My guess is, after stage2 wait and clear ISS, mostly it will miss the satellite slot registered to og2. It may have to wait couple days for second chance of orbital insertion, by then there won't be enough LOX and RP-1 will be frozen to rock solid.
Quote from: Joffan on 10/09/2012 11:06 pmThe upper stage was testing how much propellant it had remaining, as I understand it. After the "no-go for main orbit", there's nothing computationally hard about making a second decision, "go/no-go for contingency orbit", based on that same result.Yes it is. It doesn't have enough data to such a decision nor is possible to design a contingency orbit that covers all the possibilities.
The upper stage was testing how much propellant it had remaining, as I understand it. After the "no-go for main orbit", there's nothing computationally hard about making a second decision, "go/no-go for contingency orbit", based on that same result.
Especially since some number/combination of those cases include "LV has bad/degraded nav" and so not possible to safely target a circularization burn. Safer to just have go/no-go logic, as SpaceX did.
This is exactly what we're asking.The only data the stage needs is A: "Did I successfully deliver Dragon", and B: "can I execute burn 2 plan-A".If the answers are YES and NO, then the stage can execute the fallback burn 2 plan-B, which is raise the perigee. Raising the perigee to a circular orbit is a well defined maneuver and is something the GNC should be able to do - I don't see the problem with unknown initial conditions.
Quote from: Jorge on 10/10/2012 03:58 amEspecially since some number/combination of those cases include "LV has bad/degraded nav" and so not possible to safely target a circularization burn. Safer to just have go/no-go logic, as SpaceX did.That's why I brought up the context of Falcon and Delta making orbit even though there were issues with the ascent profile.In both cases you could have asked the same question - something has gone wrong, the rocket might have bad/degraded nav, or an unpredictable engine, why try to reach orbit and potentially put other assets at risk? Using the logic you present, it's better to just let them fail to make orbit and burn up - and clearly nobody chose that.I think Jotten's proposal makes perfect sense and is within what the second stage can be programmed to do without requiring real-time risk assessment.
iss can go as low as 330km, right now orbcomm apogee is at 320km, you make a circle there, it's kinda pretty close.
No, it has no way of knowing if it did A successfully.Again, LV's only have one trajectory, no decisions.No, Raising the perigee to a circular orbit is not a well defined maneuver, if you don't know where you are starting from. LV's don't move on to the next maneuver unless the first one is complete and successful. Otherwise, there are too many variations to analyze.
Ed, your definition of launch failure is pretty silly. If people used your definition as an industry standard, then no one would ever want to launch secondaries because it'd cause too much of a risk of the whole flight being labeled a failure. A bunch of performance would be just left on the table for the sole reason of avoiding the label of full "failure." Not only that, but even the Orbcomm spacecraft is still usable for its most important primary purpose, which is checking out that the spacecraft works as planned in orbit. I don't think the "partial failure" of the Shuttle's two engine-out events should be counted as launch failures, nor do I think the recent Delta IV anomaly should be counted as a launch failure, nor do I think the one Atlas V underperformance should be counted as a launch failure. They were underperforming missions that left the primary mission successful according to the customer, and that's what matters....
Also, Ed, your figures for Merlin Vac (based on Merlin 1D) thrust are out of date and much too low. The M1D-based Merlin Vac can do at least 80 tons of thrust, versus 45 you estimated here: http://www.spacelaunchreport.com/falcon9.html
Quote from: meekGee on 10/10/2012 04:11 amQuote from: Jorge on 10/10/2012 03:58 amEspecially since some number/combination of those cases include "LV has bad/degraded nav" and so not possible to safely target a circularization burn. Safer to just have go/no-go logic, as SpaceX did.That's why I brought up the context of Falcon and Delta making orbit even though there were issues with the ascent profile.In both cases you could have asked the same question - something has gone wrong, the rocket might have bad/degraded nav, or an unpredictable engine, why try to reach orbit and potentially put other assets at risk? Using the logic you present, it's better to just let them fail to make orbit and burn up - and clearly nobody chose that.I think Jotten's proposal makes perfect sense and is within what the second stage can be programmed to do without requiring real-time risk assessment.Wrong. The rocket goes until it can't go anymore. If it can't, it does burn up or go in a useless orbit.It is not within the second stage capability to be programmed or do it without a risk assessment.I am going to have to say it, you don't know what you are talking about
Is there a bright line that defines what types of decisions a launch vehicle can or should make, and those it may not or should not make?
But most importantly, planned insertion orbits were not achieved. ...Imagine that your next United or Southwest flight to Chicago was forced to land in Milwaukee or Rockford or Cleveland instead, but the airline refused to refund your ticket, calling the flight a "success" because the airplane didn't crash.
Imagine that your next United or Southwest flight to Chicago was forced to land in Milwaukee or Rockford or Cleveland instead, but the airline refused to refund your ticket, calling the flight a "success" because the airplane didn't crash.
I think the question Jim is answering is, What Would Centaur Do?Once on orbit Centaur can do some amazing stuff for a launch vehicle second stage ... it is almost like a spacecraft. Yet Centaur wouldn't be programmed to attempt any sort of contingency burn if it found itself unable to proceed with a planned burn. It is certainly no shame for SpaceX that the programming of their stage is similar to Centaur's!That said, I'm pretty sure that if the customer flying a secondary payload were to pay for it, SpaceX software engineers would be willing to implement slightly clever logic to handle a small number of pre-planned contingency cases. But the costs along the way to get that logic into the flight controller on an actual mission? The SpaceX internal review; the NASA review; the Orbcomm review; all costly and time consuming. And that's just for a contingency case. Is it really surprising Orbcomm accepted the secondary payload service as offered, i.e. "best" effort where best might not be very good?Finally, it isn't yet clear Orbcomm can get any more value from the satellite if it is in circular but low orbit than it can get from being in elliptical and low orbit. If what they really need is testing at a constellation-representative altitude, circularization doesn't get them that!
Quote from: Robotbeat on 10/09/2012 02:25 pmEd, your definition of launch failure is pretty silly. If people used your definition as an industry standard, then no one would ever want to launch secondaries because it'd cause too much of a risk of the whole flight being labeled a failure. A bunch of performance would be just left on the table for the sole reason of avoiding the label of full "failure." Not only that, but even the Orbcomm spacecraft is still usable for its most important primary purpose, which is checking out that the spacecraft works as planned in orbit. I don't think the "partial failure" of the Shuttle's two engine-out events should be counted as launch failures, nor do I think the recent Delta IV anomaly should be counted as a launch failure, nor do I think the one Atlas V underperformance should be counted as a launch failure. They were underperforming missions that left the primary mission successful according to the customer, and that's what matters....I count such failures because they are failures. If a rocket didn't do its complete job, something had to have actually failed. In all of the examples you mention above, real hardware failures happened. Engines shut down or underperformed. Propellants leaked. But most importantly, planned insertion orbits were not achieved. The latter point actually clears the most recent Delta 4 from the "failure" category because it made its orbit, thanks to the relatively benign engine problem and thanks to performance margin. Money bought that extra margin, and combined with a little luck, that money paid for success in this particular case.Imagine that your next United or Southwest flight to Chicago was forced to land in Milwaukee or Rockford or Cleveland instead, but the airline refused to refund your ticket, calling the flight a "success" because the airplane didn't crash. There has long seemed, to me, an industry bias to "over emphasize" success and, well, to just not talk about failures, allowing some of them to slip from the record books. If two rockets make it to orbit, but one suffers a problem that strands its payload short, then the record should show and count it.QuoteAlso, Ed, your figures for Merlin Vac (based on Merlin 1D) thrust are out of date and much too low. The M1D-based Merlin Vac can do at least 80 tons of thrust, versus 45 you estimated here: http://www.spacelaunchreport.com/falcon9.htmlI would like to see SpaceX numbers on that. The vacuum version of Merlin doesn't need to produce maximum thrust, it needs specific impulse. I think that SpaceX traded some thrust for ISP on Merlin (1C) Vacuum. Also keep in mind that I use metric tons (tonnes) rather than tons. - Ed Kyle
