http://www.nasawatch.com/archives/2006/05/...finally_so.html : "Yikes! ITAR issues aside, it is obvious why NASA sat on this report for a year -- it has a high embarassment coefficient."

I put this report in this place on the website because of the crucial importance of robotic rendezvous and docking to Mars sample return. Overall causes of the failure were our familiar old friends: too small a team, lack of a systems engineer, over-hasty schedule, and above all "High Risk, Low Budget Nature of the Procurement". Seems to me I've heard that song before...

I still don't see the need to do it "half blindly", without both sides communicating with each other.

So is the data ITAR sensitive because it reveals gross incompetence? Are there no lessons to be learned. Ever?

Another fine mission managed by MSFC. The wrong center was managing a mission that they had no expertise in and not smart enough to ask or listen to it.

Keith Cowing's coverage is interesting - just wait until Mark Wade has a go, too!

It seems that the root cause of this one is, once again, a lack of management quality rather than a lack of quality management - someone should have called a halt well before launch.

And *why* did the report take 18 months to escape into the wild?

Bob Shaw

They were racing the Huygens investigation release

Doug

So much for Griffin's pledge that the sock is out of the scientists mouth...or does this mean that there were no scientists involved in the Dart debunkle?

In the years leading up to the Discovery- er Columbia Explosion, we were told that Shuttle Technology was mature and there were hard design freezes. We were also told engineering degrees where not essential to managing a space enterprise, and managers started creeping in with experience making plastic diapers. Unfortunately, these same entities lack the analytical skill necessary to separate good engineers from fudge punchers...

Well - it wasn't a science mission - but at some point the role of cutting edge engineering and science merge into one fuzzy middle ground.

However - the new press/outreach guidelines from a few months ago certainly don't take "the sock" out of anyones mouth

Doug

Where is this from? Unless you are talking about the USAF, it doesn't apply to NASA.

DART was originally proposed to cost 47M dollars, and ended up costing 110M. So it may have been "faster" but it sure as hell wasn't "cheaper."

Not every failure is caused by lack of spending. I'm going with incompetence on this one.

What strikes me as simply insane is that the USAF seems to have flown largely the same mission with total success, for less money too. One agency or the other should have been told not to fly - in this case, the results indicate it should have been NASA!

Bob Shaw

Jesus Christ. http://www.space.com/news/060516_dart_mishap_update.html :

"Scott Croomes, the NASA engineer who chaired the DART Mishap Investigation Board, said the mission's failure stemmed from a combination of spacecraft navigation errors and missteps by the vehicle's NASA and contractor design teams...'What we found in the investigation is that this almost worked,' said Croomes, who also serves as test laboratory deputy manager at NASA's Marshall Space Flight Center (MSFC) in Huntsville, Alabama. 'I think there are quite a few lessons available for other programs.'...

"Orbital Sciences officials did not comment on the newly released mishap summary itself, but added that while DART did not accomplish all of its goals, the mission did yield several lessons for future autonomous spacecraft systems. 'It was a challenging, experimental mission to begin with and it was all new technology,' Orbital Sciences spokesperson Barry Beneski told SPACE.com of the DART mission. 'Finding out what went wrong and why is an important part of the process to learn so that next time the technology is tested out we'll get even better at it.'

"Croomes added NASA should not allow the DART mission mishap to prevent subsequent tests of autonomous spacecraft rendezvous systems. 'This was the first time the United States attempted an autonomous mission like this,' Croomes said. 'NASA should not shrink back from attempting things that are difficult.' "
______________________________________________

DIFFICULT!! CHALLENGING!! NEW TECHNOLOGY!!! "ALMOST WORKED"!!!! "EVEN BETTER AT IT"!!!!! Russia has been doing this routinely for THIRTY-NINE YEARS-- as part of a space program in which they practically have to send a guy out to the launch pad to light the rocket's fuse!!

Well, Russia makes sure both sides communicate with each other. That's why I don't understand this DART test - it implies that one side is silent. It's harder to implement, but what's the point? It's only needlessly complicates everything.

This is in fact not true. My very reliable sources confirm they actually use a monkey for the job. And it's not actually lighting the fuse as much as pulling an electrical switch on the launch pad. This method greatly alleviates the need to run electrical cables all the way to the firing room and all other hassles.

Er... ...yup!

And still the USAF vehicle flies...

Bob Shaw

Two of my sources E-mail me to tell me I'm partially mistaken about saying that DART would just have repeated what "the Russians have been routinely doing for 39 years." First:

"From what I know of the 'KURS' docking system the Russians used (I don't know if they still use it) this isn't a valid comparison. The way I understand things, the Russians flew their vehicles up to proximity operations space (i.e. they did the rendezvous maneuvers) from the ground, and then the KURS system only took over when they got within some prox-ops boundary (i.e. 250 m) -- at which point KURS, which is a simple computer that reacts to radar signals from the target and the chaser, fires the engines to line the vehicle up and dock it.

"Not only isn't this part very hard, but KURS also relies on intervention from the ground when it screws up, which it regularly does, or at least did during the MIR days.

"DART, on the other hand, flew all the way from launch without any possible intervention from the ground, and it flew the whole rendezvous part and the prox-ops part autonomously and bullseyed the MUBLCOM spacecraft using GPS and other onboard orbit determination stuff that has never been tried before.

"Several very smart friends and collegues of mine worked among the Orbital Sciences engineers after the DART Mission to figure out what went wrong, and their response was very similar to this report -- i.e. that this was ultimately a very successful mission on which a whole lot of very new technologies worked very well. In the end the hardware all worked, and everything was there for the mission to be successful, but some poor software testing screwed them. The fix is very simple. The system would work as-is with the software fix.

"I try not to spend much of my time apologizing for NASA, and I certainly think the ITAR crap that they are using to hide the real report here is a crock. That said, I think it is incorrect to characterize this mission as a complete failure. It was a difficult set of problems, and they almost got all of them right on the first try.

"As far as I know, the Russians have never done anything like this, and I don't even know that they can build the KURS system any more. They were having a hell of a time doing so during MIR. I recall hearing that they had to recycle the computer out of old Soyuz modules because they couldn't build them and the Ukranians didn't want to sell them any new modules for a reasonable price."

Second (from Jeffrey Bell): "Actually, they haven't [been doing what DART tried for 39 years]. Progress auto-dockings frequently fail in the later phases, and humans have to take over manual control with a Mk1 Mod0 eyeball and joysticks."
______________________________

My initial reply is that I stand partially corrected -- but the facts remains:

(1) This docking system OFTEN works fully automatically for the Russians, whereas the US hasn't yet done it successfully once -- and...

(2) It shouldn't be that much harder for a spacecraft to carry out the pre-docking rendezvous sequence autonomously, using data which it acquires from GPS rather than from ground tracking stations, and/or using autonomous optical sightings of the target in orbit (such as many manned crews have done using a sextant and computer). I still think there is no excuse for a screwup this big, and I still think that Goldin's stupid semi-religious belief in Faster-Better-Cheaper claimed still another victim this time.

This, in turn, has ignited a further multiple exchage between me and my different sources on just how innovative and how successful DART really was -- which I'm still trying to digest. More details later (I hope).