[Test Results] How much wing does your SSTO spaceplane need? Answer: Maybe none at all.

[jiminator]

A_Name,

This is an example of what can be achieved with wings:

http://i52.photobucket.com/albums/g13/GoSlash27/CamachoII_zpsxstt1iyy.jpg

This is amazing. how do you get this plane into orbit? I mean exactly? I understand you build up as much horizontal speed as possible then angle up? With the remaining air to raise apoapse as high as possible, then close the ports? I don't seem to have much success. I can hit 80k (for contracts) but my rocket fuel is gone by then

[GoSlash27]

Slashy, I noticed you used tilted adapters and nose cones, is that to get it to align with the mk3 body? Aka, does that reduce the drag you were talking about?

sardia,

That's really just for convenience around the runway. Having the nose angled that way gives me a place to mount the nose gear and having the tail mounted that way gives me a convenient spot to mount the tailplane and avoids breaking stuff when I rotate for takeoff and landing.

It doesn't particularly help or hurt the aerodynamics.

Best,

-Slashy

[GoSlash27]

jiminator,

Apologies, wikisend is acting up. I'll send you a copy and walk you through it once I'm able.

Best,

-Slashy

[Captain H@dock]

Amazing design capability. I've been playing a bit with tilting the wings, but I wasn't able to clearly show the benefits you describe, though I didn't go for such an angle. i must try harder.

Also, from your screenshot, it looks like you are flying level when switching to close cycle. I guess such a low Thrust to Drag means it's the only solution, but I've always had a hard time figuring out what the air drag trade-off was regarding the velocity vector at switching point. I usually try to be in the 10-15 degrees range, which my design can achieve without getting bellow the 1450 m/s mark. Which leads to my question: What speed/Ap do you shoot for when accelerating on such a flat trajectory in close-cycle (assuming you keep flying level, and rely on exceeding orbital velocity to 'climb' your way out of the atmosphere). edit: looking at the album, you're pitching up after the switch-over?

Also, what's the behavior when re-entering and gliding back to base? Do you have a dedicated thread, I'd be keen on trying to fly it, I'm quite curious of the low speed maneuverability.

Also, I think my 6 rapiers equivalent craft doesn't have a higher part count (+5% i'd say?). But I would need to check the number tonight.

edit2: this is why we need a very simple black-box mod.... Being able to trace pitch, Aoa, Altitude and Speed during the ascent would make it so much easier to talk spaceplane....

Edited September 4, 2015 by Captain H@dock

[Val]

edit2: this is why we need a very simple black-box mod.... Being able to trace pitch, Aoa, Altitude and Speed during the ascent would make it so much easier to talk spaceplane....

MechJeb-dev has a Flight Recorder now, but it is scale-less.

[Stoney3K]

The trick is drag management. You see... it's not the thrust to weight that keeps you from going supersonic, it's the thrust to drag. So if you build it clean you can do more with less. And the wings are critical because they allow you to keep the fuselage aligned with the airflow without falling into denser air.

Wings create very little drag even when they're angled up at a very high incidence. Fuselage parts, OTOH, create a lot of drag even when they're not misaligned very much. So it's worth it from a drag standpoint to use wings to create your lift instead of the fuselage.

The point is that there is no way to find out numeric data on your drag numbers or even the center of drag in KSP. So when it comes to drag (as well as body lift), you just need to wing it and hope your numbers are in the ballpark. All you can do is guesstimate from the length of the arrows in the F12 overlay.

We could really use an improvement on KER to show the amount of drag along with the lift, weight and thrust numbers. It would mean we can tweak our SSTO's in a more structured manner.

Also, when it comes to MechJeb, having a feature that holds either a pre-set airspeed (and adjusting pitch/vertical speed to match) or a pre-set vertical speed (and adjusting pitch/airspeed to match) would be *really* helpful for SSTO pilots.

[GoSlash27]

Stoney3k,

This is true, but we *do* know a lot about the aerodynamic properties of the parts we use and how the aerodynamics engine behaves. We have the drag box figures for all the parts, aero overlay curves for the wings and lifting bodies, and we know about how occlusion does and doesn't work.

While this doesn't give us our exact drag figures at any given moment, it does allow us to design aerodynamically clean aircraft.

Capt. H@dock,

I usually don't need to use this much wing incidence. I wanted to use a low number of wing panels to keep them from being floppy and needing a lot of strutting. The idea is to incline them as much as necessary to keep the prograde vector forward throughout the flight so the fuselage is flying with zero incidence. Once you have that, you don't need any more incidence.

While the nose is level, it is actually still climbing. When I switch to closed cycle, I'm pitching up about 15° to bring up my prograde vector.

It is very slippery on reentry, even with the speedbrakes deployed. S-turning is mandatory to get it slowed down before overshooting KSC. As you can imagine, landing speed is very low. Roll response is weak on the tanker, so it's a good idea to have the lineup sorted out early.

I'll see if I can get you a copy of the craft file this evening. Wikisend isn't cooperating with me right now.

Best,

-Slashy

Edited September 4, 2015 by GoSlash27

[Captain H@dock]

MechJeb-dev has a Flight Recorder now, but it is scale-less.

http://i.imgur.com/tpZ5EDB.png

Yes, but it needs the whole mechJeb. Which i'm not really keen on getting. I keep it as un-modded as I can (KAC, AFBW), and mechJeb is pretty much the opposite of that.

Also, when it comes to MechJeb, having a feature that holds either a pre-set airspeed (and adjusting pitch/vertical speed to match)

IIRC Concorde autopilot held M2.02 this way.

@GoSlash27 Thanks. There's no rush, i'm not able to run KSP until tonight anyway. Still, since lift varies with speed, accelerating at zero fuselage incidence should lead to more and more lift, meaning you'd have to pitch up to keep fuselage zero incidence....

[GoSlash27]

Still, since lift varies with speed, accelerating at zero fuselage incidence should lead to more and more lift, meaning you'd have to pitch up to keep fuselage zero incidence....

Capt. H@dock,

You would think that's the case, but it doesn't quite work that way. The physics engine alters the lift and drag with velocity and incidence IAW spline curves in the physics.cfg and of course lift decreases with the lower air density.

Upshot of all of this is the lift stays pretty constant through the flight.

Best,

-Slashy

[Stoney3K]

Still, since lift varies with speed, accelerating at zero fuselage incidence should lead to more and more lift, meaning you'd have to pitch up to keep fuselage zero incidence....

Lift varies with speed and wing angle of attack. When your plane is properly balanced and a wing generates more lift, it will pitch up, leading to more drag and reducing speed. This eventually ends up in an equilibrium where thrust, drag, weight and lift are balanced.

If you back off the thrust, drag will win, slowing the plane down, thereby reducing lift and dropping the nose, which will *reduce* drag and estabilish a new airspeed. Conversely, with more thrust, your plane will speed up and raise the nose by itself to maintain that equilibrium.

A common saying in real aviation is, that pitch attitude determines your airspeed, and thrust determines your rate of climb or descent. If you trim to maintain a pre-set attitude, your airspeed will stay constant, and more or less thrust means the plane will climb or descend.

At least, that's how it works in the real world with proper aerodynamics. Your mileage may vary in KSP, but a high TWR only means a lightning fast climb rate. Low TWR means slow climb and shallow angle of attack, but since that means you spend longer in the higher atmosphere to accelerate to hypersonic speeds, it would have a positive effect for an SSTO.

[jiminator]

So I built something similar...

the thing pretty much flies itself, at 2x even. there appears to be some sort of balancing act once it hits mach 5. Air depletes and thrust reduces and start losing airspeed fast. Is that when the vents close causing switch? I had enough for orbit so thanks for that. :-) I probably should have used bigger wheels, the small ones were tricky to land

Edited September 4, 2015 by jiminator

[Just Jim]

wow..... I've had terrible luck making an SSTO... but after reading this, it occurs to me my biggest problem is I keep trying to build bigger when I really need to think smaller.

Thanks to all in this thread... there is a lot of useful info to consider.

[GoSlash27]

Jiminator,

Lookin' good

The air breathing engines decrease thrust rapidly with airspeed and altitude. That's why I like to hang out in the 17-22km range to get all the speed I can from the jets before I switch to rockets. Also, you should set it up for manual switching. It auto-toggles way too early and that costs a lot of rocket DV, which translates to mass and drag.

I don't bother closing the intakes in 1.04. I haven't seen a difference in the drag.

Best,

-Slashy

[Captain H@dock]

A common saying in real aviation is, that pitch attitude determines your airspeed, and thrust determines your rate of climb or descent. If you trim to maintain a pre-set attitude, your airspeed will stay constant, and more or less thrust means the plane will climb or descend.

Are you sure this isn't just for the backside of the power curve? And I've obviously never flown supersonic, but I would consider this applies even less to faster than sound regimes...

Therefore, i would buy GoSlash27 answer. Especially since he seems to get supersonic so low. What happens if you get to 15.000 m subsonically, then level off and accelerate to 1500 m/s? (obviously this will hurt your payload fraction, but i'm interested in what pitch/AoA you would need to hold across the speed range)

Also, I haven't found KSP to model very well the backside of the power curve, but that could just be because of the wesley slow throttle response.

wow..... I've had terrible luck making an SSTO... but after reading this, it occurs to me my biggest problem is I keep trying to build bigger when I really need to think smaller.

I agree. Anybody starting on SSTO spaceplane should start small. One rapier, one set of small wings, some mk2 fuselage, and play with ascent profiles.

[xtoro]

Isn't there a debug menu option to show aero data on parts? I've never clicked it but wouldn't that give useful data?

[Starhawk]

I don't bother closing the intakes in 1.04. I haven't seen a difference in the drag.

I did a balance beam test yesterday with one intake open and one closed. The open one definitely has more drag in flight.

Whether it's enough to even be noticeable on a normal craft is unknown.

For now, I'll continue to put the intake close action into my mode switch action group since it only takes a moment for each design.

Happy landings!

[Claw]

This is exactly the kind of stuff that's awesome about this community. Great job on discussing and exploring the game, finding what works within the game, and then sharing that with everyone.

I wish we had more of this.

Cheers,

-Claw

---

I did a balance beam test yesterday with one intake open and one closed. The open one definitely has more drag in flight.

I'd actually be interested in seeing this test. Based on my knowledge of the system, it shouldn't make any difference.

[jiminator]

Actually the "intake close" action also switches the rapiers into rocket mode, so you get a twofer on that....

[Starhawk]

I'd actually be interested in seeing this test. Based on my knowledge of the system, it shouldn't make any difference.

I've been thinking about this. I wonder, does intake air have any mass?

If so, that would affect the test I did.

I took a Flea, attached a couple of short I-beams radially, then attached shock cone intakes, one at a time, to the ends of the I-beams. I rotated them to face up and put it on the launch pad. Closed one intake and launched. The craft veered toward the open intake.

I reverted. Changed which one was open and which was closed and launched. The craft veered in the other direction.

I repeated the process of changing the open intake and launching one more time. The result was as expected if the open intake had more drag or more mass.

Happy landings!

[GoSlash27]

Starhawk,

I don't know, but I bet you could test to find out. Try launching the rig with both intakes open and close one of them after they are full. If it flies straight, then you know the imbalance is due to mass.

Best,

-Slashy

[Claw]

This might be partially my fault, but you might want to take the subject of intake testing into a new thread, to not completely hijack the thread about wings.

[Starhawk]

Starhawk,

I don't know, but I bet you could test to find out. Try launching the rig with both intakes open and close one of them after they are full. If it flies straight, then you know the imbalance is due to mass.

That should have been obvious to me. I wish I was as smart as you.

I did the test as you suggested, and it seems that it's the intake air that made the difference.

No veering if I let the intake fill before closing it.

Happy landings!

edit: Sorry about the hijack. I think I'm done now.

[GoSlash27]

That should have been obvious to me. I wish I was as smart as you.

I did the test as you suggested, and it seems that it's the intake air that made the difference.

No veering if I let the intake fill before closing it.

Happy landings!

edit: Sorry about the hijack. I think I'm done now.

haha you've clearly got me confused with someone else! Bonus: You've just confirmed that the intake volume mass still applies in 1.04.

So that settles it then? There's no demonstrable benefit to closing intakes in 1.04? I hadn't tested to see, I just never noticed an advantage one way or the other.

Best,

-Slashy

- - - Updated - - -

What happens if you get to 15.000 m subsonically, then level off and accelerate to 1500 m/s? (obviously this will hurt your payload fraction, but i'm interested in what pitch/AoA you would need to hold across the speed range)

Capt. H@dock,

That's an interesting question, and one I've never attempted to answer. Seems like it'd be worth an experiment (hint hint). Subsonic is pretty slow for that altitude, so not only would you be dealing with insufficient lift, but also insufficient airflow for the engines. At the other extreme (if you get that far) you've got a whole lot of velocity and heating.

You would pretty much have to see a swing in the AoA in that test *but* that doesn't represent an actual launch profile.

The way I do it is I run the whole profile and note the AoA throughout the whole thing. if I'm carrying AoA (especially at Mach 1.1 in level flight because that's the really important part) then I need either more wing or higher incidence. Opposite means I need less.

When it's dialed in the prograde and forward markers are aligned the whole time.

Best,

-Slashy

Edited September 4, 2015 by GoSlash27

[ajburges]

I agree. Anybody starting on SSTO spaceplane should start small. One rapier, one set of small wings, some mk2 fuselage, and play with ascent profiles.

I'm more a proponent of learning on turbojets and rockets first. While less efficient, the turbojet ascent profile has wider margins.

I just can't get past the poor aesthetics of wing incidence on Mk2 and Mk1 designs though. Makes it too easy to rationalize zero incidence designs. Maybe I'll toy with some non conventional wing designs.

I tried experimenting with wing incidence on a previously working Mk3 design and found it made the craft more squirrelly. Fighting the pitch up tendency (not terribly hard, just annoying) caused me to spend too long too low and miss orbit. The wing has two pairs of engine pods and offset RCS, so fine rotation wouldn't be very easy to tune. Maybe my issue was simply that adding incidence sifted my CoL past CoM.

Has anyone had problems with changing CoL with Mk2 designs? The difference in lift vector angles should make for potentially scary CoL shifts at the stall angle. Or do you just keep you designs stable enough to be stall resistant?

[GoSlash27]

I'm more a proponent of learning on turbojets and rockets first. While less efficient, the turbojet ascent profile has wider margins.

I just can't get past the poor aesthetics of wing incidence on Mk2 and Mk1 designs though. Makes it too easy to rationalize zero incidence designs. Maybe I'll toy with some non conventional wing designs.

I tried experimenting with wing incidence on a previously working Mk3 design and found it made the craft more squirrelly. Fighting the pitch up tendency (not terribly hard, just annoying) caused me to spend too long too low and miss orbit. The wing has two pairs of engine pods and offset RCS, so fine rotation wouldn't be very easy to tune. Maybe my issue was simply that adding incidence sifted my CoL past CoM.

Has anyone had problems with changing CoL with Mk2 designs? The difference in lift vector angles should make for potentially scary CoL shifts at the stall angle. Or do you just keep you designs stable enough to be stall resistant?

ajburges,

I'd really have to see some pics of what you were working with. If you have an adequate design without incidence then adding incidence doesn't benefit you much if any.

Adding incidence is really about 1) getting the nose aligned with airflow during the critical transsonic region and 2) getting lift out of wings that are really too small for the job. It's part of an entire integrated design philosophy, not something that you can just tack onto an existing design and expect it to work better.

Best,

-Slashy