SpaceX Falcon Heavy Demo - Discussion and Speculation

[deruch]

This mission also has to function as acceptable demo for DoD, which means that one-off mods to the upper stage will be discouraged if not vetoed outright.

Sure, but we went through this cycle several times with S1 and reusability hardware.  There is a certain level of modification that is acceptable, and it's still considered "the same stage".  Like add-ons.  If it flies with an extra parachute tucked away near the top, it probably won't matter.  If it flies with a heat shield in front - it's still a pretty small change as far as the up-trip is concerned.

But clearly fairing/no fairing is important enough to matter.  So I'm guessing it's up to SpaceX to get approval from DoD for the configuration they want to fly.

If, in order to attach the heat shield, SpaceX is altering the way that the payload adapter attaches to the rocket, then haven't they totally trashed their ability to do any coupled loads analysis for vehicles launching without such alterations.

[rakaydos]

This mission also has to function as acceptable demo for DoD, which means that one-off mods to the upper stage will be discouraged if not vetoed outright.

Sure, but we went through this cycle several times with S1 and reusability hardware.  There is a certain level of modification that is acceptable, and it's still considered "the same stage".  Like add-ons.  If it flies with an extra parachute tucked away near the top, it probably won't matter.  If it flies with a heat shield in front - it's still a pretty small change as far as the up-trip is concerned.

But clearly fairing/no fairing is important enough to matter.  So I'm guessing it's up to SpaceX to get approval from DoD for the configuration they want to fly.

If, in order to attach the heat shield, SpaceX is altering the way that the payload adapter attaches to the rocket, then haven't they totally trashed their ability to do any coupled loads analysis for vehicles launching without such alterations.

If they're going to do so to accomidate reuse, eventually, isnt it better to do so now, and make this the certified setup, than certify the old setup and recertify later?

I think the biggest change the second stage needs, however, is deployable, variable drag picaX airbrakes near the bottom of the second stage, so it can come in sideways instead of nose or tail first.

[rsdavis9]

As to S2 reuse...

Maybe from early on they had already developed the S2 with heat shield and PLF integration and its just sitting on some warehouse someplace. It never got used because of lack of margins...

[Ronsmytheiii]

ULA wants to eventually use a HIAD technology derivative for the Vilcan first stage recovery, and HIAD is a NASA developed technology so no reason SpaceX can't use it, so maybe this is his Hail Mary?

https://www.nasa.gov/directorates/spacetech/game_changing_development/HIAD/

[robert_d]

This mission also has to function as acceptable demo for DoD, which means that one-off mods to the upper stage will be discouraged if not vetoed outright.

Right. But DoD should be supportive of innovation too. So the PICA-X covered legs I proposed or the upside down Dragon that was also mentioned should be fine as long as they attach in a normal manner to the standard payload adapter within a standard payload fairing. 

Is it really even clear how much the demo flight is expected to contribute to the certification process?

[meekGee]

This mission also has to function as acceptable demo for DoD, which means that one-off mods to the upper stage will be discouraged if not vetoed outright.

Sure, but we went through this cycle several times with S1 and reusability hardware.  There is a certain level of modification that is acceptable, and it's still considered "the same stage".  Like add-ons.  If it flies with an extra parachute tucked away near the top, it probably won't matter.  If it flies with a heat shield in front - it's still a pretty small change as far as the up-trip is concerned.

But clearly fairing/no fairing is important enough to matter.  So I'm guessing it's up to SpaceX to get approval from DoD for the configuration they want to fly.

If, in order to attach the heat shield, SpaceX is altering the way that the payload adapter attaches to the rocket, then haven't they totally trashed their ability to do any coupled loads analysis for vehicles launching without such alterations.

Yup.

So this "if" is what the DoD has to look at.

It's not a given that they do, and the key factor is how well you can model the effects of the change.

[ChrisWilson68]

Inside the fairing the biggest off the shelf solid upper stage FH can carry and a minimal Dragon on top of it.

It's hard for me to imagine SpaceX buying a solid upper stage.  It's just not the way they work.  They like to develop their own hardware, and they like everything to be on a path to reusability.  Especially since this is just a bonus on top of a demo, I think anything they do will be with hardware they want to develop eventually anyway.

[dror]

Elon Musk‏Verified account @elonmusk  16m16 minutes ago

 Considering trying to bring upper stage back on Falcon Heavy demo flight for full reusability. Odds of success low, but maybe worth a shot.

'Bring it back for full reusability' implies that if their plan goes well, the second stage will be retrieved on this flight

[rakaydos]

Elon Musk‏Verified account @elonmusk  16m16 minutes ago

 Considering trying to bring upper stage back on Falcon Heavy demo flight for full reusability. Odds of success low, but maybe worth a shot.

'Bring it back for full reusability' implies that if their plan goes well, the second stage will be retrieved on

So the question becomes, "what is the fastest, least intrusive changes that could theoretically make S2 reusable?"
People have suggested mounting a dragon 2 upside down, but you dont need the whole dragon, just the heat shield, heat shield-equipped legs, and the Superdracos.

They already have a design that attaches stage 2 to dragon's heat shield for "normal" flight- how hard would it be to flip that upside down and build a docking adapter out of it?

If that's possible, then S2 gains a "landing module" that is a set of deployable Superdracos and tanks, a standard Dragon2 heat shield with dragonlegs, and a disposable payload adapter between the head shield and the contents of the fairing. Throw some (titanium prototype to avoid orbital velocity burnup) grid fins near the Mvac for stability and you have all the elements that could, theoretically, land a second stage.

It wont work the first time, of course. Or the second time, probably. But eventually they'll debug it into existance.

[IainMcClatchie]

Elon has said the first stage is 75% of the vehicle cost.  I've also heard 70% and 80% from SpaceX.  I've also heard the MVac costs twice what an M1D costs, and the fairing cost $6m.  When I put all these numbers together, and try to get some gross margin between costs and price, it seems like the fairing costs almost as much as the second stage, which doesn't sound right.  I think Elon said once that he was surprised at how much the fairing cost to make.

.	No Reuse	.	First stage	reuse	1st & fairing	reuse	1st, 2nd & fairing	reuse
M1D	$  2.5	.	$  0.5	.	$  0.5	.	$  0.5
First stage remainder	$  7.0	.	$  1.0	.	$  1.0	.	$  1.0
First stage	$29.5	69%	$  5.5	30%	$  5.5	41%	$  5.5	65%
Mvac	$  5.0	.	$  5.0	.	$  5.0	.	$  1.0
Second stage remainder	$  2.0	.	$  2.0	.	$  2.0	.	$  1.0
Second stage	$  7.0	16%	$  7.0	38%	$  7.0	52%	$  2.0	24%
Fairing	$  6.0	14%	$  6.0	32%	$  1.0	7%	$  1.0	12%
Total vehicle cost	$42.5	.	$18.5	.	$13.5	.	$  8.5
Ops cost	$10.0	.	$10.0	.	$10.0	.	$10.0
Total cost	$52.5	.	$28.5	.	$23.5	.	$18.5
Price	$62.0	.	$43.5	.	$40.0	.	$38.0
Gross margin	$  9.5	15%	$15.0	34%	$16.5	41%	$19.5	51%
Price reduction	.	.	30%	.	8%	.	5%
R&D per step	$1,000	.	$1,000	.	$200	.	$500
Launches to recover R&D	105	.	182	.	133	.	167

These numbers suggest it will take well over 100 launches to pay for first stage reuse, and maybe something similar to pay for second stage reuse (or more -- I had to make almost no price reduction for second stage reuse to make the numbers look reasonable).  Fairing reuse looks a bit more promising, mostly because I'm guessing the capital costs are lower.  While hundreds of launches for payoff sounds ridiculous, in the context of launching that CommX constellation it's just a couple of years.  I think that constellation business plan is going to absolutely drive the rest of the business.

They have said they want to land ITS without legs, onto a cradle that catches it.  This is going to take some substantial development, and SpaceX needs to pay for that development somehow.  If I were them, I'd want some immediate revenue to show for it.

Bottom line: maybe they develop cradle catch for Falcon 9 stage 2 reuse.  That way, they get an immediate revenue benefit from the technology, which improves cash flow until they can get the ITS system built.

Note: The accepted wisdom is that extra weight costs more payload capacity on the second stage than on the first.  I'm not sure that applies to a landing system -- legs should cost some fixed fraction of the stage's empty mass, and the first stage empty mass is around 6x that of the second, which is around the same ratio of payload cost.  It seems like a wash to me.

[CorvusCorax]

Replying to @BadAstronomer

We can def bring it back like Dragon. Just a question of how much weight we need to add.

How dows Dragon reenter:

1. Deorbit burn
2. Reentry with PicaX Heatshield
3. Splashdown with Parachutes

Problems:

1. Stage2 is not a capsule, so aerodynamic stability during reentry, supersonic and transonic flight is a potential problem. Solution:
- Add weight at the tip (aka Pica X Heatshield + Parachute package
- Add control fins to the tail of the stage for stabilization and attitude control (I think SpaceX already has a working solution for deployable +controlable fins laying around, but they'd have to go at the bottom end of the stage instead of the convenient (but non-existing) interstage area

2. Sufficient and suitable Parachutes to slow down stage for splashdown.
Empty mass of second stage + reentry package would match that of Dragon1 very closely, they could probably use Dragon 1 parachute package as is without any large modifications.


So a quick "done in 3 months" ghetto solution for this is:
1. somehow put a Dragon1 heatshield on the top of stage2 without messing too much with the way fairing and payload adapter attaches. The trick might be to use an adapter that has the same mounts as Dragon1 trunk on the heatshield side. Payload adapter would be jetissoned prior to reentry
2. Find a way to add deployable grid fins at the bottom of the stage. If not possible, strap on TPS coated "dumb-fins" for passive stabilization to the outer mold line
3. Add a parachute package. Parachutes would be at the top for weight, but lines would hold on to the bottom of the stage, that way it comes down heatshield first and splashes down very gently, with the delicate M1Vac getting max protection.

might just work.

[CorvusCorax]

I've also heard the MVac costs twice what an M1D costs,

.	No Reuse	.	First stage	reuse	1st & fairing	reuse	1st, 2nd & fairing	reuse
M1D	$  2.5	.	$  0.5	.	$  0.5	.	$  0.5
Mvac	$  5.0	.	$  5.0	.	$  5.0	.	$  1.0

Ehm, that can't be right... There's 9 M1D on the vehicle but only 1 M1DVac. Or are you suggesting the M1DVac is worth twice as much as 9 M1D together?

[Kaputnik]

Note: The accepted wisdom is that extra weight costs more payload capacity on the second stage than on the first.  I'm not sure that applies to a landing system -- legs should cost some fixed fraction of the stage's empty mass, and the first stage empty mass is around 6x that of the second, which is around the same ratio of payload cost.  It seems like a wash to me.

It costs more because it has to be accelerated to a higher speed. So the hardware doesn't weigh any more, but if you are carrying it all the way to orbit you need more energy to get it there.

[rsdavis9]

They were considering S2 reuse for awhile before they decided it wasn't worth it from a payload penalty analysis. With falcon heavy they have the margin. Maybe they still have designs and actual hardware from the before it was scrapped.

They may already have:
a integrated heat shield and PLF.(maybe expendable PLF)
landing legs for a S2

[oldAtlas_Eguy]

A Note on retro-propulsion for S2.

At 38% throttle (lowest level) the engine burns ~6.45mt of prop per minute.

So depending on burn time needed that determines the prop required for retro-propulsion.

A 7 minute burn using 45mt of prop would also represent a DV of 7.3km/s.
A 4 minute burn using 25mt of prop -> DV of 5.6km/s
A 3 minute burn using 19mt of prop -> DV of 4.9km/s
A 2 minute burn using 13mt of prop -> DV of 3.9km/s
A 1 minute burn using 6.5mt of prop -> DV of 2.5km/s
The S2 is loaded with 220mt of prop.

So it is possible but the payload is the US (up to 45mt of prop).

The questions are:
How long is the retro burn?
How much additional DV is attributed to the atmospheric drag?
Will any prop be reserved for a powered landing (simulated)?
What would be the weight of any legs and additional ablative heat shielding?

Many questions we have without any answers. Just our own speculations.

[RobLynn]

Propulsive re-entry is the only possible answer for near-term (late summer) re-use - a 5-6km/s deltaV re-entry burn still leaves >10 tonnes? LEO payload, and all elements of the (R)LV can be recovered for reuse dropping flight costs a lot (10-15million per use assuming 10 reflights?).  Propulsive landing will be tricky with a Merlin Vac (will nozzle survive startup of severely over-expanded Merlin Vac?), as will the longer legs needed.

Could a propulsively buoyed re-entry trajectory be used to at least partially reduce the re-entry delta V required?  Essentially modulate height using M-Vac to maintain tank temperatures at survivable levels.   Even saving a few hundred m/s would be useful.

A Raptor powered upper stage would bring the payload up to a level that would make propulsive braking before re-entry more commercially attractive.  I think Elon's comment about trying recovery the second stage on the first Heavy flight this year strongly suggests that this is a development path they will pursue.

[oldAtlas_Eguy]

Updated the model.

Still need the estimator detail for drag during retro burn and for atmospheric drag for the later non-retro burn portion.

[groundbound]

Re: S2 recovery. Since no one else has mentioned it let me ask a totally ignorant question.

How does F9 S2 burnout mass compare to a reasonable safe mass for mid-air helicopter recovery?

[meekGee]

Re: S2 recovery. Since no one else has mentioned it let me ask a totally ignorant question.

How does F9 S2 burnout mass compare to a reasonable safe mass for mid-air helicopter recovery?

Estimates are 4-5 tons, which is feasible.

(Remember the dragon drop tests?)

[IainMcClatchie]

I've also heard the MVac costs twice what an M1D costs,

.	No Reuse	.	First stage	reuse	1st & fairing	reuse	1st, 2nd & fairing	reuse
M1D	$  2.5	.	$  0.5	.	$  0.5	.	$  0.5
Mvac	$  5.0	.	$  5.0	.	$  5.0	.	$  1.0

Ehm, that can't be right... There's 9 M1D on the vehicle but only 1 M1DVac. Or are you suggesting the M1DVac is worth twice as much as 9 M1D together?

My swag is that one MVac costs twice what 1 M1D costs.  The M1D cost gets multiplied by 9 and added into stage 1 costs.