Rocket pitches over uncontrollably during gravity turn.

[WhiteWeasel]

Was there any change to the aerodynamics or pod torque that I'm not aware of?

Edited July 25, 2013 by WhiteWeasel

[Anton P. Nym]

I know 0.21 changed how pod torque works but I don't have details. I gather SAS is now replaced with reaction wheels to add countervaling torque, and that ASAS is now more like a fly-by-wire system than an autopilot.

I suspect I'll have to revisit a lot of my designs when I get a chance to install 0.21.

-- Steve

[Vonar]

I've noticed that some particularly big rockets have always had a pitch over and die on you problem when doing a low gravity turn. Instead of starting at 10 km try waiting longer to 15-18 km.

[WhiteWeasel]

I've added reaction wheels, and all they do is delay the spinout.

[WhiteWeasel]

4 Reaction wheels! They do nothing!

ASAS you do not make my rocket go 180 for the lulz

[WhiteWeasel]

4 Reaction wheels! They do nothing!

ASAS you do not make my rocket go 180 for the lulz

[Eric S]

Well, the cupola went from being the strongest rotPower command pod by a large margin to being barely stronger than the Mk2 lander-can, and the command flywheels aren't as powerful as the old rotPower stuff was (even the mk1-2 went from 20 rotPower to 12.5 torque). So if 1 rotPower=1 torque, you'll need to add (30-9)/3 = 7 reaction wheel parts to get the same command authority.

[UmbralRaptor]

I would be surprised if that craft worked in 0.20.2 -- it's quite unstable aerodynamically (high drag parts near the front, low drag parts near the back).

[Ruedii]

Two things

1. Try not doing your gravity turn so rapidly. You should only pull your nose at most 10 degrees off of your current trajectory at any one time.

2. Don't put fins in the middle of your rocket. The fins need to be at the back of the rocket to keep it stable. Remember, rockets are not airplanes, they have a thrust to weight ratio well over 1.1, often exceeding 4.0. This means that they need no wings, just tail control surfaces. The purpose of the tail control surfaces is to swing the vehicle back so the nose points towards the trajectory. It does this because the center of lift is BEHIND the center of gravity.

[Anton P. Nym]

Taking another look, the design starts of slightly top-heavy (centre of mass is ahead of centre of lift/pressure), gets less so when the side boosters drop, and then gradually more so as the Big Orange Booster empties. If you got out of atmosphere in time you'd have no problem, but it looks like you're still in atmosphere when the design goes back to being top-heavy and then your rocket tries to fly like a plane... a plane without wings, that is. Adding reaction wheels above the CoM mitigates this but make the top-heaviness worse... so that's why you're seeing limited results. Add in that reaction wheels aren't as powerful as the old "shake-rattle-n-roll" SAS and I think that's why your design doesn't work in 0.21.

Removing the upper fins would help move the CoL back behind the CoM, increasing the rocket's stability. That might be enough to get you out of danger all by itself. If not, maybe try canards in place of the aft-most fins and give the new ASAS more to work with... though as I haven't played with 0.21 yet that's a guess.

-- Steve

[Vonar]

After a closer inspection I don't see any RCS on that design. Maybe add the big yellow RCS tank, and some RCS engines then try again?

[WhiteWeasel]

Would it fly better now even though it has less control authority because of less fins? And counting the insertion stage how 16 tons a heavy payload? Also all my rockets flew just fine without RCS. Infact scott manley says it's wasteful to use it like that. You only need it for docking in most cases.

Edited July 25, 2013 by WhiteWeasel

[maccollo]

Having aerospikes in an atmosphere is suicide, get rid of them. Since they have a drag value of 0.1 instead of 0.2, the rest of the rocket acts a giant sail. As soon as you deviate just a little from the motion vector your rocket will allays flip out, and recovery is impossible.

It's sort of ironic that they are absolutely useless for their intended purpose.

Just swap them out for the Lv-t30 and your rocket will fly just fine.

Edited July 25, 2013 by maccollo

[WhiteWeasel]

Having aerospikes in an atmosphere is suicide, get rid of them. Since they have a drag value of 0.1 instead of 0.2, the rest of the rocket acts a giant sail. As soon as you deviate just a little from the motion vector your rocket will allays flip out, and recovery is impossible.

It's sort of ironic that they are absolutely useless for their intended purpose.

Just swap them out for the Lv-t30 and your rocket will fly just fine.

That pretty much fixed it. ASAS isn't as reliable and have to keep my fingers on WASD all the time, but still.

[SRV Ron]

I have notice that one simple design with winglets did the flipover at 10,000 meters, totally uncontrolable. Removing the winglets made it stable. You will now have to add some counter steering when doing the gravity turn so you don't pitch over.