[FAR] Can't seem to get into orbit any more :(

[Hyperlynx]

I've played KSP for a fair while now. I started a new campaign when v1 came out, and now I just can't seem to build a rocket that gets to space. They all seem to be very unstable, with a tendency to turn end-over-end.

NOTE: I am using the Ferram Aerospace Research mod, so it's not the stock aerodynamics model. I was using it before the v1 change and things seemed to be ok, but now I'm stumped.

Basically, I'm not sure what I should be trying to do with my rocket design, and I'm not sure if my ascent profile is correct. Also, it seems to be very hard to actually fly my ascent profile correctly without the thing pitching over much too quickly, then flipping end-over-end.

Here's a two-seater, intended to take tourists into orbit:

... after SRB separation:

... and second stage only:

I've tried to keep the centre of lift behind the centre of mass. The only way I know how to do that is to add fins (which doesn't seem very rockety), and I think I've added them basically everywhere I can.

As far as ascent profile goes, I'm trying to fly directly up until I hit 50-60m/s, then pitch over 5 degrees down facing east and let gravity do its thing. What ends up happening is the rocket strays off directly vertical unless I use SAS, and when pitched over to 5 degrees it pitches over further than that and I have to fight it to stay on course. After separation, all bets are off and I generally lose control of the rocket.

I've thrust-limited each stage to around 1.3 TWR, and I'm limited the gimbal range to 25%.

So... what am I doing wrong? What should I be looking for in the vehicle designer? How can I fix up my flight plan?

Edited July 8, 2015 by Hyperlynx

[Streetwind]

The golden rule is not "fins everywhere I can". The golden rule is also not "CoL behind CoM", that's for vessels relying on lift, i.e. planes. On a rocket, you're not dealing with lift, and so you don't have a center of lift in the first place; the thing that should technically interest you is the center of pressure, which is something different, and I don't think KSP even calculates it.

The golden rocketry rule is "heavy bits at the front, draggy bits at the back". Just the way an arrow is constructed: the heavy head at the front, the draggy feathers at the back.

Fins are draggy bits, so first things first: remove them from the upper stage. Fins should never go on upper stages unless you know exactly what you're doing and have a very special purpose in mind.

Secondly, I bet that if your rocket flips over, it does so backwards, i.e. tilting towards the camera in the first screenshot. Or, if you overcorrect, it will flip to opposite way. It will never flip sideways.

How do I know that? Because there are three control axes (yaw, roll, pitch) and you only stabilize two of them with your fins (yaw and roll). Pitch is completely unstabilized, and that not only makes it easy to flip in that direction, but also the reason your manual pitchover is so violent. You need a minimum of three fins in radial symmetry to stabilize all three axes; four is better. So take the pair of fins off the bottom of the central stack, switch to quad symmetry, and put them back on. Yes, I know you have SRBs attached radially. You can shift your fins to sit diagonal, that works fine too.

Thirdly, your rocket is unsymmetric. How did I know the rocket would tilt over backwards if not steered? Because there is a canopy on one side of that inline cockpit that's not on the other side, and that canopy creates drag that pulls the rocket nose in the direction of the canopy. If you use this part in a rocket under FAR, which goes directly off of vessel shape, you have to account for this. Your other alternative is to build a tourist vessel with basic pods only, for example three side-by-side. But that also creates a draggy top stage, and you'll have to fly carefully to prevent flipovers too. EDIT: What you could also try is flipping that inline cockpit 180 degrees so that the canopy is on the belly of the rocket. Maybe you can let it help drag you into the gravity turn that way.

Finally, you want more push on launch stages than on upper stages. With your vessel, it's probably better to go with TWRs of 1.5, 1.3 and 1.0 instead of 1.3 across all three stages. DV-wise, you are almost always better served with not thrust limiting your launch stage at all. The key thing to watch out for is maintaining control over your gravity turn; a high thrust first stage with large fins doesn't like to turn much, so you have to pitch over earlier and more aggressively, which can be a scary thing if you're new to the game and may cause piloting errors. If you want to play it save and comfortable, you can go with less liftoff TWR, but it's not as fuel efficient.

EDIT: Oh, and a trick Scott Manley recently showed off in one of his videos: You want to keep more heavy bits at the front fo the rocket, but fuel tanks empty top to bottom, which equals removing heavy bits (fuel is your main heavy bits) from the top. You can counter this by constantly pumping fuel from your center stack's bottom tank to the center stack's top tank during ascent as that top tank empties. Don't get distracted from your piloting though!

Edited July 6, 2015 by Streetwind

[Wolfox]

I´m not using FAR, so my advices might be a little bit odd, but I personally would relocate tail fins from SRB´s to main rocket, to keep them producing a bit of drag even after detach of SRB´s, so your rocket keep facing "towards space". When I place winglets on SRB´s, its mainly to keep them off rest of my ship when detached. Another thing, your rocket seems to be very tall and thin, whats a different name for flying disaster. Try to keep it shorter, and more bulky. If you don´t have yet wider fuel tanks, try using two or tri-adapter and place 2-3 tanks underneath, each capped with own engine, and strut them together near the engines.

[Hyperlynx]

Thanks Streetwind, I'll try your advice. Those vertical fins on the SRBs are a recent addition; I was getting these problems almost regardless of fin configuration. I'll see how I go with the other details, though.

Wolfox, I don't have those parts yet, so it's a no-go.

e: no, it's still pitching to the right....

Oh. Well, that would do it. That probably wasn't helping my stability either!

ee: no, it's still veering to the side. Can't work out why. I assume it must be asymmetrical in some way...

eee: well, I had to have Jeb SAS the craft up, then SAS to prograde, but it's worked.

Edited July 6, 2015 by Hyperlynx

[capi3101]

I fly FAR my own self, and I do a fair number of rocket launches in addition to planes (my tech is still low end, so I can lift light stuff via plane but still rely on rockets for payloads much above ten tonnes). I pretty much agree with Streetwind's assessment of your rocket. I'll add that those AV-1Rs on the upper stage don't steer, so they don't help you keep your heading at all, and then I've found that the Delta-Deluxe Winglets aren't really the best choice for steering fins since it's only a little piece of the fin that actually steers. My suggestion would be to go with Tail Fins, or (barring that) AV-R8s, both of which are fully-deflecting surfaces. Those will help your steering out.

FAR these days voxelizes your craft and analyzes its drag profile from that data; a long, skinny craft will generate less cross-sectional drag than a wider, shorter craft. From that, I might suggest losing the SRBs entirely in favor of another lower inline stage, but before I'd do that I'll ask how far up the tech tree you are - do you have access to any 2.5 meter parts? You'll have Modular Girder Segments for sure (as they're Starting Tech), and can use those to make engine clusters for moar power on the lowest stage if necessary.

Your flight profile might also need some work - here's what I do: launch and stay vertical to 5 kilometers, then turn eastward only by about 5 degrees. Hold that until 7 kilometers, then turn another five degrees. At 9 kilometers, slowly pitch at 5 degree increments until 15 kilometers, when you should be at 45 degrees or so. Now, if you've maintained stability - and a slow pitchover is the key there - check your time to Apoapsis. If it's 35 seconds or more, slowly proceed to pitch down to 25 degrees above the horizon; at 45 seconds or more, pitch to 10 degrees above the horizon; at 55 seconds or more, pitch to the horizon and keep burning until you reach your desired Apoapsis height.

Suggested mods - Procedural Fairings (much nicer than the stock ones; using those here would compensate for the un-evenness of the Mk1 Inline Cockpit's Canopy) and TAC Fuel Balancer (this one would help with the idea of shifting fuel back into the upper stage tank; you just set it to "Transfer In" on both Oxidizer and Liquid Fuel and set the pump rate to maximum, and fly. TAC will take care of the fuel handling for you for as long as you need it).

[Hyperlynx]

Removing the SRBs and going with only a stick design I've managed to get a comms sat into space. Thanks capi, that's done the trick.

(I'm using RemoteTech, so I need the com sat for low kerbin automated stuff)

[Streetwind]

If removing radial stages got rid of your vessel instability, then your issue was flexing at the decoupler. An alternative solution may be to strut from the nosecones of the boosters to the upper stage.

[Khazar]

What is your speed when you lose control ? And what alt ? How many reaction wheels do you have ?

Also have you checked what torque do you have at each stage ?

[John FX]

I would suggest not using the inline aero cockpit for a space craft and instead choosing one that is more round if using the latest FAR.

EDIT :

If removing radial stages got rid of your vessel instability, then your issue was flexing at the decoupler. An alternative solution may be to strut from the nosecones of the boosters to the upper stage.

Of course. I didn`t see the boosters were unstrutted.

That`s your problem. Have 3+ radial fins and strut your boosters.

[Hyperlynx]

I don't yet have struts or multi-kerbal cockpits. Making do with pretty crap technology.

[Franctireur]

Hi Hyperlynx,

I'm pretty new to Kerbal, but i see something on your model and noticed myself with my own experimentation: radial parachute can be a problem.

Your radial parachute aren't symmetric on top, and i think but i'm not sure can cause some aerodynamic turbulence?

What veteran players think?

[Hyperlynx]

Aha!

I spent the $75,000 for the astronaut complex upgrade, did some EVAs and spent the science on struts. Now that I've strutted up my boosters, things are MUCH better.

Hi Hyperlynx,

I'm pretty new to Kerbal, but i see something on your model and noticed myself with my own experimentation: radial parachute can be a problem.

Your radial parachute aren't symmetric on top, and i think but i'm not sure can cause some aerodynamic turbulence?

What veteran players think?

Yes, I took that screenshot to show that I had found that problem. Unfortunately, it was not the only problem with my design. The main problem was that the radial decouplers were not keeping the boosters straight. Struts has fixed that problem.

[smunisto]

And nobody advised the user to just use the "Atmo" button. Judging by your initial screenshots, your TWR is probably way too low to take you into orbit. I bet if you press atmo - it will show values of less than 1 TWR and less than 2500 dV.

This should be your starting point. The other problems are very very minor - fins are not nearly as draggy enough to stop a rocket. Non-symmetric parachutes can lead to spin, but first look at the "Atmo" button and start adjusting your values from there. Ideally you want about 3k dV on your "Atmo" stages - i.e. the ascent stage.

[Hyperlynx]

The problem was definitely, totally that the solid boosters were slewing. Struts solved it.