Help with countering sideslip (FAR)

[TheXRuler]

So guys I need help. First off let me clarify I am not new to building spaceplanes. I have built several in my time these being my two most successful. I know I know imgur album but I just couldn't make it work properly T.T

I am currently designing this Laythe SSTO.

Javascript is disabled. View full album

Normally I would simply go for the dihedral wings that I used on the large SSTO. But I had this image in my head and wanted to build according to that, well it didn't quite work out but I still really love the design atm.

There is just one huge problem, if you have a look at the sideslip derivatives you'll see that this thing starts slipping sideways at ~7° when climbing. It flies like a dream, really (if anyone wants to test it I'll upload the file) and I would not like to change the thing to drastically, however if it can't be avoided I will of course do so.

My question to you guys, what am I doing wrong? I even tried sticking a massive (really massive, like twice the wing area) tailfin on it and that barely reduced the problem soo I'm all out of ideas.

Thanks in advance for everything

Edited October 27, 2015 by TheXRuler

[Van Disaster]

The problem is ( well a problem is, and I'm very tired here ) - is that if it starts to yaw, the AoA of the now-leading wing is quite negative ( especially with your not exactly vertical down-facing winglet things ), which means it's going to want to roll into the yaw - and so your sideslip gets worse, which puts more pressure downwards on the leading wing, etc. The obvious answer is "don't yaw" you can try brake rudders if you're really adverse to vertical fins, but your wing design is just inherently unstable so you're going to need to do extra to counter it. You will probably need pretty strong roll authority also so if it does start to roll into a sideslip you can counter that, so bigger ailerons too.

Theoretically yawing a swept wing would mean you get more lift on the forward wing - but in this case it seems the force from -ve AoA is considerably larger.

[TheXRuler]

Thanks I hadn't anticipated that the downswept winglets would be such a problem I had expected that they would act similar to a vertical stabilizer on a regular tail. Roll control and yaw don't seem to be such a problem in regular flight. It turns really well and as really good pitch authority. I can pull around 4g in a hard turn at ~220m/s. The problem only really shows up when I try climbing. At about 10° pitch it climbs reasonably well but while the plane itself flies in a straight line the velocity vector starts drifting off to the left... I guess the real problem is that I don't yet really know a lot about wing design :/

I'll fly some more tests and update the thread if anything interesting happens

[Bomaye]

Thanks I hadn't anticipated that the downswept winglets would be such a problem I had expected that they would act similar to a vertical stabilizer on a regular tail.

They will they are just too small, look on the F-15 and you see the size or the SR-71 where they are all moving.

[Keldor]

One thing you could try is tilting the back of the winglets inward. Then when you sideslip, it should tend to pull you back forward.

[Van Disaster]

Thanks I hadn't anticipated that the downswept winglets would be such a problem I had expected that they would act similar to a vertical stabilizer on a regular tail. Roll control and yaw don't seem to be such a problem in regular flight. It turns really well and as really good pitch authority. I can pull around 4g in a hard turn at ~220m/s. The problem only really shows up when I try climbing. At about 10° pitch it climbs reasonably well but while the plane itself flies in a straight line the velocity vector starts drifting off to the left... I guess the real problem is that I don't yet really know a lot about wing design :/

I'll fly some more tests and update the thread if anything interesting happens

It's hard to tell but the vertical surfaces ( and I presume rudders ) didn't look completely vertical. Drifting off as you pitch up is usually a sign there's not enough authority though, which means "make them bigger"; if you've built something that naturally unstable you need a ton of control authority, and you don't get much to work with at high altitudes. I would also make sure they are actually properly vertical too. Usually you can fix most stability problems by giving the plane bigger levers

But yeah a lot of stability issues come from the fact that craft don't actually fly perfectly forwards even when they're nominally going straight; this is before you consider what happens when they're banking in a turn. When you're building something look at it from the front at an angle & think "does that look stable if it's moving towards me?".

[TheXRuler]

Thanks everyone, that's a lot to consider, I'll try some redesigns and different flight profiles and let you know how things turned out.

[Wanderfound]

1) You have no tailfin.

A bit of anhedral on the wingtips is no substitute; barring a few high-tech differential thrust exceptions, planes need fins.

2) It's a delta design with a nuke. This means a heavily rearwards weight bias.

The power of your vertical stabiliser is in part a function of its distance from the CoM. Ergo, a rearwards weight bias requires a larger tailfin. Place it as far back as you can, and make it beefy. Also try to pull the CoM forwards as much as you can.

[TheXRuler]

Uhhh the high lord of SSTO's *.* all bow to Wanderfound.

I was wondering if that might not have a rather large impact, as it is now my fins are VERY close to the CoM. Incidentally, I had already tried vertical stabilizers on the nose, albeit very small ones. I feel confident I'll manage to get this thing working with all the input, thanks everyone, you guys are amazing I'll be sure to post the final design.

[Van Disaster]

Control authority ( for a single control ) is in essence "surface area of control * distance from CoM * maximum deflection". There's a few other considerations given the resulting force might not have a linear relation to the deflection, but in broad terms that's fine. If you're thinking of fixed surfaces you can just assume it's deflection is 1 and forget about it. Anyway the point is, if it's struggling for control in any particular direction then you don't have enough of that whole equation, so pick the least costly one to raise. Generally you don't want to raise maximum deflection if you can help it because that will cause more drag, but then so will increasing the area - it just may not increase it as much, but it will increase mass. The best one is to move away from CoM ( if you remember the basic physics of moments it's obvious why... ).

Don't put fixed vertical stabilizers in front of CoM - any yaw is going to make the nose fall off the heading, not straighten back up again. Moving ones should work but they're going to be pretty sensitive... there's no self correction due to gravity like there is in pitch. It's not the downswept fins that are making it unstable ( unless as I thought they're not vertical ), it's the anhedral wings. The fins just aren't big enough to work.

Brake rudders are a new FAR thing which will let you do completely tailless designs - but at the cost of drag obviously, as they're actual airbrakes.

Edited October 26, 2015 by Van Disaster

[TheXRuler]

Oh well there seem to be ever more problems xD the fins are indeed tilted outward by about 10° also, the I guess I should've clrified that I was sticking all moving surfaces on the nose, although I never tested it in flight. I just ran the FAR simulation and the numbers hadn't changed much so I trashed that concept, maybe I will go back to it though, so many things to try out. It also seems I had slightly misunderstood anhedral/dihedral effects, I was only considering the CoL shift, not the effect that yaw would have on a deflected wing. I guess theres quite a few things which I need to read up on >.< sadly aircraft aircraft design is a master module where i'm studying so it's gonna be a while until I really know what I'm doing

[Van Disaster]

Give the main wings dihedral instead, and make the tips twice as deep - it'll still look dramatic & be a lot more stable

Knocked something up in 0.90 because I don't have a full 1.0 install atm, principle should translate fine.

Edited October 26, 2015 by Van Disaster

[TheXRuler]

Oh .90 is fine, I am still not using 1.0.x either xD

Thanks you guys, it has been done, yesterday night the as of yet unnamed

Lathye SSTO-Spaceplane had it's maiden flight, it turns out it is even easily capable

of getting into orbit around Kerbin. Many many thanks to all of you for contributing

and being so patient with me.

This is my final version, any thoughts on a name?

[Hodo]

You can build a tailless design but it is a challenge. You will want to use the airbrakes as your yaw controls for that craft otherwise you will end up in a flat spin situation. I would place the airbrakes as close to inline with the CoM as possible, but as far out on the wings as possible to give you the most control with the least effort.

I would also reduce the sweep of the wings to give you a bit more low speed stability.

[Van Disaster]

The old FAR version he's using ( and the one from my build ) doesn't have yaw brakes. Sweep is effective dihedral though - start yawing & the outer wing unsweeps itself & presents a longer face to the airflow.