Slashdot Mirror
Saving Huygens
TazMainiac writes "This months IEEE Spectrum is running an article on how a Swedish scientist discovered that the Titan probe Cassini had a communications flaw that would cause it to lose all data sent back from the Huygens lander as it plunges into Titan's atmosphere. The problem - Doppler effect. The fix: go read the article."
It is amazing that the problem with the reciever was detected. It was more amazing to read what they went through to document and present the problem. It also says something about the relationship between NASA and it's subcontractors when they can accept a receiver design and not sign a standard non-disclosure agreement so that they can see the specific design elements. If they had done so, they would have been able to see the problem before launch. However, having read the article, the complexity of the mission is such that I am possibly more amazed that more didn't go wrong.
http://www.busyweather.com/
God bless good 'ole auntie.
:)
Saw this on Horizon yesterday evening.
Always nice to see a simple solution.
Now if only NASA could find a simple solution to conversions between imperial and metric, or not undoing bolts
so in this case the problem was indeed caused by the fact they couldn't RTFM to check the supplier had done the job correctly.
I think it is all summed up with the line "An Alenia Spazio spokeswoman said that none of the company's officials were available to comment because of a company-wide summer vacation period."
That's nice. Did they actually explain how the Doppler shift affected the BPS coding used in the Huygen's telemetry, or describe how the problem was missed, or tell the story of Boris Smeds pushed through his test and ended up modifying it on the fly? And does every reader of IEEE Spectrum get the BBC on their TV? (hint, Spectrum has a global circulation)
-- disclaimer, I edited (and did some reporting for) this story.
"Just once, I'd like to meet an alien menace that wasn't immune to bullets." -- The Brigadier, Dr. Who
While slashdot has category icons, it really REALLY needs a "hero" tag, like you'd see on Fark.
This engineer that found the problem and rallied against opposition to see that this gets fixed is, in my opinion, a total hero. The world would be a much better place if more people like him were around!
I have no problem with your religion until you decide it's reason to deprive others of the truth.
It's actually fairly common...when you have a bunch of smart people working on difficult problems they tend to breeze through the easy parts, and don't necessarily double check each other's work because of "Jim has 2 PhD's...he'll get that right" syndrome.
Reminds me of taking caculus exams...it was always something dumb like switching +/- or "1+1=3" that I got wrong...not the partial differentials.
It just shows that no matter how smart you are, if you hurry and don't pay attention to every trivial detail you'll make mistakes.
If God had had a computer it would have taken him 7 months to create the earth...if he even bothered to do it at all.
Were you doing SDR in 1997 when Cassini was launched? Were you doing it in 1987 when Cassini was being designed?
All the bits don't come to earth because Cassini doesn't have continuous data transmission to the Earth. That would be extremely expensive. DSN time is charged out the wazoo. I don't know Cassini specs, but most missions plan on recording data and shipping it back to Earth when DSN time is allocated. It isn't continous. DSN has other things to do.
You want to send the raw analog signals ("verbatim"?) back to Earth for signal processing in real time? That doesn't even really make sense.
I had a friend key up a dead carrier on 446 MHz while standing at the side of the road, put my car radio in SSB mode (which makes the dead carrier sound like a plain sine wave) and then I drove past him at around 100 km/hr.
At that speed, it causes a total shift of around 80 Hz, which is easily heard by ear. Quite cool.
Boris was key to finding the problem, but the solution was mostly a trajectory one. The article has a rather mangled explanation of the trajectory changes, and it treats the people who worked on it rather anonymously.... just calling them 'navigators'. The fact is that the trajectory changes done to save the probe mission were far from trivial.
Boris deserves a lot of the credit for saving Huygens, but several other people deserve credit but have been rather anonymous outside of ESA/JPL. 'Saving' Huygens was a team effort, and a lot of people played a part. There are a handful of other key people that the article doesn't mention at all.
Also there are a few factual errors in the article.... NASA couldn't simply sign a NDA to get the specs for the receiver, and there was a lot of effort by NASA to get the specs. Even after the problem was detected, Alenia resisted sharing information for many months.
There are 10 types of people in this world, those who can count in binary and those who can't.
If Boris Smeds was a hero for the Cassini-Huygens Mission, Ann Harch was a heroine for the $150 Million Near Earth Asteroid Rendezvous(NEAR) spacecraft mission to asteroid Eros. When an engine burn to reduce the spacecraft's approach speed went awry, the spacecraft tumbled violently and contact to it was lost for 36 hours. When they regained contact, the spacecraft was rushing at a great speed that it would pass Eros about two weeks ahead of schedule. The spacecraft had to be given new commands within 12 hours or it would miss clicking Eros altogether in the flyby. In the nerve-racking moments, Ann Harch and her colleagues rewrote new command sequences for the original program that took them 6 months to write. In a nail-biting finish the commands reached 8 minutes before the Eros flyby, just in time to enable capturing images of asteroid Eros. Ann Harch was later honored for her efforts with an asteroid being named after her.
How a down-to-the-wire computer fix at Cornell enabled a troubled spacecraft to take images of an asteroid
CU astronomer who helped save mission receives a celestial gift
Boris Smeds did a great job in replacing lots of expensive tests with a series of trivial, yet critical tests.
Why weren't simple tests like these used while the spacecraft was on the ground?
These are obvious problems. When you take a transmitter and throw it into a planetary descent, this is what should be expected.
It is shocking to me that a transceiver pair isn't tested by the team assembling the spacecraft before launch!
If it can be tested in 2 days when it's in space, 48 light-speed minutes away, why can't it be tested on the ground, fully assembled?
Engineering isn't a science, but I expect that engineers desigining projects like this should be using thorough unbiased scientific testing, not only thorough design.
If they slip up like this in non-destructive tests, one has to wonder about how tests on the resistance to physical damage are carried out?
Do they simply make assumptions that all nuts & bolts are manufactured to spec? Do they assume that all parts will withstand the forces that they are requesting in spec sheets?
How can a cost of hundreds of millions of dollars be justified in creating such craft, when basic, inexpensive testing isn't being carried out? If the test would cost 30,000$ (a few days of well-paid outside experts time, plus expenses and travel), as a critical portion of a 300,000,000$ mission, how is it not done?
The only agreement that should be accepted by an agency purchasing a part is that they won't use the specifications of that part to replicate exactly the same device. I'm sure that they paid a high premium for the transceiver. Why wouldn't they have access to the documentation and spec sheets? This use of NDA's is dangerous.
I agree the recovery was a team effort, but the fact remains that what Smeds did was a rarity: a singular individual effort that, if it hadn't occured, would have resulted in disaster. Thus we felt he deserved some serious kudos and so the article focused on him.
I'm in no way deingrating the amazing and creative work that the trajectory guys did. But think of it like this: If any one of those guys were absent from the project, because of a sabbatical, or, God forbid, an accident, chances are that the mission still would have been salvaged.
The same cannot be said about Smeds during the period between being told to do a test and coming back to ESA with the results -- it's fair to say that many, if not most, engineers would have just developed a carrier wave only test as originally planned, or wouldn't have bothered to persist with the more complex test after being turned down (after all, who's looking to get into trouble to do extra work?), or might not have had the insight required to modify the test on-the-fly when the downlink started showing problems.
The situation is analogous to Apollo in some ways: a lot of people helped design and build the LEM (Tom Kelly is one of my personal engineering heroes), but John Houbolt deserves his place in history for pushing NASA onto the LOR architecture path in the first place.
As for what we said about NASA and the NDA, I'll just have to say we stand by Oberg's reporting. But if you have something that shows we really did get it wrong, I'd be more than happy to look at it and print a correction if warranted.
"Just once, I'd like to meet an alien menace that wasn't immune to bullets." -- The Brigadier, Dr. Who