Author Topic: NASA defends decision to restart RS-25 production, rejects alternatives  (Read 54951 times)

Offline RocketGoBoom

  • Full Member
  • **
  • Posts: 241
  • Idaho
  • Liked: 211
  • Likes Given: 215
What I find disturbing is that the fifth set of engines are only required in 2027, so there will only be four SLS flying in the next 11 years?  :o
I was under the impression that they'd be aiming at a somewhat higher launch cadence.  :-\

I will be shocked if they even need 4 sets of engines for the entire program life.
My bet is 2 sets and done.

Offline Dante80

  • Full Member
  • ****
  • Posts: 873
  • Athens : Greece
  • Liked: 774
  • Likes Given: 491
The baseline is a production rate of 2 engines per year.
Here is the project breakdown according to the JOFOC.

Its unclear how much the per engine cost would be after certification, assuming that a production rate like that holds.


Online AncientU

  • Senior Member
  • *****
  • Posts: 4305
  • Liked: 2578
  • Likes Given: 3604
Will the new production line be able to produce 8 engines per year instead of 0.75 per year?
Is that another billion dollar effort?

Did anyone ever produce 8 RS-25s in a year?
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline Chalmer

  • Member
  • Posts: 94
  • Copenhagen
  • Liked: 27
  • Likes Given: 28
Will the new production line be able to produce 8 engines per year instead of 0.75 per year?
Is that another billion dollar effort?

Did anyone ever produce 8 RS-25s in a year?

I dont know about ever. I think it is highly unlikely that 8 RS-25s will be produced in a year at any point in time.

Quote from: Page 6 in the Justifacation document
Engine production rate shown is set at two engines per year, which is the SLS Program baseline steady state need and therefore establishes the manufacturing infrastructure and labor force size.

So if production need to be higher, more money is needed. But it dosent since steady state is 2 per year. Which also says something about expected long term flight rate i guess.
« Last Edit: 01/07/2016 12:21 PM by Chalmer »

Offline Dante80

  • Full Member
  • ****
  • Posts: 873
  • Athens : Greece
  • Liked: 774
  • Likes Given: 491
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

So, the factory can hold 10/12 engines inside, in various stages of production/delivery.

I might have this wrong.
« Last Edit: 01/06/2016 12:44 PM by Dante80 »

Offline gospacex

  • Senior Member
  • *****
  • Posts: 3017
  • Liked: 524
  • Likes Given: 597
On the positive side, once the US Government pays to restart production of the RS25 as well as updating the design for modern fabrication methods and a fewer part count, it will be another option on the shelf for future launch vehicle designs.  A lot of people have said the RS25 is an amazing engine, it is just expensive.  Well, these investments will eventually make RS25 a lot cheaper to some future user if they step forward.

This "it will be cheaper" thing tends to, you know, fail to happen.

A semi-random example: ULA was supposedly created to make launches cheaper. After which they become more expensive.

Offline Chalmer

  • Member
  • Posts: 94
  • Copenhagen
  • Liked: 27
  • Likes Given: 28
On the positive side, once the US Government pays to restart production of the RS25 as well as updating the design for modern fabrication methods and a fewer part count, it will be another option on the shelf for future launch vehicle designs.  A lot of people have said the RS25 is an amazing engine, it is just expensive.  Well, these investments will eventually make RS25 a lot cheaper to some future user if they step forward.

This "it will be cheaper" thing tends to, you know, fail to happen.

A semi-random example: ULA was supposedly created to make launches cheaper. After which they become more expensive.

Theres that, but also it seems like bad business acumen to invest 0.7-1 Billion to save maybe 20 million per engine on six engines. They have to build 35-50 engines for it to make sense (break even).

Offline notsorandom

  • Full Member
  • ****
  • Posts: 1675
  • Ohio
  • Liked: 370
  • Likes Given: 87
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

So, the factory can hold 10/12 engines inside, in various stages of production/delivery.

I might have this wrong.
A rate of 2 a year is a head scratcher. It literally doesn't add up. One SLS needs 4 engines, they want to get into a launch cadence of at least once a year. A rate of two engines a year means a launch every two years. If the contract really does call for two engines a year then eventually they will have to ramp up capability or reduce the flight rate.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 7726
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2043
  • Likes Given: 4743
Thanks for the article great Chris. :) I find it ironic that NASA has to justify or explain anything about SLS or the RS-25. It doesn't matter, Congress ordered it to be built end of story.... Did we forget how hard Charlie tried to slow walk it in the first place? Flight rate really? NASA will just continue to splice in an Orion/SLS to Mars video promo each time a Commercial Cargo and upcoming Commercial Crew flight to ISS and no one will be the wiser... ;)
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline Dante80

  • Full Member
  • ****
  • Posts: 873
  • Athens : Greece
  • Liked: 774
  • Likes Given: 491
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

So, the factory can hold 10/12 engines inside, in various stages of production/delivery.

I might have this wrong.
A rate of 2 a year is a head scratcher. It literally doesn't add up. One SLS needs 4 engines, they want to get into a launch cadence of at least once a year. A rate of two engines a year means a launch every two years. If the contract really does call for two engines a year then eventually they will have to ramp up capability or reduce the flight rate.

It will be some time before launches catch up though.

The contract seems to be for 6 engines, the first two of which will be delivered for integration in 2022. By the time they are used all (2027), a new contract placed before 2022 would have another 10 engines ready by then (I don't count 2 engines that will remain for spares).

If we assume 1 SLS launch per year (from 2028 onwards), then launches will catch up with production in..2032 (EM-10 or EM-11).
« Last Edit: 01/06/2016 04:00 PM by Dante80 »

Online oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 2821
  • Florida
  • Liked: 1160
  • Likes Given: 76
If SLS operates for as long as STS (30 years) and after 2027 launches at a rate of 2 per year, that is only an additional 20 years and 40 SLS launches (first year is 2017-2018+30 is 2047).

4 engines per launch is only 160 engines total produced over 20 years. At a price of ~$50M per engine that is a total of $8B in engines or $400M per year in contract spending to AJR.

$1.5B over 11 years is only a $136M per year. This is something but not nearly enough to keep AJR afloat all by itself. It helps but without other work AJR could still fold. AJR has been loosing business lately. This is the troubling item not the price or the duration but AJR's health as a business.

Current engine lines of AJR:
RS25 restarting 2/yr at $57M per engine
RS68A ~9/yr dropping to only 3/yr in 2018-2019 at ~$20M per engine
RL-10 ~10/yr increasing by 2/yr for EUS in 2017 and then possibly dropping to only 2 in 2022 if ULA decides to use a different engine for ACES at $10M per engine

Current LV engine revenue = $416M
In 2027 LV engine revenue = $194-294M (depends on ULA's decision on the engine used in ACES, $40M for 4 RL-10s vs $10M for a single BE-3)

This is not encouraging for AJR health in the business of engine manufacture for LVs.

LV engine revenue as a percentage of all AJR business is shrinking.
This could mean that overheads rise making the 2/yr manufacture of RS-25 rise in costs sharply by 2022 when additional engine contracts are made.

Edit: I included RS-68's in 2027. With Vucan/ACES operating the complete DeltaIV line even for DIVH would stop flying in 2025. So the revenue in 2027 for LV engines could be solely from SLS use of 4 RS-25 and 4 RL-10s per flight at $134M /yr for 2 RS-25 and 2 RL-10s. No other US launch vehicle would be using AJF engines. Also the build rate is set at 2/yr because that was the tooling and rates needed to supply STS with new engines for their flight rate of 3 STS per year using 9 engines per year (reuse rate of 5 flights per engine require a build rate of 2 /yr).
« Last Edit: 01/06/2016 04:38 PM by oldAtlas_Eguy »

Online Lars-J

  • Senior Member
  • *****
  • Posts: 3234
  • California
  • Liked: 2539
  • Likes Given: 1510
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

5 years to build an engine!?!?!

Offline Dante80

  • Full Member
  • ****
  • Posts: 873
  • Athens : Greece
  • Liked: 774
  • Likes Given: 491
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

5 years to build an engine!?!?!

From the JOFOC.



ARJ hand builds their engines (at least, that's what I knew they did in the past).

Offline bad_astra

  • Full Member
  • ****
  • Posts: 1772
  • Liked: 192
  • Likes Given: 150
We could be building F1's if we were going to go resurrecting hardware.
"Contact Light" -Buzz Aldrin

Online Lars-J

  • Senior Member
  • *****
  • Posts: 3234
  • California
  • Liked: 2539
  • Likes Given: 1510
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

5 years to build an engine!?!?!

From the JOFOC.



ARJ hand builds their engines (at least, that's what I knew they did in the past).

 :o This is the production process I imagine for a 5 year build: (see image) It's no wonder the surviving RS-25's are treated like precious commodities.

(Meanwhile, another domestic liquid engine producer builds over 100 engines per year)
« Last Edit: 01/06/2016 05:19 PM by Lars-J »

Online oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 2821
  • Florida
  • Liked: 1160
  • Likes Given: 76
To be more precise. One engine takes about 5 years to make. The factory has the capacity to start working on two new engines per year. This means that after 5 years (and assuming that each year you add two new engines to production), 2 engines are delivered for integration. And two for every year after that.

5 years to build an engine!?!?!

From the JOFOC.



ARJ hand builds their engines (at least, that's what I knew they did in the past).

 :o This is the production process I imagine for a 5 year build process: (see image) It's no wonder the surviving RS-25's are treated like precious commodities.

(Meanwhile, another domestic liquid engine produces over 100 engines per year)
In an article several years ago (I could not find it) the wood form used to bend the tubing for the engine bell for RL10 was shown. The RL-10 is still manufactured in almost the same way its was done in the 1960's the RS-25 manufacturing tech is the 1970's. 2015 manufacturing tech is a lot different which includes 3D printing of very complex parts in a highly repeatable level of quality manufacturing. I wonder how many parts (especially those no longer having a supplier) will be manufactured using 3D printing?

Offline shooter6947

  • Member
  • Posts: 86
  • Idaho
  • Liked: 69
  • Likes Given: 188

 :o This is the production process I imagine for a 5 year build: (see image) It's no wonder the surviving RS-25's are treated like precious commodities.


Except that they're not.  5 years hand-building an engine might have even come close to making sense when those engines were being reused and reflown on multiple missions.  Now those precious jewels are going to be dropped into the Atlantic like the Heart of the Ocean from Titanic (spoiler alert).

Offline TrueBlueWitt

  • Space Nut
  • Full Member
  • ****
  • Posts: 1971
  • Mars in my lifetime!
  • DeWitt, MI
  • Liked: 59
  • Likes Given: 67
I'm thinking block 2 should have 3 RS25 and a pair of RTLS FH for boosters.

Offline notsorandom

  • Full Member
  • ****
  • Posts: 1675
  • Ohio
  • Liked: 370
  • Likes Given: 87
It will be some time before launches catch up though.

The contract seems to be for 6 engines, the first two of which will be delivered for integration in 2022. By the time they are used all (2027), a new contract placed before 2022 would have another 10 engines ready by then (I don't count 2 engines that will remain for spares).

If we assume 1 SLS launch per year (from 2028 onwards), then launches will catch up with production in..2032 (EM-10 or EM-11).
There is a mismatch in when SLS is expected to begin a 1+ launch a year cadence. Most preliminary manifests I've seen have the start of that happening in 2021-2024. There are currently enough engines to support 3 SLS flights (with a set of 4 backups). At 2 engines a year starting in 2022 and using a 2023 start date for that launch rate it might looks something like:
2018 SLS-1 (12 engines left)
2022 14 engines
2023 SLS-2 (12 engines left)
2024 SLS-3 (10 engines left)
2025 SLS-4 (8 engines left)
One can see where this is heading. In a short time the surplus of engines dwindles and the launch rate either needs to go down or engine production increased. Starting the 1 launch a year cadence in 2027 only delays this. Furthermore by the end of the 2020s NASA wants to be surge launching at a grater than a once a year average for its Mars plans.

Online rcoppola

  • Full Member
  • ****
  • Posts: 1946
  • USA
  • Liked: 1181
  • Likes Given: 460
These engines are highly complex for a number of reasons. Certainly reuse was one requirement that drove the design but  so was the fact that this engine's operational envelope occupied sea-level all the way through to orbital. This alone drove a number of requirements wrt two separate, independently controlled pre burners, closed-loop among others. (As far as I can tell from reading up on it)

Interestingly, the RS-25 was a jump off from the original J2. And they spent Millions updating that to the J2X. (now mothballed, but I do understand the thrust Vs isp of that decision)  I am a bit concerned that the SLS has been designed around a complex, very expensive sustainer engine. I understand why they did it but always believed that in the background, they would create new requirements to allow the RS25 to be simplified both design and manufacturing wise.

Seeing as now these engines are part of a 2 stage expendable configuration, did the SLS Program Office provide a new set of requirements for the RS-25? (i'm sure they did but to what extent?)

Would it not have been more efficient in the long run to use the RS-25 as a bases to create a less complex and easier to manufacturing engine using today's techniques? Taking a few extra years to re-certify but coming out with a much less complex and easier engine to reproduce would have been worth it IMO.

I could be way off base here but it seems to me that NASA asks what it will cost to do X and then expends the funds accordingly. Did they push back and say something to the effect that AJR needs to spend some time and come up with a proposal that gets them to say...30Million an engine with production taking no longer then 2.5 years per? Or is this program so locked in now, with so much funding that this just gets lost as a rounding error over so many years. I'm an SLS supporter but this just doesn't seem to be the way to initiate a program that is expected to last for 3 decades...

So that's my issue. I'm ok with AJR being sole sourced  since the entire core is built around their engine but did anyone at NASA push back against their response? Is there absolutely no leverage in the system to allow for this outside of opening it up for competition? Or is this really our only option for many, many legacy reasons.
Sail the oceans of space and set foot upon new lands!
www.linkedin.com/in/rvcoppola/

Tags: