Side Boosters Destroying Main Engines

[Xylenox]

Hello,

I have a problem with my side boosters. Whenever they detach, they always seem to destroy one of the main engines. How can I prevent this? Thanks in advance.

 

 

[Snark]

1 hour ago, Xylenox said:

I have a problem with my side boosters. Whenever they detach, they always seem to destroy one of the main engines. How can I prevent this?

What this means is that you've got two problems:

1. Something (likely aerodynamic force) is causing them to go inwards when ejecting, rather than harmlessly outwards which is what you want.
2. It's taking a long time for your center core to move far enough forwards, relative to the ejected boosters, that it is no longer in danger of collision.

It would help if you could post a screenshot of your ship-- that would let us give you specific advice, i.e. "oh, you're doing this!  Change that thing to be this way instead", you get the idea.

However, in the absence of a screenshot, we can offer some general advice.  There are various things you can do to help the problem.

1. Mount your radial decouplers as high as possible on the radial booster.  Or, put another way:  after you've placed the decouplers on your central core, you want to place the booster as low down on the decoupler as you can, so that the decoupler is near the front of the booster.  Specifically, you want the attachment point to be above the radial booster's center of mass.  Why is that?  Because that way, when the decoupler fires, its ejection force will tend to make the booster rotate outwards (because it's kicking the booster's nose outwards), which will then cause aero forces to move the booster away from your central core.  It's possible, for example, that your current design might have the decoupler attached below the booster's center of mass, which is bad-- that' because it's kicking the back of the booster outwards, which points the booster inwards, which makes aero forces slam it against your central core, which is bad.
2. Mount the radial boosters as low as possible on the central core.  In general, what happens when you eject radial boosters is that they start to fall behind your craft.  That's because they're no longer thrusting, but your central core is thrusting.  If you can mount the boosters really low on your central core, it means that the central core doesn't have to travel very far before it "clears" them, i.e. it gets far enough ahead so that the back of the central core is in front of the forward tip of the boosters.  By that time, the boosters are behind you, so it doesn't matter if they move inwards and smash into each other.  Think of it as a "race":  the radial boosters move inwards to smash your central core, but your central core moves forward to get out of the way.  By mounting the radial boosters nice and low, you're stacking the deck in favor of your central core "winning the race" and getting out in front of them before they have a chance to wreak mischief.
3. Use a more heavy-duty radial decoupler.  If you have a big booster, you may want to consider using the Hydraulic Detachment Manifold instead of a more lightweight decoupler like the TT-38K.  The bigger decoupler is heavier, yes, but it also has a stronger ejection force and does a better job of kicking the booster outwards.
4. Use sepratrons.  These are little surface-mountable tiny SRBs.  You can attach them to your radial boosters in strategic locations, and set them up in the staging UI so that they activate at the same time you activate the radial decouplers.  Set them up so that their thrust moves the front end of the boosters away from the central core.
5. If the boosters are big, add tiny fins on the front end, angled slightly outwards, so that the aero force on the fin will pull the nose of the ejected booster outwards and away from the central core.

I find that in my own gameplay, 99% of the time it's sufficient just do do #1 and/or #2 above.    The latter options also can work, but I generally only use them in odd "edge cases" (i.e. unusual ship designs) where the first two don't do the trick.

[Capt. Hunt]

if you try number one, make sure you're using the dry COM rather then the COM when loaded with solid fuel.  You won't be jettisoning those boosters until they're all used up.  I usually just use #4, lots and lots of sepatrons.  Make sure you have rockets on both ends of the booster for balance.  Set the bottom rockets to have less then 1 unit of fuel, so they just have enough kick to clear, but give the top rockets just a hair more fuel so they angle the boosters away.

[A_name]

As usual, @Snark has completely killed the thread with his thorough response. I will add however a note of caution that imparting too much outward rotation to the boosters can also cause the butt end to hit your central stack, especially on longer boosters that have the center of mass further from the bottom. Because of this, my go-to method is usually to use separatrons to try and have the boosters gain distance from the central stack while reducing rotation in any direction to the minimum, so that the boosters maintain parallel to the stack as long as possible. Aero forces will eventually win, but hopefully by that time the stack is far away enough so that it no longer matters.

 

Edited February 21, 2018 by A_name

[Geonovast]

Just now, A_name said:

As usual, @Snark has completely killed the thread with his thorough response.

I've often wondered if his keyboard has heatsinks.

[Not Sure]

2 hours ago, Snark said:

If the boosters are big, add tiny fins on the front end, angled slightly outwards, so that the aero force on the fin will pull the nose of the ejected booster outwards and away from the central core.

Snark on top of it every time lol. But this method is my favorite. The cheapest fins on the top of the booster will do wonders at a super low cost.

[Xylenox]

4 hours ago, Snark said:

Mount your radial decouplers as high as possible on the radial booster.

Thanks! This one works like a charm!

[Spricigo]

4 hours ago, Snark said:

Mount your radial decouplers as high as possible on the radial booster.  Or, put another way:  after you've placed the decouplers on your central core, you want to place the booster as low down on the decoupler as you can, so that the decoupler is near the front of the booster.  Specifically, you want the attachment point to be above the radial booster's center of mass.  Why is that?  Because that way, when the decoupler fires, its ejection force will tend to make the booster rotate outwards (because it's kicking the booster's nose outwards), which will then cause aero forces to move the booster away from your central core.  It's possible, for example, that your current design might have the decoupler attached below the booster's center of mass, which is bad-- that' because it's kicking the back of the booster outwards, which points the booster inwards, which makes aero forces slam it against your central core, which is bad.

I get that the idea there is to point the booster outwards. Since the tendency is that after the separation the aerodynamic forces will force the booster to move that way.

However, we don't want the booster too rotate to much, otherwise, while  indeed the top part rotate outward, the bottom part rotate inwards and may hit the core. Thus, My suggestion is that the attachment point need to be above the CoM of the booster, but not "as high as possible" just "slightly".

 

IME works like a charm.

Edited February 21, 2018 by Spricigo

[Kryxal]

I usually use the spaced decouplers and one sepratron mounted so it thrusts up and a bit in, works pretty well.  Actually, once I have it working well, I turn it into a subassembly, and put a strut from the nosecone (which is what the sepratron is actually mounted on) to the main stack.

[Kerbal101]

Noisecones  Nosecones play some role too. Some of them may cause air to push the booster tops away from the center. Its all about rotation

Edited February 22, 2018 by Kerbal101
too much noise in my head :P

[Snark]

3 hours ago, A_name said:

I will add however a note of caution that imparting too much outward rotation to the boosters can also cause the butt end to hit your central stack, especially on longer boosters that have the center of mass further from the bottom.

48 minutes ago, Spricigo said:

However, we don't want the booster too rotate to much, otherwise, while  indeed the top part rotate outward, the bottom part rotate inwards and may hit the core.

Yah, that can happen.  Depending on how the ship is built, though, it may or may not matter.  For example, that bit of advice about "mount the boosters as low as possible on the central core"?  For me, that generally means that the bottom half of the boosters are sticking down below the bottom of the central core.  My radial decouplers are mounted right at the very bottom of the central stack, and the boosters are mounted as low as possible on that.

So, it doesn't actually matter if the butt end gets kicked inwards, because there's nothing in the middle down there but empty space.

In short, my rockets never get their butts kicked. 

(sorry, couldn't resist)

[Geonovast]

27 minutes ago, Kerbal101 said:

Noisecones play some role too

Do noisecones still work if you play with KSP muted?

25 minutes ago, Snark said:

(sorry, couldn't resist)

 

[Kerbal101]

@Geonovast That was supersonic.

[mikey117]

One thing I tend to do when separating boosters, is throttle down quite a bit before the boosters run out of fuel, then as soon as I stage the decouplers ramp up to full throttle, tends to stop any collisions when boosters don't get pushed away too much. 

Also make sure you are bang in the middle of the prograde marker when decoupling, generally wont destroy the engine, but I've lost a few fins due to this.

[CSE]

It can help, too, if you design the side boosters to burn for almost as long as the core stage, so that your rocket is at high altitude when you stage the boosters away. That way you're in very thin air and have much lower aerodynamic forces to cope with. It's difficult to do this with KSP solid boosters, which have very short burn times, but easier with a Falcon / Delta Heavy style where the  boosters are very similar to the core. Definitely consider this if you're playing in an up-scaled solar system.

On 21/02/2018 at 9:58 PM, Snark said:

My radial decouplers are mounted right at the very bottom of the central stack, and the boosters are mounted as low as possible on that.

Ariane 5's boosters are mounted quite a long way back on the core, or the core tapers quite early, and I wonder if it's partly this.