Jump to content

Design Challenge: Orbital Shuttle


Recommended Posts

I've considered using rocket designs where the large expensive engines are jettisoned and recovered to maximize recovery costs without losing too much payload to recover the entire booster (I'll probably do this with some heavy-lift vehicles eventually). The above approach is interesting, and making it fully recoverable would probably be rather challenging but might be doable.

The biggest problem I can see when translating that idea to KSP is separating the skirt with the engines. Stock in-line decouplers have a solid collision mesh even though they appear hollow (a property shared by the toroidal fuel tanks, in fact). I have discovered this the hard way by attempting to drop some toroidal tanks and a decoupler whilst reusing the same engine. It did not go well. I don't know if modded decouplers all have the same issue though, and I may do some testing to try to find out.

Link to comment
Share on other sites

21 minutes ago, eloquentJane said:

I've considered using rocket designs where the large expensive engines are jettisoned and recovered to maximize recovery costs without losing too much payload to recover the entire booster (I'll probably do this with some heavy-lift vehicles eventually). The above approach is interesting, and making it fully recoverable would probably be rather challenging but might be doable.

The biggest problem I can see when translating that idea to KSP is separating the skirt with the engines. Stock in-line decouplers have a solid collision mesh even though they appear hollow (a property shared by the toroidal fuel tanks, in fact). I have discovered this the hard way by attempting to drop some toroidal tanks and a decoupler whilst reusing the same engine. It did not go well. I don't know if modded decouplers all have the same issue though, and I may do some testing to try to find out.

Yes, this makes it tricky, as sections attached to separate decouplers automatically detach upon staging, no matter how well you strut them together.

The solution, I think, would be to use a single radial decoupler and build the skirt in mirror symmetry around it. Then set the decoupler's force to zero and throw some sepratrons on there. To keep it balanced, you can throw another radial decoupler on the opposite side and just strut stuff to it.

Link to comment
Share on other sites

17 minutes ago, sevenperforce said:

The solution, I think, would be to use a single radial decoupler and build the skirt in mirror symmetry around it. Then set the decoupler's force to zero and throw some sepratrons on there. To keep it balanced, you can throw another radial decoupler on the opposite side and just strut stuff to it.

I've thought about doing that. The issue I've had with that approach is the difficulty of building a nice-looking engine skirt. I think the best parts for it aesthetically would probably be wing segments in most cases, but that might cause lift problems. I might give it a try anyway though at some point, it's an interesting enough concept.

Link to comment
Share on other sites

31 minutes ago, eloquentJane said:

I've thought about doing that. The issue I've had with that approach is the difficulty of building a nice-looking engine skirt. I think the best parts for it aesthetically would probably be wing segments in most cases, but that might cause lift problems. I might give it a try anyway though at some point, it's an interesting enough concept.

How about a Mk3 to Mk2 Adapter or a C7 Brand Adapter (Slanted)? Arrange those radially and you should be set.

Link to comment
Share on other sites

21 minutes ago, sevenperforce said:

How about a Mk3 to Mk2 Adapter or a C7 Brand Adapter (Slanted)? Arrange those radially and you should be set.

The nosecone parts directly above the engines isn't the problem. The issue is as follows:

  • There is a central engine which must not be jettisoned, and several engines attached radially around it which will be jettisoned
  • The radial engines need nosecones (not difficult to find)
  • They also need some sort of connecting structure to ensure that all radial engines act as a single vessel after separation
  • The connecting structure can't be a single adapter or decoupler part because it has to surround the central engine and be jettisoned with the radial engines

Hopefully that gives you a better idea of where the difficulty lies. I think constructing a shell out of wing parts is probably the way to go, but I don't know how much of an effect that would have on the rocket's flight due to lift.

Edited by eloquentJane
Link to comment
Share on other sites

4 hours ago, sevenperforce said:

Anybody want to try building this?

Gave it a shot. Tried, failed.... but the concept seems promising....

I kinda lost control on the way up. Pictured is pre separation, separation moment, and a nice shot with the chute coming out with the launch vehicle in the background.... Will need to play around with the angles to get it straight, and possibly swap to outer LV series engines for vectors, and add a lot more fuel to the orbiter. Anyway, is a start on the idea....

Also, pics are twilight, sorry.

cl4qiZu.pngJfY1KT6.png
iBg3s6P.png

Link to comment
Share on other sites

Experimented around with a blended wing/lifting body design, going for a pair of smaller external tanks instead of a larger one to counter the offset thrust problem
bkn7xFT.png
11 Kerbals to LKO with room for 5 tons of cargo or an internal tank for extended range. Went with RAPIERS for versatility/atmo cruising in case of LZ overshoot/ emergency abort/etc., but it would work just as well with LV-T30s instead.
 

5 hours ago, sevenperforce said:

Anybody want to try building this?

PpJQKCB.png
The short Mk3 Cargo Bay works well for this.

Link to comment
Share on other sites

12 hours ago, SuicidalInsanity said:

Experimented around with a blended wing/lifting body design, going for a pair of smaller external tanks instead of a larger one to counter the offset thrust problem

11 Kerbals to LKO with room for 5 tons of cargo or an internal tank for extended range. Went with RAPIERS for versatility/atmo cruising in case of LZ overshoot/ emergency abort/etc., but it would work just as well with LV-T30s instead.

Very nice. I'd be a little concerned about differential yaw between the two boosters in an IRL configuration, but having the right attachment points should help with that.

I wonder what kind of size we'd need in this configuration to take a crew of 7 to the ISS and back. Would three Raptor engines (one on the orbiter and one on each booster) be enough? Lack of vacuum optimization is a bit of a problem, but this could be remedied either by giving the central Raptor a longer but deconverging engine bell (like an SSME) and/or adding a small pair of vacuum-optimized OMS engines for orbital insertion assist and circularization.

The other problem is whether the Raptor can throttle deeply enough to land the two boosters on their tails.

13 hours ago, SuicidalInsanity said:

PpJQKCB.png
The short Mk3 Cargo Bay works well for this.

Oh wow, that's beautiful! Very nice. So can the cargo bay be jettisoned as whole without decouplers, then?

What kind of performance difference do you get by adding fuel tanks to the cargo bay skirt (either for crossfeed or simply to fuel the quad)? And if you add a probe core, can you successfully land the quad by itself?

Link to comment
Share on other sites

A decoupler is still needed. The trick is flipping the decoupler upside down so it'll detach whatever is attached to it, but still remain connected to the rocket. Cargo bay is connected to the decoupler, decoupler is connected to the bottom node of the central engine. Decoupler gets offset into the engine root, and cargo bay gets offset until it's flush with the bottom of the fueltank.

Launching it straight up resulted in a max apoapsis of around 1.25 million meters. Replacing the Rockomax adapter engine mounts with C7 1.25-2.5m fueltanks and removing the X-200-32 from the main tank for same total weight/fuel carried increased the max apoapsis to around 1.75 million meters, so there's definite performance improvements from fuelling the skirt.

As for the engine quad/skirt assembly itself, it's aerodynamically stable and will fall engines first, so a powered landing is possible, but it needs at least a FL-T200's worth of fuel per engine to do so,

Link to comment
Share on other sites

1 hour ago, SuicidalInsanity said:

A decoupler is still needed. The trick is flipping the decoupler upside down so it'll detach whatever is attached to it, but still remain connected to the rocket. Cargo bay is connected to the decoupler, decoupler is connected to the bottom node of the central engine. Decoupler gets offset into the engine root, and cargo bay gets offset until it's flush with the bottom of the fueltank.

Lovely. I figured there was a way to do that.

1 hour ago, SuicidalInsanity said:

Launching it straight up resulted in a max apoapsis of around 1.25 million meters. Replacing the Rockomax adapter engine mounts with C7 1.25-2.5m fueltanks and removing the X-200-32 from the main tank for same total weight/fuel carried increased the max apoapsis to around 1.75 million meters, so there's definite performance improvements from fuelling the skirt.

As for the engine quad/skirt assembly itself, it's aerodynamically stable and will fall engines first, so a powered landing is possible, but it needs at least a FL-T200's worth of fuel per engine to do so,

You can, of course, reserve only two engines for landing and allow the other two to burn out at separation. I'm guess the maximum performance would come from carrying the majority of the fuel in the skirt and crossfeeding it to the core engine so the core is fully-fueled at separation, but that might push the limits of what can wrap around the core without a collision at separation.

Link to comment
Share on other sites

24 minutes ago, sevenperforce said:

I'm guess the maximum performance would come from carrying the majority of the fuel in the skirt and crossfeeding it to the core engine so the core is fully-fueled at separation

This is essentially identical to conventional asparagus staging, except with a single separated stage rather than two or more.

Edited by eloquentJane
Link to comment
Share on other sites

8 minutes ago, eloquentJane said:

This is essentially identical to conventional asparagus staging, except with a single separated stage rather than two or more.

Yes, precisely.

The thought being that if you can combine the higher efficiency of asparagus staging with a single separation event, you can have the closest possible performance to a refuel-and-refly system. Parallel staging has a much higher efficiency than serial staging, high enough that everything can be overengineered for rapid reuse and still have a good enough payload fraction to act as a crew shuttle.

Link to comment
Share on other sites

On 3/14/2017 at 7:57 AM, sevenperforce said:

In particular, I like the idea of a parallel-stage-to-orbit design, where all the launch engines fire on the pad together, with the orbiter engines continuing to fire through stage separation (think Space Shuttle). This introduces additional complications, particularly with respect to configuration.

I fear I'm being a little dense, but not sure I quite understand your definition of "parallel stage to orbit."  It sounds to me like it's something like this:

  • Not an SSTO - i.e., at least one stage must peel off at some point before reaching LKO
  • Every engine you've got is ignited at liftoff
  • All components (both the orbiter and whatever booster stages you have) must be recovered intact. 

So by that definition, the Space Shuttle was not a PTSO because (1) the external fuel tank burned up, and (2) the OMS engines were not fired before reaching space.  (I guess also the SRBs splashed down, which would not fit your challenge requirements but might fit a broader definition of recoverable).  

Is that correct?  

Link to comment
Share on other sites

1 hour ago, Aegolius13 said:
  • Not an SSTO - i.e., at least one stage must peel off at some point before reaching LKO

Not an SSTO is generally correct but not necessarily. I could, for example, make a spaceplane that launches upon the back of a larger spaceplane, and the whole device acts as an SSTO. The large spaceplane is the launch vehicle and will return to the ground immediately after a de-orbit burn, whilst the smaller spaceplane has extra fuel (perhaps it drained a tank on the larger plane during launch) and thus has a range beyond the low initial parking orbit (it could, for example, go to the Mun and back). Alternatively, a design like the first R7 rocket which launched Sputnik may be considered a PSTO (but not an SSTO), as all engines ignite on the ground and then the boosters are jettisoned while the core stage continues to fire until orbital insertion is completed.

1 hour ago, Aegolius13 said:

Every engine you've got is ignited at liftoff

Every launch engine. Not necessarily every engine on the entire vehicle. As far as I can tell, a PSTO design requires that there are a minimum of two stages, and both stages have at least one engine which ignites at liftoff and continues to burn until either that stage is jettisoned or until the vehicle reaches orbit. OMS engines don't need to ignite on the ground, nor would jet engines if the vehicle takes off vertically (the one I've been working on for this challenge has both of these; the OMS engines are just used in space and the jet engines are for assistance after re-entry).

1 hour ago, Aegolius13 said:

All components (both the orbiter and whatever booster stages you have) must be recovered intact. 

That's a feature of this challenge, not a requirement for any vehicle to be considered a PSTO design. The space shuttle was a PSTO, just without total recoverability. Similarly, the proposed rocket derived from the Saturn V first stage (discussed earlier in this thread) would probably originally have been designed so that the F1 engines that are separated mid-flight would be expendable (since at that time it was unlikely that the technology would exist to enable recovery of them).

Link to comment
Share on other sites

Here's my entry:

VjpX3Gr.png

Lots of mods involved, including @Nertea's full set, FAR, StageRecovery and Procedural Wings.

The ship has a MkIV main body, bimodal engines and LF/O RCS system, no monoprop needed.

HydroLOX boosters augmented with SRBs allow a decent TWR at launch (1.6, without SRB is 1.1)

LES is guaranteed by ejection seats supplied by the appropriate mod

 

Full test mission report here: http://imgur.com/a/iIdAt

Link to comment
Share on other sites

On 3/17/2017 at 6:51 PM, eloquentJane said:

Not an SSTO is generally correct but not necessarily. I could, for example, make a spaceplane that launches upon the back of a larger spaceplane, and the whole device acts as an SSTO. The large spaceplane is the launch vehicle and will return to the ground immediately after a de-orbit burn, whilst the smaller spaceplane has extra fuel (perhaps it drained a tank on the larger plane during launch) and thus has a range beyond the low initial parking orbit (it could, for example, go to the Mun and back).

They don't even have to be two spaceplanes. Skylon is supposed to employ a kicker stage in its cargo bay for BLEO missions like GTO comsats. In this instance you have a two-stage vehicle (Skylon and the kicker stage) but orbit is reached in a single stage.

Then again, Skylon could potentially carry a larger payload in a suborbital hop, but give its payload a high enough apogee to circularize. In this case it would be back to a TSTO (two stage to orbit), since orbit isn't actually reached by the first stage.

On 3/17/2017 at 5:30 PM, Aegolius13 said:

I fear I'm being a little dense, but not sure I quite understand your definition of "parallel stage to orbit."  It sounds to me like it's something like this:

  • Not an SSTO - i.e., at least one stage must peel off at some point before reaching LKO
  • Every engine you've got is ignited at liftoff
  • All components (both the orbiter and whatever booster stages you have) must be recovered intact. 

So by that definition, the Space Shuttle was not a PTSO because (1) the external fuel tank burned up, and (2) the OMS engines were not fired before reaching space.

As Jane said, recoverability is just an element of this challenge, not anything necessary to the definition of a PSTO.

Not every engine has to be ignited at launch, and engines ignited on the pad don't necessarily have to be used for orbital insertion; technically, it's enough that some of the engines ignited on the pad reach orbit. The Space Shuttle and the R-7 were both PSTOs; the Atlas D was right on the cusp of being a PSTO depending on what you consider an orbit.

A crossfed Falcon Heavy without its second stage could easily launch a Dragon into orbit as a PSTO, but without enough residuals to recover the core.

On 3/17/2017 at 6:51 PM, eloquentJane said:

The proposed rocket derived from the Saturn V first stage (discussed earlier in this thread) would probably originally have been designed so that the F1 engines that are separated mid-flight would be expendable (since at that time it was unlikely that the technology would exist to enable recovery of them).

During the Saturn V era, recovering the four F-1s on the skirt would have been more feasible with midair helicopter catch than by propulsive landing. I don't know when the first inflight-restartable kerolox engine was designed, but that tech was in its infancy if it existed at all, and it would not have been possible to downthrottle the F-1 low enough. Nor, for that matter, was there sufficient computer technology for autonomous propulsive landing a la Dragon 2 or Falcon 9.

A simpler option was to simply chute them down and let them drop in the ocean. The F-1 was a pretty hardy beast and could have been refurbished and reflown more easily than the SSMEs were, even after a dunk in the Atlantic. A Saturn 1D-based PSTO launcher with 4/5 recoverable engines would have been able to outperform the Shuttle system in cost, frequency, and pretty much every way other than downmass.

On 3/18/2017 at 11:15 PM, Hesp said:

Here's my entry:

VjpX3Gr.png

Lots of mods involved, including @Nertea's full set, FAR, StageRecovery and Procedural Wings.

The ship has a MkIV main body, bimodal engines and LF/O RCS system, no monoprop needed.

HydroLOX boosters augmented with SRBs allow a decent TWR at launch (1.6, without SRB is 1.1)

LES is guaranteed by ejection seats supplied by the appropriate mod

Good lord that is beautiful!

How large are the prop tanks on the orbiter itself (and, for that matter, where are they located)? I can see from your gallery that staging happens around 1.7 km/s so there has got to be pretty substantial fuel reserves on board. Are the wings wet (i.e., fuel-carrying)?

Using recoverable onboard solids to boost launch TWR is a nice trick! I'm guessing the losses from carrying the empty solid boosters up and down are minimal. One of the ideas I like is a blended air-augmentation-and-solids launcher, where a liquid booster is shrouded by an annular solid-fueled rocket ignited at launch; the solids burn away to leave behind an air augmentation shroud.

Link to comment
Share on other sites

1 hour ago, sevenperforce said:

Good lord that is beautiful!

How large are the prop tanks on the orbiter itself (and, for that matter, where are they located)? I can see from your gallery that staging happens around 1.7 km/s so there has got to be pretty substantial fuel reserves on board. Are the wings wet (i.e., fuel-carrying)?

Using recoverable onboard solids to boost launch TWR is a nice trick! I'm guessing the losses from carrying the empty solid boosters up and down are minimal. One of the ideas I like is a blended air-augmentation-and-solids launcher, where a liquid booster is shrouded by an annular solid-fueled rocket ignited at launch; the solids burn away to leave behind an air augmentation shroud.

Thanks!

The tanks for LF/O are in the main fuselage (the mk4 system holds the equivalent of mk1 tanks at the side of the structure) and in the tail. Some other fuel (mostly LF) is held in the wings. I'll give you the exact figures when I get home. By the way, the tanks are huge if the purpose is limited to LKO operation, I only used 30% for launch and 145km orbital insertion. The trick for this ship is lift! I could almost fly it up like a spaceplane, following a very shallow gravity turn. I'm curious to see what is its payload to LKO

The Hammers are a quick and dirty solution for an easy liftoff. I feel that the first stage can still be optimized, though. Before adding them, I reached LKO with LH2 boosters alone, even with an anemic 1.1 TWR off the pad.

If you look carefully, you can see a 2.5m docking port on the back of the vessel. I see it as the right place for the SRB stage. After decoupling, it would be free and ready for a rendezvous with an interplanetary transfer stage (Nuclear/ion/VASIMR engines and tanks, hab and science facilities, life support, etc). The cargo bay would still be helpful to haul smaller craft, like a lander or some probes.

--

I don't think your idea would be feasible. After all, the airbreather engines are good only for the first 20-25km. If you cover the first half with solid rockets, you're bringing a lot of complication in for little returns.

Anybody tried a RAPIER booster stage (like 4 rapiers with 2.5m tanks and shock cone on top)? Maybe I'll give it a try this evening, if the numbers are right.

Edit: If i remember well, Nertea's mod has a 2.5m LH2 rapier engine. I'll definitely pick that one for the booster!

 

Edited by Hesp
Link to comment
Share on other sites

2 hours ago, Hesp said:

I don't think your idea would be feasible. After all, the airbreather engines are good only for the first 20-25km. If you cover the first half with solid rockets, you're bringing a lot of complication in for little returns.

I probably was unclear. The idea is to have a pure liquid rocket engine at the center with a small shroud packed full of solid fuel. The liquids combined with the solids get you off the pad, and when the solids burn out there's enough forward airspeed for air augmentation through the shroud to boost the thrust of the liquid engine. Allows you to use a smaller liquid engine than you'd otherwise need.

2 hours ago, Hesp said:

Anybody tried a RAPIER booster stage (like 4 rapiers with 2.5m tanks and shock cone on top)? Maybe I'll give it a try this evening, if the numbers are right.

Edit: If i remember well, Nertea's mod has a 2.5m LH2 rapier engine. I'll definitely pick that one for the booster!

TWR would be horrible, I think.

Link to comment
Share on other sites

this my "Kama-LS" PSTO system. It have 2 cheap boosters, 2 cheap (juno) atm.engine and 1 cheap rocket engine (trust limiter is half),

It can deliver 2-3 tonn payload (comm sputnik) to LKO

Boosters return system with 2 junos (and with its big wing are good glide)

Main ship land with it wings to runway after orbit (but as always i get tooo much delta for orbital flight... but... as i say before it depends of payload)

Deep night right now in my country, so I make video for this ship only tomorrow or at weekend... so sorry for this and good night.

69f775ee3131.png

booster looks like cruise missile. not a problem to do this in RL

099ab94724de.png

construction of main ship is simple, but I made it nose with heatshield for more realism

81d84b63a2a9.png

the only thing is not good - the dimension between booster can be mach more , or it can be burned with main engine..(in real life). But I think I make it also in RSS for real severe test, but later (still not lose hope for FAR update for 1.2.2)

ps; sorry for my poor English

9c1bb410ade9.png

 

76d9e41aac55.png

 

Edited by *MajorTom*
Link to comment
Share on other sites

18 hours ago, sevenperforce said:

TWR would be horrible, I think.

TWR on the pad with only rapier boosters (Cutlass dual-mode engine LH2 / LH2+LOX from Nertea's Mk4) is around 0.8.

I used a half filled twin-boar as a liquid booster on the free docking port. It gets ditched really soon (recovered 10km from KSC with parachutes), only to get the ship supersonic when the rapiers start to shine.

I'll be able to adjust the maximum payload to LKO tuning the thrust and fuel amount in the twin-boar. :) Forgot to take screens during testing, maybe I'll attach some pics this evening

 

Link to comment
Share on other sites

15 hours ago, *MajorTom* said:

this my "Kama-LS" PSTO system. It have 2 cheap boosters, 2 cheap (juno) atm.engine and 1 cheap rocket engine (trust limiter is half),

It can deliver 2-3 tonn payload (comm sputnik) to LKO

Boosters return system with 2 junos (and with its big wing are good glide)

Main ship land with it wings to runway after orbit (but as always i get tooo much delta for orbital flight... but... as i say before it depends of payload)

Deep night right now in my country, so I make video for this ship only tomorrow or at weekend... so sorry for this and good night.

booster looks like cruise missile. not a problem to do this in RL

construction of main ship is simple, but I made it nose with heatshield for more realism

the only thing is not good - the dimension between booster can be mach more , or it can be burned with main engine..(in real life). But I think I make it also in RSS for real severe test, but later (still not lose hope for FAR update for 1.2.2)

ps; sorry for my poor English

 

Looks great, but I can't tell -- do the orbiter's engines fire from the very start? If not, it's not a PSTO.

Link to comment
Share on other sites

4 hours ago, sevenperforce said:

Looks great, but I can't tell -- do the orbiter's engines fire from the very start? If not, it's not a PSTO.

of course all engines (and even juno strats at same time then i hit "space" button :)) ) , only one engine is shut down - sputnik payload. (i record video now)

ps: and concept of return back has it prototype - maybe you know about Baikal boosters for our Angara rocket ... it is the sort of same

Edited by *MajorTom*
Link to comment
Share on other sites

55 minutes ago, *MajorTom* said:

of course all engines (and even juno strats at same time then i hit "space" button :)) ) , only one engine is shut down - sputnik payload. (i record video now)

ps: and concept of return back has it prototype - maybe you know about Baikal boosters for our Angara rocket ... it is the sort of same

Took a closer look, and I see how it works now. Nice concept!

Link to comment
Share on other sites

On Tuesday, March 14, 2017 at 1:51 PM, sevenperforce said:

Yeah, it would have been tough. Atlantis was already gearing up for another mission, so it was available. It would have needed to be pushed through to launch, skipping 90% of preflight checks, and launch and rendezvous. Then they would have had to actively hold the two ships in a bay-to-bay perpendicular orientation (since neither had a Canadarm for grappling) while they did repeated EVA from Atlantis to Columbia and back, one crew member at a time. Doable, but just barely.

Not to mention only 7 seats in the shuttle. Atlantis would have to have compensated for that. like leaving some crew on the ground to increase chance of survival for those who are not strapped in.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...