Question..maybe to the devs, if the curve is not showing wrong..

[Duke-49th]

After I saw this Video right now:

You can see the curve, its going parabolic from the begin on..no long climbing before the turn...and as I can see it, the rocket starts infact to turn very early. And here we are talking about the heavy Ariane IV

And same for some other videos..

I ask myself why we need to lift so many kilometers straight before we can make a stable turn in ksp?

If we do it for example in the height of 2 kilometers, espacially large rockets are going unstable and tips over to the side and many times become uncontrollable.

Why is that? I mean..the Earth is way larger than Kerbin. But still they are not going 10-20 kilometers or more until they start to turn. No matter how big the thrust is (well..at least almost no matter^^)

Do I miss something or is it the lack of not being able to simulate anything so accurate like in reality? Or what is the reason for it?

Thanks

[regex]

Why is that?

As I understand it, the stock aerodynamic model doesn't support it makes Kerbin's lower atmosphere complete pea soup and has an unrealistic drag model that prevents such launch profiles. Try using FAR; you'll have a lot more success with realistic launch profiles.

[DaveofDefeat]

Because the aerodynamics in ksp is terrible and the lower atmosphere in the game is like soup, so the most efficient way to launch is to get out of the lower atmosphere as fast as possible. Real life is a bit different.

[Guest]

KSP atmosphere simulation is completely wrong. If you use FAR, you'd have to follow such a launch trajectory, too.

[VirtualCLD]

I'm not sure I fully understand. Even with my heavy lift rockets, I start turning shortly after clearing the scafolding, but it's very gradual. I try to time it so that I hit 45 degree pitch at ~15km and 10 degrees at 35km. I perform a slow turn all of the way from from launch to orbit and I seem to have more fuel that way then waiting before turning. The trick is not turning too fast either, because you end up wasting fuel due to aero drag.

[el_coyoto]

Kerbin has a quite different atmosphere than Earth which make it seem like pea soup at low altitude. (due I think to how drag is modeled in the game)

This is why KSP rockets HAVE to vertical climb a bit before gravity turning, whereas Earth rockets gravity turn almost immediately.

It's a trade-off between losing Delta-V to gravity (vertical climb) or losing it to air friction (gravity turning too low).

Earth rockets are aerodynamic, so they don't lose too much dV because of air friction, compared to KSP rockets.

Thus, for an optimal ascent profile, Earth rockets gravity turn much earlier to DV minimize losses to gravity.

A bit off-topic, but one of the reasons I like the Kerbal Engineer mod so much is because it displays your terminal velocity and atmospheric efficiency.

Hope this helped!

EDIT : ninja'd 4 times lol

EDIT : Cool video, I really have a soft spot for the Ariane 5 and its massive look!

Edited November 13, 2013 by el_coyoto

[K^2]

Yup. It's the drag model, as a few people mentioned. Specifically, in KSP drag coefficient is mass times a constant. For a real rocket, drag coefficient is independent of mass, but grows with speed. So as a fraction of rocket's mass, it starts out really small, and increases dramatically as the rocket picks up speed. Because of that, it's far more advantageous to do a gravity turn really early.

[Franklin]

I'm not sure I fully understand. Even with my heavy lift rockets, I start turning shortly after clearing the scafolding, but it's very gradual. I try to time it so that I hit 45 degree pitch at ~15km and 10 degrees at 35km. I perform a slow turn all of the way from from launch to orbit and I seem to have more fuel that way then waiting before turning. The trick is not turning too fast either, because you end up wasting fuel due to aero drag.

Yep, exactly. While the common belief with KSP's current drag model is that you need to punch it out of the soup as quickly as possible and then crank your tilt later, it's not really necessary, or rather the degree in which this is done isn't as chunky as the "floor it until 14km, then immediately shift 45Ëš, keep punching it til you've hit a 100km apoapsis" recipe that gets batted around. Few really adhere to the terminal velocity table, and a lot of everyone's circularization burn could be mitigated by elongating their launch tilt. Lots of wasted fuel during what most consider a 'standard Kerbin launch pattern'.

To add weight to the argument, everyone should download MechJeb (if only once) and test out its ascent guidance tool with its 'maintain terminal velocity' option ticked, for a 'perfect' launch. And note its automatic path settings. It'll cap your launch speed to 100-175m/s right up until it begins its 45Ëš tilt at 6.9km, and slowly banks for the 90Ëš apoapsis throughout the rest of the ascent. So much fuel is saved with this process.

[Duke-49th]

A bit off-topic, but one of the reasons I like the Kerbal Engineer mod so much is because it displays your terminal velocity and atmospheric efficiency.

EDIT : ninja'd 4 times lol

Good call..I will take a look at it. I have mechjeb installed...but cannot remember to have these two values in any window.

Btw...ninja'd? What does this mean? (sry..im german..english is not my mothertongue)

Yep, exactly. While the common belief with KSP's current drag model is that you need to punch it out of the soup as quickly as possible and then crank your tilt later, it's not really necessary, or rather the degree in which this is done isn't as chunky as the "floor it until 14km, then immediately shift 45Ëš, keep punching it til you've hit a 100km apoapsis" recipe that gets batted around. Few really adhere to the terminal velocity table, and a lot of everyone's circularization burn could be mitigated by elongating their launch tilt. Lots of wasted fuel during what most consider a 'standard Kerbin launch pattern'.

To add weight to the argument, everyone should download MechJeb (if only once) and test out its ascent guidance tool with its 'maintain terminal velocity' option ticked, for a 'perfect' launch. And note its automatic path settings. It'll cap your launch speed to 100-175m/s right up until it begins its 45Ëš tilt at 6.9km, and slowly banks for the 90Ëš apoapsis throughout the rest of the ascent. So much fuel is saved with this process.

Ahhh...thats what maintain terminal velocity does I never saw a difference...but to be honest..most of my rockets are so tall and heavy that it is already hard to get 200m/s before Im at 5-10km high. (And I know at least for KSP that you are wasting fuel if you go higher speed..as the density of the air let the rocket accelerate less...kind of ISP...dont know how to explain that in english terms)

I always feel like the ascent Pilot is wasting fuel. I only use it if I launch several rockets..like satellites for Remote Tech...or if Im really lazy^^

And also yes...I use FAR. But as Im always got problems before without it, Im still afraid start to turn in a low height I just recognized that I need a bit less fuel and fairings/cones make more sense with it (at least that is how I understand it).

I'm not sure I fully understand. Even with my heavy lift rockets, I start turning shortly after clearing the scafolding, but it's very gradual. I try to time it so that I hit 45 degree pitch at ~15km and 10 degrees at 35km. I perform a slow turn all of the way from from launch to orbit and I seem to have more fuel that way then waiting before turning. The trick is not turning too fast either, because you end up wasting fuel due to aero drag.

Depends on your rockets If you have a very ( I mean really VERY) tall and heavy rockets...lets say 500 tons or more and almost the size of the VAB buliding..your not able to turn in the same low height as with a lets say 30 tons lightweight rocket with a size of about 10feet, where you can probably start turning immediately after launch

And thanks to all for the answers! It helps me to understand why that is so. - I already know the problem and Im able to handle it.

Im just wondering why, after seeing this video. I always thought it was normal to launch rockets straight until a big altitude. So its only in KSP...alright^^

Edited November 15, 2013 by Duke-49th

[bgraves]

Ninja'd means that while they were typing their message, someone else already posted the same thing.

EDIT: Ironically, I thought someone was going to ninja me on posting what ninja'd means.

[EtherDragon]

You can see the curve, its going parabolic from the begin on..no long climbing before the turn... (more words) ... Why is that? ... Thanks

There are several reasons, one of which was pointed out.

Comparing Kerbal Space Program to real world rocket science really can only go as far as the physics and basic concepts - the rest is very different due to several factors.

1. Kerbin is about 1/10th the size, but has the same G force.

2. Kerbin's lower atmosphere is much denser than Earths.

3. The entire Kerbol system is smaller overall.

4. Rocket engines in Kerbal Space Program are less efficient than real world rockets.

5. Real rockets execute a gravity turn by doing nothing... seriously - they actually just sort of start tipping over to follow prograde naturally. Kerbin's atmosphere, size, etc... makes this pretty much impossible to simulate.

In short, the Kerbol system is good enough to learn on, but not at all realistic with regard to size, density, etc... (except for the orbital physics, which is really well done).

[numerobis]

The main reason we don't turn off the pad is that we don't care if our spent boosters land on KSC -- or our failed rocket, for that matter. Ariane loses a few m/s by not going straight right at the very start, but gains by having failure not destroy Guyana.

There is a problem where the drag model is basically equivalent to us always flying big flat pancakes, no matter how sleek we make our rocket. That increases the cost of ensuring range safety. But if we had to choose between paying 100 m/s or killing people (and thus losing the game), we'd pay 100 m/s.

[KvickFlygarn87]

Ether: Rockets' gravity turns aren't automatic; they are constantly managed by onboard systems.

[Kasuha]

Ether: Rockets' gravity turns aren't automatic; they are constantly managed by onboard systems.

Being managed by onboard systems rather than manually means automatic.

I'd call them natural, though. True gravity turn means the rocket is burning exactly prograde except for initial small kick to the side and gravity is what makes the trajectory turn. Of course the trajectory is carefully calculated and minor deviations are fixed by slight changes to engine thrust or gimbal.

See also: Gravity turn (Wikipedia)

The rocket begins by flying straight up, gaining both vertical speed and altitude. ... Losses associated with this slowing are known as gravity drag, and can be minimized by executing the next phase of the launch, the pitchover maneuver, as soon as possible.

...

The pitchover maneuver consists of the rocket gimbaling its engine slightly to direct some of its thrust to one side. This force creates a net torque on the ship, turning it so that it no longer points vertically.. The pitchover angle varies with the launch vehicle ... for some vehicles it is only a few degrees while other vehicles use relatively large angles (a few tens of degrees). After the pitchover is complete, the engines are reset to point straight down the axis of the rocket again.

...

After the pitch over, the rocket's flight path is no longer completely vertical so gravity acts to turn the flight path back towards the ground.

As you can see, even real rockets do the pitchover maneuver after varying length of vertical ascent. And in KSP, the atmosphere is particularly thick.

[LethalDose]

This is a really interesting thread. Because of the thickness of Kerbin's lower atmosphere, is it worthwhile to add aerodynamic structures (e.g. winglets or delta wings) to rockets to reduce the load on or assist SAS/RCS systems during launch?

With more control and stability (and maybe even a SMALL bit of lift), could a gravity turn in low Kerbin orbit become more feasible?

[numerobis]

A gravity turn is mostly feasible in stock KSP. It's just we don't have any UI to help us design a rocket that has the right thrust profile. With a lot of either trial and error, or calculation on the side, you can do everything as described in that wiki page: go straight up for a bit, then a controlled pitch over, then burn prograde all the way up to orbit. Static fins will keep you pointing prograde in the lower atmosphere; higher up you need control input, either gyros or vectoring thrust.

[Duke-49th]

This is a really interesting thread. Because of the thickness of Kerbin's lower atmosphere, is it worthwhile to add aerodynamic structures (e.g. winglets or delta wings) to rockets to reduce the load on or assist SAS/RCS systems during launch?

With more control and stability (and maybe even a SMALL bit of lift), could a gravity turn in low Kerbin orbit become more feasible?

I almost use AV-8B "Winglets" to make tall rockets more stable. Or to prevent them from start to spin.

But you need plenty of them. And still then they are not very efficient. But better than nothing of course.

Ahhh...thats what maintain terminal velocity does I never saw a difference...but to be honest..most of my rockets are so tall and heavy that it is already hard to get 200m/s before Im at 5-10km high. (And I know at least for KSP that you are wasting fuel if you go higher speed..as the density of the air let the rocket accelerate less...kind of ISP...dont know how to explain that in english terms)

I always feel like the ascent Pilot is wasting fuel. I only use it if I launch several rockets..like satellites for Remote Tech...or if Im really lazy^^

And also yes...I use FAR. But as Im always got problems before without it, Im still afraid start to turn in a low height I just recognized that I need a bit less fuel and fairings/cones make more sense with it (at least that is how I understand it).

Ok...this does not work. It still continues to accelerate until 400m/s and more. In other words..it still accelerates "infinite"..

But to be honest..I start to more launch rockets manual as MechJeb is helpful for save time and workload. But is not very efficient for fuel and not very precise. Also with tall rockets they start to going crazy when the gravity turn starts.

I like Mechjeb..but control anything manual is much more precise and safe and fuel saving. Except landing...im too stupid or not enough trained to do a good landing. But still it sometimes do crazy things to my rockets and dont know why. So sometimes its still more safe to land it by hand^^

Edited November 20, 2013 by Duke-49th

[numerobis]

Ok...this does not work. It still continues to accelerate until 400m/s and more. In other words..it still accelerates "infinite"..

Assuming you're talking about the "limit to terminal velocity" setting, what it actually does is limit your *vertical* velocity to terminal velocity. Works great if you have a high TWR at low altitude; past about 10km or so, you don't want to go quite so fast. There's another setting, for maximum acceleration, which is useful to set if you have high TWR.

Having high TWR isn't optimal for mass; you can generally do better with more fuel and less engines. If you have low TWR, then you can run at full throttle the whole time, without a problem.