How to handle fairing lift?

[Dorlan]

Pretty much what the title says. Sometimes I need (want) to put awkward, draggy payloads into orbit or on another body. Sometimes, these payloads are really big, and draggy enough that I can't keep the tail draggier than the nose with the payload exposed. That makes it fairing time, and that opens up an entire new galaxy of problems. The 3.75m fairing generates a ridiculous amount of body lift -- enough to backflip my rockets. To have a prayer of getting even a relatively awkward, large payload into orbit, like, say, this space station core:

Javascript is disabled. View full album

I have to build a custom lifter (as opposed to the fleet of lifters I've already built) and revert a dozen times until I get the ship perfectly balanced in terms of CoM and fly it exactly right. Anything less than perfection will result in a dazzling display of aerobatics, but nothing is going to space today. The other two images in the album show the insanity that is the body lift of the 3.5m fairing. The lift is so incredibly strong that it can backflip a rocket when its nose and prograde marker are only 40 degrees above the horizon. It can backflip a stack consisting of a Mammoth with 2 Twin Boars pushing on the other end and, after it does so, when the rocket starts to fall fast enough it will flip it back upright.

How am I supposed to handle this? I have to be doing something wrong, but what? 2.5m and 1.25m fairings don't give me this trouble. Relatively aerodynamic payloads don't give me this trouble. Only the 3.75m fairing does this, and I'm at a loss as to how to deal with it without engaging in the fun-killing grind of reverting until the stars align and the payload makes it to orbit. Any ideas would be greatly appreciated.

Thanks!

Edited October 11, 2015 by Dorlan

[ForScience6686]

I have not experienced this issue and use the large fairing often. I didn't think they created lift due to the round shape.

- - - Updated - - -

I have not experienced this issue and use the large fairing often. I didn't think they created lift due to the round shape.

I can't see your ship very well but it looks like an odd design. Is the rocket tank off set?

[zolotiyeruki]

The trick is to keep your nose as close to prograde as possible. Start your gravity turn early (around 100m/s), then stick close to that marker. The whole unbalanced lift thing only gets out of control once you've exceeded your craft's ability to counteract it with engine gimballing and control surfaces.

[Dorlan]

Nope, everything is straight stack-node connected from top to bottom, save the fins and the (not pictured) booster stages. Nothing is offset, CoM is in line with the center of the ship, and RCS Build Aid is showing no thrust offset arrows in the VAB. The fairing is kind of an odd shape, it has to be to accommodate the four arms on the station, but I've tried dozens of different configurations for the fairing and they all result in that immense body lift vector coming off the fairing base.

I've also tried forcing KSP to rebuild my PartsDatabase.cfg and getting a new copy of physics.cfg from Steam, on the chance that maybe one of those two files had picked up some bunk values. That didn't fix things, either.

[ForScience6686]

In that case I would agree with above. I think the lift you are seeing is caused by the air stream hitting the side of the rocket. Check if you have the same lift while flying at prograde.

[Dorlan]

Flying prograde is what I'm trying to do. In anything except a rocket with a 3.75m fairing on it, I can let gravity do the work. I start my turn between 50 and 100 m/s depending on the rocket, relax, and watch the show. I maybe have to touch the controls 10 times total during a bad ascent, and that counts staging. For whatever reason, that fairing overcomes gravity pulling the nose down with its lift, even when the engines are fully gimbaled, reaction wheels are running full power, and all control surfaces are working in concert to keep that nose down.

To keep the thing from backflipping, I have to take a much more active approach to flying it. I'm a keyboard jockey, so I have to tap the d key to keep the nose down. If I'm too aggressive, I'll push the nose down too far too early, and that's a completely different way to fail to go to space today. If I leave too much of a gap between keypresses, it's too late; the lift builds up beyond the point of no return, the fairing takes control of the ship and protracted full yaw won't stop that nose from going back the way it came.

That's what I meant about having to fly it perfectly. I did get the station into orbit via launch, but it took way more tries than it should have.

[MockKnizzle]

Maybe just don't build stations that wide. ¯\_(ツ)_/¯ I find that it's often easier to just pack unwieldly payloads in multiple separate pieces and then assemble them on-site.

[MarvinKitFox]

The problem is that you are flying sideways!

Don't blame the fairing, blame the pilot/navigator who thinks it is a good idea to aim 15 degrees off prograde while flying supersonic!

[Harry Rhodan]

Maybe just don't build stations that wide. ¯\_(ツ)_/¯

Or at least don't put the widest section on the top.

[pandaman]

It looks like you're having the same issue I get on occasion. And yes it does seem to be pilot error not helped by having a less than perfect design.

If you have the tech unlocked try locking SAS to prograde immediately after making your initial 'tweak' from vertical. You will probably need to keep adjusting the throttle to avoid going too fast low down, and maybe adjust gimbal limits a bit too, but it's easier to control than adjusting attitude manually all the way.

[tseitsei]

As others have said earlier: It's a piloting issue that only gets amplified because of wide payload.

You should (almost) never point your craft outside the yellow prograde circle during atmospheric ascent. That is what causes extra drag and extra bodylift and makes your craft flip.

From your last 2 screenshots we clearly see that you are way outside the yellow circle and on the last one you have as much as 15 degrees of AoA!! Thats way too much. You should try to keep that AoA (angle of attack. The angle between your prograde vector and the direction you are currently pointing your nose at) at minimum. That way you wont get high drag/bodylift and wont flip your ship...

[Dorlan]

The problem was that the lift was pushing the nose out of the prograde marker. I really mean it when I said it was violent. In the second screenshot in the album, for example, the nose was inside the back side of the prograde ring just prior to the screenshot being taken. I normally keep the nose pointed between the center and the back(towards vertical) edge of the prograde ring during initial ascent. In the time between releasing the "D" key and reaching and pressing the F1 key with my left hand, that's how far the body lift from the fairing pushed the nose back.

Regardless, it seems that if I make my fairings just wide enough to accommodate the parts of the payload as they occur, even if that makes the end result look like I've strapped a water tower to a rocket engine or I'm trying to fly a "this end towards space" sign, the excessive lift problem largely goes away. So, I can keep lofting awkwardly shaped things, I just need to shroud them in an equally awkwardly-shaped fairing. Solved!

As thanks, please accept this picture of the Flying Toadstool. Yes, it made it to orbit.

[ForScience6686]

I would add more fin. Don't be afraid to use wings, you are after all lifting a giant payload, use giant fins. That should help stabilize your dart and give you more control authority.

[Zephram Kerman]

One more possibility. Although I don't think this is relevant in your case, check it out anyway. Sometimes my larger payloads wobble around inside the fairing. If this happens, the CoM is not where it was designed to be, causing phantom control forces, sometimes oscillation, and always eventual 'beautiful spirals'.

The cure I've found is to strut the payload to the fairing somewhere near the nose. To do this requires that trick of strutting to a temporary beam, then deleting the beam.

But, what I think is really going on is the diameter of the payload fairing is much larger than the diameter of the launcher. This would amplify any drag-forward design issue. It can be a little bit larger depending on the length. (That's why your toadstool is stable.) But not this much. I suspect if you upgrade the launcher to something wider, like the NASA parts, things might get better.

Edited October 11, 2015 by Zephram Kerman
early morning grammar fail

[Jouni]

There are two problems I can see. First, the shape of the fairing is bad. Guess which of these two fairings makes the rocket behave better:

Second, the TWR may be too high. The faster you fly, the worse all aerodynamic problems become.

[Curveball Anders]

Sometimes my larger payloads wobble around inside the fairing. If this happens, the CoM is not where it was designed to be, causing phantom control forces, sometimes oscillation, and always eventual 'beautiful spirals'.

The cure I've found is to strut the payload to the fairing somewhere near the nose. To do this requires that trick of strutting to a temporary beam, then deleting the beam.

Thank you, that solved an annoying mystery for me.

I solved my invisible/internal wobble by just strutting the payload better, just as if launched without any fairing.

But strutting it to the fairings is likely even better.

[p1t1o]

Fairings were pi**ing me off too, until Ferram helped me realise that fairings are not intended to reduce drag and make a rocket more stable - they are intended to protect the payload from airflow. A big fairing over a big payload will always be draggy, you can minimise it, but *reducing* drag is not what they are designed to do.

[DancesWithSquirrels]

I use the trick (someone demonstrated it on a video, I would credit them if I could remember who) of creating hardpoints.

* Build the stack without the fairing, using a decoupler smaller than the booster diameter.

* Add some small 'towers' of small cubic struts (3x or 4x symmetry, 3 or 4 cubics stacked) to the top of the booster under the planned fairing location, outside the decoupler and near the edges of the booster. They just have to be a little longer than the fairing [base] thickness.

* Then insert and construct the fairing, smooth is better even at the cost of a bit of fairing weight.

* The cubic strut towers should now protrude through the fairing base, and you can solidly strut from the hardpoints to your payload.

Ends up solid as a rock.

Edited October 12, 2015 by DancesWithSquirrels

[cantab]

In your original design you have a payload fairing that bulges out to about twice the diameter of the launcher. What were you expecting with a shape like that?!

Of course sometimes we want to launch unaerodynamic payloads. Often the simplest solution is to put whacking great fins on the bottom of the rocket.