Author Topic: Testing upper stage propulsive entry survivability  (Read 5493 times)

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2556
  • Liked: 288
  • Likes Given: 266
Thinking about how SpaceX approached the issue of first stage recovery testing, is there any merit in doing the same with the second stage? That is, use margin on an operational launch, using existing hardware, to see how far through entry the vehicle can get.

Using the numbers of Ed Kyle's site, I get a delta V of about 11,000m/s for the second stage.
If there were no payload at all (hypothetical scenario), I estimate you would have around 5,300m/s to deorbit and slow the entry. That's quite a lot, but is it enough?

We know what combination of speed/altitude the first stage can survive, but I lack the knowledge to translate that information into the relevant conditions for the upper stage.

Secondly, does it seem likely that the second stage might be more robust than the first, from an entry standpoint?

So my main question is... does the forum think that there is anything to be learned from attempting partial propulsive entry testing? And are there any particularly small payloads manifested that would be potential opportunities (like CASIOPE was).
Waiting for joy and raptor

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 7256
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 4157
  • Likes Given: 2840
Re: Testing upper stage propulsive entry survivability
« Reply #1 on: 02/20/2017 04:28 PM »
I think the theory is the numbers miss by so much that there's no merit in trying to actually recover. But you may be on to something in that there may be things that can be learned by higher speed reentries even if the stage isn't recovered. But what? What does SpaceX need to know that it hasn't gotten from S1 reentry and the low density heat shield work NASA already did?

If we can spot something?
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Offline envy887

  • Full Member
  • ****
  • Posts: 1335
  • Liked: 553
  • Likes Given: 289
Re: Testing upper stage propulsive entry survivability
« Reply #2 on: 02/20/2017 05:30 PM »
It would be interesting to see if (or how long) the vac nozzle can survive hypersonic entry aero loads. Even after launching a very light LEO payload and propulsive deceleration it will still be travelling somewhere around Mach 10 at entry interface.

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2556
  • Liked: 288
  • Likes Given: 266
Re: Testing upper stage propulsive entry survivability
« Reply #3 on: 02/20/2017 05:46 PM »
Where I would see the value in this experiment is in determining the survival point of the stage. Start the entry burn just before you think the stage is about to be lost, and make a long, throttled back burn to see how long the stage can survive. There is still a lot to be learned in hypersonic retropropulsion.
Waiting for joy and raptor

Offline whitelancer64

Re: Testing upper stage propulsive entry survivability
« Reply #4 on: 02/20/2017 05:54 PM »
Recovering the second stage with retropropulsion isn't going to work, too much fuel is needed. Just stick a heat shield (perhaps expandable) on the nose.
"One bit of advice: it is important to view knowledge as sort of a semantic tree -- make sure you understand the fundamental principles, ie the trunk and big branches, before you get into the leaves/details or there is nothing for them to hang on to." - Elon Musk
"There are lies, damned lies, and launch schedules." - Larry J

Offline R7

  • Propulsophile
  • Senior Member
  • *****
  • Posts: 2738
    • Don't worry.. we can still be fans of OSC and SNC
  • Liked: 942
  • Likes Given: 663
Re: Testing upper stage propulsive entry survivability
« Reply #5 on: 02/20/2017 07:17 PM »
S2 with proper heat shield doesn't even have to do atmospheric entry burn S1 does.
AD·ASTRA·ASTRORVM·GRATIA

Offline S.Paulissen

  • Full Member
  • ***
  • Posts: 336
  • Boston
  • Liked: 257
  • Likes Given: 372
Re: Testing upper stage propulsive entry survivability
« Reply #6 on: 02/22/2017 04:10 AM »
S2 with proper heat shield doesn't even have to do atmospheric entry burn S1 does.

Is there evidence for this assertion?  How much would the shield mass?  Is that factored into sim?  What are peak breaking gees?  Just curious.
"An expert is a person who has found out by his own painful experience all the mistakes that one can make in a very narrow field." -Niels Bohr
Poster previously known as Exclavion going by his real name now.

Offline R7

  • Propulsophile
  • Senior Member
  • *****
  • Posts: 2738
    • Don't worry.. we can still be fans of OSC and SNC
  • Liked: 942
  • Likes Given: 663
Re: Testing upper stage propulsive entry survivability
« Reply #7 on: 02/22/2017 08:31 AM »
1. Is there evidence for this assertion? Is that factored into sim?
2. How much would the shield mass?
3. What are peak breaking gees?

1. Rational reasoning. Reusable S2 has to cope with heat and stresses while decelerating from orbital speed. Heat shield is given, more on this below, so whatever thermal environment the S1 couldn't cope without powered breaking was, S2 can do it. S2 is more stubby which adds to structural strength. And it has to be aerodynamically stable during reentry. Didn't early F9 S1 reentry attempts fail mainly because there were no means to control the attitude, no RCS no fins, so it didn't stabilize and kept tumbling until aerodynamic forces broke it?

2. There are calculations in other threads estimating Dragon heatshield weighing little over 200kg. Area is known and PICA-X density is known, it's very light stuff. Then there's of course some additional support structure but it's still a total of how many hundreds of kgs, not tons. Relying on that mass to decelerate from orbital speed to terminal velocity vs. hypersonic retropropulsion isn't much of a contest.

3. Hard to say for sure because need to know things like S2 ballistic coefficient, drag coefficient and whether it uses any lift or does purely ballistic reentry. BC would likely to be smaller than any returning SCs so far. IIRC low BC means greater deceleration in upper atmosphere but lower peak deceleration because velocity when entering the quickly thickening lower atmosphere is smaller. Read about this in some paper analysing ICBM RV trajectories, fun!  ;)

AD·ASTRA·ASTRORVM·GRATIA

Offline dorkmo

  • Full Member
  • ****
  • Posts: 579
  • Liked: 269
  • Likes Given: 694
Re: Testing upper stage propulsive entry survivability
« Reply #8 on: 02/23/2017 01:10 AM »
would the center of gravity be a problem?

Offline envy887

  • Full Member
  • ****
  • Posts: 1335
  • Liked: 553
  • Likes Given: 289
Re: Testing upper stage propulsive entry survivability
« Reply #9 on: 02/23/2017 03:27 AM »
S2 will have a ballistic coefficient similar to Apollo unless it does a lifting entry, in which case the heat shield needs to extend down one side but the BC can be 4 or 5 times less.

With chines, a full length shield, and an engine flap it could enter like ITS, with a very high alpha, lots of lift, and quite low ballistic coefficient.

Offline corneliussulla

  • Full Member
  • *
  • Posts: 126
  • Liked: 28
  • Likes Given: 10
Re: Testing upper stage propulsive entry survivability
« Reply #10 on: 02/23/2017 10:00 AM »
The ITS is effectively a 2nd stage re entry vehicle. A similar type vehicle could be built for launch on falcon I would think. In a world where ITS refuelling exists refuelling a notional falcon 2nd stage for deorbit would be a minor affair I would think. ITS tankers could be parked in orbit to enable a great number of falcon 2nd stages to return to earth for each ITS tanker launch. If ITS tanker will be available in next 6 years this would seem the most logical way forward rather than trying to make the existing 2nd stage survivable

Re: Testing upper stage propulsive entry survivability
« Reply #11 on: 02/23/2017 09:52 PM »
Seems like a good thread to re-post this old 2011 animation showing S2 re-entry, landing. But would it be able to land back at the cape? Would it have to orbit first?


Offline speedevil

  • Full Member
  • *
  • Posts: 193
  • Fife
  • Liked: 82
  • Likes Given: 70
Re: Testing upper stage propulsive entry survivability
« Reply #12 on: 02/24/2017 02:15 AM »
Seems like a good thread to re-post this old 2011 animation showing S2 re-entry, landing. But would it be able to land back at the cape? Would it have to orbit first?
Second engine cutoff is around 600 seconds into flight, at ~8km/s. Neglecting stuff, that's probably over 1/3*8*600 = 1500km or so downrange.
In order to get back to the takeoff site without orbiting, you're going to need to do something wacky - as you are at this point in orbit. Something like a quite substantial burn to intersect the atmosphere, use the atmosphere to kill most of the velocity, burn back up ballistically to the takeoff site, entry trim burns perhaps, and then entry and landing burns like F9 S1.

This is going to need a _lot_ more delta-v than waiting around a day, perhaps with a trim manoever to adjust phasing and going straight in for a landing without the intervening~1500km ballistic hop needing several kilometers/s worth of fuel.
Unless of course you add high performance wings.
(which would be insane)

Online Lars-J

  • Senior Member
  • *****
  • Posts: 2577
  • California
  • Liked: 2026
  • Likes Given: 1134
Re: Testing upper stage propulsive entry survivability
« Reply #13 on: 02/24/2017 05:41 PM »
Seems like a good thread to re-post this old 2011 animation showing S2 re-entry, landing. But would it be able to land back at the cape? Would it have to orbit first?

Short answer: A reusable 2nd stage could only make it back to the launch site after one orbit. (it might have to wait more orbits, perhaps as long as a day)

Offline mvpel

  • Full Member
  • ****
  • Posts: 1095
  • New Hampshire
  • Liked: 1263
  • Likes Given: 1595
Re: Testing upper stage propulsive entry survivability
« Reply #14 on: 02/24/2017 09:37 PM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
"Ugly programs are like ugly suspension bridges: they're much more liable to collapse than pretty ones, because the way humans (especially engineer-humans) perceive beauty is intimately related to our ability to process and understand complexity. A language that makes it hard to write elegant code makes it hard to write good code." - Eric S. Raymond

Offline Jcc

  • Full Member
  • ****
  • Posts: 703
  • Liked: 144
  • Likes Given: 53
Re: Testing upper stage propulsive entry survivability
« Reply #15 on: 02/26/2017 01:24 PM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.

They have clearly determined that there is no way to make it work for Falcon, but it is already an integral feature for ITS. It works (hopefully) with ITS by scaling to 10x more thrust in the first stage and over 20x more thrust in the 2nd stage/payload, on orbit refueling, all composite structure, low bulk density for the nearly empty returning vehicle, etc.  you could call these non-trivial changes.

Offline deruch

  • Full Member
  • ****
  • Posts: 1221
  • California
  • Liked: 978
  • Likes Given: 1029
Re: Testing upper stage propulsive entry survivability
« Reply #16 on: 03/08/2017 09:35 PM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together. 
« Last Edit: 03/08/2017 09:35 PM by deruch »
In theory, there is no difference between theory and practice.  But, in practice, there is.  --Jan van de Snepscheut

Offline S.Paulissen

  • Full Member
  • ***
  • Posts: 336
  • Boston
  • Liked: 257
  • Likes Given: 372
Re: Testing upper stage propulsive entry survivability
« Reply #17 on: 03/09/2017 04:06 AM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together.

Perhaps your reputation alone can attest to this, but I would like to see this claim substantiated with some evidence. 
"An expert is a person who has found out by his own painful experience all the mistakes that one can make in a very narrow field." -Niels Bohr
Poster previously known as Exclavion going by his real name now.

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 9447
  • Liked: 331
  • Likes Given: 416
Re: Testing upper stage propulsive entry survivability
« Reply #18 on: 03/09/2017 04:24 AM »
I vaguely remember a Gary Hudson concept where the upper stage would return engine first, with the engine firing during re-entry. Apparently, this would be sufficient for the stage to survive the re-entry heating.

Offline JamesH65

  • Full Member
  • ****
  • Posts: 443
  • Liked: 233
  • Likes Given: 6
Re: Testing upper stage propulsive entry survivability
« Reply #19 on: 03/09/2017 10:10 AM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together.

Perhaps your reputation alone can attest to this, but I would like to see this claim substantiated with some evidence.

Indeed - I've not seen any claims or data on this before.

Offline mvpel

  • Full Member
  • ****
  • Posts: 1095
  • New Hampshire
  • Liked: 1263
  • Likes Given: 1595
Re: Testing upper stage propulsive entry survivability
« Reply #20 on: 03/09/2017 12:18 PM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together.

Per the public statements of Elon Musk:

Quote from: Elon Musk via Motley Fool
As Elon Musk explains, "The boost stage [of a Falcon 9 rocket] is about 70% of the cost of the rocket ... it's sort of on the order of $30 to $35 million dollars."

It seems far-fetched to me that the Mvac would cost in excess of 9x more than the SL Merlin when those 9 Merlins plus legs, fins, and some longer tanks rolled out with the same equipment as the upper stage tanks is 70% of the $43 - $50 million cost of the rocket.
"Ugly programs are like ugly suspension bridges: they're much more liable to collapse than pretty ones, because the way humans (especially engineer-humans) perceive beauty is intimately related to our ability to process and understand complexity. A language that makes it hard to write elegant code makes it hard to write good code." - Eric S. Raymond

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2556
  • Liked: 288
  • Likes Given: 266
Re: Testing upper stage propulsive entry survivability
« Reply #21 on: 03/09/2017 08:51 PM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together. 

Niobium must be really expensive stuff.

Source?
Waiting for joy and raptor

Online HMXHMX

  • Full Member
  • ****
  • Posts: 1334
  • Liked: 789
  • Likes Given: 216
Re: Testing upper stage propulsive entry survivability
« Reply #22 on: 03/09/2017 10:42 PM »
I vaguely remember a Gary Hudson concept where the upper stage would return engine first, with the engine firing during re-entry. Apparently, this would be sufficient for the stage to survive the re-entry heating.


My approach was a bit more subtle than stated here; my idea dates from the early 1980s when I was looking at recovery of a lifting-ballistic VTOL, base-first, and proposed firing a small centrally-located thruster or gas generator to push off the main re-entry shock.  That dramatically lowers the conductive, radiative and convective heating of the base.  Applied to the F9 S2, it might be possible to operate the engine's GG only, perhaps with additional H2O injected to increase mass flow and to lower temperatures, and then to exhaust the flow out of the main bell.  But the large bell would almost certainly need to be jettisoned first.  This concept is informed by some of the wind tunnel work done by NASA Langley on supersonic retropropulsion, which can be seen on youtube videos.

Offline deruch

  • Full Member
  • ****
  • Posts: 1221
  • California
  • Liked: 978
  • Likes Given: 1029
Re: Testing upper stage propulsive entry survivability
« Reply #23 on: 03/11/2017 12:27 AM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together.

Perhaps your reputation alone can attest to this, but I would like to see this claim substantiated with some evidence.
Was based on a series of comments from a former SpaceX employee who worked on the MVac.  @gongora put them all together in a comment in the Merlin 1D thread (quoted pertinent bits below).  Most specifically that at the time he worked there, it took 1-2 days to produce an M1D vs. 18-21 days for an MVac. due to its complexity.  My understanding of the disparate costs wasn't based on materials cost but man-hours.

The Reddit comments section is here: https://www.reddit.com/r/spacex/comments/5h94xv/picture_of_a_mvac_engine_sitting_inside_its/

Some interesting notes from a former employee who used to work on the MVac engines.  He left in the Fall of 2015 so some things may be a little out of date.
Comment by Foximus05 on whether employees floated between different tasks:
Quote
Not when I was there. You might float if someone was behind and needed help, but there was a set tam that only did MVAC, only did M1D's, only did octaweb, etc. the M1D guys might move around from lowers to uppers, or chambers, but not Mvac, because it required so much more attention to detail.
Comment by Foximus05 on assembly times for the engines:
Quote
When I left it was a day or two for an M1D (dependant on parts) Vs 18-21 days for an MVAC. Mvac is a lot more complex, has more systems and has a bunch of made on assembly parts
Comment by Foximus05 on M1D vs. MVac
Quote
Very. MVAC contains more systems that M1D's have inside the octaweb, along with some control valves for the second stage. The chamber and a few other parts are the only similarities. Its in the same class, but its like comparing a Small Block Chevy V8 to a Ferrari engine.
Comment by Foximus05 on M1D vs. MVac:
Quote
MVAC was around ~400 pounds heavier than M1D, sans vacuum nozzle. But that was before they went to FT and I left.
In theory, there is no difference between theory and practice.  But, in practice, there is.  --Jan van de Snepscheut

Offline stcks

  • Member
  • Posts: 46
  • Liked: 39
  • Likes Given: 53
Re: Testing upper stage propulsive entry survivability
« Reply #24 on: 03/11/2017 12:54 AM »
Comment by Foximus05 on M1D vs. MVac
Quote
Very. MVAC contains more systems that M1D's have inside the octaweb, along with some control valves for the second stage. The chamber and a few other parts are the only similarities. Its in the same class, but its like comparing a Small Block Chevy V8 to a Ferrari engine.

Pardon the stupid question, but is this common for upper stage engines in general? Or is this just unique to MVac vs M1D? What are the reasons why MVac should be so much more complicated than M1D?

Offline mattstep

  • Member
  • Posts: 30
  • Iowa, USA
  • Liked: 13
  • Likes Given: 686
Re: Testing upper stage propulsive entry survivability
« Reply #25 on: 03/11/2017 01:23 AM »
I can certainly see why they've tabled this for the time being - they're already recovering 90% of the engines and 70% of the cost of the launch vehicle, and bringing down that last engine and its tankage is clearly a very tricky proposition.
1 MVac costs more than all 9 of the SL Merlins put together.

Perhaps your reputation alone can attest to this, but I would like to see this claim substantiated with some evidence.
Was based on a series of comments from a former SpaceX employee who worked on the MVac.  @gongora put them all together in a comment in the Merlin 1D thread (quoted pertinent bits below).  Most specifically that at the time he worked there, it took 1-2 days to produce an M1D vs. 18-21 days for an MVac. due to its complexity.  My understanding of the disparate costs wasn't based on materials cost but man-hours.

The Reddit comments section is here: https://www.reddit.com/r/spacex/comments/5h94xv/picture_of_a_mvac_engine_sitting_inside_its/

Some interesting notes from a former employee who used to work on the MVac engines.  He left in the Fall of 2015 so some things may be a little out of date.
Comment by Foximus05 on whether employees floated between different tasks:
Quote
Not when I was there. You might float if someone was behind and needed help, but there was a set tam that only did MVAC, only did M1D's, only did octaweb, etc. the M1D guys might move around from lowers to uppers, or chambers, but not Mvac, because it required so much more attention to detail.
Comment by Foximus05 on assembly times for the engines:
Quote
When I left it was a day or two for an M1D (dependant on parts) Vs 18-21 days for an MVAC. Mvac is a lot more complex, has more systems and has a bunch of made on assembly parts
Comment by Foximus05 on M1D vs. MVac
Quote
Very. MVAC contains more systems that M1D's have inside the octaweb, along with some control valves for the second stage. The chamber and a few other parts are the only similarities. Its in the same class, but its like comparing a Small Block Chevy V8 to a Ferrari engine.
Comment by Foximus05 on M1D vs. MVac:
Quote
MVAC was around ~400 pounds heavier than M1D, sans vacuum nozzle. But that was before they went to FT and I left.

I don't think we have enough information here to make a cost comparison. There is nothing in those quotes about numbers of people on each team to make a comparison on total labor hours to produce each engine.

Offline Adaptation

  • Full Member
  • *
  • Posts: 159
  • Liked: 39
  • Likes Given: 37
Re: Testing upper stage propulsive entry survivability
« Reply #26 on: 03/11/2017 08:24 AM »
"Really tempting to redesign upper stage for return too (Falcon Heavy has enough power), but prob best to stay focused on the Mars rocket" -Elon

https://twitter.com/elonmusk/status/755167487017291776

SpaceX is also experimenting with an upper stage raptor for F9/FH.

https://www.defense.gov/News/Contracts/Contract-View/Article/642983

There is no real rush on this, core block 5 will be more or less finalized but new upper stages will be needed continually for the foreseeable future.  A reusable version could be developed years down the line.   It would likely be raptor based and still be expendable for F9 the reusable version would probably only fly on FH.

The current Merlin sized raptor was mostly 3d printed so perhaps machining and manufacture time wont be such a big deal.  Or maybe if the Vulcan reuse system of just returning the engines with an inflatable heat shield works well SpaceX could copy that for the upper stage.

Online saliva_sweet

  • Full Member
  • ***
  • Posts: 376
  • Liked: 314
  • Likes Given: 910
Re: Testing upper stage propulsive entry survivability
« Reply #27 on: 03/11/2017 09:55 AM »
it took 1-2 days to produce an M1D vs. 18-21 days for an MVac. due to its complexity.  My understanding of the disparate costs wasn't based on materials cost but man-hours.

That's flawed because you assume the same number of people working on both. It would make sense for them to proprtion their workforce such that they can produce 9 M1Ds during the time it takes to make an MVac.

Offline Kaputnik

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 2556
  • Liked: 288
  • Likes Given: 266
Re: Testing upper stage propulsive entry survivability
« Reply #28 on: 03/11/2017 11:56 AM »
Of course an Mvac *must* function flawlessly every time, whereas there is some margin for failure on the other engines. Would enhanced QC play a part?

Instinctively, though, it seems likely that Mvac is simply produced in smaller numbers by a smaller team.
Waiting for joy and raptor

Tags: