Controlling huge rockets

[Hyperspace Industries]

I'm starting to occasionally build huge 5m rockets which thrust vectoring can't control, and vernor thrusters are never aligned perfectly and can't control roll, so I end up spinning like a top.

What do you recommend?

Are there any helpful mods I can use for bigger liquid fuel oxidizer rcs, or should I just pack monoprop and tweakscale some rcs thrusters?

[KerikBalm]

2 hours ago, Hyperspace Industries said:

I'm starting to occasionally build huge 5m rockets which thrust vectoring can't control?

If the engines can lift the stack, their thrust vectoring can control the stack...

I don't see how scaling up the stack results in a problem, are you sure you aren't just building them in an unstable way?

[Streetwind]

I'm with @KerikBalm here. There's no reason large rockets should be less stable than small ones. If you're having issues, you're probably building it in a way that invites stability issues at any size.

Perhaps you could share a screenshot of your launch vehicle, and we can take a look at it? Or tell us what you need lifted, and we can tell you how we'd design the lifter.

Some rules of the thumb you can follow to improve the stability of your rockets:

• Heavy stuff at the front, draggy stuff at the back (AKA the golden rule of rocketry). Following this rule gives you what is called passive stability. A passively stable rocket is one that wants to fly perfectly straight without any fins and any gimbals to help it. You should always aim for passive stability in anything you build. In an ideal world, fins are an unnecessary crutch, and engine gimbals are meant to force a passively stable rocket to change directions, not to keep an unstable one straight.
• If you're just on the edge of instability, you can employ fuel tank priority settings to make your fuel tanks drain bottom-first, thus shifting the CoM upwards durign flight and leaning into the rule above.
• Fins may be a crutch, but if you must use fins, mount four of them (not three, not six - exactly four) and align them to the cardinal directions. If you also have side boosters, mount them 45° offset instead. The boosters don't care (at least when they're just boosters without control authority). The fins do.
• Don't overdo control authority. If you have SAS on and your rocket oscillates more and more, you have too much authority from gimbals and/or active steering fins, and they are fighting each other into a feedback loop. Turn them down or switch them off.
• Try for a single-stack design. KSP's radial decouplers are bendy things, and will cause engines mounted on side boosters to thrust off center ever so slightly. It's not so bad when they are used just for thrust, as the imparted forces will cancel each other out between mirrored boosters. But as soon as you try to use them for control authority, you have a potential recipe for things to go haywire.
• If you must use side boosters or multiple cores, don't mount fins on them, especially not active steering fins; and turn off all gimbals on them.
• Try for a single-size design. Don't change diameter along the length of the rocket, except to go from smaller top to larger bottom. The most stable designs avoid even that. The biggest sin is going from large to small and then large again; this will not only cause extra drag, but also introduces a pair of weak(er) joints that is liable to wobble. If you wish to mount undersized engines in the middle of a larger stack, use Making History's engine plates, or surface attach them while building an interstage with the help of a fairing base.
• Try for smooth stacks. Don't mount anything on the outside that doesn't absolutely need to be there during atmospheric flight. Payload bays, interstage fairings, and even structural tubes let you hide whatever you like inside the smooth stack.
• Avoid part clipping. It works in 99% of all cases, and in the other 1%, it induces phantom forces that make your rocket shake, bend, and flip. Murphy's Law dictates that you are the 1%.
• Try for sane acceleration values. A surface take-off TWR of 1.4 to 1.5 is recommended. Not only do such rockets need less control authority to steer, but they also experience much lower maximum dynamic pressures while transsonic than high-TWR designs.
• Fly smooth, even turns with the help of SAS Hold Prograde. Once you're past 2km in altitude or so and happy with your heading, you shouldn't need to touch anything but spacebar and your throttle control to make it into orbit. Every manual control input is a potential source for instability.

You may not always be able to follow all of these principles, but at minimum you should be aware of them, and compromise on them knowingly, rather than accidentally.

I've never played with autostruts. Not once. My rockets almost never have fins, and their gimbals are turned down to 25% or locked outright. They still fly straight and true, and don't wobble or drift around randomly. That is the power of proper construction.

 

Edited January 6, 2022 by Streetwind

[Jetpackmaniac]

In Stratzenblitz75’s spin launcher video he used a mod that allowed him to use stock liquid fuel engines as RCS thrusters. Might have to get a bit creative to get roll control but it could work. Unfortunately I can’t remember the name of the mod right off and I don’t have time to go back and watch the video. 

[Geonovast]

1 hour ago, Hyperspace Industries said:

and vernor thrusters are never aligned perfectly and can't control roll,

Make sure angle snap is on when you place them, and simply offset them so their COT isn't through the middle of the rocket, and you can control roll with them just fine.

 

[Hyperspace Industries]

23 minutes ago, KerikBalm said:

If the engines can lift the stack, their thrust vectoring can control the stack...

I don't see how scaling up the stack results in a problem, are you sure you aren't just building them in an unstable way?

It's likely I might be. Probably that and too low twr.

[not giving a name]

 

2 hours ago, Jetpackmaniac said:

In Stratzenblitz75’s spin launcher video he used a mod that allowed him to use stock liquid fuel engines as RCS thrusters. Might have to get a bit creative to get roll control but it could work. Unfortunately I can’t remember the name of the mod right off and I don’t have time to go back and watch the video. 

the mode used in the video is throttle control avionics

[Snark]

Moving to Gameplay Questions.

[Axelord FTW]

2 hours ago, not giving a name said:

 

the mode used in the video is throttle control avionics

The KAL1000 can achieve the same result, kinda. It can't be controlled by SAS and MJ, for that matter, but you can get HUGE torque by manual inputs.

[Hyperspace Industries]

18 minutes ago, Streetwind said:

I'm with @KerikBalm here. There's no reason large rockets should be less stable than small ones. If you're having issues, you're probably building it in a way that invites stability issues at any size.

Perhaps you could share a screenshot of your launch vehicle, and we can take a look at it? Or tell us what you need lifted, and we can tell you how we'd design the lifter.

Some rules of the thumb you can follow to improve the stability of your rockets:

• Heavy stuff at the front, draggy stuff at the back (AKA the golden rule of rocketry). Following this rule gives you what is called passive stability. A passively stable rocket is one that wants to fly perfectly straight without any fins and any gimbals to help it. You should always aim for passive stability in anything you build. In an ideal world, fins are an unnecessary crutch, and engine gimbals are meant to force a passively stable rocket to change directions, not to keep an unstable one straight.
• If you're just on the edge of instability, you can employ fuel tank priority settings to make your fuel tanks drain bottom-first, thus shifting the CoM upwards durign flight and leaning into the rule above.
• Fins may be a crutch, but if you must use fins, mount four of them (not three, not six - exactly four) and align them to the cardinal directions. If you also have side boosters, mount them 45° offset instead. The boosters don't care (at least when they're just boosters without control authority). The fins do.
• Don't overdo control authority. If you have SAS on and your rocket oscillates more and more, you have too much authority from gimbals and/or active steering fins, and they are fighting each other into a feedback loop. Turn them down or switch them off.
• Try for a single-stack design. KSP's radial decouplers are bendy things, and will cause engines mounted on side boosters to thrust off center ever so slightly. It's not so bad when they are used just for thrust, as the imparted forces will cancel each other out between mirrored boosters. But as soon as you try to use them for control authority, you have a potential recipe for things to go haywire.
• If you must use side boosters or multiple cores, don't mount fins on them, especially not active steering fins; and turn off all gimbals on them.
• Try for a single-size design. Don't change diameter along the length of the rocket, except to go from smaller top to larger bottom. The most stable designs avoid even that. The biggest sin is going from large to small and then large again; this will not only cause extra drag, but also introduces a pair of weak(er) joints that is liable to wobble. If you wish to mount undersized engines in the middle of a larger stack, use Making History's engine plates, or surface attach them while building an interstage with the help of a fairing base.
• Try for smooth stacks. Don't mount anything on the outside that doesn't absolutely need to be there during atmospheric flight. Payload bays, interstage fairings, and even structural tubes let you hide whatever you like inside the smooth stack.
• Avoid part clipping. It works in 99% of all cases, and in the other 1%, it induces phantom forces that make your rocket shake, bend, and flip. Murphy's Law dictates that you are the 1%.
• Try for sane acceleration values. A surface take-off TWR of 1.4 to 1.5 is recommended. Not only do such rockets need less control authority to steer, but they also experience much lower maximum dynamic pressures while transsonic than high-TWR designs.
• Fly smooth, even turns with the help of SAS Hold Prograde. Once you're past 2km in altitude or so and happy with your heading, you shouldn't need to touch anything but spacebar and your throttle control to make it into orbit. Every manual control input is a potential source for instability.

You may not always be able to follow all of these principles, but at minimum you should be aware of them, and compromise on them knowingly, rather than accidentally.

I've never played with autostruts. Not once. My rockets almost never have fins, and their gimbals are turned down to 25% or locked outright. They still fly straight and true, and don't wobble or drift around randomly. That is the power of proper construction.

 

Thanks! Yeah, it's mostly a few designs that aren't working, I'll go make some better ones.