5 ms·

I worked on one component of the spacecraft which was a derivative of something we've built may times before. However, the test program was entirely unique to

I worked on one component of the spacecraft which was a derivative of something we've built may times before. However, the test program was entirely unique to Europa Clipper, and most of the cost was in this bespoke testing. The use of a "heritage" component served mostly to lower risk; it did not save much cost overall.

consumer451 2y ago
Oh cool, someone that actually worked on it! I was just thinking about how much pressure there must be on everyone involved in Discovery Class missions. People's entire professional careers, billions of dollars, so much at stake! Is the pressure something significant, or is it spread across so many people that there is little trouble sleeping at night?
- mandevil 2y ago
  The one that gets to me is the Huygens part of the Cassini-Huygens mission. Cassini was launched October 15th, 1997, Huygens separated from the Cassini carrier on December 25th 2004, and landed on Titan on January 14th, 2005. So it was in space for over 7 years before it got it's 90 minutes of time on the surface. That was actually towards the upper-end of expected lifetime on Titan- somewhere between 30-90 minutes was the expected lifetime of the lander(1). So you build and test and test and test (2) for years before launch, then the probe travels for seven years in outer space, and then it gives you an hour and a half of data and that's it. Your entire life for ... 15 years? Resolved in 90 minutes. On the shorter end of probe lifespan, it would have died before the first signal even reached Earth! 1: Officially I believe the expectation was "3 minutes" but that was a deliberate under-promise so that a success could be declared as long as they got any message at all from the lander on the surface: I have second-hand accounts that 30 minutes was what the scientists considered the minimum. 2: Even with all that testing, disaster almost struck. It wasn't until after the launch that someone realized even all of this testing had missed something important. The radio communications between Cassini and Huygens would be affected by the Doppler shift of Huygens hitting Titan's atmosphere, which would be unpredictable changes to velocity. After launch they had to rejigger when Huygens would be launched to a time when the signals would be perpendicular to the direction of travel so the shift wouldn't affect the radio waves so much that the Cassini receiver firmware (which could not be modified after launch) could still detect the signals. And also with all of that testing, ESA's instructions to the Cassini probe missed turning on one channel on the receiver and so half of the pictures that Huygens transmitted had nothing listening in and were lost.
  - consumer451 2y ago
    Thanks for that. It made me wonder: who was it? Were they in the shower, did they awake from a nightmare? Here is an HN post from 2014, "How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)" [0] Since the link has rotted away, here is the archive link to the IEEE story. [1] [0] https://news.ycombinator.com/item?id=7472495 https://news.ycombinator.com/item?id=7472495 [1] https://archive.is/3oj6P https://archive.is/3oj6P (archive.org is still not working reliably)
  - JumpCrisscross 2y ago
    > Your entire life for ... 15 years? Resolved in 90 minutes I ultimately decided against pursuing a career in aerospace engineering after talking to engineers who worked a similar time frame on a project only to watch it get killed in 30 seconds' debate in Congress.
- hydrogen7800 2y ago
  >Is the pressure something significant, or is it spread across so many people that there is little trouble sleeping at night? When you are on contract for something, you deliver to the contract, and are done when you successfully meet the customer's requirements. So in that sense you don't have the same exposure to the program risks. However, I've been part of one-off science missions before, and there is a different feeling beyond the contract obligations, though it's certainly abstracted through the many layers of sub-, sub-contracts.
mandevil 2y ago
And can introduce unusual failure cases for these bespoke missions. Mars Observer was lost in flight to Mars three decades ago, probably because of the inappropriate reuse of a satellite rocket engine. (1) The space environment out around Jupiter is really quite different from the environment that the JWST is facing around E-S L2 or what the Parker Solar Probe is facing right near the Sun. Even if the component is spec'd to handle the environment, you need to have actual educated humans (read: expensive labor) determine what those conditions will be, and then verify that the part will meet it, and that's where the money goes- to pay all of those humans. If you built even 15 Europa Clippers the cost per-item would come down enormously (because all of those people's work could be re-used), but since the 1970's NASA has not had the budget for multiple probes per missions. So every mission is bespoke, and has to be done again completely from scratch. 1: The engine was normally used for circularizing the orbit of a geosynch comm satellite, so within a few hours of flight. For doing a Mars Insertion burn it needed to sit fueled for months in outer space, which was not appropriately tested, and probably the fuel tank exploded in flight because of that.
- quotemstr 2y ago
  Yeah. I'm surprised that we don't have a standard deep space probe bus by now in at least serial production, at least for orbiters if not landers, rovers, and such. Each mission has unique requirements, but since payload mass costs are coming down, ISTM it should be possible to create a standard buss that meets most requirements most of the rime, even if it's heavier than a bespoke effort for any one mission.
  - accrual 2y ago
    There are somewhat standard busses. Though you're correct, many use custom busses for their specific missions. https://en.wikipedia.org/wiki/Comparison_of_satellite_buses https://en.wikipedia.org/wiki/Comparison_of_satellite_buses For example, the SSL 1300 apparently has hosted 118 satellites so far: https://en.wikipedia.org/wiki/SSL_1300 https://en.wikipedia.org/wiki/SSL_1300 Though maybe the distinction between "satellite" and "spacecraft" bears importance here.
  - 2y ago
throwup238 2y ago
There’s also politics at play. Public space agencies need to keep the absolute number of failures down to keep funding flowing, even if it costs absurd amounts to do it due to rapidly diminishing returns. This is especially important to flagship missions. When I took Ae105 at Caltech, the NASA MSL project manager explained it like this (I remember the numbers he used clearly): a mission might cost $500 million with an 80% chance of success, or they can spend twice as much to increase the chance of success to 95% by investing a lot more in upfront testing and R&D. Now, the smart thing to do - given a billion dollar budget - is to take that first option because if it fails you can try again and the probability of both attempts failing is only 4%, compared to 5% for the expensive single mission. Then you’ve got an 80% chance of having $500 million left over for a different mission. The public and decision makers react irrationally to any failure, putting funding for other missions and the entire program in jeopardy. NASA and ESA have to make some extremely suboptimal decisions to make sure that funding doesn’t get catastrophically cut. The above is the example the instructor used to easily illustrate his point but he said the real numbers are even more stark. Often times the cost savings of just building a second copy of the payload along with the first means it costs $600 million for the first attempt, and only $200 million for the second (the cost of the launch vehicle and keeping people on staff), saving hundreds of millions overall.
- frickinLasers 2y ago
  It's really frustrating that there's not a way to educate the public on simple concepts like this. You'd think the out-in-the-open development approach of SpaceX, for instance, would make it blatantly obvious how much money they're saving by permitting failure...yet the news continually spins their less-than-perfectly successful test launches as undesirable. And space spending is perhaps one of the least consequential areas of government where this failure of the herd to comprehend reality applies.
  - downvotetruth 2y ago
    The public are smarter than you give them credit for. For the most part space missions are contests between governments and those governed with science as a consolation price. If you are part of team A do you want to have to come up with reasoning on why you lost? No. Does anyone care if a billionaire's toy gets blown up? No. The obvious choice to eliminate all risk would be not to play, but that is not an option; the other is to delay and delay some more. Thus, SLS.

consumer451 2y ago

Oh cool, someone that actually worked on it! I was just thinking about how much pressure there must be on everyone involved in Discovery Class missions. People's entire professional careers, billions of dollars, so much at stake! Is the pressure something significant, or is it spread across so many people that there is little trouble sleeping at night?

mandevil 2y ago
The one that gets to me is the Huygens part of the Cassini-Huygens mission. Cassini was launched October 15th, 1997, Huygens separated from the Cassini carrier on December 25th 2004, and landed on Titan on January 14th, 2005. So it was in space for over 7 years before it got it's 90 minutes of time on the surface. That was actually towards the upper-end of expected lifetime on Titan- somewhere between 30-90 minutes was the expected lifetime of the lander(1). So you build and test and test and test (2) for years before launch, then the probe travels for seven years in outer space, and then it gives you an hour and a half of data and that's it. Your entire life for ... 15 years? Resolved in 90 minutes. On the shorter end of probe lifespan, it would have died before the first signal even reached Earth! 1: Officially I believe the expectation was "3 minutes" but that was a deliberate under-promise so that a success could be declared as long as they got any message at all from the lander on the surface: I have second-hand accounts that 30 minutes was what the scientists considered the minimum. 2: Even with all that testing, disaster almost struck. It wasn't until after the launch that someone realized even all of this testing had missed something important. The radio communications between Cassini and Huygens would be affected by the Doppler shift of Huygens hitting Titan's atmosphere, which would be unpredictable changes to velocity. After launch they had to rejigger when Huygens would be launched to a time when the signals would be perpendicular to the direction of travel so the shift wouldn't affect the radio waves so much that the Cassini receiver firmware (which could not be modified after launch) could still detect the signals. And also with all of that testing, ESA's instructions to the Cassini probe missed turning on one channel on the receiver and so half of the pictures that Huygens transmitted had nothing listening in and were lost.
- consumer451 2y ago
  Thanks for that. It made me wonder: who was it? Were they in the shower, did they awake from a nightmare? Here is an HN post from 2014, "How a Swedish engineer saved a once-in-a-lifetime mission to Titan (2004)" [0] Since the link has rotted away, here is the archive link to the IEEE story. [1] [0] https://news.ycombinator.com/item?id=7472495 https://news.ycombinator.com/item?id=7472495 [1] https://archive.is/3oj6P https://archive.is/3oj6P (archive.org is still not working reliably)
- JumpCrisscross 2y ago
  > Your entire life for ... 15 years? Resolved in 90 minutes I ultimately decided against pursuing a career in aerospace engineering after talking to engineers who worked a similar time frame on a project only to watch it get killed in 30 seconds' debate in Congress.
hydrogen7800 2y ago
>Is the pressure something significant, or is it spread across so many people that there is little trouble sleeping at night? When you are on contract for something, you deliver to the contract, and are done when you successfully meet the customer's requirements. So in that sense you don't have the same exposure to the program risks. However, I've been part of one-off science missions before, and there is a different feeling beyond the contract obligations, though it's certainly abstracted through the many layers of sub-, sub-contracts.

mandevil 2y ago

And can introduce unusual failure cases for these bespoke missions. Mars Observer was lost in flight to Mars three decades ago, probably because of the inappropriate reuse of a satellite rocket engine. (1) The space environment out around Jupiter is really quite different from the environment that the JWST is facing around E-S L2 or what the Parker Solar Probe is facing right near the Sun. Even if the component is spec'd to handle the environment, you need to have actual educated humans (read: expensive labor) determine what those conditions will be, and then verify that the part will meet it, and that's where the money goes- to pay all of those humans. If you built even 15 Europa Clippers the cost per-item would come down enormously (because all of those people's work could be re-used), but since the 1970's NASA has not had the budget for multiple probes per missions. So every mission is bespoke, and has to be done again completely from scratch. 1: The engine was normally used for circularizing the orbit of a geosynch comm satellite, so within a few hours of flight. For doing a Mars Insertion burn it needed to sit fueled for months in outer space, which was not appropriately tested, and probably the fuel tank exploded in flight because of that.

quotemstr 2y ago
Yeah. I'm surprised that we don't have a standard deep space probe bus by now in at least serial production, at least for orbiters if not landers, rovers, and such. Each mission has unique requirements, but since payload mass costs are coming down, ISTM it should be possible to create a standard buss that meets most requirements most of the rime, even if it's heavier than a bespoke effort for any one mission.
- accrual 2y ago
  There are somewhat standard busses. Though you're correct, many use custom busses for their specific missions. https://en.wikipedia.org/wiki/Comparison_of_satellite_buses https://en.wikipedia.org/wiki/Comparison_of_satellite_buses For example, the SSL 1300 apparently has hosted 118 satellites so far: https://en.wikipedia.org/wiki/SSL_1300 https://en.wikipedia.org/wiki/SSL_1300 Though maybe the distinction between "satellite" and "spacecraft" bears importance here.
- 2y ago

throwup238 2y ago

There’s also politics at play. Public space agencies need to keep the absolute number of failures down to keep funding flowing, even if it costs absurd amounts to do it due to rapidly diminishing returns. This is especially important to flagship missions. When I took Ae105 at Caltech, the NASA MSL project manager explained it like this (I remember the numbers he used clearly): a mission might cost $500 million with an 80% chance of success, or they can spend twice as much to increase the chance of success to 95% by investing a lot more in upfront testing and R&D. Now, the smart thing to do - given a billion dollar budget - is to take that first option because if it fails you can try again and the probability of both attempts failing is only 4%, compared to 5% for the expensive single mission. Then you’ve got an 80% chance of having $500 million left over for a different mission. The public and decision makers react irrationally to any failure, putting funding for other missions and the entire program in jeopardy. NASA and ESA have to make some extremely suboptimal decisions to make sure that funding doesn’t get catastrophically cut. The above is the example the instructor used to easily illustrate his point but he said the real numbers are even more stark. Often times the cost savings of just building a second copy of the payload along with the first means it costs $600 million for the first attempt, and only $200 million for the second (the cost of the launch vehicle and keeping people on staff), saving hundreds of millions overall.

frickinLasers 2y ago
It's really frustrating that there's not a way to educate the public on simple concepts like this. You'd think the out-in-the-open development approach of SpaceX, for instance, would make it blatantly obvious how much money they're saving by permitting failure...yet the news continually spins their less-than-perfectly successful test launches as undesirable. And space spending is perhaps one of the least consequential areas of government where this failure of the herd to comprehend reality applies.
- downvotetruth 2y ago
  The public are smarter than you give them credit for. For the most part space missions are contests between governments and those governed with science as a consolation price. If you are part of team A do you want to have to come up with reasoning on why you lost? No. Does anyone care if a billionaire's toy gets blown up? No. The obvious choice to eliminate all risk would be not to play, but that is not an option; the other is to delay and delay some more. Thus, SLS.