Author Topic: What could push SpaceX to reconsider F9 second stage reusability?  (Read 17238 times)

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6234
  • Liked: 4057
  • Likes Given: 5570
Elon Musk has stated that F9 second stage reusability was not expected -- but with qualifiers that GTO/high energy orbits were too challenging.

Quote
I don't expect the Falcon 9 to have a reusable upper stage, just because the - with a kerosene-based system, the specific impulse isn't really high enough to do that, and a lot of the missions we do for commercial satellite deployment are geostationary missions. So, we're really going very far out. These are high delta-velocity missions, so to try to get something back from that is really difficult. But, with the next generation of vehicles...

http://shitelonsays.com/transcript/elon-musk-at-mits-aeroastro-centennial-part-1-of-6-2014-10-24

Lar began an interesting argument (really a discussion since Moderators don't argue) with this statement:
Quote
50 F9 or 15ish FH launches a year may be reason to revisit the decision not to do the engineering for second stage reuse

http://forum.nasaspaceflight.com/index.php?topic=36594.msg1319114#msg1319114

This is an interesting topic in and of itself, so here's a new thread to discuss it.

So the two questions are:
1) What would justify reconsideration of the expendible second F9 stage 'decision'?
2) What technology would be needed to enable cost-effective reuse?
 
These are F9/FH questions and not about the MCT.

My perspective:
Quote
A reusable second stage that returns the fairing and dispenser would make lots of sense when there are so many identical launches.  The fairing could open on one side as done on STS, or hinge back fully and then re-close.  Expendible fairings, dispensers, second stages launched week after week will be prohibitively expensive -- and the fuel to de-orbit all of this hardware (with the exception of the fairing as currently used) will need to be in the mass budget anyway.

Note: This same argument could be made for a reusable tanker second stage...

http://forum.nasaspaceflight.com/index.php?topic=36594.msg1319607#msg1319607


« Last Edit: 01/21/2015 08:19 PM by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1789
  • Bracknell, England
  • Liked: 549
  • Likes Given: 70
First requirement is that it is economically justified. Development cost < savings in launch costs. For a 3 year payback, 50 launches /year and $10M savings / launch that seems perfectly possible.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1789
  • Bracknell, England
  • Liked: 549
  • Likes Given: 70
Second requirement is for it to be technically feasible. It is known to be a hard problem, while there is no obvious reason that SpaceX cannot get it to work there may be all sorts of non-obvious things that make it uneconomic.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1789
  • Bracknell, England
  • Liked: 549
  • Likes Given: 70
Third requirement is there are not other better things that the SpaceX engineers can be working on. That is the opportunity cost.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1789
  • Bracknell, England
  • Liked: 549
  • Likes Given: 70
Fourth requirement is that SpaceX are not working on something else with equivalent functionality. E.g. a SFR.

Offline RocketmanUS

  • Senior Member
  • *****
  • Posts: 2226
  • USA
  • Liked: 71
  • Likes Given: 31
First get the F9 1st stage back and reuse
Second launch and get back ( land-landed ) and reuse Dragon ( with super Draco )

Learn from them.

Then make a reusable 2nd stage.

If it could have enough performance for Dragon cargo/crew/lab and or enough performance for some LEO payloads.
( More for a single payload )
( Just needs to lower cost enough to increase profit a little. )

And that would be a prototype for a more cost effective next generation two stage RLV.

Could also be used to test return for reuse from GTO on FH for next generation two stage RLV.

Could give them some data for their possible future MCT.

I don't see it being made just for the F9/FH. Their are new methods of manufacturing that are lowering production cost that the 2nd stage can be cheap enough to just make a new one.

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6234
  • Liked: 4057
  • Likes Given: 5570
First get the F9 1st stage back and reuse
Second launch and get back ( land-landed ) and reuse Dragon ( with super Draco )

Learn from them.

Then make a reusable 2nd stage.

If it could have enough performance for Dragon cargo/crew/lab and or enough performance for some LEO payloads.
( More for a single payload )
( Just needs to lower cost enough to increase profit a little. )

And that would be a prototype for a more cost effective next generation two stage RLV.

Could also be used to test return for reuse from GTO on FH for next generation two stage RLV.

Could give them some data for their possible future MCT.

I don't see it being made just for the F9/FH. Their are new methods of manufacturing that are lowering production cost that the 2nd stage can be cheap enough to just make a new one.

I agree that once you've developed second stage reusability, its application could be used elsewhere, even where it wasn't originally deemed useful like GTO.  The F9/FH constraint on OP was just to avoid another MCT thread.
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline cambrianera

  • Full Member
  • ****
  • Posts: 1426
  • Liked: 311
  • Likes Given: 250
A recovery kit with a mass of 400 kg or less.
Oh to be young again. . .

Offline nadreck


1) What would justify reconsideration of the expendible second F9 stage 'decision'?
2) What technology would be needed to enable cost-effective reuse?
 


1) a) (give us #2) a new development that (economically) requires significantly less TPS weight for re-entry
and
b) there have to be enough launches foreseen to justify the effort

2) a) (see 1a above) ok the lowest hanging fruit just might be a redesign of the shape of the 2nd stage (along with increasing its weight) and targeting the FH as the only cost effective lower stage
b) returning additional hardware beyond 2nd stage (dispenser, faring etc) can only fit with a tail first re-entry which to me suggests compromising the engine or going for a different engine (methalox plug nozzles around main TPS)
c) 2nd stage needs solar panels and more power storage plus greatly upgraded avionics/rcs and of course landing gear

Implications of 2a - if you think 1st stage core reuse gets down to a per use cost of $10M and you believe the goal of reuse of 2nd stage is to get a per use cost of $2.5M (assumptions: first stage cores with engines cost $50M and 2nd stages with all hardware for satellite launch and deployment cost $15M) so an F9 fully expendable is $65M per launch, an F9R1 (reusable 1st stage) is $25M, an F9HR1 is $45M, and an F9HR2 (2nd stage reusable as well) is $32.5M per flight.  I say based on this that there is NO case for a reusable F9 2nd stage if the only place it can be used is on the FH.  So eliminating a significantly heavier 2nd stage, we are left with, IMNSHO, a complete breakthrough in TPS being required and still having to give up 50% of the payload to add the other necessaries to bring the 2nd stage (Without faring and other payload hardware) back.  I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

I believe, given today's TPS technology that a returnable 2nd stage needs re-enter tail first and needs to be designed from the ground up (this would protect cargo bay style faring and it's contents).
It is all well and good to quote those things that made it past your confirmation bias that other people wrote, but this is a discussion board damnit! Let us know what you think! And why!

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6234
  • Liked: 4057
  • Likes Given: 5570
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Online oiorionsbelt

  • Full Member
  • ****
  • Posts: 1360
  • Liked: 607
  • Likes Given: 1365
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.
But his development team is busy with a reusable 2nd stage for BFR, so another team would be needed F9 FH or BFR would be delayed. He probably has the best team available, already working on a reusable 2nd stage for BFR. I'll take an earlier BFR over reusable 2nd stage for F9 FH any day.

Offline MikeAtkinson

  • Full Member
  • ****
  • Posts: 1789
  • Bracknell, England
  • Liked: 549
  • Likes Given: 70
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.

It depends on if they are mass limited or volume limited. If mass limited, then it might carry 2/3 the satellites at 2/3 the cost. So no net saving.

IMO they will be volume limited, so if they can launch the same number of satellites then there be a big win.

Offline nadreck

I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.

How if it costs more per launch (because you use an extra 2 cores up) to recover the 2nd stage.

Even if you could get the hypothetical reusable 2nd stage to work with a minimal payload on an F9 instead an FH you would still have to increase the number of 1st stage cores. Presuming 10 launches per recoverable 1st or 2nd stage and that any launch costs $12.5M in fixed costs, refurbishing/inspecting, etc (and I think that is truly optimistic but not impossible), then if a 1st stage costs $50M to build and a 2nd stage costs $15M then it would cost $25M to launch an F9R1 (first stage only reusable) and $12.5M to launch an F9R2(both stages recovered) so if your payload is less than half what it would be on an F9R1 then you are throwing money away even if it cost you nothing to develop F9R2.   

It is all well and good to quote those things that made it past your confirmation bias that other people wrote, but this is a discussion board damnit! Let us know what you think! And why!

Offline jzjzjzj

  • Member
  • Posts: 29
  • Liked: 14
  • Likes Given: 4
Third requirement is there are not other better things that the SpaceX engineers can be working on. That is the opportunity cost.

I'd say this is it. SpaceX will move as quick as they can to achieve the ultimate goal ("enabling people to live on other planets") given the human resources they have while making just enough money to cover the expenses.

Offline Burninate

  • Full Member
  • ****
  • Posts: 1134
  • Liked: 351
  • Likes Given: 73
Take one of three options for reentry: Supersonic Retropropulsion, Magnetoshell Aerocapture, or Hypersonic Inflatable Aerodynamic Decelerators.

Mix and match descent and landing options, which are much less difficult on Earth than on Mars.

If you double the weight of the 5T upper stage using this hardware, you're reducing the payload to LEO by 5T.

...

I don't see that it's very necessary though.  The upper stages only require 1/9th of the engine hardware and 1/4th of the tank hardware as the lower stage.  Mass production is already fairly competitive with reusability.  If F9E can operate as wastefully as it does right now and still make a profit, and the lower stages can be made reusable, then the upper stage reusability will only become worth considering when launch costs are a small fraction of what they are for F9E.
« Last Edit: 01/21/2015 11:06 PM by Burninate »

Offline schaban

  • Full Member
  • *
  • Posts: 118
  • Liked: 22
  • Likes Given: 22
First requirement is that it is economically justified. Development cost < savings in launch costs. For a 3 year payback, 50 launches /year and $10M savings / launch that seems perfectly possible.

But in 8 years they should have raptor based fully reusable rocket. Or be close to it.
So it might be profitable to develop reusable 2nd stage for Falcon but in the same time it does increases risk to their strategic product.

The only way I see them interested if they come up with minimal updates to existing 2 stage.
Something as simple as developing payload adapter to combine heat shield and parachute compartment. Would be used on selected missions, similarly to legs.

Offline Zed_Noir

  • Senior Member
  • *****
  • Posts: 2489
  • Canada
  • Liked: 374
  • Likes Given: 582
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.

If SpaceX uses the Moog EELV secondary payload adapter (ESPA) rings to mount Elonsats of less than 182 kg each. Then in theory you could stack 12 rings to lift 72 Elonsats per flight in the standard SpaceX PLF. Mass of the Elonsats and ESPA rings is about 15 tonnes.

So in theory a total of 56 flights will complete the Elonsat constellation at a tempo of about 10 flights annually, also the number of annual Elonsat replacement flights after constellation deployment.

So the number of flights annually to send up Elonsats will be too low to justified development of a reusable 2nd stage.

Presuming the eventual Elonsat will be similar in size and mass to the Orbcomm-2 sats.

Offline TrueBlueWitt

  • Space Nut
  • Senior Member
  • *****
  • Posts: 2005
  • Mars in my lifetime!
  • DeWitt, MI
  • Liked: 78
  • Likes Given: 72
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.

If SpaceX uses the Moog EELV secondary payload adapter (ESPA) rings to mount Elonsats of less than 182 kg each. Then in theory you could stack 12 rings to lift 72 Elonsats per flight in the standard SpaceX PLF. Mass of the Elonsats and ESPA rings is about 15 tonnes.

So in theory a total of 56 flights will complete the Elonsat constellation at a tempo of about 10 flights annually, also the number of annual Elonsat replacement flights after constellation deployment.

So the number of flights annually to send up Elonsats will be too low to justified development of a reusable 2nd stage.

Presuming the eventual Elonsat will be similar in size and mass to the Orbcomm-2 sats.

Math of this is eluding me..  72 Sats per launch.. 720 satellite constellation = How many total flights?

Offline Zed_Noir

  • Senior Member
  • *****
  • Posts: 2489
  • Canada
  • Liked: 374
  • Likes Given: 582
I believe I could make a good business case for working on bringing the costs of the 2nd stage, faring and satellite support hardware that would be lost, down over spending the same capital on F9 2nd stage reuse.

This is the crux of the issue... how low can these costs be driven. 

If SpaceX is flying 20 ElonSats per trip, this requires 200 second stages/fairing pairs/dispensers (100 if double the number can be launched each time) to get the constellation established without considering replacements.  ISTM that the cost of 200 M1D vacs alone would possibly fund the development effort.

If SpaceX uses the Moog EELV secondary payload adapter (ESPA) rings to mount Elonsats of less than 182 kg each. Then in theory you could stack 12 rings to lift 72 Elonsats per flight in the standard SpaceX PLF. Mass of the Elonsats and ESPA rings is about 15 tonnes.

So in theory a total of 56 flights will complete the Elonsat constellation at a tempo of about 10 flights annually, also the number of annual Elonsat replacement flights after constellation deployment.

So the number of flights annually to send up Elonsats will be too low to justified development of a reusable 2nd stage.

Presuming the eventual Elonsat will be similar in size and mass to the Orbcomm-2 sats.

Math of this is eluding me..  72 Sats per launch.. 720 satellite constellation = How many total flights?

Musk was talking about 4025 Sats for his constellation. 648 is the number for the OneWeb constellation.

See this thread
« Last Edit: 01/22/2015 02:14 AM by Zed_Noir »

Offline A_M_Swallow

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8479
  • South coast of England
  • Liked: 349
  • Likes Given: 148
Can the second stage engine be used as the main engine of the capsule?
If so they can be merged into a single spacecraft.

Tags: