Stratolaunch: General Company and Development Updates and Discussions

[ThePhugoid]

...many inert systems do not give one crap what propellant choice you made...

This crucial point isn't entirely true. A hydrolox tank will be twice as large as a methalox tank (assuming fixed GLOW), meaning it will weigh more (pressure vessel mass is linear in volume), have much heavier engines (fixed GLOW means fixed thrust, but LH2 engines have much lower TWR), and require more TPS area, and LH2 needs extra insulation to prevent freezing air out on everything.

The heavier tanks and engines and TPS and insulation means the wings have to be larger to land at the same speed, and the landing gear have to be heavier to support the greater dry mass. The larger heavier empty vehicle needs larger control surfaces for aerodynamic flight, a larger RCS system for control in vacuum, and so on.

The increased tank and engine mass spirals out into other vehicle components, and your LH2 vehicle ends up 4 points worse in mass fraction, mostly eating the 5 points lower mass fraction you gained by switching for the higher ISP. This is readily apparent in upper stages, but those usually get away with very light low thrust engines since they don't start subsonic and deep in the atmosphere.

And on top of that you now have to deal with aerial fueling/topping of a really finicky deep cryogen instead of two soft cryogens. Hydrogen is a giant PITA, and it's hardly clear that it's the best choice for this application. It may win out in the trade space, but that's not at all obvious.

Alright, I'll give you that some of those systems will care about the larger volume of the tanks, but this is still missing the forest for the trees.  The point is that they have to exist, and the systems that supplement them then exist, and all of that piles together to eat away at a very precious structural ratio that you can't afford to give away.  You can't propose starting out with a system with that many mass sinks that still has to start out at ~95% PMF.  Again, I am not saying Hydrogen is the best choice, I'm saying it's the least bad choice for this particular application.

[Archibald]

And then I would argue, 87% is not enough of a gain compared to 95%, I mean not with LH2 - because LH2 tanks are such a giant PITA - in structure and draggy shape. They just exactly negate the PMF gain compared to storable / kerosene / methane.

Now if there was a "magic propellant" that could lower a SSTO PMF from 95% to 87% without the flaws of LH2 - that is, with reasonnable tanks - now that would make one hell of a difference, indeed.

When you think about it, 95% to 87% is only a 8% difference. Knee-jerk reaction "oh, 8%, that's not very much of a difference."  Bad logic ! Because there is a logarithm stuck in the rocket equation, the difference between 95% and 87% PMF turns a 9.5 km/s, orbital, kerosene-fueled SSTO into a miserable (barely 6 km/s  = suborbital ) unuseful boondoggle. I mean, the difference is huge. The rocket equation is quite counter-intuitive, misleading, and unforgivable. Only 2% drop in PMF can ruin the day.

Now see the A380 and 787 programs: between the drawing board and the first prototype, the weight budget creeped up by 5% or 10% or even more. Airliners can tolerate that, but SSTO would be toast...

First time I toyed with SSTO, Excel, delta-v and PMF, I was quite surprised to see the delta-v collapse so fast.
Build your SSTO a little too heavy, only a little below the planned PMF, and the delta-v goes down the drain very, very quickly. A 2% drop can ruin the day.

[ThePhugoid]

When you think about it, 95% to 87% is only a 8% difference. Knee-jerk reaction "oh, 8%, that's not very much of a difference."  Bad logic ! Because there is a logarithm stuck in the rocket equation, the difference between 95% and 87% PMF turns a 9.5 km/s, orbital, kerosene-fueled SSTO into a miserable (barely 6 km/s  = suborbital ) unuseful boondoggle. I mean, the difference is huge. The rocket equation is quite counter-intuitive, misleading, and unforgivable. Only 2% drop in PMF can ruin the day.

I lost which way you're arguing because we're in complete agreement here - 8% worth of PMF on day 1 is invaluable.  You need that to cram in the inert items whether they care about tank volume or not.  And at the end of the day, the draginess and volume of larger LH2 tanks do not substantially cripple an outer moldline that already needs a lot of wing area, lift, drag, and controllability that characterize a reusable system as this.

[Patchouli]

I remember reading the F9 first stage is barely capable of SSTO with sea level optimized nozzles so maybe hydrogen is not be needed.
An air launched vehicle with a more efficient engine such as the AR-1 or BE-4 and a high altitude nozzle might be able to carry a  useful payload as a single stage.

[niwax]

I remember reading the F9 first stage is barely capable of SSTO with sea level optimized nozzles so maybe hydrogen is not be needed.
An air launched vehicle with a more efficient engine such as the AR-1 or BE-4 and a high altitude nozzle might be able to carry a  useful payload as a single stage.

It pretty much follows the numbers laid out here. F9 first stage has a crazy high mass fraction somewhere around 95-97%.

[Elmar Moelzer]

If I take my BFS- Tanker based SSTO dry weight (55 tonnes) and shrink that down to max out the 230 tons GLOW, I get 11 tonnes dry weight for the space plane as a crude estimate.
Assuming a near vac Isp of the Raptor at 370, one could get a DV of about 9450 m/s with a 6 ton payload.
So theoretically, a methalox SSTO spaceplane could deliver a 6 ton payload to the ISS (or equivalent orbit), unless I made a mistake somewhere. Not a whole lot of wiggle room, but it could be plausible. Of course that would require them to develop a Raptor equivalent engine from scratch, which by itself is quite a challenge. On the other hand, they should have lots of experience with composites, which could help with dry weight.

[Elmar Moelzer]

Did another calculation with hydrolox. I am assuming and engine (or multiple engines) with enough thrust and about 450 sec Isp with a dry weight of about 6 tonnes (there are different engine combinations that could make this work). The tankage should be less than 8 tonnes. Wings about half of that at 4 tonnes. Add 3 tonnes for TPS landing gears and other things, we get about 21 tonnes dry weight. With 6 tonnes payload we get again about 9450 m/s DV.

Overall, I think that the hydrolox version has more margins for structures and more off the shelf engine options.

[jbenton]

Stupid question, but I'll ask anyways:

Can we better guess what propellant they intend to use just by the look of the boosters and spaceplane? I mean STS and Delta IV prefer to use orange foam (though white paint was an option early on in the program, only more weight) to keep the hydrolox cool. In renderings of the Advanced Boosters using the F-1B (for SLS) They had the part of the exterior protecting the Lox as foam and the part covering the RP-1 as white paint. (Admittedly, Centaur also uses white paint)

In any case, what do we think of the thermal properties of the exteriors suggested in the renderings? Maybe it could help us figure what they have in mind. Only thing that seems clear to me is that they're doing liquids and not solids.

[Markstark]

Stupid question, but I'll ask anyways:

Can we better guess what propellant they intend to use just by the look of the boosters and spaceplane? I mean STS and Delta IV prefer to use orange foam (though white paint was an option early on in the program, only more weight) to keep the hydrolox cool. In renderings of the Advanced Boosters using the F-1B (for SLS) They had the part of the exterior protecting the Lox as foam and the part covering the RP-1 as white paint. (Admittedly, Centaur also uses white paint)

In any case, what do we think of the thermal properties of the exteriors suggested in the renderings? Maybe it could help us figure what they have in mind. Only thing that seems clear to me is that they're doing liquids and not solids.

Another possibility is learning what test stand they plan on using at Stennis.

[Robotbeat]

<snip>
Just eyeballing the space plane it looks like it would need a booster stage or at least some additional fuel capacity, no?

In theory you can have wing tanks for kerosene during ascent.

It depends on how much of the airframe is tankage for propellants. Maybe X-15 style drop tanks if additional propellants is needed.

Doubt these are kerosene, given the scale of the vehicles plus the supposed goal of SSTO for the winged booster.  They are most likely hydrogen.

There are other options than kerosene and hydrogen, you might have heard of some upcoming rockets featuring methane. Methane does really seem to be the right choice (IMO) for a space plane.

Based on their payload mass targets, Hydrogen is the only option to get there.

Based on what exactly? All the Hydrogen SSTO's out there? Making a hydrogen SSTO is not appreciatively easier than making a kerosene or methane SSTO. All have their pros and cons.

No, based on physics.  The ability to succeed with SSTO means pulling out all the stops in every performance parameter you can within the design in both mass fraction as well as propulsion.  You have to do the crazy efficient propellant mass fraction no matter the propellant choice, but with hydrogen you can get an extra 30% in specific impulse over methane.  This fact, combined with their aggressive payload targets on both the cargo launchers as well as the SSTO spaceplane, leads me to assume hydrogen.

ie “how to tell someone doesn’t fully grok the rocket equation...”

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The thing that might change this is if they use oxygen-rich hydrolox for the beginning part of the trip then switch to stoich or hydrogen rich later on. (And airlaunch has mass limitations... But also volume and boiloff limitations!)

But generally speaking, hydrolox is significantly WORSE than methane/LOx or kerolox for SSTOs. I wish more people understood this instead of just slavishly and naively assuming the somewhat better Isp matters but mass fraction (which is way worse) doesn’t. Isp is not synonymous with performance!

[Patchouli]

ie “how to tell someone doesn’t fully grok the rocket equation...”

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The thing that might change this is if they use oxygen-rich hydrolox for the beginning part of the trip then switch to stoich or hydrogen rich later on. (And airlaunch has mass limitations... But also volume and boiloff limitations!)

But generally speaking, hydrolox is significantly WORSE than methane/LOx or kerolox for SSTOs. I wish more people understood this instead of just slavishly and naively assuming the somewhat better Isp matters but mass fraction (which is way worse) doesn’t. Isp is not synonymous with performance!

Gravity losses early on are a big issue along with tank size so the best solution might be tri propellant engine along the lines of the RD-701 as used on MAKS.
http://www.astronautix.com/r/rd-701.html
http://www.buran.ru/htm/molniya6.htm

A quick and dirty solution might be a J-2S or SSME with TAN.

[HMXHMX]

Just to throw out an observation, unattached to any judgement for or against, I wonder if the expendable Kraken might be an expendable SSTO.  I look at the image below and offer the following comments.

1) We've heard that Strato is developing their own LH2 engine (or at least I have, don't know if there is a prior reference in this thread) but no mention of a smaller upper-stage engine.

2) The image hints at – to me, faintly – a large LH2 tank and a smaller LO2 tank forward.

3) If one wanted to stage something, they could stage the wing after it helps them perform the gamma turn.

4) The 3 metric ton performance seems pretty low for the GLOW that the Roc can deploy but would reasonably match the payload performance of a one stage LH2 expendable.

[Hobbes-22]

I've put the Medium and Heavy images side by side, and added my guesses to what is what.
Surprisingly little commonality. Not even the same aft section on the first stage, different second stages, different first stage lengths.

 

[ThePhugoid]

ie “how to tell someone doesn’t fully grok the rocket equation...”

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The thing that might change this is if they use oxygen-rich hydrolox for the beginning part of the trip then switch to stoich or hydrogen rich later on. (And airlaunch has mass limitations... But also volume and boiloff limitations!)

But generally speaking, hydrolox is significantly WORSE than methane/LOx or kerolox for SSTOs. I wish more people understood this instead of just slavishly and naively assuming the somewhat better Isp matters but mass fraction (which is way worse) doesn’t. Isp is not synonymous with performance!

Le sigh.

Your argument holds up for a cargo-carrying tube launched from the ground without people on board with no plans for reentry, but that isn't what this is.  I don't know where this spectrum you quote where hydrocarbons win out all around could come from otherwise.  I also never said hydrolox is the only way to do SSTO, only that it's probably the choice that won out for this particular vehicle.

Go do a bottoms up mass allocation across the three choices for a horizontally hung spaceplane with wings, TPS, and with people on board. You'll find the answer isn't so black and white and these proven SSTO rules you quote go right out the window.

[ncb1397]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

[ThePhugoid]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

This. Your first paragraph, all on point. Academics are great for first order estimation but can easily get overruled or rewritten when a program takes off and reality sets in.

[envy887]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

On the other hand, Atlas having a higher delta-v despite having a much lower vacuum ISP kinda exactly proves Chris's point...

(BTW "proportional to" doesn't mean the factor of proportionality is equal to 1)

[ncb1397]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

On the other hand, Atlas having a higher delta-v despite having a much lower vacuum ISP kinda exactly proves Chris's point...

(BTW "proportional to" doesn't mean the factor of proportionality is equal to 1)

http://www.mathwords.com/i/inverse_variation.htm

As far as the Atlas booster having higher delta-v, it is:

1.)not a large amount
2.)offset by the Delta IV CBC having a 1.41 thrust to weight ratio while the Atlas V CCB booster is 1.28. Lower thrust to weight would mean higher gravity losses, eating into the ~100 m/s.
3.)If you made a 90% scale RS-68A like engine to match thrust to weight numbers between Atlas/Delta, you could expect about a 10% dry mass reduction on the engine or 660 kg. This would push the delta v performance of Delta IV CBC to 8830 m/s, 21 m/s lower than Atlas V CCB.
4.)If you look at sensitivity to payload, that 21 m/s gets smaller and smaller as you add payload. At 6000 kg, we are looking at a 8078 m/s vs 8085 m/s or Delta IV CBC losing by 7 m/s.

[ThePhugoid]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

On the other hand, Atlas having a higher delta-v despite having a much lower vacuum ISP kinda exactly proves Chris's point...

(BTW "proportional to" doesn't mean the factor of proportionality is equal to 1)

Yes, however this form of stage-only dV analysis doesn't take into account other factors, such as losses, the workload of the upper stages, and how much payload is eventually inserted.

[su27k]

Physics says (conventional) hydrolox is bad for SSTO. Physics says hydrogen is the lowest density liquid there is, and density is proportional to thrust and inversely proportional to dry mass.

The inverse proportion law is a good hypothesis but it doesn't seem to apply to actual rockets. For instance, if we take the Delta IV CBC with a dry mass of 26,000 kg and a propellant load of 200,400 kg, we could surmise the dry mass of a corresponding kerolox booster like the Atlas V CCB. With a propellant load of 284,089 and 2.87x the fuel density, the 7.7:1 fuel:dry mass of the Atlas V CCB should be 22.1:1 or a dry mass of 12,854 kg. Actual dry mass is 21,054 kg.  Something that seems to fit actual real life rockets of which there are myriad examples seems to suggest a more complicated relationship than a 1:1 relationship between volume and dry mass. For instance, keeping volume fixed, but varying mass of the load probably has structural implications.

edit: We should also look at single stage performance of the hydrolox CBC and the kerolox CCB using their vacuum isp numbers.

Delta IV CBC: 8738 m/s
Atlas V CCB: 8851 m/s

Surprisingly close.

If I'm not mistaken, LM didn't put a lot of effort into optimizing Atlas V CCB, it doesn't use common bulkhead for example.