Author Topic: F9 Second Stage Reusability  (Read 99465 times)

Offline meekGee

  • Senior Member
  • *****
  • Posts: 7894
  • N. California
  • Liked: 4141
  • Likes Given: 843
Re: F9 Second Stage Reusability
« Reply #160 on: 04/05/2017 05:56 AM »
This is obviously not a finished design of the reusable second stage because there's no landing legs.
That image appears to be from this 2008 paper, which puts it back in the parachute recovery era (no legs). Jon Goff had some posts on Selenian Boondocks about it back in the day:
http://selenianboondocks.com/2008/11/fun-spacex-paper-and-presentation/
http://selenianboondocks.com/2008/12/falcon-ix-upper-stage-recovery-kremlinology/

I don't know how applicable that image is to the current Falcon 9 upper stage, but it may be one of the simpler recovery options.

wow, time flies....   2008....

From the paper: "Consistent with SpaceX’s corporate philosophy of rapid and continuous improvement..." 

Easy words to say, but looking back..  From Falcon 1, Kestrel and Merlin 1A (ablative) - boy they ain't kidding.

"Cum continua emendationem et corporatum rebus philosophia est celeri consistent"
ABCD - Always Be Counting Down

Offline MP99

Re: F9 Second Stage Reusability
« Reply #161 on: 04/05/2017 06:38 AM »


The problem remains how to you ensure that a rear heavy cylinder stays nose down when every slightest disturbance wants to push it nose up, if not into an active end over end tumble.

"Drag" something behind it.

Reminder that the concepts we've seen for Magnetoshell AeroCapture have this trailing behind on a tether. The analogy with a parachute is very strong, and could provide a very stable way to keep the stage orientated nose first.

Because this generates drag magnetically, at altitudes where the stage experiences negligible drag, and therefore no heating, all heat is produced behind the stage and dissipated harmlessly.

The two big questions are whether this can provide enough drag to slow the stage so that the stage experiences low heating (like S1 entry burn), and how low into the atmosphere it can continue operating.

Another advantage is that the level of braking is adjustable, which allows to target a particular landing distance downrange in the face of differing atmospheric conditions.

Cheers, Martin

PS I also wonder if the same tether might provide a good mounting point for a parasail to land the stage on an airbag, like the fairings. This could compensate for lack of steerability by MAC.


Offline john smith 19

  • Senior Member
  • *****
  • Posts: 6406
  • Everyplaceelse
  • Liked: 870
  • Likes Given: 5510
Re: F9 Second Stage Reusability
« Reply #162 on: 04/05/2017 09:48 AM »
The problem remains how to you ensure that a rear heavy cylinder stays nose down when every slightest disturbance wants to push it nose up, if not into an active end over end tumble.

"Drag" something behind it.
That's what my petals do create drag...
Part of why a Shuttlecock is stable is because of its mass distribution. Let's see how this works for your idea.

Merlin is reported to be about 470Kg. The counter balance would presumably be a layer of PICAX on the nose. PICAX is much lighter than AVCOAT c0.27 g/cm^3  IE 2.7Kg/sq. m.
F9 is 5.2m in diameter so that's about 21 sq. metres, needing a mass of more than 22Kg/sq. m of PICAX to keep it nose heavy. That means a layer 8cm thick would be needed to be nose heavy.

If that was the whole story. But then we have the mass of the propellant. Since the stage is under deceleration it will be urged to the back bulkheads of the two tanks.

And you have put your "petals" on the back end, making it back end heavier still.
Does this still sound like a viable idea to you?

Keep in  mind the angles being talked about are no more than about 5deg below the local horizontal.

The trade offs in system design are tricky. It's not between drag/control surface weight and propellant. It's between drag/control surface weight and propellant and TPS mass. The "exchange rates" between the three are a PITA to calculate without detailed mass and thermophysical properties.

The easy one is every m/s you can shave off in the upper atmosphere means so many Kg of TPS you don't have to apply to resist such a high level of heating in the lower, denser atmosphere (provided your drag device can survive the heating). But control surfaces are measured in Kgs of mass, not m/s

But every Kg you add means another Kg of dead weight that builds up terminal velocity the Merlin will have to cancel.

So to be a net win your drag/control surface (and its control system) has to trade better than 1:1 in terms of mass in order to lower the propellant needed to cancel the terminal velocity of the stage at landing. Anything less and you might as well just leave the TPS layer in place (if that is adequate to do the job).
« Last Edit: 04/05/2017 09:49 AM by john smith 19 »
"Solids are a branch of fireworks, not rocketry. :-) :-) ", Henry Spencer 1/28/11  Averse to bold? You must be in marketing."It's all in the sequencing" K. Mattingly.  STS-Keeping most of the stakeholders happy most of the time.
So you're going to Mars to seek a better life. What does that mean to you? Always spot a fanbois by how they react to their idols failures.

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2806
  • Liked: 454
  • Likes Given: 397
Re: F9 Second Stage Reusability
« Reply #163 on: 04/05/2017 09:58 AM »
F9 is 3.6m diameter, not 5.2m (your number probaboy refers to the PLF, not the stage itself).
Waiting for joy and raptor

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 6406
  • Everyplaceelse
  • Liked: 870
  • Likes Given: 5510
Re: F9 Second Stage Reusability
« Reply #164 on: 04/05/2017 10:15 AM »
F9 is 3.6m diameter, not 5.2m (your number probaboy refers to the PLF, not the stage itself).
Oops. Yes I pulled the fairing number by mistake.

That gives about 10.75 sq. m. of nose, which means the PICAX has to cover it at about 43Kg/Sq. m. IE about 17cm thick to balance the 470Kg of a Merlin in the back.

One point I should have mentioned is that from previous work the maximum heating is normally on the lip of the cylinder rather than its face (and that ignores the details around the Payload Attach Fitting as well). This is quite a tricky location to mount TPS. 


"Solids are a branch of fireworks, not rocketry. :-) :-) ", Henry Spencer 1/28/11  Averse to bold? You must be in marketing."It's all in the sequencing" K. Mattingly.  STS-Keeping most of the stakeholders happy most of the time.
So you're going to Mars to seek a better life. What does that mean to you? Always spot a fanbois by how they react to their idols failures.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8176
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2330
  • Likes Given: 5481
Re: F9 Second Stage Reusability
« Reply #165 on: 04/05/2017 11:27 AM »
The problem remains how to you ensure that a rear heavy cylinder stays nose down when every slightest disturbance wants to push it nose up, if not into an active end over end tumble.

"Drag" something behind it.
That's what my petals do create drag...
Part of why a Shuttlecock is stable is because of its mass distribution. Let's see how this works for your idea.

Merlin is reported to be about 470Kg. The counter balance would presumably be a layer of PICAX on the nose. PICAX is much lighter than AVCOAT c0.27 g/cm^3  IE 2.7Kg/sq. m.
F9 is 5.2m in diameter so that's about 21 sq. metres, needing a mass of more than 22Kg/sq. m of PICAX to keep it nose heavy. That means a layer 8cm thick would be needed to be nose heavy.

If that was the whole story. But then we have the mass of the propellant. Since the stage is under deceleration it will be urged to the back bulkheads of the two tanks.

And you have put your "petals" on the back end, making it back end heavier still.
Does this still sound like a viable idea to you?

Keep in  mind the angles being talked about are no more than about 5deg below the local horizontal.

The trade offs in system design are tricky. It's not between drag/control surface weight and propellant. It's between drag/control surface weight and propellant and TPS mass. The "exchange rates" between the three are a PITA to calculate without detailed mass and thermophysical properties.

The easy one is every m/s you can shave off in the upper atmosphere means so many Kg of TPS you don't have to apply to resist such a high level of heating in the lower, denser atmosphere (provided your drag device can survive the heating). But control surfaces are measured in Kgs of mass, not m/s

But every Kg you add means another Kg of dead weight that builds up terminal velocity the Merlin will have to cancel.

So to be a net win your drag/control surface (and its control system) has to trade better than 1:1 in terms of mass in order to lower the propellant needed to cancel the terminal velocity of the stage at landing. Anything less and you might as well just leave the TPS layer in place (if that is adequate to do the job).
John, you keep "conveniently ignoring" that I stated that the hypergolic tanks, legs and their tanks and actuators are all in the nose... So you have "no definitive"  CoG to speak of... My grid-fin idea  on S1 worked out pretty good, wouldn't you say? Lets hear your great concept...
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8176
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2330
  • Likes Given: 5481
Re: F9 Second Stage Reusability
« Reply #166 on: 04/05/2017 11:45 AM »
And of course we have this from a few years back.  Be interesting to know the full history of the image.
https://forum.nasaspaceflight.com/index.php?topic=36440.msg1308198#msg1308198
Thanks for posting. These short-small petals may work for controllabilty, but "may not" protect the engine from entry plasma and "may need" to be made longer and around the entire stage with gap seals just like my shuttlecock concept...
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 760
  • Liked: 192
  • Likes Given: 225
Re: F9 Second Stage Reusability
« Reply #167 on: 04/05/2017 12:38 PM »
If that was the whole story. But then we have the mass of the propellant. Since the stage is under deceleration it will be urged to the back bulkheads of the two tanks.
Au contraire. The stage is under deceleration, so the propellant (if any remains) will be urged to the forward bulkheads of the two tanks.

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: F9 Second Stage Reusability
« Reply #168 on: 04/05/2017 12:46 PM »
And of course we have this from a few years back.  Be interesting to know the full history of the image.
https://forum.nasaspaceflight.com/index.php?topic=36440.msg1308198#msg1308198
Thanks for posting. These short-small petals may work for controllabilty, but "may not" protect the engine from entry plasma and "may need" to be made longer and around the entire stage with gap seals just like my shuttlecock concept...

That depends on the angle of attack, and the angle of plasma departure. Shuttle flew a very high AOA for high lift and drag. F9 US could generate some lift at a very low AOA (or just dive straight into the atmosphere on a ballistic return) and the engine would just sit in the wake. Sure, it gets pretty hot back there, but not hot enough to bother niobium. The other engine parts would need some TPS.

Offline envy887

  • Senior Member
  • *****
  • Posts: 2946
  • Liked: 1375
  • Likes Given: 915
Re: F9 Second Stage Reusability
« Reply #169 on: 04/05/2017 12:48 PM »
...
If that was the whole story. But then we have the mass of the propellant. Since the stage is under deceleration it will be urged to the back bulkheads of the two tanks.
...

Uhh.... do you get pinned back into your seat when you brake your car? That's not how inertia defies deceleration, at all.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8176
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2330
  • Likes Given: 5481
Re: F9 Second Stage Reusability
« Reply #170 on: 04/05/2017 01:00 PM »
And of course we have this from a few years back.  Be interesting to know the full history of the image.
https://forum.nasaspaceflight.com/index.php?topic=36440.msg1308198#msg1308198
Thanks for posting. These short-small petals may work for controllabilty, but "may not" protect the engine from entry plasma and "may need" to be made longer and around the entire stage with gap seals just like my shuttlecock concept...

That depends on the angle of attack, and the angle of plasma departure. Shuttle flew a very high AOA for high lift and drag. F9 US could generate some lift at a very low AOA (or just dive straight into the atmosphere on a ballistic return) and the engine would just sit in the wake. Sure, it gets pretty hot back there, but not hot enough to bother niobium. The other engine parts would need some TPS.
As I stated my petals are "controllable independently" so that the shape of the shuttlecock can be actively "morph" into any shape for lift, side to side, roll etc... during flight... The stage "will change into a lifting body during entry" then into a ballistic terminal flight to land nose down...... "The petals are "not" passive"...
« Last Edit: 04/05/2017 01:07 PM by Rocket Science »
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 6406
  • Everyplaceelse
  • Liked: 870
  • Likes Given: 5510
Re: F9 Second Stage Reusability
« Reply #171 on: 04/05/2017 05:43 PM »
John, you keep "conveniently ignoring" that I stated that the hypergolic tanks, legs and their tanks and actuators are all in the nose... So you have "no definitive"  CoG to speak of... My grid-fin idea  on S1 worked out pretty good, wouldn't you say?

Actually I just looked at the picture of the petals you posted and wondered how exactly were they going to support the whole loaded weight of a stage sitting on them?

But you're right putting a bunch of hypergolic tanks and landing legs in the nose will shift the Cg.

That will put the nozzles of the super dracos pointing into the airstream but I presume you have a plan to protect them till they get nearer the ground.

Quote from: Rocket Science
Lets hear your great concept...
Funny you should say that as I posted it a few days ago.

TL:DR Sub orbital velocity, near single orbit eliminates retro burn. Large nose grid fins and TPS. If the grid fins size and mass trades for reduced TPS but propellant for hoverslam also has to be factored in to ensure a net win.
"Solids are a branch of fireworks, not rocketry. :-) :-) ", Henry Spencer 1/28/11  Averse to bold? You must be in marketing."It's all in the sequencing" K. Mattingly.  STS-Keeping most of the stakeholders happy most of the time.
So you're going to Mars to seek a better life. What does that mean to you? Always spot a fanbois by how they react to their idols failures.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8176
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2330
  • Likes Given: 5481
Re: F9 Second Stage Reusability
« Reply #172 on: 04/05/2017 06:40 PM »
John, you keep "conveniently ignoring" that I stated that the hypergolic tanks, legs and their tanks and actuators are all in the nose... So you have "no definitive"  CoG to speak of... My grid-fin idea  on S1 worked out pretty good, wouldn't you say?

Actually I just looked at the picture of the petals you posted and wondered how exactly were they going to support the whole loaded weight of a stage sitting on them?

But you're right putting a bunch of hypergolic tanks and landing legs in the nose will shift the Cg.

That will put the nozzles of the super dracos pointing into the airstream but I presume you have a plan to protect them till they get nearer the ground.

Quote from: Rocket Science
Lets hear your great concept...
Funny you should say that as I posted it a few days ago.

TL:DR Sub orbital velocity, near single orbit eliminates retro burn. Large nose grid fins and TPS. If the grid fins size and mass trades for reduced TPS but propellant for hoverslam also has to be factored in to ensure a net win.
They don't support the stage, they all slide back in unison as "a ring of control surfaces" prior to entry...
Sub orbital means your payload now needs an assist-motor and now you take a hit for that. So you plan on using MVac to land, how? Give us a sketch...
« Last Edit: 04/05/2017 08:39 PM by Rocket Science »
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline punder

  • Full Member
  • ****
  • Posts: 443
  • Liked: 343
  • Likes Given: 207
Re: F9 Second Stage Reusability
« Reply #173 on: 04/05/2017 08:34 PM »
I posted this at ars technica, but thought I might also subject it to the flames here.   :D

Apologies if it's been done to death already.

-----

An expedient way to add reusability to the US would be to make a bolt-on "module" containing the complete recovery package--heat shield, reentry RCS, guidance, retracted landing legs, SuperDracos, and prop tanks. The module would mount on the stage exactly as the payload adapter or Dragon, so no major redesign of the stage would be necessary. It would place all of the recovery mass at the front of the stage. Reentry would be nose-first, with no "swoop of death" because the stage would land upside down. The module could be omitted if max performance was required, just as the boosters are still occasionally expended for big payloads.

The design would share one issue with other nose-first reentry designs: mounting the payload above (or through) the heat shield. The Shuttle had hatches in its heat shield for connections to the external tank, and a similar arrangement might work for Falcon.

Offline mme

  • Full Member
  • ****
  • Posts: 1094
  • Santa Barbara, CA, USA, Earth, Solar System, Milky Way Galaxy, Virgo Supercluster
  • Liked: 1320
  • Likes Given: 3455
Re: F9 Second Stage Reusability
« Reply #174 on: 04/05/2017 08:42 PM »
I posted this at ars technica, but thought I might also subject it to the flames here.   :D

Apologies if it's been done to death already.

-----

An expedient way to add reusability to the US would be to make a bolt-on "module" containing the complete recovery package--heat shield, reentry RCS, guidance, retracted landing legs, SuperDracos, and prop tanks. The module would mount on the stage exactly as the payload adapter or Dragon, so no major redesign of the stage would be necessary. It would place all of the recovery mass at the front of the stage. Reentry would be nose-first, with no "swoop of death" because the stage would land upside down. The module could be omitted if max performance was required, just as the boosters are still occasionally expended for big payloads.

The design would share one issue with other nose-first reentry designs: mounting the payload above (or through) the heat shield. The Shuttle had hatches in its heat shield for connections to the external tank, and a similar arrangement might work for Falcon.
That's what I was thinking here:
Because of the time constraints, I proposed that they could try a system on the Falcon Heavy Demo mission with 4 legs in the payload fairing. Splay them out and perform re-entry from LEO. Spreading the heat over the larger area would be interesting and might inform the development of a fairing/heat-shield combo to protect the second stage.
It might be something like a 3 part fairing with two parts recoverable and the 3rd part sliding into position downward to protect the stage.
How it would land? Deployable paraglider 'chute like the fairing onto a "bouncy castle" (air-bag like) barge.
 
That gives me an idea (sorry if it's been suggested.) Behold, my first NFS Rube Goldberg device.  It must be contagious:

Rocket LEGO Landing Element.  A landing module that connects to the PAF and is used as a PAF for the "real" payload.  It includes integrated SDs and landing legs and it's mass shifts the COM forward so it reenters "head first." No idea how to add a heat shield or control the trajectory w/o some sort of fins.

Similar idea could use a controllable parafoil instead of propulsive landing (or maybe with propulsive assist.)

Basically I like the idea of the PAF as a PAF so that there is only one S2 and you plug on the "reuse option."  Some assembly and lots of hand waving required.
But you said it better. :)
Space is not Highlander.  There can, and will, be more than one.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1697
  • Bracknell, England
  • Liked: 460
  • Likes Given: 55
Re: F9 Second Stage Reusability
« Reply #175 on: 04/05/2017 08:57 PM »
I posted this at ars technica, but thought I might also subject it to the flames here.   :D

Apologies if it's been done to death already.

-----

An expedient way to add reusability to the US would be to make a bolt-on "module" containing the complete recovery package--heat shield, reentry RCS, guidance, retracted landing legs, SuperDracos, and prop tanks. The module would mount on the stage exactly as the payload adapter or Dragon, so no major redesign of the stage would be necessary. It would place all of the recovery mass at the front of the stage. Reentry would be nose-first, with no "swoop of death" because the stage would land upside down. The module could be omitted if max performance was required, just as the boosters are still occasionally expended for big payloads.

The design would share one issue with other nose-first reentry designs: mounting the payload above (or through) the heat shield. The Shuttle had hatches in its heat shield for connections to the external tank, and a similar arrangement might work for Falcon.

A good solution. As you say mounting the payload through the heat shield is likely to be difficult. It shares that problem with other nose first designs, the payload adapter might have to be thrown away.

I'm not sure it is worth making the module bolt-on. Only reusable upper stages will be reused (by definition) so the module is only needed on reusable upper stages. It would be slightly more efficient to design the module integrated into the upper stage and the existing stage would need no changes, it could just go on being flown on expendable missions. Making it a bolt-on module would require minor mods to the upper stage, for attachment points, power and signalling, it would be better in my opinion not to have to make them for stages that are not going to be reused.

Although a good long term solution, I don't see how they could do all the engineering, manufacturing and testing of this "module" within 6 months.

Online AncientU

  • Senior Member
  • *****
  • Posts: 5192
  • Liked: 3109
  • Likes Given: 4429
Re: F9 Second Stage Reusability
« Reply #176 on: 04/05/2017 09:04 PM »
You are assuming starting now; they didn't have that constraint.
That's how.
« Last Edit: 04/05/2017 09:04 PM by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8176
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2330
  • Likes Given: 5481
Re: F9 Second Stage Reusability
« Reply #177 on: 04/05/2017 09:11 PM »
I posted this at ars technica, but thought I might also subject it to the flames here.   :D

Apologies if it's been done to death already.

-----

An expedient way to add reusability to the US would be to make a bolt-on "module" containing the complete recovery package--heat shield, reentry RCS, guidance, retracted landing legs, SuperDracos, and prop tanks. The module would mount on the stage exactly as the payload adapter or Dragon, so no major redesign of the stage would be necessary. It would place all of the recovery mass at the front of the stage. Reentry would be nose-first, with no "swoop of death" because the stage would land upside down. The module could be omitted if max performance was required, just as the boosters are still occasionally expended for big payloads.

The design would share one issue with other nose-first reentry designs: mounting the payload above (or through) the heat shield. The Shuttle had hatches in its heat shield for connections to the external tank, and a similar arrangement might work for Falcon.
Is it like the decent stage on my Dragon 2 lunar lander with a heat shield?
« Last Edit: 04/05/2017 09:11 PM by Rocket Science »
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline smfarmer11

  • Member
  • Posts: 72
  • Liked: 21
  • Likes Given: 52
Re: F9 Second Stage Reusability
« Reply #178 on: 04/05/2017 09:16 PM »
A kind of crazy idea would be to elongate the fairing and coat one shell in TPS, then instead of jettisoning it move it to cover the second stage after payload deployment. Re- enter using that as an ablator testing biconic re-entry and saving the half the fairing.

Offline punder

  • Full Member
  • ****
  • Posts: 443
  • Liked: 343
  • Likes Given: 207
Re: F9 Second Stage Reusability
« Reply #179 on: 04/05/2017 09:27 PM »
Is it like the decent stage on my Dragon 2 lunar lander with a heat shield?

Here is my quick-n-dirty cad model of the landed stage. So, yeah, sort of!

Tags: