NASA has confirmed to Hyperbola that it has started to spend the $1 million of American Reinvestment and Recovery Act funds it has for human rating studies. With human rating being such a divisive topic at the moment - is Atlas or Delta OK, is commercial up to scratch - the synopsis makes interesting reading
NASA told Hyperbola: "[T]he procurement you identified is related to the $1 million of Recovery Act funds allocated for human rating requirements development. The procurement is for only part of the $1 million, not all of it."
How much of the $1 million is being spent is not being disclosed but the synopsis showing that Wyle Integrated Science and Engineering is getting the human rating contract does says "The human system integration requirements developed under this task order shall be based on a review of existing Human Rating requirements".
Go through to the extended portion of this blog post to see the full procurement synopsis text
NASA told Hyperbola: "[T]he procurement you identified is related to the $1 million of Recovery Act funds allocated for human rating requirements development. The procurement is for only part of the $1 million, not all of it."
How much of the $1 million is being spent is not being disclosed but the synopsis showing that Wyle Integrated Science and Engineering is getting the human rating contract does says "The human system integration requirements developed under this task order shall be based on a review of existing Human Rating requirements".
Go through to the extended portion of this blog post to see the full procurement synopsis text
The full synopsis text is:
"The purpose of this task order is to develop a set of human system integration requirements for application to commercial spacecraft in support of NASA's Commercial Crew and Cargo Program (C3PO). The human system integration requirements developed under this task order shall be based on a review of existing Human Rating requirements, Spaceflight Human Systems Standards, Constellation Program requirements, Commercial Crew and Cargo Program Office operational concepts and requirements, and the Johnson Space Center Space Life Sciences Directorate Human Interface Design Handbook."
Despite all the allegations made about the safety of Ares I crew launch vehicle Hyperbola has heard from various quarters that the Constellation programme team was taking safety very seriously. Requirements over and above past programmes were being employed contributing to the ballistic cost growth
Looking at the synopsis text above you get the feeling that the debate - between Constellation and commercial supporters about what is a satisfactory level of rating - has got to the point where that key decision to resolve impassable management logjams has been taken: bring in the consultants
We hear lots about safety factors of one point this and one point that but in this blogger's humble opinion there is no substitute for shakin, bakin and freezin components, subsystems and whole systems far beyond the heinous launch and in orbit environment conditions to know the likelihood of failure is sufficiently low
But that is expensive. Test and fly is another approach we hear about but my engineering lecturers always told us that decisions late in a programme cost a lot more than earlier ones - you may have heard that too ;-)
There is no way round it making a machine safe enough for people is tough and costly and the outcome of this sub-$1 million study by Wyle could lead to many millions more in expense for whatever launch system administrator Bolden opts for
"The purpose of this task order is to develop a set of human system integration requirements for application to commercial spacecraft in support of NASA's Commercial Crew and Cargo Program (C3PO). The human system integration requirements developed under this task order shall be based on a review of existing Human Rating requirements, Spaceflight Human Systems Standards, Constellation Program requirements, Commercial Crew and Cargo Program Office operational concepts and requirements, and the Johnson Space Center Space Life Sciences Directorate Human Interface Design Handbook."
Despite all the allegations made about the safety of Ares I crew launch vehicle Hyperbola has heard from various quarters that the Constellation programme team was taking safety very seriously. Requirements over and above past programmes were being employed contributing to the ballistic cost growth
Looking at the synopsis text above you get the feeling that the debate - between Constellation and commercial supporters about what is a satisfactory level of rating - has got to the point where that key decision to resolve impassable management logjams has been taken: bring in the consultants
We hear lots about safety factors of one point this and one point that but in this blogger's humble opinion there is no substitute for shakin, bakin and freezin components, subsystems and whole systems far beyond the heinous launch and in orbit environment conditions to know the likelihood of failure is sufficiently low
But that is expensive. Test and fly is another approach we hear about but my engineering lecturers always told us that decisions late in a programme cost a lot more than earlier ones - you may have heard that too ;-)
There is no way round it making a machine safe enough for people is tough and costly and the outcome of this sub-$1 million study by Wyle could lead to many millions more in expense for whatever launch system administrator Bolden opts for
on November 23, 2009 7:20 PM | Reply
Human rating was a term invented in the early space age when NASA was converting ballistic missiles to vehicles for carrying crews. Since multiple ICBMs would be launched in an attack, their reliability was not a big issue. So they clearly had to be upgraded considerably for crew applications.
As even Mike Griffin noted in testimony to Congress a couple of years before he became administrator, a vehicle designed for carrying billion dollar satellites is going to be as reliable as possible. While a launch escape system is standard as the ultimate safety redundancy for crews, if there were any mods to the rocket that could make it more reliable for humans, they would be made as well for its unmanned satellite task.
"...no substitute for shakin, bakin and freezin components, ... But that is expensive."
But that is exactly what is done for the big satellite launchers. In addition, for the EELVs there has now been many actual flights. It is only through multiple flights that reliability statistics become "reliable". As Jeff Greason has pointed out, the flight rate for the Ares I would be so low that it would never graduate from the infant mortality phase of vehicle development and would never reach the reliability level that theoretical calculations by NASA predict.
"Test and fly is another approach we hear about but my engineering lecturers always told us that decisions late in a programme cost a lot more than earlier ones - you may have heard that too "
On the other hand, if one tries to eliminate every possible problem at the design phase, one would do studies forever and never build anything. I think most of us have also heard that there must be an acceptance that real world testing will bring out flaws, especially at the overall system level, that simply will never be found with just component tests, paper studies and simulations. Systematic expansion of the flight envelope, and fixing problems as they appear, has been very successful in development of safe aircraft. There's no reason the same approach should not work with launch vehicles.
on November 26, 2009 7:10 AM | Reply
Clark has hit the nail on the head, the synopsis plus Clark cover it well.
But the devil is in the details.
So I would add:
1 an escape system needs a trigger. Which is strain gauges, accelerometers, temperature/pressure sensors etc. Ares 1-X had 700 sensors scattered all over it. Since as Clark pointed out, the flight rate would never build up a sufficient experience-base, those sensors should be there for every flight.
2 another principle applied to manned flights is 'triple-redundancy'. Wherever possible. This is not done for unmanned vehicles.
3 finally a top-to-bottom analysis pinpoints 'single-point' failure modes. Even on today's Shuttle there are hundreds of those. Flight-management must sign off on each before every flight.
These items, not used for unmanned vehicles, are in addition to applying the best possible engineering in the first place.
And by-the-way, it would be easy to spend a billion or two for each vehicle you wanted to do it to.