How to create four parallel linkages?

[jwatte]

I want to use four parallel engines from a big fuel tank, using a one-to-four adapter.

Then I want to join them up again into the big tank below them.

However, I can only get one engine to "click" into the socket, and the others just sit there.

The end result is a very weak coupling and the whole thing collapses on the launchpad. (Even if I add a lot of steel bars)

How can I make all four of the engines "click" into the adapter at the same time? If that can't be done, what's a good alternative?

[Roastduck]

The first main problem I see is that you have no decouplers. Even that one linkage will never separate, except maybe by accident (structural failure). Try putting a 1.25m decoupler on the bottom of each engine, then the quad adapter *should* attach to all of them

[Superfluous J]

Short answer: You can't.

Long answer: Yoooooooooooooooou Caaaaaaaaaaaaaaan't.

Longer answer: The tree structure for craft files makes it so you can never have a closed loop. Your best bet is to run a girder down the middle of those engines and connect that way. Then use struts so your rocket doesn't snap in half.

And as RoastDuck says, you need a decoupler too

[Master Tao]

Are you certain 5thHorseman? Try this technique by Kasuha from a couple weeks ago:

Okay, you got me here. I was experimenting with it when I started using Imgur and direct upload to album never worked reliably for me. So I settled with uploading the list, then converting it to album using a button on the screen. But for some reason it stopped working (at least for me) some time ago. Apparently the functionality of these two approaches was switched somehow.

Thanks for pointing it out, I'd probably never consider trying that rejected method again.

http://imgur.com/a/gd7EY

[Murph]

In short, it can't really be done, and I believe that it's a fundamental limitation in the Unity physics engine. Even with decouplers added below the engines, only 1 decoupler will be attached to the lower multi-coupler, resulting in a very weak and off-center linkage. You might get away with it by generous use of struts around the weakness, but I'd just avoid that type of design in the first place.

The good alternative is just to use a single larger engine, and only use the multi-couple pieces facing downwards (usually only at the very bottom of a stack).

[sumghai]

The first main problem I see is that you have no decouplers. Even that one linkage will never separate, except maybe by accident (structural failure). Try putting a 1.25m decoupler on the bottom of each engine, then the quad adapter *should* attach to all of them

You've correctly surmised that the OP is missing decouplers between stages, however, decouplers alone won't solve the issue.

The key thing to note is that KSP uses a tree hierachy to "build" its vessels, such that each part can only have *one* point of attachments - the parallel design the OP is trying to build simply isn't possible.

My only suggestion is for the OP to consider an alternative upper stage design that does not use multi-point connections.

[Box of Stardust]

Actually, there is a technique. Learned it over in the Stock Construction Techniques Repository (or whatever it's called). If you're going to use decouplers (I assume you would), place the decouplers under the engines. Then under those, place docking ports. Then, under a single engine, connect another docking port (in the docking position, naturally). Latch your coupler under that one while lining it up with the other three engines. Then place the docking ports on the bottom coupler.

The docking ports will be all docked upon loading. Hope this helps!

[Superfluous J]

While clever, that whole "use docking ports" is kludgy and promotes wobbly rockets that are more expensive (which matters now, at least somewhat). The central-column + struts is simple, effective, and cheap.

[Box of Stardust]

I assumed that from the number of Kerbodyne tanks, cost wouldn't really be much of a factor. And if cost wasn't rebalanced again, it doesn't cost all too much if I remember correctly. I was specifically taking a look at how much this technique would cost earlier with .24 on the horizon at the time.

I've used the technique myself, and will admit it does require strutting the stage to the lower stage. But only a few well placed struts. Or depending on the rocket.

... But in the case of OP, yeah... I think just doing a central-column might be better. Looking at it again, Kerbodyne tanks have a lot of room to symmetrically place the 909s. And with that much space, there's definitely more than enough room to put, say, a tiny decoupler in the middle and just connect down with an I-beam.

Although I can't imagine why four LV-909s instead of just a Poodle. That would solve everything.

Edited July 18, 2014 by Box of Stardust

[Superfluous J]

Although I can't imagine why four LV-909s instead of just a Poodle. That would solve everything.

That question is being begged quite strongly. I was looking at the problem itself and not the craft. OP, why ARE you using 4 LV-909s? They're not the worst engine out there but they're not so great you'd want to go through all this effort to stack them so. I've only put this work into LV-Ns and 48-7Ss.

[Claw]

The docking port trick is about the only way to directly connect all four.

Another technique is to use cubic struts to radially mount the 909s. Then you can use a larger decoupler to connect the two stages directly. It won't look quite the same, but it's also a popular way to build.

[Oafman]

Yeah not sure what you have above that in the picture, but you definitely have a 3.75m tank at least, so I'm doubtful that 4 x 909s is going to make it move very quickly.

Even if that engine setup suits your needs, a permanent goal to keep in mind when you're in the VAB is reducing your parts count, so if there's a situation where one part can do a similar job to many parts, you should always go with the one. And a Poodle would mean none of these complicated adaptor solutions are required. Just a Poodle, and a decoupler, and you're good to go

Oh, and some struts. Lots of struts.

[jwatte]

Even with decouplers added below the engines, only 1 decoupler will be attached to the lower multi-coupler

Yes that's what I'm seeing. Thanks everyone for verifying it isn't possible.

The reason there were no de-couplers was that it's easier to illustrate the problem -- only one motor will be sheathed, showing that there's only one connection point. If I put four de-couplers on, four motors will be sheathed, but it still doesn't snap in more than one location, and still collapses on the launchpad, except the cause is much less obvious to see.

As for answering the higher-level problem: I actually want to do this with the nukes, but as they were blowing up on launchpad, I figured I'd get a less volatile engine working first, before going back to that.

I had already tried strutting -- eight "200 I-beams" in turn clad with M-2x2 structural panels, but that didn't brace it enough.

I admire the cleverness of the docking port design, but I've found docking ports to be even wobblier than decouplers.

In general, the physics engine is more wobbly than "real life" as far as I can tell. Mild Steel has a shear strength of at least 50,000 pounds per square inch; alloy steels much higher. The connecting area of even a small decoupler or other ring connection would be in the hundreds of square inches, but they behave like bubblegum and duct tape. I think this is because KSP uses the default 32-bit game physics engine of Unity?

[Claw]

As for answering the higher-level problem: I actually want to do this with the nukes, but as they were blowing up on launchpad, I figured I'd get a less volatile engine working first, before going back to that.

The other problem you'll run into using nukes in a quad configuration is making sure that the engine shrouds don't destroy the engines. You have to rotate the engines so that the shells of the shroud sort of point at each other. That why when they come off, they destroy each other instead of the engines/ship.

~Claw

[Box of Stardust]

In that case, if you're going to use LV-Ns, you're going to have to use the 'manual coupling' technique, the one where you place cubic octagonal struts symmetrically + radially underneath the tank. Flip them over to clip inside the tank if you want. In any case it'll give you nodes to attach your engines to (and fuel feeds normally), plus the center node will be open to attach a decoupler to, which you should extend down with an I-beam or similar part. It'll require vertical struts between the two stages to keep it stable, but it'll work.

[Superfluous J]

I will again suggest using 4 cubic octagonal struts to "radially" connect your LV-Ns to the bottom of your rocket, and then a small decoupler in the middle between them all, attached to a girder that sticks down to the stack below. I think it looks just fine and the only thing you're missing is shrouds. Needs a couple struts to keep it from wobbling but what rocket doesn't?

[jwatte]

How do I attach the octagonal struts to the bottom of the tank? As opposed to the regular I-beams, they don't seem to want to stick (or multiply by the N-way selector.)

[jwatte]

I appreciate your advice. I'm trying to build that now, and I'm still having some problems.

First, I realized that the cubic octagonal strut does attach to the sides like an I-beam, but the regular octagonal strut does not. Confusing!

Second, I can't figure out how to properly strap so that it actually "glues" to both ends. I have six I-beams from the big tanks below sticking up (, and I attach six radial decouplers and six I-beams from above (A), and make them meet in the middle. Turn on the MIG and jam it on there, and it would weld them just fine :-) However, on the launchpad, these two different I-beams just cheerfully glide through each other, with no bonding action at all. Same thing if I try girders in the same configuration.

So, how do I make the strapping "stick" in this configuration? Ideally, with some way of releasing it when done with the bottom monster stage?

[Superfluous J]

I appreciate your advice. I'm trying to build that now, and I'm still having some problems.

First, I realized that the cubic octagonal strut does attach to the sides like an I-beam, but the regular octagonal strut does not. Confusing!

Second, I can't figure out how to properly strap so that it actually "glues" to both ends. I have six I-beams from the big tanks below sticking up (, and I attach six radial decouplers and six I-beams from above (A), and make them meet in the middle. Turn on the MIG and jam it on there, and it would weld them just fine :-) However, on the launchpad, these two different I-beams just cheerfully glide through each other, with no bonding action at all. Same thing if I try girders in the same configuration.

You're making it too hard. All you need is the central column and a few struts (the ones that are like bars that can be any length). Here's an actual ship I sent (it's a refueller) to Jool recently. It didn't shake at all:

[Blind Dead McJones]

I built a rocket exactly like that and it works fine:

Notice the decouplers. Put those on with symmetry enabled before you add anything else and you should be fine for making the connection. You'll definitely need struts to keep it from breaking in half, though.

The decouplers tend to explode rather than just decouple, but it doesn't damage the payload. Oh, and I built this one during 23.5 and it worked then. I tested it again in .24 which is where I got those pics.

So yes, this is possible, or at least it's not impossible.

Edited July 19, 2014 by Blind Dead McJones

[jwatte]

Ah, the straps with little fasteny bits, instead of I-beams. I'll try that!

[Box of Stardust]

You realize that makes the point of decoupler moot though, right?

[J3ansley]

I regularly do what Blind Dead McJones does. The trick is to go back to single symmetry when you place the upside down four to one adapter.

I generally don't bother with making sure the 4 nukes are oriented correctly so they always do damage when activated due to the ejection of the shroud. But it seems to not affect anything.