart" is strongest when the flight manifest is smaller.   As of right now, the manifest looks much larger than it did in the last few years.    We shall see how things turn out.

[deadman1204]

True.   But accountants and CEOs know that going from a company that builds and launches expendables, to a company that develops, tests, flys, recovers and refurbishes/remanufactures boosters requires considerable hiring, retraining, and investment in physical infrastructure.   Elon has tweeted that reuse development cost something like 1 billion dollars.

So as you say, if a recovered set of BE-4 engines costs $X and a new set costs $2X or $3X it might makes sense to recover them.   But I don't see the parents of ULA allowing for a billion dollar investment in "smart" reuse.    A steel man version of their logic might go like this:   

* We don't think we will save 1 billion dollars over the life of Vulcan
* We think we can charge the full cost of the Vulcan to most of our customers.   
* A lower cost Vulcan will end up having a lower price to our customers, our profit per launch won't change
* Our customers prefer/demand first time use rockets
* We estimate we need a fleet of 30 engines to service our yearly launch rate.   Not very many fewer than if we expend
* We can save 100s of millions of dollars by NOT hiring more engineers to work on "smart"

So, disclaimer here, this is my estimation.   I could well be wrong.   The case for not doing "smart" is strongest when the flight manifest is smaller.   As of right now, the manifest looks much larger than it did in the last few years.    We shall see how things turn out.

Recover is about far more than just saving money. In the past, ULA ignored the other advantages when they weren't needed. A major one is launch cadence. ULA is looking at SMART again because it will help support the much higher launch cadence the amazon contract will allow. Look at Faclon 9, ignoring money spaceX can turn around a booster much faster than it can make a new one from scratch.

[JayWee]

* Our customers prefer/demand first time use rockets

https://twitter.com/Dr_ThomasZ/status/1512459452938534916

[meekGee]

Of course the problem, once again, is that after several years of expensive effort, SMART will be at about the same (or maybe inferior) place that SpaceX's re-flyable Falcon first stages have been at for several years now. And - SpaceX will be moving on to a new, larger vehicle that re-uses BOTH stages and can carry a lot more.

Of course, Starship has to work, but I would suggest that praying for SpaceX to fail is not a great business plan.

The company that succeeds in being real competition for SpaceX will have to be highly innovative and fast.

Yup exactly.
SMART's biggest risk is that it works as advertised.

ULA needs an existential crisis in order to make some really hard decisions and all it's getting is measures that prolong the saga.

[JayWee]

Also, SpaceX/BO/RocketLab approach - land the whole thing - is ultimately scalable. Falcon 9 experience enabled Starship.
SMART is not.

[DanClemmensen]

Recover is about far more than just saving money. In the past, ULA ignored the other advantages when they weren't needed. A major one is launch cadence. ULA is looking at SMART again because it will help support the much higher launch cadence the amazon contract will allow. Look at Faclon 9, ignoring money spaceX can turn around a booster much faster than it can make a new one from scratch.

If you are an accountant, you convert this cadence into money for the comparison. If you can manufacture and deliver boosters fast enough at lower cost than recovery, then recovery is uneconomic. In fact, higher production tends to reduce unit cost.  I'm  not arguing against reuse: F9 shows that it can be cost-effective, but it comes down to money.

[baldusi]

The problem with SMART is that it might allow Vulcan to stay competitive through 2027/8, a most. But ESA has made it clear that Ariane 6 is an intermediate step to a reusable Ariane 7. New Glen is partly reusable with enough headroom for full reusability. SpaceX has a partly reusable F9 and is working on a fully reusable StarShip. RocketLabs is working on the partially reusable Neutron. The Chinese are working towards partial reusability of the LM-8 at first.
Please remember that launch contracts are usually done 18 to 36 months before launch. And they have a full manifest until 2026, at the very least.  But for the 2027~30 launches, they must start selling them by 2025, a period when StarShip and New Glenn should be kinda proven (if they do work). So, they could have a couple of "bad" years, but by 2028/30, they will need to be fielding a new launcher or an evolution of Vulcan with fully reusable first stage (plus a plan for a reusable second).
So ULA must have a plan to have a partly reusable rocket with headroom for full reusability by 2028/30. I'm pretty sure they are doing the trades with a small team, let's say 10 people. Since they know they have to commit to an architecture by 2025 at the latest.

[JayWee]

So ULA must have a plan to have a partly reusable rocket with headroom for full reusability by 2028/30. I'm pretty sure they are doing the trades with a small team, let's say 10 people. Since they know they have to commit to an architecture by 2025 at the latest.

But does it make sense to compete against NG? When they will use BO engines?

[spacenut]

A few years ago, BE-4 was supposed to cost $6 million each and Merlin was supposed to be around $1 million each.  So, that is $9 million vs. $12 million.  Seems to me that would make Smart reuse pay for itself.  Using a helicopter might cost a little more in fuel, but faster in time back to land.  So, it might be worth it for Blues constellation as well as their other contracted launches.  Hopefully they will develop it. 

[Surfdaddy]

The problem with SMART is that it might allow Vulcan to stay competitive through 2027/8, a most. But ESA has made it clear that Ariane 6 is an intermediate step to a reusable Ariane 7. New Glen is partly reusable with enough headroom for full reusability. SpaceX has a partly reusable F9 and is working on a fully reusable StarShip. RocketLabs is working on the partially reusable Neutron. The Chinese are working towards partial reusability of the LM-8 at first.
Please remember that launch contracts are usually done 18 to 36 months before launch. And they have a full manifest until 2026, at the very least.  But for the 2027~30 launches, they must start selling them by 2025, a period when StarShip and New Glenn should be kinda proven (if they do work). So, they could have a couple of "bad" years, but by 2028/30, they will need to be fielding a new launcher or an evolution of Vulcan with fully reusable first stage (plus a plan for a reusable second).
So ULA must have a plan to have a partly reusable rocket with headroom for full reusability by 2028/30. I'm pretty sure they are doing the trades with a small team, let's say 10 people. Since they know they have to commit to an architecture by 2025 at the latest.

So are you saying that the horse might be about to leave the barn?

[baldusi]

So are you saying that the horse might be about to leave the barn?

I'm just saying that ULA is composed of some of the most knowledgeable managers and engineering on the world of aerospace. And they can see the writing on the wall. They know their market and their customers. And the did noticed how NASA, DoD, NRO and all relevant clients have accepted flying on reused boosters after about five years of reuse operations. So, the future is pretty clear.
They might have a huge liability for taking the hard and audacious decisions that need to be made, because of their parent companies. But neither is really ready to compete in a commercial market (just watch Boeing Space and Launch performance with SLS and Starliner). Their only chance to stay relevant is that ULA succeeds. And the recent Kuiper order might have give them enough revenue to self finance such development.
Please not that at this time having a fully reusable booster is de rigueur, but the upper stage is still being traded all the way. Look at Neutron vs StarShip vs NewGlenn, and we still don't know that the European, Japanese, Chinese and the rest are trading.
So ULA will probably have a follower approach, where they take the best practices and apply them to a completely new design. Their MLP based launch site gives them a lot of flexibility. They have already mastered all relevant propellants, and with SMART they will learn a lot about atmospheric interface. In other words, they do have a chance.
I'm not saying it's simple nor that they will succeed. But as a follower who caters to premium clientele, they do have a fighting chance.

[su27k]

I'm just saying that ULA is composed of some of the most knowledgeable managers and engineering on the world of aerospace. And they can see the writing on the wall.

I'm not sure they can. To fix the problem you first have to admit you have a problem, I see no sign that they think being expendable is a problem. A good first step would be recant the Sowers excel and the whole "first stage needs 10 flights to break even" non-sense, i.e. the thing started this thread in the first place, they're not doing that. In fact I think the Amazon deal would embolden them to stick to "expendable is best" camp, from their POV they won that with expendable, and reusable Falcon 9 didn't get any launches.

[Surfdaddy]

I'm just saying that ULA is composed of some of the most knowledgeable managers and engineering on the world of aerospace. And they can see the writing on the wall.

I'm not sure they can. To fix the problem you first have to admit you have a problem, I see no sign that they think being expendable is a problem. A good first step would be recant the Sowers excel and the whole "first stage needs 10 flights to break even" non-sense, i.e. the thing started this thread in the first place, they're not doing that. In fact I think the Amazon deal would embolden them to stick to "expendable is best" camp, from their POV they won that with expendable, and reusable Falcon 9 didn't get any launches.

Tory Bruno is an engineer and a very smart guy. He will understand the situation well. The question is how much support he can drum up from his parent companies.

[Coastal Ron]

I'm just saying that ULA is composed of some of the most knowledgeable managers and engineering on the world of aerospace. And they can see the writing on the wall.

I'm not sure they can. To fix the problem you first have to admit you have a problem, I see no sign that they think being expendable is a problem. A good first step would be recant the Sowers excel and the whole "first stage needs 10 flights to break even" non-sense, i.e. the thing started this thread in the first place, they're not doing that. In fact I think the Amazon deal would embolden them to stick to "expendable is best" camp, from their POV they won that with expendable, and reusable Falcon 9 didn't get any launches.

Tory Bruno is an engineer and a very smart guy. He will understand the situation well. The question is how much support he can drum up from his parent companies.

Unfortunately while Tory Bruno may hold the title of President and Chief Executive Officer (CEO) of United Launch Alliance, he does not control what gets funded by ULA - ULA's parents control that money.

I would agree that ULA has great engineers, and I cite their papers and presentations frequently. I truly believe that if Congress would have directed NASA to return to the Moon back in 2010, didn't impose the SLS and Orion MPCV, and allowed NASA to solicit the best industry ideas, that ULA could have been a big contributor to that effort and we could have been back on the Moon by now.

But great ideas don't get funded, great potential businesses do. And it is not clear to me that implementing SMART on Vulcan makes good business sense.