Getting to orbit without SAS

[messiahua]

I've always used SAS, but right now I'm trying to get relatively simple unmanned rocket to orbit completely without SAS using gravity turn to finish one contract.

I've tried dozens and dozens of times with different design changes. I've tried lowering speed, switching to less powerful boosters, more fins/winglets, but it's unstable even at launch and rocket tends to tilt in one direction without me pressing any buttons.

I don't understand why, it is perfectly symmetrical, nothing special at all. If I try to correct course manually it is hard and very imprecise, at some point (usually when I detach boosters) I completely lose control and can't do anything. Right now it looks like this:

I've search all possible information, read several tutorials and can't get any solution. I'm really tired of this. What's wrong with this design?

Edited October 19, 2016 by messiahua

[Vanamonde]

This vehicle is more than heavy enough to flex under its own weight. As soon as it takes off, that compression is released and it's going to swing in some random direction as the structure rebounds. Mounting it on launch clamps would help a little, but really, you're trying to avoid the thing SAS was made to avoid, so without SAS, it's just plain going to be a fight to keep it on course. 

[DD_bwest]

going without sas can be a real challenge, but it is certainly alot easier now with the fuel flow options.  draining your tanks bottom to top

taking advantage of the new auto strut and rigged atrachment, if your total part count is low, you can split your orange tank into smaller ones and have them drain upwards, steadily raising your center of gravity higher, making it more stable.   

also as mentioned above, launch clamps will help alot. when that thing loads, it will bend and flex.

 

i also tend to find SDB's to be chaotic, making it harder to control them, and they dont follow your prograg as well when not using sas. but that may be something to do with my designs.

also if you tip a little in the wrong direction, like enough for a 3 to 4 degree inclination. it can be easier to fix that little bit further into your gravity turn when your thrualst vector is more horizontal.    atleast i find that easier

[FullMetalMachinist]

The short answer is that your rocket is not aerodynamically stable. 

I know you said that you've already read some tutorials, so I hope that I'm not giving you something you already know, but this is a good tutorial for flipping rockets. 

So basically your CoM is too far backwards because the combined mass of those two Hitchhikers is only 5 tons, much less than what that tank and engine masses. This means that it's stable if it's flying backwards. 

In your case I would get rid of the side boosters and add a second in-line stage. This will not only shift the CoM forwards slightly, but also give the fins at the back a longer lever arm to try and force it to fly front first. 

[bewing]

It's also a lot easier to keep things flying straight, if you start them out flying straight. Try launching straight up. Forget the gravity turn. Go straight up until you are at 42km or so, and then you can start turning. Turning and steering is destabilizing by its very nature.

Yes, I know it's less efficient ... but a little extra fuel is worth it if it actually works.

[Allocthonous]

Steering really is wobble-inducing. I've had some success launching satellites in the 1 ton range using the largest solid booster and a fairing without SAS, by using launchpad clamps or supports or whatever they're called and tilting the whole rocket by 5 or 10 degrees eastward. In theory the concept would scale up, but you might need more oomph to your engines to keep the gravity turn shallow enough. Fins also help.

[Plusck]

That's a Mainsail, isn't it?

Odd thing I've found is that thrust on a Mainsail is most definitely not centred. So without SAS, it's probably just pushing you constantly off course.
In any event you shouldn't need a Mainsail - the Skipper is the better engine if you don't need all that thrust.

 

Second thing, don't Hitchhiker cans pass fuel through? You could mount them under the orange tank and solve a lot of your weight-distribution problems. You don't appear to have any engine in the payload so it doesn't really matter if the empty orange tank stays attached, does it?

 

And final thing - in KSP 1.0.5, during testing I found that those fins were actually slightly worse at keeping a prograde heading than the very basic, cheap fins. They have more surface area, but in your case it's still a very small surface area compared to the size of the rocket.

[messiahua]

Thanks everyone for answers. Here are my comments:

8 hours ago, DD_bwest said:

but it is certainly alot easier now with the fuel flow options.  draining your tanks bottom to top

I didn't know about this feature. I've searched and found that to use it you need to enable "Advanced tweakables" in settings. I've tried some more variations and found two solutions:

1. Split fuel in two stages. As far as I see it's more stable solution, but not very optimal financially, because you need to use decoupler and second engine.

2. Lower fuel flow priority for higher tank, which I guess is the best solution.

That's how it looks right now:

 

At last I can get to orbit! The only problem is strange constant instability at launch, when rockets slowly tilts in one specific direction (west, south-west) and I'm using Skipper engine. It's hard to keep ascend at 90 degrees. I've also tried Launch Stability Enhancer, but it doesn't help.

FullMetalMachinist and DD_bwest gave the best answers IMHO.

Edited October 20, 2016 by messiahua

[Muskrat]

The tilting off launch is because your not moving fast enough for the aero parts to stabilize you. Think of it like trying to stand a pin on it's head. Rockets are stable in flight with higher drag at the back, but with the higher mass at the front. This is the opposite that's needed for very low speed stability.

Most use SAS to mainstain stability during this part of flight (until reaching 50-100kph). You could possibly trying a quick burning, high thrust booster stage to get you to an aerodynamically stable speed quickly. I haven't done this, it's just a wild idea I'm throwing out there.

 

Is your goal really "no sas" or is it minimal steering (true gravity turn)? If the later you might consider just this small use of SAS.

[Plusck]

4 hours ago, messiahua said:

...and I'm using Skipper engine...

Yes I see that now. My bad.

[Vaporized Steel]

Not only is the COM to far down, the hitchikers that are lightweight in comparison do induce drag. And drag is not displayed in the VAB by the aerodynamics display. But you will see its effect if you press F12 in fight which opens the aerodynamics forces overlay.
If alot of red lines start to visualize themselves on the top of the rocket when you turn with F12 pressed in flight you'll know you have drag issues. And because theres less mass at the top and thus the COM is far down the drag works like a lever.

Because the COM is far down the Engine gimballing has less effect, so you have no authority to combat the drag while turning.
And because the fins are even higher (probably just under the COM) you probably have similar flight characteristics if you pull them of completely.

That^ and everything everyone else has said.

Edited October 20, 2016 by Vaporized Steel