• 0
Wcmille

Choosing the Right Radial Decoupler

Question

At present, I use the small decoupler for small radial stages and the larger one with standoff for larger stages. Doesn't feel like good engineering to me.

Cost aside, what does math say about when you should pick each of the five radial decouplers?

Share this post


Link to post
Share on other sites

Recommended Posts

  • 1

Without cost as a factor here are my thoughts:

Small, flat decoupler: I use this every/anywhere I can, and usually only on rockets. It hardly weighs anything, fits on almost every surface, and can hold on to any other part. I use this part 95% of the time.

Medium decoupler with standoff: I only use this for the standoff feature, so that I can clear bulging stacks or other radially attached parts. I use this part 4% of the time.

Large, flat decoupler: It's too heavy, decouples (by default) too forcefully, and does nothing that the small decoupler can't with a little thoughtful strutting. I'm curious about whether or not parts jiggle as much when held by it, but the weight of the decoupler always keeps me from investigating (or caring). I never use this part.

Small hardpoint: I only use this on airplanes, but very rarely since I almost never need to jettison parts from them. I use this part <1% of the time.

Large hardpoint: Again, only on airplanes, even more rarely than the small ones. I use this part <1% of the time.

Edited by arise257

Share this post


Link to post
Share on other sites
  • 1

From my bit of experience it's not so much which one you use but the location of both the decoupler and the object on the decoupler that are really important.

Share this post


Link to post
Share on other sites
  • 0

There's no significant difference in rigidity: it doesn't take much load and you have to reinforce the decoupler with struts. Guess what? Small decoupler+struts is just as stable as large decoupler+struts.

Unless you have special needs, the cheapest / most lightweight solution will be best.

 

Share this post


Link to post
Share on other sites
  • 0

I have literally no idea what you would use the Hydraulic Detachment Manifold for. 

Share this post


Link to post
Share on other sites
  • 0
1 hour ago, Foxster said:

I have literally no idea what you would use the Hydraulic Detachment Manifold for. 

It's easily one of the most useless parts in the game. They should either eliminate it or make the joint connection so stiff and stable that you'd never need to add struts -- and make it lighter.

Share this post


Link to post
Share on other sites
  • 0

For decoupling really heavy radial boosters you generally need sepratron anyway, so the higher ejection force is moot.

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, Foxster said:

I have literally no idea what you would use the Hydraulic Detachment Manifold for. 

 

31 minutes ago, Temstar said:

For decoupling really heavy radial boosters you generally need sepratron anyway, so the higher ejection force is moot.

I use the Hydraulic Detachment Manifold for really heavy radial boosters, i.e. Big Orange Tank or bigger, also Kickback SRBs.

It's not a must-have; it's possible to use the smaller decoupler.  But I find that the extra ejection force is handy for getting those big boosters clear of the central ship.  Yes, it's heavy at 0.4 tons.  But the things I use it on are so big and heavy themselves (e.g. a Big Orange Tank, with a Skipper or Mainsail, plus nosecone, etc.) that it's a relatively small price to pay; as a percentage, it's only a very small fraction of the total mass.  And it's very low-hassle.

Conversely, I never use sepratrons.  I just don't like 'em.  They boost part count, they require too much tedious fiddling to get placed right, I have to worry about their exhaust heating up and destroying the central core... yes, they're an alternative to the Hydraulic Detachment Manifold, but I find the manifold to be quicker and less hassle to use.

Share this post


Link to post
Share on other sites
  • 0

I tend to use HDMs when sticking on massive KWR tapered booster tanks that utterly obscure lesser decouplers. Also those things are huge and I want the big boom to shift them the heck away

Share this post


Link to post
Share on other sites
  • 0
5 minutes ago, Snark said:

 

I use the Hydraulic Detachment Manifold for really heavy radial boosters, i.e. Big Orange Tank or bigger, also Kickback SRBs.

It's not a must-have; it's possible to use the smaller decoupler.  But I find that the extra ejection force is handy for getting those big boosters clear of the central ship.  Yes, it's heavy at 0.4 tons.  But the things I use it on are so big and heavy themselves (e.g. a Big Orange Tank, with a Skipper or Mainsail, plus nosecone, etc.) that it's a relatively small price to pay; as a percentage, it's only a very small fraction of the total mass.  And it's very low-hassle.

Conversely, I never use sepratrons.  I just don't like 'em.  They boost part count, they require too much tedious fiddling to get placed right, I have to worry about their exhaust heating up and destroying the central core... yes, they're an alternative to the Hydraulic Detachment Manifold, but I find the manifold to be quicker and less hassle to use.

I suppose that's one of the cool things about KSP: There is no one right way to do something. You can develop you own "style" of craft building. 

Share this post


Link to post
Share on other sites
  • 0
5 hours ago, Snark said:

 

I use the Hydraulic Detachment Manifold for really heavy radial boosters, i.e. Big Orange Tank or bigger, also Kickback SRBs.

It's not a must-have; it's possible to use the smaller decoupler.  But I find that the extra ejection force is handy for getting those big boosters clear of the central ship... Conversely, I never use sepratrons.  I just don't like 'em.  They boost part count, they require too much tedious fiddling to get placed right, I have to worry about their exhaust heating up and destroying the central core... yes, they're an alternative to the Hydraulic Detachment Manifold, but I find the manifold to be quicker and less hassle to use.

Does anyone else just add tiny fins at the front of boosters they need to clear outward? You can even control the speed at which they peel outward from the stack by increasing the angle of incidence of those fins.

Share this post


Link to post
Share on other sites
  • 0

I must admit I use the HDMs all the time. A bit embarrassing I know because I'm adding weight uselessly, but I feel a bit cheaty using the small ones when they'll be handling huge forces.

Also they do help to ensure that the bits they attach get forced clear (though this can be a bad thing if the attached stack is too tall, because it generally means that the bottom of the stack is forced clear but the top is now angled towards my rocket, and that + atmosphere can be fatal to my core stack...).

So after reading this thread, I guess I'll feel less like a cheat using tiny decouplers to hold huge stacks together.

Share this post


Link to post
Share on other sites
  • 0

I must say I always use the small one. Burt again, I don't do much regular stages rockets (I prefer recoverable SSTO rockets). I only detach lateral payloads such as satellites.

In beta 0.9, I used a lot the medium ones, but never the big ones.

 

Share this post


Link to post
Share on other sites
  • 0
8 hours ago, arise257 said:

Does anyone else just add tiny fins at the front of boosters they need to clear outward? You can even control the speed at which they peel outward from the stack by increasing the angle of incidence of those fins.

Yes, I do this at times when I absolutely need a bit of separation. Most of the time though, I make sure the boosters are centered with their empty center of mass on the decouplers, ensuring that at decoupling, they simply 'slide' down with hardly any torque.

Also, I am using the hardpoints too, both small and big. Small ones make great wing-like connectors for small planes. The large ones I have been using for big radial boosters, because of the more streamlined looks, but also because I can place two or three in a row, disable staging of the outer two, and it looks and works like one part, while allowing a LOT more clearance for those huge Mammoth based boosters I sometimes use.

Share this post


Link to post
Share on other sites
  • 0
3 hours ago, Plusck said:

(though this can be a bad thing if the attached stack is too tall, because it generally means that the bottom of the stack is forced clear but the top is now angled towards my rocket, and that + atmosphere can be fatal to my core stack...)

Except that it's a good thing for tall stacks, because you mount the decouplers above the CoM of the radial stack, so that the top is forced clear, angling the booster away from your rocket, so that atmosphere helps move the jettisoned booster away from the core stack, right?  ;)

Share this post


Link to post
Share on other sites
  • 0
49 minutes ago, Snark said:

Except that it's a good thing for tall stacks, because you mount the decouplers above the CoM of the radial stack, so that the top is forced clear, angling the booster away from your rocket, so that atmosphere helps move the jettisoned booster away from the core stack, right?  ;)

Well you would, but then that would be illogical because the forces that the radial stack are transmitting to the central stack are far from the CoT. You could strut it at the bottom, and the game would be happy with that, but a single strut would not be a logical way of transmitting the weight of the stack before liftoff and its thrust afterwards. ;)

Share this post


Link to post
Share on other sites
  • 0
36 minutes ago, Plusck said:

Well you would, but then that would be illogical because the forces that the radial stack are transmitting to the central stack are far from the CoT.

I'm having trouble following the reasoning here.  What does it matter where those forces are relative to the CoT?  CoT is irrelevant, as long as it's on the CoM axis.

CoT can be relevant for off-axis thrust, such as you can get with airplanes/spaceplanes sometimes, and is especially relevant for shuttles, for which it's the bane of their existence.  However, we're talking about radially-symmetric vertical-takeoff rocket designs here, where the CoT is always located on the center axis of the rocket, so it doesn't matter where it is.

The CoT of the radial boosters does have some relevance in terms of its radial distance from the attachment point-- not in terms of the overall flight behavior of the rocket (it's irrelevant for that) but in terms of mechanical strain (a CoT that's very far radially outward from the booster's attachment point has a bigger lever arm and therefore generates more deformation torque to contend with).  However, if we're just talking about adjusting the vertical position of the radial decoupler up and down, it doesn't affect the radial component of CoT displacement, and is therefore not an issue here.

The fact is, mounting the radial decoupler as high above the radial booster's CoM as possible will do the best job of getting the booster to peel out and away from the central stack, by making it rotate outwards.  I do this all the time, and it works great, and have never had a single problem with it.  I've seen this advice echoed by other folks in the forums as well, so I'm not the only one advising this.

44 minutes ago, Plusck said:

You could strut it at the bottom, and the game would be happy with that, but a single strut would not be a logical way of transmitting the weight of the stack before liftoff and its thrust afterwards.

Well, as with any big radial boosters (regardless of decoupler placement), flexing at the joint can be an issue, and the solution is usually a strut.  But again, I'm having trouble following your argument here.  A single strut works great to keep boosters stable.  I do it all the time.  How is it not a "logical" way?  (In any case, the lion's share of the structural coupling is borne by the decoupler; the strut is just there to give a second contact point to prevent flexing around the joint.)

And it's irrelevant* whether the strut is at the top or the bottom of the radial stack.  It works fine either way, as long as the strut is placed far enough away from the attachment point (i.e. the radial decoupler) to give it a decent lever arm to work with.  I usually end up placing the strut on top, but that's just because it's more convenient in the editor to get at (it can be awkward to maneuver the camera underneath the ship to allow bottom placement), not because it's somehow more "logical".

(* re: "irrelevant" strut placement:  yes, it can make minor differences based on drag or mass added by the strut, but this is such a tiny contribution that I find it's not worth worrying about.  Strut mass and drag becomes a real concern only if you have a whole lot of them, which is usually an indicator of an iffy design anyway.)

Share this post


Link to post
Share on other sites
  • 0
40 minutes ago, Snark said:

I'm having trouble following the reasoning here.  What does it matter where those forces are relative to the CoT?  CoT is irrelevant, as long as it's on the CoM axis.

CoT can be relevant for off-axis thrust, such as you can get with airplanes/spaceplanes sometimes, and is especially relevant for shuttles, for which it's the bane of their existence.  However, we're talking about radially-symmetric vertical-takeoff rocket designs here, where the CoT is always located on the center axis of the rocket, so it doesn't matter where it is.

The CoT of the radial boosters does have some relevance in terms of its radial distance from the attachment point-- not in terms of the overall flight behavior of the rocket (it's irrelevant for that) but in terms of mechanical strain (a CoT that's very far radially outward from the booster's attachment point has a bigger lever arm and therefore generates more deformation torque to contend with).  However, if we're just talking about adjusting the vertical position of the radial decoupler up and down, it doesn't affect the radial component of CoT displacement, and is therefore not an issue here.

A strut is essentially a steel (or other) rod. As such, logically, it has high tensile strength, a bit less compressive strength (but still decent) and much less shear strength.

Now if your thrust is perfectly aligned up the centre of your booster and your core stack - at all times - then the only forces on the strut will be compressive and tensile. So no great problem.

However, that is not the case, even for an SRB. Whenever you make the slightest course correction (not to mention all the buffetting and turbulence during the atmospheric ascent) there is going to be a shear force. And the further your decoupler is from the CoT, the greater the shear forces on your strut. Any difference in thrust vectoring between your central stack and the booster is going to magnify those forces. A single metal rod would never be used to hold the bottom of a booster to the core stack. You'd want to attach the booster very firmly at the bottom so that it doesn't budge in any direction, and then strut it at the top so that it doesn't flex along its length. Doing the opposite means that your (necesarily slightly flexible) booster structure is carrying all of the force of the engines. To me, that just doesn't seem logical.

Share this post


Link to post
Share on other sites
  • 0
18 hours ago, Snark said:

 

I use the Hydraulic Detachment Manifold for really heavy radial boosters, i.e. Big Orange Tank or bigger, also Kickback SRBs.

It's not a must-have; it's possible to use the smaller decoupler.  But I find that the extra ejection force is handy for getting those big boosters clear of the central ship.  Yes, it's heavy at 0.4 tons.  But the things I use it on are so big and heavy themselves (e.g. a Big Orange Tank, with a Skipper or Mainsail, plus nosecone, etc.) that it's a relatively small price to pay; as a percentage, it's only a very small fraction of the total mass.  And it's very low-hassle.

Conversely, I never use sepratrons.  I just don't like 'em.  They boost part count, they require too much tedious fiddling to get placed right, I have to worry about their exhaust heating up and destroying the central core... yes, they're an alternative to the Hydraulic Detachment Manifold, but I find the manifold to be quicker and less hassle to use.

I also used to use the HDMs (sometimes Tweakscaled to 200%) along with sepratons with big radial boosters, but now I use Space Y's  aerodynamics ones with built-in sepratons.  The Triple-Packed one especially just about launches big SRBs away, and they'll even handle stacks with multiple 5m tanks and a 5m Mammoth all by themselves.  Along wth KJR, I don't even need to strut big SRBs and only have to put a couple or so on 5m+ radial stacks.

Share this post


Link to post
Share on other sites
  • 0
48 minutes ago, Plusck said:

A strut is essentially a steel (or other) rod. As such, logically, it has high tensile strength, a bit less compressive strength (but still decent) and much less shear strength…

Ah, you're assuming (KSP) struts are like (real-world) rods! I believe it is a mistake to do so, and struts are an abstraction with a single scalar breaking strength. Has anyone done any investigation to determine whether this is the case?

After all, it's reccommended to use struts to connect and stiffen multi-panel wings (coplanar with the panels), and axial connections would be worse than useless for that task.

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, Plusck said:

<IRL explanation of what force would be on a strut reinforcing a radial booster>

To me, that just doesn't seem logical.

So correct me if I'm wrong here, but I'm getting the impression that your argument is based entirely around what would work in real life, and how real rockets are designed?  In other words, essentially what you're doing is adding a role-playing element to the game?

In that case, we're talking at cross-purposes.  I'm talking about KSP, not IRL.  :)  Specifically, I'm talking about what the game actually does (which is what I suspect most folks in these forums are primarily interested in.)  In that world, "not logical" equates to "wouldn't work in the game," which is why your post was confusing me-- because manifestly this does work in the game, and is in fact exactly the kind of situation why Squad added struts to the game in the first place.

The fact is, KSP makes lots and lots and lots of simplifying assumptions that do not reflect real life.  Necessarily so.  Many of these simplifications are for playability; others are to help with performance and/or code complexity.  Examples abound:  Patched conics.  Lack of weather.  Physicsless parts.  100% reliable parts.  The tree structure of rockets.  Heat modeling.  Aero modeling.  Toy planets at 1/10th scale.  And on and on.

Some of these simplifications means that KSP is way easier than real life.  (For example, physicsless parts let you put stuff asymmetrically on a rocket without screwing it up.  Fragile components can withstand a Mach 3 airstream unscathed, no fairing needed except to help with drag.  Heat shields provide 100% protection to everything inside their shock cone-- no worries about surface components getting fried by heat radiating from the incandescent bow-wave plasma. And so forth.)

Some of these simplifications make KSP harder than real life.  For example, it would be great if I could mount a radial booster using two decouplers (one at the top, one at the bottom) for strength.  But I can't, because of the "tree" model that KSP applies to all parts.  (We've all been there and tried to do that, amirite?)

So that's why we use struts with radial boosters:  because the game forces us to do that, since there's no other option.  (Well, unless you just use no struts at all and accept some bending at the joint, but that seems even less real-life-like than the strut option.)

In general, I don't bother too much in KSP with worrying about "would this work in real life".  It's a fantasy, albeit a plausible one.  It's true that I try to avoid ship designs that feel like blatant exploits (which is why I tend not to use clipping, for example), but beyond that, I don't worry much about it.  The moment I set foot on a 600km-radius planet that has full Earth gravity, "IRL" went out the window.  If I started worrying about whether a strut "would" work, I'd also have to worry about the 1,001 other things that are even less plausible and which I accept routinely without a qualm.

So I use struts as they're intended in the game: to reinforce things.  Yes, it looks like a single steel bar, and IRL a single steel bar would be a bad choice.  But then, IRL I'd have a whole bunch more structural options than that single-steel-bar.  So as far as I'm concerned, that's a game playability feature, not an exploit:  the strut is a stand-in for "some kind of reasonable IRL reinforcement," if you like.  If you disallow struts for reinforcing boosters, then where do you allow them?

There's nothing wrong with deciding to role-play that "this isn't 'logical'" and adjusting your ship designs accordingly... but we all have our personal preferences that other players likely don't share, so it's worth mentioning that explicitly if discussing it in the forums.  ;)  One person's "logical" is another's "illogical."  I think clipping is an exploit, so I don't use it.  Plenty of other people think clipping is a highly valuable design feature of the game, and use it all the time, and nothing wrong with that, either.

Since definition of "exploit" or "logical" tends to be highly subjective and individual, therefore, I usually try to confine my comments in the forums to "what actually works", since there's no arguing with that.  :)

Share this post


Link to post
Share on other sites
  • 0
49 minutes ago, Snark said:

Since definition of "exploit" or "logical" tends to be highly subjective and individual, therefore, I usually try to confine my comments in the forums to "what actually works", since there's no arguing with that.  :)

Of course, yes. But my very first post in here admitted to using HDMs all the time because the tiny decoupler feels "cheaty". I then said "You could strut it at the bottom, and the game would be happy with that, but a single strut would not be a logical way of transmitting the weight of the stack before liftoff and its thrust afterwards", which was again, I thought, a clear indication that what is most efficient in the game can look a bit "off" when you think about it.

And the OP was also clearly making a distinction between what works in the game (standoff decoupler with large stages) and what feels wrong when you think about the implications IRL:  "Doesn't feel like good engineering to me".

So since this thread is essentially about "what you use" with the subtext of "how do you feel about it", I think it is precisely the sort of thread where "highly subjective and individual" comments are welcome, even if people argue with them. ;)

And no, I didn't get around to the OP's question about what math says you should use.

Share this post


Link to post
Share on other sites
  • 0
1 hour ago, Plusck said:

Of course, yes. But my very first post in here admitted to using HDMs all the time because the tiny decoupler feels "cheaty". I then said "You could strut it at the bottom, and the game would be happy with that, but a single strut would not be a logical way of transmitting the weight of the stack before liftoff and its thrust afterwards", which was again, I thought, a clear indication that what is most efficient in the game can look a bit "off" when you think about it.

Well, sure, and that's the rub-- I totally missed that.  I tend to think of the word "logical" in concrete terms, i.e. game reality, so I interpreted "would not be logical" as "would not work in the game" and became very confused thereby.  Other than the "little bit cheaty" comment, the rest of the entire sequence of posts was about practical matters, and so I treated the "cheaty" comment as a parenthetical aside that wasn't really the main point of the exchange.

TL;DR:  language can be tricky sometimes, best to be very explicit.  :)

Share this post


Link to post
Share on other sites
  • 0
8 hours ago, Snark said:

The fact is, mounting the radial decoupler as high above the radial booster's CoM as possible will do the best job of getting the booster to peel out and away from the central stack, by making it rotate outwards.  I do this all the time, and it works great, and have never had a single problem with it.  I've seen this advice echoed by other folks in the forums as well, so I'm not the only one advising this.

It does work, but sometimes too good. Intertia can carry a heavy booster parallel to the center stage for longer than you may wish, especially if you're getting into the thinner upper atmosfere and drag doesn't counter it enough anymore. Any kind of rotation while the CoM of the booster keeps up with the center stage can make the other end of the booster hit the center, with catastrophic results.

With lighter boosters, it's the drag that sometimes kills you, as the smallest rotation putting the booster nose out of prograde can initiate a sudden flip in midair, again crashing it into other parts of your craft.

And sometimes, and I admit this may be a personal thing I have with some of my designs, the shape of a complex craft really needs my boosters to let go with an absolute minimum of lateral or rotational movement.

I've had very good results just taking the trouble to figure out the exact CoM of my booster assemblies, and placing that on the decouplers. Most of the time I can even completely dial down the ejection force of the decouplers, cause they just 'slide' down without endangering the rest of the craft, and whatever tight space I left in the configuration of the craft for the exhaust is enough for the booster to fall through. They tend to start rotating only a good distance down/away from my craft. It's very worth the extra effort, and very satisfying to watch.

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.