Quadcoupler + Staging

[Pupu]

Hello, i got some problem with constructing rockets with bi-couplers or quad-couplers. I want to get four nuclear engines on the tank and use the quadcoupler. I cant put the decoupler directly on the 4 engines (for staging). So i use another quadcoupler at the end of the four engines and attach the decoupler to that quadcoupler.

see image:

So it looks weird cause only one engine seems to be attached to the 2nd quadcoupler. At launch the quadcoupler breaks and i think its cause it isnt fully attached. Am I right or is it normal that the other three engines are visible?

[nadreck]

You need the smaller decoupler under each engine then the lower upside down quadcoupler

[TMS]

Yeah, lots of problems with the node system doing what you want to do. You can't do a 1-2-and-2-1 node attachment (or a 1-4-and-4-1 in your case). Each node can only have one parent, so it won't attach properly. Only one of your four parts will be attached at both ends.

Not quite sure where your decouplers are in that shot. Even if you put decouplers at the bottom of each, you'll also have a problem when they blow their shrouds into each other and destroy your engines. The joys of KSP. You can get around that with a quicksave and reload.

Also, four NERVAS are fuel inefficient for most loads.

[LtKraftKrackers]

You need the smaller decoupler under each engine then the lower upside down quadcoupler

i've actually tried this before. what ends up happening it that the Nervas fairings are expelled with such force that they destroy the other engines. really, if these werent nervas on your image, i would agree, just place decouplers on the bottom of each engine and then place the quad adapter. at the moment of staging they should just all slide off. but the fairings on the Nervas are expelled with such force that the engines are destroyed.

[Monkeh]

What you need to do is rotate the engines so the seams of the fairings are all pointing outwards. Add 4 symmetry decouplers to the bottom and then the adaptor on those.

You can also use docking ports as well, they'll attach at all 4 points when the ship 'drops' on the launchpad.

Good luck.

Edited November 4, 2013 by Monkeh

[nadreck]

I had success with it in 0.21.1 and managed to launch 4 landers with that sort of arrangement - haven't tried it in .22 as I now sling my Nerva's out on all four sides of my landers. I may be trying it again at some point again to send a fuel depot to Joolian space.

[Pupu]

Ah thx for the advices with the 4 docking ports and decouplers. dont know why i put the quadcoupler direct at the engine. maybe i should go to bed

i used 4 atomic engines, because the burntime gets too long. and so its less accurate. but yeah maybe 2 engines are enough.

[Monkeh]

There also seems to be a certain lack of fuel for them to burn as well

Long burn time accuracy can be improved by burning half the estimated time in front of the node and half behind. So an estimated burn of 10 minutes should be started at t -5minutes and finished at t +5minutes.

[J.Random]

I would say: scratch this design altogether. If my math is correct, your last stage weighs some 30 tons, give or take, and 9 tons of this weight are engines.

[capi3101]

I'll tell you right now - you do need to go ahead and add the decouplers and arrange the LV-N sheaths like everybody's saying. You then need to add a clamp-o-tron to the bottom of the decouplers, and then pick up construction with four more clamp-o-trons below those (so that the top of the rocket is "docked" to the bottom part).

"Why would I do that, capi?", says you. Well, it has to do with the branching nature of .craft files. Long and the short of it is that you can't form loops and expect your rocket to behave the way you want - you can't branch off into four separate stacks (which is what the quad-adapter does) and then later rejoin those stacks. What winds up happening is that the .craft file will ignore three of the reconnects, you wind up with a wobbly rocket (think insufficient strutting) and you can look forward to seeing a nice big explosion when you finally decouple. Like this one:

No actual damage to the craft in this case, but seeing this will still make you jump a bit.

The clamp-o-tron trick is the only way I know to get around this; basically, the game treats the lower half of the rocket as a separate craft. When the whole thing is loaded on the pad and the physics engine releases, the docking ports automatically clamp together with a rock-solid connection, just as they would with a multi-port docking setup. When you're ready to utilize the stage, you just hit the decouplers like normal; the docked docking ports slough off with the rest of the lower stage and you can go on your merry way.

Provided you aim those sheaths the right way, of course......

Actually, I've pretty much given up on the notion of using quad-adapters in my designs (they'd be a lot more useful if you could form loops with them). What I do instead is stick the NERVAs on the ends of BZ-52 Radial Attachment points, then use a long girder segment or two and an adapter to attach it to the rest of the rocket. Like this:

With sufficient strutting, this will stay attached to a booster (quite well, actually). You also avoid headaches with the sheaths that way, since you make the connection by the girder and not by the engines themselves.

Edited November 4, 2013 by capi3101

[Pupu]

Ok i abondoned the idea with the quadcoupler and attached 2 of the enngines radial to the main fuel tank:

Added some more boosters.

Now i get into kerbin orbit ( i bet its not very fuel efficent ) without using fuel from the atomic engines. i ll try that quad-port-coupling next time ;-) .

[Superfluous J]

Actually, I've pretty much given up on the notion of using quad-adapters in my designs (they'd be a lot more useful if you could form loops with them). What I do instead is stick the NERVAs on the ends of BZ-52 Radial Attachment points, then use a long girder segment or two and an adapter to attach it to the rest of the rocket. Like this:

I get the general concept but don't see the purpose of using BZ-52s and you covered them with fuel tanks in your image. Could you post another with just the bare essentials of your design?

[Pupu]

i like that idea using the girder segment to bypass the engines. If its stable enough at launch its really way more easier.

[capi3101]

I'll have to try to take a better screenshot when I get an opportunity. Incidentally, that setup (an X200-8 in the center and six X200-8s outboard) gives 1,939.29 m/s of delta-V for an 86-tonne payload. The TWR isn't too bad; 0.26 or thereabouts. Still not great, but what can you expect from NERVAs?

The BZ-52s work kinda like Tail Connectors; they allow fuel flow and allow you to define symmetrical attachment points easily. The big advantage of the BZ-52 over a tail connector (aside from being a smaller part) is that's it's mass is a tenth of the mass of a Tail Connector (always a good thing).

[Smidge204]

I believe I managed to figure out capi's secret recipe:

Pretty efficient, especially when you consider that the X200-8 holds twice the fuel as the four FL-T100s which would be normally used in a quad stack for the same height. Yay part reduction! Of course there's no reason why you can't use something other than LV-Ns.

=Smidge=

[capi3101]

Beat me there...I'd even prepared a nice little step-by-step imgur album for it. Behold the power of the BZ-52...

Javascript is disabled. View full album

I'll point out with my specific design that had I used an X200-16 in the center, with the six outboard X200-8s there's the same amount of fuel as an orange fuel tank. Much more compact, much more structurally sound IMHO.

For those of you who may doubt that little thing is structurally stable, I invite you to look at my entries in the Konstellation Space Program challenge. I put a 65 tonne lander on top of this, a huge booster below it, and it makes orbit fine (with that particular booster, it's never this part that fails).

[kirkmegna]

You could also just build your centre stack using four FL-T800 tanks. Since each holds 360L, having 2 stacks of 4 is the same as having 1 big orange tank.

[Giggleplex777]

This: http://forum.kerbalspaceprogram.com/threads/43621-How-to-have-engine-clusters-in-upper-stages