Vulcan SMART Reuse

[jimvela]

f the few cases where a customer has explicitly requested re-usability, SMART does not help.  There are no fight-proven tanks in SMART, since they are new for every mission.

True, but that's not even the worst part.
The tanks are chump change compared to the rest of the re-integration of used engines back into a flight ready stage.

Re-use of expensive engines (in particular, for expensive engines) has to be in a system that re-flies the integrated system.

SpaceX has this in great shape with Falcon.  Likely much better if SH/Starship work out.

BO has a better idea about the reuse economics with New Glenn than the absolute pig that is SMART. IF they can get New Glenn to fly at better than a snails' pace and without extensive redesign after they enter the learning (launching) phase.

ULA / SMART , bad as it is, is still a better idea than the whole rest of the expendable only world.

[laszlo]

f the few cases where a customer has explicitly requested re-usability, SMART does not help.  There are no fight-proven tanks in SMART, since they are new for every mission.

...Re-use of expensive engines (in particular, for expensive engines) has to be in a system that re-flies the integrated system.

Why?  If the cost of re-integration with a new vehicle is low compared to the cost of fabricating the engines from scratch, wouldn't engine-only re-use actually make more sense for expensive engines than cheap ones?

[dglow]

f the few cases where a customer has explicitly requested re-usability, SMART does not help.  There are no fight-proven tanks in SMART, since they are new for every mission.

...Re-use of expensive engines (in particular, for expensive engines) has to be in a system that re-flies the integrated system.

Why?  If the cost of re-integration with a new vehicle is low compared to the cost of fabricating the engines from scratch, wouldn't engine-only re-use actually make more sense for expensive engines than cheap ones?

From all accounts the BE-4s SMART would recover cost ULA less than the RL-10s they always expend.

[laszlo]

f the few cases where a customer has explicitly requested re-usability, SMART does not help.  There are no fight-proven tanks in SMART, since they are new for every mission.

...Re-use of expensive engines (in particular, for expensive engines) has to be in a system that re-flies the integrated system.

Why?  If the cost of re-integration with a new vehicle is low compared to the cost of fabricating the engines from scratch, wouldn't engine-only re-use actually make more sense for expensive engines than cheap ones?

From all accounts the BE-4s SMART would recover cost ULA less than the RL-10s they always expend.

How many RL-10s would a re-usable Centaur cost?

That still doesn't answer the question of why re-use of expensive engines has to be in a system that re-flies the integrated system.

[dglow]

f the few cases where a customer has explicitly requested re-usability, SMART does not help.  There are no fight-proven tanks in SMART, since they are new for every mission.

...Re-use of expensive engines (in particular, for expensive engines) has to be in a system that re-flies the integrated system.

Why?  If the cost of re-integration with a new vehicle is low compared to the cost of fabricating the engines from scratch, wouldn't engine-only re-use actually make more sense for expensive engines than cheap ones?

From all accounts the BE-4s SMART would recover cost ULA less than the RL-10s they always expend.

How many RL-10s would a re-usable Centaur cost?

ULA have never discussed a reusable Centaur, and balloon tanks don’t seem fit for the task. From what I’ve read $25M is the starting price for an RL-10.

[TrevorMonty]

ULA have never discussed a reusable Centaur, and balloon tanks don’t seem fit for the task. From what I’ve read $25M is the starting price for an RL-10.

If I'm RIGHT that is NASA price for SLS from long time ago. The price ULA pays for latest version on Vulcan is confidential, good luck finding a price but i doubt it is >$5M.

[dglow]

ULA have never discussed a reusable Centaur, and balloon tanks don’t seem fit for the task. From what I’ve read $25M is the starting price for an RL-10.

If I'm RIGHT that is NASA price for SLS from long time ago. The price ULA pays for latest version on Vulcan is confidential, good luck finding a price but i doubt it is >$5M.

Interesting. Why do you suspect an RL-10 for ULA will cost 20% of what NASA paid?

[JEF_300]

ULA have never discussed a reusable Centaur, and balloon tanks don’t seem fit for the task. From what I’ve read $25M is the starting price for an RL-10.

If I'm RIGHT that is NASA price for SLS from long time ago. The price ULA pays for latest version on Vulcan is confidential, good luck finding a price but i doubt it is >$5M.

Interesting. Why do you suspect an RL-10 for ULA will cost 20% of what NASA paid?

I mean, NASA did order an entirely new variant of the engine with the intention of only using it on like 7 launches. Granted, that's 4 engines per flight, but by the end of next year ULA will probably have already flown more RL10s on Vulcan-Centaur than will ever fly on EUS. So it was always gonna be at least a bit more expensive.

[dglow]

ULA have never discussed a reusable Centaur, and balloon tanks don’t seem fit for the task. From what I’ve read $25M is the starting price for an RL-10.

If I'm RIGHT that is NASA price for SLS from long time ago. The price ULA pays for latest version on Vulcan is confidential, good luck finding a price but i doubt it is >$5M.

Interesting. Why do you suspect an RL-10 for ULA will cost 20% of what NASA paid?

I mean, NASA did order an entirely new variant of the engine with the intention of only using it on like 7 launches. Granted, that's 4 engines per flight, but by the end of next year ULA will probably have already flown more RL10s on Vulcan-Centaur than will ever fly on EUS. So it was always gonna be at least a bit more expensive.

Good points. But sufficient for 80% savings?

[TrevorMonty]

This was posted by Tory years ago. RD180 is $25m so use that as point of reference for booster cost. Unfortunately no comparison between 1st and 2nd stage costs.   If RL10 is close to $25m they must be using lot gold in the Avionics. Vulcan RL10 will be considerably cheaper as it uses lot of 3D printing and they are ordering a lot more. Tank construction has also been automated.
https://pbs.twimg.com/media/B8xjOj-CEAAk5lx.jpg:orig

[tbellman]

Good points. But sufficient for 80% savings?

My understanding is that the RL10C-3 is still assembled and brazed manually.  Lot's of touch labour.

And importantly, consider the full situation:

1. Customer says "we want, no, need you to develop, qualify, test, and build this special-purpose version of your engine."

2. Customer's project doesn't care about costs; their financer will just appropriate more money from the taxpayers.

3. Supplier has a virtual monopoly on the kind of engines requested (US-made high-I_sp hydrolox).

As supplier, you can then set basically any price you want.  If the customer thinks the price is not reasonable, then you just gently remind them that you are under no obligation to give an offer at all if they insist on a lower price, but they are welcome to turn to your non-existing competition.

[dglow]

Good points. But sufficient for 80% savings?

My understanding is that the RL10C-3 is still assembled and brazed manually.  Lot's of touch labour.

I've read the Centaur's V's RL-10CX described as being "partially 3D-printed."

And importantly, consider the full situation:

1. Customer says "we want, no, need you to develop, qualify, test, and build this special-purpose version of your engine."

2. Customer's project doesn't care about costs; their financer will just appropriate more money from the taxpayers.

3. Supplier has a virtual monopoly on the kind of engines requested (US-made high-I_sp hydrolox).

As supplier, you can then set basically any price you want.  If the customer thinks the price is not reasonable, then you just gently remind them that you are under no obligation to give an offer at all if they insist on a lower price, but they are welcome to turn to your non-existing competition.

This is all true and why I doubt that, even for a volume order, these engines are selling for $5M. I'd put them nowhere below $15M.

[tbellman]

My understanding is that the RL10C-3 is still assembled and brazed manually.  Lot's of touch labour.

I've read the Centaur's V's RL-10CX described as being "partially 3D-printed."

Yes.  They are different engines.  One costs a lot more to build than the other, and NASA has chosen to expensive one for EUS.

As supplier, you can then set basically any price you want.  If the customer thinks the price is not reasonable, then you just gently remind them that you are under no obligation to give an offer at all if they insist on a lower price, but they are welcome to turn to your non-existing competition.

This is all true and why I doubt that, even for a volume order, these engines are selling for $5M. I'd put them nowhere below $15M.

NASA is buying a unique, and expensive version, developed especially for them, in low quantity.  They are in a very bad bargaining position, and have repeatedly shown that they don't even care^† about how much money is shoveled to their suppliers and contractors for this project (SLS).

ULA is buying a different version, one that is designed to be much cheaper to manufacture, in larger quantities.  They also have the very credible threat that if they don't get a good price from Aerojet-Rocketdyne, they won't be able to compete on the launch market and will then go out of business, and AJR will lose all sales to ULA.

Why would you think there is not a huge difference in the price AJR charges NASA and ULA for the two different engines?

(I myself have experienced more than one case where suppliers who felt secure in their position overcharged us.  Not until we a) showed that we knew what the competition was charging, and b) made very clear we were willing to go through the pains of changing to those other suppliers (government procurement rules can be a pain sometimes), did they drop their prices.  In one case by a factor four.)


† _{GAO/OIG^‡ complains that NASA's accounting obscures the SLS costs; NASA refuses to change.  GAO/OIG complains that contractor performance is poor, but that NASA even so awarded them maximum allowed bonus; NASA ignores the complaints.}

‡ _{I don't remember which of GAO or OIG that have made these complaints.}

[whitelancer64]

In 2015-2018, ULA was considering using the BE-3U engine for ACES, which eventually was cut down to the Centaur V. So presumably, the RL-10CX is cheaper than the BE-3U price Blue Origin was pitching ULA at the time.

[dglow]

Why would you think there is not a huge difference in the price AJR charges NASA and ULA for the two different engines?

Great points, all of them. But don't turn around the argument, my case is that @TrevorMonty's assertion of "<$5M" per Centaur V RL-10C is far off-base.

What do you think?

[FutureSpaceTourist]

When will Atlas retire? Are you confident SMART reuse will survive the first attempt and will that engine section be reused?

https://twitter.com/torybruno/status/1834637807534391649

No. The plan is for the first flight test to be all about verifying a clean separation. The first recovered engine will be cut up to understand its robustness.

[TrevorMonty]

When will Atlas retire? Are you confident SMART reuse will survive the first attempt and will that engine section be reused?

https://twitter.com/torybruno/status/1834637807534391649

No. The plan is for the first flight test to be all about verifying a clean separation. The first recovered engine will be cut up to understand its robustness.

Given they are testing separate engine pod may not have HIAD. Really don't won't to spend money on HIAD if separation is in doubt.

Development process is likely to follow F9R and Electron step by step approach.

[meekGee]

Given they are testing separate engine pod may not have HIAD. Really don't won't to spend money on HIAD if separation is in doubt.

Development process is likely to follow F9R and Electron step by step approach.

And that right there is why they are where they are.

Someone who's acting like their livelihood depends on moving fast (aka "urgency") - that's exactly what they don't do.  Even for something as limited as SMART, they assume separation can be made to work, so absolutely develop and install HIAD even before separation is proven.

That's why Starship is flying before the heat shield is figured out, and even NG is trying to land on the first try.

ULA is just comatose in comparison.

[TrevorMonty]

When will Atlas retire? Are you confident SMART reuse will survive the first attempt and will that engine section be reused?

https://twitter.com/torybruno/status/1834637807534391649

No. The plan is for the first flight test to be all about verifying a clean separation. The first recovered engine will be cut up to understand its robustness.

Given they are testing separate engine pod may not have HIAD. Really don't won't to spend money on HIAD if separation is in doubt.

Development process is likely to follow F9R and Electron step by step approach.

And that right there is why they are where they are.

Someone who's acting like their livelihood depends on moving fast (aka "urgency") - that's exactly what they don't do.  Even for something as limited as SMART, they assume separation can be made to work, so absolutely develo.amd install HIAD even before separation is proven.

Based on your comment SpaceX should've put legs on first F9 V1.1. Catch SS booster on last flight.

[meekGee]

Given they are testing separate engine pod may not have HIAD. Really don't won't to spend money on HIAD if separation is in doubt.

Development process is likely to follow F9R and Electron step by step approach.

And that right there is why they are where they are.

Someone who's acting like their livelihood depends on moving fast (aka "urgency") - that's exactly what they don't do.  Even for something as limited as SMART, they assume separation can be made to work, so absolutely develo.amd install HIAD even before separation is proven.

Based on your comment SpaceX should've put legs on first F9 V1.1. Catch SS booster on last flight.

No, you're taking my statement to an absurd (as usual) so you can argue with a strawman.

There's a difference between putting the legs on as soon as they're ready (even before reentry is proven) to holding the first flight until all pieces (including legs) are ready.

Holding back HIAD until separation is proven, "to save money" is not even financially responsible, unless your planning horizon is only end-of-quarter.