Attach things to multiple couplers / mounts / pylons

[UncleManuel]

Hi there!

Yes, another newbie-question (aaand I was drawn into KSP because of Scott Manley and Tim Dodd)...  

I'm using the sandbox mode with all available parts and experiment with different things to get a feel for the game mechanics. One thing bugs me: structural stability of attached boosters. Obviously external things only attach to the selected coupler/mount/pylon and you have to stabilize them with at least 1 strut. But is there a way to actually connect f.e. a very long SRB to two pylons for structural stability? 

 

Bonus question: how do I safely eject multiple boosters (or to be more precise: NOT to collide with my rocket)? Especially the very long SRBs tend to tilt inwards after decoupling...

[AHHans]

Hi, welcome to the forums.

1 hour ago, UncleManuel said:

But is there a way to actually connect f.e. a very long SRB to two pylons for structural stability?

No.

...

O.K. O.K. I'll (try to?) explain that a bit: the way that crafts are modeled internally means that every part can have only exactly one upstream (in the direction of the root part) connection / part. I.e. from every part there can be only one path to the root part. Connecting an SRB via two decouplers (or pylons or whatever) to the same central core would mean that that there are two paths from the SRB to the root part, so that is not possible. The struts are more a variation on that rule than a violation of it: they are attached to one part, and then they point in a certain direction and transfer forces to the first part that they find.

If you don't like the looks of the normal struts, then you can try using autostruts. (You need to activate "Advanced Tweakables" in the settings if you haven't done so already.) If you do so then I suggest to use "autostrut to grantparent", the other settings can be dangerous under certain conditions. But in general, there is a reason why "more struts" is the got-to solution to many KSP design problems.

1 hour ago, UncleManuel said:

Bonus question: how do I safely eject multiple boosters (or to be more precise: NOT to collide with my rocket)? Especially the very long SRBs tend to tilt inwards after decoupling...

I usually first try to place the decoupler in a way so that it tilts the top of the SRBs outwards so that the airflow will push the SRBs away from the core. Often I also attach small airfoils (typically the "Basic Fin") to give the air some more area to push against. A "simpler", more reliable but also more expensive method is to attach sepratrons to the SRBs that will push them away from the core when staged.

[Foxster]

A simple way to get a booster to peel away is to mount the decoupler closer to the top than the middle - say 2/3 or 3/4 of the way up the booster. This tends to throw the top out on separation and the booster clears the craft's main body. 

You can place the decoupler, attach the booster, translate the decoupler up and then the booster back down. 

Oh and you will almost certainly then need a strut at the bottom. 

Edited November 16, 2019 by Foxster

[UncleManuel]

On 11/15/2019 at 9:45 PM, AHHans said:

No.

...

O.K. O.K. I'll (try to?) explain that a bit: the way that crafts are modeled internally means that every part can have only exactly one upstream (in the direction of the root part) connection / part. I.e. from every part there can be only one path to the root part. Connecting an SRB via two decouplers (or pylons or whatever) to the same central core would mean that that there are two paths from the SRB to the root part, so that is not possible. The struts are more a variation on that rule than a violation of it: they are attached to one part, and then they point in a certain direction and transfer forces to the first part that they find.

Ah ok, I understand. So it's basically a programming limitation of the physics engine. Makes sense now... 

And I've already discovered that you can switch tools and MOVE already attached parts. Very handy in case of the positioning of boosters. But for the heavy boosters the ejection force still is not enough in some flight profiles. I'll experiment with some options - it's called "sandbox" for a reason...

 

[Gargamel]

If you are having issues getting your boosters to clear your main core, slap a couple sepatrons near the nose cone of the booster.   One on each side of the booster, pointing towards the main core.  Have them fire when you stage the decouplers.    Might require a little tweaking to get the intended behavior, but they will push the boosters away from the core.

[Snark]

On 11/17/2019 at 10:28 AM, UncleManuel said:

But for the heavy boosters the ejection force still is not enough in some flight profiles.

Yeah, the decouplers only have a limited amount of ejection force, and it doesn't scale up well-- especially for atmospheric flight (e.g. the lower parts of the ascent phase), where aerodynamic forces tend to be very large, and dwarf any contribution from the decoupler.

Sepratrons can be good for this.  Another option is to mount some small fins (e.g. Basic Fin) on the nose of the radial boosters, angled slightly outwards so that when the booster is released, aero force on the fins will pull the booster nose away from the central core, and then aero forces do your work for you by moving the booster rapidly away from your central stack.

[Fierce Wolf]

21 hours ago, Snark said:

Sepratrons can be good for this

Place 2 sepratron on each booster (in the middle of it), pointing at your vessel. Be careful staging those.

 

And there! no more worries about accidentally destroying your engines. Plus: they look professional

[UncleManuel]

54 minutes ago, Fierce Wolf said:

Plus: they look professional

Isn't that the whole purpose of KSP? 

 

And yes, even one sepatron at 45° (nose cone) is enough to safely separate even the heavy S3 boosters with the Mammoth engine pack... 

 

And I've got the hang of the autostrut feature. Works well for keeping even heavy things stable - yes, I'm really using S3 tanks with the Mammoth engine pack as boosters... 

[Snark]

3 hours ago, Fierce Wolf said:

Place 2 sepratron on each booster (in the middle of it), pointing at your vessel. Be careful staging those.

This can work.  However, I'd actually suggest placing the sepratrons on the front of the booster, rather than the middle.

That way, they'll tend to rotate the booster by shoving its nose away from the central core.  This is really handy, because as soon as the booster is angled away from the central stack, aero forces will hit the side of the booster, generating outward "lift" and very quickly shoving the booster outwards. It allows even a small sepratron to move a really big booster out of the way:  aero forces are big.  This aerodynamic effect can be much stronger than just the thrust of the sepratron itself.  It's a really handy "force multiplier".

(This only applies if you're still in a fair amount of atmosphere at separation time, of course.  If you're already well over 25 km when you separate, then the aero forces are so small that it won't make that much of a difference.)

The other advantage of putting the sepratrons up at the nose of the booster is that you can usually get by with just one per booster, rather than needing a pair-- put it on the nose, so that its exhaust goes over the top.

(One thing to be aware of, with sepratrons, is that they have an exhaust that's hot enough to be destructive.  This has the potential to actually destroy the central core in some circumstances.  It's pretty short-range, though, so the main thing is just to be careful not to mount the sepratrons too close to the center stack.)

[UncleManuel]

On 11/19/2019 at 10:12 PM, Snark said:

It allows even a small sepratron to move a really big booster out of the way:  aero forces are big.  This aerodynamic effect can be much stronger than just the thrust of the sepratron itself.  It's a really handy "force multiplier".

I haven't thought of the aero drag during separation yet. Interesting. I've only noticed it when I oversteered my MK3 spaceliner (spaceshuttle clone) and the whole thing tumbles because the wings and/or the huge fuselage is catching aero forces... 

But as I said, one sepatron on the nosecone of the booster pointing at the vessel is mostly enough to create an outward momentum. I usually separate when the boosters are empty (separation and next stage ignition in one go)... 

[Snark]

1 hour ago, UncleManuel said:

I haven't thought of the aero drag during separation yet. Interesting. I've only noticed it when I oversteered my MK3 spaceliner (spaceshuttle clone) and the whole thing tumbles because the wings and/or the huge fuselage is catching aero forces... 

The other nice thing about it is that if you happen to have fourfold radial symmetry on your boosters, this technique makes a really nice Korolev cross. 

[TanDeeJay]

On 11/16/2019 at 6:45 AM, AHHans said:

O.K. O.K. I'll (try to?) explain that a bit: the way that crafts are modeled internally means that every part can have only exactly one upstream (in the direction of the root part) connection / part. I.e. from every part there can be only one path to the root part. Connecting an SRB via two decouplers (or pylons or whatever) to the same central core would mean that that there are two paths from the SRB to the root part, so that is not possible. The struts are more a variation on that rule than a violation of it: they are attached to one part, and then they point in a certain direction and transfer forces to the first part that they find.

I did a bit of an experiment to test this.  While it is not possible to have multiple connection points as the OP asked when constructing in the VAB/SPH, I have found that if you have 2 ships that have 2 docking ports each, and you are careful in how you place those docking ports, then when in space (or on the ground...) if you align both pairs of docking ports on both ships correctly, then when you dock, both pairs of docking ports dock.   You have to undock both pairs before you can separate.  I'm not sure how the game deals with the grandparent/root relationship in this situation though.  My son had an idea of using an arrangement like this for increasing the structural integrity of space stations.

[AHHans]

1 hour ago, TanDeeJay said:

I'm not sure how the game deals with the grandparent/root relationship in this situation though.

In one of the change-logs of the more recent patches (i.e one of the 1.7.x or 1.8.x patches) they mentioned something about secondary docking ports becoming primary docking ports when the primary is undocked. So I think that the first pair of docking ports to make contact will become the primary thus defining the root and grandparent relationships, and the other docking ports become secondary and are probably more implemented like (auto-)struts. (But I can also be wrong.)

[TanDeeJay]

On 11/22/2019 at 11:39 PM, AHHans said:

In one of the change-logs of the more recent patches (i.e one of the 1.7.x or 1.8.x patches) they mentioned something about secondary docking ports becoming primary docking ports when the primary is undocked. So I think that the first pair of docking ports to make contact will become the primary thus defining the root and grandparent relationships, and the other docking ports become secondary and are probably more implemented like (auto-)struts. (But I can also be wrong.)

That sounds logical.  I tried launching a multi-fueltank module for my spacestation with 4-way radial symetrical docking ports at the top and bottom.  Each fuel tank was individually launchable including RCS with thrusters, reaction wheels, and a RC unit and with Snr docking ports either end to plug it into my space station.  I had 3 launched and connected in an L shape then I launched a 4th tank and tried to dock it with 2 ports on one of the existing tanks and 2 ports on the other to complete a square.  not quite sure what happened, but only 3 pairs of the ports connected.  the 4th pair seemed to be slightly mis-aligned, so I used the dock-rotate mod to try and straigten things out, but I moved things too far and I managed to get the ports hooked on their outside edge, kind of jambed... I kept trying to fix things with dock-rotate, but I eventually discovered that one of the other (previously docked) ports had become detached from its fuel tank, and was just floating in space.

I came to the conclusion it is either a bad design arrangement, or I had the wrong approach when docking.  I docked the usual way of aligning 2 ports using the Docking port alignment indicator mod, which meant the 2 ports on the side which were supposed to dock with the other tank on the side were sliding together sideways, which was when they initally got the side edge collision that probably pushed things out of alignment. perhaps I should have come in such that the ports approached at a 45 degree angle.

[StrandedonEarth]

One other tip with separating boosters: it helps to have the bottom of the boosters slightly below the bottom of the core, to prevent a bottom strike as the booster rotates away. This is more important with the longer, heavier, more cumbersome boosters 

[linuxgurugamer]

58 minutes ago, StrandedonEarth said:

One other tip with separating boosters: it helps to have the bottom of the boosters slightly below the bottom of the core, to prevent a bottom strike as the booster rotates away. This is more important with the longer, heavier, more cumbersome boosters 

That actually depends on the location of the decoupler, and any separatrons. What I’ve found is the mount the SRB so that the decoupler is very slightly below the CoM of the SRB when it is empty.  That way the whole srb is pushed away with very little rotation