Software woes keep NASA's new crewed missions grounded The United States Government Accountability Office has found that NASA's return to crewed space exploration will likely not commence in 2018, as planned, and will probably slip into 2019. And familiar technology integration challenges are partly to blame. The Office on Thursday published a Report to Congressional Committees ( …
Sounds like a square canister into round hole problem all over again...
"if suppliers deliver on time and the ESA stops finding dodgy welds and other defects."
Soooo... were there no penalties in the contract for not delivering on time? And letting the CEO's wife's nephew do the welding?
Can you imagine telling Elon Musk he has dodgy welds in his spacecraft? There Would Be Pain.
"each program must integrate its own hardware and software individually, after which EGS is responsible for integrating all three programs’ components into one effort at Kennedy Space Center"
I'm reading about Apollo (again) and the main lesson was they ALWAYS had some bloke paid to be on top of shit like this, trying to predict problems like this and proactively dealing with it before it hit the fan. He had the authority to say "big 'ol nope!" and get things fixed.
THAT NASA's long, long dead, however.
Then they got rid of Von Braun as being not quite American enough :-(
The ideal (which I think Shuttle lacked) is someone with the breadth of skills to understand (broadly) all of it and the authority to trade off margins on one system with those on another to ensure the whole thing gets the job done.
But the SW? FFS
There is no way on Earth (or anywhere else) that the task SLS (or Orion) has to fly is anywhere near as complex as Shuttle flight control. Apollo ran on 72KB of code. Shuttle on a couple of MB, written in a HLL. Orion (and SLS) has no wings, no control surfaces and no landing gear.
Apollo ran on 72KB of code. Shuttle on a couple of MB, written in a HLL. Orion (and SLS) has no wings, no control surfaces and no landing gear.
Yeah, but you've forgotten to allow for the Javascript framework.
"Soooo... were there no penalties in the contract for not delivering on time?"
There probably are*. But just how is collecting a penalty bring you back the lost time? **
* I don't know the contracts, obviously. But I do know a thing or three about EU procurement rules. Penalties look pretty nice on paper. However, the bar for invoking them is surprisingly high. Formulating terms of tender is an art in itself and requires a lot of both technical and legal knowledge, including stuff that isn't yet defined in codes and current court rulings that haven't been incorporated in the rule book yet.
It boils down to this: if you really know enough about anything you want to procure in order to make the terms of tender and the resulting contract airtight, watertight, foolproof, etc. - then you know enough about the damn thing to make it yourself. From scratch.
** Cancel the contract? Sure. Don't collect €€€, and start the procurement process again. Yeah, that'll save time. And if it's something highly specialised you're after, end up with a couple of bids by the same handful of companies as in round one. Only higher this time. And maybe with the company you've fired as a (possibly hidden) subcontractor. And quite possible having to pay the company you've fired damages for lost profit because of something the terms of tender doesn't cover, bacause see above.
Gene Kranz himself said in an interview that the large heavy lift rocket was the wrong approach. Its so expensive that they'll only every be able to afford one a year .
A better solution would have been a cheaper smaller rocker and assembling missions in orbit. That give you multiple launches, in space time for crew and experience at on orbit assembly. All stuff which is valuable skill and can be transferred to the inevitable change of program when a new administration comes in.
It gets back to Cost/Pound to orbit. Why does Elon think Heavy is a good idea its about lofting the weight fewer flights. As much as the fantasy of on-orbit assembly is its not as easy as it looks and has the added weights of interfaces.
There is an important place for heavy lift, think Mars or Moon re-supply will all be about moving enough supplies out of the gravity well cheaply enough.
Now if SpaceX can land all three boosters of a Falcon Heavy and reuse them ten times, the cost is going to plunge.
If the US launched one of each of it's current big launchers (Atlas V, Delta IV Heavy, F9 and Antares) it could put 77 tonnes in LEO within 1 week.
That capability exists right now. With no R&D. Just buy the launches
When SX's SLC 40 is working again that will rise to 99tonnes and (assuming a good launch) FH will increase that to 141 tonnes.
ULA is planning to phase out Atlas and Delta to go to Vulcan. Vulcan is expected to be 5.4m in diam. ULA have said they have looked at launching an Atlas V (3.81m in dia) with a 7.2m payload fairing. A ratio of about 1.889. On a 5.4m upper stage that would mean Vulcan could launch a 10.2m dia PLF (the baseline 8.4m SLS dia is well within its capability).
Most actual mass of big payloads is usually the propellant to get them where they are going. So with Vulcan operational you could probably launch any payload NASA could come up with on empty tanks and load propellant on orbit.
The real issue is this. SLS is being run out of Marshall at Alabama (Von Braun's old centre). They are trying to prove they can still direct and run a programme to specify an ELV designed to NASA specifications and operated by NASA for NASA payloads.
I'd suggest the evidence is that they can't. In a normal USG agency HQ would shut down the operation and either set up elsewhere or end in house provision, but NASA does not have control over wheather it can shut its own centres. Congress took that power so they would not create unemployment in the congresscritters home states. I don't think they feel the need to do with with any other Federal agency.
f the US launched one of each of it's current big launchers (Atlas V, Delta IV Heavy, F9 and Antares) it could put 77 tonnes in LEO within 1 week.
That capability exists right now. With no R&D. Just buy the launches
If you've designed your payload in 75 tonne chunks, to launch it on multiple rockets which have a smaller payload capacity, you have to go back to the drawing board and completely redesign your payload (e.g. space ship/station) into smaller sections.
In fact 90% of most of the payloads SLS is expteced to carry (because NASA doesn't have the budget to actually design anything for it except Orion so far) are propellant. So your "75 tonne chunk" is in fact a 10 tone payload, with a 5 tonne (empty) tank. The rest is propellant for the burns to get it to the Moon/Mars/Europa etc.
Propellant is the easiest thing to sub-divide, provided you're OK with on orbit transfer.
BTW NASA's Chief Technologist Robert Braun estimated better on orbit propellant management and transfer could cut the mass to LEO needed for a Mars mission by 60%, the biggest single factor in mission mass reduction.
Propellant is the easiest thing to sub-divide, provided you're OK with on orbit transfer.
And how hard can pumping pressurised cryogenic liquid oxygen and hydrogen be? After all, we do it on earth all the time. All we have to do is figure out a way to launch all those thousands of tons of pipes, tanks, pumps, monitoring, valves, spark-free electrical and hydraulic machinery that clutter up launch pads into orbit, and then make them work in microgravity, (I'm sure systems designed to handle cryogenic liquids on earth will work just as well when it's all floating around in zero-viscosity bubbles and globules.)
Admittedly it'll be a complete waste of time because there's no point humans going to Mars, obviously, but it'll keep a lot of aerospace engineers off the streets and their kids in college for a few more years, so it's all win really.
"NASA's also been very optimistic: despite knowing that space projects often experience cost blowouts, the reserve budget for the SLS is just two per cent of the its total."
2% would be way too small for comparably mundane projects like building a house. (Rule of thumb: 10%. With the estimate based on proper plans. Not including any changes to the design.)
My take from this is that NASA wasn't optimistic - NASA was desperately making the numbers fit. Politics and engineering really don't mix that well.
...rocket science is hard. But also worth the wait, as today's pictures from Saturn's rings so amply demo[n]strate.
Actually the pictures from Cassini perfectly demonstrate what a complete waste of time and money this whole boondoggle is. NASA wouldn't have even started bending metal if they'd tried to design this whole hypersonic pork barrel for the amount the entire, 20 year Cassini project cost, and if it ever produces 1% of the actual science of Cassini, I'll eat a life size replica of Ralph Lorenz's experiment ; on Huygens.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
