Help! Spaceplane Tumbles Like Mad!

[JYNg]

Hi guys,

I was trying to design a spaceship/spaceplane hybrid that is similar to Dyna-Soar. I've managed to get it into LKO, rendezvous with my space station and return to Kerbin.

It survived the re-entry pretty well, but the issue came right after the re-entry: prograde pointed right towards the ground, and my ship started tumbling like a washing machine. 

Would really appreciate some help if somebody could guide me on what's wrong with the plane. (I've checked the COM and COL, COL is constantly behind the COM even when the tank is empty. 

Thanks!

Edited June 30, 2016 by JYNg
Spelling correction

[Van Disaster]

You still have the problem of no pitch lever arm - you can rely on the cockpit reaction wheels somewhat ( from the look of the craft you're currently relying on them completely ) but that only lasts as long as you have electric power, so it's generally always best to seek an aerodynamic solution. I use FAR so things are slightly different, but I'd recommend something like this - notice how far the elevons are from CoL.

For some reason the CoL ball didn't show up at all, but it's about 1/3d forwards of the rear of the centre tank, so roughly where it is in your shot too.

As a side note I just put that into orbit using a single large SRB

Edited July 2, 2016 by Van Disaster

[bewing]

Your pitch control comes from the tailfins mounted at the sides. When empty, your CoM is too close to those fins, and they don't have any leverage to keep the nose up. Maybe you can move those fins back, or maybe forward, or maybe move them back and add another pair at the front end.

[Van Disaster]

Add fixed wings and *then* pitch controls. Tilting the entire wing doesn't change pitch unless you've mostly got body lift, and in that case they'd be at the back where the tail is ( for the same reason the tail is at the back ).

If the wings are using those canards but they're unmoving, then you have no pitch control at all.

[Van Disaster]

Picture CoM as where the craft body is resting on a fulcrum, and each control surface as a force pushing up/down on the craft body through the middle of the surface ( overall CoL also ).

[Snark]

Also, note that having the CoL behind CoM doesn't necessarily mean you're stable.  If your CoM is near the back of the plane, you're likely to have stability problems regardless of where your CoL is.  Could you post a picture of your plane in the VAB  (just the "plane" part, without all the boosters and such) with the CoM display turned on?

Also, just to check:  make sure that your vertical stabilizer is enabled for yaw control only.  Especially make sure that it has roll authority turned off.  Vertical stabilizers with roll enabled cause a lot of problems with spaceplanes-- they end up fighting themselves and you lose a lot of the stability benefit.

[FullMetalMachinist]

2 hours ago, JYNg said:

@Snark I may have found one of the problems. 

The last I checked in VAB the CoL was right behind the CoM (that was when the plane was vertical). Then when I opened the same plane in SPH again I noticed the CoL is now slightly in front of the CoM.

What happens is that the VAB uses "airflow" that is coming straight down from the ceiling to calculate the position of the CoL marker. However, the SPH uses "airflow" at a 5 degree angle of attack, with it going from the front to the back and slightly up. 

If you use the rotate gizmo on the root part and rotate your ship slightly nose down, then you should see the same thing you saw in the VAB. 

Also, what @bewing said, empty is more important, because that's the state it'll be in when it reenters. 

[JYNg]

Thanks for the suggestion. I'll work on an improved model based on them. 

[JYNg]

@Snark I may have found one of the problems. 

The last I checked in VAB the CoL was right behind the CoM (that was when the plane was vertical). Then when I opened the same plane in SPH again I noticed the CoL is now slightly in front of the CoM.

[bewing]

But is that when it is full of fuel, or when it is empty? Empty is a much more important scenario.

[JYNg]

@bewing @FullMetalMachinist

The CoM when it's empty is roughly at the same location. 

[JYNg]

14 hours ago, Van Disaster said:

You still have the problem of no pitch lever arm - you can rely on the cockpit reaction wheels somewhat ( from the look of the craft you're currently relying on them completely ) but that only lasts as long as you have electric power, so it's generally always best to seek an aerodynamic solution. I use FAR so things are slightly different, but I'd recommend something like this - notice how far the elevons are from CoL.

So I've worked on a solution based on the feedback from this post and I've managed to come up with the following craft:

Kind of similar to what @Van Disaster did, but the wings ended up in a shuttlecock-like configuration to get the CoL further behind. Tried flying the plane on a jet engine and it works relatively stable compared to the earlier design.

But unfortunately it introduced another problem. The lift generated from the wings became much larger, so even the slightest tip of the rocket during takeoff would result in the whole rocket making a 180-degree turn, sending Jeb right back onto Kerbin then to Kingdom Come.

Oh well, that's the whole essence of KSP: keep running into walls until we get it right. This is going to take me a while to solve it.

Side Note 1:

I guess this is why Dyna-Soar, SpaceShipOne and SpaceShipTwo have swept-back wings, to move the CoL further back I guess.

Side Note 2:

And this is probably why Boeing X37-B has a fairing covering it. The aerodynamics is a pain to deal with during a vertical launch.

Side Note 3:

@Van Disaster I didn't know a single SRB is powerful enough to launch it into LKO!

[JYNg]

Finally got the design to work! Special thanks to @Van Disaster, @bewing, and @FullMetalMachinist for the useful feedback!

Here's an album of the successful launch: