Author Topic: ULA and Boeing Unveil the Atlas V Configuration for the CST-100 Starliner  (Read 23279 times)

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5476
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 575
  • Likes Given: 676
My question is, why did the original design of CST-100 with the stumpy adapter pass muster?  They wind tunnel tested it some time back and it was made the baseline design.  What changed, and why?

Maybe the use of the 422 configuration gives them plenty of performance margin, so ULA was free to address a concern they had about Centaur (even though it involved a low probability failure mode) in a way that -- by their calculations -- increased the overall likelihood of mission success.

In that context the first questions to ask might be:

- When was the concern raised?
- Who raised the concern and what response did it get initially?
- Was a decision to proceed without the skirt ever made, and then reversed?
-- sdsds --

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 421
  • Liked: 245
  • Likes Given: 33
Pure speculation:

Perhaps the trajectory design changed after the original OML was baselined, which would have resulted in a different velocity/altitude profile, meaning different shock wave behavior.  If the issue involved is shock wave reattachment, then we're not talking a simple analysis, and I'm not surprised that something like this would show up later in the game.

Wind tunnel testing is helpful, but scaling factors mean that you can't test everything exactly like it will be on the full size vehicle.  Ultimately, some things are different.  To fill in the gaps, you have to do CFD analyses, which are labor intensive and take time.

Offline woods170

  • IRAS fan
  • Senior Member
  • *****
  • Posts: 6945
  • IRAS fan
  • The Netherlands
  • Liked: 2438
  • Likes Given: 738
You're looking at the difference between designing a spacecraft to go on top of a pre-existing launch vehicle design that is outside of the original use concept vs contemporaneous spacecraft and launch vehicle design.
My question is, why did the original design of CST-100 with the stumpy adapter pass muster?  They wind tunnel tested it some time back and it was made the baseline design.  What changed, and why?

 - Ed Kyle

We don't know if it passed muster.  Nothing could have changed, just more testing was done
We know that multiple configurations of the stack have been tested in the wind-tunnel. The one shown in the image up-thread was the original configuration with no solids on the stack. That quickly changed to a configuration with one solid on the stack followed by two solids on the stack. Somewhere in-between CST-100 gained a perforated ring around the SM circumference for aerodynamic stability and finally the skirt was added to have the shockwave re-attach to the stack well below Centaur.
So yeah, likely much more testing done once the first couple of rounds were complete.

I'm not so much troubled by the found aerodynamic stability and aero-acoustic issues, but by the overweight issue. How did Boeing manage to get the spacecraft so much heavier that it required not one but two additional solids on Atlas 5? Haven't heard much details about that.
« Last Edit: 10/18/2016 07:21 AM by woods170 »

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 421
  • Liked: 245
  • Likes Given: 33
Mass issues are common in aerospace.  Even Shuttle had them.  When you design a new vehicle from the ground up, you estimate as best you can, and you include a growth allowance to account for things you haven't designed yet.  You design a big-picture concept, you do some analysis, you refine details, you do more analysis, you add more detailed design, lather, rinse, repeat.

The problem is that every little screw, every little threaded insert, every inch of wire, every wire clip, every thousandth of an inch of wall thickness has mass, and it all adds up.  Now, were your estimates three years ago good enough?  Was your mass growth allowance enough?  What if you identified a problem late in the game that required a redesign?  Is it mass neutral?

Don't forget also that you have to make design trades throughout the development life cycle.  A big one fighting against mass is cost.  Lighter weight components are more expensive to design.  They are more expensive to analyze.  They are more expensive to manufacture.  Decisions have to be made based on these mass-cost trades, but the information you have early in the design stage is different from the information you have later on, so which fixed budget do you blow, your fiscal budget or your mass budget?

Offline SgtPoivre

  • Member
  • Posts: 58
  • Paris - France
  • Liked: 11
  • Likes Given: 0
Mass issues are common in aerospace.  Even Shuttle had them.  When you design a new vehicle from the ground up, you estimate as best you can, and you include a growth allowance to account for things you haven't designed yet.  You design a big-picture concept, you do some analysis, you refine details, you do more analysis, you add more detailed design, lather, rinse, repeat.

The problem is that every little screw, every little threaded insert, every inch of wire, every wire clip, every thousandth of an inch of wall thickness has mass, and it all adds up.  Now, were your estimates three years ago good enough?  Was your mass growth allowance enough?  What if you identified a problem late in the game that required a redesign?  Is it mass neutral?

Don't forget also that you have to make design trades throughout the development life cycle.  A big one fighting against mass is cost.  Lighter weight components are more expensive to design.  They are more expensive to analyze.  They are more expensive to manufacture.  Decisions have to be made based on these mass-cost trades, but the information you have early in the design stage is different from the information you have later on, so which fixed budget do you blow, your fiscal budget or your mass budget?
Also in these early design phases there is a huge pressure to "sell" your concepts and consequently to take optimistic assumptions and minimize your mass margins.
As most conceptual design remain only paper this is usually not an issue!
« Last Edit: 10/18/2016 05:39 PM by SgtPoivre »

Offline Lars-J

  • Senior Member
  • *****
  • Posts: 3159
  • California
  • Liked: 2488
  • Likes Given: 1460
Mass issues are common in aerospace.  Even Shuttle had them.  When you design a new vehicle from the ground up, you estimate as best you can, and you include a growth allowance to account for things you haven't designed yet.  You design a big-picture concept, you do some analysis, you refine details, you do more analysis, you add more detailed design, lather, rinse, repeat.

The problem is that every little screw, every little threaded insert, every inch of wire, every wire clip, every thousandth of an inch of wall thickness has mass, and it all adds up.  Now, were your estimates three years ago good enough?  Was your mass growth allowance enough?  What if you identified a problem late in the game that required a redesign?  Is it mass neutral?

We all know this (or most of us do) - but this is why you always design with mass margins. Clearly something out of the ordinary must have happened here to require *two* solid boosters instead of zero.

And those added booster's aren't "free", they do add to the launch cost and decrease the safety by some amount. Boeing seems happy with the tradeoffs, it seems.
« Last Edit: 10/18/2016 06:44 PM by Lars-J »

Offline SWGlassPit

  • I break space hardware
  • Full Member
  • ****
  • Posts: 421
  • Liked: 245
  • Likes Given: 33
We all know this (or most of us do) - but this is why you always design with mass margins. Clearly something out of the ordinary must have happened here to require *two* solid boosters instead of zero.

And those added booster's aren't "free", they do add to the launch cost and decrease the safety by some amount. Boeing seems happy with the tradeoffs, it seems.

Ultimately, you have to decide what the mass growth allowance is at the beginning of your project.  You have to pick a number, and in the aerospace world, that number unfortunately can't be very big.  In a perfect world, that number is enough.  Things happen during design/analysis cycles.  Surprises show up.  Engineers aren't clairvoyant.

Offline arachnitect

  • Member
  • Full Member
  • ****
  • Posts: 1438
  • Liked: 384
  • Likes Given: 455
I can't remember CST ever being "officially" on Atlas 401/402. When those concepts were going around, they were also circulating artwork with CST on Delta IV M+ and Falcon 9.

I think CST was on 412 when the ULA/Boeing partnership was formalized, and the switch to 422 came not long afterwards.

Offline vapour_nudge

  • Full Member
  • ***
  • Posts: 376
  • Australia
  • Liked: 140
  • Likes Given: 244
I can't remember CST ever being "officially" on Atlas 401/402. When those concepts were going around, they were also circulating artwork with CST on Delta IV M+ and Falcon 9.

I think CST was on 412 when the ULA/Boeing partnership was formalized, and the switch to 422 came not long afterwards.
From the complete opposite perspective, perhaps that played a big part in the choice of the Atlas. If they went with the F9, they can't increase mass past a point. You can add an extra solid or even two to a 412 & get a 422 or 432. Delta would need two or four. The Atlas can be tailored somewhat if mass increases. A bonus over the other LVs
IIRC the Atlas OSIRIS-REx mission started with a 401 & grew to a 411
« Last Edit: 10/19/2016 02:08 AM by vapour_nudge »

Offline Lars-J

  • Senior Member
  • *****
  • Posts: 3159
  • California
  • Liked: 2488
  • Likes Given: 1460
I can't remember CST ever being "officially" on Atlas 401/402. When those concepts were going around, they were also circulating artwork with CST on Delta IV M+ and Falcon 9.

I think CST was on 412 when the ULA/Boeing partnership was formalized, and the switch to 422 came not long afterwards.
From the complete opposite perspective, perhaps that played a big part in the choice of the Atlas. If they went with the F9, they can't increase mass past a point. You can add an extra solid or even two to a 412 & get a 422 or 432.

Little did they know at the time that F9 v1.1 and then v1.2 (and soon v1.3?) would appear...  ;D But yes, you do have a point, "dial-a-rocket" buys you flexibility.

Offline muomega0

  • Full Member
  • ****
  • Posts: 854
  • Liked: 64
  • Likes Given: 1
I can't remember CST ever being "officially" on Atlas 401/402. When those concepts were going around, they were also circulating artwork with CST on Delta IV M+ and Falcon 9.

I think CST was on 412 when the ULA/Boeing partnership was formalized, and the switch to 422 came not long afterwards.
it was unlikely that the 0 SRB version would fly a capsule because of launch capacity, TMK.

Atlas Generic Lift Capacity              the Apollo CM was 5,557kg       
401    411   421    431                    Wiki list Orion at 10,387
4750 5950 6890 7700 (kg)

Edit:  The wiki page is wrong.  The two engine Atlas V has *not* flown.  NROL-35 -->541

Centaur_(rocket_stage)
Quote from: centaurwiki
The Atlas V rocket currently uses the Common Centaur variant.[10] In 2014, on the NROL-35 mission, Atlas V's Common Centaur first flew in a reengined configuration with an RL10-C-1 replacing its previous RL10-A-4-2. This engine is meant to be common between Centaur and the Delta Cryogenic Second Stage to reduce costs.[11][12] RL10-A-4-2 will continue to be used on some future flights. Atlas V launches using the Dual Engine Centaur configuration must use RL10-A-4-2 because the new engine is too wide to accommodate two side-by-side.[12] To date, all Atlas V launches have used the Single Engine Centaur variant, however CST-100 Starliner and Dream Chaser missions will require the dual engine variant, because it allows a "flatter" trajectory safer for aborts.
As on Titan-Centaur, Atlas V 500 launches encapsulate the upper stage inside the payload fairing, to reduce aerodynamic loads. Atlas V 400 flights carry the fairing on top of Centaur, exposing it to the air.

Adding margin(SRBs) helps to obtain future 'market' share. 

"The baseline, at least for NASA missions, is two flights a year for crew,"
said John Mulholland, vice president and program manager of commercial programs at Boeing Space Exploration. "So augmenting that plan with some cargo missions, or extending crew missions beyond NASA, will be important for us and will help drive the overall price down to get that efficiency of scale.

Inside the crew cabin, though, the displays, seats and life support system will be removed and replaced with a cargo pallet able to handle at least 2,500 kg (5,500 lb.) of food, clothing and hardware for the orbiting outpost."

Solids are common with Vulcan v0, so the efforts are more aligned with the current and future LVs, not to mention Cislunar1000.   NASA gets shuttle derived SLS and the to be retired Atlas....
« Last Edit: 10/19/2016 06:58 PM by muomega0 »

Offline Welsh Dragon

  • Full Member
  • *
  • Posts: 176
  • Liked: 164
  • Likes Given: 20
The two engine Centaur first flew in 2014.  All previous launches were the single engine variant.
Incorrect. The majority of Centaurs were two engine flights, starting all the way back in the 60s. The single engine model first flew on Atlas III, in 2000 (?). If you are talking about Atlas V Centaurs, all have been single engine.

Offline muomega0

  • Full Member
  • ****
  • Posts: 854
  • Liked: 64
  • Likes Given: 1
The two engine Centaur first flew in 2014.  All previous launches were the single engine variant.
Incorrect. The majority of Centaurs were two engine flights, starting all the way back in the 60s. The single engine model first flew on Atlas III, in 2000 (?). If you are talking about Atlas V Centaurs, all have been single engine.
Yes, many two engine Centaurs!  Sorry for the confusion.  The wiki page incorrectly states two engine Atlas V flew.   Thanks.
needed to add Atlas V....typed too quick...


The two engine {Atlas V reengined} Centaur first flew in 2014.  All previous {Atlas V }launches were the single engine variant.
Quote from: centaurwiki
The Atlas V rocket currently uses the Common Centaur variant.[10] In 2014, on the NROL-35 mission, Atlas V's Common Centaur first flew in a reengined configuration with an RL10-C-1 replacing its previous RL10-A-4-2. This engine is meant to be common between Centaur and the Delta Cryogenic Second Stage to reduce costs.[11][12] RL10-A-4-2 will continue to be used on some future flights. Atlas V launches using the Dual Engine Centaur configuration must use RL10-A-4-2 because the new engine is too wide to accommodate two side-by-side.[12] To date, all Atlas V launches have used the Single Engine Centaur variant, however CST-100 Starliner and Dream Chaser missions will require the dual engine variant, because it allows a "flatter" trajectory safer for aborts.
As on Titan-Centaur, Atlas V 500 launches encapsulate the upper stage inside the payload fairing, to reduce aerodynamic loads. Atlas V 400 flights carry the fairing on top of Centaur, exposing it to the air.

NASA gets shuttle derived SLS and the to be retired Atlas....

Wrong like most of your posts
Atlas V has yet to fly dual engines. The  Common Centaur variant is a single engine.  All Atlas Centaurs, Atlas I's and Atlas II's had dual engines.  The single engine Centaurs were introduced during Atlas III flights and all Atlas V's are single engine to this point.

And NASA does not get Atlas
Atlas V has yet to fly dual engines.   Got it.    That's not a good thing, right?
When was the last time any dual engine Atlas flew?

True..NASA does not *get* Atlas (nor Vulcan) mixing solids and crew....The reason is that it would require a substantial number of flights to test all the possible flight conditions and its not worth the effort given that LOC is so small during ascent *and* the LV will be retired.   Simply redesign Vulcan v0 without solids and start on the path towards reuse, and only certify Vulcan, not Atlas.   My guess is that ULA would jump at the chance.., no?  Oh..I see now...Vulcan has to have solids so it 'looks like'  Atlas.  Bravo. 

BTW.  Corrections always welcome....

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30805
  • Cape Canaveral Spaceport
  • Liked: 9049
  • Likes Given: 292
Atlas V has yet to fly dual engines.   Got it.    That's not a good thing, right?
When was the last time any dual engine Atlas flew?

True..NASA does not *get* Atlas (nor Vulcan) mixing solids and crew....The reason is that it would require a substantial number of flights to test all the possible flight conditions and its not worth the effort given that LOC is so small during ascent *and* the LV will be retired.   Simply redesign Vulcan v0 without solids and start on the path towards reuse, and only certify Vulcan, not Atlas.   My guess is that ULA would jump at the chance.., no?  Oh..I see now...Vulcan has to have solids so it 'looks like'  Atlas.  Bravo. 

BTW.  Corrections always welcome....

Again, all wrong. and what is your point?

Flying dual engines is not a big deal.  2004 was the last time they flew
There is no need to test all possible flight conditions.  It has never been done for any vehicle, manned or unmanned.
There is no issue with mixing crew and solids
There is no sense in only certifying Vulcan, not Atlas since Atlas will be flying crew for many years before Vulcan is available
and finally, Vulcan doesn't need solids for CST-100
« Last Edit: 10/19/2016 08:04 PM by Jim »

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 7870
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 4688
  • Likes Given: 3169
let's all be excellent to each other.
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Offline FutureSpaceTourist

  • Senior Member
  • *****
  • Posts: 3092
  • UK
    • Plan 28
  • Liked: 2054
  • Likes Given: 696
Quote
Hardware!  Structural test article for #Starliner #Atlas LV adapter

https://twitter.com/torybruno/status/861615045729439744

Offline FutureSpaceTourist

  • Senior Member
  • *****
  • Posts: 3092
  • UK
    • Plan 28
  • Liked: 2054
  • Likes Given: 696
Quote
Hardware!  Structural test article for #Starliner #Atlas LV adapter

https://twitter.com/torybruno/status/861615045729439744

Quote
Great hardware designed by @ulalaunch engineer Ed Walton and built by our excellent Decatur facility!
CC: @barbegan13

https://twitter.com/tylerrogers82/status/861625539114618880

Offline starbase

  • Member
  • Posts: 13
  • Liked: 1
  • Likes Given: 14
I wonder why the Atlas V lifting the Starliner is called the 422 variant, since it doesn't use any payload fairing. Wouldn't be Atlas V 022 more appropriate in this case?

Offline hkultala

  • Full Member
  • ****
  • Posts: 530
  • Liked: 122
  • Likes Given: 43
I wonder why the Atlas V lifting the Starliner is called the 422 variant, since it doesn't use any payload fairing. Wouldn't be Atlas V 022 more appropriate in this case?

5XX means fairing that encapsulates the whole centaur, 4XX means fairing on top centaur, exposing Centaur.

In the Starliner configurtion, the Centaur is not inside a failring, so 4XX.

Offline starbase

  • Member
  • Posts: 13
  • Liked: 1
  • Likes Given: 14
I wonder why the Atlas V lifting the Starliner is called the 422 variant, since it doesn't use any payload fairing. Wouldn't be Atlas V 022 more appropriate in this case?

5XX means fairing that encapsulates the whole centaur, 4XX means fairing on top centaur, exposing Centaur.

In the Starliner configurtion, the Centaur is not inside a failring, so 4XX.

Right, I got the picture. However, according to ULA the first digit refers to the fairing diameter, not if the Centaur is encapsulated or not (which is a natural byproduct of the fairing variant used of course). So if they use no fairing at all it's kind of misleading to designate the variant 422 imho.

Source: http://www.ulalaunch.com/products_atlasv.aspx
« Last Edit: 05/29/2017 02:28 PM by starbase »

Tags: