SpaceCraft Spinning Uncontrollably!

[MattyAB]

Hey Everyone,

So I know there are already posts about this kind of thing, but most of them are fairly old and none of the solutions in them work for me

My spacecraft has 3 engine stages: 4 Thumpers, then 4 T-30s and finally 1 more T-30. The craft flies fine in the first stage, only needing a little amount of adjustment to keep it centred, but as soon as the second stage starts, the Craft spins uncontrollably. No amount of counter-thrust or fins can solve it. Here is a screenshot of the Craft at the moment.

How can I stop this and keep my Craft under control?

Thanks!

[Snark]

What's your ascent profile like? Are you going too fast while you're still in atmosphere? A few things to be aware of:

• The amount of drag shoots up dramatically when you pass Mach 1. Need to avoid going a lot faster than 300ish m/s until you're above 10km. If your rocket is aerodynamically unstable, then it will get dramatically harder to control when it starts going really fast.
  
• Your rocket is short and wide. That means the fins aren't placed very far behind, which will limit their effectiveness. A taller, skinnier rocket will allow the fins to get more leverage.
  
• If you're using T-30 engines, they don't have any gimbal ability, which means they're not helping to steer the ship. Consider switching to T-45s, they can gimbal and will really help steer the ship (at least while it's under thrust).
  
• As far as I can tell from looking at your ship, you don't have any reaction wheels. Combined with your lack of gimbaling engines and steerable fins, that means that the only torque authority on your ship is the Mk1 command pod up top, which is puny. You simply have way too little torque to steer something huge like that ship.
  
• You're making your life difficult by building tall stacks of 1-ton tanks. Reason: Each engine is pulling fuel from its tank stack. It will start pulling from the topmost tank first, then when that goes dry, it starts draining the one below that, and so forth. This means that it's losing the forwardmost mass first, which means that as it burns fuel, the CoM is moving rapidly backwards on the ship. Pretty soon this will move your CoM behind your center of drag, which will make the ship aerodynamically unstable, and hilarity ensues. What you need to do is to replace that stack of 1-ton tanks with a single 4-ton tank. It will be the same size and shape as the stack-of-4, but it drains its mass uniformly rather than from the top down, so you won't have that stability problem.
  

A side comment on your design: You appear to be trying to use 2 sets of radial decouplers for each stage; for example, you appear to be trying to hold on each Thumper with two radial decouplers, one positioned above the other. KSP doesn't work like that. Except for struts and fuel ducts, every part can only have one attachment point. When you put multiple decouplers on there like that, the thing being attached (e.g. Thumper) is actually being held by only one of the decouplers. The other decoupler is just sitting there and isn't actually connecting anything to anything; it's nothing but dead weight and drag, you can delete it. I assume that you did this to try to make the rocket stiffer; if that's what you want, use a strut.

• 
  

Edited September 1, 2015 by Snark

[Warzouz]

True. Try to build more vertically. It might wobble a bit, but you'll gain in stability. You even have fins quite high, which is not too good either.

I'm used to build rockets I can't even grab the top of it in the VAB.

PS : Looking at you design, you want to go to LKO, not farther (or maybe doing some Mun flyby ?). In that case, your ship is overbuilt. Use smaller boosters which pushes a stacked 2 stage rocket.

Your heat shield is misplaced and unusefull, and I don't know why you added some heat management parts on you command pod. It's only needed when using Nukes. If you can EVA, put the stack decoupler right under the command pod (and EVA to grab your science). If you can't, add 2 radial chutes to help landing your science equipement safely. You also may need a battery

EDIT : Ascent profile is also important. SAS may be doing worse then without. There ar many videos which explains how todo it. It's quite hard to do, but when you get it, flying is easy. SAS shouldn't be activated before 30km, especially if ship is wobbleing. But on light ships like yours it's less important than on big ones.

Also remember that flying small ships can be harder than bigger ones. Light ships tends to flip quickly on first trouble as bigger ones turn slowly so you have time to compensate before it's too late.

Edited September 1, 2015 by Warzouz

[ghpstage]

It looks like its flipping when you drop the SRB stage because the drag on the radial liquid stages is way too high up, almost certainly putting the centre of drag in front of the Centre of mass. When this happens you get an unstable rocket that, left to its own devices, will flip over and fly backwards.

Also, you want to swap at least the central, if not all of the LV-T30s with LV-T45s. The gimbals they come with will amaze you, swap them all out and they should even be able to handle the instability of this design without other changes.

In future to provide more useful screenshots, include CoM and CoL showing for each stage in the VAB, and one at the point you lose control during flight.

Edited September 1, 2015 by ghpstage

[Stoney3K]

What's your ascent profile like? Are you going too fast while you're still in atmosphere? A few things to be aware of:You're making your life difficult by building tall stacks of 1-ton tanks. Reason: Each engine is pulling fuel from its tank stack. It will start pulling from the topmost tank first, then when that goes dry, it starts draining the one below that, and so forth. This means that it's losing the forwardmostmass first, which means that as it burns fuel, the CoM is moving rapidly backwards on the ship. Pretty soon this will move your CoM behind your center of drag, which will make the ship aerodynamically unstable, and hilarity ensues. What you need to do is to replace that stack of 1-ton tanks with a single 4-ton tank. It will be the same size and shape as the stack-of-4, but it drains its mass uniformly rather than from the top down, so you won't have that stability problem.

The CoM of a stack of 1-ton tanks an and a 4-ton tank are exactly the same. A 4-ton tank will also drain top to bottom and has the same amount of mass (full or empty) as a stack of 4 1-ton tanks.

Check it for yourself in the VAB, the dry CoM of both setups as well as the wet CoM should be exactly identical.

In real life it should NOT be and the big tank would drain more uniformly, but in KSP, this is the case.

Having at least 2 separate tanks is even beneficial because you can transfer fuel back up when one tank is drained.

As far as the spin thrust is concerned: Try shutting off your engines and see if you can stabilize your craft. If that is the case, your outboard liquid boosters are producing torque, probably because they are not in line with the core radial. You may have placed them on the decouplers with a slight offset.

Edited September 1, 2015 by Stoney3K

[tater]

It doesn't look like a rocket, for one. It's too short. The fins are not actually helping. Why the radiators on the capsule? You don't need so many engines, they are mass you are lifting for no gain.

You want the CM higher. Think about a dart, weighted at the nose. If you put the weight on the back, it will flip.

[jiminator]

ditch all the fins, there should only be 1 set as low as possible on the main rocket.

2: fix the symmetry. grab one 2 second stage at the collection point, pull it off. delete the others. use pair symmetry to place it once and then a second time at 90 degrees. use fuel lines to connect the first set to the second set, and again to connect the second set to the main stack. congrats, asparagus staging.

arrange everything to fire at once, then the boosters to drop second stage, then the 1st asparagus set to drop 3rd stage then the final asparagus stage to drop 4th stage.

the main stack also should be split up into stages. 2/3 for the bottom, and a small fuel efficient engine for the top 1/3

[Kelderek]

The CoM of a stack of 1-ton tanks an and a 4-ton tank are exactly the same. A 4-ton tank will also drain top to bottom and has the same amount of mass (full or empty) as a stack of 4 1-ton tanks.

...

Having at least 2 separate tanks is even beneficial because you can transfer fuel back up when one tank is drained.

Looking at his design, I'm guessing this is early career and probably doesn't have the bigger tanks yet, and probably can't transfer fuel yet either. Engine choices may be limited too.

[Snark]

The CoM of a stack of 1-ton tanks an and a 4-ton tank are exactly the same.

Correct, if all the tanks are full. And also correct, if all the tanks are empty.

A 4-ton tank will also drain top to bottom and has the same amount of mass (full or empty) as a stack of 4 1-ton tanks.

Yes, it has the same mass. But it does not drain "top to bottom". KSP does not model the mass distribution inside a partially full tank-- for example, a half-empty 4-ton tank doesn't have the fuel sitting in the bottom half of the tank. The center of mass of each individual tank is always in the center of that tank.

So when you have a 4-ton tank, its CoM stays in the same place as it drains.

But when you have a stack of four 1-ton tanks, this is not the case, because they don't all drain at the same time. First the top tank drains, then the next one down, and so forth. So the CoM of the stack moves downwards as the stack drains, unlike with the 4-ton tank.

Check it for yourself in the VAB, the dry CoM of both setups as well as the wet CoM should be exactly identical.

Yes, that's true for dry and it's true for wet, but it's not true for half-full.

[Stoney3K]

Yes, it has the same mass. But it does not drain "top to bottom". KSP does not model the mass distribution inside a partially full tank-- for example, a half-empty 4-ton tank doesn't have the fuel sitting in the bottom half of the tank. The center of mass of each individual tank is always in the center of that tank.

So when you have a 4-ton tank, its CoM stays in the same place as it drains.

True, but that has no effect on the craft CoM, which still moves down linearly as the mass of either of the tanks drain. There's still a heavy engine at the bottom which moves the center of mass down as anything on top of it becomes lighter.

Build a craft in the VAB with a single FL-T800 tank loaded for 50% with fuel, and an engine at the bottom. Check the CoM. Next, build one with four FL-T100's that are full and four that are empty. The CoM should be in the same place (even though real rockets will have the CoM higher on a single tank).

Yes, that's true for dry and it's true for wet, but it's not true for half-full.

Both of these points are the most important factors for craft stability. What happens in between is largely unimportant, unless it's a discrete step when a tank is empty (which it definitely is not).

[tbarcello]

True, but that has no effect on the craft CoM, which still moves down linearly as the mass of either of the tanks drain. There's still a heavy engine at the bottom which moves the center of mass down as anything on top of it becomes lighter.

Build a craft in the VAB with a single FL-T800 tank loaded for 50% with fuel, and an engine at the bottom. Check the CoM. Next, build one with four FL-T100's that are full and four that are empty. The CoM should be in the same place (even though real rockets will have the CoM higher on a single tank).

Both of these points are the most important factors for craft stability. What happens in between is largely unimportant, unless it's a discrete step when a tank is empty (which it definitely is not).

Should i draw you a picture to demonstrate how CoM moves up, when radial tanks with engines connected far below the next stage of a craft?

If u insist...

http://imgur.com/rodtkr9

Blue is the rest of craft, the next stage after radials, which are red with green heavy engines.

When u attach radials to the CoM of next stage, then you can empty them, and the CoM will move down till staged.

Edited September 2, 2015 by tbarcello

[Stoney3K]

I was not referring to radial tanks, I was just referring to the mass distribution of a single tank during ascent.

The CoM will always move towards the heaviest part of the craft -- if you have a core stage that is full with radial tanks underneath that are draining, it will move up because the heaviest part becomes the core stage.

It doesn't matter whether the core stage or the radials are made up of a single tank or out of separate tanks.

[tbarcello]

I was not referring to radial tanks, I was just referring to the mass distribution of a single tank during ascent.

The CoM will always move towards the heaviest part of the craft -- if you have a core stage that is full with radial tanks underneath that are draining, it will move up because the heaviest part becomes the core stage.

It doesn't matter whether the core stage or the radials are made up of a single tank or out of separate tanks.

Oh yeah?

Continue education then...

http://imgur.com/iX0vYja

On top is single tank half full.

On bottom 4 small tanks, 2 empty.

Imagine green is your center of lift.

Jeb will tell you whether it matters later. Or Jeb's widow...

[Stoney3K]

Oh yeah?

Continue education then...

http://imgur.com/iX0vYja

On top is single tank half full.

On bottom 4 small tanks, 2 empty.

Imagine green is your center of lift.

Jeb will tell you whether it matters later. Or Jeb's widow...

You might want to make some more decent drawings (or, screenshots) that may get your point across better.

I tested it: Yes, a stack of tiny tanks drains downward faster than a single tank, reaching its peak of CoM aft when the bottom-most tank is still full, and from there, the CoM will shift back up because the command pod is heavier than the empty weight of the bottom tanks.

This is also what would happen in real life if the LiquidFuel and Oxidizer were to be stored in a uniform mix. However, they are not, for that exact reason which really messes up your rocket's stability (if your center of mass moves behind the center of drag, flip flop, game over).

There is a way to mitigate this in the early stages of the game though: Store LiquidFuel and Oxidizer as separate tanks like in the real world. It'll cost you some tankage weight and make a bigger rocket, but when your bottom tanks are beginning to drain, the center of mass will be BETWEEN them, not near the center of the bottom tank.

[Snark]

True, but that has no effect on the craft CoM, which still moves down linearly as the mass of either of the tanks drain.

Except that it does. The CoM of the overall craft is the vector sum of "mass * position" of all the components in it.

That is, suppose you want to find the vertical position (i.e. on the y axis) of a craft's center of mass, and it consists of n components each of which has a mass m and a vertical position y for its own CoM, then you can find it thus:

yCoM = (Σ mn*yn) / (Σmn)

What this means is:

1. If you move the CoM of any part of the whole, it moves the CoM of the whole.

2. If you change the mass of any part of the whole, it moves the CoM of the whole (unless that part's CoM is exactly on the CoM of the whole)'

There's still a heavy engine at the bottom which moves the center of mass down as anything on top of it becomes lighter.

Sure, the engine moves the CoM downwards. But that's beside the point. The point is, a half-drained stack of 1-ton tanks (with two empty tanks sitting on top of two full ones) will have a lower CoM than a single half-drained 4-ton tank does. And therefore, a craft containing the former will have a lower CoM than a craft containing the latter, irrespective of whatever other mass there is. Of course, how much lower it is will depend on the rest of the craft. But the effect is not zero.

Build a craft in the VAB with a single FL-T800 tank loaded for 50% with fuel, and an engine at the bottom. Check the CoM. Next, build one with four FL-T100's that are full and four that are empty. The CoM should be in the same place (even though real rockets will have the CoM higher on a single tank).

Nope. If you put the four full FL-T100s on the bottom and the four empty on the top, it will have a lower CoM than the 50%-full FL-T800.

...Update: Went ahead and verified it in-game. Here's a screenshot. The FL-T800 at left is 50% full. The assembly at right has four empty FL-T100 tanks on top of four full ones.

...note that the one on the right has a significantly lower CoM, despite the fact that the stack of FL-T100's is actually significantly taller than the lone FL-T800.

Edited September 3, 2015 by Snark

[tbarcello]

You might want to make some more decent drawings (or, screenshots) that may get your point across better.

I tested it: Yes, a stack of tiny tanks drains downward faster than a single tank, reaching its peak of CoM aft when the bottom-most tank is still full, and from there, the CoM will shift back up because the command pod is heavier than the empty weight of the bottom tanks.

This is also what would happen in real life if the LiquidFuel and Oxidizer were to be stored in a uniform mix. However, they are not, for that exact reason which really messes up your rocket's stability (if your center of mass moves behind the center of drag, flip flop, game over).

There is a way to mitigate this in the early stages of the game though: Store LiquidFuel and Oxidizer as separate tanks like in the real world. It'll cost you some tankage weight and make a bigger rocket, but when your bottom tanks are beginning to drain, the center of mass will be BETWEEN them, not near the center of the bottom tank.

Let's keep real life for real life and KSP for KSP. If you can't understand my drawings, then, probably, I shouldn't even try to convince you, because wth would I do that, right? Either sit on those until you understand, or at least ask questions, that will help you understand, or just read the comment from Snark. I don't argue with you, I'm educating you. If you want to believe the earth is flat - not much i can or want to do. I mean jesus!

[Claw]

Before you guys hijack the thread further with debates...

The way KSP models fuel tanks is indeed the fact that the fuel mass is distributed equally within the fuel tank. The CoM of a given fuel tank does not shift downward as fuel drains. This is a separate issue from where the CoM of the craft moves, which will be dependent upon the craft configuration.

That being said, the above posted picture, repeated here, is the truth of the matter with the lego style fuel tanks in KSP. An even better way to demonstrate this is to do the same thing without the engine attached, since having an attached engine changes the results.

Plop down a single FL-T800 and tweak the fuel up and down. Then compare to a stack of FL-T100s emptying from top to bottom, and where the CoM is lowest. This is a typical cause of early rocket instability (in KSP), and why fins are often needed.

There are many ways around this (some of which are already stated), but I request that you guys focus on helping the new player with his or her specific problem, rather than getting into technical debates inside a request for help. (There's this forum, Development Discussion, and even Science Labs for technical discussion.)

Cheers,

~Claw

[tbarcello]

Sorry, I guess that's my fault, KSP gets me too excited sometimes. I really believe, that if I'll play Call of Duty someday I'll be hospitalized.