E. F. Kranz Posted February 14, 2012 Share Posted February 14, 2012 Alright, so I\'ve been struggling with this since I\'ve been constructing larger rockets - how in the hell to fly them!?!The problem is: making the turn.Depending on the rocket, I might begin to roll over to the horizon at 20km, or 40km, etc.I prefer a nice, smooth, controlled transition.What I get is absolute chaos.With larger (i.e. - longer) rockets, the horizon roll is basically impossible. As soon as I tip it over from vertical, the thing goes into an insane head over heels tumbling. No (reasonable) amount of RCS thrusters can keep it under control. I end up either throttling down and fighting it out, or ditching the offending stage before it\'s depleted.So I ask you, rocket scientists. What causes this, and how do I avoid it?Is it a weight distribution thing? (i.e. - lots of empty tanks on top) Is it a matter of speed? (i.e. - velocity, momentum, all that good stuff) Or am I just a terrible pilot? Link to comment Share on other sites More sharing options...
Kosmo-not Posted February 14, 2012 Share Posted February 14, 2012 Put some vectored thrust LFEs on it. This situation can be caused by winglets high up on the rocket. Best thing to do is nose over very gradually. Link to comment Share on other sites More sharing options...
Tosh Posted February 14, 2012 Share Posted February 14, 2012 More likely it\'s well-known \'wobbling\' that cause this. Ya see, large rockets are rather \'soft\' and do bend like crazy . I found no way to avoid this tumbling when I flew heavy rockets. I ended up adding lots of winglets at the bottom and installing ASAS to control them; and performing most of maneuvers in lower atmosphere while winglets are still effective. Link to comment Share on other sites More sharing options...
Kosmo-not Posted February 14, 2012 Share Posted February 14, 2012 Tosh said: More likely it\'s well-known \'wobbling\' that cause this. Ya see, large rockets are rather \'soft\' and do bend like crazy . I found no way to avoid this tumbling when I flew heavy rockets. I ended up adding lots of winglets at the bottom and installing ASAS to control them; and performing most of maneuvers in lower atmosphere while winglets are still effective.Sounds like you need to install some struts. Link to comment Share on other sites More sharing options...
Xivios Posted February 14, 2012 Share Posted February 14, 2012 Struts, gimballing engines, symmetry, SAS, ASAS - with devices that can use it (moving fins, RCS, gimballing engines), and a tapered drag form - by which I mean, the bottom of the rocket suffers from more drag than the top - will all contribute to safe, stable flight. Well, as safe and stable as can be expected from trash cans filled with boom and Jeb at the helm. Link to comment Share on other sites More sharing options...
Tosh Posted February 14, 2012 Share Posted February 14, 2012 Kosmo-not said: Sounds like you need to install some struts.I did. The rockets I flew would just fell apart without \'em =P Link to comment Share on other sites More sharing options...
foamyesque Posted February 14, 2012 Share Posted February 14, 2012 It\'s the drag model, which gives empty tanks less drag. On multistage liquids, this messes you up pretty badly. I would recommend winglets at the bottom as a general-purpose solution to tumbling. Link to comment Share on other sites More sharing options...
E. F. Kranz Posted February 14, 2012 Author Share Posted February 14, 2012 Thanks for your input, people.It\'s not wobble. I don\'t use solid boosters, and I always have a mix of gimballed and fixed engines.I use control surfaces, mostly biased towards the bottom of the stack, but with canards on the top. RCS is abundant.Here\'s my thoughts:I suspect this is a physics/design problem, as opposed to a flight control problem. I just don\'t know enough to know whether the game is accounting for these things.Consider:At launch, a well conceived stack is akin to an arrow. Of more or less equal distribution of weight; the tail drag accounting for the payload (arrow) at the end.Unlike an arrow, however, my rocket drains weight at a more or less constant rate, beginning from top to bottom.So at altitude, having exhausted - let\'s say - 60% of its fuel, the rocket is now more akin to a bowling pin. Heavily weight-biased towards the bottom, virtually no weight in the middle, and somewhat heavy again at the top. When turned horizontal (that is, perpendicular to gravity), the bottom section of this bowling pin carries more kinetic energy, causing the structure to begin a tumble in that direction.Proposed solution:A fuel distribution system that draws fuel at an equal rate from both the top at the bottom and delivers it to the vehicle\'s center of gravity.[EDIT] - Actually, since you\'ll be pumping fuel from the bottom towards the center (to the distribution manifold) and from the center towards the bottom (to the engine), you\'ll need to pump fuel from the bottom even faster. Link to comment Share on other sites More sharing options...
Kosmo-not Posted February 14, 2012 Share Posted February 14, 2012 Could you post your craft file here? We could then take a look at it and come back with some good critiques. Link to comment Share on other sites More sharing options...
Zephram Kerman Posted February 14, 2012 Share Posted February 14, 2012 foamyesque said: It\'s the drag model, which gives empty tanks less drag. On multistage liquids, this messes you up pretty badly. I would recommend winglets at the bottom as a general-purpose solution to tumbling.What I did was to cluster one boost stage around the top stage, ignite the top stage engine at liftoff, and pump fuel into it until the boost stage is empty. That way, all engines are running at liftoff and none are wasted. It also has the effect of reducing weight near the top of the craft, because the stages are basically lateral instead of vertical. It sortof works, usually-ish. Link to comment Share on other sites More sharing options...
khyron42 Posted February 23, 2012 Share Posted February 23, 2012 In real-world rocket design, the key to a rocket being stable in atmosphere is to have Center of (drag) Pressure lower than Center of Gravity. That makes the aerodynamic resistance act as a stabilizer instead of trying to flip the rocket over.With KSP\'s drag model, just building a rocket that tapers towards the top does this fairly well most of the time by keeping CoG from shifting downwards too much, while giving a very bottom-of-rocket CoP due to larger numbers of parts clustered at the bottom. You just have to be high enough when you shed those lower-rocket parts that drag isn\'t much of an issue anymore. Link to comment Share on other sites More sharing options...
Recommended Posts