Author Topic: F9 Second Stage Reusability  (Read 139604 times)

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #420 on: 04/19/2017 03:01 PM »
As a very basic first step toward second-stage reuse, SpaceX could simply equip one of the two standard payload adapters with spray-on or stick-on TPS, spin-stabilize the stage for re-entry, and let it come down in a ballistic trajectory. It would still impact the ocean at terminal velocity, but it would be a demonstration of re-entry.

Chutes could be added for a survivable splashdown...they would only run around 50 kg...but I don't know that there's any space for them to fit. Maybe around the engine, if there's room? There's not much room there.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8502
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2587
  • Likes Given: 6337
Re: F9 Second Stage Reusability
« Reply #421 on: 04/19/2017 03:24 PM »
As a very basic first step toward second-stage reuse, SpaceX could simply equip one of the two standard payload adapters with spray-on or stick-on TPS, spin-stabilize the stage for re-entry, and let it come down in a ballistic trajectory. It would still impact the ocean at terminal velocity, but it would be a demonstration of re-entry.

Chutes could be added for a survivable splashdown...they would only run around 50 kg...but I don't know that there's any space for them to fit. Maybe around the engine, if there's room? There's not much room there.
If you are going with this approach, why not allow the heat shield to drop down on retaining straps (similar to Mercury) allowing landing bags to inflate and deploy for a soft landing on land, rather than a splash down and salt water incursion...
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #422 on: 04/19/2017 03:37 PM »
As a very basic first step toward second-stage reuse, SpaceX could simply equip one of the two standard payload adapters with spray-on or stick-on TPS, spin-stabilize the stage for re-entry, and let it come down in a ballistic trajectory. It would still impact the ocean at terminal velocity, but it would be a demonstration of re-entry.

Chutes could be added for a survivable splashdown...they would only run around 50 kg...but I don't know that there's any space for them to fit. Maybe around the engine, if there's room? There's not much room there.
If you are going with this approach, why not allow the heat shield to drop down on retaining straps (similar to Mercury) allowing landing bags to inflate and deploy for a soft landing on land, rather than a splash down and salt water incursion...
Wind is a problem there. The second stage can come down on chutes, splash down, and tip over without being damaged because it is far shorter than the first stage, but if it came down on land it would likely cartwheel due to residual horizontal velocity and rip itself apart. Plus, there's the issue of the landing site; bringing down a spent stage on land requires a very large exclusion zone because there's really no way to control exactly where it comes down.

Offline adrianwyard

  • Full Member
  • ****
  • Posts: 960
  • Liked: 176
  • Likes Given: 227
Re: F9 Second Stage Reusability
« Reply #423 on: 04/19/2017 03:39 PM »
Parafoil + mid-air helicopter recovery still seems appealing to me - if the total reusable-S2 weight can be kept down. Set it down on a 'bouncy castle' or the far superior - and fun - ball pit in order to deal with any swinging or rotation that's induced as the speed lowers. (I'm only 90% joking.)
« Last Edit: 04/19/2017 04:34 PM by adrianwyard »

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8502
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2587
  • Likes Given: 6337
Re: F9 Second Stage Reusability
« Reply #424 on: 04/19/2017 03:57 PM »
As a very basic first step toward second-stage reuse, SpaceX could simply equip one of the two standard payload adapters with spray-on or stick-on TPS, spin-stabilize the stage for re-entry, and let it come down in a ballistic trajectory. It would still impact the ocean at terminal velocity, but it would be a demonstration of re-entry.

Chutes could be added for a survivable splashdown...they would only run around 50 kg...but I don't know that there's any space for them to fit. Maybe around the engine, if there's room? There's not much room there.
If you are going with this approach, why not allow the heat shield to drop down on retaining straps (similar to Mercury) allowing landing bags to inflate and deploy for a soft landing on land, rather than a splash down and salt water incursion...
Wind is a problem there. The second stage can come down on chutes, splash down, and tip over without being damaged because it is far shorter than the first stage, but if it came down on land it would likely cartwheel due to residual horizontal velocity and rip itself apart. Plus, there's the issue of the landing site; bringing down a spent stage on land requires a very large exclusion zone because there's really no way to control exactly where it comes down.
A deploy-able tripod gear may stop a tip-over if you release the the chutes or parafoil and a landing can be made at one of the many dry lakes sites out west... (quick sketch below)
« Last Edit: 04/19/2017 06:09 PM by Rocket Science »
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #425 on: 04/19/2017 05:14 PM »
A deploy-able tripod gear may stop a tip-over if you release the the chutes and a landing can be made at one of the many dry lakes sites out west... (quick sketch below)
If you have large enough landing legs and you cut the chutes at precisely the right moment, then you might be able to land it without tipover. If wind speed is low enough.

The dry lakes out west have pretty high winds, usually, and you still need a large exclusion zone in case re-entry ends up lower than expected.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 8502
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2587
  • Likes Given: 6337
Re: F9 Second Stage Reusability
« Reply #426 on: 04/19/2017 05:23 PM »
A deploy-able tripod gear may stop a tip-over if you release the the chutes and a landing can be made at one of the many dry lakes sites out west... (quick sketch below)
If you have large enough landing legs and you cut the chutes at precisely the right moment, then you might be able to land it without tipover. If wind speed is low enough.

The dry lakes out west have pretty high winds, usually, and you still need a large exclusion zone in case re-entry ends up lower than expected.
Agreed, that's when you choose to de-orbit when landing conditions are within limits. NASA landed "Stardust" out in Utah and was considering Orion to land at KSC with a parafoil...
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline adrianwyard

  • Full Member
  • ****
  • Posts: 960
  • Liked: 176
  • Likes Given: 227
Re: F9 Second Stage Reusability
« Reply #427 on: 04/19/2017 05:28 PM »
The stock S2 can spin up to 5 rpm. Not sure if that's enough, or if it can maintain that all the way through re-entry though:
http://www.spacex.com/sites/spacex/files/falcon_9_users_guide_rev_2.0.pdf Section 3.7

Here's some modest steps you could follow to investigate S2 return across four test missions:

+ On a Dragon/FH test mission*, add Pica-X to the nose of S2, spin it up and see how close to Earth it gets.
+ If it gets low enough, add grid fins to the next flight, confirm they can null the spin and fly to the pickup co-ordinates.
+ If that works, add a parafoil and see how it flies.
+ If all is well, attempt to capture the next one with a helicopter.
_______
* I say Dragon mission because the payload mount for a satellite needs to be dealt with. I guess you could build up the Pica-X shield and add a hinged trap door that closes over the payload mount. But that's beginning to sound too complicated if we're just proving the concept.
« Last Edit: 04/19/2017 05:32 PM by adrianwyard »

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #428 on: 04/19/2017 05:34 PM »
The stock S2 can spin up to 5 rpm. Not sure if that's enough, or if it can maintain that all the way through re-entry though:
I'm sure it could spin faster if it wanted to; that limit is for customers. The cold gas thrusters could spin up the stage well beyond its structural capacity if that was desired.

Quote
Here's some modest and steps you could follow to investigate S2 return across four test missions:

+ On a Dragon mission*, add Pica-X to the nose of S2, spin it up and see how close to Earth it gets.
+ If it gets low enough, add grid fins to the next flight, confirm they can null the spin and fly to the pickup co-ordinates.
+ If that works, add a parafoil and see how it flies.
+ If all is well, attempt to capture the next one with a helicopter.
_______
* I say Dragon mission because the payload mount for a satellite needs to be dealt with. I guess you could build up the Pica-X shield and add a hinged trap door that closes over the payload mount. But that's beginning to sound too complicated if we're just proving the concept.
Elon said he'd like to test it out on the Falcon Heavy inaugural launch, which will definitely feature a payload fairing and adapter rather than the Dragon. But the payload fairing is pretty tough; if you attached Pica-X panels to the surface and painted the adapter attachment points with ablative paint, it might be enough. Recall that peak heating is on the edge of a heat shield; the center gets less heat than any other point.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #429 on: 04/20/2017 03:41 PM »
If they wanted to go full-propulsive-landing on the F9 second stage, the only way to do it without significant modifications to the outside of the stage would be something like this:



Basically they would be designing a replacement payload adapter with all the recovery hardware on it, including expendable payload attachment adapter, heat shield, landing legs, bipropellant tanks, pressurant tanks, and SuperDracos. Note the high cosine losses on the SuperDracos.

Might be simpler to jettison the heat shield completely, though that goes against rapid reuse.
« Last Edit: 04/20/2017 03:43 PM by sevenperforce »

Offline adrianwyard

  • Full Member
  • ****
  • Posts: 960
  • Liked: 176
  • Likes Given: 227
Re: F9 Second Stage Reusability
« Reply #430 on: 04/20/2017 03:51 PM »
The stock S2 can spin up to 5 rpm. Not sure if that's enough, or if it can maintain that all the way through re-entry though:
I'm sure it could spin faster if it wanted to; that limit is for customers. The cold gas thrusters could spin up the stage well beyond its structural capacity if that was desired.

Quote
Here's some modest and steps you could follow to investigate S2 return across four test missions:

+ On a Dragon mission*, add Pica-X to the nose of S2, spin it up and see how close to Earth it gets.
+ If it gets low enough, add grid fins to the next flight, confirm they can null the spin and fly to the pickup co-ordinates.
+ If that works, add a parafoil and see how it flies.
+ If all is well, attempt to capture the next one with a helicopter.
_______
* I say Dragon mission because the payload mount for a satellite needs to be dealt with. I guess you could build up the Pica-X shield and add a hinged trap door that closes over the payload mount. But that's beginning to sound too complicated if we're just proving the concept.
Elon said he'd like to test it out on the Falcon Heavy inaugural launch, which will definitely feature a payload fairing and adapter rather than the Dragon. But the payload fairing is pretty tough; if you attached Pica-X panels to the surface and painted the adapter attachment points with ablative paint, it might be enough. Recall that peak heating is on the edge of a heat shield; the center gets less heat than any other point.
If the changes to S2 are just Pica-X around the payload mount hardware and spinning up for entry, I guess there's no harm in trying. My guess is it would eventually conduct too much heat to internal structure, but even if it failed it could be useful to know how long it survived.

If spin stabilization is indeed the answer to S2's CG/CP challenges, then we can imagine grid fins maintaining that spin (windmilling) once in the sensible atmosphere. They'd be deep in the wake, so wouldn't get too hot.

Offline adrianwyard

  • Full Member
  • ****
  • Posts: 960
  • Liked: 176
  • Likes Given: 227
Re: F9 Second Stage Reusability
« Reply #431 on: 04/20/2017 03:58 PM »
If they wanted to go full-propulsive-landing on the F9 second stage, the only way to do it without significant modifications to the outside of the stage would be something like this:



Basically they would be designing a replacement payload adapter with all the recovery hardware on it, including expendable payload attachment adapter, heat shield, landing legs, bipropellant tanks, pressurant tanks, and SuperDracos. Note the high cosine losses on the SuperDracos.

Might be simpler to jettison the heat shield completely, though that goes against rapid reuse.

The space-cadet in me would prefer to see this approach, but it's hard to see how in practical terms it's better than mid-air capture...

Online envy887

  • Senior Member
  • *****
  • Posts: 3570
  • Liked: 1758
  • Likes Given: 1124
Re: F9 Second Stage Reusability
« Reply #432 on: 04/20/2017 03:59 PM »
If they wanted to go full-propulsive-landing on the F9 second stage, the only way to do it without significant modifications to the outside of the stage would be something like this:



Basically they would be designing a replacement payload adapter with all the recovery hardware on it, including expendable payload attachment adapter, heat shield, landing legs, bipropellant tanks, pressurant tanks, and SuperDracos. Note the high cosine losses on the SuperDracos.

Might be simpler to jettison the heat shield completely, though that goes against rapid reuse.

Legs can go straight through the heatshield, see Dragon 2. So can the landing thrusters, they just need to blow off a tile when they start up.

Shouldn't need to add more than 1/2 a stage diameter to the height of the stage to fit everything in.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #433 on: 04/20/2017 04:40 PM »
If the changes to S2 are just Pica-X around the payload mount hardware and spinning up for entry, I guess there's no harm in trying. My guess is it would eventually conduct too much heat to internal structure, but even if it failed it could be useful to know how long it survived.
Seems like this would satisfy Elon's "we want to try second stage recovery" claim without incurring substantial development cost and without endangering FH launch qualification. The initial tests of first-stage return were never intended to be actually recoverable; they were intended to test supersonic retropropulsion and in-atmo retrograde engine restarts, primarily.

Quote
If spin stabilization is indeed the answer to S2's CG/CP challenges, then we can imagine grid fins maintaining that spin (windmilling) once in the sensible atmosphere. They'd be deep in the wake, so wouldn't get too hot.
Ooooh, now here's an idea: what about autorotation? With the larger grid fins planned for Block 5, supposedly capable of getting an L/D ratio of 1:1 on the first stage, autorotating the second stage could really dramatically reduce its terminal velocity...perhaps even to the point of being able to land on a "bouncy castle" without RUD. The grid fins would be able to guide it down to a pinpoint trajectory and then allow autorotation to reduce the terminal velocity to something survivable.

EDIT: back of the envelope...terminal velocity for S1 is slightly subsonic...S2 masses less than 25% of S1...estimating similar drag...grid fin L/D for S2 would be 4:1...

...I'm estimating an impact velocity with autorotation on the order of 20 m/s. If it comes down on the heat shield/payload adapter, that's not half bad. I wish we knew how much cushioning the bouncy castle would provide. Does anyone know if the cold gas thrusters can provide any meaningful prograde thrust?
« Last Edit: 04/20/2017 04:56 PM by sevenperforce »

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #434 on: 04/20/2017 05:00 PM »
Legs can go straight through the heatshield, see Dragon 2. So can the landing thrusters, they just need to blow off a tile when they start up.

Shouldn't need to add more than 1/2 a stage diameter to the height of the stage to fit everything in.
The existing S2 is more than three times the height of the Dragon 2, so those stubby little legs wouldn't provide nearly enough stability to avoid tip-over. And blowing a hole in the heat shield doesn't fit with reuse plans.

Offline Digitalchromakey

  • Member
  • Posts: 23
  • Phuket
  • Liked: 5
  • Likes Given: 2
Re: F9 Second Stage Reusability
« Reply #435 on: 04/20/2017 05:17 PM »
If they wanted to go full-propulsive-landing on the F9 second stage, the only way to do it without significant modifications to the outside of the stage would be something like this:



Basically they would be designing a replacement payload adapter with all the recovery hardware on it, including expendable payload attachment adapter, heat shield, landing legs, bipropellant tanks, pressurant tanks, and SuperDracos. Note the high cosine losses on the SuperDracos.

Might be simpler to jettison the heat shield completely, though that goes against rapid reuse.
Have been thinking about an approach similar to this, however perhaps everything necessary for a S2 landing could be contained within one self contained 'standalone' module (minus the PICA heat shield) fixed between a standard S2 and the payload adaptor, also with grid fins fitted to this same module, first using the Merlin 1D/grid fins  for a stable re-entry similar to S1, then only once the stage has significantly slowed (perhaps after another short Merlin 1D burn) inverting the S2 for a RTLS Super Draco powered landing.

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2821
  • Liked: 466
  • Likes Given: 430
Re: F9 Second Stage Reusability
« Reply #436 on: 04/20/2017 05:34 PM »
EDIT: back of the envelope...terminal velocity for S1 is slightly subsonic...S2 masses less than 25% of S1...estimating similar drag...grid fin L/D for S2 would be 4:1...

Erm, that's not how it works. You're working off mass and drag. But you need lift and drag.

An object with a L/D of 4:1 is essentially a crude glider. Think Shuttle orbiter, or jump chute. It doesn't matter how many grid fins you stick on the back of a rocket, it's not going to turn into a glider.
Waiting for joy and raptor

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #437 on: 04/20/2017 05:42 PM »
EDIT: back of the envelope...terminal velocity for S1 is slightly subsonic...S2 masses less than 25% of S1...estimating similar drag...grid fin L/D for S2 would be 4:1...

Erm, that's not how it works. You're working off mass and drag. But you need lift and drag.

An object with a L/D of 4:1 is essentially a crude glider. Think Shuttle orbiter, or jump chute. It doesn't matter how many grid fins you stick on the back of a rocket, it's not going to turn into a glider.
Autorotation works a little differently. I don't know whether the AOA of the grid fins can be adjusted in real-time as readily as the collective and pitch on a chopper, but it's worth investigation.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 873
  • Liked: 249
  • Likes Given: 267
Re: F9 Second Stage Reusability
« Reply #438 on: 04/20/2017 05:58 PM »
Another thought: the S2 is coming in from orbit, so there's no penalty for RTLS. If fitting landing legs behind the plasma shield is a problem, the S2 could simply come in and land upside-down on SuperDracos in a fixed cradle with a blast trench underneath. It would reduce plume impingement/wash dramatically and really simplify the recovery hardware needed for the S2.

Have been thinking about an approach similar to this, however perhaps everything necessary for a S2 landing could be contained within one self contained 'standalone' module (minus the PICA heat shield) fixed between a standard S2 and the payload adaptor, also with grid fins fitted to this same module, first using the Merlin 1D/grid fins  for a stable re-entry similar to S1, then only once the stage has significantly slowed (perhaps after another short Merlin 1D burn) inverting the S2 for a RTLS Super Draco powered landing.
Heat shield is a non-negotiable. Hypersonic retroprop won't work at orbital speeds.
« Last Edit: 04/20/2017 05:59 PM by sevenperforce »

Offline adrianwyard

  • Full Member
  • ****
  • Posts: 960
  • Liked: 176
  • Likes Given: 227
Re: F9 Second Stage Reusability
« Reply #439 on: 04/20/2017 06:10 PM »
... Block 5, supposedly capable of getting an L/D ratio of 1:1 on the first stage ...

Here's Musks's comment:
Quote
the new grid fins should be capable of taking a scorching and being fine. And they'll also have significantly more control authority, so, that should improve reusability of the rocket. It will improve the payload to orbit by being able to fly at a higher angle of attack and use the aerodynamic element of the rocket to effectively glide like a big cylinder. It does have an L/D of roughly 1 if flown at the right angle of attack, but you need control authority, particularly pitch control authority, that's higher than we currently have to achieve that.
It's hard to tease this apart, but I think he's talking about increasing payload to orbit by by 'flying' the first stage which has a L/D of 1 at certain (high) speeds. Not clear if the gain comes from going uphill (engine gimbaling), or RTLS landing (less prop needed for boostback because cross range 'flying' with grid fins is increased). TL;DR I don't think this helps S2 landing scenarios.
« Last Edit: 04/20/2017 06:11 PM by adrianwyard »

Tags: