In praise of the new aerodynamics and rocket flipping

[Voidryder]

I find craft that wobble/bendiness combined with hyperactive SAS and extreme gimbal makes craft more inclined to fall over, so I tweak gimbal down to .90 values and I use the basic fins with struts to add truss structures to my longer craft.

I swear decouplers and fairing bases are made of rubber.

Sal, you add single-track railway truss bridges to your rockets? I think I might know why you're having difficulty achieving orbit.

[Wheffle]

I remember going through a short rage-phase after installing FAR in alpha/beta because my rockets were flipping, which forced me to learn more about aerodynamics. It just takes some time for people to adapt to new rules, I think. In the end, for me a large part of KSP is about discovering parts of how the real world works by crashing goofy rockets, which is why the new aero system is awesome. Add simple aerodynamics to the list of things I understand a lot better because of this awesome game.

[Hotaru]

I was very happy when I tried out 1.0.2 stock aero to find that my old Mechjeb ascent profile from FAR in 0.90 worked pretty much unmodified. I am a little discouraged at how hard it is to build finless rockets, even with Mechjeb controlling the ascent (real launch vehicles haven't had fins since the 60s) but on the whole I find that just about anything roughly rocket-shaped works fine so long as I stick some fins on it and limit the acceleration to 15 m/s^2 for the first 20,000 meters or so of ascent. KJR helps with the noodle-rocket problem, but honestly I was getting along just fine without it until my rockets developed an annoying habit of blowing up on the pad (not sure what exacerbated that issue, but KJR solved it--the reduced wobble was an added bonus).

All in all I'm having fun with the new aerodynamics. Rockets have to make some kind of realistic sense, airplanes handle MUCH better--coming from years of X-Plane, I could hardly fly planes at all in 0.90 stock--everything just behaves the way I always expected it to behave in the first place, only now I don't need mods to make it happen.

I do wish they'd make fuel tanks drain in the opposite order, though. Or just have them drain evenly, like with jets.

[Aerindel]

The 1.02 areo is one of the few things I'm actually happy about since 1.0. Rockets have never been this easy to fly before.

[Yakuzi]

We need a center of pressure marker in the vab....

I suggested this before the release of 1.0, as OP's observations also happened in bottom heavy craft in 0.90. tetryds explained that a Center of Aerodynamics (your center of pressure I believe), would be quite valuable to have. However, the CoA is dynamic as it depends on speed, altitude and AoA, meaning that we would need a "sweeping" analysis tool like in FAR in the VAB/SPH. I'm all for a feature like this, since you would be able to evaluate the flight characteristics of your craft and identify and fix any potential major aerodynamic issues before launch. IN combination with n fuel management system that includes CoM shift evaluation during fuel usage would be great additions for designing craft. Unfortunately the developers have mentioned on multiple occasions that "starving" the player of information is somehow "fun". On the other hand, they've changed their mind on a deltaV readout, so we might get lucky on a Center of Aerodynamics as well.

If not, there's always FAR, which is far better than the current indecisosphere that is stock 1.0.0-1.0.2. Let's hope the stock aerodynamic/re-entry heat model won't need any more fixing and adjustments after the next KSP update, so we can focus on making persistent craft designs.

Edited May 16, 2015 by Yakuzi

[Cdr_Zeta]

Hello,

I still have major problems with this and I consider myself a veteran KSP player...new players are having a hard time with this and getting into ALOT of frustration; I contend there is still major problems with the current drag, fuel, and whatever to cause rockets to flip over; it's rediculous and NEEDS to be fixed; not improved; FIXED.

That being said, as a vet KSP player I cant say I like or dislike it; it is a matter of solving a problem.

I saw one streamer solve it in a certain way that honestly...I cant understand it - and NO FINS...so I am like ya - ok this really isnt right either...but is it ?!

I dont know cause I havnt tried it myself...

Not sure how to put a spoiler in here...

So I probly will give this ingenious solution a try tonite.

We always take a setp forward and a few steps back; Squad has taken a 'few' steps forward and a step back here; we are all re-learning...

Cmdr Zeta

[archnem]

Yah, I really fail to see what the wailing and gnashing of teeth is all about. I admit when I first started playing with the new aero, I'd get bad rocket wobble and flips and got a bit frustrated. You know what I did then? I RELEARNED HOW TO BUILD AND FLY ROCKETS. And guess what? Now I'm loving the new aero! Planes fly much more realistically, and rockets are much easier to get into orbit. Know what the secret was? Ease off the throttle!! You can't simply leave all your rockets at 100% throttle the whole way anymore. Add ease off the flight controls too! Takes a gentle touch now. Add some fairings or fins if you like for super-easy mode. Once I made the adjustment, things are way easier now than they ever have been.

[jmbailey2000]

I just wish we had more tools to control fuel flow in the rockets. The way LFO flows in the tanks, it's as if they want the flips to happen. As if kerbal rockets were designed to fly doing stunts...

If someone hasn't mentioned it already, try TAC Fuel Balancer or GPOSpeedFuelPump. Fuel pump work pretty well since you can set different levels on different tanks to control the flow.

[Nurph]

Been playing since 0.17 and very much in the 'enjoy' camp with the new aero. My only gripe with it would be that their needs to be some kind of penalty for experiencing high G's and parts breaking off due too extreme drag forces, like if I deploy some air-brakes going Mach 4 and the G-meter maxes out, those air-brakes should get torn off.

I've had some rockets flip out and manage to even recover from some flips and keep going. Hands-off gravity turns are (somewhat) possible. A general rule I've followed is never exceed 2 G's of acceleration and start your gravity turn at ~60 m/s (yank it over 5 to 10 degrees), which is usually less 500 m above the pad. The soup-o-sphere has been banished to the nether realms, hath no fear of early gravity turns as long as thou hath decent TWR. Just the other day I built a fully recoverable launcher capable of lifting 15 tons to LKO and perform powered landing at KSC. Just start simple and build your way up.

Y'all just bein' silly

[tg626]

Why, are you trying to fly your rocket sideways?

No.

Knowing the CP would allow you to know your Static Margin and therefore the stability of the craft.

(You do know what Center of Pressure is, right?)

http://exploration.grc.nasa.gov/education/rocket/rktcp.html

http://ftp.demec.ufpr.br/foguete/bibliografia/tir-33_CP.pdf

[eddiew]

So thank you, Squad, for introducing that sort of necessary thinking into the game.

I quite agree and well said

To everyone struggling with stock aero; it is not broken, and it is not flipping your rockets for "no reason". There are two primary questions to ask when a rocket is unstable:

1) where is the centre of mass at the time the rocket flips?

2) where is the centre of pressure/lift at the time the rocket flips?

Every single time the answer will be that the CoM has gotten behind the CoL. Go back to the VAB, drain some fuel, starting from the tank furthest away from your engine, and see where the CoM is going during flight

[little square dot]

i must admit that sometimes i also feel like blaming everyone else except me. Usually a temporary feeling :-)

haha.. love the honesty.

I'm pretty sure we all go there from time to time.

[ToddEth]

My first flight in 1.0.x and the rocket flipped and went crazy spinning but managed to get control back and just get in to orbit but not enough fuel to get back so back to sandbox to do a little testing on a few designs using only the parts I had unlocked and a few changes here and there and I was golden to complete a rescue mission. it made me think outside the norm and I learnt how to adapt.

This is what all the people whining about aero and rockets flipping need to understand the game is about challenges and the rise to overcome those challenges.

If you don't like it then mod it to your tastes, if there is not a mod that can do that then write one, if you can not write one then learn how to, if you can not be bothered then sorry to say this is not the game for you.

If you are offended by what I have said then tough luck.

[Fleb]

I spent quite some time looking at real rockets and trying to replicate them a long time ago, so I haven't had any problems adapting either.

[Temeter]

No.

I brought rockets into space that looked like a giant over the top morning star (the one you hit people with). I'm not sure a CP does matter very much in KSP, at least if I understand the concept correctly.

[martinborgen]

what's with this flipping thing? I never had any big problems with it! Happened once or twice, but only minor nuisance.

[Yakuzi]

I quite agree and well said

To everyone struggling with stock aero; it is not broken, and it is not flipping your rockets for "no reason". There are two primary questions to ask when a rocket is unstable:

1) where is the centre of mass at the time the rocket flips?

2) where is the centre of pressure/lift at the time the rocket flips?

Every single time the answer will be that the CoM has gotten behind the CoL. Go back to the VAB, drain some fuel, starting from the tank furthest away from your engine, and see where the CoM is going during flight

While this is applies to most craft, it's a bit more complicated. Center of Lift is not the same Center of Aerodynamics (CoA), so having CoM before/above CoL won't always result in non-flipping. CoA represents the combined forces of center of lift and center of drag. Drag has become more important in 1.0.x, and if its center (together with CoL) moves before/above the CoM you get the "throwing a dart backwards" behavior.

For those who never/hardly experienced flipping: you can induce this behavior by making a long bottom heavy rocket with lots of draggy parts on the top. You can add as many winglets on the bottom (CoM before/above CoL, full fuel and empty) but your rocket will flip. Alternatively you can look at the craft from people struggling with flipping, I'm sure most of them have applied the CoM before/above CoL principle if they've ever played KSP, however their CoA will move above CoM just before their craft starts to flip out.

Squad should really introduce a CoA view in the VAB/SPH (see my post earlier in this thread).

Edited May 17, 2015 by Yakuzi

[Majorjim!]

I built a rocket the other day that ingenuously (I thought) siphoned the fuel from the payload into the main launcher, to give my side boosters about 30 more seconds of thrust on ascent. My trial runs with the rocket had shown that I'd save total dV if they could burn a little longer, and why add more fuel when I didn't actually need to? Once in orbit, I could just transfer fuel back from the remaining center stack before ditching it to fall back to Kerbin.

This rocket, which in early testing had been pretty rock solid on the ascent, started flipping like a baton no matter how well I kept it in the airflow, no matter how many fins I put on the bottom, and no matter how much I tried to correct manually.

Now at this point, from what I see in so many posts in this forum, most people would have thrown their arms up and complained that KSP was killing their fun. Instead, I looked at it, thought about it, and realized my problem. By emptying the fuel out of the top of the rocket, I was making the whole thing EXTREMELY bottom heavy. I was - to continue to overuse an overused analogy - throwing a dart backward. Putting the feathers on the arrow head won't make it fly backwards, to overuse another analogy.

So I undid my change, found a better way to add a little more time to my early boost stage, and the rocket made orbit with the payload intact. I could even deviate significantly from prograde very low in the atmosphere and at high speed, all because I (re)applied a little aerodynamics know-how.

So thank you, Squad, for introducing that sort of necessary thinking into the game.

Very well said. The aero changes where much needed. I discovered as you did the importance of weight distribution in a rocket. Once you do that it's as easy as before.

[moogoob]

I for one like the new aero. Launch times are no longer boring - they are often now chaotic and exciting. Sometimes it's a minor oscillation that can be corrected by shutting down a gimbal or something (a similar problem was experienced by the Saturn V on a few launches), sometimes you really bork up the aerodynamics and your rocket snaps in half.

Payloads now have to be designed such that they're either aerodynamic or have to fit under a reasonable fairing. I can go on. I like the engineering challenge!

[LordCurlyton]

What would be nice now that I think of it would be to have an option to see CoM/CoL/CoT in game instead of just during construction. I think being able to see what is happening at the point of failure (flipping) would help immensely, especially if its not entirely obvious why a rocket which has gone straight as an arrow suddenly inverts. For me, it just seems I try to go too fast (above 250-275) with my rockets and it flips, but usually rights itself quickly as it slows down, allowing me to recover. I suspect the magical SAS may also have a hand.

[waterlubber]

I only had flipping issues in 1.0.2.

The problem is, rockets are too wobbly, so we need to built outwards. The best thing to do IMHO is match your payload width to your lifter width.

[RogalDorn]

So nice to see someone put thought into this problem! I was thinking I was the only person out there who was totally comfortable with none of my old designs working anymore. I was half tempted to throw a second engine on the top of my ships and attach with fuel likes so I can just keep on burning when I flipped!!! Instead I just became a better person and did what I had to do!

[luizopiloto]

What would be nice now that I think of it would be to have an option to see CoM/CoL/CoT in game instead of just during construction. I think being able to see what is happening at the point of failure (flipping) would help immensely, especially if its not entirely obvious why a rocket which has gone straight as an arrow suddenly inverts. For me, it just seems I try to go too fast (above 250-275) with my rockets and it flips, but usually rights itself quickly as it slows down, allowing me to recover. I suspect the magical SAS may also have a hand.

Squad changed drag/lift values in 1.0.1/2... In a try to "fix" some reentry problems ppl was having (because of wrong reentry profiles ppl was using)... but this unbalanced the game... Lots of ppl complained about this change, but have seen no response from Squad until now, as they're gone on vacation...

Edited May 18, 2015 by luizopiloto

[Wallygator]

CoA, CoL, CoP, CoM... dear mother of...

back in the the 60's the rule of thumb in model rocketry was that the centre of pressure must be at least 1.5 times the body tube diameter behind the enter for mass to assure stability (maybe I am not remembering it clearly enough, but those of you in the know will know what i mean).

I think we (ok me) are inundated with too many variations... and the terminology for rockets v aeroplanes might be getting a bit too commingled.

Or... thinking has just moved on, and I'm now a Dynasoar.

[PB666]

A thousand and one threads complaining about rockets flipping for no reason and finally someone understands the problem. Hurrah!

Im tired of people complaining, I went through career mode hard up to level 4 and had none of the problems people complain about early game career.

Everyone seems to think the game is broken if they cant push their rocket to Mach 1 before 10,000 m and have never read spit about MACH 1, or maintaining control during flight, some drag needs to be at the back to feather if you put all the drag at the front and all the mass at the back, you can be certain that when near MACH dynamic density isoquants begin pushing the nose of the craft it will tip over. We've known for 5 months that aero was coming, plenty enough time to start playing with new parts or making new parts.

https://en.wikipedia.org/wiki/Area_rule

https://en.wikipedia.org/wiki/Supersonic_speed

https://en.wikipedia.org/wiki/Wave_drag

https://en.wikipedia.org/wiki/Sears%E2%80%93Haack_body

https://en.wikipedia.org/wiki/Area_rule

If you want to surpass .85 Mach at high atmospheric pressures you need to design for it. You can't argue the games a failure because a previous version converted mass into drag and you could launch any blinking profile that you wanted.

And lets end the complaints about early game part shortages when there is 200 sci available with 5 km of KSC. Just gather some Sci, tech up and stop barking about a little more work in career mode to get parts so that you can speed through the lower atmosphere at a wasteful MACH 3.

Im glad to see the new aero, I hope we will see a revolution of new more realistic part designs (not squad retextures but de-novo parts) designed to deal with many atmospheric flight scenarios.

https://en.wikipedia.org/wiki/Scale_height

5000M up decrease by e = 2.718 fold