Is asparagus the best staging system? (might contain science)

[allmappedout]

I always tend to think of Asparagus staging as compounding the maximal rocket equation to smaller segments of time (t) - ie: you want to reduce as mass as quickly as possible (hence dumping it as soon as it's done) whilst keeping constant (or increasing) thrust. Thus I guess the idealised solution would be having an infinite number of infinitely small tanks which weighed an infinitesimal amount, st at each delta t, these tanks were discarded as they were emptied. Obviously this is impossible, and hence delta t is longer between drops, but the idea is the same.

Maybe I'm overcomplicating things...

[Pbhead]

There are other momentum effects that impact it. And the other factors I mentioned.

If it was "really easy to fix" I'm pretty sure NASA would be doing it.

There is a reason why I only responded to that section of your post. other factors harder, yes. Spiraling because your fuel flow is not symmetrical? make it symmetrical.

I built a stack like this in 0.17 but it was unstable because the game didn't account for the vectored thrust being ahead of the center of mass (so the vectoring was reversed). Has this been corrected in the current version?

That is a bummer. Because really. in real life, a chariot style space ship would require significantly less superstructure than traditional push from the bottom, because you can use much much lighter materials if the force is tension, as opposed to compression.

Extra superdeduper important if you have frail life forms on board which cant be too close to radiation sources requiring the engines to not be close to the crew compartments.

[Eric S]

It's not used in real-life because conservation of angular momentum would result in an ever-increasing, unstoppable roll of the craft as the fuel spiraled inward to feed the core stage.

I disagree, though I'll gladly listen to counterarguments. The angular momentum is minimal. Yes, you're causing angular acceleration of a portion of the fuel supply as fuel enters the fuel line. However, you're also causing angular deceleration of an equal amount of the fuel supply as it exits the fuel line. There will be a brief moment where there is only acceleration as the fuel line fills, and a brief moment where there is only deceleration when the tank is empty but the fuel line isn't. That doesn't sound like ever-increasing, and hardly sounds unstoppable.

I think part of our disagreement is that you're basing this on it spiraling in, which it doesn't do. It moves circularly, stops (or at least makes a 90 degree turn dumping it's angular energy into whatever forced it to turn 90 degrees), then moves directly inward. I understand conservation of angular momentum, I just don't think it applies here the same way you do.

I do agree with the danger of it. The engineering challenge impossible? Difficult, yes, but I don't think I'd call it impossible since that's very similar to the external tank on the space shuttle, not to mention the fact that the Falcon 9 Heavy is scheduled to test soon (later this year, if I remember correctly), and while not true asparagus staging, has many of the same technical challenges, though at a smaller fuel flow rate since each fuel line is feeding half of the center stack at most, instead of half the center stack plus two other complete stacks.

[RoboRay]

There is a reason why I only responded to that section of your post. other factors harder, yes. Spiraling because your fuel flow is not symmetrical? make it symmetrical.

Keeping it symmetrical is a huge challenge, even if you use an "H" booster configuration instead of a spiral one. Attitude changes and steering of the craft will contribute to momentary changes in fuel distribution, which could result in a feedback loop that increases the deviation. Even more of a problem is that any variation in fuel consumption by different engines would result in different fuel flow quantities in different directions, allowing conservation of angular momentum to do it's thing anyway.

I assure you that even the fuel-flow considerations of asparagus-stalk boosters are not "really easy to fix" in real-life with simple ideas.

I disagree, though I'll gladly listen to counterarguments.

I really wish the forum-monster hadn't ate the last big thread on this, because it addressed all these issues and with the math to back it up.

As to me, I'm about to get off work so I'm done with forums for the day.

Edited May 9, 2013 by RoboRay

[AmpsterMan]

I always tend to think of Asparagus staging as compounding the maximal rocket equation to smaller segments of time (t) - ie: you want to reduce as mass as quickly as possible (hence dumping it as soon as it's done) whilst keeping constant (or increasing) thrust. Thus I guess the idealised solution would be having an infinite number of infinitely small tanks which weighed an infinitesimal amount, st at each delta t, these tanks were discarded as they were emptied. Obviously this is impossible, and hence delta t is longer between drops, but the idea is the same.

Maybe I'm overcomplicating things...

I've always thought this was the reason that radial fuel flow was more efficient. It makes a lot of sense really.

lately I have only been using a Falcon Heavy approach to radial staging and it works well for most instances.

[Anglave]

Ya, I realized that at the time, but basicly the thinking was this: If i figure out how many solid rocket boosters a solid rocket booster can lift, then while at launch that first solid rocket booster does nothing but neutralize the 'weight' of the other solid rocket boosters... but now i have solid rocket boosters higher up, which will contribute to canceling out the effect of dropping mainsails with the asparagus.

You'e still using boosters to lift boosters, rather than a more efficient thrust source. Basically, because of their high thrust but low efficiency, boosters are only valuable in your first stage; getting off the pad. In virtually every other application, you'd be better off lifting more engine or more fuel.

[Eric S]

I always tend to think of Asparagus staging as compounding the maximal rocket equation to smaller segments of time (t) - ie: you want to reduce as mass as quickly as possible (hence dumping it as soon as it's done) whilst keeping constant (or increasing) thrust.

Right. single-stack staging can't do this because they more stages they add, the more dead engine weight they're pushing, so it hits diminishing returns from staging much faster than asparagus staging does. Heck, I've got one Asparagus design that's ten stages including the center stage but not the stage on top of the center stage. Not a practical vehicle, but it was designed for a "go really fast" type of challenge.

I really wish the forum-monster hadn't ate the last big thread on this, because it addressed all these issues and with the math to back it up.

Which isn't to say that it didn't overlook something, but it's more likely that I did, and I really want to know what I did.

[Pbhead]

You'e still using boosters to lift boosters, rather than a more efficient thrust source. Basically, because of their high thrust but low efficiency, boosters are only valuable in your first stage; getting off the pad. In virtually every other application, you'd be better off lifting more engine or more fuel.

Weight to payload ratio is important... but, for me at least, payload to lag ratio is more important.

Also... Is it possible to asparagus ion engines? I cant seem to figure it out...

[Temstar]

Weight to payload ratio is important... but, for me at least, payload to lag ratio is more important.

In that case you still should use asparagus staging, since the highest "delta-V/part" ratio parts are big orange tanks and mainsail, and the most efficient way to use these two is asparagus staging.

[TheDarkStar]

Also... Is it possible to asparagus ion engines? I cant seem to figure it out...

Like RCS, xenon gas (ion engine fuel) does not have liquid fuel flow mechanics. If it's anywhere on the craft, a tank can give fuel to an ion engine. So no, you can't asparagus stage it very well.

[magnemoe]

Weight to payload ratio is important... but, for me at least, payload to lag ratio is more important.

Also... Is it possible to asparagus ion engines? I cant seem to figure it out...

Yes you can, downside is that you have to transfer fuel manually to inner stage and drop the outer. However its pretty pointless. most of the mass of an ion probe is engines and solar panels and you main problem is low twr, you might want to drop empty tanks, going close to the sun you might also want to drop unneeded solar panels.

[Akalaël]

I figured that I could use asparagus style for interplanetary trips.

Pic of my hily modded save :

8 orange tank (only one attached for the moment), feeding the core of the space ship, then drop 2 by 2 during the trip...

I did not try the system yet. Do you think it will be efficient ?

[Eric S]

Yes you can, downside is that you have to transfer fuel manually to inner stage and drop the outer. However its pretty pointless. most of the mass of an ion probe is engines and solar panels and you main problem is low twr, you might want to drop empty tanks, going close to the sun you might also want to drop unneeded solar panels.

An alternative is to disable all fuel tanks except those you plan on dropping next. Taking 15 seconds to turn on new tanks probably won't significantly affect hour long burns.

[magnemoe]

An alternative is to disable all fuel tanks except those you plan on dropping next. Taking 15 seconds to turn on new tanks probably won't significantly affect hour long burns.

An point, however as the empty fuel tanks for xenon is less than half the weight of the smallest decopler I don't think this is so practical.

Using larger ion systems from mods and it start making more sense.

[Plur303]

It's not used in real-life because conservation of angular momentum would result in an ever-increasing, unstoppable roll of the craft as the fuel spiraled inward to feed the core stage.

And moving that much fuel from tank to tank to tank while feeding engines consuming vast quantities of it is pretty much an impossible engineering challenge.

Not to mention that disconnecting fuel lines while engines are burning is a rather dangerous activity.

The drag induced by an asparagus setup isn't any greater in real-life than any other lateral booster arrangement, and those are used extensively.

There are other momentum effects that impact it. And the other factors I mentioned.

If it was "really easy to fix" I'm pretty sure NASA would be doing it.

The new Falcon Heavy from SapceX is using fuel crossfeeding to pump fuel to the center stage and expend the outer 2 stages first.

[Nimaci]

The new Falcon Heavy from SapceX is using fuel crossfeeding to pump fuel to the center stage and expend the outer 2 stages first.

To avoid a lot of the technical issues the F9Hs boosters will feed propellant directly to the engines of the core stage (as well as their own). The initial concept was for each booster to feed the 3 core stage engines closest to it while the core stage fed its centre 3 itself, but with the new octagonal engine configuration for the F9v1.1 I'm not sure how this plan has changed, but I'm guessing the boosters will be feeding 4 core stage engines each, leaving only a single core engine to be fed by the core until separation.

Elon will gain more grey hairs before the first F9H test flight, and a few more during it

[numerobis]

Also... Is it possible to asparagus ion engines? I cant seem to figure it out...

IIRC you can tweak a setting in the resources file to make xenon flow in stacks like liquid fuel rather than simultaneously from everything like electricity. This came up in a challenge to get the minimum solar apoapsis, which requires a 28 km/s capacity.

[Plur303]

IIRC you can tweak a setting in the resources file to make xenon flow in stacks like liquid fuel rather than simultaneously from everything like electricity. This came up in a challenge to get the minimum solar apoapsis, which requires a 28 km/s capacity.

It's far easier to just disable flow on tanks you don't want to consume. Considering how slowly Xenon is used this should not be a very big pain.

[RoboRay]

The new Falcon Heavy from SapceX is using fuel crossfeeding to pump fuel to the center stage and expend the outer 2 stages first.

The Falcon Heavy will do limited cross-feeding directly to some (not all) of the core engines... it's not an asparagus-stalk booster design.

But yeah, it's the first to actually attempt anything resembling this kind of stuff in flight. We'll have to wait and see if they manage to pull it off.

[ThePyrateCaptain]

Shouldnt be called asparagus staging, more like wile e coyote rocketry.

[RoboRay]

It was named "asparagus" long before the KSP community discovered it. Ed Keith is credited with the name, which is used in "Orbital Mechanics: Theory and Applications" by Tom Logsdon in 1997 to refer to the technique.

Edited May 15, 2013 by RoboRay

[Temstar]

Ed Keith is credited with the name, which is used in "Orbital Mechanics: Theory and Applications" by Tom Logsdon in 1997 to refer to the technique.

Well that's a shame, since Mikhail Tikhonravov came up with it first in 1947, he even designed a R-3 derived launch vehicle using the idea:

Too bad everyone except Korolev thought he was crazy. Korolev took on his idea of "packet rocket" (but not the cross feeding) and hence the now familiar shape of the R-7 / Soyuz rocket family's first stage.

Edited May 15, 2013 by Temstar

[RoboRay]

Well that's a shame, since Mikhail Tikhonravov came up with it first in 1947

Yep. All he needed was to give it a flashy name and make a hilarious yet educational video game where the rules gloss over its inherent problems.

[AngelLestat]

You only need to know some basics.

The effiency of a rocket depends on 4 factors:

-on the efficiency (ISP) of the engines that you use at different heights.

-the drag that you get taking in count your speed at low altitudes.

-the amount of death mass you carry in each stage.

-the monetary cost of all the fuel and parts that you cant recover.

The best way to not carry death mass is drop tanks, if they are attaching at side you cant decoupling 1 at the time that will be the best case, to keep balance you do it by 2, but 2 orange tank simultaneously in some cases is not the best, so to improve this you can use smallers tanks with only 1 big engine like in this picture.

Beneficts: you dont carry extra engines to drop it later.

Cons: you consume fuel very fast so the isp is not the best.

This craft is just an example for death mass effiency and nothing more.

[madoxlabs]

Not knowing a lot about rocket design myself, I have been struggling with creating proper efficient rockets. It always bothered me that, with 'normal' SaturnV type staging, you throw away engines each time and need to carry new engines in each stage. I've been playing with throwing away parts from the top down instead of bottom up to save on engines.

Don't know if its good or not yet though. There is a lot of great info in this thread for me to work with.