First Delta IV Heavy launch

[Ottawan]

But isn't that where the "risk" factor comes in?

When do you man-rate a booster? When it's at 98% or 99%?

What could the Atlas have been at when Glenn went up?

Back in the early 60's you lost one astronaut if the launch failed. Lately it's been seven.

I seriously think that the decision not to use the Delta as a "man-rated" booster is that the PTB have already decided to use STD (SRB's) to boost the CRV into LEO.

Sorry, I sound like Robin Williams in "Good Morning, Vietnam"

[Johno]

Although I am a layman, it occurs that the problem here is that there are no guarantees.  To unilaterally declare a vehicle safe for humans must be a subjective judgement unless you have specific knowledge of the future. 

You can't have that, mainly because of quantum.  However, I digress.

We can talk about 1 failure in 50 all we like, but at the end of the day, some failures are inherently unpredictable, and some predictable failures simply don't materialise.  That's why you get issues like unrated boosters with no failures, and man-rated boosters (not mentioning any particular cases), oh, causing 14 deaths over the lifetime of the system . . .

[Bob B.]

But isn't that where the "risk" factor comes in?

Since 100% reliability is impossible, risk will always be present.

Let’s look at a numerical example …

Say a launch vehicle has a unit cost of $185 million and a reliability of 90%.  Every 100 launches has a total cost of $18.5 billion, out of which 90 payloads will be delivered to orbit.  Thus our cost per payload delivered is $206 million. 

We’ll assume that for an extra $15 million per unit we can increase the reliability to 98%.  It will now cost $20 billion to successfully delivery 98 payloads, thus our cost per payload is $204 million.  It is therefore more economical to build the more reliable vehicle.

Each extra percent of reliability costs more than the previous percent (law of diminishing returns).  We got 8% more reliability for only $15 million per vehicle, but let’s say it will cost us $5 million per unit more to get another 1% reliability.  It will now cost $20.5 billion to deliver 99 payloads to orbit for a per payload cost of $207 million. 

For a cargo vehicle the extra reliability doesn’t make economic sense.  It is cheaper just pay for the loss of the extra payloads.  We therefore go into standard production with a vehicle know to be less reliable than it could be.

However, if this vehicle is going to carry a crew then we want to build in the extra reliability regardless of cost.  Our main driver is now safety, not economics.  We therefore pay the extra $5 million per unit to do the things we deliberately chose not to do in the standard cargo vehicle.

When do you man-rate a booster? When it's at 98% or 99%?

This I don't know.  There may be a fuzzy line between reliable enough and not reliable enough.  I know that if I was going to ride on a rocket I wouldn't want to know there was something that could have been done that wasn't simply because it was more economical to allow me to die.

[Bob B.]

We can talk about 1 failure in 50 all we like, but at the end of the day, some failures are inherently unpredictable, and some predictable failures simply don't materialise. That's why you get issues like unrated boosters with no failures, and man-rated boosters (not mentioning any particular cases), oh, causing 14 deaths over the lifetime of the system . . .

You are absolutely right.  You never know what will really happen; you can only make estimates based on the best information available.  Engineering economics looks largely at probabilities.  Just because something is probable doesn't mean it will happen that way.  Nonetheless, these probabilities form the basis for many business decisions.

edit spelling

[Ottawan]

   Some day I gotta learn to use that "Quote" thingy . .

Good points both Johno and Bob B. but myobservation still remains . . . .

There is no thought being given to man-rating the Delta, or any other rocket for that matter, because the decision seems to have been made to use Shuttle Derived Technology.

I was scared to bejeezus when I saw the CEV being launched on a SRB. I noticed the Launch Escape System but still. . . . .

I am afeared that NASA is falling back into the "Goldin Age" of "Better, Faster, Cheaper".

I would like to see "Bestest, Safest, Whogivesaratsassaboutthecostest".

But that's not gonna happen.

When this Administration goes, this Vision for Space Exploration goes.

I've been living the frustration since 1972 guys . . . . that's thirty-three fricken years!

You think I am going to count on a political system to get us back to the moon?

I'm starting to learn Chinese

[Bob B.]

I was scared to bejeezus when I saw the CEV being launched on a SRB. I noticed the Launch Escape System but still. . . . .

The SRB is actually a very reliable motor.  Since its redesign following the Challenger accident, 178 of them (2 per launch) have been flow without incident.  That's a pretty impressive safety record.  Furthermore, the SRB problem on Challenger wouldn't have been so disastrous had the motor not been mounted immediately next to 700 tons of liquid hydrogen and oxygen.  This arrangement has been eliminated in the new CLV design.

[skyjim]

I guess that one set of requirements is extensive onboard sensors and a fault evaluation system which can utilize the data from them to make split-second abort decisions.  Not a trivial bit of R&D to be sure, but I suspect that this and other tasks needed to human-rate (We can't really say "man-rate" any more, right?) a vehicle like Delta IVH aren't going to be any costlier than the CLV development.  New upper stage, modify SSME to air-start, devise roll control for single SRB, full aero, thermal, and vibe testing regimen, and a new fault detection system for the LES, not to mention pad development - somebody is going to have to explain how this is going to be cheaper than a crewed EELV Heavy.  I don't believe that's the issue, though, and I wish the NASA people who claim this would stop.  It's a political call.   

I understand why they are going with the design they've chosen for CLV - it's OK technically and superb politically - but I admit that I'm still prejudiced against flying people on solids.  One of those unforeseen failure modes could cause  sudden catastrophic failure.  Improbable?  Sure.  But I still see that Titan IV, and that Delta II a decade or so ago, at the instant when an SRB went BOOM, and I'm still not convinced that any LES is going to get a crew clear of that sort of failure without smashing them to jelly.

 As I've stated elsewhere, though, I'm willing to be educated...   

Jim

[DonPMitchell]

And of course you have to be able to throttle back thrust on a manned rocket, unless you want to squash people when the tanks near empty.  I assume that's possible, but the rocket may not be originally designed to do it.

The first chimp we sent into space got exposed to 20 gees, due to a miscalculation on how much throttling had to be done.

[Bob B.]

And of course you have to be able to throttle back thrust on a manned rocket, unless you want to squash people when the tanks near empty. I assume that's possible, but the rocket may not be originally designed to do it.

The SSME is already made to throttle; I know it can at least throttle down to 65%.  And even if it isn't throttled, it looks like the maximum acceleration at stage 2 cutoff will be around 4.5 g.

Stage 1, on the other hand, obviously can't be throttled since it is solid propellant.  Solid motors are funny in that their thrust varies over the burn as the area and shape of the burn surface changes.  I have no idea what the SRB thrust is at burnout.  The SRBs are designed to give a burst of maximum thrust right at liftoff and then drop of to about 2/3rds thrust after 50 seconds.  If I assume the burnout thrust is equal to the average thrust, then the maximum acceleration is about 3.9 g.

According to this page, the CLV is designed for a maximum of 4.00 g during the accent burn.  This is about the same maximum acceleration as the Saturn V.

[evancise]

I thought y'all might find this interesting...

I'm quoting the following information from a class I took a couple years ago called, "Human Space Flight: Mission Analysis and Design."  The handouts are copyrighted to the Teaching Science and Technology, Inc., 2002.  That should cover the legal aspects....

NASA has requirements for 4 basic types of missions, based on NASA document JSC-28354:

• Earth to Orbit vehicles (ETO)
• Space Stations (SS)
• Crew Return Vehicles (CRV) [non-routine entry]
• Beyond Earth Orbit Vehicles (BEO)

The requirements are divided into three areas:

• General requirements
• Safety and reliability requirements
• Human-in-the-loop requirements

General Requirements

Requirement 1 (design for human space flight) - The vehicle shall be designed, built, inspected, tested, and certified specifically addressing the requirements for human-rating.

Requirement 2 (aerospace design standards) - The vehicle design, manufacture, and test shall comply with JSCM 8080.5 and applicable military standards.

Requirement 3 (crew habitability) - The vehicle crew habitability and life support systems shall comply with NASA Standard 3000 and NASA Space Flight Health Requirements for crew habitability and life support systems design.

Requirement 4 (flight test) - A successful, comprehensive flight test program shall be completed to validate analytical math models, verify the safe flight envelope, and provide a performance data base prior to the first operational flight with humans on board.

Requirement 5 (proximity operations) - Spacecraft operations in proximity or docking with a crewed vehicle shall not pose a hazard to either vehicle.  Provisions shall be made to enable abort, breakout, and separation by either vehicle without violating the design and operational requirements of either vehicle.  Uncrewed vehicles must permit safety critical commanding from the crewed vehicle.

Safety & Reliability Requirements

Requirement 6 (crew survival) - The program shall be designed so that the cumulative probability of safe crew return over the life of the program exceeds 0.99.  This will be accomplished through the use of all available mechanisms including mission success, abort, safe haven, and crew escape.

Requirement 7 (crew survival) - A crew escape system shall be provided on ETO vehicles for safe crew extraction and recovery from in-flight failures across the flight envelope from prelaunch to landing.  The escape system shall have a probability of successful crew return of 0.99.

Requirement 8 (aborts) - For ETO vehicles, abort modes shall be provided for all phases of flight to safely recover the crew and vehicle or permit the use of the crew escape system.  For BEO missions, spacecraft and propulsion systems shall have sufficient power to fly trajectories with abort capabilities and provide power and critical consumables for crew survival.  Trajectories and propulsion systems shall be optimized to provide abort options.  When such options are unavailable, safe haven capabilities shall be provided.

Requirement 9 (flight termination) - A flight termination (range safety) system is required for ETO vehicles (and BEO vehicles launched intact) not demonstrating aircraft-like reliability to provide for safe recovery of the crew.

Requirement 10 (failure tolerance) - All critical systems essential for crew safety will be two-fault tolerant.  If not practical, no single failure shall cause loss of crew.

Requirement 11 (reliability verification) - Vehicle reliability shall be verified by test backed up with analysis at the integrated system level prior to first crewed flight and verified by flight-based analysis and system health monitoring for subsequent flights.

Requirement 12 (software reliability) - The performance and reliability of all critical software shall be tested on a flight equivalent avionics testbed across the entire flight envelope.  Independent Verification and Validation (IV&V) methods shall be used to confirm the integrity of the software testing process.

Human-in-the-Loop Requirements

Requirement 13 (crew role and insight) - The vehicle shall provide the flight crew on board the vehicle with proper insight, intervention capability, control over vehicle automation, authority to enable irreversible actions, and critical autonomy from the ground.

Requirement 14 (crew role and insight) - The flight crew shall be capable of taking manual control of the vehicle during all phases of flight.  The vehicle shall exhibit Level 1 handling qualities as defined by the Cooper-Harper Rating Scale.  (Qualitative scale used by test pilots to rate the flying qualities of an airplane: 1 is excellent requiring no pilot compensation, 10 has major deficiencies resulting in loss of control during some phases of flight.)

Requirement 15 (crew role and insight) - The spacecraft displays and controls design shall be based on a detailed function and task analysis performed by an integrated team of human factors engineers with spacecraft displays and controls design experience, vehicle engineers, and crew members.

Requirement 16 (task analysis) - The mission design, including task design and scheduling, shall not adversely impact the ability of the crew to operate the vehicle.


So, after all that - the number that y'all are looking for is 0.99 for the reliability of the entire system.

[Bob B.]

Excellent, thank you.

[SpaceChem]

Thank you, evancise!  I think this answers a lot of questions.