What is up with K.S.P.?

[Shade Mourning]

I have the latest update of k.s.p. and i'm building various space planes that would work in real life and no matter whether i have an s.a.s. unit on the vehicle and turned on or not my plane ends up doing a back flip and nose dives into the ground. is there something i'm doing wrong? is it the version? or does k.s.p. just hate space planes? can anyone help me please? just to be clear it's during take off i get wheels up and maybe two hundred feet up then it flips and nose dives.

Edited March 2, 2017 by Shade Mourning

[Foxster]

The most basic thing to check is where the Centre of Mass is in relation to the Centre of Lift. 

These can be shown by clicking the appropriate button on the panel at the bottom left during assembly. 

You want the CoM a little way in front of the CoL at all times, when full of fuel and when empty.  

Edited March 2, 2017 by Foxster

[Streetwind]

This tutorial thread with lots of pictures should get you started.

It is outdated in terms of the atmospheric model used - KSP has improved a lot in that regard since the guide was written - but every piece of actual advice given there still holds completely true today.

 

[bewing]

Flipping, in general, is usually caused by "too much drag ahead of the CoM, and not enough behind the CoM". Adding more control surfaces can help. But basically we'd need pix and/or a vanilla .craft file in order to make concrete suggestions.

 

[Snark]

+1 to @bewing's excellent advice.  Your CoM is too far to the back.  Yes, we'll need to see a screenshot to know for sure, but that's usually the problem when people have flipping spaceplanes.  One of the reasons why that tends to be such a problem is that the heaviest part of a spaceplane is usually the engines, so if you put those all at the back of the plane, it tends to result in a CoM that's way in the back.

Which means that trying to fly a plane like that nose-first is kind of like trying to throw a badminton birdie feathers-first.  It just doesn't work.

One common solution to this problem is to design the plane so that the engines are mounted on the sides of the fuselage, near the middle of the plane, rather than sticking them at the back.  This helps to move the CoM forward and reduce the flipping problem.  Of course, it's not a silver bullet, and like everything in KSP, it matters how you do it, which is why we need to see a screenshot of your plane in order to give relevant advice on how to fix your problem.

 

9 hours ago, Foxster said:

the most basic thing to check is where the Centre of Mass is in relation to the Centre of Lift.

Careful, here-- you're not wrong (because yes, you do want the CoL behind the CoM, and it would be a problem if it were in front), but it's not telling the full story.  "CoL behind CoM" is certainly necessary, but it's far from sufficient.

When a plane (or any craft) flips, it's important to understand the real reason.  The real reason is that the center of dynamic pressure is behind the CoM.  Not the center of lift.  You can have a craft whose CoL is behind the CoM, and still flips.

A plane that has its CoM in the back is highly likely to flip, even if its CoL is even further back.  That's because the long, lightweight, draggy fuselage sticking out in front is moving the center of dynamic pressure way forward, even though it's not affecting the SPH's CoL display since it's not a "lifting surface".  Whereas, there are other things that do affect the CoL display marker that are irrelevant to your stability.  For example, if your plane is unstable on the pitch axis, then adding a vertical tailfin at the back will do nothing to help that-- but you'll notice that when you stick that vertical stabilizer on the back, it moves the CoL rearwards, possibly leading the player to a false sense of security.

Sorry, I don't mean to nitpick.  My concern is simply that this "CoL behind CoM" advice tends to be so widespread in the forum, and presented so authoritatively (as if it tells the whole story), that it leads to common misconceptions about how aerodynamic stability works, and thence to player frustration.

It's not bad advice-- it's good, necessary advice, in fact.  CoL should be behind CoM.  But the danger is that this advice oversimplifies, by leading people to think that "as long as I put CoL behind CoM, I'm fine", when that's not the case.

[Foxster]

I know what you are saying but I was keeping it simple and giving something the OP could check and perhaps eliminate as an issue pretty quickly. 

[sal_vager]

It'd help a lot if we could see these craft, and their CoL and CoM.

[Snark]

4 minutes ago, Foxster said:

I know what you are saying but I was keeping it simple and giving something the OP could check and perhaps eliminate as an issue pretty quickly.

Yup, and I totally get the desire to keep it simple.

Just sayin' that it may be worth accompanying "CoL behind CoM" advice with a simple disclaimer, such as "this is a starting point and isn't the whole story.  Make sure CoL is behind CoM, and see if that fixes it.  If it doesn't, come back and we'll talk."

 

By the way, @Shade Mourning, you might want to consider editing the title of this thread to something informative like "My spaceplanes keep flipping" instead of just "what's up with KSP", because that will help people help you.  The current title doesn't really help people browsing this forum to know what you're asking about.  (In fact, "what's up with KSP" probably applies to just about any thread in Gameplay Questions.)  It's up to you, of course, but in general I expect you'll get better results with a more relevant title.

[Shade Mourning]

@Snark I was intending to use this thread as a general question board for if/ when I can't find the answer, though I will perform the change. And from what I understand from your first paragraph is that I should balance my CoM between front and back then add CoL just aft of that. Or have I read and understood what you said wrong?

And thank you to everyone who take the time to respond to my questions, which admittedly probably could have been solved by just some adjustments.

[Kryxal]

If you can't find the answer, a new, specific thread is generally a good idea.  Mind you, the answer is likely out there, but if you don't know what to look for, it can be tricky to find...

[Snark]

On 03/03/2017 at 1:51 AM, Shade Mourning said:

@Snark I was intending to use this thread as a general question board for if/ when I can't find the answer, though I will perform the change.

It's up to you.  The main thing is, the Gameplay Questions sub-forum isn't really designed for open-ended long-running discussions.  It's more geared towards a scenario towards,

1. a person asks a specific question
2. people answer the question
3. when the OP decides it's answered, they mark it as such
4. people see that it's "done", (usually) stop posting, and life goes on

Main thing is that you're more likely to get helpful answers if you ask targeted questions with explicit titles.  Best way to understand that is to put yourself in the shoes of the folks who may help you:  specific titles help them in a lot of ways, which makes them more likely to take the time to help you.

Specific examples in spoiler section for why it's generally more effective to make "targeted question" posts instead of an open-ended thing.

Spoiler

• Different people know about different things.  If I see a question whose title says "spaceplanes", and the Helpful Person doesn't know anything about spaceplanes, then the Helpful Person just skips over the question and can spend the time more productively looking at a question from someone else asking a question the H.P. does know something about.
• Easier to tell whether someone actually needs assistance.  If it's a specific question, and it gets answered, and the OP marks it as such, then the Helpful Person can see that it's done and doesn't need to spend any time reading it or trying to answer things.
• Easier to keep mental track of what open issues are out there.  If there are a dozen open threads whose titles are just variations on "Help!", it's hard to keep them straight.
• Quicker to zero in on what needs answering and what's already been said.  Specific-question threads tend to be short (i.e. don't have tons of posts in them), which means it's fast for the Helpful Person to read the OP and then skim through the previous posts in the thread.  Whereas if you have a long-ongoing thread, then your "latest question" will be buried somewhere in the middle and is a pain to find, which makes it hard for the Helpful Person to know "what's even being asked, here?" since they can't just go to the OP to find it.

You can post however you like... it's just that if you make it really easy and quick and convenient for people to help you, then you're more likely to get help. 

On 03/03/2017 at 1:51 AM, Shade Mourning said:

And from what I understand from your first paragraph is that I should balance my CoM between front and back then add CoL just aft of that. Or have I read and understood what you said wrong?

No, you've pretty much got it right!

Here's the summary:

• There are three things you have to care about.  Center of mass (CoM), center of lift (CoL), center of dynamic pressure (CoDP).
• You have to keep the CoM in front of both the CoL and the CoDP.
• CoM and CoL both have helpful graphical indicators available in the SPH, so that's easy.
• Unfortunately, there is no graphical indicator for CoDP (nor is there ever likely to be one, for various reasons that for brevity's sake I won't go into here, but can be summarized as "because it's really hard, verging on impossible").  So you have to visually estimate that.
• "Move your CoM forward" is generally the best answer when you have aerodynamic instability problems.

On 03/03/2017 at 1:51 AM, Shade Mourning said:

And thank you to everyone who take the time to respond to my questions, which admittedly probably could have been solved by just some adjustments.

No problem at all, it's what this forum is for.  And "could have been solved by just some adjustments" describes pretty much everything in KSP, including all the questions posted in this forum, so don't feel bad.  Knowing what those adjustments may be is the hard part, which is why we have a Gameplay Questions forum in the first place. 

[Shade Mourning]

@Snark ok I'll keep that in mind. And thanks for your help and clarifying, I'm fairly smart but this IS rocket science so I feel pretty dumb

[Snark]

1 hour ago, Shade Mourning said:

this IS rocket science so I feel pretty dumb

 

Ha!  Don't feel bad, we've all been there. 

[Jaeleth]

I find this thread very interesting. I've been designing KSP aircraft for a long time and, in fact, the CoL behind the CoM does not solves the problems always but if you have a fairly pointy nose and the wings backwards it should do, like Snark says at a point, it is somewhat intuitive. Use this aviation motto: "if it looks good, it'll fly good".

Most important also and, a real design issue, is inflight displacement of CoM. This is usually due to fuel being spent or cargo being dropped. The trick here is to put as much fuel mass as close as possible to empty CoM, either by a central fuel tank or wings. But on cargo aircraft, SSTO's or shuttles, which have to be balanced for flight with and without payload, it is useful to place a fuel tank as far ahead as you can so that you can transfer fuel to it, in flight, in order to move the CoM further ahead in case of need. Of course, always design your craft to be flyable with zero fuel... 

The Placement of CoM too further ahead makes it also "nose heavy" and hard to land and take off. For those cases consider the use of canards.

Also, sometimes, you may find it difficult to take off even when there is plenty of lift. Check placement of landing gear, ideally, the main gear (rear, on a tricycle gear), should be as close as possible to the CoM as long as the craft doesn't tends to flip back when landed...

Hope to have contributed, somehow.

 

Edited March 5, 2017 by Jaeleth

[GoSlash27]

On 3/3/2017 at 3:51 AM, Shade Mourning said:

I should balance my CoM between front and back then add CoL just aft of that

@Shade Mourning,

 Yes. That will eliminate problems with both the center of lift and center of pressure (which isn't shown in the VAB). Additionally, you should make sure that the CoM doesn't shift as the fuel drains. It's a bit of a balancing act (no pun intended).

Best,
-Slashy

 

[Kryxal]

I had a recent contract to test a RAPIER in solar orbit, and made a SSTO to do so ... don't forget that if the CoM shifts some on you, you can move fuel between tanks to balance.  That's what I ended up doing on reentry to fix a trim problem, it was nose-heavy enough that I couldn't hold a very high AoA to bleed off speed in thinner atmosphere.