ATK push for US space flight independence via Liberty

[ugordan]

Ok, so in a sense, what you're saying is that liquids are more riskier and less reliable (based on complexity of the system) than solids based on a general need to check liquids proir to commit?

No, that's not what he's saying at all. The ability to check liquids prior to launch commit does not imply they are less safe.

[Adam K]

Ok, so in a sense, what you're saying is that liquids are more riskier and less reliable (based on complexity of the system) than solids based on a general need to check liquids proir to commit?

No, that's not what he's saying at all. The ability to check liquids prior to launch commit does not imply they are less safe.

More Risk doesn't imply less safe.  The inherent design of a liquid engine along with the turbines, pumps, implies there is more moving parts and therefore more risk.  If you could prove that all parts would work all the time, there would be no need (but a desirement to "because you can") to test the engines prior to lift-off.  Right now, there is a need to do so.

[wolfpack]

Fellas, this is about Liberty, not general liquid vs. solid debate. Liberty is an inline, 2 stage rocket with a solid fueled 1st stage. The question is whether or not this is the best design from an engineering standpoint. In my opinion it is not, but I'm open to being educated on the subject.

[Downix]

The Atlas solids include thrust termination, note.

I haven't been able to find any information about this, except for a note in a January 2012 ULA presentation that said it might consider using thrust termination as a range safety method, rather than the current use of explosive destruction.  There was no mention of solids in the presentation.
www.aiaa.org/WorkArea/DownloadAsset.aspx?id=7543

Interestingly, this ULA presentation shows Dream Chaser using a solid boosted Atlas.

 - Ed Kyle

I'll have to check, but I recall that they developed a thrust termination modified solid under the OSP program for Atlas. Was not a significant modification, but IIRC it was tested and was ready. I'll have to dig that out.

[strangequark]

More Risk doesn't imply less safe.  The inherent design of a liquid engine along with the turbines, pumps, implies there is more moving parts and therefore more risk.  If you could prove that all parts would work all the time, there would be no need (but a desirement to "because you can") to test the engines prior to lift-off.  Right now, there is a need to do so.

You're falsely equating # of moving parts with complexity. To take an example, jet engines have a substantial number of moving parts, whereas a scramjet has virtually none, and yet we can only just barely get them to work. Solid rocket motors are not simple, just because they don't have moving parts.

Liquids give you the ability to test and characterize the idiosyncrasies of the engines you will actually be flying, and to shutdown right up to the moment of launch. With solids, you have no such luxury, and have to get by with subscale testing and modeling, and hope that the actual flight unit will be in family. There is a need to test both types, but with solids you can only test a similar device, not the one you actually fly. Couple that with the fact the many of the failure modes of a solid are, shall we say, less than graceful, and you can understand the inherent reluctance than many have in using them for a manned system.

[mmeijeri]

I don't have a problem with the specifics of the rocket itself.  It is a powerful two-stage rocket created by assembling existing, or largely existing, proven and reliable systems.

I believe this is incorrect. The 5 seg solids are not "largely existing", neither are the avionics, and the upper stage and its engine need a substantial redesign. What is reused is the workforce and tooling. That's not insignificant, but you are putting a very positive spin on it.

[Downix]

I don't have a problem with the specifics of the rocket itself.  It is a powerful two-stage rocket created by assembling existing, or largely existing, proven and reliable systems.

I believe this is incorrect. The 5 seg solids are not "largely existing", neither are the avionics, and the upper stage and its engine need a substantial redesign. What is reused is the workforce and tooling. That's not insignificant, but you are putting a very positive spin on it.

They have already done three test firings on the 5-segment solids, so I would classify them as largely existing. The Avionics were tested as well. The upper stage engine has already been tested in the configuration planned for. The upper stage does not require a significant redesign either, it does need stiffening but not significantly.

I'm not seeing a game stopper for this, in fact I'm seeing a very good design, which surprised me as I was not an Ares I fan.

[Jim]

The Avionics were tested as well.

Are they using Ariane avionics.

[wolfpack]

I don't have a problem with the specifics of the rocket itself.  It is a powerful two-stage rocket created by assembling existing, or largely existing, proven and reliable systems.  It would be built in existing factories by existing workers, and be transported, assembled, and launched using largely existing infrastructure.  Its systems would continue to be used by other rockets, helping defray costs.

But the question is this - best design for what?  Liberty lifts twice as much mass as its ISS commercial crew competitors.  It would, as a result, seem to be a costlier solution for that application.  It can only compete on costs, it seems to me, if it also takes over the ISS cargo work, which would turn it into a sort of mini-shuttle.  ATK/EADS have, of course, made just such a proposal. 

Liberty and its spacecraft seem to offer more capability than its competitors.  But can NASA use that capability?

 - Ed Kyle

Good info.

There are varying ways to interpret "best", I should have realized. Best ?= performance, best ?= safety, best ?= cost, etc.

I'm biased towards the safety aspect when it comes to HSF. So, for me, best = safest. All we need to do, in my opinion, for Commercial Crew is get somewhere between 5 to 7 bodies to LEO. I'd like to see LOCV risk < 1/1000. Can Liberty do that? Is it better or worse than other designs? How do we measure that?

Here's what I found from digging through the CCiCap documents.

-- CCiCap

Safety and Mission Assurance

The Participant shall describe its approach and management processes (e.g., test operations, range, ground, and flight), for safety reliability, maintainability, supportability, quality, and software assurance.  When discussing the approach and management processes, describe the interactions between the safety function, engineering function, and project management.  Include a discussion of the proposed approach for assuring the safety and survivability of crewmembers and passengers including methods of assessing risk to human life, giving consideration to potential customer standards (such as CCT-PLN-1120, CCT-REQ-1130, CCT-STD-1140, CCT-STD-1150, SSP 50808, and industry equivalents). The Participant shall also describe the technical and programmatic risks associated with the proposed CCiCap effort with particular attention to ensuring safety of crewmembers and the public.

6.1.2   Initial Evaluation – NASA will then conduct an evaluation of proposals that are compliant with this Announcement to assess the effectiveness of the Technical Approach and Business Information in accordance with Table 6-1 below. Additionally, proposals will be evaluated for confidence, where NASA will assign a rating of High, Medium or Low likelihood of successful performance of the proposed CCiCap activities for both Technical Approach and Business Information in accordance with Table 6-2. The Technical Approach and Business Information will be evaluated separately with an effectiveness rating and confidence rating for each.  NASA will evaluate proposal content for both the base and optional periods to determine the extent to which the proposals meet the intent of the goals outlined in Section 3.


-- CCT-PLN-1120

4.1   System Safety and Reliability
The Commercial Partner will incorporate safety and reliability into the design process to identify and eliminate catastrophic hazards.  If elimination is not practical, the Commercial Partner will put in place controls that reduce the risk to an acceptable level for crew safety and mission success.  These controls will be verified.

The safety and reliability processes provide an integrated, systematic, and comprehensive approach, which can be used to determine the need for design changes and safety measures, such as failure tolerance, based on an understanding of the risk to the crew due to individual hazards, as well as the system as a whole.  These methods are used to balance risks and trades by identifying hazards and developing hazard controls based on assessments of crew risk and rankings of risk scenarios.  In this way, the level of failure tolerance and other safety measures can be selected commensurate with the risk posed to the crew.  This includes the evaluation of hardware and software capabilities, limitations, and interdependence, as well as environmental and human factors relevant to safety and mission success.

The safety and reliability processes should:
•   Emphasize the use of industry standard safety and reliability analyses (e.g., hazard analysis, failure modes and effects, criticality analysis, etc.) throughout the system life-cycle (i.e., design, manufacture, and operation) of the Commercial Partner’s CTS, as well as during any changes to the baseline design and/or operation.
•   Provide detailed plans for the communication and acceptance of risk to stakeholders and/or appropriate control boards to ensure residual risk is appropriately managed. 
•   Include methodologies for performing qualitative assessments of the probability of occurrence and the severity of hazards, including detailed definitions for likelihood and severity categories that will be used by the Commercial Partner to ensure a uniform method of assessing hazards.
•   Provide for closed-loop tracking and verification of hazard controls.
•   Include a plan for communication and approval of safety and reliability analyses.

That's a lot of verbiage, but no numbers. I'm not even seeing a requirement for applicants to use a common risk assessment methodology. Maybe it works out that way via "industry standards" (which I don't know as I'm not in the industry), but maybe not.

[sdsds]

ATK said that the abort system would be designed to clear any first stage failure

That is a remarkable goal! Is that exactly what they have said? It has never come across to me as having been quite that directly stated! Did they say "would be" designed (future tense) or "has been" designed (past tense)? So ... do they have a solution to the problem, or just an admirable intent to address it?

[wolfpack]

ATK said that the abort system would be designed to clear any first stage failure

That is a remarkable goal! Is that exactly what they have said? It has never come across to me as having been quite that directly stated! Did they say "would be" designed (future tense) or "has been" designed (past tense)? So ... do they have a solution to the problem, or just an admirable intent to address it?

A thought I had while recovering from a hangover on the treadmill this morning (don't ask ...

Would it be possible to REDUCE thrust from the 1st stage during an abort by explosively severing the nozzle (similar to the Shuttle SRB just prior to splashdown)? Would that buy some margin for the escaping spacecraft? First-order analysis says "yes" (it should just cause significant under-expansion of the exhaust flow and thus reduce thrust), but higher-order analysis is skeptical (dynamic loads, under-expanded plume impingement on hardware in the +X direction).

I could see it working. I could also see it creating more problems than it solves. Or even being a wash in terms of risk, thus why invest the R&D.

[Downix]

ATK said that the abort system would be designed to clear any first stage failure

That is a remarkable goal! Is that exactly what they have said? It has never come across to me as having been quite that directly stated! Did they say "would be" designed (future tense) or "has been" designed (past tense)? So ... do they have a solution to the problem, or just an admirable intent to address it?

A thought I had while recovering from a hangover on the treadmill this morning (don't ask ...

Would it be possible to REDUCE thrust from the 1st stage during an abort by explosively severing the nozzle (similar to the Shuttle SRB just prior to splashdown)? Would that buy some margin for the escaping spacecraft? First-order analysis says "yes" (it should just cause significant under-expansion of the exhaust flow and thus reduce thrust), but higher-order analysis is skeptical (dynamic loads, under-expanded plume impingement on hardware in the +X direction).

I could see it working. I could also see it creating more problems than it solves. Or even being a wash in terms of risk, thus why invest the R&D.

ATK developed a way to shut down, then restart a solid motor back in '03:

http://www.thefreelibrary.com/ATK+Successfully+Conducts+Its+First-Ever+Start-Stop-Start+Solid...-a0131555528

The technology used likely could also be used to allow for a termination to allow a safer abort.

[Robotbeat]

ATK said that the abort system would be designed to clear any first stage failure

That is a remarkable goal! Is that exactly what they have said? It has never come across to me as having been quite that directly stated! Did they say "would be" designed (future tense) or "has been" designed (past tense)? So ... do they have a solution to the problem, or just an admirable intent to address it?

A thought I had while recovering from a hangover on the treadmill this morning (don't ask ...

Would it be possible to REDUCE thrust from the 1st stage during an abort by explosively severing the nozzle (similar to the Shuttle SRB just prior to splashdown)? Would that buy some margin for the escaping spacecraft? First-order analysis says "yes" (it should just cause significant under-expansion of the exhaust flow and thus reduce thrust), but higher-order analysis is skeptical (dynamic loads, under-expanded plume impingement on hardware in the +X direction).

I could see it working. I could also see it creating more problems than it solves. Or even being a wash in terms of risk, thus why invest the R&D.

ATK developed a way to shut down, then restart a solid motor back in '03:

http://www.thefreelibrary.com/ATK+Successfully+Conducts+Its+First-Ever+Start-Stop-Start+Solid...-a0131555528

The technology used likely could also be used to allow for a termination to allow a safer abort.

COMPLETELY different scale. Orders and orders of magnitude different. Might as well talk about a firecracker company that has done the same.

[cleonard]

ATK said that the abort system would be designed to clear any first stage failure

That is a remarkable goal! Is that exactly what they have said? It has never come across to me as having been quite that directly stated! Did they say "would be" designed (future tense) or "has been" designed (past tense)? So ... do they have a solution to the problem, or just an admirable intent to address it?

A thought I had while recovering from a hangover on the treadmill this morning (don't ask ...

Would it be possible to REDUCE thrust from the 1st stage during an abort by explosively severing the nozzle (similar to the Shuttle SRB just prior to splashdown)? Would that buy some margin for the escaping spacecraft? First-order analysis says "yes" (it should just cause significant under-expansion of the exhaust flow and thus reduce thrust), but higher-order analysis is skeptical (dynamic loads, under-expanded plume impingement on hardware in the +X direction).

I could see it working. I could also see it creating more problems than it solves. Or even being a wash in terms of risk, thus why invest the R&D.

If I remember correctly the only way to a abort a burning solid is to depressurize it.  This is generally done explosively and basically unzips a good portion of the pressure chamber.  The burning rate of solid propellants is a strong function of the pressure.  Remove that pressure and it slows down a lot.  Breaking off the nozzle would be a good start, but I think that a larger depressurizing vent would need to be created.

The root of the problem is that anything like this only deals with benign solid failure modes.  The catastrophic failure of a chunk of fuel breaking off and blocking the nozzle can not be mitigated.  That pressure vs burn rate makes the blockage turn into a near instantaneous large explosion of the solid booster containment vessel.  There is no time for any launch abort system to do anything.  If a large fast moving piece of the booster hits the crew compartment it's game over.   

[Robotbeat]

Or, burning pieces of propellant hit the parachutes (much bigger targets)...

[wolfpack]

ATK developed a way to shut down, then restart a solid motor back in '03:

http://www.thefreelibrary.com/ATK+Successfully+Conducts+Its+First-Ever+Start-Stop-Start+Solid...-a0131555528

The technology used likely could also be used to allow for a termination to allow a safer abort.

burn rate = a*Pc^n, right? "a" and "n" are propellant-specific, so what works for one motor might not work for another. But, yes, in theory rapidly dropping the chamber pressure should stop combustion.

Example a, n values are 5.6059* (pressure in MPa, burn rate in mm/s) and 0.35 respectively for the Space Shuttle SRBs, which gives a burn rate of 9.34 mm/s at the average chamber pressure of 4.3 MPa.*

I am assuming Liberty values would be about the same. How much would the chamber pressure need to be reduced to seriously moderate thrust? How quickly can that process take place?

*http://www.braeunig.us/space/propuls.htm

[wolfpack]

Breaking off the nozzle would be a good start, but I think that a larger depressurizing vent would need to be created.

I should have been more specific. I meant break off the nozzle extension. Leave the throat. The shock at the throat would keep the chamber pressure constant, would it not? Basically I was just trying to make it a really inefficient motor in a big hurry, not actually stop combustion.

Agree on the catastrophic mode of nozzle blockage. In all fairness to ATK, though, I am forced to acknowledge that that has never happened neither in a test firing of a Shuttle SRB nor in the flights of 270 motors. That's a sizable enough statistical database to have a decent confidence interval.

[cleonard]

Or, burning pieces of propellant hit the parachutes (much bigger targets)...

Assuming that you have enough deltav in the launch abort system you could at least get far enough away from the shower of burning fuel.  I'm not sure how much deltav is needed to clear a worst case scenario, but I'd guess that it must be at least 500 m/s or more if it is such an issue.

[Downix]

Breaking off the nozzle would be a good start, but I think that a larger depressurizing vent would need to be created.

I should have been more specific. I meant break off the nozzle extension. Leave the throat. The shock at the throat would keep the chamber pressure constant, would it not? Basically I was just trying to make it a really inefficient motor in a big hurry, not actually stop combustion.

Agree on the catastrophic mode of nozzle blockage. In all fairness to ATK, though, I am forced to acknowledge that that has never happened neither in a test firing of a Shuttle SRB nor in the flights of 270 motors. That's a sizable enough statistical database to have a decent confidence interval.

It has happened with the related Titan 7-segmented SRB however, so is not unheard of.

[wolfpack]

Or, burning pieces of propellant hit the parachutes (much bigger targets)...

Assuming that you have enough deltav in the launch abort system you could at least get far enough away from the shower of burning fuel.  I'm not sure how much deltav is needed to clear a worst case scenario, but I'd guess that it must be at least 500 m/s or more if it is such an issue.

Probably some "devil-in-the-details" with regards to what the worst case is. I think you have to employ statistical analysis on the ballistic trajectories of both the escaping spacecraft and the liberated propellant. Add some guard bands (the devil lives here BTW), then figure out the probability of them being in the same place at the same time.