Author Topic: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video  (Read 160593 times)

Online QuantumG

  • Senior Member
  • *****
  • Posts: 8712
  • Australia
  • Liked: 3649
  • Likes Given: 857
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #160 on: 07/24/2014 05:46 AM »
I did leave the lawyering to the lawyers.  I asked my patent attorney for his opinion prior to posting.

Heh, did he give you that horrible attempt at prior art too? How much did he charge you to read the patent?

I hear those things are awfully loud. It glides as softly as a cloud. What's it called? Monowhale!

Offline HMXHMX

  • Full Member
  • ****
  • Posts: 1519
  • Liked: 1357
  • Likes Given: 346
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #161 on: 07/24/2014 06:28 AM »
I did leave the lawyering to the lawyers.  I asked my patent attorney for his opinion prior to posting.

Heh, did he give you that horrible attempt at prior art too? How much did he charge you to read the patent?



Nothing.  He used to work for me and wrote the book.  There are numerous other examples of prior art, none of which are quoted by the patent.  It is utterly indefensible on several grounds – and I do know something about the prior art here.

But we are straying too far off topic.

Online QuantumG

  • Senior Member
  • *****
  • Posts: 8712
  • Australia
  • Liked: 3649
  • Likes Given: 857
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #162 on: 07/24/2014 06:41 AM »
Nothing.  He used to work for me and wrote the book.

Oh, Pat.

Quote from: Grandpa Simpson
Why the fax machine is nothing but a waffle iron with a phone attached!

That's how every discussion about patents seems to go on the Internet.
I hear those things are awfully loud. It glides as softly as a cloud. What's it called? Monowhale!

Offline deruch

  • Senior Member
  • *****
  • Posts: 2355
  • California
  • Liked: 1896
  • Likes Given: 4437
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #163 on: 07/24/2014 07:47 AM »
Repairing the CRS-3 landing was restoring a painting that had been put through a shredder.  Repairing this video would be trying to restore a painting where the painter was constantly being interrupted by a toddler drawing all over their canvas as they were working.

I really like the shredder description, but the rest falls a bit flat.  The real point is that the CRS-3 repair was about recovering an underlying picture who's data was garbled in transmission.  In this video, the data is transmitted without error, but rather there is an obstruction between the underlying picture and its recording.  Fixing it would be more like taking a cubist painting and trying to recreate the view the artist was looking at when they painted it.  But yeah, different tools and skills would be needed.
« Last Edit: 07/24/2014 07:48 AM by deruch »
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline rpapo

  • Cybernetic Mole
  • Full Member
  • ****
  • Posts: 1198
  • Michigan, USA
  • Liked: 631
  • Likes Given: 486
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #164 on: 07/24/2014 11:21 AM »
One tidbit I just noticed.  If you replay the landing video and take the timestamps just before and after the middle fadeout (between the retro burn and the landing burn), you see 15:24:17 and 15:25:39, or only one minute and 22 seconds.  That's a mighty short time period, which implies that the retro burn is occurring at a much lower altitude than I (at least) would expect.  Even if the rocket were still descending at 1000mph after completion of the retro burn, and averaged around 700 mph during the descent (faster at higher altitudes, slower near the end due to thickening air), that works out to only about 15 miles or so.

I am not a rocket scientist, nor an aeronautical engineer.  If my napkin calculation is wrong, please explain.
An Apollo fanboy . . . fifty years ago.

Offline Dudely

  • Full Member
  • ***
  • Posts: 312
  • Canada
  • Liked: 109
  • Likes Given: 92
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #165 on: 07/24/2014 11:29 AM »
Guys, we really need to lay off the patent discussion, as it's a pretty clear-cut issue. Not only is it not valid, but I'm almost sure they will be doing this operation just a few times, and always in international waters.

Kinda hard to prosecute.

Offline cscott

  • Senior Member
  • *****
  • Posts: 2949
  • Liked: 2056
  • Likes Given: 664
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #166 on: 07/24/2014 12:56 PM »
One tidbit I just noticed.  If you replay the landing video and take the timestamps just before and after the middle fadeout (between the retro burn and the landing burn), you see 15:24:17 and 15:25:39, or only one minute and 22 seconds.  That's a mighty short time period, which implies that the retro burn is occurring at a much lower altitude than I (at least) would expect.  Even if the rocket were still descending at 1000mph after completion of the retro burn, and averaged around 700 mph during the descent (faster at higher altitudes, slower near the end due to thickening air), that works out to only about 15 miles or so.

I am not a rocket scientist, nor an aeronautical engineer.  If my napkin calculation is wrong, please explain.

I think the real question is: what was the timestamp of stage separation?  We know there was some RCS maneuvering immediately after sep.  Did the retro burn immediately follow that, or did they allow themselves to "drift" downrange?  (And if they were targetting RTLS, would they have drifted, or done the burn right away?)

Offline rpapo

  • Cybernetic Mole
  • Full Member
  • ****
  • Posts: 1198
  • Michigan, USA
  • Liked: 631
  • Likes Given: 486
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #167 on: 07/24/2014 01:12 PM »
One tidbit I just noticed.  If you replay the landing video and take the timestamps just before and after the middle fadeout (between the retro burn and the landing burn), you see 15:24:17 and 15:25:39, or only one minute and 22 seconds.  That's a mighty short time period, which implies that the retro burn is occurring at a much lower altitude than I (at least) would expect.  Even if the rocket were still descending at 1000mph after completion of the retro burn, and averaged around 700 mph during the descent (faster at higher altitudes, slower near the end due to thickening air), that works out to only about 15 miles or so.

I am not a rocket scientist, nor an aeronautical engineer.  If my napkin calculation is wrong, please explain.

I think the real question is: what was the timestamp of stage separation?  We know there was some RCS maneuvering immediately after sep.  Did the retro burn immediately follow that, or did they allow themselves to "drift" downrange?  (And if they were targetting RTLS, would they have drifted, or done the burn right away?)
I think (I don't know) that they skipped the boostback burn altogether on this flight.  So the stage separation and reorientation would have occurred at about the three minute mark, followed by a controlled drift (keeping the stage orientation steady with RCS) for twelve whole minutes, arcing up, and then down.  The reentry burn would have occurred soon after atmospheric drag started to make itself felt.

Looking back at the launch videos (presumably the same clock and same camera), the timer at stage separation was approximately 15:17:50.  So retro burn was about 6:30 afterwards.

Both readings are approximate, since the timer at stage separation was hard to read (white on white), and when we see the retro burn, it had already started.
An Apollo fanboy . . . fifty years ago.

Online dcporter

  • Full Member
  • ****
  • Posts: 871
  • Liked: 246
  • Likes Given: 406
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #168 on: 07/24/2014 01:14 PM »
I did leave the lawyering to the lawyers.  I asked my patent attorney for his opinion prior to posting.
Heh, did he give you that horrible attempt at prior art too? How much did he charge you to read the patent?

Nothing.  He used to work for me and wrote the book.  There are numerous other examples of prior art, none of which are quoted by the patent.  It is utterly indefensible on several grounds – and I do know something about the prior art here.

But we are straying too far off topic.

Maybe Commercial Space Flight General or Space Policy? I would be interested in seeing this discussion continue…

Offline cscott

  • Senior Member
  • *****
  • Posts: 2949
  • Liked: 2056
  • Likes Given: 664
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #169 on: 07/24/2014 01:15 PM »
1. So, no external video? <Groan>. What about the video from the NASA plane which filmed the hypersonic “re-entry”?
I agree that I'm hoping to see this at some point.  But I don't think SpaceX "owes it to me".
Come to think of it, you don't even need a surface ship in order to get video. [...] Get a few quad-copters on a submarine
Radio waves don't propagate underwater.  (Except for VLF.)

I don't think you need a quadcopter when you have Elon's personal jet in the area.

2. Did they do any boost-back on this flight? Anyone know the co-ordinates of the splashdown?
The nautical keep out area was posted (I'm too lazy to look it up, it might have been in L2) and it was indeed significantly closer to shore than on the previous flight.  We also had real time flight tracking of the various aircraft involved in observing the flight, and the area they were canvassing was consistent with the keep out area.  That's in the public thread, you can go find it.

It's clear there was a retro burn, it's on the video.  Is that "boost back"?  Well, it returned the stage closer to shore than previously.  There might be an earlier burn (either planned or executed), and even if the retro burn was "boost back", it might not be "precision boost back" -- as someone else has noted, when you write the control software you can specify free variables that aren't strictly controlled (in order to save propellant, improve precision on other variables, etc).  Rotation is typically one of them -- no one cares what direction the logo on the rocket is facing when it lands.  We don't know what other free variables were specified on this landing attempt.

Wouldn't they want to demonstrate that they could fly the F9 S1, precisely, and using the same mass-vs-time profile as an S1 that had to carry prop reserve for boost-back?

No, that's what F9R is for.  These landing attempts are about supersonic retropropulsion and reentry heating on the structure, which is the flight regime F9R can't easily test.  (Propellant quantities can be computed, they don't need a test flight for that.)

That suggests to me that that the next one will travel a significant distance cross-range (instead of boost-back), and that perhaps there's some uninhabited area (as opposed to the relatively densely populated Space Coast?) that SpaceX can touch down on.

Crazy talk.  The progression between cassiope, CRS-3 and ORBCOMM flight has been landing closer to shore each time.  And take a look at the map sometime -- the Atlantic isn't filled with tiny islands.  It's empty out there (and the water is *deep*).

3. Plus, I think they'd have to demonstrate more than pin-point landing accuracy.

Sure, that's what telemetry is for.  They know exactly what their error terms are, at every point and for every variable they care about.

(EDIT: Couldn't all three S1 cores separate at the same time? [...])
That would defeat the point.  FH is a three-stage rocket.  The two boosters are the first stage.  Getting rid of a stage has a large negative impact on performance; study the rocket equation.

4. I don't understand why a floating platform would be uneconomical.

Understanding economics would be helpful.  But briefly: additional facilities add overhead, cores which exist but aren't current flying add overhead and initial capital cost, twice the flights (if you're flying the cores back) mean half the useful money-earning life of the rockets, it costs a lot to take things apart and put them back together, and there's always the chance that you put them together wrong, the sea atmosphere is corrosive => at sea operations are especially expensive, the nautical facilities involved have to be built on quite a large scale in order to get acceptable stability, etc.

None of this says it's *impossible*, or that there isn't a way to close the economic case ("big dumb barge", eg), but it's certainly not obvious.  If they plan on doing this, you'd certainly start seeing a gradual development program, because the whole thing is certainly not trivial to make work.  Perhaps the flights after CRS-4 will reveal the start of this development program.  If I were to bet, though, I'd bet "no".
« Last Edit: 07/24/2014 01:20 PM by cscott »

Offline meekGee

  • Senior Member
  • *****
  • Posts: 8780
  • N. California
  • Liked: 4968
  • Likes Given: 904
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #170 on: 07/24/2014 05:50 PM »
One tidbit I just noticed.  If you replay the landing video and take the timestamps just before and after the middle fadeout (between the retro burn and the landing burn), you see 15:24:17 and 15:25:39, or only one minute and 22 seconds.  That's a mighty short time period, which implies that the retro burn is occurring at a much lower altitude than I (at least) would expect.  Even if the rocket were still descending at 1000mph after completion of the retro burn, and averaged around 700 mph during the descent (faster at higher altitudes, slower near the end due to thickening air), that works out to only about 15 miles or so.

I am not a rocket scientist, nor an aeronautical engineer.  If my napkin calculation is wrong, please explain.

I think the real question is: what was the timestamp of stage separation?  We know there was some RCS maneuvering immediately after sep.  Did the retro burn immediately follow that, or did they allow themselves to "drift" downrange?  (And if they were targetting RTLS, would they have drifted, or done the burn right away?)
I think (I don't know) that they skipped the boostback burn altogether on this flight.  So the stage separation and reorientation would have occurred at about the three minute mark, followed by a controlled drift (keeping the stage orientation steady with RCS) for twelve whole minutes, arcing up, and then down.  The reentry burn would have occurred soon after atmospheric drag started to make itself felt.

Looking back at the launch videos (presumably the same clock and same camera), the timer at stage separation was approximately 15:17:50.  So retro burn was about 6:30 afterwards.

Both readings are approximate, since the timer at stage separation was hard to read (white on white), and when we see the retro burn, it had already started.

You are right.  I care less about how long it was since stage separation - what matters is the backwards timeline, working from splashdown back up to the retro-burn, and they are indeed awfully close, if the timer is consistent.

1 minute and 22 seconds at terminal velocity is not much at all.
ABCD - Always Be Counting Down

Offline Mike_1179

  • Member
  • Full Member
  • ****
  • Posts: 563
  • New Jersey
  • Liked: 231
  • Likes Given: 52
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #171 on: 07/24/2014 06:13 PM »
Wasn't this a more lofted trajectory than normal?  Is it possible they didn't need a large boost-back because the first stage went higher and less downrange than previous?

Offline rpapo

  • Cybernetic Mole
  • Full Member
  • ****
  • Posts: 1198
  • Michigan, USA
  • Liked: 631
  • Likes Given: 486
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #172 on: 07/24/2014 06:18 PM »
Wasn't this a more lofted trajectory than normal?  Is it possible they didn't need a large boost-back because the first stage went higher and less downrange than previous?
Frankly, I'm not sure they did any boostback at all.  They ran one seriously lofted trajectory this time.
An Apollo fanboy . . . fifty years ago.

Offline cscott

  • Senior Member
  • *****
  • Posts: 2949
  • Liked: 2056
  • Likes Given: 664
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #173 on: 07/24/2014 06:50 PM »
1 minute and 22 seconds at terminal velocity is not much at all.

I believe the retro burn began shortly after stage separation; we don't see the start of the burn on the video.  If makes sense to null out the horizontal velocity as soon as possible.  My guess is that the 1-2 minute glide is basically the performance/safety margin for the boostback; if reentry drag were lower or mass was higher etc the glide would be even shorter.  I bet there's a good reason why the retro burn wants to distribute the thrust over as long a time period as possible; otherwise they'd relight all nine engines and get their delta-v 3x quicker.  (For one, the nozzles will become more efficient as the stage re-approaches the ground.)

It's interesting that they still need to turn off the engines completely for 1-2 minutes.  I'm sure they'd get rid of the second relight of the center engine if they could. 

The description SpaceX posted on the video clearly states "restart main engines twice".  I suppose it's possible that the first restart is a boost back burn right after stage sep, and the second is the retro burn shown in the video... and then the center engine does not shut down but continues to run throttled down until the start of the "landing burn" (which is just the throttle-up point, not a relight).

If that were the case, I'd say that the first boost back burn would be designed to put the stage on a ballistic trajectory that will still impact safely off shore if something goes wrong (like the engine fails to relight, or the stage breaks up during reentry max Q).  The "retro" burn might be primarily designed to move the trajectory on-shore over the landing zone and serves as the final engine check and decision point.  This would be broadly similar to the Dragon v2 landing profile, which contains a higher altitude "engine check" before the final brown-pants landing burn.

ps. we know SpaceX has video footage from that camera from liftoff through staging and then to landing.  They probably have their reasons for not releasing all of it.  I bet they would have preferred not to reveal even the brief footage of the "retro" burn we got, but felt they needed to do so to explain the ice buildup.

UPDATE: as mwfair says in a different thread:
The flight radar track shows that F900 was 336 km downrange from the pad.  [...]  I'll have to run the numbers, but that seems pretty close to a ballistic trajectory.
If it's a ballistic trajectory and the closer return was due solely to the lofted launch, then most of my reasoning above is bogus.  Oh, well.
« Last Edit: 07/24/2014 09:49 PM by cscott »

Offline mwfair

I didn't see any aggregated listing of time and altitude for 1st stage (or second), so here is my analysis of the videos:

time    alt  speed range  note
         km   mps   km
launch video
15:14:30    0   0   0       
+1:00    13  450     2.5     
+2:20    54  1400    21     
+2:25                40    sep
+2:40    112 1700    51    MVAC   
+4:30    240 1800    130   NewHampshire   
+5:30    334 2200    209     
+6:30    420 2700    305     
+7:30    511 3600    460     
+8:45    600 5700    150???     
+9:30    620 7200    1000  2ndStage cutoff
landing video                   
+9:13     start of music 
+9:23     ?   ?    ?         retro burn 
+9:49     ice
+11:09    ?   ?    ?         oceanvisible   
+11:12    ?   ?    ?         landing RCS
+11:20    ?   ?    ?         roll stopped   
+11:27    ?   ?    ?         legs   
+11:33    ?   ?    ?         full plume 
+11:36    ?   0    330     engine out 
+11:39    ?   0    330     splash subsides
+11:41    ?   0    330     start tipover   
+11:43    ?   0    330     kaboom 
« Last Edit: 07/25/2014 03:10 AM by mwfair »
Mike Fair

Offline cscott

  • Senior Member
  • *****
  • Posts: 2949
  • Liked: 2056
  • Likes Given: 664
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #175 on: 07/24/2014 09:08 PM »
+11:12                     landing RCS

On the video this is titled "landing burn".  Don't know if it's worth trying to hyperparse wording.

If you use the (pre) and (/pre) tags around your table the columns will line up better.

Offline mwfair

On the video this is titled "landing burn".
I don't believe the wording in the video, at least not precisely.  Obviously an engine is burning.  But there is a big difference in the flame size at +11:12 vs +11:33 , and the earlier one has a pulsating pattern.
Also, time to decelerate from terminal velocity to hover is T = V / a where a = (F-W)/m = (F/m - g) => T = V/(F/m - g) .   Guesses of V=100m/s, F = 600 kN, m = 18000 kg yields deceleration of 23m/s2 and firing time of 3 seconds, which is consistent with my parsing of the video.  However, the video label "landing burn" occurs 24 seconds before splashdown.

EDIT: changed embarrassing algebra and unit and arithmetic mistakes.  original had a = 30g, should be 23 m/s2 = 2.3g.
« Last Edit: 07/25/2014 06:39 PM by mwfair »
Mike Fair

Offline meekGee

  • Senior Member
  • *****
  • Posts: 8780
  • N. California
  • Liked: 4968
  • Likes Given: 904
Re: Falcon 9 v1.1 ORBCOMM - First Stage Ocean Landing Video
« Reply #177 on: 07/25/2014 03:58 AM »
Oh, one thing I don't think we knew before - the re-entry burn uses 3 engines.

We knew the boost-back burn uses 3 (from the video, and it's clear you need to turn back as fast as possible, and the stage is still heavy) and we knew the landing burn uses 1 (empty stage, hover-slam) but the secret-sauce burn is a little less secret now.  We know when it occurs, and we know it uses 3 engines.
ABCD - Always Be Counting Down

Offline mwfair

Not sure if anyone has done a terminal velocity calculation.
Drag = Cd * 0.5 * rho * V^2 * Area_wetted.    Long cylinder has Cd = 0.8.  rho = 1.2 kg/m3.  Area = length * pi * D^2 = 400 sq.meters.   Thus V_terminal = sqrt ( 2 * Weight / Cd * rho * A_wetted).   empty F9 masses 18 tons = 180 kN.    Vt = about 40 m/s.  Seems slow to me.
Mike Fair

Offline mwfair

We knew the boost-back burn uses 3 (from the video, and it's clear you need to turn back as fast as possible, and the stage is still heavy) and we knew the landing burn uses 1 (empty stage, hover-slam) but the secret-sauce burn is a little less secret now.  We know when it occurs, and we know it uses 3 engines.
A reddit user TheVehicleDestroyer ran the numbers more than a week ago. The best fit for the available data is a ballistic trajectory from separation. No boostback.
The video does look like 3 engines are burning, at least at time Separation+6:00.  However, we don't know how long it burned.  It also seems clear that it was not a 'boost-back', since range kept increasing. 
Perhaps they executed a short burn for the benefit of the WB-57?
What is the maximum altitude at which the atmosphere would support icing?
Mike Fair

Tags: