Author Topic: Configurations for reusable PSTO shuttle  (Read 3516 times)

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Configurations for reusable PSTO shuttle
« Reply #20 on: 03/17/2017 05:16 PM »
Speaking of parallel stages...does anyone remember this bad boy?  :D
A beauty to be sure. Of course, nowadays we don't really need to have the booster glide in for a landing when we can simply have it land on its tail. The only trouble, then, is getting the right thrust balancing with crossfeed and a safely recoverable crew-carrying shuttle.

Forget the booster. The ITS ship is similar in size, shape, mass, and thrust to the S-1C. Replace the fixed ring of 6 RVacs with a detachable ring of 10 or 12 SL Raptors and you have a modern S-1D. Plus it already has reentry and landing designed in.

The toroidal tank is a good idea, but it only needs to be big enough to fly the ring back... Maybe 50t of methalox.
It's an interesting idea that could stand to be explored, for sure.

To start with there, are a few different ways that the engines could be plumbed, all with their own advantages and disadvantages. The original S-1D proposal lacks any tankage on the skirt, plumbing solely from the main tank, but this won't allow a propulsive landing. It would still be a possibility if we added something like superdracos for the landing, which could do double duty in a launch assist role, but that might be more trouble than it's worth.

Adding just enough tankage to the skirt for a propulsive landing (as you suggest) would require crossfeed, as the engines would be plumbed from the core tank during ascent and then plumbed from the auxiliary tankage during landing. The complexity of crossfeed can be avoided if the auxiliary tankage is made slightly larger and plumbed to two designated landing engines for both ascent and landing, with the remaining skirt engines plumbed solely from the core tank for ascent, without crossfeed.

The next option is by far the simplest, as far as plumbing is concerned: the tank in the skirt carries enough fuel for the skirt engines during both ascent and landing. This means there are no fuel lines crossing between stages, which is obviously advantageous for system simplicity and reuse.

The final option is to crossfeed the core engine from an even larger skirt tank, so that the core tank is still full at separation. This results in earlier staging, a smaller terminal stage, and the maximum performance optimization, but it does involve a burgeoning first stage and added plumbing complexity. Another advantage here is that the core stage will have a T/W ratio greater than 1 at launch, giving it 0/0 LES functionality.

Possible configurations:



There are other considerations as well, like control. The Saturn V had a fixed core engine and gimbals on the outer four engines, but this would need gimbal on the core engine. The skirt engines could be either partially or completely fixed, depending on which ones would be used for landing.

Another issue is altitude compensation on the core engine. One option is to use a SL Raptor and simply take the performance hit; after all, even the SL Raptor has a specific impulse better than the MVac. Another option would be to use a lengthened engine bell which can be used at sea level, like the SSMEs. The final option would be to add a lower-thrust vacuum-optimized OMS cluster to assist during orbital insertion and circularization.

Finally, this overall staging configuration lends itself really ridiculously well to air augmentation, since there is already a metal fairing wrapped around the core engine. A simple aluminum shroud boosts static thrust by 15% and goes up to a 50% boost around Mach 1. If we did go with the initial option (skirt engines plumbed directly from the core tank and auxiliary liquid tanks+thrusters for the landing) then these could be used as part of the launch assist to get up to a reasonable fraction of Mach 1.

Curious to see what overall configuration results in the lowest overall launch vehicle size but can still take a crew of 7 to the ISS and bring them back. I'm thinking it could be significantly smaller than the S-1D; methalox has a more oxygen-rich mixture ratio than kerolox, and the Raptor takes densified propellants, so the higher specific impulse is really going to shine.

Online envy887

  • Senior Member
  • *****
  • Posts: 2233
  • Liked: 969
  • Likes Given: 644
Re: Configurations for reusable PSTO shuttle
« Reply #21 on: 03/18/2017 01:33 AM »
Curious to see what overall configuration results in the lowest overall launch vehicle size but can still take a crew of 7 to the ISS and bring them back. I'm thinking it could be significantly smaller than the S-1D; methalox has a more oxygen-rich mixture ratio than kerolox, and the Raptor takes densified propellants, so the higher specific impulse is really going to shine.

Hmmm.... yeah. S-1D is monstrously over-sized for that. Raptor is too large for a upper stage sized solely for taking 7 to the ISS, or for landing anything remotely that size, it can't throttle low enough. Even Merlin only throttles down to 32 tonnes-force. So an auxiliary landing thruster setup is pretty much required. The upper stage could maybe use a production version of the 1,000 kN demo Raptor engine.

But because the upper stage always ends up heavier, the 1.5 STO does worse in my calculations than a TSTO. It always needs 20 to 40% greater GTOW mass to deliver the same payload and return. The larger tank drag area doesn't reduce the TPS requirements enough to offset the extra fuel and thrust mass needed to put that tank it in orbit.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Configurations for reusable PSTO shuttle
« Reply #22 on: 03/20/2017 04:11 PM »
S-1D is monstrously over-sized for that. Raptor is too large for a upper stage sized solely for taking 7 to the ISS, or for landing anything remotely that size, it can't throttle low enough. Even Merlin only throttles down to 32 tonnes-force. So an auxiliary landing thruster setup is pretty much required. The upper stage could maybe use a production version of the 1,000 kN demo Raptor engine.
Raptor has half the thrust of the F-1 to begin with, so depending on whether it used a 4+1 configuration as with the S-1D or a 2+1 configuration like the Atlas D, we would be looking at launch mass between 30% and 50% of an S-1D. Raptor will be able to downthrottle to 20% of max thrust.

Quote
Because the upper stage always ends up heavier, the 1.5 STO does worse in my calculations than a TSTO. It always needs 20 to 40% greater GTOW mass to deliver the same payload and return. The larger tank drag area doesn't reduce the TPS requirements enough to offset the extra fuel and thrust mass needed to put that tank it in orbit.
Well, that depends greatly on whether crossfeed is employed. And staging would typically happen high enough that tank drag is no longer a problem.

This presents a fairly straightforward multivariable optimization problem, actually. Probably easier to approach iteratively, at least initially. I'll try taking the numbers from the ITS system as a base reference point.

Consider a PSTO launcher powered by three Raptor engines (two parallel and one sustainer), sized for carrying ten passengers to orbit. Dragon 2's dry mass is 6.4 tonnes; this includes the trunk and SuperDracos but we'll allow that for conservatism. Scaling gives us 9.14 tonnes, plus 200 kg per passenger (to allow for reasonable additional payload), so our minimum payload is 11.14 tonnes. I'll call it 12.

Three Raptors boast 9,150 kN of SL thrust, allowing us a launch mass of 666 tonnes with a brisk 1.4 T/W ratio. The ITS, which stages at 2.4 km/s, requires a hefty 470 tonnes of propellant for RTLS, corresponding to a reserve dV of 3.3 km/s. If we go with an S-1D or Atlas approach and crossfeed fuel from the core to the parallel engines, holding only enough fuel in the skirt for landing, then we'll need m0/m1 = 2.7 or so on the return alone. We don't really know the T/W ratio of the Raptor but if we guess at 200, that gives us an engine mass of 1.55 tonnes per Raptor. I can already tell that our vehicle T/W ratio at touchdown, even at minimum throttle, would be well over 100, which is utter nonsense.

So instead, let's suppose we fuel the parallel Raptors from larger tanks on the skirt alone, with no crossfeed either way. Let us allow 1.5 km/s of gravity and atmospheric drag; this means that 2.4 km/s at staging costs us 3.9 km/s. Our vehicle drinks up 464 tonnes of fuel to get there. Two thirds of this (309 tonnes) will be borne by the skirt.

Propellant fraction on the ITS booster is 0.96. This will be lower for us, due to the square-cube law, but it's fine for an approximation since that includes engine mass. Crunching the numbers, and getting tanks large enough to carry 309 tonnes of propellant plus landing reserves requires a dry mass of 14.9 tonnes plus 3.1 tonnes of engine. Given that the two Raptors cannot downthrottle below 124 tonnes combined thrust, that's a toasty 7.75 T/W. It would definitely need some dedicated landing thrusters.

What about the second/sustainer stage? At launch it massed 314 tonnes. With 12 tonnes of payload and 1.55 tonnes of engine, that's about 12 tonnes of tankage and 288 tonnes of fuel. At 2.4 km/s staging, this has dropped to 133 tonnes of fuel. The SL Raptor will give 361 s of specific impulse in space, so it could reach orbit with 34.5 tonnes of residuals, enough for about 1 km/s of landing reserves. With one SL raptor and a dry mass of 25.55 tonnes, that's a landing TWR of 2.4, so it would need dedicated landing thrusters as well.

Offline Rocket Science

  • Senior Member
  • *****
  • Posts: 7730
  • NASA Educator Astronaut Candidate Applicant 2002
  • Liked: 2047
  • Likes Given: 4761
Re: Configurations for reusable PSTO shuttle
« Reply #23 on: 03/20/2017 04:55 PM »
I would start with X-34B X-37B as a hypothetical example.  It is a 5-ish tonne spacecraft that reenters and flies back.  What if a version also provided some appreciable ascent delta-v, allowing use of someone's already-developed recoverable first stage?  The result would be a reusable Agena type stage/spacecraft bus.
I like it. But where do you put it? Slung alongside to allow parallel thrust? And what kind of engine would it take?
It would serve as a serial second stage.  As for engine, I would start by looking at storables, since spacecraft need propellant on-board for months or years at a time.

I'm suggesting an upgrade of the rocket in the first photo, which boosted an Agena that also served as the spacecraft bus in orbit.  These things flew every couple weeks or so.  First, replace the Thor first stage by a Falcon 9-like first stage that is recovered.  Second, replace Agena by something that looks like X-37B.

 - Ed Kyle
I agree Ed about a X-37B like 2nd stage, when I suggested it as part of my "Flyback Falcon 9" several years back...
https://forum.nasaspaceflight.com/index.php?topic=27477.0
« Last Edit: 03/20/2017 04:56 PM by Rocket Science »
"The laws of physics are unforgiving"
~Rob, Physics instructor, aviator, vintage auto racer

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Configurations for reusable PSTO shuttle
« Reply #24 on: 03/20/2017 05:19 PM »
I agree Ed about a X-37B like 2nd stage, when I suggested it as part of my "Flyback Falcon 9" several years back...
https://forum.nasaspaceflight.com/index.php?topic=27477.0
I feel like the wings on the second stage would make it dangerously draggy on the front end during ascent, given that we'd invariably be using propulsive landing on the first stage.

Online envy887

  • Senior Member
  • *****
  • Posts: 2233
  • Liked: 969
  • Likes Given: 644
Re: Configurations for reusable PSTO shuttle
« Reply #25 on: 03/27/2017 02:50 PM »
...
Three Raptors boast 9,150 kN of SL thrust, allowing us a launch mass of 666 tonnes with a brisk 1.4 T/W ratio...

What are you trying to optimize for here? This vehicle is substantially larger than Falcon 9 with its GLOM of about 580 tonnes. If you plug the same assumptions about GLOM, TWR, ISP, and PMF into an SSTO model, it will get very nearly the same mass to the ISS. So the extra flyback vehicle doesn't really help that much (because ground-lighting the orbiter engine effectively reduces that vehicles propellant fraction), unless you crossfeed (which effectively increases the propellant fraction, at the expense of the booster - where it's much less critical).

Crossfeed kind of sucks for a recoverable suborbital booster like Falcon Heavy - it's a kludge to allow a smaller upper stage, at the expense of a really hot downrange entry for the booster (with the advantage of sharing a stage with F9). But cross-feeding a suborbital booster into an orbital stage works really well. It massively improves the mass fraction of the orbital stage, where mass fraction is much more important.

With that in mind, I suggest an near-biamese configuration that would look a lot like a tanker ITS and a crew ITS launching while docked together. The tanker feeds both it's own and the crewed vehicle's engines until separation, at which point the full tanks of the crewed vehicle take it to orbit and the tanker boosts back to the landing site. Scaled down to a single Raptor per vehicle it would gross 532t at liftoff and put 10t to ISS orbit, reserving 1700 m/s for booster landing and 440 m/s for orbiter landing. And that's 10 tonnes of actual cargo, not Dragon+cargo.
« Last Edit: 03/27/2017 08:31 PM by envy887 »

Tags: