Taking off from the runway

[Benno]

Hi all,

Well, I'm not a complete noob at Kerbal (have landed some craft on the Mun, drove under a Mun Arch, put a rover on Duna, all selft built), but spaceplanes have been in my too hard basket for some time. But, I've made a few small ones and gotten them to fly, but never in to orbit and always somewhat unstable. But, I'm planning my first mission to Eve, and as part of that mission I wanted a craft capable of exploring Eve, landing and capturing some CO2 from the atmosphere, and bringing it back to orbit for conversion to oxygen so my little guys don't suffocate (I'm using ioncross lifesupport and Kethane). I figured a kethane powered space plane was the most viable option for doing this given Eve's gravity and atmosphere, but am happy to land it near a ground based fuelling station before heading back in to space.

Anyhow, that's all a very long way off, because I'm yet to actually get my plane off the runway in less than 10 separate pieces, let alone assemble the mighty colony ship. The problem is that soon after firing up the engines, it starts pulling to one side (the left), rolls and crashes. SAS/ASAS helps a little, but not much. It crashed rather quickly when using the two kethane jets, but can sometimes make it to the runway when using the nuclear engine.

I've been pulling my hair out for countless hours now trying to get off the ground, have tried different balancing of COM and COL, single sets of landing gear, doubled up sets, angled sets, multiple sets attached to wings or fuselage, putting the gear on backwards, tail fins, no tail fins, different flaps settings etc. No matter what I do, it always pulls off quite sharply after a little while, but it seems to be more reliable with less thrust or with the nuclear engine only, but of course then it doesn't go fast enough to get off the ground before the end of the runway. Below is my latest incarnation (i hope the steam links work) with lots of landing gear, which seems to help. Flaps are usually set to 1 as the problem seems worse at higher settings. The two kethane jets are stapped underneath the wings, as this seemed to work better than above the wings or on the wing tips.

I should also mention that I'm using FAR, as that seemed to make things easier a while back. I'd love to be able to refine it so that it flies better and is stable at speed and altitude, but obviously I need to figure out how to get off the ground before that's even possible.

Can anyone help? What am I doing wrong?

Edited July 25, 2013 by Benno
Answered!

[Snooze]

You have far too many wheels, for one. Two, you should stick the nose wheel (if you're insisting on a tricycle configuration) somewhere on the fuselage that isn't the cockpit.

[Johnno]

Can anyone help? What am I doing wrong?

Judging purely from pictures (I don't have all those mods) I'd say your CoM/CoL are too far forward. What's likely happening is that once you get lift your rear wants to take off but can't, presses down on your forward gear which buckles minimally, making your plane turn.

To be able to take off without issues your rear landing gear has to be quite close to your CoM, the further behind it is the higher lift/speed you need to be able to take off.

Try moving your CoM (and CoL) closer to the rear and move the rear landing gear toward them.

[K^2]

Your CoL needs to be closer to CoM. Having it too far behind makes it difficult to take off. The other problem is landing gear, as Snooze said. It's ok to have too much, but the further back the rear-most gear is the harder it is to rotate for takeoff. Your rear-most landing gear should be only slightly behind CoM as well.

[AletzN1]

Well it might be that you have very little lift or you dont have enough control surfaces to actually get yourself pointing up, because you are at 100 m/s most aircraft go up at 40-80 m/s even a super heavy bomber I made took off at 50 m/s

When an airplane goes to the side is pretty much because they are going to fast and should pull up earlier. You should first try putting more wings on it :/ See if that works

(Note: I am not an experienced FAR user)

For spaceplanes if you are just starting you shouldnt use FAR, First get used to ksp default aerodynamics and THEN try FAR

[AletzN1]

Also your back-wheels are too behind you should place them more close to the CoM so you can point up with less than 100 m/s thats what my heavy bomber uses and it takes off at pretty slow speeds

Just as Johnno said.

[TMS]

Yeah, the gear should act as a pivot point and allow the plane to lever upwards

I always find that having multiple sets of gear never really works for me. Strip it back to the minimum.

[HoY]

Also too much mono propellant and not enough Kethane storage! Kethane doesn't need fuel lines it routs itself like monoprop, so you can add more tanks wherever it's convenient. I'm not familiar with that blue monoprop tank at the back but I assume its O2? Will you need that much between the ground base and station in orbit? Or is that to bring it up to convert it?

[Grand Lander]

Okay, just a suggestion (I only make atmospheric aircraft) But I find having the mass and lift (icons) on top of each other make a nice stable craft.

Also it maybe rolling over to it side due to there being more mass at the front of the craft and the instability of the configuration of the wheel (<---Thought).

I hope these suggestions some how help you out, or zero out the different possible variables involved.

[mellojoe]

Also, try to have your rear landing gear slightly higher than your front gear. This causes the entire ship to lean back while sitting on the runway. It makes takeoff a breeze.

((edit))

Also, have an even number of tailfins. I don't know why but it seems to help. As does slightly angling your wings or tailfins. Sometimes this helps stabilize the left and right forces.

[Rune]

Forget what everybody says: they are right, but it's not your problem.

Your problem is actually structural rigidity: you may have the wheels pointing straight in the SPH, but what good is that if as soon as you get to the runway the whole plane starts flexing around. Actually, putting the wheels on pylons is even worse for that, since most supposedly structural parts have the rigidity of a rubber band, especially around their attachment points. If you don't believe me, try lifting again, but this time with the camera pointing at the wheels during takeoff, and enjoy the landing gear dance.

My suggestion is to place them on the heaviest parts of the plane (i.e: the fuel tanks that make up the fuselage), and then strut the whole thing firmly, till you can be sure than applying a few tons of force (your engines on) won't make the landing gear bend out of position. Preferably, with no parts between them and the plane.

Also, more lift makes you lift off slower, but I think that's kind of obvious (lift is proportional to speed).

Oh, and if you can't pull up (which looking at the plane, I'm not sure you could), then try moving the rear landing gear closer to the center of mass: think levers and Archimedes, and you will know why that works.

Rune. FAR has nothing to do with this, BTW.

[espm400]

Although it's true that you might want to move your CoL and landing gear (you want your rear gear very close behind your CoM) a bit closer to your CoM as not only will it help you do your take-off roll at a lower speed (thus reducing your time on the runway), but IIRC as you hit supersonic speeds, your CoL actually moves rearward which will will make your spaceplane harder to control.

As for the problem you actually mentioned (one that happens with a lot of my planes), and that no one save Rune seems to want to address, I would suggest re-mapping your rover controls (also controls your landing gear) from the WASD keys to IJKL instead. Not only will this make driving rovers easier (you're not trying to pitch your craft whilst turning or accelerating), but with your RCS off, you're only moving the wheels (or in this case, landing gear) and not using the torque from the command pod. That's how I solved it for my planes.

Whomever that was that suggested not using FAR until your used to the stock drag model, I disagree. Considering that down the road we will probably see a stock drag model similar to FAR, why not get used to it. Also, the stock model doesn't work properly when using planes a FAR is much closer to a real life analogue than stock (that is unless I'm wrong and jets like the SR-71 actually pointed their nose 40* in the air to maintain level flight above 60,000' @ Mach 3+).

EDIT: Rune is right as well (forgot to mention), you could probably go with more structural rigidity (MOAR STRUTS).

EDIT 2: Just to illustrate the difference that the positioning of your landing gear makes, take these two examples:

a) This is one of my early SSTO Craft. Notice how far back the gear are from the CoM. My take-off roll in this craft happens at about 180-200 m/s

This is the B-1B Lancer I've been working on (keeping the swing wings rigid is a massive problem, but that's neither here nor there). Notice how close the rear landing gear are to the CoM. The take-off roll in this craft happens at 50-80 m/s , despite having considerably less thrust.

Edited July 23, 2013 by espm400

[Benno]

Wow. Thanks for all the tips everyone. There seemed to be a few common themes:

1) Move the CoM back a bit

2) Move the CoL closer to the CoM

3) Move the wheels closer to the CoM

Yes, the pivoting makes perfect sense now that I think about it. My plane was definately too heavy at the front with that fuel filled fuselage, and the back was mostly likely trying to lift first and placing pressure on the front. So, I've taken on board all of your advice:

and, tada!

I'm airborne! And, it's even pretty stable. Managed to get it up to Mach 2.5 without flipping.

Thanks everyone for the excellent advice. Now, getting it into orbit.... For some reason it insists on levelling out at 12-15km. But, that's another topic I think.

Thanks again everybody.

Benno