SpaceX Falcon Heavy - Psyche - KSC LC-39A - 13 October 2023 (14:19 UTC)

[Starship_SpaceX]

[LouScheffer]

This is for planetary only

A.  The trajectory to orbit remains the same and on the same azimuth (easy for the range)
b.  Atlas can do the steering autonomously and continuously but only only every fifth minute of the window is analyzed and used as a planned launch attempt.  This way the spacecraft can handle the number of variations in the trajectory.
c.  SpaceX does not want deal with all the different trajectories from different azimuths and flight software loads.  For M2020, this at minimum would be 12 attempt X 21 days = more than 250.

I don't think (A) is technically correct.   See the NASA diagram below - the needed azimuth depends on the time of launch and will vary over the window.  Maybe ULA always *launches* to the same azimuth, for the convenience of the range.  But then after launch either the first or second stage needs to modify the azimuth to hit the correct spot.  The time to second ignition definitely varies.

Falcon 9 too much work pre launch and I doubt SpaceX wants to deal with that many software loads.
Plus stage recovery would be in jeopardy

I suspect this is the main reason for SpaceX instantaneous windows for planetary launches.  As time progresses though the window, the launch azimuth changes.  But this means the trajectories of the returning boosters changes as well.  That's 3x the trajectories to analyze, plus I suspect (but do not know for sure) there is more overhead in getting return trajectories approved, as they fly much closer to people and other valuable assets.

When SpaceX does interplanetary missions with ASDS landings, there would be even stronger reasons for an instantaneous window, namely that the needed barge location will vary with launch time within the window.  Since the barge cannot move on this time scale, they will just pick one time and try again the next day if they miss it.

[Jim]

I don't think (A) is technically correct.   See the NASA diagram below - the needed azimuth depends on the time of launch and will vary over the window.  Maybe ULA always *launches* to the same azimuth, for the convenience of the range.  But then after launch either the first or second stage needs to modify the azimuth to hit the correct spot.  The time to second ignition definitely varies.

I stated that only the trajectory to orbit remained the same.  Made no claim about the rest of the trajectory.

I made a similar diagram and stated that upperstage pointing is different

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2542985#msg2542985

[LouScheffer]

I don't think (A) is technically correct.   See the NASA diagram below - the needed azimuth depends on the time of launch and will vary over the window.  Maybe ULA always *launches* to the same azimuth, for the convenience of the range.  But then after launch either the first or second stage needs to modify the azimuth to hit the correct spot.  The time to second ignition definitely varies.

I stated that only the trajectory to orbit remained the same.  Made no claim about the rest of the trajectory.

I made a similar diagram and stated that upperstage pointing is different

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2542985#msg2542985

I'm afraid I still don't understand your point.  Depending on when in the window you launch, you need a different inclination parking orbit.  So I don't see how the trajectory to orbit can be the same everywhere within the window - to get to a different inclination, you need a different trajectory.  Maybe the first stage trajectory could be the same (and the second stage yaws), or the first stage could yaw during flight (I think SpaceX does this in polar trajectories from the Cape), but somehow the rocket needs to get into an orbit with varying inclination.  (Of course you also need varying delay to trans-planetary ignition, too, but you can't hit the right spot by *only* varying the delay.)

[Jim]

I don't think (A) is technically correct.   See the NASA diagram below - the needed azimuth depends on the time of launch and will vary over the window.  Maybe ULA always *launches* to the same azimuth, for the convenience of the range.  But then after launch either the first or second stage needs to modify the azimuth to hit the correct spot.  The time to second ignition definitely varies.

I stated that only the trajectory to orbit remained the same.  Made no claim about the rest of the trajectory.

I made a similar diagram and stated that upperstage pointing is different

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2542985#msg2542985

I'm afraid I still don't understand your point.  Depending on when in the window you launch, you need a different inclination parking orbit.  So I don't see how the trajectory to orbit can be the same everywhere within the window - to get to a different inclination, you need a different trajectory.  Maybe the first stage trajectory could be the same (and the second stage yaws), or the first stage could yaw during flight (I think SpaceX does this in polar trajectories from the Cape), but somehow the rocket needs to get into an orbit with varying inclination.  (Of course you also need varying delay to trans-planetary ignition, too, but you can't hit the right spot by *only* varying the delay.)

The parking orbit can remain the same but the start time and duration of the upper stage second burn varies as well as the attitude and steering for the burn. 

[Jim]

I can't show the actual diagrams, but this should help.  The second one is an enlargement from the first and I added another trajectory.  The first has a list of the milestones.  The park orbit is the same for much of the launch period, but there was a second park orbit at a different inclination for the last few days of the period and the extension.   But this applies to both park orbits.
Milestone 8 is the start of the 2nd upperstage burn at the beginning of the launch window.   The start of the burn at the end of the window is at 8b

[LouScheffer]

The parking orbit can remain the same but the start time and duration of the upper stage second burn varies as well as the attitude and steering for the burn. 

Interesting - I thought this technique went out with Apollo.  Clearly from the diagram above, the *optimum* azimuth changes continuously with the launch time.  And for a long window (Apollo's were > 4 hours), you *need* to change the azimuth - anything else is too inefficient.

But for short windows (maybe 2 hours or less) you can use a parking orbit targeted to the center of the window, then execute an out-of-plane injection burn (using somewhat more delta-V) to compensate.  Apollo was forced to use this strategy (in addition to varying the azimuth) since they wanted to do the trans-lunar injection either on the second or third orbit, chosen *after* they launched.  So they used a single parking orbit, chosen half-way between the two options, and corrected on the TLI burn.  The penalty is not too bad since this was only +- 45 minutes from the ideal parking orbit, but the cost grows rapidly after that (see Apollo lunar landing launch window: The controlling factors and constraints, Figure 11.)

So if you have enough margin on your mission, this is one way to use it.  Use a constant parking orbit for ease of launch analysis, and correct during the injection burn.