Posted by
Hemos
on from the good-week-end-for-NASA dept.
lonedfx writes "After nine delays, STS 103 has launched and its crew should service the Hubble space telescope in the next few days. Hopefully Hubble will be soon back to give us great pictures to stare at. "
Re:Lest We Forget...
by
Anonymous Coward
·
· Score: 4
"Except that I read somewhere that, unofficially, the shuttles' software has never been able to handle a year change correctly, anyway..." You are right, the software cannot handle a year-end rollover. The navigation routines that tell the orbiter where to de-orbit perform their calculations in the M50 (Mean of Aries 1950) coordinate system. This system is an inertial coordinate system, which means that it is referenced to a fixed point in space. Now, in order to deorbit into the right window, we need to tell the orbiter how the earth is positioned within this inertial reference frame. To accomplish this we need to use a Rotation Nutation Precession (RNP) matrix to transform the orbiter's position in M50 coordinates to a position in Earth Centered Earth Fixed (ECEF) coordinates. The RNP matrices are computed on the ground for various times during the year and each are only good for about 3-months on either side of their designed time. If a launch slips over the year end, the RNP matrix is no longer any good, no matter how stale it is (1 day or 1 month or [1 year - 1 hour]). The good news is that we have now designed an improvement to the flight software that calculates the RNP matrix on-board. This would be able to get us out of this predicament, but I would venture to guess that the program still would never have a mission that would cross the year-end boundary in orbit. Actually, we could probably get out of it by patching a new RNP while in orbit, but that is a risk that the program is most likely unwilling to take. - These views are mine and do not necessarily represent those of my employer.
Purpose? I don't think so.
by
Tau+Zero
·
· Score: 3
Its original aim -- to spot extra solor planets
BZZZT!Sorry, but thanks for playing.The biggest original purpose of the Hubble Space Telescope was to determine the value of the Hubble Constant by calculating the luminosity of Cepheid variable stars in galaxies too distant for ground-based telescopes to perform the work.Extra-solar planets cannot be reliably detected by their reflected light at this time (only one freakish case has allowed this), and Hubble has not been involved in the detection of these planets; it has been done with ground-based telescopes collecting extremely precise spectrometry data. -- The Karma Century Club is taking new members.
-- Time is Nature's way of keeping everything from happening at once... the bitch.
I don't strap my car to two solid rocket boosters and send it into orbit, at least not regularly. This is a very bogus comparison, the tolerances (e.g. heat, stress) required for normal operation of a car, and the tolerances required for a man rated launcher are not comparable.
You also don't baby your car the way that NASA babies it's space shuttles, nor do you meticulously repair the maybe slightly defective parts the way NASA does with the space shuttle. Both vehicles are designed for certain limits, and both operate within those limits.
Unfortunately is bad design thats causing the delays, the shuttle design was a compromise (much like the current ISS debacle) due to repeated budget cuts. Initially design targets were to have something that could fly and turn around again in a matter of days or weeks, they were looking for a shuttle that could fly (at a minimum) twenty or thirty times a year. What they ended up with falls very short of the mark, something that was actually more expensive to launch than a normal rocket. The shuttle is an aging launcher, with design compromises at every turn, it shows.
Compromise of the initial design vision isn't what's delaying the shuttle in any way. If the shuttle was built to be used 20-30 times a year, it still would undergo the safety checks that NASA is giving it now, and if they wiring was bad on one of those shuttles, I'm sure NASA would take the time to check each one, and it would probably still have an excessive number of miles of wiring. What's causing the delay here isn't the design of the shuttles, whatever form that design may be, but NASA's anality (analism, analness?) about safety. Rightly so, I might add, since protecting the lives of the passengers should be the top priority.
Hopefully, VentureStar won't be as big a disappointment, although personally I still think they should have gone with the Delta Clipper (at least they had a working prototype in hardware).
Yeah, and in the meantime, what do you use? The current 'bad design' that's really only failed once in 20 years of usage, right? NASA's not going to make the mistake again of stopping all space flight for development of a launching mechanism when they have a perfectly good one. And given the VentureStar or the Delta Clipper's design snafus, I'm glad they're still researching it.
No way, we've already stretched the design lifetime well beyond the inital estimates.
Then how can it be a bad design? It may not fulfill the initial design vision, but as anyone who's ever worked in design can tell you, the vision changes as you work on a project. The fact that the shuttles have lasted this long with only problems to the launch device, not the orbiter itself, shows exactly how great and durable a design it actually is.
Odd... it seems to me that every time I turn around, Hubble has broken again... are we attacking the symptoms here or the disease, so to speak?
You might want to learn a little more about the Hubble mission. When this thing was put together, several servicing missions were anticipated to replace components as the wore out, and to upgrade or swap out instruments. These missions are planned on a three year cycle.
The ORIGINAL plan was to bring the Hubble back to earth for refitting every 5 years, however the ability of shuttle crews to do in-orbit servicing has made this unnecessary.
IN FACT, the Hubble has been one of the greatest successes of the entire space program.
This mission not only includes replacement of gyroscopes (that lasted longer than originally planned), but upgrading some instruments and the main Hubble computer system.
Venture Star killed the real competition
by
Tau+Zero
·
· Score: 5
Back in the days of the unjustly-maligned Strategic Defense Initiative, there was a group of people who were charged with getting SDI satellites into orbit. They had advantages over NASA in this regard. They were tied to neither the existing launch-vehicle fleet nor the high-cost aerospace contractors; this let them examine things afresh. They took out a clean sheet of paper and tried to design, not a vehicle, but a program for getting a vehicle.
What they wound up with looked very different from the Space Scuttle and VultureStar. It was a squarish bullet, covered in thermal-blanket material originally developed for the Shuttle. It was not terribly fussy about its engines; it could have flown on J2's or RL-10's. The innovations were several:
It was designed to land tail first, under power. This takes advantage of the engines, which are along for the ride, to provide landing capabilities. This eliminates the need to be able to glide subsonically; the glideslope, flare and landing maneuver required by Shuttle (and VentureStar) is unnecessary.
It did not have wings. This saved a great deal of weight in the airframe.
Its landing gear was a system of struts and pads. No wheels required. This saved more weight.
The pilot went away also. When so many missions are just putting unmanned birds or cargo in orbit, why carry people along all the time?
The vehicle was to be called the Delta Clipper, or DC-1.
The development program was very innovative: build a little, fly the results, roll the lessons learned back into the next generation. The first vehicle (low-altitude atmospheric testing, designed to prove some of the required maneuvers for takeoff, landing and aborts) was the DC-X. The second-generation, subscale, orbital (with no payload) vehicle was to be the DC-Y; it would have tested fuel tankage, weight-saving and thermal-protection systems.
The total cost of DC-X and DC-Y was to be less than one year's budget for the Shuttle program.
SDIO borrowed stuff from everywhere to build DC-X. They got 4 RL-10's on loan from Rocketdyne, had the aeroshell built by Scaled Composites (Burt Rutan's outfit), and reprogrammed an airliner autopilot to fly the bird. DC-X was a phenomenal success, proving everything it was set out to do. And then SDIO, shutting down and getting outside of their bailiwick (which was NOT to develop commercial spacecraft launchers), turned the program and the prototype over to NASA.
NASA completed the scheduled test flights and then crashed and burned the prototype when someone neglected to reconnect a landing-gear unlock line before flight. Accident? Deliberate? No one's talking.
After the destruction of the DC-X, NASA let a contract for the development of a successor to the Shuttle. The developers of the DC-X had a bid in, but the contract was awarded to a company whose vehicle:
Had no development record;
Could not be delivered for many years longer;
Had a much more expensive development program.
On the other hand it took off vertically like Shuttle, landed on (and required) a runway like Shuttle, and required a new engine development program. The winner was not the low bidder. Can you say "more pork"?
When the winner of the contract was announced, the counsel for NASA was present. This was apparently to keep the DC-1 proponents from getting the idea of suing to either get the contract or find out what funny business had gone on. In the mean time, the Shuttle and its standing army of maintenance people are still working, and there's a lucrative R&D contract for the VentureStar (even though it's having serious difficulties with its composite LH2 tanks delaminating). It's great for everyone except the taxpayer and people who might benefit from flying satellites cheaper; IOW, it sucks. -- The Karma Century Club is taking new members.
-- Time is Nature's way of keeping everything from happening at once... the bitch.
Does anyone know if the sort of problems the Hubble is having are "routine" or if this was a problem inherent in the design of the telescope? It seems (and maybe I'm just too influenced by the media) that this project has been beset by problems from the very start. Too bad too...its images are breathtaking.
If you want the whole story, I highly recommend reading Eric Chaisson's The Hubble Wars ; he was a senior scientist on the project during the pre-launch and commissioning, and is a tremendously detailed but engaging scientific writer. Read it, and you'll never look at NASA the same way again. (For a similar perspective on people in space, read Dragonfly.)
The problems with Hubble are too many to enumerate here, but they begin with the overselling of the shuttle's capabilities, i.e. flight rate and cost. (In the 1970s, they would have laughed at the idea of a six-month delay in launching a servicing mission.) The Hubble was also beset by requirements that they borrow tech from the military spysat side, but without classified knowledge about the limitations of that tech. The closed procurement process probably factored in the misshapen mirror.
But it is also crystal-clear from the book that NASA fumbled the PR. First they dissembled about the problems, then they labeled it a complete failure. Chaisson and others desperately tried to show that it could do real science even with the astigmatism, and they succeeded. And they came up with a correction, and NASA got on board with installing the fix. Since the fix, it has performed at or above expectations.
The gyros were known to have a limited lifespan, and having them replaced was always a possibility. The telescope is happily waiting in safe mode for its systems to be repaired. The shuttle repair mission was moved up, but then it was held, and held, and held again, while the gyros began to fail. Having Hubble offline is extremely disappointing, but this particular problem is really not to be compared with a design flaw. In fact, with this mission and the 3B repair mission in '01, we can probably expect Hubble to outperform its expected 15-year on-orbit lifespan.
(If anyone can recommend a decent book from outside Chaisson's perspective, I'd like to hear of it, just to hear the other side of the story.) ----
-- lake effect weblog {Network engineer in Chicago--looking for work!}
The Hubble was also beset by requirements that they borrow tech from the military spysat side, but without classified knowledge about the limitations of that tech. The closed procurement process probably factored in the misshapen mirror.
It most certainly did. The mirror on the Hubble was built by Perkin-Elmer, in a plant devoted to spysats. NASA was not allowed access to the plant to verify the correct figuring of the mirror, and an error by P-E in the construction of the test gear caused it to be built to an incorrect focus. As a cost-saving measure, the entire Hubble was not checked for proper focus before launch (Perkin-Elmer was assumed to know what they were doing). In contrast, the backup mirror contract was let to Kodak; after the problem with the P-E mirror was discovered after launch, Kodak's mirror was checked and found to be flawless.
I guess "open-source" isn't just good for software, it's good for space projects too. -- The Karma Century Club is taking new members.
-- Time is Nature's way of keeping everything from happening at once... the bitch.
Chips with smaller feature sizes are usually more susceptible to radiation. Space qualified, radiation hardened chips are often several generations behind commercial chips. Most vendors have bailed out of the milspec/space market. The profits (if any) are too small in comparison to the commercial market. ESA designed a space qualified version of the SPARC and Sandia National Laboratory is working on a space qualified Pentium.
Well seeing the space shuttle take off from 300 miles away is a religious experience. First you see the cirrus clouds way of in the distance get faint red, then you see a huge ball of fire rise right under the red clouds, illuminating them from underneath and arcing to the right. Then it punches through the clouds, flares up, and spits out two smaller balls of fire, the boosters. The white ball of fire produced by the main engines keeps burning forever and arcs right until it looks horizontal but really you're seeing the curvature of the earth. Then it becomes just another star, like Battlestar Galactica. I got the whole thing on Realvideo. It's extremely rare to get a perfectly clear night in Fl*rida in December.
Slashdot Mirror
"Except that I read somewhere that, unofficially, the shuttles' software has never been able to handle a year change correctly, anyway..." You are right, the software cannot handle a year-end rollover. The navigation routines that tell the orbiter where to de-orbit perform their calculations in the M50 (Mean of Aries 1950) coordinate system. This system is an inertial coordinate system, which means that it is referenced to a fixed point in space. Now, in order to deorbit into the right window, we need to tell the orbiter how the earth is positioned within this inertial reference frame. To accomplish this we need to use a Rotation Nutation Precession (RNP) matrix to transform the orbiter's position in M50 coordinates to a position in Earth Centered Earth Fixed (ECEF) coordinates. The RNP matrices are computed on the ground for various times during the year and each are only good for about 3-months on either side of their designed time. If a launch slips over the year end, the RNP matrix is no longer any good, no matter how stale it is (1 day or 1 month or [1 year - 1 hour]). The good news is that we have now designed an improvement to the flight software that calculates the RNP matrix on-board. This would be able to get us out of this predicament, but I would venture to guess that the program still would never have a mission that would cross the year-end boundary in orbit. Actually, we could probably get out of it by patching a new RNP while in orbit, but that is a risk that the program is most likely unwilling to take. - These views are mine and do not necessarily represent those of my employer.
That's a Galileo project picture. Some Hubble shots are available at http://marvel.stsci.edu/top.html.
--
The Karma Century Club is taking new members.
Time is Nature's way of keeping everything from happening at once... the bitch.
I don't strap my car to two solid rocket boosters and send it into orbit, at least not regularly. This is a very bogus comparison, the tolerances (e.g. heat, stress) required for normal operation of a car, and the tolerances required for a man rated launcher are not comparable.
You also don't baby your car the way that NASA babies it's space shuttles, nor do you meticulously repair the maybe slightly defective parts the way NASA does with the space shuttle. Both vehicles are designed for certain limits, and both operate within those limits.
Unfortunately is bad design thats causing the delays, the shuttle design was a compromise (much like the current ISS debacle) due to repeated budget cuts. Initially design targets were to have something that could fly and turn around again in a matter of days or weeks, they were looking for a shuttle that could fly (at a minimum) twenty or thirty times a year. What they ended up with falls very short of the mark, something that was actually more expensive to launch than a normal rocket. The shuttle is an aging launcher, with design compromises at every turn, it shows.
Compromise of the initial design vision isn't what's delaying the shuttle in any way. If the shuttle was built to be used 20-30 times a year, it still would undergo the safety checks that NASA is giving it now, and if they wiring was bad on one of those shuttles, I'm sure NASA would take the time to check each one, and it would probably still have an excessive number of miles of wiring. What's causing the delay here isn't the design of the shuttles, whatever form that design may be, but NASA's anality (analism, analness?) about safety. Rightly so, I might add, since protecting the lives of the passengers should be the top priority.
Hopefully, VentureStar won't be as big a disappointment, although personally I still think they should have gone with the Delta Clipper (at least they had a working prototype in hardware).
Yeah, and in the meantime, what do you use? The current 'bad design' that's really only failed once in 20 years of usage, right? NASA's not going to make the mistake again of stopping all space flight for development of a launching mechanism when they have a perfectly good one. And given the VentureStar or the Delta Clipper's design snafus, I'm glad they're still researching it.
No way, we've already stretched the design lifetime well beyond the inital estimates.
Then how can it be a bad design? It may not fulfill the initial design vision, but as anyone who's ever worked in design can tell you, the vision changes as you work on a project. The fact that the shuttles have lasted this long with only problems to the launch device, not the orbiter itself, shows exactly how great and durable a design it actually is.
Odd... it seems to me that every time I turn around, Hubble has broken again... are we attacking the symptoms here or the disease, so to speak?
You might want to learn a little more about the Hubble mission. When this thing was put together, several servicing missions were anticipated to replace components as the wore out, and to upgrade or swap out instruments. These missions are planned on a three year cycle.
The ORIGINAL plan was to bring the Hubble back to earth for refitting every 5 years, however the ability of shuttle crews to do in-orbit servicing has made this unnecessary.
IN FACT, the Hubble has been one of the greatest successes of the entire space program.
This mission not only includes replacement of gyroscopes (that lasted longer than originally planned), but upgrading some instruments and the main Hubble computer system.
What they wound up with looked very different from the Space Scuttle and VultureStar. It was a squarish bullet, covered in thermal-blanket material originally developed for the Shuttle. It was not terribly fussy about its engines; it could have flown on J2's or RL-10's. The innovations were several:
- It was designed to land tail first, under power. This takes advantage of the engines, which are along for the ride, to provide landing capabilities. This eliminates the need to be able to glide subsonically; the glideslope, flare and landing maneuver required by Shuttle (and VentureStar) is unnecessary.
- It did not have wings. This saved a great deal of weight in the airframe.
- Its landing gear was a system of struts and pads. No wheels required. This saved more weight.
- The pilot went away also. When so many missions are just putting unmanned birds or cargo in orbit, why carry people along all the time?
The vehicle was to be called the Delta Clipper, or DC-1.The development program was very innovative: build a little, fly the results, roll the lessons learned back into the next generation. The first vehicle (low-altitude atmospheric testing, designed to prove some of the required maneuvers for takeoff, landing and aborts) was the DC-X. The second-generation, subscale, orbital (with no payload) vehicle was to be the DC-Y; it would have tested fuel tankage, weight-saving and thermal-protection systems.
The total cost of DC-X and DC-Y was to be less than one year's budget for the Shuttle program.
SDIO borrowed stuff from everywhere to build DC-X. They got 4 RL-10's on loan from Rocketdyne, had the aeroshell built by Scaled Composites (Burt Rutan's outfit), and reprogrammed an airliner autopilot to fly the bird. DC-X was a phenomenal success, proving everything it was set out to do. And then SDIO, shutting down and getting outside of their bailiwick (which was NOT to develop commercial spacecraft launchers), turned the program and the prototype over to NASA.
NASA completed the scheduled test flights and then crashed and burned the prototype when someone neglected to reconnect a landing-gear unlock line before flight. Accident? Deliberate? No one's talking.
After the destruction of the DC-X, NASA let a contract for the development of a successor to the Shuttle. The developers of the DC-X had a bid in, but the contract was awarded to a company whose vehicle:
- Had no development record;
- Could not be delivered for many years longer;
- Had a much more expensive development program.
On the other hand it took off vertically like Shuttle, landed on (and required) a runway like Shuttle, and required a new engine development program. The winner was not the low bidder. Can you say "more pork"?When the winner of the contract was announced, the counsel for NASA was present. This was apparently to keep the DC-1 proponents from getting the idea of suing to either get the contract or find out what funny business had gone on. In the mean time, the Shuttle and its standing army of maintenance people are still working, and there's a lucrative R&D contract for the VentureStar (even though it's having serious difficulties with its composite LH2 tanks delaminating). It's great for everyone except the taxpayer and people who might benefit from flying satellites cheaper; IOW, it sucks.
--
The Karma Century Club is taking new members.
Time is Nature's way of keeping everything from happening at once... the bitch.
Does anyone know if the sort of problems the Hubble is having are "routine" or if this was a problem inherent in the design of the telescope? It seems (and maybe I'm just too influenced by the media) that this project has been beset by problems from the very start. Too bad too...its images are breathtaking.
If you want the whole story, I highly recommend reading Eric Chaisson's The Hubble Wars ; he was a senior scientist on the project during the pre-launch and commissioning, and is a tremendously detailed but engaging scientific writer. Read it, and you'll never look at NASA the same way again. (For a similar perspective on people in space, read Dragonfly.)
The problems with Hubble are too many to enumerate here, but they begin with the overselling of the shuttle's capabilities, i.e. flight rate and cost. (In the 1970s, they would have laughed at the idea of a six-month delay in launching a servicing mission.) The Hubble was also beset by requirements that they borrow tech from the military spysat side, but without classified knowledge about the limitations of that tech. The closed procurement process probably factored in the misshapen mirror.
But it is also crystal-clear from the book that NASA fumbled the PR. First they dissembled about the problems, then they labeled it a complete failure. Chaisson and others desperately tried to show that it could do real science even with the astigmatism, and they succeeded. And they came up with a correction, and NASA got on board with installing the fix. Since the fix, it has performed at or above expectations.
The gyros were known to have a limited lifespan, and having them replaced was always a possibility. The telescope is happily waiting in safe mode for its systems to be repaired. The shuttle repair mission was moved up, but then it was held, and held, and held again, while the gyros began to fail. Having Hubble offline is extremely disappointing, but this particular problem is really not to be compared with a design flaw. In fact, with this mission and the 3B repair mission in '01, we can probably expect Hubble to outperform its expected 15-year on-orbit lifespan.
(If anyone can recommend a decent book from outside Chaisson's perspective, I'd like to hear of it, just to hear the other side of the story.)
----
lake effect weblog
{Network engineer in Chicago--looking for work!}
Chips with smaller feature sizes are usually more susceptible to radiation. Space qualified, radiation hardened chips are often several generations behind commercial chips. Most vendors have bailed out of the milspec/space market. The profits (if any) are too small in comparison to the commercial market. ESA designed a space qualified version of the SPARC and Sandia National Laboratory is working on a space qualified Pentium.
Mea navis aericumbens anguillis abundat
Well seeing the space shuttle take off from 300 miles away is a religious experience.
First you see the cirrus clouds way of in the distance get faint red, then you see a huge ball of fire rise right under the red clouds, illuminating them from underneath and arcing to the right. Then it punches through the clouds, flares up, and spits out two smaller balls of fire, the boosters. The white ball of fire produced by the main engines keeps burning forever and arcs right until it looks horizontal but really you're seeing the curvature of the earth. Then it becomes just another star, like Battlestar Galactica. I got the whole thing on Realvideo. It's extremely rare to get a perfectly clear night in Fl*rida in December.