SpaceX F9 : USA 328-331 / Globalstar FM15 : SLC-40 : 19 June 2022 (04:27 UTC)

[SPKirsch]

Cross-post:

https://twitter.com/julia_bergeron/status/1538863103701569536

Current recovery positions: Doug with fairings and B1060 on ASOG (Starlink), Finn with JRTI and B1061 (Globalstar-2) and Bob steaming along with fairings.

[jcm]

This is interesting but I really wish people would do some error estimates.
What is the uncertainty on the "54 degrees" measured from the launch info?  Plus or minus 0.1 deg? or plus or minus 1 deg? or what?

A reasonable request, but tough.  For practice, I tried to guess the inclination of a GPS launch from the barge position.  In this case we know the real answer, so we can check.

From the FCC notice, the locations are:
Cape 28 29 11N, 80 32 51W
ASDS 32 49 43N, 75 59  8W

Using the spherical geometry website: https://www.movable-type.co.uk/scripts/latlong.html
We find the initial bearing is 40.95 from North and hence 49.05 from equator

Assuming an instantaneous dV, and an orbit of 533 km (7598 m/s), we find the components in the launch frame:
X = 7598 * cos(49.05 degrees) = 4980 m/s
Y = 7598 * sin(49.05 degrees) = 5739 m/s

Next, we need to correct for Earth rotation, which will add to the X component, giving
X = 4980 + 40000000/(24*3600)*cos(28.5 degrees) = 5387 m/s

Then we re-find the azimuth
atan(5739/5387) = 46.81 degrees from equator and hence azimuth = 43.19

Next, an orbit's azimuth depends on its latitude.  The usual equation is:
azimuth = asin(cos(inc)/cos(lat))

We instead solve for inclination:
sin(azimuth) = cos(inc)/cos(lat))
cos(inc) = sin(azimuth)*cos(lat)
inc = acos(sin(azimuth)*cos(lat))

So we get:
inc = acos(sin(43.19 degrees)*cos(28.5 degrees)) = 53.02 degrees

But this GPS launch was known to be 55.0 degrees.  So we seem to be about 2 degrees off.  Either I made a mistake (entirely possible) or perhaps the extended nature of the launch maneuver causes the difference.

Probably more accurate is differential, which is what I think the original poster did, comparing it to previous launches, their barge location, and their inclination.   If you go through the same exercise with the GlobalStar launch (see below), you get about 0.16 degrees less inclination.  This makes sense since the landing locations were very similar.  This would say the mystery orbit inclination would be about 54.84 degrees.  But the error margin is hard to say.

Second stage burns are even guess-ier.  The duration is hard to estimate from the camera, the startup and shutdown transients are a big part of the dV, and we don't know what throttle settings are used.

---- Same calcs for GlobalStar:

Step 1 : get long, lat of ASDS and Cape from notice
https://apps.fcc.gov/oetcf/els/reports/STA_Print.cfm?mode=current&application_seq=115185&RequestTimeout=100Connection
   Cape: 28 29 11N, 80 32 51W
  ASDS: 32 52 26N, 75 53 58W

Step 2 : find bearing, get 41.16 from north = 48.84 from equator

Step 3 : Consider launch as a single impulse.  V for a 533 km orbit is 7598

Step 4: Find components
X = 7598 * cos(48.84) = 5001
Y = 7598 * sin(48.84) = 5720

Step 5: Add Earth rotation:
X = 5001 + 40000000/(24*3600)*cos(28.5) = 5408
Inertial inclination = atan(5720/5408) = 46.61 degrees; azimuth=43.39

Step 6: Convert azimuth at a latitude to inclination
azimuth = asin(cos(inc)/cos(lat))
sin(azimuth) = cos(inc)/cos(lat))
cos(inc) = sin(azimuth)*cos(lat)
inc = acos(sin(azimuth)*cos(lat))
inc = acos(sin(43.39 degrees)*cos(28.5 degrees)) = 52.86 degrees

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

[Alexphysics]

The theory that they're some sort of prototype Starlink 2.0 satellites that they're launching without anyone knowing falls itself apart when you consider the fact that they'd have to be approved by the FCC. If there were any Starlinks in this flight of any kind they would have to be of the older generation and/or be for a government agency which wouldn't necessarily need FCC approval. That or whoever proposes that theory has to include SpaceX comitting an illegal action that could entail the removal of their rights to operate the Starlink constellation.

Not necessarily, SpaceX didn't file anything when they started flying v1.5, what can be flown under their existing Gen1 license is not clearly defined. Should be obvious that if they did fly Gen2 prototype, it'll still be using the same spectrum as Gen1, this would eliminate most of the concerns from FCC. The altitude and inclination of USA 320 is pretty close to one of the Gen1 orbits as well (97.5 degrees vs 97.6 degrees, 540km vs 560km)

Or they could sidestep FCC by asking their DoD customer such as SDA to classify the launch as part of the DoD program. For example it's quite possible that they're using Starlink v2 bus for their SDA missile warning satellites, so they could ask SDA to authorize a test launch or two of this bus as part of the SDA constellation program. This way SDA gets the peace of mind that their missile warning satellite would actually work once launched, and SpaceX gets to test Starlink v2 bus early, win-win.

SpaceX did file an updated constellation planning to add v1.5 satellites and needed special permission to launch the laser link satellites on Transporter-1 for the polar shells.

[Barley]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

The great circle calculation makes the assumption that the drone ship is directly under the launch trajectory. 

A landing booster has down range and cross range capability.  The extreme is a RTLS, but for anything but a completely maxed out launch there is some.  The cosine losses of a dogleg during the reentry burn moving the landing point on the order of 10km would be quite small.  For lighter launches (but still too heavy for RTLS) hundreds of km of cross range could be possible.

As you hint there may choose a dogleg. Non-orbital mechanic reason to move the drone ship could include avoiding shipping or air lanes, strong currents such as the gulf stream, a jet stream or other weather.  Or they might not bother to optimize the landing position to six decimal places because it really doesn't matter.

Add: Also they might be trolling.  The offer price for twitter shows us that Mr. Musk is not averse to encoding messages in less significant digits.  Digits that had far more value than minutes of latitude.

[JayWee]

https://twitter.com/planet4589/status/1538960051414646796

Space-Track confirms the presence of four secret payloads on the Globalstar Falcon 9 launch  - USA 328 to USA 331, catalog 52889 to 52892, orbital data not available. One piece of debris, probably a Starlink-style tension rod?

[TimTri]

https://twitter.com/planet4589/status/1538960051414646796

Space-Track confirms the presence of four secret payloads on the Globalstar Falcon 9 launch  - USA 328 to USA 331, catalog 52889 to 52892, orbital data not available. One piece of debris, probably a Starlink-style tension rod?

Here we go! Seems like they did end up flying the same amount of secret Starlink satellites they already sent up on Transporter-3. Let’s hope we get some TLEs soon!

[zubenelgenubi]

From SeeSat-l, observation of the 3rd second stage burn:

Lick Observatory Public Programs Telescope Operator (PPTO) Keith Wandry forwarded some images of a rocket burn / fuel dump that was taken last night at Lick Observatory on 6/18/2022 at approximately 23:15 Pacific Daylight Time (6/19 06:15 UT) from near the 40" Nickel Telescope at 37.34302, -121.63717, 4200 ft MSL,  The plume is approximately Azimuth 72.9° (NE), Alt 8.1° in the constellation Pegasus.
<snip>
https://www.flickr.com/photos/xb70man/albums/72177720299927045

[LouScheffer]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

The locations in the notices are precise, but I don't think they are very accurate.  They each state "within 40.5 nautical miles" after the coordinates.  Plus they specify the droneship, and the accompanying ship, at the same coordinates, though we know the ship stands off by about 4(?) miles during the actual landing.  So I suspect the location cannot be deduced to anywhere near 100 meters from the notice.   Internal to SpaceX, of course, the location of the droneship is VERY well known.   That's because the rocket lands on a specific GPS determined spot, and the droneship had better be there.  Since the booster almost always lands in the circle on the deck, the location must be known to a few meters, or better.

[jcm]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

The locations in the notices are precise, but I don't think they are very accurate.  They each state "within 40.5 nautical miles" after the coordinates.  Plus they specify the droneship, and the accompanying ship, at the same coordinates, though we know the ship stands off by about 4(?) miles during the actual landing.  So I suspect the location cannot be deduced to anywhere near 100 meters from the notice.   Internal to SpaceX, of course, the location of the droneship is VERY well known.   That's because the rocket lands on a specific GPS determined spot, and the droneship had better be there.  Since the booster almost always lands in the circle on the deck, the location must be known to a few meters, or better.

Yes , good point.  So a few miles - say  5km - corresponding to more like a 1 degree uncertainty. At which point the claimed change in
inclination is no longer significant.

[JohnLloydJones]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

The locations in the notices are precise, but I don't think they are very accurate.  They each state "within 40.5 nautical miles" after the coordinates.  Plus they specify the droneship, and the accompanying ship, at the same coordinates, though we know the ship stands off by about 4(?) miles during the actual landing.  So I suspect the location cannot be deduced to anywhere near 100 meters from the notice.   Internal to SpaceX, of course, the location of the droneship is VERY well known.   That's because the rocket lands on a specific GPS determined spot, and the droneship had better be there.  Since the booster almost always lands in the circle on the deck, the location must be known to a few meters, or better.

I presume that, at some point, F9 changes over from aiming at where the droneship is supposed to be to using some reference on the ship for final guidance. I have no idea how close it would have to be to switch over.

[cwr]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty.  Usuing the droneship hazard area only, we can get perhaps about a 0.1 degree confidence on its alignment, giving an uncertainty in the intended trajectory. I'm not sure though what the
added uncertainty is for slop from intended to actual. It does look a lot better determined than I would have guessed before doing the math.
gives

The locations in the notices are precise, but I don't think they are very accurate.  They each state "within 40.5 nautical miles" after the coordinates.  Plus they specify the droneship, and the accompanying ship, at the same coordinates, though we know the ship stands off by about 4(?) miles during the actual landing.  So I suspect the location cannot be deduced to anywhere near 100 meters from the notice.   Internal to SpaceX, of course, the location of the droneship is VERY well known.   That's because the rocket lands on a specific GPS determined spot, and the droneship had better be there.  Since the booster almost always lands in the circle on the deck, the location must be known to a few meters, or better.

I presume that, at some point, F9 changes over from aiming at where the droneship is supposed to be to using some reference on the ship for final guidance. I have no idea how close it would have to be to switch over.

That is not the way it works.
LouScheffer accurately summarized the process.

Both ASDS and F9 booster endeavor to arrive at the same GPS coords at the time of landing.

For much of the entry and landing the F9 is targeting a spot displaced from those coords so
that if something goes wrong it will land in the water not on the ASDS.
If you watch the landing video from the F9 you can see it switch from that target to the agreed coords
for the ASDS.

Carl

[OneSpeed]

Well, we can simplify the request to 'what is the uncertainty on the launch azimuth'.   The FCC position is given to 1" of lat/lon corresponding
to 31 metres, but I am skeptical the droneship is held to that degree of accuracy. Let's say it's good to 100m. Taking that and 655 km from the pad gives 0.01 degree uncertainty on the azimuth, which is pretty good.  But it's clear (e.g. from Raul's LHA maps) that there
was something of a dogleg, and that introduces a larger uncertainty. ...

From the Mission Control Audio, there wasn't so much a dogleg as a gradual curve to the south. This should be expected, because the ground track is part of a great circle route, which over an entire orbit is more like a distorted sinusoid.

Raul was kind enough to include a link to his GlobalStar Google Map, and I've taken the liberty of superimposing some ground tracks that conform to the orientation of the primary LHA.

The locations in the notices are precise, but I don't think they are very accurate.  They each state "within 40.5 nautical miles" after the coordinates. ...

They might not be very accurate, but the shape of the ground track should at least be a curve.

[Orbiter]

There was a dogleg. You can see it in the streak shots.

[OneSpeed]

There was a dogleg. You can see it in the streak shots.

So the Mission Control Audio webcast is wrong? I can see an initial roll to the azimuth, but no actual dogleg (a sharp yaw).

[su27k]

The theory that they're some sort of prototype Starlink 2.0 satellites that they're launching without anyone knowing falls itself apart when you consider the fact that they'd have to be approved by the FCC. If there were any Starlinks in this flight of any kind they would have to be of the older generation and/or be for a government agency which wouldn't necessarily need FCC approval. That or whoever proposes that theory has to include SpaceX comitting an illegal action that could entail the removal of their rights to operate the Starlink constellation.

Not necessarily, SpaceX didn't file anything when they started flying v1.5, what can be flown under their existing Gen1 license is not clearly defined. Should be obvious that if they did fly Gen2 prototype, it'll still be using the same spectrum as Gen1, this would eliminate most of the concerns from FCC. The altitude and inclination of USA 320 is pretty close to one of the Gen1 orbits as well (97.5 degrees vs 97.6 degrees, 540km vs 560km)

Or they could sidestep FCC by asking their DoD customer such as SDA to classify the launch as part of the DoD program. For example it's quite possible that they're using Starlink v2 bus for their SDA missile warning satellites, so they could ask SDA to authorize a test launch or two of this bus as part of the SDA constellation program. This way SDA gets the peace of mind that their missile warning satellite would actually work once launched, and SpaceX gets to test Starlink v2 bus early, win-win.

SpaceX did file an updated constellation planning to add v1.5 satellites and needed special permission to launch the laser link satellites on Transporter-1 for the polar shells.

SpaceX filed an updated constellation plan so that they can move all the satellites from 1200km to 550km, in the same filing they added polar shells, this is a change of all the shells of the constellation, it's not specific to a satellite design. The designation "v1.0" or "v1.5" did not appear in this filing, what FCC cares most is the orbit and spectrum, not some satellite design.

FCC didn't approve this new constellation plan until April 2021, so when SpaceX wants to fly some Starlink to polar orbit on Transporter-1 in Jan 2021, they're flying to an orbit not approved by FCC yet, this is why they need to ask special permission. It has nothing to do with the fact that the satellites flown to polar orbits are v1.5, they could very well fly v1.0 on Transporter-1 and they'd still need to ask for permission. In fact we don't know that those flown on Transporter-1 are v1.5's, they're likely prototypes since they were all deorbited in less than a year.

I believe the first official v1.5 launch is Group 2-1 on September 13, 2021, with Elon Musk's confirmation: https://twitter.com/elonmusk/status/1436541063406264320

[su27k]

https://twitter.com/planet4589/status/1538960051414646796

Space-Track confirms the presence of four secret payloads on the Globalstar Falcon 9 launch  - USA 328 to USA 331, catalog 52889 to 52892, orbital data not available. One piece of debris, probably a Starlink-style tension rod?

Just realized the significance of the number 4: It's the # of missile warning satellites SDA contracted SpaceX to build: https://spacenews.com/spacex-l3harris-win-space-development-agency-contracts-to-build-missile-warning-satellites/

The Space Development Agency awarded  $193.5 million to L3Harris and $149 million to SpaceX to build four satellites each to detect and track ballistic and hypersonic missiles.

[gongora]

They aren't the SDA satellites, those will launch later.

[su27k]

They aren't the SDA satellites, those will launch later.

Well they're not the ones specified in the SDA launch contract, those goes to near polar orbit at 950km. But it's a awful big coincidence here don't you think? Seems to me this could be some kind of testing for SDA satellites.

[gongora]

But it's a awful big coincidence here don't you think?

I really don't.  The government launches lots of different satellites.

[Galactic Penguin SST]

https://twitter.com/cgbassa/status/1539044005564100609?s=21&t=m1HdleuTxMXnV1glSXGX3g