Asymetric design stablity

[lazarus1024]

Okay, I've built some mighty big interplantary ships in my day (not to boost or anything). Probably the bigger ones were in the 400-500t range.

The issue I have is balancing them. Even using bicouplers and docking collars getting Orange tanks to dock up to each other perfectly symetrically with both collars engaging has proven to be hard. All of my designs have been pusher designs for ALL of my craft with the exception being that sometimes my landers use puller/skycrane designs for A) landing base modules [always], descent stages on landers (rarely, but it happens) or C) Ascent stages on landers (often).

Would a puller design balance an asymetrically put together ship better than a pusher design would??? Or would it still have crazy spin/torque?

[superm18]

much better ballance!

i used pusher design in the past, but now i'm all pulling

[Rascal Nag]

The idea that pulling something rather than pushing it will increase stability is a common misconception. At least if the pulled object is rigidly attached to whatever is pulling it. And since we don't have ropes, this means pulled rockets in KSP will always have the same problems as a pushed one. And even if we did have ropes, it would not be wise to use them in rocket assembly. What if you're towing an object via rope in order to avoid CoM issues, but then you slow down? Suddenly a piece of your own rocket is accelerating towards you, which won't end well.

Wikipedia actually has an article dedicated to this very idea:

http://en.wikipedia.org/wiki/Pendulum_rocket_fallacy

[bsalis]

\At least if the pulled object is rigidly attached to whatever is pulling it. And since we don't have ropes, this means pulled rockets in KSP will always have the same problems as a pushed one.

Problem is, the docking ports are not that rigid. At least the regular ones.

[superm18]

sr. docking port are rigid and works very good when pulling

[Akalaël]

I prefer pulling than pushing, but seems the same in the facts.

An other surprising thing : I saw and tested some sideway docking, it works actually pretty well (don't know the mass you can put actually...)

So now, I use the tree ways : pull, push and sideway at once :

But the fact rest the same : if you're unbalanced, it's just impossible !

[Promii]

Pendulum rocket fallacy only applies if your craft is rigid. Pushing a big train of docked modules can be like trying to push a string.

[lazarus1024]

Yeah I think my biggest issue is that my design is typically 3 orange tanks with 2.5m NERVAs on the end of each. One center, two side tanks with each docked to the other in seperate launches. The side tanks get jettisoned once down to 1/3rd fuel remaining in the tanks (then all is transfered to the center tanks and the side tanks and engines are kicked loose).

I think I need to go back to my prior design of a seperate drive section of NERVA engines (1.5m jobs since I am back to all stock), but maybe with the fuel tanks either docked out from a node connector up front on seperatrons (and/or spin up the ship before seperating tanks) or else a semi-pusher design. From front to rear of mission section, drive section, fuel section. Maybe with a few of the engines of decouplers as well so that once I burn through some of my fuel and eject tanks I can eject engines to balance TWR and increase efficiency.

[Tokay Gris]

And even if we did have ropes, it would not be wise to use them in rocket assembly. What if you're towing an object via rope in order to avoid CoM issues, but then you slow down? Suddenly a piece of your own rocket is accelerating towards you, which won't end well.

Ehm... That would be a funny design for a rocket.

How do you slow down? You turn around and blast, so you are still stuck with pulling. Or pushing.

Pulling doesn't change to pushing and vice versa.

Unless you use engines in both directions.