space plane control issues on approach

[ForScience6686]

Good morning everyone,

I've spent my weekend developing a space plane to go out to duna and return my crew. My question comes down to wing placement. I was understanding that a col above the com would at stability and self righting characteristics for level flight. My setup has the main wings mounted low and angled up as well as a fairly high incidence. My thoughts were that this would help stabilize re entry and keep it fairly easy to maneuver, but it's been a bit tricky on final approach. I was also wandering about the stability of a v shape tail instead of the traditional fashion.

One last thought was my tail and wing sections being the same size, delta wings, but my col and com are properly placed. I figure with the v shape tail, I should only really experience a partial loss of yaw control.

I will try and get pics later, but for now any ideas on the setup I'm using?

Edited October 19, 2015 by ForScience6686

[Snark]

I've found that it's generally not worthwhile to try to put the CoL below (I.e.ventral to) the CoM. Theoretically that would help with level-flight stability, but only on the roll axis, which I have never found to be a problem. The only stability that's ever an issue, in my experience, is pitch stability, for which all that matters is the fore-to-aft position of CoL along the ship's axis, relative to CoM. And putting the CoL below has the potential to make it harder to keep the center of drag and of thrust coaxial with the CoM, which can complicate control.

I generally prefer to keep my wings on-line with the ship (I.e. symmetric about the axis), with CoL just slightly behind the CoM. Then put some control surfaces at front and/or back of the ship for controlling pitch. For aerobraking or landing, just pitch the ship up.

[ForScience6686]

So I have it backwards, I currently have col above com. I am currently experiencing roll that is hard to control and keep level. This had them lead to over compensation and to much turn which generally ends in a flat spin.

[The_Rocketeer]

Wing placement is pretty complicated. I've found that an/dihedral wing arrangements tend to add uncertainty to steering input, causing wandering and drifting on other axial rotations than the one you want, even when control surfaces are configured properly. They can also leave the lift platform with 'flat spots' where it tends to settle somewhat off-upright. Also, V-shape tails can cause similar unwanted pitching and rolling issues when yawing, which the SAS fails to properly correct because you're triggering it's 'stability assist' mode through control input.

The relative size of your main- and tail-wings probably doesn't have much to do with it if, as you've said, your CoM/CoL are correctly placed. You're just halfway between a commercial jet arrangement and a true delta wing

On the whole, I find T-tails or conventional tailplanes (tailwings and stabiliser) better. Also, for your main-wing, you could try a hi/lo biwing where the lo wing is anhedral and hi wing is dihedral, so that they taper towards each other at the tips. That should balance out any steering vaguery.

Edit: another thing to note is the position of your dry CoM compared to wet CoM, and also it's generally a good idea to disable engine vectoring if you're having control issues - rely on the control surfaces to keep you pointed right.

[Snark]

So I have it backwards, I currently have col above com. I am currently experiencing roll that is hard to control and keep level. This had them lead to over compensation and to much turn which generally ends in a flat spin.

Sorry, brain fart, I said it backwards in my last post (should have said "dorsal" and "above" instead of "ventral" and "below".) But the main point holds.

I've never had even the slightest problem with roll stability, so I'm really curious about your ship design. Could you post some pictures?

[ForScience6686]

I think I'm still a bit confused here on wings. Most training planes use a high wing setup and acrobatics will use a low mount. So with that said, wouldn't a higher col give more stable flight? As for my craft, flight up to space is fairly stable, it's the final approach that I seem to have the most instability, but if I get to the run way I have plenty of glide and control to land under 60m/s.

[Snark]

I think I'm still a bit confused here on wings. Most training planes use a high wing setup and acrobatics will use a low mount. So with that said, wouldn't a higher col give more stable flight? As for my craft, flight up to space is fairly stable, it's the final approach that I seem to have the most instability, but if I get to the run way I have plenty of glide and control to land under 60m/s.

Pictures = king hint, hint...

Just to be clear, by "final approach" you mean landing? And what sort of problems, exactly, are you having?

[The_Rocketeer]

Trainers typically have a hi-wing because it's safer - you get a better view of everything (except overhead), plus there's a bit of extra roll wiggle-room without getting a wingtip groundstrike. Acrobatic aircraft come in all shapes and sizes - the most manoeuvreable are still biplanes.

Ascent is high-speed high-power high-thrust, so the more subtle effects of your aerodynamics get streamlined into fairly constant linear forces that are easily dealt with by the SAS, and by the time you reach space they're not relevant anymore. Landing is fundamentally less stable - real landings are after all just controlled stalls.

Stability when flying slowly is all about having a good aerodynamic platform, whereas climbing is just about not having too much drag in crazy places. Also, when returning your plane is much lighter so it 'floats' as a body rather than 'hanging' on the wings - like a feather on the wind, rather than like parasail. You can see any issues in your lift platform will become much more apparent when landing - it's not that the plane is good during ascent, it's just that u don't notice it's flaws.

EDIT:

Don't underestimate the importance of the CoT to this either. In your plane I'm guessing the CoT is inline with the CoM and at the back, but in a trainer it would be in front. The relatively high drag of the mainwing would cause a natural pitch-up moment at high throttle settings, another safety advantage for stall recovery. The relative positions of CoT and CoL around the CoM create pitch moments - this is how commercial airliners are able to fly despite the engines being almost at the bottom of the aircraft, under the wings.

Edited October 18, 2015 by The_Rocketeer

[interface]

I like angled twin tail fins, yes they do create pitch or off axis yaw, if you actually use yaw for anything... I do not, I roll the aircraft and use pitch to turn SSTO as it is long and built to go straight, slightly nose up, very fast.

Going straight slightly nose up, very fast, twin tail fins are nice way to get CoL behind a CoM that is pretty far back on the aircraft, due to something like a bunch of nuke engines hanging off the back. Generally I use 4 'vertical' tail fins. 2 on top are large, angled less than 45 from vertical. 2 on bottom are smaller but longer, angled outward more horizontal than vertical, but still balancing the top vertical fins. Longer at the root but shorter root to tip, to allow aircraft to pitch up and take off from a runway without hitting them.

I find canards are very useful if you can actually get the CoL to accept them. Do not be afraid to have empty locked tanks at front and back that are aerodynamic structure not fuel. if your tanks are all near the center wet and dry CoM are much easier to manage.

[ForScience6686]

Here's the pics, a few things to consider before commenting. This plane is designed to fly out to duna and return my crew, it is launched partially filled and then refills in orbit before heading out and again before launching from Duna surface. The boosters are not staged together as they have separate purposes, top booster fired ~300m/s to break mach1 for useful thrust. Bottom booster fires ~1000-1100m/s to help with low twr of nukes, it must burn out before jet flame out. The ride up after that is a bit slow with the nukes. Wet and dry CoM are almost on top of each other. I am going to go try a re entry again and see how I fare, I am wondering now if it is just pilot error and too much stick, but let me know what you guys think.

- - - Updated - - -

Whoops:

http://steamcommunity.com/sharedfiles/filedetails/?id=537752768

http://steamcommunity.com/sharedfiles/filedetails/?id=537752768

[ForScience6686]

Ok, a couple test glides in, and I get this weird roll issue around mach1 on the descent, but once down to below 200m/s stability comes back.

[ForScience6686]

I am gonna go with pilot error mostly, too much input (joystick sensitivity now decreased), and bad high speed maneuverability due to the high surface area. I can bring it down just fine with SAS and hands off. I mostly run into problems when trying to make lateral glide path adjustments at high speed. I am gonna call it answered and leave it to pilot error.

[Alshain]

Well the wing angle could also be causing stability issues.

[ForScience6686]

After a bit of testing it can handle its basic purpose, a glide back home, I will just have to be sure I don't have to change heading much. I have plenty of glide on it to make up ground.

I figured the wing angle would help with re entry, though I guess it probably doesn't account for much. I am happy with it after the tests tonight. I wasn't planning on doing much atmosphere maneuvers, so I hope I won't need to.

[The_Rocketeer]

You might find the re-entry/landings a little easier if you shift the CoM a little further forward so the plane is slightly nose-heavy. Of course if you don't already have strong pitch authority you'll also need to beef that up.

Having the CoM and CoL too close together means you don't get a natural tendency to accelerate 'forward' while in freefall, which makes individual wings, and the plane as a whole, much more likely to stall. The reasons why are a bit technical (combination of aerodynamic flow and drag-/stall-induced rotation leverage), but basically having a CoM a) too rearward and too near the CoL are not particularly stable arrangements.

[ForScience6686]

I'll take that into account for the next ship. This one does like to pitch up so I can see where moving the com forward would help.

Since it is already at my station, I'm going to continue on with its mission. How well do you think this will work on duna? I have a total of about 5500 dv, so it should meet the requirements. I'm just not sure how well the nukes will perform on duna.

[The_Rocketeer]

Better than Kerbin. Duna's atmo pressure is similar to Kerbin's upper atmo, so you can expect higher thrust and ISP.

However, unless you're planning on chuting in you might find yourself needing to make a 'higher-than-normal'-speed landing due to lower drag and weaker lift. On the plus side, all aero forces are weaker on Duna so overall stability should be improved.

[ForScience6686]

No chutes on board, but I was hoping the air brakes would be sufficient. I can safely land on kerbin at about 50-60 m/s, so how much higher ate we talking about?

[The_Rocketeer]

It depends on how lift-y your plane is. Basically there's less air to fly on, so your stall speed will be higher - you need to go faster to generate the same amount of lift. Whatever your glide-speed on Kerbin, I'd reckon on it being up to twice as fast on Duna. Of course that doesn't mean you necessarily have to touch down at that speed, you can always come in low before making a steep pitch-up flare manouvre to kill your forward speed and flop into a touchdown - risky, but workable.

[ForScience6686]

That does sounds fun to try. My fleet in route to duna uses various methods for landing, this being the first plane I'll being sending out to another planet. My transfer to duna will be right on dv as I'm getting closed to the worst time to be heading out, but should arrive in time for an optimal window to return.

How's the terrain for landing? I've managed landings on kerbins hills but with lower gravity at duna I'm a bit concerned I won't be able to rely on wheel brakes as much.

[eddiew]

Well the wing angle could also be causing stability issues.

I thought that... when rolling with the right wing down, that's going to make the left tail more flat - which I believe will result in more lift, which would try to turn the ship back up again. Suspect it's too aggressive and inducing oscillations though.

[ForScience6686]

I have seen the oscillations, if to much input at high speed will cause control issues followed by the oscillation which I am guessing is due to too little yaw control.