Artemis 1 Launch Window?

[Nick]

Does anyone know whether the Artemis 1 mission will have a launch window, or whether it can launch at any time (subject to range constraints,, etc., etc.)

I gather that the mission is targeting a 22 November launch, and am starting to wonder about making the trip over The Pond to watch the launch, Covid restrictions permitting.

If there's a launch window of x days, where x is a relatively small number, I'd make sure I was in the Titusville area for that length of time. But if the mission had a very short window and then would slip a month while the Moon trundled round its orbit, then the trip is likely not a goer, as I wouldn't want to spend the money or the time twiddling my thumbs, especially in the run-up to Christmas.

My guess would be that heading for a halo orbit with no landing issues is not time crtical, but that is just a guess. There could be many things I've not thought of that impose time constraints.

Any comments appreciated.

[Starlab90]

https://www.nasaspaceflight.com/2021/06/artemis-1-sls-core-stage-lift/

It has been a couple of months, but Phillip Sloss has provided the dates for the first 3 possible launch periods near the end of the article linked above.

The Artemis 1 mission management team is currently working through flight readiness analysis cycles, calculating lunar launch opportunities and vehicle performance margins. Currently, the first launch period available for an NET late-November 2021 readiness date is Launch Period 15, which opens on November 23 and runs through December 10. Following that, Launch Period 16 is December 21 through January 3, 2022, with Launch Period 17 running January 17 through January 30.

The daily launch windows for those periods have not been provided, but expect them to be mostly at night.

[Nick]

Many thanks.

A night launch of that beast would be something to see. The first Shuttle launch I saw was at night - STS-116 in December 2006. A long duration shot I took of the arc from ignition to MECO still has pride of place on my wall...

Why would the daily windows be at night? Just curiosity really: I'd just  to understand.

[TJL]

I seem to recall when I made plans to see the Apollo 15 launch, that the launch window was 7/26 and 7/27. After that it was moved out a month to late August. But that had to do with lighting conditions for the lunar landing. Those that witnessed Apollo launches were very lucky in that IIRC the only delay once the countdown started was Apollo 9...a 3 day postponement due to "colds". Apollo 14 and 17 had "delays" but launched on the scheduled date.

[slobber91]

For those wishing some additional insight into how the daily launch window for a lunar mission out of KSC is determined, the first half of the instructional film "Launch Windows for Lunar Landing" (put together by the NASA Mission Planning and Analysis Division in 1967) is one of the better descriptions I've encountered.  While the insertion tracking constraints on launch azimuth are probably no longer applicable (with TDRS now available), I believe the rest of the constraints described in the film would generally apply for Artemis 1 daily launch windows.

Of course the monthly launch window constraints described in the second half of the film (dictated primarily by surface lighting and transfer orbit plane location relative to lunar landing sites) would not apply to Artemis 1.

"Launch Windows for Lunar Landing" may be viewed at the following link:


 

[CreponChris]

If they wanted to launch with an ecliptic inclination as low as possible (around 5 degrees) on the 22nd of November 2021, that would occur at about 0710 UTC by my calculations. Although, I've heard there are advantages for lunar missions to not doing that. But I've done it in Orbiter 2016 with the SLS Block 1 and gotten to the Moon (performed a TLI) that way with total delta-V (post-LEO) of around 2.35 km/s

[Starlab90]

If they wanted to launch with an ecliptic inclination as low as possible (around 5 degrees) on the 22nd of November 2021, that would occur at about 0710 UTC by my calculations. Although, I've heard there are advantages for lunar missions to not doing that. But I've done it in Orbiter 2016 with the SLS Block 1 and gotten to the Moon (performed a TLI) that way with total delta-V (post-LEO) of around 2.35 km/s

There are restrictions on the SLS trajectory that you probably violated to do that. SLS propulsion elements provide more performance than Orion and some parts of SLS hardware can allow. I've posted about this before, but it has been a long time.

[CreponChris]

If they wanted to launch with an ecliptic inclination as low as possible (around 5 degrees) on the 22nd of November 2021, that would occur at about 0710 UTC by my calculations. Although, I've heard there are advantages for lunar missions to not doing that. But I've done it in Orbiter 2016 with the SLS Block 1 and gotten to the Moon (performed a TLI) that way with total delta-V (post-LEO) of around 2.35 km/s

There are restrictions on the SLS trajectory that you probably violated to do that. SLS propulsion elements provide more performance than Orion and some parts of SLS hardware can allow. I've posted about this before, but it has been a long time.

Well, that is interesting. Thanks. There are a couple PDF's I've found related to the Artemis I, https://ntrs.nasa.gov/api/citations/20205005150/downloads/AAS20649_Artemis1_Trajectory_Design_Optimization.pdf that one I think being the most relevant. I'll have a look at the article already linked to in this thread and if you ever want to point me toward threads you've already contributed to that would be appreciated. Don't want to hijack this one :]

[Starlab90]

Well, that is interesting. Thanks. There are a couple PDF's I've found related to the Artemis I, https://ntrs.nasa.gov/api/citations/20205005150/downloads/AAS20649_Artemis1_Trajectory_Design_Optimization.pdf that one I think being the most relevant. I'll have a look at the article already linked to in this thread and if you ever want to point me toward threads you've already contributed to that would be appreciated. Don't want to hijack this one :]

If you look, all the authors of that particular paper are from JSC. JSC is responsible for the Orion trajectory once SLS is finished. MSFC is responsible for the SLS part of the trajectory, though obviously SLS trajectory design must meet Orion requirements. Look for papers with MSFC authors, or with both MSFC and JSC authors. There are multiple papers like that out there, and they can give you more insight into what I was talking about above. MSFC and JSC work very close together on these trajectories.

Also, welcome to the forum!

[Stan-1967]

Does anyone know whether the Artemis 1 mission will have a launch window, or whether it can launch at any time (subject to range constraints,, etc., etc.)

The answer to your question is complicated & interesting.  For starters the linked YouTube video in post #4 is a good primer.  The paper linked by CreponChris in post#7 get down into the nitty gritty details about what constraints limit the launch window, as well as all other critical events that will happen on any of the Artemis missions.  After reading the whole thing, I wish I would have started with pages 17&18 to refresh my mind to all the acronyms.  Page 13 gives the actual constraint for the daily launch window.

https://ntrs.nasa.gov/api/citations/20205005150/downloads/AAS20649_Artemis1_Trajectory_Design_Optimization.pdf

A couple of things I noted:
The video says there are two daily launch windows.  The linked paper says there is one.  I think that is because one of the daily launch windows is not feasible to reach from the launch site and have the spacecraft arrive at the right time for the TLI burn.  The spaceship needs to circularize the orbit before the TLI burn, & this time & geography constraint mean the only meaningful launch window is the one on the opposite side of the earth from the launch site on the day of launch.

Since Artemis 1 is not targeting a landing, there will be 1 theoretical opportunity to launch every day.  There may be other constraints that eliminate some of those windows.  Things like spacecraft orientation to the sun ( thermal management, communication management) as well as constraints on the arrival time & location back on Earth, launch azimuth restrictions, and lots of other things I cannot imagine right now with my limited knowledge. 

I did note that launch periods 15, 16, & 17 all occur within 4 days of a full moon.  I'm guessing there are good reasons for that.

[slobber91]

One additional good reference for at least Apollo lunar landing launch windows (and a good companion to the film) is the Apollo Flight Journal article on launch windows (linked below).  It provides more details and lots of good stats on all of the actual Apollo lunar mission launch window times and durations, TLI locations, etc. (it shows that TLI for all Apollo missions except Apollo 17 was performed at what is referred to in the film to as the "Pacific" antipode).

While a lot of this is specific to Apollo, it still provides an idea of some of the considerations which must be taken into account when determining lunar launch windows in general. 

https://history.nasa.gov/afj/launchwindow/lw1.html

[c4fusion]

A couple of things I noted:
The video says there are two daily launch windows.  The linked paper says there is one.  I think that is because one of the daily launch windows is not feasible to reach from the launch site and have the spacecraft arrive at the right time for the TLI burn.  The spaceship needs to circularize the orbit before the TLI burn, & this time & geography constraint mean the only meaningful launch window is the one on the opposite side of the earth from the launch site on the day of launch.

I read the paper to mean that there is one perfectly optimal instantaneous launch window per day, but there is more possible launch windows per day but they are less optimal.

Also the video did say that even thought there is 2 day (technically there is far more if you wait in orbit), only one can really be selected if you want to land within a certain area, since the burn over Pacific will take you to the northern hemisphere, while the burn over Atlantic will take you over the southern hemisphere.

[slobber91]

On page 6 of the "Artemis I Trajectory Design and Optimization" paper referenced in Reply 7 and Reply 9, it is stated that a perigee raising maneuver occurs 45 minutes after core stage separation and that the TLI burn then occurs prior to next perigee passage at a true anomaly of between -40deg and -90deg.  This would seem to imply that perigee of the parking orbit at the time of TLI would be roughly in the vicinity of the launch site (KSC at ~80deg W longitude), and therefore that the Artemis I TLI location (and that of the lunar antipode) lies 40 to 90 deg west of KSC, roughly in the range 120deg W to 170deg W longitude.

Based on this, Artemis I seems to be planning to use a Pacific injection.  This range of TLI longitudes also has a good amount of overlap with the range of TLI longitudes for Apollos 8 and 10-16, which all occurred between 142deg W and 142deg E (per the Apollo TLI Statistic table in the Apollo Flight Journal article I linked to in Reply 10).

Attempting to summarize constraints and observations from the Artemis I paper, the Apollo launch window film, and the Apollo Flight Journal launch window essay:

1) The Trans-Lunar Injection maneuver must occur at (or at least very close to) the location of the lunar antipode for the time of arrival at the moon
 
2) The launch from KSC must therefore result in a parking orbit which intersects with this antipode location

3) Because of the limited range of launch azimuths available out of KSC, the resulting parking orbit will only be able to intersect the antipode when it is located either in a limited region of the Atlantic or a similarly limited region of the Pacific (with the exact location depending on lunar declination and launch azimuth).  The launch time is therefore limited to times when the antipode is located in one of these two regions.

4) Both Artemis and Apollo appear to prefer (or are constrained) to perform TLI only when the antipode is located in the Pacific (a Pacific injection).  Therefore, in most cases, the launch time is further limited only to a time when the antipode is located in the Pacific region.

5) The longitude range in which the Artemis I TLI burn can occur (and therefore the range of possible antipode longitudes) appears to be ~50 deg wide in longitude, centered somewhere around 150 deg W, with TLI occurring between ~45 and ~90 minutes after launch.

To circle back and try to answer Nick's original question,

Why would the daily windows be at night?

I put together four diagrams, based on the above constraints & observations, which try to illustrate, at least at a top level, the likely launch lighting at KSC, given a particular lunar phase at moon arrival, assuming a median antipode (and TLI) location at 150 deg W.  These are shown in Cases 1-4 attached at the bottom of this post.

Each diagram shows the earth, moon, and sun locations for a particular lunar phase, as well as earth and moon terminators, as viewed looking down on the earth’s north pole.  The diagrams attempt to show the situation at the time of TLI (~70 minutes after launch)

Case 1 (New Moon at lunar arrival):  the antipode (and thus TLI) location is in the vicinity of local midnight on earth (remember that at new moon, the local time at the location of the antipode will always be midnight, regardless of where on earth the antipode is).  Since KSC is ~70 deg east of this location, local time at the launch site when TLI occurs would be ~4:40am.  Launch time would be ~70 min earlier, or ~3:30am (=> launch in darkness) 

Case 2 (First Quarter at lunar arrival):  local time at the antipode is seen to be roughly 6am; therefore, local time at KSC at TLI is seen to be around 10:40am, (=> launch in daylight).

Case 3 (Full Moon at lunar arrival):  antipode is located at local noon; therefore, local time at KSC is seen to be around 4:40pm (=> launch in daylight).

Case 4 (Third Quarter at lunar arrival):  local time at the antipode is roughly 6pm, and time at KSC would therefore be 10:40pm (=> launch in darkness)

How does this match up with actual Apollo launch lighting conditions for Apollo 8, 10-16?  Since Apollo lunar landings required morning lighting conditions at the landing site (and since all landing sites were on the near side of the moon), all Apollo lunar arrivals occurred after new moon and prior to full moon, or somewhere between the conditions in Case 1 and Case 3 as shown in the above diagrams.  Based on the diagrams for Case 1, 2 and 3, we would predict all Apollo launches using Pacific injection would occur in daylight, and this was indeed the case, with the earliest launch (Apollo 8) occurring at 7:51am local, and the latest launch (Apollo 14) occurring at 4:03pm local.

Next, let's examine the Artemis I launch windows provided in Reply 1.  If we assume lunar arrival occurs a couple of days after launch, the lunar phase at arrival for Launch Period 15 (23 Nov – 10 Dec) would range from just before third quarter to just before first quarter; i.e. it starts between Case 3 and Case 4, and ends between Case 1 and Case 2.  The diagrams predict the launch site would be in darkness for the majority of the Launch Period 15, but a launch at the end of the period might occur just after sunrise.

For Launch Period 16 (21 Dec – 3 Jan), moon phase at arrival would range from just after full moon to just after new moon, covered mostly by Case 3, Case 4 and Case 1.  The diagrams again predict KSC to be in darkness at launch time for essentially the entire period.

The launch site lighting "predicted" by the Case1 - Case 4 diagrams therefore appears to be consistent with actual Apollo launch lighting conditions as well as the fact that that Artemis I launches occurring near the end of 2021/start of 2022 will most likely to occur at night.  Keep in mind that the diagrams are intended to provide only a top level illustration of the connection between lunar phase at arrival and launch time lighting, and may be off by several hours from actual conditions depending on actual TLI longitude (which can apparently vary by around +/- 25 deg), season, actual elapsed time between launch and TLI, and other simplifying assumptions that I’ve made.

[MaxTeranous]

A magnificent and informative post slobberg1, thank you for taking the time.

[Reynold]

Great post, slobberg1.  I have a follow up question, you mention that Apollo went for morning lighting for landing, which makes sense since they had to see topography to avoid boulder fields and such.  The Artemis landings are targeting locations at the South Pole, would lighting conditions there always be "morning like" since the sun is at a low angle?  Or is there any precession/seasonality that would give slightly better lighting conditions at certain times of the month or year? 

[slobber91]

Great post, slobberg1.  I have a follow up question, you mention that Apollo went for morning lighting for landing, which makes sense since they had to see topography to avoid boulder fields and such.  The Artemis landings are targeting locations at the South Pole, would lighting conditions there always be "morning like" since the sun is at a low angle?  Or is there any precession/seasonality that would give slightly better lighting conditions at certain times of the month or year? 

Since the angle of tilt of the moon’s rotation axis relative to the sunline is only 1.5 deg, it is true that lighting would be “morning light” for a good part of the year at the lunar south pole.  But even with only 1.5 deg of axis tilt, winter solstice for the lunar southern hemisphere will still put the sun 1.5 deg below the horizon at the south pole; therefore, a landing site right on the lunar south pole will still be in continuous shadow for some period of time around this solstice.  And summer solstice at the South Pole would result in more reasonable shadow lengths due to the slightly higher sun angle.  Because of this, it wouldn’t surprise me if lunar southern hemisphere summer was preferred for south pole landings, and if there were constraints against south pole landings for some window around winter solstice.

But also keep in mind that the need to have shadows long enough to permit observation of terrain relief was only one of the considerations driving the morning landing requirement for Apollo.  Just as important were:

1) The need to have the landings occur down-sun (sun on the back of the LM) so the astronauts weren’t staring into the sun during the landing, and,

2) The need to have sunlit landmarks early along the powered descent ground track, which the astronauts could time crossing over in order to manually confirm their trajectory (e.g. crater Maskelyn W for Apollo 11)

Since Apollo missions utilized an east-to-west orbit around the moon, a morning landing ensured all of these conditions were met.

I don’t know if Artemis missions will still need landings to occur down-sun and sunlit landmarks to be available along the descent path, but if we assume for the moment that they will, one can make some educated guesses of how this might drive monthly south pole lunar landing windows, depending on how the lunar landing orbit is oriented relative to the earth-moon axis.  I’ve attached diagrams that show this relationship for three different lunar landing orbit orientations.  To simplify things, I’ve ignored earth & moon axis tilt and lunar declination which I don’t think would significantly change the fundamentals.  All three examples are shown at the optimal time during the lunar month for which both the down-sun landing and sunlit landmark requirements.

Example 1 shows the case where the earth-moon axis is parallel to the plane of the lunar landing orbit, and where the ascending node of this orbit is on the earth facing side of the moon; in this case the best time of month to perform the landing would seem to be around new moon.  This provides a full half-orbit of sunlit landmarks prior to landing, and sun is behind the lander at the time of landing at the south pole.   One disadvantage of this case is that the lander would be behind the moon and therefore not in direct radio contact with the earth during all except the very end of powered descent.  This might not be an issue if a relay through the Lunar Gateway or some other satellite in a halo orbit was available.

Example 2 is similar to Example 1, but in this case the descending node of the lunar landing orbit is on the earth facing side of the moon.  In this case, the ideal time of month to perform the landing in order to meet the down-sun and sunlit landmark requirements would be around the time the moon is full.  Example 2 has an advantage over Example 1 in that the entire powered descent and landing would occur in direct view of earth (although preps prior to descent initiation might occur behind the moon).

In Example 3, the earth-moon axis is perpendicular to lunar landing orbit plane (with descending node of the orbit on the sun-facing side of the moon).  Optimal time of month for this orbit in order to meet the down-sun and sunlit landmark requirements is seen to be around the time the moon is at First Quarter.  Example 3 has an advantage over both Examples 1 &2 in that the orbit allows continuous line-of-sight communication with earth throughout the entire orbit.

From my hazy understanding of the Near-Rectilinear Halo Orbit to be used for Lunar Gateway (and assuming the lander would be departing from the Lunar Gateway), I think the lunar landing orbit used for Artemis may end up being closest to Example 3 (or something between Examples 3 and 2), but I may be completely wrong on that.

I’ll reiterate that I don’t know if Artemis descent and landing design will require either down-sun landing or sunlit landmarks along the descent path; it seems to me that use of modern automated vision systems for a computer controlled descent/landing might eliminate the need for either or both of these requirements (and possibly even the need for the landing site to be sunlit).  If so, none of the monthly landing windows described in my examples would be applicable to Artemis.  It may therefore be better to instead consider them as examples of how Apollo might have approached the problem of a South Pole lunar landing.

Finally, to try and keep this post at least somewhat related to the original topic of the thread, which was Artemis I Launch Windows, I wanted to remind readers that Artemis I (and I believe Artemis II) are is performing an insertion into a Distant Retrograde Orbit at moon arrival (not a lunar landing), and therefore any Artemis lunar landing timing constraints would not seem to be applicable to the launch window for these two this mission.

[Steven Pietrobon]

Finally, to try and keep this post at least somewhat related to the original topic of the thread, which was Artemis I Launch Windows, I wanted to remind readers that Artemis I (and I believe Artemis II) are performing an insertion into a Distant Retrograde Orbit at moon arrival (not a lunar landing), and therefore any Artemis lunar landing timing constraints would not seem to be applicable to the launch window for these two missions.

Artemis II won't be going into DRO. It will do a Lunar flyby around the Moon, since nearly all the delta-V of Orion will be used to complete TLI.

[kdhilliard]

Artemis II won't be going into DRO. It will do a Lunar flyby around the Moon, ...

So Artemis I will enter a DRO, but Artemis II won't.

...since nearly all the delta-V of Orion will be used to complete TLI.

Because its payload will be greater than Artemis I's?

And what about Artemis III?  It enters lunar orbit.  Is it that its NRHO is an easier destination than a DRO?

[DanClemmensen]

Artemis II won't be going into DRO. It will do a Lunar flyby around the Moon, ...

So Artemis I will enter a DRO, but Artemis II won't.

...since nearly all the delta-V of Orion will be used to complete TLI.

Because its payload will be greater than Artemis I's?

And what about Artemis III?  It enters lunar orbit.  Is it that its NRHO is an easier destination than a DRO?

I think they chose the "free return" orbit for Artemis 2 because it is formally the qualification test flight for human-rating for both Orion and for SLS.  Qualification test flights traditionally use a small crew and a conservative mission profile. So, no Lunar orbit, and the mission is mostly focused on testing the Orion.

[Proponent]

But why lunar DRO for Artemis 1?  That orbit hasn't otherwise been part of the program since the Asteroid Rendezvous and Redirect Mission was abandoned.  Why not go to NRHO?