Quote from: Jim on 11/21/2023 08:05 pmQuote from: mn on 09/26/2023 01:45 amCan they do that for this mission? or does this mission have an instantaneous window and if it is instantaneous, exactly why, (Jim is claiming that it is instantaneous due to the inability to do RAAN steering, I'm not yet convinced)You don't need to be convinced. The facts are the facts.There are multiple questions here and I'd like separate them and get more info. (This forum is not just about believing, it's also about learning and understand why things are so)It is a fact that the F9 cannot do RAAN steering. (I have no reason to doubt this, what I am trying to understand why it makes a difference).If I understand correctly RAAN steering allows the vehicle to autonomously calculate a new trajectory if/when T0 is changed, is that understanding correct? or does RAAN steering mean something else?Can SpaceX ground computers calculate a new trajectory if they change the T0 and upload that new trajectory to the vehicle, is that possible or not? and if not, why not? (they change T0 on starlink missions often, but obviously Psyche is different)(There should be plenty of time between a new T0 being selected and the actual launch, since as discussed, SpaceX does not change T0 after start of fueling. Is that part correct or not?)Is the FH performance limited so that it could not have pushed Psyche to the correct orbit if T0 was even a little bit moved? (within the desired safety performance margin).TIA
Quote from: mn on 09/26/2023 01:45 amCan they do that for this mission? or does this mission have an instantaneous window and if it is instantaneous, exactly why, (Jim is claiming that it is instantaneous due to the inability to do RAAN steering, I'm not yet convinced)You don't need to be convinced. The facts are the facts.
Can they do that for this mission? or does this mission have an instantaneous window and if it is instantaneous, exactly why, (Jim is claiming that it is instantaneous due to the inability to do RAAN steering, I'm not yet convinced)
1. Can SpaceX ground computers calculate a new trajectory if they change the T0 and upload that new trajectory to the vehicle, is that possible or not? and if not, why not? (they change T0 on starlink missions often, but obviously Psyche is different)2. Is the FH performance limited so that it could not have pushed Psyche to the correct orbit if T0 was even a little bit moved? (within the desired safety performance margin).
(they change T0 on starlink missions often, but obviously Psyche is different)
Quote from: mn on 11/21/2023 09:52 pm1. Can SpaceX ground computers calculate a new trajectory if they change the T0 and upload that new trajectory to the vehicle, is that possible or not? and if not, why not? (they change T0 on starlink missions often, but obviously Psyche is different)2. Is the FH performance limited so that it could not have pushed Psyche to the correct orbit if T0 was even a little bit moved? (within the desired safety performance margin).1. Trajectories are not calculated but constructed and then validated by running 1000's of simulations. Delta II got around it by having two trajectories available that were 40 minutes apart. Time to reload the new trajectory software and valid it. The two trajectories fly out at different azimuths but end up at the same end point. Not worth it for Falcon 9.2. It has nothing to do with performance, but burning at the right place in orbit. Delta II (and Delta IV before ULA common avionics) and Falcon 9 fly the same trajectory regardless of the launch time. The second burn of the upperstage is going to burn at the same time interval after the first burn and pointing in the same direction with respect to the LHLV.Atlas knows that when it delays from the planned T-0, that the second burn is going to occur sooner (because the earth rotated during the delay) and may have to point slightly different.The same applies to ISS orbit, Atlas could launch earlier or later after the optimum time (the amount is depended on available excess performance). Atlas would steer left or right to get "underneath" the ISS orbit.Because of the infinite number of trajectories for a planetary window. ULA would only run sims on 5 minute intervals throughout the window and the launch would be limited to only those times. This is to reduce the number of trajectories that the spacecraft and tracking stations have to deal with and analyze.
Quote from: Jim on 11/21/2023 10:26 pmQuote from: mn on 11/21/2023 09:52 pm1. Can SpaceX ground computers calculate a new trajectory if they change the T0 and upload that new trajectory to the vehicle, is that possible or not? and if not, why not? (they change T0 on starlink missions often, but obviously Psyche is different)2. Is the FH performance limited so that it could not have pushed Psyche to the correct orbit if T0 was even a little bit moved? (within the desired safety performance margin).1. Trajectories are not calculated but constructed and then validated by running 1000's of simulations. Delta II got around it by having two trajectories available that were 40 minutes apart. Time to reload the new trajectory software and valid it. The two trajectories fly out at different azimuths but end up at the same end point. Not worth it for Falcon 9.2. It has nothing to do with performance, but burning at the right place in orbit. Delta II (and Delta IV before ULA common avionics) and Falcon 9 fly the same trajectory regardless of the launch time. The second burn of the upperstage is going to burn at the same time interval after the first burn and pointing in the same direction with respect to the LHLV.Atlas knows that when it delays from the planned T-0, that the second burn is going to occur sooner (because the earth rotated during the delay) and may have to point slightly different.The same applies to ISS orbit, Atlas could launch earlier or later after the optimum time (the amount is depended on available excess performance). Atlas would steer left or right to get "underneath" the ISS orbit.Because of the infinite number of trajectories for a planetary window. ULA would only run sims on 5 minute intervals throughout the window and the launch would be limited to only those times. This is to reduce the number of trajectories that the spacecraft and tracking stations have to deal with and analyze.Thanks for the detailed response, (now I have to read it a few more times to try to understand it)
Quote from: mn on 11/21/2023 10:35 pmQuote from: Jim on 11/21/2023 10:26 pmQuote from: mn on 11/21/2023 09:52 pm1. Can SpaceX ground computers calculate a new trajectory if they change the T0 and upload that new trajectory to the vehicle, is that possible or not? and if not, why not? (they change T0 on starlink missions often, but obviously Psyche is different)2. Is the FH performance limited so that it could not have pushed Psyche to the correct orbit if T0 was even a little bit moved? (within the desired safety performance margin).1. Trajectories are not calculated but constructed and then validated by running 1000's of simulations. Delta II got around it by having two trajectories available that were 40 minutes apart. Time to reload the new trajectory software and valid it. The two trajectories fly out at different azimuths but end up at the same end point. Not worth it for Falcon 9.2. It has nothing to do with performance, but burning at the right place in orbit. Delta II (and Delta IV before ULA common avionics) and Falcon 9 fly the same trajectory regardless of the launch time. The second burn of the upperstage is going to burn at the same time interval after the first burn and pointing in the same direction with respect to the LHLV.Atlas knows that when it delays from the planned T-0, that the second burn is going to occur sooner (because the earth rotated during the delay) and may have to point slightly different.The same applies to ISS orbit, Atlas could launch earlier or later after the optimum time (the amount is depended on available excess performance). Atlas would steer left or right to get "underneath" the ISS orbit.Because of the infinite number of trajectories for a planetary window. ULA would only run sims on 5 minute intervals throughout the window and the launch would be limited to only those times. This is to reduce the number of trajectories that the spacecraft and tracking stations have to deal with and analyze.Thanks for the detailed response, (now I have to read it a few more times to try to understand it)Thanks again.1. I understand why the departure burn will be at a different MET (and perhaps slightly different angle), simple and obvious. But I'm stuck on why F9 or FH cannot do it - or 'not worth it', if you say 'this mission has a instantaneous window because they cannot do RAAN steering', and now it's down to a different timing for the 2nd stage 2nd burn, something that SpaceX can easily calculate in advance for several T0 times and then select the right one and upload it to the vehicle before launch. (perhaps in the days of Delta II uploading new parameters was a big deal and allowing the launch vehicle to do it autonomously was a big advantage, but it is hard to imagine that is a big deal today). And it seems changing the 2nd stage burn timing doesn't even change any performance requirements, so that is not a limiting issue2. If the difference is just the timing, why would the Atlas V with new avionics fly a different trajectory in such a case? (as you seem to imply, though not really relevant to our discussion)3. While I am not a rocket scientist, I'm quite certain that the timing of the 2nd stage 2nd burn is not related to RAAN steering)Re: launch to ISS, I understand that the LV needs to steer to get to the correct plane, (essentially flying a slightly curved trajectory). that is what I understand RAAN steering is for, but is SpaceX not capable of doing that calculation on the ground in advance and uploading to the vehicle when selecting a new T0?And here I also understand that performance is a significant limiting factor, the curved trajectory wastes alot of performance, so it's a question of performance margin that determines how far out of plane you can launch and still get to the correct final orbit (and for SpaceX, that margin means potentially switching from RTLS to ASDS to expended, which they would likely not do, so the question boils down to how far out of plane can they launch and still do the same planned landing)Bottom line TL DR: What exactly is SpaceX not capable of doing that leads us to this being an instantaneous window.
Bottom line TL DR: What exactly is SpaceX not capable of doing that leads us to this being an instantaneous window.
Quote from: mn on 11/22/2023 03:57 pmBottom line TL DR: What exactly is SpaceX not capable of doing that leads us to this being an instantaneous window.If they needed RAAN steering they would certainly be capable of getting it right. But they haven't needed it.As I understand it, subcooled propellants have limited "shelf life" after being loaded (they start warming and expanding which means reduced propellant mass and reduced performance). So once they start loading propellant they're going to have to either launch close to the originally planned T0, or detank and rechill - they have likely concluded that adding RAAN steering won't give them enough flexibility in being able to move T0 at the last minute to be worth the additional complexity in the firmware.
1. I understand why the departure burn will be at a different MET (and perhaps slightly different angle), simple and obvious. But I'm stuck on why F9 or FH cannot do it - or 'not worth it', if you say 'this mission has a instantaneous window because they cannot do RAAN steering', and now it's down to a different timing for the 2nd stage 2nd burn, something that SpaceX can easily calculate in advance for several T0 times and then select the right one and upload it to the vehicle before launch. (perhaps in the days of Delta II uploading new parameters was a big deal and allowing the launch vehicle to do it autonomously was a big advantage, but it is hard to imagine that is a big deal today). And it seems changing the 2nd stage burn timing doesn't even change any performance requirements, so that is not a limiting issue
2. If the difference is just the timing, why would the Atlas V with new avionics fly a different trajectory in such a case? (as you seem to imply, though not really relevant to our discussion)
Re: launch to ISS, I understand that the LV needs to steer to get to the correct plane, (essentially flying a slightly curved trajectory). that is what I understand RAAN steering is for, but is SpaceX not capable of doing that calculation on the ground in advance and uploading to the vehicle when selecting a new T0?
Quote from: launchwatcher on 11/22/2023 05:16 pmQuote from: mn on 11/22/2023 03:57 pmBottom line TL DR: What exactly is SpaceX not capable of doing that leads us to this being an instantaneous window.If they needed RAAN steering they would certainly be capable of getting it right. But they haven't needed it.As I understand it, subcooled propellants have limited "shelf life" after being loaded (they start warming and expanding which means reduced propellant mass and reduced performance). So once they start loading propellant they're going to have to either launch close to the originally planned T0, or detank and rechill - they have likely concluded that adding RAAN steering won't give them enough flexibility in being able to move T0 at the last minute to be worth the additional complexity in the firmware.We have gone over this many times. They can and have several times delayed the start fueling to try to find a opening in the weather. But this is obviously only possible if the launch window is not instantaneous, hence the question and discussion.
Quote from: mn on 11/22/2023 03:57 pm1. I understand why the departure burn will be at a different MET (and perhaps slightly different angle), simple and obvious. But I'm stuck on why F9 or FH cannot do it - or 'not worth it', if you say 'this mission has a instantaneous window because they cannot do RAAN steering', and now it's down to a different timing for the 2nd stage 2nd burn, something that SpaceX can easily calculate in advance for several T0 times and then select the right one and upload it to the vehicle before launch. (perhaps in the days of Delta II uploading new parameters was a big deal and allowing the launch vehicle to do it autonomously was a big advantage, but it is hard to imagine that is a big deal today). And it seems changing the 2nd stage burn timing doesn't even change any performance requirements, so that is not a limiting issueBecause trajectory design is not automated. There is a lot of back and forth between the spacecraft and launch vehicle to get a trajectory design. The spacecraft has to provide targets for every launch time. Delta II had two attempts per day at different azimuths. SpaceX doesn't want to spend the time designing all these trajectories and working with the range on all the different azimuths. it is the equivalent of the work required for dozens of missions*. It doesn't provide much increase in launch probability when once the vehicle is tanked, it has only the ability to launch within 10 minutes of the planned T-0. Very few countdown delays happen before tanking. * NASA get charged extra for planetary missions for this work, even with one launch attempt per day. Number of trajectories means daily attempts x launch period days. Atlas can fly from the same azimuth, so from a range POV (this also includes all analyses), it is the same boost trajectory for each day and most days of the launch period (there might be another azimuth to extend the launch period). What changes is the coast periods duration and second burn parameters (analyses have to account for the differences this causes).There is a performance difference when you launch in a planetary window. It is a bell curve (launch time vs performance required.) and the cut off is vehicle performance capability. This relates to upperstage second burn duration and steering.
If trajectory design is not automated and involves so much more work on multiple parties, how does the Atlas V do it autonomously?
Re: performance curve, so that is something I'd love to know, if we ignore the 'is it worth the effort' part of the equation, what kind of launch window can the FH potentially have for the Psyche mission before it hits the limits on this curve?
Re: ISS, true it is rarely worth it, but half hour is not nothing, if we are dealing with bad weather and we see that it will clear up at 15 minutes past the optimal T0, why wouldn't they delay start of fueling and try to launch 15 minutes later. (and like I said, if that involves all this analysis by many parties, how can Atlas V do it?)
(and that half hour you mention is based on what? different vehicles will have different margins? And F9 will have different margins depending on landing plan, obviously not going to offload anything, just work within the available margins)