A mistake like that on the way to Mars would be fatal
|
|||||||||||||||||||
|
| |||||||||||||||||||
|
« Smart Parking Systems Make It Easier to Find a Parking Space | Main | Anatomy of a Crash-Test Dummy »
A mistake like that on the way to Mars would be fatal
TrackBack URL for this entry:
http://blogs.spectrum.ieee.org/cgi-bin/mt/mt-t.fcgi/3429
This page contains a single entry from the blog posted on October 4, 2007 8:51 PM.
The previous post in this blog was Smart Parking Systems Make It Easier to Find a Parking Space.
The next post in this blog is Anatomy of a Crash-Test Dummy.
Many more can be found on the main index page or by looking through the archives.
Comments (4)
While the story is very interesting, I found utterly disturbing the biased and unnecessary comments introduced by the author.
I think as engineers we should be well past the point of publicly mocking others for their mistakes. No matter how obvious these facts look now, I think they were not trivial or the author would have pointed it out as soon as he heard about the problems.
Space technologies are complex, lots of thought is put on each and every action that is taken, they often manage to make the impossible happen. The morale of this story should be that even when you prepare for everything something unexpected can happen, and we can be happy that the people on board had the appropriate training to mitigate the impact of the problem.
Let's look towards the future, and when we talk about ISS, let's forget about differences between US, EU, Russia, Japan, ... we are all on the same spacecraft and we should be rowing together.
Posted by Miguel E. Gil
|
October 17, 2007 9:30 PM
Posted on October 17, 2007 21:30
I am inclined to agree that the tone of the article was inappropriate. Mr. Putin's recent comments regarding the cold war seem to be reflected in the attitudes of some on this side of the Atlantic as well.
Power quality problems on the ground can be complicated by many factors, and politics is usually one of them. Someone's credibility or even their job is at stake, and they want an answer that will make them seem to be effectively dealing with the situation. Early on, in extreme situations, without adequate data, lots of wild explanations can surface when equipment malfunctions that need to be logically eliminated (usually with simple procedures at the start, including visual check-out and continuity tests, then with more complex methods and technologies).
I recall a major power plant with technicians afraid of working at a particular control panel due to recurrent electrical shocks. The potential need for a massive model of the power system had been raised. It turned out, fortunatly before lots of money was unnecessarily invested, that the panel ground bus simply hadn't been connect to the ground wire entering the panel.
Throwing mud across the Atlantic isn't going to reduce tensions and make problem solving more effective. (Is that a motive when various groups are supposedly cooperating, but are really competing for prestige and the opportunity to get or award the next contract?) I certainly hope that greater professionalism develops before lives are at stake on an interplanetary mission. Did someone forget to install a power quality monitoring device with data that is regularly transmitted to earth on the power buses on the station going to key computer systems? (Tsk, tsk... the Russians wouldn't have been so quick to blame "dirty" power if that data had been available for the time when the computers malfunctioned, with a valid GMT time stamp, of course.)
As for the triply redundant computer network with one line capable of shutting all three down, was it really all that insane? A power glitch that could kill one or more critical computers might normally be perceived to be less dangerous than a brief loss of attitude control relative to a very brief power quality event, particularly if you assume that other systems are doing their job properly, such as those regulating atmosphere.
Inertia tends to give massive objects an element of stability that is not sensitive to sudden changes occuring in the few cycles or less of most power quality problems. The Russians should have known if Voltage instability or transients could have affected their computer's operation and what level of Voltage instability would have been required to cause a problem, so power quality data would have eliminated the issue regarding the solar panel produced power.
One is still given pause to question why an FMEA didn't conclude that one line failing due to corrosion, a common problem in any environment, could cripple the attitude control computer system. Is the international space station team backsliding on FMEAs (Failure Mode and Effects Analyses)? Too bad the article didn't investigate who had responsibility to evaluate the reliability of the attitude control computer system, and whether the FMEA related to that system identified the problem. If not, someone needs to review the related documentation. If the FMEA did identify the problem, why didn't someone fix it before the station got off the ground?
Engineering should be about solving problems, including those that are administrative, when necessary, not hurling mud.
Posted by StealthSquirrel
|
October 18, 2007 2:44 AM
Posted on October 18, 2007 02:44
On the way to Mars, radiation may be a hazard eclipsing computer reliability risk.
Posted by Ed
|
October 18, 2007 5:12 AM
Posted on October 18, 2007 05:12
I'm sure making a space station isn't comparable to making a sand castle at the beach. If I were versed in aerospace engineering, somebody somewhere would have told me that water condensation is a big problem in a closed capsule with a regulated environment housing humans. My concern is that they used a power coupling switch that is susceptible to corrosion in the first place. How often do they have to plug and unplug that thing? As much as a laptop you take to and from work? Wouldn't those computers be permanently coupled? Why couldn't they seal the coupling conductors from both air and water?
Posted by Brian Croner | October 30, 2007 1:02 PM
Posted on October 30, 2007 13:02