SpaceX F9 : Starlink v1.0 L19 : CCSFS SLC-40 : 15/16 Feb 2021 (0359 UTC)

[yokem55]

Here is a comparison of the S1 telemetry from Starlink L18 and L19.

1. The result of the boost phase is identical in terms of velocity and altitude gained.
2. The L19 entry burn acceleration reduces by about a quarter from 399-401s, and continues at that level until MECO2 at 405s.
3. This difference may well correspond with a single early engine shutdown.

I tried to find the graph from L5 last March (the last engine out mission), but I didn't see it in a cursory scroll through that thread. Any idea where that one might be?

[mikelepage]

Not sure how to assess whether it's real or just an artefact of the video feed, but just looking at the feed from the drone ship, the "rocket glow" seems to fluctuate in brightness at about 2 cycles per second for 5ish repetitions, before a final flare.

Could perhaps indicate the stage was on fire and spinning wildly out of control (at about 2Hz) before it hit the water.

[OneSpeed]

I tried to find the graph from L5 last March (the last engine out mission), but I didn't see it in a cursory scroll through that thread. Any idea where that one might be?

https://forum.nasaspaceflight.com/index.php?topic=50372.msg2059817#msg2059817

Of course it was only one engine out of 9, so the difference was more subtle.

[klod]

https://twitter.com/elonmusk/status/1361544540411355137

Yeah. Active fairing half recovered though.

Sounds like good fairing splash down and Elon acknowledges landing failure, a lot better than the weeks it took for B1056 and 11 months (and counting) for B1048.

What does Elon mean by saying "Active"?

[FutureSpaceTourist]

https://twitter.com/julia_bergeron/status/1361536442057515010

Farwell B1059. Starlink L19 is to be the final mission for this booster and OCISLY will return with an empty deck.

Mission details: nasaspaceflight.com/2021/02/starli…

https://twitter.com/spacecoast_stve/status/1361552097733468161

Falcon 9 B1059 streaks off into the night on what would turn out to be its last mission.

She had a good run, having launched 6 missions - 2 Dragons, SAOCOM, NROL-108, and a total of 118 Starlinks. Farewell, good booster.

nasaspaceflight.com/2021/02/starli…

https://twitter.com/spacecoast_stve/status/1361553928849096704

Have you ever wondered if you can get a streak shot with your phone? Well, you can!

I used Slow Shutter Cam for iOS, sort of as an afterthought, and now I know it can work. Next time, I'll tell my phone not to lock after 2 minutes, so it doesn't cut the streak off suddenly.

[FutureSpaceTourist]

Photos from SpaceX website by Ben Cooper

[Jansen]

Yeah. Active fairing half recovered though.

Sounds like good fairing splash down and Elon acknowledges landing failure, a lot better than the weeks it took for B1056 and 11 months (and counting) for B1048.

What does Elon mean by saying "Active"?

There are two halves to a fairing set. One half has     hardware to facilitate separation and is designated as active, while the other is passive.

The two halves of the fairing are fastened by mechanical latches along the fairing vertical seam. To deploy the fairing, a high-pressure helium circuit releases the latches, and four pneumatic pushers facilitate positive-force deployment of the two halves. The use of all-pneumatic separation systems provides a benign shock environment, allows acceptance and preflight testing of the actual separation system hardware, and minimizes debris created during separation.

[eeergo]

Here is a comparison of the S1 telemetry from Starlink L18 and L19.

1. The result of the boost phase is identical in terms of velocity and altitude gained.
2. The L19 entry burn acceleration reduces by about a quarter from 399-401s, and continues at that level until MECO2 at 405s.
3. This difference may well correspond with a single early engine shutdown.

Thanks for the analysis.

Cross-posting from a budding discussion in "the other" Starlink thread:

The just-launched mission has also seen some delays seemingly related to something on the vehicle - and appears to have suffered an engine loss *during ascent* that led to loss of control during the subsequent burn, not being a "life leader" stage at its 5th flight (however amazing that statement reads anyway).

The telemetry comparison with the previous Starlink mission does not support a theory about the engine loss during ascent, rather during an entry burn: https://forum.nasaspaceflight.com/index.php?topic=52895.msg2192524#msg2192524

It doesn't mean it's not worth being thoroughly checked though, landing reliability is very important for the launch cadence.

Fair enough, but the pre-MECO "bucket" (in the interval [145,157] s) does look much more out of family than other recent flights (around 15-20% inferior), in spite of less noisy TM. See, for instance:

https://forum.nasaspaceflight.com/index.php?topic=46524.msg2167636#msg2167636
https://forum.nasaspaceflight.com/index.php?topic=52045.msg2170673#msg2170673

I would hesitate to claim "the telemetry comparison does not support a theory about engine loss during ascent" until post-MECO MET with the available data. That is not to say it unequivocally supports it either, so I should have better qualified that section of my statement.

I would also argue checking the engine issue thoroughly is a necessity regardless of the importance of recovery towards future launch cadence, since it could as well have popped up during ascent (even if it didn't in this situation, which again is debatable), unless the failure signature they have is exclusive of reentry conditions, if there's such a thing.

What do you make of that lower-than-expected (?) 12-second period? Nothing out of the ordinary was clearly visible in the webcast since they were showing a very wobbly tracking shot during that time.

[kdhilliard]

Here is a comparison of the S1 telemetry from Starlink L18 and L19.

1. The result of the boost phase is identical in terms of velocity and altitude gained.
2. The L19 entry burn acceleration reduces by about a quarter from 399-401s, and continues at that level until MECO2 at 405s.
3. This difference may well correspond with a single early engine shutdown.

Both L18 and L19 execute a g-limiting throttle reduction about 10 s prior to MECO.
I assume the spike in 18's acceleration plot immediately prior to that is an artifact of noisy telemetry.
But what do you make of 19's ~ 1.6X  greater reduction in acceleration?

Attached acceleration graphs:
* Crop of the final 30s of powered booster ascent.
* Crop of entry burn.

Aside:
Is this throttle reduction done by shutting down the center engine?
Have we ever actually seen this in video, or are the plumes too diffuse by then?

[aviators99]

If there was a loss of a single engine during entry burn, would that likely cause entry heat damage?  Clearly, the booster made it down, as we could see the landing burn.  Do we think that SpaceX decided not to attempt landing prior to the landing burn?  If so, what would be the purpose of the burn?  To avoid an higher impact force that might damage OCISLY, even from a translated X,Y?  I understanding this is all conjecture, at this point.

[Vettedrmr]

No, we saw reflections of something burning while it descended, and the explosion on impact.  Whether it was the single engine landing burn or not we don't know.  It also was changing in brightness about twice per second, so I'm wondering if it wasn't in a spin of some kind.

Hope to learn more in the near future.

Have a good one,
Mike

[RocketLover0119]

No, we saw reflections of something burning while it descended, and the explosion on impact.  Whether it was the single engine landing burn or not we don't know.  It also was changing in brightness about twice per second, so I'm wondering if it wasn't in a spin of some kind.

Hope to learn more in the near future.

Have a good one,
Mike

I think the stage was on fire as it plunged, then the larger flash was impact. The seagulls also were startled when that bright flash occurred, but luckily they didn’t turn into seafood 

I almost think the stage may have almost fully either exploded or was engulfed in flames, hence the telemetry drop out, but I hope we know something official soon.

[dondar]

Uninformed speculation: Off-center plasma is due to incorrect attitude.  The grid fins deployed but didn't seem to be moving after that.  Perhaps another hydraulic pump failure?

it was not only plasma. It was low pressure fuel leak (fire?). There was way too much material for plasma glow.

[RocketLover0119]

Uninformed speculation: Off-center plasma is due to incorrect attitude.  The grid fins deployed but didn't seem to be moving after that.  Perhaps another hydraulic pump failure?

it was not only plasma. It was low pressure fuel leak (fire?). There was way too much material for plasma glow.

That was for sure not plasma, looked liquidy almost, I think a leak or fire

[gaballard]

It was at least somewhat plasma... you see the same flashes and colors in the fairing re-entry video. You don't see bright blue jets like this from the Merlins during entry burns.

If one engine of three failed during re-entry, you'd probably end up with this situation; increased drag as the stage gets into the thicker atmosphere, followed by plasma heating as the protective exhaust "sheath" isn't fully enveloping the vehicle anymore.

Seeing as we saw pieces ablating from the far-right engine (from the camera's perspective), followed by plasma heating on that side, I think it's extremely likely that engine failed during re-entry, leading to loss of vehicle.

[LouScheffer]

Here is a comparison of the S1 telemetry from Starlink L18 and L19.

1. The result of the boost phase is identical in terms of velocity and altitude gained.
2. The L19 entry burn acceleration reduces by about a quarter from 399-401s, and continues at that level until MECO2 at 405s.
3. This difference may well correspond with a single early engine shutdown.

Both L18 and L19 execute a g-limiting throttle reduction about 10 s prior to MECO.
I assume the spike in 18's acceleration plot immediately prior to that is an artifact of noisy telemetry.
But what do you make of 19's ~ 1.6X  greater reduction in acceleration?

Attached acceleration graphs:
* Crop of the final 30s of powered booster ascent.
* Crop of entry burn.

The end of the first stage burn is interesting.  The peak acceleration is 3.2Gs, so even with an engine out, the stage should be able to manage 2.85 Gs.  But it doesn't even do that.  I can think of two explanations:
(1) The acceleration of L19 was a little higher all throughout the trajectory.  If there is some sort of terminal guidance for stage 1 (to help hit the barge) maybe it backed off the acceleration the last 10 seconds to hit the target.
(2) The final reduction in acceleration is open loop.  Perhaps they simply throttle all engines back to 80%.  Then if one engine shut down at this time, you'd see a reduction from 2.8 to 2.5 Gs as observed.

If (2) is the explanation, it could account for both events.  Perhaps an engine went out of envelope when throttled down.  It might make sense to shut it off, since it's not absolutely needed and SpaceX would not want a bad engine to destroy others.  But for recovery, they'll lose the booster if doesn't work (only 3 engines can restart, and the increased gravity losses from a two engine burn would not leave enough fuel for landing). So it makes sense to ignore limits and give it a try.  But then it failed partway through the re-entry burn.

[AstroDave]

Even with unfortunate booster failure, does recovery of large grid fins warrant exploration. Yes, they could be damaged, but it seems like a lot of titanium to let go of. Even if just one was reusable, it would seem to be of benefit.

[RocketLover0119]

Even with unfortunate booster failure, does recovery of large grid fins warrant exploration. Yes, they could be damaged, but it seems like a lot of titanium to let go of. Even if just one was reusable, it would seem to be of benefit.

Likely won’t go after them, in past failures they’ve just had to let them go. Pricey, but it’s just what is safest sometimes to let it go.

[FutureSpaceTourist]

https://twitter.com/djsnm/status/1361720223766700032

Comparison of booster accelerations around time of entry burn - blue is last night's flight, red is from 2 weeks ago.
It's not a full 33% reduction in thrust so it wasn't a complete engine failure.
Afterwards you can see aerodynamic drag rises faster before loss of telemetry

twitter.com/bexben9/status/1361720718426337283

Decreased drag likely due to control system not aiming for the droneship after a detected anomaly

https://twitter.com/djsnm/status/1361721803320696832

It's consistent with it simply being deeper in the atmosphere at a higher speed

[cpushack]

Even with unfortunate booster failure, does recovery of large grid fins warrant exploration. Yes, they could be damaged, but it seems like a lot of titanium to let go of. Even if just one was reusable, it would seem to be of benefit.

It would depend on the water depth, recovery costs get exponentially more expensive the deeper the water is.  But if its shallower and easy to find, then the cost of recovery may be cheaper then the cost of making new ones.  Grid fins are something that doesn't seem to be a limit though (they probably have more grid fins then boosters)