Live data from Hacker News

How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

spectrum.ieee.org

21–30 of 52 posts

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#21

Perhaps I missed it when I read the article: Why did he originally have the hunch that the radio wouldn't work when Doppler shifted? Also, I guess something like Manchester encoding wasn't use because of the data rate limitations? IIRC Manchester encoding is self-clocking, which seemed like the issue here.

He didn't have a hunch, he just knew that the standard comprehensive test was skipped, which is why the standard test procedure only detected a problem, and he had to end early and go off book in order to reveal the doppler problem.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#22
post #15

Earlier quoted context omitted.

It lends human interest. Humans are inherently tribalistic, and one of the key ways we understand and categorize other humans is according to tribal affiliations.

This is an aspect of humanity that should be fought, not embraced and exploited.

I strongly disagree. Tribalism is what allowed humans to survive into the modern age, and is a force that in this era of individualism keeps societies from unraveling. Humans will be tribal, it's just a matter of how tribal lines are drawn. In the era of globalization, tribal lines are increasingly being drawn along lines of wealth, class, and education which is in my opinion worse than drawing them along geographical boundaries.

In practice, the alternative to geographic tribalism isn't some broadly-inclusive egalitarianism. It's relatively wealthy and educated folks like us receding into circles with other relatively wealthy and educated folks.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#23
Hopefully everyone realized that this article is from 2004 and Huygens completed its descent onto Titan in January 2005. The problem discussed in the article was successfully fixed, saving the mission, but an unrelated problem (also with the telemetry) means a modest chunk of data was lost:

> Huygens was programmed to transmit telemetry and scientific data to the Cassini orbiter for relay to Earth using two redundant S-band radio systems, referred to as Channel A and B, or Chain A and B. Channel A was the sole path for an experiment to measure wind speeds by studying tiny frequency changes caused by Huygens's motion. In one other deliberate departure from full redundancy, pictures from the descent imager were split up, with each channel carrying 350 pictures.

>As it turned out, Cassini never listened to channel A because of an operational commanding error. The receiver on the orbiter was never commanded to turn on, according to officials with the European Space Agency. ESA announced that the program error was a mistake on their part, the missing command was part of a software program developed by ESA for the Huygens mission and that it was executed by Cassini as delivered.

>The loss of Channel A means only 350 pictures were received instead of the 700 planned. All Doppler radio measurements between Cassini and Huygens were lost as well. Doppler radio measurements of Huygens from Earth were made, though not as accurate as the expected measurements that Cassini would have made; when added to accelerometer sensors on Huygens and VLBI tracking of the position of the Huygens probe from Earth, reasonably accurate wind speed and direction measurements could still be derived.

http://en.wikipedia.org/wiki/Huygens_(spacecraft)#Channel_A_...

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#24
post #20
post #2

Amazing work by a tenacious and talented engineer. Great story. (Question: What does his nationality have to do with it though? Seems like an odd addition to the headline.)

Exactly my thinking. Its rare to see American person did X or American company did Y. But it may be because I am on the inside looking out.

Disagree with your point about nationalities, not "American" per say.

Here's an example: look at the coverage of the missing Malaysia airlines plane.

Every single article I've read starts with this: "[insert nationality] experts have [insert what they've done]".

EDIT: Of course, that's necessary for journalism. It's information. How else are you going to distinguish between experts?

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#25

Hopefully everyone realized that this article is from 2004 and Huygens completed its descent onto Titan in January 2005. The problem discussed in the article was successfully fixed, saving the mission, but an unrelated problem (also with the telemetry) means a modest chunk of data was lost: > Huygens was programmed to transmit telemetry and scientific data to the Cassini orbiter for relay to Earth using two redundant…

Well, being pedantic for a moment, the problem never was fixed. The flightpath of Cassini was simply altered so that there was negligeable Doppler shift of the radio signal from Huygens, meaning that the inability to handle a Doppler shift.

Happily enough this was possible to do without using too much fuel, and indeed Cassini had more fuel available than planned, as the original launch and early course corrections had been so precise that fuel planned for course correction burns was available for other uses. One of the consequences of all of this is that Cassini is still operational today, more than 5 years passed the planned end of it's mission.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#26

Earlier quoted context omitted.

The French do the same. I see nothing wrong with it.

The whole set of top management taking off at one time? Maybe for a remote startup company, but c'mon as a taxpayer it must be infuriating.

As a taxpayer, you don't think about being infuriated as you're on the beach yourself during your own summer vacation.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#27

Perhaps I missed it when I read the article: Why did he originally have the hunch that the radio wouldn't work when Doppler shifted? Also, I guess something like Manchester encoding wasn't use because of the data rate limitations? IIRC Manchester encoding is self-clocking, which seemed like the issue here.

Take this with a grain of salt because it's been a while since I've done telecom/telemetry stuff, but even with a Manchester-encoded signal, one of the first steps in the demodulation process is clock recovery from the signal. This usually involves some sort of PLL (phase-locked loop), which has parameters to control it's freq tracking range. However there are trade offs for widening the range, in terms of time-to-lock and increased probability of locking onto noise etc. so for the original application of the system in earth satellites, they probably used a tight range to improve performance.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#28
> In proposing this more complex test with simulated telemetry, Smeds "had to argue with those who didn't think it was necessary," recalled JPL's Mitchell. Smeds was persistent and continued championing the test even after it was initially rejected. In the end, with the backing of Sollazzo and Huygens's project scientist, Jean-Pierre Lebreton, Smeds's plan was accepted because it was easy to do, even though hardly anybody seemed to think it was worth doing. On such seeming trivia US $300 million missions can turn: the simpler carrier-signal-only test, Mitchell noted, would never have uncovered any problems.

Is there any place you can go to escape the forces of willful ignorance?

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#29
post #22

Earlier quoted context omitted.

This is an aspect of humanity that should be fought, not embraced and exploited.

I strongly disagree. Tribalism is what allowed humans to survive into the modern age, and is a force that in this era of individualism keeps societies from unraveling. Humans will be tribal, it's just a matter of how tribal lines are drawn. In the era of globalization, tribal lines are increasingly being drawn along lines of wealth, class, and education which is in my opinion worse than drawing them along geographica…

Very much agree. Taken to the extreme, complete suppression of tribalism would not only be extremely difficult but also extremely harmful to society, as even the love between children and parents will be destroyed in favor of egalitarian love. Our biological nature causes children to prefer their parents more than any other parents, and in doing so, prefer their siblings (their parent's other children), more than any other children, and prefer their parents' siblings and parents more than any other elderly persons and adults. It is this nature, as it radiates outwards, that causes one to feel greater love to those who are similar than those who are different.

I don't want to live in a world where children are taught to betray their parents in favor of their state, where children are taught to report on their parents for their crimes, where brothers treat each other as strangers, where husband betrays wife and vice versa, at the same time. I've already read about it in 1984.

The harm in suppressing this aspect of humanity comes from having to involve the state. It takes the state to educate children to report on their parents crimes. It takes the state to force children to attend schools to study the same subjects. It takes the state to extract wealth from parents to distribute it to other children. It takes the state to enact taxes on children to distribute their income amongst other parents. Taken to the extreme - you'll get a very big, very powerful state, because it must involve itself in everyone's personal relationships, to ensure equality amongst all. And with a big state, you get corruption, you get misallocations of capital, you get a concentration of power, all of which would lead to a state that would monitor its citizens every communication, to protect its power in actuality, and to maintain equality and protect its citizens in name. And, as you reduce every person's natural relationships, people will form relationships with others, leading to divisions not amongst genetic or even geographic lines, but amongst other arbitrary lines such as class and wealth, government and private. As a group of people gain control of state, the actions of the state will begin to benefit some, but not others, arbitrarily, so that in actuality equality (in wealth, status, whatever) will always only be an illusion, and inequality will always manifest itself in other ways, so that all those resources supporting a very big state to maintain equality, are mostly wasted, except in so far as it provides an illusion of equality, which it can maintain for only so long.

Oh, wait.

Re: How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)

#30

Earlier quoted context omitted.

The whole set of top management taking off at one time? Maybe for a remote startup company, but c'mon as a taxpayer it must be infuriating.

As a taxpayer, you don't think about being infuriated as you're on the beach yourself during your own summer vacation.

It is the same in Sweden. Most of the country shuts down during summer and we're fine with it because we do too have our vacations right then.
Post reply on HN