Author Topic: Atlas V and Centaur Q&A  (Read 212196 times)

Offline Kayla

  • Extreme Veteran
  • Full Member
  • ***
  • Posts: 396
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #40 on: 02/06/2007 12:51 AM »
Quote
bombay - 4/2/2007  11:24 PM

What of the possibility of using the current Delta II 2nd and 3rd stages on Atlas V with a modified interstage adapter?  Is that being considered?  What am I missing in my thought process?    

Prior to ULA the Atlas and Delta programs were in staunch competition, hindering such combinations.  One of the many benefits of ULA are the numerous potential combinations (such as your suggestion) that are now available.

Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #41 on: 02/06/2007 01:45 AM »
Quote
Kayla - 4/2/2007  7:32 PM

From its inception WBC was freed of this limitation.  Originally the assumption was that the tank would be built at Michoud where three 5.4m demonstration hoops were assembled and lap welded together to demonstrate the planned construction using thin (0.080") al-li sheet stock.  With the formation of ULA, ULA now has the option of using either Michoud or Decatur for construction.  Either location has the ability to friction stir weld large diameter tanks and then ship the final structures to the Cape or Vandenberg.

WBC is designed to merge the extremely efficient monocoque, common bulkhead tank from the historic Centaur with modern materials and friction stir welding spear headed for space application by the external tank and Delta programs.  This combination, along with increased propellant load allows the WBC to increase the mass fraction from the current Centaur's 0.91 to 0.94!  The following paper provides additional information.
http://www.lockheedmartin.com/data/assets/12382.pdf
Very interesting!  I'm surprised that lap joints are used given that they are analyzed in shear, not tension or compression where the allowables are much higher.

I'm guessing that the shear allowable for Al-Li is comparable to other 2XXX series aluminums: Fsu = 35 ksi and the welded allowable: Fsu = 20 ksi.  Realizing that fric. stir welding produces excellent as welded allowables, I would guess that in an analysis, 80% of 35 ksi would conservatively be used.  Lets say 30 ksi is used for the as-welded shear allowable.

The common bkhd concept would utilize a delta p between the LH2 tank and LOX tank.  Lets say maximum flight pressure in the tank is 50 psi (this is probably low).  So here's a simple analysis using ONLY pressure loading.  

I'll spare everybody the math, but the longitudinal line load would be 1800 lb/in and the stress would be 22.5 ksi using .080" mat'l.  Using a man-rated f.s. of 1.4, the m.s. would be -0.05 (ult).

Throw in S.V. bearing loads, interstage adapter, fairing, and bending loads and all other dynamic and transient axial loading, you could see where the guage thickness would need to be much thicker, which in-turn would alter mass ratio.

The .94 appears to be a goal, not a given.  




Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30380
  • Cape Canaveral Spaceport
  • Liked: 8691
  • Likes Given: 283
Re: Atlas V and Centaur Q&A
« Reply #42 on: 02/06/2007 01:49 AM »
Quote
bombay - 5/2/2007  9:45 PM

The common bkhd concept would utilize a delta p between the LH2 tank and LOX tank.  Lets say maximum flight pressure in the tank is 50 psi (this is probably low).  So here's a simple analysis using ONLY pressure loading.  


That is too high

Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #43 on: 02/06/2007 02:12 AM »
Quote
Jim - 5/2/2007  8:49 PM

Quote
bombay - 5/2/2007  9:45 PM

The common bkhd concept would utilize a delta p between the LH2 tank and LOX tank.  Lets say maximum flight pressure in the tank is 50 psi (this is probably low).  So here's a simple analysis using ONLY pressure loading.  


That is too high
Well what is it?

Offline Antares

  • ABO^2
  • Senior Member
  • *****
  • Posts: 5190
  • Done arguing with amateurs
  • Liked: 342
  • Likes Given: 208
Re: Atlas V and Centaur Q&A
« Reply #44 on: 02/06/2007 02:48 AM »
Quote
Okay to GTO, Atlas 552 with WBC is 12.9 mT. Delta Heavy is 13.1 mT. Close but Delta Heavy still has a slight 200 kg edge.

I know for the baseline Centaur, xx2 is not optimal for GTO.  The deadweight of the extra engine holds you back.  Dual Engine Centaur is better for LEO, which incidentally AV is yet to fly.

Quote
The Centaur has to be kept "in stretch" or under pressure at all times. It just means it has to be monitored. Most of the pressure monitoring is automated (use to require a tech onsite at all times).
Tank watch is now fully automated.  Warnings get sent to pagers, email and cell phones.  Gotta love technology.
If I like something on NSF, it's probably because I know it to be accurate.  Every once in a while, it's just something I agree with.  Facts generally receive the former.

Offline Gus

  • Member
  • Posts: 62
  • Liked: 2
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #45 on: 02/06/2007 04:13 AM »
Quote
Jim - 4/2/2007  3:16 PM

Quote
bombay - 4/2/2007  6:05 PM

Quote
yinzer - 4/2/2007  4:52 PM

Quote
bombay - 4/2/2007  9:58 AM
You have to ask the question, why did Atlas V settle on 12.5 ft dia. versus going to 17 ft like the Delata IV?  Existing Titan tooling and factory size pretty much dictated the choice to use a 12.5 ft. diameter Atlas V booster.

I'm curious as to how existing Titan tooling lead to the 12.5 ft diameter Atlas V booster.  All Titan stages (except the Centaur G, which doesn't apply in this case) were 10 ft diameter.
Titan IV tooling could accomodate up to 12.5 ft diameter without significant remods.

The Titan tooling is not used for the Atlas V.  All new tooling was created.  The domes and panels are produced by subcontractors.  Welding is only done in Denver.

The Titan weld cell in the second floor factory is used to weld barrels for Atlas V.  The new tooling is some handling equipment to handle skin panels and prep joints for weld.  All this will move to Decatur with an emphasis on the friction stir welded longitudinal welds.  The circumferential welds are still conventional arc welds.  The weld cell could accomodate up to a 13 foot diameter but constraints of C-5 and Antonov constrained the diameter to 12.5.

Offline Gus

  • Member
  • Posts: 62
  • Liked: 2
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #46 on: 02/06/2007 04:26 AM »
Quote
bombay - 4/2/2007  9:02 PM

Quote
Kayla - 4/2/2007  7:32 PM

Atlas V's 12.5' diameter was driven by the need to air lift the long booster tank.

Titan Centaur was 14' dia. and air lifted, though obviously not as long as the booster.

The Titan Centaur was derived from the Shuttle Centaur which was cancelled days after Challenger.  14 feet was the diameter that could be accomodated in the cargo bay of the shuttle.  It is also the reason many spacecraft require 14 ft or 4m fairings.  Although 5m fairings are offered, the majority of the Atlas manifest requires the 14ft PLF.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 30380
  • Cape Canaveral Spaceport
  • Liked: 8691
  • Likes Given: 283
Re: Atlas V and Centaur Q&A
« Reply #47 on: 02/06/2007 10:30 AM »
15 feet was the diameter of the shuttle payload bay and the Shuttle/Titan Centaur was closer to 15 than 14.   The 15 foot diameter payload bay was the reason for the 16.7 foot Titan IV fairing.  This translated into the 16.7 foot diameter Delta -IV (who still uses the same T-IV fairing along with a new composite 5m)) and why Atlas-V needed a 5.4 meter fairing

The 4m fairings are for "small" payloads that were Atlas II and Delta II class(GPS, DMSP, Comsat, etc)

The 5M is for T-IV class payload (NRO, MSL, Ariane Comsat, etc)

Offline quark

  • Veteran
  • Full Member
  • ****
  • Posts: 444
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #48 on: 02/09/2007 02:37 AM »
Quote
kevin-rf - 5/2/2007  2:28 PM

Okay to GTO, Atlas 552 with WBC is 12.9 mT. Delta Heavy is 13.1 mT. Close but Delta Heavy still has a slight 200 kg edge.

So for GTO missions developing Atlas 552 WBC would make more sense than developing Atlas V Heavy which gunther lists as 12.6 mT to GTO.
(That is assuming five solids, a single common core, and a WBC costs less than three common core Atlas booster and a standard Centuar).

Thanks

The WBC is about 4 times the cost to finish Atlas V HLV.  The AV HLV is essentially developed having completed CDR and qualification testing.  All that remains is buying and installing additional ground support equipment and finishing up a handful of unique components like separation motors for the LRB's.

Offline quark

  • Veteran
  • Full Member
  • ****
  • Posts: 444
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #49 on: 02/09/2007 02:43 AM »
Quote
Gus - 5/2/2007  10:13 PM

The Titan weld cell in the second floor factory is used to weld barrels for Atlas V.  The new tooling is some handling equipment to handle skin panels and prep joints for weld.  All this will move to Decatur with an emphasis on the friction stir welded longitudinal welds.  The circumferential welds are still conventional arc welds.  The weld cell could accomodate up to a 13 foot diameter but constraints of C-5 and Antonov constrained the diameter to 12.5.

All the booster weld tooling was new for Atlas V.  It is in the same location on the factory floor as the old titan stuff---but it is new.  All the Atlas V booster welds are currently conventional TIG welds.  The move Decatur offers an opportunity to convert the longitudinal welds to friction stir.  The circumferential welds would remain conventional (like they are on DIV).

Offline meiza

  • Expert
  • Senior Member
  • *****
  • Posts: 3069
  • Where Be Dragons
  • Liked: 2
  • Likes Given: 3
Re: Atlas V and Centaur Q&A
« Reply #50 on: 02/09/2007 12:16 PM »
Quote
quark - 9/2/2007  3:37 AM

Quote
kevin-rf - 5/2/2007  2:28 PM

Okay to GTO, Atlas 552 with WBC is 12.9 mT. Delta Heavy is 13.1 mT. Close but Delta Heavy still has a slight 200 kg edge.

So for GTO missions developing Atlas 552 WBC would make more sense than developing Atlas V Heavy which gunther lists as 12.6 mT to GTO.
(That is assuming five solids, a single common core, and a WBC costs less than three common core Atlas booster and a standard Centuar).

Thanks

The WBC is about 4 times the cost to finish Atlas V HLV.  The AV HLV is essentially developed having completed CDR and qualification testing.  All that remains is buying and installing additional ground support equipment and finishing up a handful of unique components like separation motors for the LRB's.

Interesting. What kind of sums are we talking about here? 800 million for WBC?
What are the main changes to current boosters if they are to be used in the heavy configuration? How much components can be manufactured in advance for application to both configurations? Engines and tanks? I understand that currently Delta "customizes" every booster, but Atlas doesn't? Would this be true if there was an Atlas V heavy?

Offline Kayla

  • Extreme Veteran
  • Full Member
  • ***
  • Posts: 396
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #51 on: 02/09/2007 02:49 PM »
Atlas HLV was a requirement from the beginning of the Atlas program as was commonality of the common core boosters (CCB) and Centaur across all of the Atlas V options.  Numerous times it was considered giving up some commonality to shave a few pounds of hardware weight and hence gain performance.  Luckily the decision was made each time to maintain commonality across the whole program, 4XY, 5XY and HLV.  This commonality allows the Atlas program to remanifest payloads late in the process to satisfy customer needs without maintaining a lot of inventory.  This flexibility is a huge benefit for the customer as well as providing significant cost savings.

Offline Kayla

  • Extreme Veteran
  • Full Member
  • ***
  • Posts: 396
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #52 on: 02/09/2007 03:32 PM »
Quote
bombay - 5/2/2007  8:45 PM

Quote
Kayla - 4/2/2007  7:32 PM

From its inception WBC was freed of this limitation.  Originally the assumption was that the tank would be built at Michoud where three 5.4m demonstration hoops were assembled and lap welded together to demonstrate the planned construction using thin (0.080") al-li sheet stock.  With the formation of ULA, ULA now has the option of using either Michoud or Decatur for construction.  Either location has the ability to friction stir weld large diameter tanks and then ship the final structures to the Cape or Vandenberg.

WBC is designed to merge the extremely efficient monocoque, common bulkhead tank from the historic Centaur with modern materials and friction stir welding spear headed for space application by the external tank and Delta programs.  This combination, along with increased propellant load allows the WBC to increase the mass fraction from the current Centaur's 0.91 to 0.94!  The following paper provides additional information.
http://www.lockheedmartin.com/data/assets/12382.pdf
Very interesting!  I'm surprised that lap joints are used given that they are analyzed in shear, not tension or compression where the allowables are much higher.

I'm guessing that the shear allowable for Al-Li is comparable to other 2XXX series aluminums: Fsu = 35 ksi and the welded allowable: Fsu = 20 ksi.  Realizing that fric. stir welding produces excellent as welded allowables, I would guess that in an analysis, 80% of 35 ksi would conservatively be used.  Lets say 30 ksi is used for the as-welded shear allowable.

The common bkhd concept would utilize a delta p between the LH2 tank and LOX tank.  Lets say maximum flight pressure in the tank is 50 psi (this is probably low).  So here's a simple analysis using ONLY pressure loading.  

I'll spare everybody the math, but the longitudinal line load would be 1800 lb/in and the stress would be 22.5 ksi using .080" mat'l.  Using a man-rated f.s. of 1.4, the m.s. would be -0.05 (ult).

Throw in S.V. bearing loads, interstage adapter, fairing, and bending loads and all other dynamic and transient axial loading, you could see where the guage thickness would need to be much thicker, which in-turn would alter mass ratio.

The .94 appears to be a goal, not a given.  
 

Im very glad that people are truly thinking and asking pointed questions!  It is this kind of dialog that keeps everyone on their toes.

When working with thin material lap welding has numerous advantages.  First layup accuracy can be greatly relaxed.  With a butt weld of 0.08 or thinner material any offset will greatly affect the joint strength.  On the other hand with lap welding, the overlap is measured in many tenths of inches or more.  Secondly, with the lap weld one can have multiple parallel welds.  These two benefits have been used for almost 5 decades on the Atlas and Centaur tanks using CRES (steel) on material thicknesses ranging from 0.013 to ~0.04.  

A lot of material work has been done under contract to NASA on various aluminum alloys, friction stir welded joints and the material properties at various temperatures as a prelude to future light weight tanks.  Some of this was presented last year at STAIF 2006 and more will be presented next week at STAIF 2007.  For example Al 2024 had average ultimate tensile strength of 85.6 KSI at LH2 temperatures with 100% weld strength using FSW.

The referenced 0.94 mass fraction is the result of a bottoms up estimate of the weight of a 161,000 lb propellant load WBC, shorter versions with less propellant will have slightly lower mass fractions.  The stage dry weight is composed of the tank structure, dry support structure, engines, avionics, propulsion-sub systems, etc.  On the current Centaur, the propellant tanks represents less than 1/3 of the dry mass of the stage, the rest of the weight coming from these other components.  For WBC Atlas will use most of the existing hardware, reintegrated onto this new large diameter tank.  About 2/3s of the weight of WBC is existing hardware and thus the weight is very well under stood. The dry structure will be built using material and processes currently used on Centaur, once again with minimal uncertainty in the weight.  This only leaves the tank as a potential weight risk.  The material and dome spinning work being done for NASA, along with the internally funded actual construction of 5.4m diameter cylinders is designed to directly reduce the uncertainty in the tank weight and manufacturability.  The WBC is on track to retire the leading edge risks associated with weight, manufacturing and integration.



Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #53 on: 02/09/2007 03:39 PM »
The sum that is often quoted for HLV development is $200M.  If the 4x figure is correct for WBC development, $800M would be in the ballpark.

I suspect if mega funds are going to be invested in something, it will be (and should be) in the U.S. development of the RD-180 given the recent events of Sea Luanch and the potential impact on Atlas V.

The various configurations of the CBC for Delta IV doesn't seem to be very common compared to the CCB for Atlas V.

Online kevin-rf

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8332
  • Overlooking the path Mary's little Lamb took..
  • Liked: 961
  • Likes Given: 225
Re: Atlas V and Centaur Q&A
« Reply #54 on: 02/09/2007 03:47 PM »
Quote
bombay - 9/2/2007  10:39 AM
I suspect if mega funds are going to be invested in something, it will be (and should be) in the U.S. development of the RD-180 given the recent events of Sea Luanch and the potential impact on Atlas V.

Your assuming the RD-171 is at fault, Chris hinted in the Sea Launch thread it might be something else.
I just saw some idiot at the gym put a water bottle in the pringles holder on the treadmill.

Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #55 on: 02/09/2007 04:10 PM »
"When working with thin material lap welding has numerous advantages. First layup accuracy can be greatly relaxed. With a butt weld of 0.08 or thinner material any offset will greatly affect the joint strength. On the other hand with lap welding, the overlap is measured in many tenths of inches or more. Secondly, with the lap weld one can have multiple parallel welds. These two benefits have been used for almost 5 decades on the Atlas and Centaur tanks using CRES (steel) on material thicknesses ranging from 0.013 to ~0.04."

Yes, the current Centaur has lap joints with parallel running rows of welds.  The big difference is that the stainless rows of welds, except the row that seals the joint, are not continuously running.  Also the resistance welding process is used.  So from a structural integrity perspective, the material efficiency in the spotweld rows is not greatly affected and properly spaced spots in shear are very efficient and strong.  The same can not be said for lap joint in shear with continuously running buttwelds.

In addition, the mock analysis that I did a few posts ago was incorrect, I used 12 ft as the tank diameter instead of 17.5 ft.  So the m.s. would be -0.67 ultimate not -0.05.  The point of this is that .080" thick wall appears to be the thickness needed to produce a .94 mass fraction rather than what is actually needed to provide structural integrity.  Something in the neighborhood of .150" thick or more would be the more appropriate thickness needed to provide positive margins.    

Offline Kayla

  • Extreme Veteran
  • Full Member
  • ***
  • Posts: 396
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #56 on: 02/09/2007 05:38 PM »
We did not set off to develop a 0.94 mass fraction vehicle.  We calculated what it took to carry the loads, pressure and structural, for the missions that we were expecting to fly.  For a purely in-space stage that is carried aloft in the PLF 0.04" is adequate.  For an upper stage that also needs to carry a PLF and Payload 0.08" is what was required.  From this we calculated a stage weight, hence the final result.

Offline jongoff

  • Recovering Rocket Plumber/Space Entrepreneur
  • Senior Member
  • *****
  • Posts: 5989
  • Lafayette/Broomfield, CO
  • Liked: 1938
  • Likes Given: 667
Re: Atlas V and Centaur Q&A
« Reply #57 on: 02/09/2007 05:44 PM »
Bombay,
Quote
In addition, the mock analysis that I did a few posts ago was incorrect, I used 12 ft as the tank diameter instead of 17.5 ft.  So the m.s. would be -0.67 ultimate not -0.05.  The point of this is that .080" thick wall appears to be the thickness needed to produce a .94 mass fraction rather than what is actually needed to provide structural integrity.  Something in the neighborhood of .150" thick or more would be the more appropriate thickness needed to provide positive margins.

While you might be right, I'm kinda doubtful myself.  The guys studying the WBC have been doing manufacturing development work on the FSW joints for a couple of years now.  I would assume that they can tell how thick the thing needs to be with adequate margins by now.  Typically, if my back of the envelope calculations are a factor of two off from something that you can only assume have had many engineer months or years put into them, I assume that I made a math error, or that there's something subtle that I'm missing.

Now, on the other hand, asking questions and picking their brains to see if they'll let you on to what that math error might be, or what subtlety you might be missing is of course both fair game and a good idea.

~Jon

Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #58 on: 02/09/2007 08:27 PM »
With $800M or so (assumed) worth of development work to be done on the WBC, I wouldn't anticipate that too much analytical work went into the initial concept.  

The fact is that the WBC joints were mentioned to be a lap joints.  What I'm presenting, using "assumed" pressure loads as I implied, is based soley on standard mechanics of materials principles for a thin-walled pressurized cylinder with the relevant failure mode for a welded lap joint in tension being shear - nothing more than that!!  
 
FSW lap joints have only been recently proven, unlike buttwelded joints.

Offline bombay

  • Regular
  • Full Member
  • **
  • Posts: 299
  • Liked: 0
  • Likes Given: 0
Re: Atlas V and Centaur Q&A
« Reply #59 on: 02/09/2007 08:41 PM »
"We did not set off to develop a 0.94 mass fraction vehicle. We calculated what it took to carry the loads, pressure and structural, for the missions that we were expecting to fly. For a purely in-space stage that is carried aloft in the PLF 0.04" is adequate. For an upper stage that also needs to carry a PLF and Payload 0.08" is what was required. From this we calculated a stage weight, hence the final result."

My computer is acting all goofy, so I have to reply this way rather than the normal way - Sorry!

Anyway, I got the .94 "goal" from the paper that I believe you (Kayla) attached, where it showed it as a goal in one of the charts.  

You mentioned before that the WBC would benefit from increased propellant loading.  What pressures in the tank are we talking about?  Just a ballpark figure!

Tags: