Author Topic: F9 Second Stage Reusability  (Read 193162 times)

Offline meekGee

  • Senior Member
  • *****
  • Posts: 8282
  • N. California
  • Liked: 4494
  • Likes Given: 872
Re: F9 Second Stage Reusability
« Reply #260 on: 04/09/2017 01:35 AM »
Returning S2 will be 'similar to Dragon' in that it comes in from orbital velocity and will involve the use of Pica-X heat-shield material, but there are also many differences. The shape and mass characteristics are totally different.

Are you thinking S2 comes down nose first ~vertically? The stock S2 is tail (engine) heavy but I suppose a big parafoil stored in the nose will weigh something, perhaps enough to make it fly nose first. Would it be a lifting-entry a la Dragon for steering?

Or... the ITS fans would be happy if it had TPS on the side and it came in ITS/shuttle-style! But then you'd need a flap or something to protect the MVac - as recent posts have suggested.

Looks like guesstimate of S2 dry mass is <5mT before re-use hardware is added. At what mass does helicopter capture become too hairy, I wonder.

Edit to add:

A couple of years ago I threw out the possibility of landing something S2-like (ESA's IXV) at very high speed on a runway. FWIW: http://forum.nasaspaceflight.com/index.php?topic=36856.msg1335557#msg1335557

Well, if you were to put Dragon, S2, and S1 next to each other, which one is the odd man out in terms of proportion?  Of Mass?

S2 has a c.g problem, but it's not that acute, since it's only one engine in the back (~500kg?), and all the reusability hardware will go in the front. (Heat shield, parachute).

Perhaps a small drag device can help shift the center of drag backwards, and also allow for active steering so that parachute deployment occurs at a precise location.  (Or alternatively, the whole 9 yards, SuperDraco and legs, as some have proposed)
ABCD - Always Be Counting Down

Offline deruch

  • Senior Member
  • *****
  • Posts: 2250
  • California
  • Liked: 1753
  • Likes Given: 3779
Re: F9 Second Stage Reusability
« Reply #261 on: 04/09/2017 01:53 AM »
Elon's latest on twitter:
Quote
Fairing is ~$5M, but that should be reusable this year. Am fairly confident we can reuse upper stage too by late next year to get to 100%.

I'm floored.
1) Does this mean that the Block 5 design freeze slated for this year is NOT a design freeze?
Surprise!  Hey....wait....why is no one surprised?
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28177
  • Minnesota
  • Liked: 7977
  • Likes Given: 5323
Re: F9 Second Stage Reusability
« Reply #262 on: 04/09/2017 01:54 AM »
Elon's latest on twitter:
Quote
Fairing is ~$5M, but that should be reusable this year. Am fairly confident we can reuse upper stage too by late next year to get to 100%.

I'm floored.
1) Does this mean that the Block 5 design freeze slated for this year is NOT a design freeze?
Surprise!  Hey....wait....why is no one surprised?
Maybe it's part of Block 5, now.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline deruch

  • Senior Member
  • *****
  • Posts: 2250
  • California
  • Liked: 1753
  • Likes Given: 3779
Re: F9 Second Stage Reusability
« Reply #263 on: 04/09/2017 02:15 AM »
I'm floored.
1) Does this mean that the Block 5 design freeze slated for this year is NOT a design freeze?
Surprise!  Hey....wait....why is no one surprised?
Maybe it's part of Block 5, now.
If so, I would have much larger concerns about SpaceX meeting their schedule goals for Commercial Crew certification.  I could sort of see it as part of a Block 5.5 update that included planned changes/updates to the upper stage to address some DoD needs related to direct GEO insertion that they will eventually fold in after a while.  But, I can't see them including the mods I expect will be needed in time to achieve their stated flight rate goals for CC prior to their deadlines.
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline Space Ghost 1962

  • Senior Member
  • *****
  • Posts: 2783
  • Whatcha gonna do when the Ghost zaps you?
  • Liked: 2903
  • Likes Given: 2249
Re: F9 Second Stage Reusability
« Reply #264 on: 04/09/2017 02:25 AM »
Here's what you can do:
 * add autonomous cold gas thruster packs
 * add autonomous drag devices
 * means to link/coordinate them
 * payloads that attaches to the payload adapter. Possibly a TPS?
 * stuff that only becomes active following primary mission
 * surface application TPS of marginal weight/drag to primary mission

What you can't touch:
 * F9US systems including propulsion/RCS
 * Flight guidance/abort
 * Payload adapter
 * Separation
 * Autonomous operation (other than location/means/attitude/orientation of US disposal)
 * Any other use of propellants/systems/structural

To play this game, the means would be to "ride down" through EI while retaining stability/guidance, avoiding hot spots ("rotisserie?"), and maximizing drag as high as possible, to avoid the sudden "fist" of dynamic pressure waiting at lower altitudes.

The passive stability will be engine first, with high stagnation pressure on the MVac nozzle. If you keep it axially stable, all the thrust will ride through the same load paths that the engine fires through, and the niobium nozzle extension will heat and radiate like with engine operation. The shock wave will buffet well ahead of the nozzle, and the real questions are: a) can you keep it stable through to transonic, and b) can the nozzle survive the impulse of hitting the pressure transient to the sensible atmosphere.

After that point, various drag devices could slow the stage to ocean impact of the resultant "test article".

That's one way to keep to a realistic AF demo mission and a permittable attempt at US recovery.

This is the kind of thing that would be most likely to be attempted.

SX knows a lot about engines as they hit the sensible atmosphere nozzle first. And the loads/heating. And the props required to stabilize (perhaps from fairing recovery?). They could and probably have simulated this.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 8282
  • N. California
  • Liked: 4494
  • Likes Given: 872
Re: F9 Second Stage Reusability
« Reply #265 on: 04/09/2017 03:41 AM »
I'm floored.
1) Does this mean that the Block 5 design freeze slated for this year is NOT a design freeze?
Surprise!  Hey....wait....why is no one surprised?
Maybe it's part of Block 5, now.
If so, I would have much larger concerns about SpaceX meeting their schedule goals for Commercial Crew certification.  I could sort of see it as part of a Block 5.5 update that included planned changes/updates to the upper stage to address some DoD needs related to direct GEO insertion that they will eventually fold in after a while.  But, I can't see them including the mods I expect will be needed in time to achieve their stated flight rate goals for CC prior to their deadlines.

Almost everything in the "reusability kit" doesn't affect the up-mission.

The only thing that can be an issue is the heat shield, but who said they won't have that option in block 5?  (as an optional component).

If it's in the design when block 5 is "frozen", then it's good.

Then everything else is a per-mission add-on.  They can refrain from adding it on manned flights until it's proven itself enough times with unmanned flights.
ABCD - Always Be Counting Down

Offline ArbitraryConstant

  • Full Member
  • ****
  • Posts: 1761
  • Liked: 435
  • Likes Given: 245
Re: F9 Second Stage Reusability
« Reply #266 on: 04/09/2017 05:03 AM »
Here's what you can do:
 * add autonomous cold gas thruster packs
 * add autonomous drag devices
 * means to link/coordinate them
 * payloads that attaches to the payload adapter. Possibly a TPS?
 * stuff that only becomes active following primary mission
 * surface application TPS of marginal weight/drag to primary mission
How about using the real upper stage to carry a reentry demonstrator upper stage on top of the payload adapter and inside the fairing? Doesn't need a real Merlin or anything like that, just a needs to reproduce the weight distribution. Add TPS and hypergolic propulsion and so forth to the demonstrator. There's no way a Falcon Heavy couldn't handle launching that.
« Last Edit: 04/09/2017 05:05 AM by ArbitraryConstant »

Offline codevalley

  • Member
  • Posts: 1
  • australia
  • Liked: 0
  • Likes Given: 0
Re: F9 Second Stage Reusability
« Reply #267 on: 04/09/2017 06:46 AM »
Can I point out that a) the S2 CG problem, b) engine heat-shielding problem, c) in-atmosphere vacuum engine thrust problem, d) super draco weight penalty, and/or a mother of all hover-slam problems, are all alleviated by 3D printing an annular aerospike nozzle in place of the Merlin 1D vac bell nozzle.

S2's CG stays aft with S2 reentering engine-first, as a truncated aerospike provides natural heatshielding for the vehicle. Also, the vacuum/sea-level problem is now solved as aerospikes have good performance from vacuum all the way to sea-level, making for an efficient braking and landing burns. Super Dracos are thus not needed, as the aerospike replaces one large combustion chamber with essentially a dozen smaller chambers on its annulus, meaning we can throttle deeper by simply not lighting the full annulus. Deeper throttling removes the need for a 30g hover-slam. Finally, the CG need only move a little to provide some lift, permitting steering in the atmosphere not dissimilar to how ITS will do it, so no grid fins either.

S2 and aerospikes are a match made in engineering heaven.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1761
  • Bracknell, England
  • Liked: 530
  • Likes Given: 70
Re: F9 Second Stage Reusability
« Reply #268 on: 04/09/2017 07:16 AM »
How about using the real upper stage to carry a reentry demonstrator upper stage on top of the payload adapter and inside the fairing? Doesn't need a real Merlin or anything like that, just a needs to reproduce the weight distribution. Add TPS and hypergolic propulsion and so forth to the demonstrator. There's no way a Falcon Heavy couldn't handle launching that.

I was going to suggest the same. They probably have several non-functional Merlin sitting around. Fuel in the reentry demonstrator would be used to add mass, so as to demonstrate FH full payload - which could be vented before reentry.


Offline Lars-J

  • Senior Member
  • *****
  • Posts: 4045
  • California
  • Liked: 3338
  • Likes Given: 2103
Re: F9 Second Stage Reusability
« Reply #269 on: 04/09/2017 07:29 AM »
S2 and aerospikes are a match made in engineering heaven.

A good match perhaps, but not a heavenly one, because an aerospike nozzle will cause an isp/efficiency loss for the 2nd stage during ascent (where it is most needed).

Aerospike nozzles are never going to be as efficient in vacuum as vacuum optimized bell nozzles.

Online rakaydos

  • Full Member
  • ****
  • Posts: 411
  • Liked: 163
  • Likes Given: 3
Re: F9 Second Stage Reusability
« Reply #270 on: 04/09/2017 03:18 PM »
Wacky idea but... Could Stratalaunch perform air-recovery of a large parachuting payload like S2?

Offline Space Ghost 1962

  • Senior Member
  • *****
  • Posts: 2783
  • Whatcha gonna do when the Ghost zaps you?
  • Liked: 2903
  • Likes Given: 2249
Re: F9 Second Stage Reusability
« Reply #271 on: 04/09/2017 09:09 PM »
Here's what you can do:
 * add autonomous cold gas thruster packs
 * add autonomous drag devices
 * means to link/coordinate them
 * payloads that attaches to the payload adapter. Possibly a TPS?
 * stuff that only becomes active following primary mission
 * surface application TPS of marginal weight/drag to primary mission

How about using the real upper stage to carry a reentry demonstrator upper stage on top of the payload adapter and inside the fairing?

Yes - this would have three benefits,  a) no mods at all to the FH/F9US just to the unfueled (kerolox) payload, b) as many mods to payload "stage" as you wish, and c) a significant scale payload in weight/bulk to demo.

Quote
Doesn't need a real Merlin or anything like that, just a needs to reproduce the weight distribution. Add TPS and hypergolic propulsion and so forth to the demonstrator.
If necessary.

The point you are making would be to treat it as an depleted stage would arrive on orbit, but at launch as an ordinary payload for launch. BTW - likely inside the fairing upside down.

Quote
There's no way a Falcon Heavy couldn't handle launching that.

Absolutely. And easy to build from stores on hand. Also, easy to validate as a payload because you already have all of the particulars modelled, just not as a payload. And it wouldn't be impossible as a payload to be processed, because you are not expecting to use it as a stage at all, just a payload.

Online adrianwyard

  • Full Member
  • ****
  • Posts: 962
  • Liked: 177
  • Likes Given: 228
Re: F9 Second Stage Reusability
« Reply #272 on: 04/09/2017 10:04 PM »
Wacky idea but... Could Stratalaunch perform air-recovery of a large parachuting payload like S2?

No need to think that wacky when it comes the Falcon second stage: the Erickson S-64 Skycrane has an advertised max payload of 9mT. I doubt you'd want to catch 9 tonnes in mid-air, but it a Falcon second stage might be doable. (For arguments sake, call it ~5 mT.)

I recently learned that mid-air capture is simplified by fitting the target with a parafoil that's designed to leave it with a fairly high horizontal speed. Or course that's not helpful if you want to land, but for mid-air capture it means smaller gusts have relatively less effect on it's path due to its momentum. The same is true for the helicopter that's matching it's speed. Plus the helicopter downwash trails the target rather than interfering. Smoother, more predictable motion makes for easier capture.
« Last Edit: 04/09/2017 10:18 PM by adrianwyard »

Offline DAZ

  • Full Member
  • *
  • Posts: 131
  • Everett WA
  • Liked: 103
  • Likes Given: 1
Re: F9 Second Stage Reusability
« Reply #273 on: 04/10/2017 12:27 AM »
SpaceX is not known with starting with a Rube Goldberg solution. They may ultimately end up with a Rube Goldberg solution but that’s generally not where they start. Most of the solutions suggested here somewhat resemble a Rube Goldberg solution.

SpaceX usually starts with the minimum conceivable starting point and builds on a minimum solution for each failure point. So Elon Musk has indicated that they may try some kind of a Hail Mary recovery test. So the question really is what is the absolute minimum you would need to do this test?

1st off you need a reaction control system for pointing the retro burn and reentry. The US already has this as part of its existing functions. Next, it needs to be able to restart the engine for the deorbit burn. This would require a small amount of reserve fuel but otherwise, the US can already do this also. So the minimum experiment would be perform the deorbit burn and keep the engine bell into the wind, as it were.

The above would assume that SpaceX knows next to nothing about reentry for their 1st try. Most likely they know a little bit more than that and would decide to add a few other items to the US. For example, they might need some heat shielding around the engine and possibly some SPAM near the base of the US. They may also choose to add some additional reaction mass to the cold gas thruster system.

This is obviously not the final recovery solution but a starting point for experimentation. The question, in this case, would be what is a successful test? A successful test would seem to be that the stage survives intact until at least the high supersonic altitude. The next parameter may be that it is approximately where they predicted it to be, for example within 10 miles of their aim point. The most important parameter is can they gather telemetry to determine what the failure was if it isn’t intact.

After the 1st test, they can then evaluate and make changes as need be. For example, do they need more heat shielding, was there enough cold gas for control or do they need to add more, was the control and the deceleration adequate or do they need to add something like a HIAD. If they need to add a HIAD they could possibly add this around the payload adapter in the space reserved for SpaceX that the payload is not to encroach upon. When they get an intact stage that survives predictably into the high supersonic they can then add a drag shoots and a Parafoil for eventual recovery. No use wasting any of the above if they can’t get the stage down to high supersonic velocities. After that, they can decide how to ultimately recover the stage whether by aircraft snatching it out of the sky (with either a fixed wing cargo aircraft or a heavy lift helicopter) or possibly landing on some kind of a “bouncy castle”.

This is essentially the same way they approached S1 recovery. It requires the minimum of modifications with the minimum of additional costs. As with the S1, they are experimenting with something that is going to be thrown away anyway. This could ultimately lead to a very inexpensive way (not only monies but potential payload mass to orbit) to recover the US. Even if they can only recover 50% of the upper stages that they launch the savings for the upcoming CommX project could be enormous.

Offline Ludus

  • Full Member
  • ****
  • Posts: 1069
  • Liked: 458
  • Likes Given: 228
Re: F9 Second Stage Reusability
« Reply #274 on: 04/10/2017 01:42 AM »
Wacky idea but... Could Stratalaunch perform air-recovery of a large parachuting payload like S2?

No need to think that wacky when it comes the Falcon second stage: the Erickson S-64 Skycrane has an advertised max payload of 9mT. I doubt you'd want to catch 9 tonnes in mid-air, but it a Falcon second stage might be doable. (For arguments sake, call it ~5 mT.)

I recently learned that mid-air capture is simplified by fitting the target with a parafoil that's designed to leave it with a fairly high horizontal speed. Or course that's not helpful if you want to land, but for mid-air capture it means smaller gusts have relatively less effect on it's path due to its momentum. The same is true for the helicopter that's matching it's speed. Plus the helicopter downwash trails the target rather than interfering. Smoother, more predictable motion makes for easier capture.

That makes sense. Also, S2 is returning from orbit so there's more choice about returning it offshore from the launch site but not very far, so a helicopter like a S-64 has the range to go out, snag it mid air and carry it back to the launch site reprocessing building.

This is pretty similar to what ULA has in mind to develop for Vulcan engines by the mid 2020s. So Elon's take is they should be able to do that in a year or so. ;)

Offline CharlieWildman

  • Member
  • Posts: 52
  • A small island north of Seattle
  • Liked: 76
  • Likes Given: 33
Re: F9 Second Stage Reusability
« Reply #275 on: 04/10/2017 01:56 AM »
Back to the heat shield for a moment... How about just having the heat shield come up to the end of the thrust chamber and jettisoning the nozzle extension. 2nd stage would return engine first.

CG  (I assume) would much closer to the heat shield than it would be if it were on the nose of the 2nd stage.  Aerodynamic center of pressure (I assume) would be way behind the CG. Jettisoning the huge bell would hopefully reduce the likelihood of it's area causing the 2nd stage to swap ends at hypersonic speeds.

Curious what you guys think.
A small amount of knowledge can be a dangerous thing.

Online Robotbeat

  • Senior Member
  • *****
  • Posts: 28177
  • Minnesota
  • Liked: 7977
  • Likes Given: 5323
Re: F9 Second Stage Reusability
« Reply #276 on: 04/10/2017 02:45 AM »
Let's all review the only real info we've had on 2nd stage reuse:

Compare with:

(and the slides here: http://www.spacex.com/sites/spacex/files/mars_presentation.pdf )

...so what's the simplest way to reconcile these two? ITS EDL and 2nd Stage reentry/landing are both fairly similar to each other. PICA shielding (or related). legs from the bottom on the side where the engine bell is.
Differences:

1) The F9 2nd stage isn't shown changing its orientation from a nose-first to tail-first configuration.ITS is shown to transition from side-first to tail-first, and in the presentation and later info, Musk makes clear it uses tail flaps like Shuttle (except two of them so you can use them to steer).

2) 2nd Stage nose is blunt, since nominally it is streamlined behind a Dragon (or fairing, I suppose) whereas ITS is more rounded/pointy.

3) ITS isn't shown with a retractable nozzle like F9 2nd stage is.

4) 2nd Stage uses auxillary thrusters, maybe like Superdracos, to land, whereas ITS uses Raptors.

5) The legs are different.

I expect that IF SpaceX pursues VTVL for the upper stage (which you would think they would, since just about everything suggests that direction), they would use ITS' version for 1), 3), and 5) while using F9's version for 4) and possibly 2).

Otherwise, they skip VTVL and instead just do body flaps and PICA-X with a parachute for splashdown recovery.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline docmordrid

  • Senior Member
  • *****
  • Posts: 4819
  • Michigan
  • Liked: 1919
  • Likes Given: 1
Re: F9 Second Stage Reusability
« Reply #277 on: 04/10/2017 05:56 AM »
Or a bouncy castle bounce-down.
DM

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 7462
  • Everyplaceelse
  • Liked: 1119
  • Likes Given: 7502
Re: F9 Second Stage Reusability
« Reply #278 on: 04/10/2017 06:57 AM »
How about using the real upper stage to carry a reentry demonstrator upper stage on top of the payload adapter and inside the fairing? Doesn't need a real Merlin or anything like that, just a needs to reproduce the weight distribution. Add TPS and hypergolic propulsion and so forth to the demonstrator. There's no way a Falcon Heavy couldn't handle launching that.
That's very attractive, and fits into the time frame given the they probably have at least one on hand.

One constraint no one has mentioned is the stress levels during flight.

All stages have a maximum survivable g level in both +ve and -ve values.  I  think S1 was rated at -5g but I've no idea what S2 is.

In principle the ideal deceleration is full engine thrust for the minimal period but the stage is so light that's impossible (I think that period could be less than 1 second). So it's all down to how deep you can throttle.

One other idea that's not been suggested would be pulsing the main engine with multiple re-starts.
I don't think pulsing the main engine on a stage has ever been tried anywhere, but I may be wrong.

The key issue is the stage is designed to certain g levels and. It's not that exceeding them will probably wreck the stage for reuse (it probably will) but SX have a lot of data and done a lot of analysis on the present structure. Exceeding those current limits will mean a detailed re-analysis of the stage to confirm it would survive. It probably wouldn't so then they'd be into a re-design to figure out what parts need strengthening.
BFS. The worlds first Methane fueled FFORSC engined CFRP structured A380 sized aerospaceplane tail sitter capable of flying in Earth and Mars atmospheres. BFR. The worlds biggest Methane fueled FFORSC engined CFRP structured booster for BFS. First flight to Mars by end of 2022. Forward looking statements. T&C apply. Believe no one. Run your own numbers. So, you are going to Mars to start a better life? Picture it in your mind. Now say what it is out loud.

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 3660
  • Sunnyvale, CA
  • Liked: 2246
  • Likes Given: 2866
Re: F9 Second Stage Reusability
« Reply #279 on: 04/10/2017 10:58 AM »
What makes this so fascinating is Musk saying they may try to bring the FH demo S2 back this summer! No-one expected that. Presumably that rules out anything but the simplest modifications. And definitely not a Raptor.

...

This line of reasoning is flawed; you are assuming that EM didn't decide to bring back the second stage until that event 1-2 weeks ago.  Us not knowing about his plan (and it is likely much more mature than he is letting on) doesn't have any bearing on him having a plan. 

His time constraints as you've outlined them don't exist... FH has been on drawing boards without a paying customer for years.  IIRC, five years ago, he asked -- if you could launch anything into space, what would it be -- he was referring to the FH demo launch.  The 'silly' payload and the hardware to return the second stage was probably ready last year when AMOS disrupted all plans.

I agree with all that.

It's not like Musk was telling us each week "This week we continued to do no work on upper stage recovery" until last week.

All we had was one data point that said that at a particular time some time ago their current thinking was that they were de-emphasizing upper-stage reusability.  We really don't know whether that meant nobody working on it or a significant team working on it but considering it risky so not the most likely plan.  And we don't know if a week after they spoke they changed their minds and decided to go full speed ahead on upper stage reusability again.

They also might have purposely been sandbagging to avoid bad publicity.  Imaging they never really stopped working on upper stage reusability at all, but they wanted the press off their back, and no stories about how they had promised upper stage reusability but failed to deliver.  They might just say they weren't planning for it, which is technically correct because they have a plan to succeed even without upper stage reusability, but, really, it was mainly just to lower expectations.

Tags: