I’ve rediscovered asparagus staging! And have officially given up on SSTO rockets.

[GoSlash27]

14 minutes ago, herbal space program said:

So even if money is the key consideration asparagus can still be still the way to go. I guess it depends on how expensive the engines you drop are. I imagine that Reliant and Swivel-based stages could often be more economical to just drop.

HSP,

 The Reliant and Skipper are both cheap disposables and the Twin Boar is even better. In most cases, it's cheaper to dispose of SRBs.

Best,
-Slashy

[numerobis]

Fuel lines cost 150 per asparagus stage, no matter how big it is (and mass 50kg and incur a certain amount of drag). The fact that the cost doesn't scale makes asparagus staging attractive for lower stages, which are huge, and not as much for upper stages until you have a large spacecraft.

For cost, it's a lot cheaper to run SRBs than liquid-fuel rockets, for equivalent delta-v. They are a lot heavier though, so if you care about mass, stay away from SRBs.

You often need some thrust vectoring as well with that solid-fuel stage; for a heavy lift payload, the thrust-vectoring engine might burn enough fuel to be worth outfitting the SRB stage with fuel tanks and fuel lines. That's how the current best entry in the cheap and cheerful challenge is constructed.

One thing that really reduces the cost of staging is the hard point decoupler. A separable asparagus stage with the usual radial decoupler costs 750 funds, but with the hard point it's just 210 funds.

[magnemoe]

Uses crossfeed for my initial Mun and Minmus landing missionsas they have to use 1.25 meter stacks.
Else I only used it on an Eve lander and accent craft. 
Don't see much need otherwise, an 3.75 meter stage with large SRB can take over 200 ton to 1800 m/ and the payload circulates. 
If I need to lift something heavier I probably use it. 
On downside with aspargus is that it makes the rocket shorter, back before 0.90 this was critical as high rockets collapsed, however in 1.x an long rocket is more aerodynamic. 

I have started using SSTO rockets for crew transfers, else I only use disposable. 
I tend to use the lander or transfer stage as upper stage for the last 5-800 m/s, often with only 1200 m/s fuel on ship and fill tank in orbit. 
The long burn time prevents an suborbital recovery and an SSTO will be far larger, its also hard to land accurate enough. 

Also that I custum make rockets, look at weight, select fitting engine fuel and srb as needed

I have two SSTO for crew transfer as they are shorter and a fixed design. 

[KerikBalm]

Asparagus staging is mass efficient contrary to some of what has been said. Linear staging involvs significant amounts of engine mass being just "deadweight" if an engine isnt burning during a burn, thats inefficient. Asparagus has all engines burning all the time during a burn. Outer engines need to be optimised for high pressure (theyll discard before the air gets thin) inner engines burn at both high and low pressure... Meaning aerospikes are good for core stages.

asparagus encouragd many stacks (hence the name) , which is inefficient aerodynamically... And thus fell out of favor with 1.0's better aero model.

i still see a lot of crossfeeding designs.... Just more modest... Like two side stacks feeding into a core.

my 3 kerbal eve lander was asparagus staged... 2 drop tanks stacks (in one variant with aerospikes under them for TWR) feeding a pair of fueltanks+vector stacks, feeding an aerospike core.

it also had 1 linear stage, a terrier + fuel + mk1 crew cabin + mk1 pod and small nose cone on top of the aerospimecore stage.

it was more asparagus staged thannot.

i disagree with the OP though aboutthe mass penalties for making a ssto recoverable (Irl its prohibitive, in ksp, its easy)

in career i very quickly transition to ssto only designs, both airbreathing and rocket only designs

[HebaruSan]

Inspired by comments in this thread, I tried some experiments with asymmetrical asparagus staging, in which each stage is one booster and they drop in a clockwise pattern around the core, with mixed results. My strategy was to try to point each engine through what would be the center of mass when it was the "current" stage, just like Orbiter One does (see collapsed spoiler below, since it's easier to paste a bunch of Steam URLs than make an imgur album). This was complicated somewhat by the fact that the central engine was also thrusting the whole time, of course, so the net thrust vector didn't quite correspond exactly to the booster's orientation.

Positive: It can work. My test launchers all got to orbit. Smaller ones (Swivel as central engine) did somewhat better (pictured below is the largest test, with Skippers on each column).

Negative: There are serious challenges during both design and flight. I underestimated just how mobile the center of mass can be, so while I was able to optimize away the torque at the beginning of each stage, the craft became harder and harder to control as each stage burned on. It ended up in some kind of southwesterly orbit because I was not able to turn eastward, but oddly it still performed a pretty clean gravity turn in that direction, which suggests that rotating it on the pad might be an effective workaround.

Evaluation: Not recommended for practical use. The time and resources you'll spend trying to fly the thing (to say nothing of cosine losses) will far outweigh any hypothetical efficiency gains over symmetrical asparagus, unsurprisingly. Still a fun exercise, though.

Spoiler

 

[genericeventhandler]

I once toyed with the idea of dropping tanks from the bottom. 

 

[DarkHawkGames]

15 hours ago, LaytheDragon said:

Behold, Banana Staging! (Please, if you can, help me find a better name for it) 

If I may make a suggestion, Rolos staging? Like, the candy? If it ditches tanks like I think it does, dropping them off in a row one at a time.

[magnemoe]

18 minutes ago, genericeventhandler said:

I once toyed with the idea of dropping tanks from the bottom. 

 

An combination of twisted candle who is serial Asparagus with radial engines offset on each stage. Should work well with the Thud
I tend to use drop tanks on landers, especially during the early game going to Mun or Minmus, two drop tanks on lander takes me from circulation to start of decent. 
Using two or four as I tend to have four side tanks for wider base and engine in center. 


 

[Van Disaster]

 

- wow that was some time ago oO ( 2013, it seems ). Old-style pancake asparagus was never really friendly with FAR, but crossfeeding 2-4 booster stages most definitely was and still is. I always assumed "asparagus" was only a description of a rocket ( in KSP terms ) where one set of boosters fed another set of boosters rather than the core.

Edited June 30, 2016 by Van Disaster

[Frybert]

I've been playing around with vertical asparagus for a while. Don't have any designs to show at the moment but I do have some ideas for when I finally return to my pc.

[Warzouz]

18 hours ago, Green Baron said:

When it comes to putting something heavy up, and doing it fast and efficient, asparagus is mandatory. The simplest is 3 cores, like described by Pecan. It reduces unneeded mass and prolongs burntime. Combined with the right engine choice (rocket parts pack(s)) it's hard to imagine anything more efficient.

SSTOs are nice, but 200t 80t payload into orbit is hard work ... (nothing against hard working people :-))

 

Hard for SSTO spaceplanes. Ultra easy for SSTO rockets.

I've build my 600T LKO launcher in 10min from my 300T launcher (even I never used this 600T in my career ; I don't have such heavy payload). First test to LKO : success. Just strap more engines and fuel, check for TWR to be in the sweet spot (same for all rockets) check for delta-V. Launch. The only limitation for SSTO rockets is you computer capacity to handle many parts.

[THX1138]

On 29/06/2016 at 2:32 PM, Brainlord Mesomorph said:

please explain.

Isn't asparagus better than stacked stages?
(seems logically better, all engines on from liftoff, all stages fully fueled after separation) 

what would be better?

Not all engines have good efficiency at different altitudes so running the same engines from start to finish isn't necessarily very efficient anyway. If you're going to stack, you can use more efficient engines for the different altitudes.

That said, I usually use the Mammoth engines in a single SSTO from start to finish and recover them at the end, usually at >95% cost but I'll reload if it doesn't go according to plan!

On 29/06/2016 at 1:48 PM, Brainlord Mesomorph said:

Recovery is a good idea and IRL the recovery of lower stages is cost effective and definitely helps a space program get more bang for its buck.

In KSP however, we have the physics bubble, which makes lower stage recovery impossible.  The only way around this is the SSTO rocket, carry the whole rocket into space, and deorbit it later hopefully to the KSC.  If you get to the KSC Peninsula you can get 90% value back, and all it costs is fuel.

It makes sense.  At least on paper.  But somehow, I kept coming in over budget.  So I had to take second look at how this actually works.

To make a rocket SSTO and recoverable you have to add a lot to it.  A probe core, parachutes, RCS, SAS, additional fins, heat shield, landing gear the list goes on, it can cost up to 20,000 funds just in additional parts.  (“That’s OK,” you say, “I’m going to get all that money back when I recover the rocket.” In theory yes, but we’ll get back to that in a moment.) it also takes additional fuel to get all the way to orbit and still have enough fuel left for a deorbit maneuver, and you don’t get any of that money back.  Plus all the extra equipment and fuel is costing you payload capacity, my guess is that a stripped down version of the same booster (fuel tank, engine and fins only) designed for suborbit, easily lifts 10 to 15% more payload than its SSTO counterpart.

But still it’s OK assuming you get back to the KSC and nothing goes wrong.  (But this is KSP and that’s a ridiculous assumption.) What can go wrong?  I’ve been using as SSTO rockets for over a year now, and a lot can go wrong.

1.       What if you miss the KSC?  You might come down on the water, can your rocket survive that?  Or in the black mountains to the west, and you’ll lose it.

2.       It tumbles during reentry and the probe core overheats, you lose it.

3.       They are often so large that they block their own solar panels, so they go dead in orbit and become a large expensive piece of orbital debris.  Or go dead during reentry.  You lose it again.

4.       You overload it, and it barely gets the payload into orbit and comes down on the far side of the planet: 30% recovery, you lose money.

Because recoverable boosters are so much more expensive to build and take so much more fuel, by my guestimate, if you lose one out of seven SSTO rockets, you give back all the profit you made recovering the other six! (and I certainly was losing 1 out of 7)

So (in KSP’s physics/economics model) disposable boosters are actually cost-competitive, a lot easier to build and fly (just jettison and forget) and take a huge variable out of your budget. (IMHO: SSTO rockets just aren’t worth the extra trouble)

And since we’re throwing the rocket away, there is no reason not to go for multi-staging, its cheaper, and since asparagus staging is the more advanced form of multi-staging, my rocket designs have now come full circle.

I’ve rediscovered asparagus staging!  

If you reuse the same SSTO again and again, it's easier to predict performance and avoid losing the SSTOs.

[Blaarkies]

I felt the same about vert SSTOs suddenly burning up in the last updates. So my go-to SSTO was a Twin-Boar, with fuel tanks on top, ending with a 2.5m fairing(inside is all the probe core and rcs, etc.) It could haul 13ton payloads into orbit and pinpoint land back at KSC(by using the Trajectories mod). Launching upwards with no fins require prograde SAS control probes and the strong gimbaling of the twin-boar.

But recently it doesn't work anymore. After lots of tricks and ideas:
-com-88 attenna into the wind is supposed to stop heating a bit because some bug...it didnt do a thing
-2.5m heatshield onto a structural truss on the bottom of the engine...nothing, everything exploded before the ablator depleted
-decrease payload and burn 500m/s at 35km...mostly stopped the BOOM, but "At what price?!"
...
Then the eureka moment(i am glad that word has a "K" in it)
The rocket as usual, but add fins to the bottom(enough to point forward when almost empty). On top of the fairing plate, is a medium docking port. On that docking port is a 2.5m heatshield(rounded side pointing down towards engine), another medium docking port is attached to that heatshield. On top of that docking port comes the payload. The reasoning is:
-Get to orbit using only LF+OX.
-Drop the payload anywhere
-Use the now open docking port to dock to space station and transfer the spare fuel(leave about 500LF and matching amount of OX in the SSTO)
-Undock. Now decouple the heatshield thing. Bump it around it dock it to the SSTO
-You now have a nose heatshield
-Burn retrograde for 100m/s to fall back to KSC, keep aiming prograde in SURFACE(!) mode
-When close enough, pop a drogue chute. When the SSTO get pulled around, pop the other bigger chutes.
-Some thrusting at the last moment make it all go easier(even in water)

It is like 32 000f per SSTO, but I haven't had any damage on them for 12 launches thus far, and it costs less than 5000f per launch. No way disposable SRBs will beat that!

*I do admit...I lost three of the prototypes in testing before I figured out the current build, but this build hasn't crashed...yet

[Corona688]

18 hours ago, LaytheDragon said:

Behold, Banana Staging! (Please, if you can, help me find a better name for it) @HebaruSan

 

17 hours ago, Corona688 said:

drop-tank.

 

15 hours ago, LaytheDragon said:

I am testing the radial engines version, but with Banana Staging.

Banana tank?

[LaytheDragon]

Here, this is Banana Staging in its purest form:

Banana Staging doesn't appear to be any different than Stack Asparagus except the top tanks come off first, rather than the bottom.

But I could see the Banana Staged Drop Tanks being useful, as they allow only one drop tank at a time to be dropped rather than two, adding more flexibility (and also making it inline) to drop tanks. Here is that Imgur album from before, for those who missed it.

 

[panzer1b]

The main issue with reusable boosters is that in KSP fuel is extremely expensive compared to engines.  In reality raw fuel is a mere fraction of the cost, and unless we are talking about SRBs (most of which are specifically designed to be cheap and disposable), engines in general are the majority of the cost of a rocket (or at least considerably higher then any other part of the rocket).  This was the whole idea behind the space shuttle, which while way less fuel efficient then a conventional rocket, was able to recover the entire engine array as well as the electronics and control systems (+ the actual structure).  It was found to work because needing extra fuel for ascent was not much compared to the ability to recover the shuttle.

Now in KSP the best thing ive found is SSTOs that are designed like aircraft.  In reality this doesnt work since we do not yet have a engine with high enough ISP to pull off the circularization burn, but in KSP a properly built MK3 fuselage craft can send as much as 3 orange tanks (the most ive ever managed) into orbit without that much fuel useage, and unlike a rocket, deorbit and land more or less exactly where you need it to (its possible with rocket but very hard to land perfectly on KSC).

That said, i never bothered with SSTO rockets (aside from trying out the concept to see if it could be done like space-X did with landing on a barge) because they require alot of engineering, and the launch is somewhat time consuming and difficult.  Not saying that isnt part of the fun of KSP, but myself im far more interested in actually doing something in orbit or on somne other planet, and i view rockets as a means to an end, essentially no point especially since i play sandbox.  Its quite an accomplishment to land a SSTO rocket on a boat (or any other relatively small target like the helipad), but aside from that it makes neither monetary sense in KSP and it requires alot more time commitment in both rocket design and the entire launch sequence.

As for asparagus, it works but the new aero drag doesnt make it as attractive as it used to be, and the only time i use it myself is when im trying to launch something excessively heavy and dont feel like actually spending the time to engineer a proper launch stage.  Most of my launchers are 2 stage with 3 engines, 2 boosters and 1 central engine, since they loift the majority of payloads i need and are very quick to make not to mention super stable.

[Green Baron]

1 hour ago, Warzouz said:

Hard for SSTO spaceplanes. Ultra easy for SSTO rockets.

I've build my 600T LKO launcher in 10min from my 300T launcher (even I never used this 600T in my career ; I don't have such heavy payload). First test to LKO : success. Just strap more engines and fuel, check for TWR to be in the sweet spot (same for all rockets) check for delta-V. Launch. The only limitation for SSTO rockets is you computer capacity to handle many parts.

Another limitation is my sense of aesthetics ...

But of course, you are right.

 

[eataTREE]

52 minutes ago, panzer1b said:

they require alot of engineering, and the launch is somewhat time consuming and difficult.

As a Misson Controller with a job, a wife, and a family, this is the deciding factor for me. I only have a certain amount of real-world time to get that damn payload up there or it's going to have to wait until next weekend. The traditional semidisposable rocket is quick to get to orbit, almost always on the first try; the SSTO requires a lengthy flight through the atmosphere, not to mention the endless tuning and repeat attempts required to actually get it into space. And then there's re-entry. I put SSTOs in the "neat, but impractical" category.

[Violent Jeb]

[redacted, done]

Edited August 16, 2016 by Violent Jeb

[wumpus]

4 hours ago, Warzouz said:

Hard for SSTO spaceplanes. Ultra easy for SSTO rockets.

I've build my 600T LKO launcher in 10min from my 300T launcher (even I never used this 600T in my career ; I don't have such heavy payload). First test to LKO : success. Just strap more engines and fuel, check for TWR to be in the sweet spot (same for all rockets) check for delta-V. Launch. The only limitation for SSTO rockets is you computer capacity to handle many parts.

Two issues for SSTO rockets:

In career, that's a lot of money you have to have in reserve for your SSTO (sure, you'll get it back, but you have to put up the payment first).

Second, and most importantly: how close to you land to the pad?  With mechjeb, its too easy.  You just hit "land on pad" and collect your 100%.  I'm curious how many players can hit close enough to the pad that they wouldn't benefit from dropped kickback stages (bunched onto two or less decouplers, natch).

SSTO rockets (or S[l]STO, Single liquid Stage to Orbit) have the advantage that you only have to perform one landing.  SSTS/TSTO (single stage to space, two stage to orbit) allows recovery, but requires two landings (a docking isn't faster than a landing.  Even thought they will be *very* close).  This makes them inconvenient to fly, if easier to make.  Maybe just use stage recovery (or flight manager and don't bother getting the first stage into orbit).

[gchristopher]

 

I'm reading a lot of statements like:

4 hours ago, panzer1b said:

The main issue with reusable boosters is that in KSP fuel is extremely expensive compared to engines.

Numbers! Will someone please post some numbers for 1.1.3 to back up all of these arguments?

For a given payload size, the question of "cost to LKO" is can absolutely be answered with a single number in Funds (), giving a Fund/Ton price ratio. This is easily answered by "who can post the best price ratio."

 

5 hours ago, Warzouz said:

Hard for SSTO spaceplanes. Ultra easy for SSTO rockets.

Agreed. The 1.0 aero changes made fitting things in cargo bays almost mandatory for spaceplanes, while you can much more easily handle odd-sized payloads on rockets using fairings. Prior to 1.0, SSTO spaceplanes using airhogging were by far the most cost-efficient option. I don't know what wins in 1.1.3.

 

40 minutes ago, wumpus said:

In career, that's a lot of money you have to have in reserve for your SSTO (sure, you'll get it back, but you have to put up the payment first).

 

Yep, that's a very good point.

 

43 minutes ago, wumpus said:

Second, and most importantly: how close to you land to the pad?  With mechjeb, its too easy.  You just hit "land on pad" and collect your 100%.  I'm curious how many players can hit close enough to the pad that they wouldn't benefit from dropped kickback stages (bunched onto two or less decouplers, natch).

For a given rocket or similar ones, you only need to know where to put your reentry periapsis and have a little dv left to catch minor overshoots of KSC. Once you know the answer, it never changes and is easy to do manually. And Mechjeb's bad enough at its early guesses that you'll probably need to fly a lot of the landing by hand anyway. The one place where Mechjeb absolutely beats a human pilot is doing a soft touchdown with rockets instead of using a lot of parachutes. That's the real savings: use MJ and you don't need any chutes.  

 

 

 

[GoSlash27]

4 hours ago, eataTREE said:

As a Misson Controller with a job, a wife, and a family, this is the deciding factor for me. I only have a certain amount of real-world time to get that damn payload up there or it's going to have to wait until next weekend. The traditional semidisposable rocket is quick to get to orbit, almost always on the first try; the SSTO requires a lengthy flight through the atmosphere, not to mention the endless tuning and repeat attempts required to actually get it into space. And then there's re-entry. I put SSTOs in the "neat, but impractical" category.

I absolutely love SSTO spaceplanes... for the missions they are most adept at handling. Shuttling crew between KSC and orbit, delivering fuel/ monoprop to a station in LKO, that sort of thing. They require a lot of time/ effort to perfect, but once that's done they don't need redesigned and those missions are my most common.
 I always use disposable vertical staged lifters for actual cargo. Tailoring the lifter to the payload is much simpler and the enginerding is a lot more straightforward.
 Of course... this all depends on how you play the game. I'm a big fan of modular designs and orbital assembly... so this approach works well for me.

Best,
-Slashy

[THX1138]

5 hours ago, eataTREE said:

As a Misson Controller with a job, a wife, and a family, this is the deciding factor for me. I only have a certain amount of real-world time to get that damn payload up there or it's going to have to wait until next weekend. The traditional semidisposable rocket is quick to get to orbit, almost always on the first try; the SSTO requires a lengthy flight through the atmosphere, not to mention the endless tuning and repeat attempts required to actually get it into space. And then there's re-entry. I put SSTOs in the "neat, but impractical" category.

I think you could save more time by designing a couple of SSTOs and reusing them with your different playloads. You can put them in your subassemblies section of the VAB. Click the arrow in the top left hand corner of the VAB and then on the green icon in case you don't know where that section is.

[Draco T stand-up guy]

On 30/06/2016 at 8:53 AM, Gaarst said:

I find strange that your heavy lifter got smaller with asparagus than otherwise. Of course it depends on how it is designed in the first place, but I remember going the other way round (getting lighter) when switching all my asparagus rockets to more conventional types back in 1.0.

I've had both happen.

One when the asparagus got way out of hand for lifting a tiny amount (IIRC, it had 8 boosters around the core with another 8 onioned lifting a mk1 cmd pod and two man passenger cabin to Munar orbit).

One where I went from a super tall 2.5m that could barely lift off even with the SRBs on the side to a core + 4 asparagus lifter that had less stages and less fuel while lifting about 4 times as much and costing far less.

 

[magnemoe]

2 hours ago, Green Baron said:

Another limitation is my sense of aesthetics ...

But of course, you are right.

 

This offend my sense of aesthetics, during launch however that is 5 minutes of standardbase2c many years of life. 

It need 4 km/s to reach orbit, 2 of them is in the base itself, base can land and take off from medium bodies without refueling, the large fuel capacity is also nice then mining and resupplying