[1.3.1] Ferram Aerospace Research: v0.15.9.1 "Liepmann" 4/2/18

[AdmiralTigerclaw]

Awesome.

Yeah I get you on the aircraft design. Engineering learns it one way, pilots learn another. But this will help a lot for those of us who are pilot/sim-pilot minded. I'm sure we'll all be

with the results.

[localSol]

Hooray and congrats to ferram and a.g. on another great release!

Thanks to this discussion thread, I used flaperons for the first time last week. And it looks like I also used elevons without knowing it.

[Surefoot]

Reporting for the new 0.10 version - everything looks good on my side, my Laythe jet flies normally again (i updated KJR at the same time, maybe it's that too), and hypersonic flight is now a bit different i notice less pitch down, more resistance to inputs (my design is stable though). Pretty happy with it, didnt notice anything unusual with my launchers and spaceplanes.

This flies wonderfully thanks to Ferram:

Happy B.K. at Laythe north pole

[Camacha]

Is there a special manual around on how to effectively use the simulation functionality of FAR in the VAB/SPH?

Readme.txt only contains general hints and I guess it might be helpful to have an explanation of what you can actually achieve with the simulation tool.

Yes, I would also like this very much. I am getting to a point where I use more and more proper science to work out missions and parameters and I feel that I am guestimating still a bit too much when it comes to FAR.

[TeeGee]

Yeah well I always install then uninstall this mod because it turns my SSTO's into death traps (unless you install B9 which I refuse to do). The planes I built that got into space wouldn't take off the runway early, regardless of airspeed, and when they flameout, become absolute death traps because they are dual engine planes. So I reach 30 km then I get an asymmetric flameout and the plane wigs out and I can't recover her until I hit thicker air. It also takes longer for me to reach escape velocity in comparison to stock kerbal regardless if I use nosecones or not.

It's just infuriating that I WANT to use this mod, it's great for rockets, but makes plane building and design stock craft ridiculously limited. I dunno if it's the fact that I part clip my wings or if I just don't know how to use the mod properly, but whenever I install it, I immediately uninstall it because it makes the game limit your design options. And forget the fact that suface ASS with Mechjeb will do anything with FAR, the plane doesn't know how to respond when I activate mechjeb in atmo.

[asmi]

Hey ferram4!

First of all I'd like to apologize you for the mod screw-up that happened a bit earlier. It's taken care of now.

But as you've said, very few people are posting here because they don't have a problem, and unfortunately this time I won't be an exception - I still have BobCat's Soyuz TMA descend module capsule wanting to turn upside down during descent - even when under a fully deployed chute! This just lead me to land on the hatch with soft-landing engines pushing me towards the ground Can you please take a look?

[ferram4]

@Surefoot; That's because in the previous version bodies made more lift than they should have at hypersonic speeds. That's been fixed now.

@TeeGee: Well, you're installing something that changes the aerodynamics a lot, so that's to be expected. But here are some possible solutions:

For taking off:

If it simply won't pitch up, that means you need to move the landing gear forward or you need to increase your pitch authority--that can be accomplished with more control surfaces or moving the CoL closer to the CoM.

If it does pitch up, but doesn't take off, then you need more lifting surfaces or you've got some type of interaction between the surfaces that is robbing them of lift--biplanes don't make twice the lift of monoplanes, they only make 1.1 - 1.2 times the lift, but twice the drag.

For flying:

Stop climbing to the highest altitude you can get away with immediately. The atmosphere isn't made of jello anymore, you don't need to be afraid of flying low with your intakes in no danger of flaming out.

The Flight Data GUI has a nice little indicator showing how close you are to a flame out; perhaps that would be useful.

Jet engines won't bring you to orbital velocity anymore; you need to light off rockets and hop out of the atmosphere instead of intake-spamming around the planet until you can spend 10 m/s to circularize.

Yes, the design choices are somewhat limited with this mod, but complaining about that is like complaining that KSP limits you to having to attach engines at the bottom of your rocket; you can attach the engines anywhere else, but it will have very nasty consequences of doing so, either in wobbliness, control or Isp losses. Similarly, in FAR, you can build planes that look nothing like real planes that still fly, but there will be many adverse consequences to those design choices. You're complaining that your design actually influences your aerodynamics, rather than having almost no effect as in stock KSP.

@asmi: If I recall correctly, the Soyuz decent module has its CoM in a very strange place; at least when I've used it I've always seen the camera act like the CoM is beneath the module itself.

If that's not the cause, make sure that it's properly oriented when coming in, with the heat shield at a proper weight. If those aren't set properly, I can guarantee it won't behave like the real thing.

Edited November 11, 2013 by ferram4

[asmi]

@asmi: If I recall correctly, the Soyuz decent module has its CoM in a very strange place; at least when I've used it I've always seen the camera act like the CoM is beneath the module itself.

If that's not the cause, make sure that it's properly oriented when coming in, with the heat shield at a proper weight. If those aren't set properly, I can guarantee it won't behave like the real thing.

it's CoM is offset, but only by 0.4 meters - like it is in real life, that allows for controlled descent, but it's still is inside the module like, well, it should. Even when it's going like 10 m/s under fully deployed chute, it still turns upside down unless locked in place by SAS & constant manual steering input. I just don't understand what kind of forces could possibly force the capsule to rotate in such a way - certainly at 10 m/s aerodynamic forces are simply can't be THAT high to turn ~2 t module.

[asmi]

And I've noticed that CoL in SPH is going crazy:

[ferram4]

That's strange. The last time I've tested it it came down quite nicely. I can look into it more though.

The CoL is just showing where the drag is being applied at that moment; angle it the other way and it will appear on the other side of the vessel, and at no angle of attack it should be directly below the capsule. What you're seeing is intended, but I think the way that the CoL is representing it is wrong (or at least unclear).

Edited November 11, 2013 by ferram4

[asmi]

That's strange. The last time I've tested it it came down quite nicely. I can look into it more though.

Thank you - I appreciate it!

The CoL is just showing where the drag is being applied at that moment; angle it the other way and it will appear on the other side of the vessel, and at no angle of attack it should be directly below the capsule. What you're seeing is intended, but I think the way that the CoL is representing it is wrong (or at least unclear).

I may sound a bit dense, but drag is being applied to... the empty air? If my understanding of the laws of physics is correct, CoL couldn't possibly be outside the vessel since you can't apply a force to the object at a point that is outside of that very object...

[ferram4]

Yeah... that sounds wrong. The center of pressure can end up outside the vessel, but I've never heard of the center of lift being outside it. I'll look into it.

[15nelsoc]

Code refactoring by sarbian for future interfacing with MechJeb

If you actually manage to get this to work properly with MechJeb some time in the future, that would be absolutely amazing.

[Camacha]

If you actually manage to get this to work properly with MechJeb some time in the future, that would be absolutely amazing.

Well, I do not really care for MechJeb, but if that means easier interoperation of FAR with things like Kerbal Engineer and the Realism mods I am all for it.

[taniwha]

According to sarbian, the changes involve interfaces to FAR's code that permit MJ (and presumably other mods) to query FAR about the vessel without actually affecting anything.

[TeeGee]

@TeeGee: Well, you're installing something that changes the aerodynamics a lot, so that's to be expected. But here are some possible solutions:

For taking off:

If it simply won't pitch up, that means you need to move the landing gear forward or you need to increase your pitch authority--that can be accomplished with more control surfaces or moving the CoL closer to the CoM.

If it does pitch up, but doesn't take off, then you need more lifting surfaces or you've got some type of interaction between the surfaces that is robbing them of lift--biplanes don't make twice the lift of monoplanes, they only make 1.1 - 1.2 times the lift, but twice the drag.

For flying:

Stop climbing to the highest altitude you can get away with immediately. The atmosphere isn't made of jello anymore, you don't need to be afraid of flying low with your intakes in no danger of flaming out.

The Flight Data GUI has a nice little indicator showing how close you are to a flame out; perhaps that would be useful.

Jet engines won't bring you to orbital velocity anymore; you need to light off rockets and hop out of the atmosphere instead of intake-spamming around the planet until you can spend 10 m/s to circularize.

Yes, the design choices are somewhat limited with this mod, but complaining about that is like complaining that KSP limits you to having to attach engines at the bottom of your rocket; you can attach the engines anywhere else, but it will have very nasty consequences of doing so, either in wobbliness, control or Isp losses. Similarly, in FAR, you can build planes that look nothing like real planes that still fly, but there will be many adverse consequences to those design choices. You're complaining that your design actually influences your aerodynamics, rather than having almost no effect as in stock KSP.

My planes fly. Don't kid yourself; I understand how to pitch up an aircraft and have them fly in your program.

I've built dozens of SSTO's that even have 7,000 m/s delta v that make orbit WITHOUT intake spamming and can make an apoapsis of 80 km, so I think I am pretty experienced in 'spaceplane building 101'. The problem I was having was pitching UP but not taking OFF even with high airspeed. When I left the end of the runway I was airborne and climbed normally until I hit thinner air at 30km then had a flameout. As soon as I veered from my prograde vector a few dozen degrees off, the plane went into seizure. All I needed to do was throttle down, the problem is that Kerbal doesn't warn me when I am close to engine flameout until it flames out! I normally throttle down at that height, but I use the flame out as the *time to throttle down* cue.

#1. Asymmetric flameouts in REALITY DONT DO THAT. So don't blame my piloting or design skills. Vertical stabilizers at those high speeds should keep my plane from flat spinning during asymmetric flameout. This thing literally frisbeed.

#2. Air intake at high altitudes are NOT an issue in real life. Jets can function fine at super high altitudes, its just that there is a problem with high TEMP (which saber engines have resolved).

#3. Air intake DRAG does not balance with air intake AMOUNT. So your system punishes use of a lot of air intakes with increased drag but still keeps the stock air intakes unit scoop volume? How is that fair? This is why I complain. Unless you are using saber engines with B9 pack, you wont get into orbit with a stock SSTO very easily, especially SMALL sstos (low mass). 'Cause whenever a low mass ssto reaches thin air and flames out, that plane is basically a coffin.

[a.g.]

I may sound a bit dense, but drag is being applied to... the empty air? If my understanding of the laws of physics is correct, CoL couldn't possibly be outside the vessel since you can't apply a force to the object at a point that is outside of that very object...

Some points:

First, according to the laws of physics, the point of application of a force can be moved to anywhere along the line of the force without any change in the effect. The indicator chooses the point that is closest to the center of the vessel.

Second, the indicator is actually representing the derivative of the force and torque, not the plain value. Specifically, the arrow represents the derivative of the force, and the position is chosen so that this change in the force, if applied there, would cause the right change in torque. When the linear force change becomes very low, the lever arm to produce the necessary torque grows a lot.

[ferram4]

@TeeGee: Alright, so what does this SSTO look like? And what counts as "high airspeed" during takeoff? 80 m/s? 120 m/s? 150? Have you tried adding flaps to your SSTO to give it a little extra lift?

You're saying that you know how to fly planes with FAR, but you insist on putting your vehicle into a position where it has no room for error; if pitching it up a few degrees is enough to cause a flameout, then you're way to close to the flameout threshold. And it's true, KSP doesn't warn you about a flameout, which is why my plugin has a reading that you can use in one of the GUIs; I've been annoyed by this and thought of it already.

1. Yeah, I know; I'm looking at ways to make the jets flameout more gracefully and evenly. There's no reason that the air from one side of the plane should be feeding the engine on the other side. That said, vertical stabilizers aren't as effective at high supersonic speeds; perhaps you need larger ones.
   
2. Well, it's implemented is actually correct. The air intake and fuel consumption values point to it being the ratio for complete burning of Jet-A. When you're trying to run at full throttle with not enough air you're pouring too rich a mixture into the engine, interrupting combustion. Realistically, the engine control system should handle that for you, but this is KSP, where everything is done manually.
   
3. The stock engines are tweaked to make less thrust with FAR; the air intakes provide the same amount of air and the same amount of drag. I don't see what you're complaining about, since you can fly much higher with FAR than in the stock game. I don't know about not getting to orbit easily; the stock Aeris 4A is easily capable of it with FAR.

Why are you climbing until you reach 30km? Why not try to gain speed at 20km - 25km and then when the engines start to run out of breath pull up and not have to worry about switching over to rockets until 30km?

[Hodo]

*cut out the self entitled BS*

#1. Asymmetric flameouts in REALITY DONT DO THAT. So don't blame my piloting or design skills. Vertical stabilizers at those high speeds should keep my plane from flat spinning during asymmetric flameout. This thing literally frisbeed.

#2. Air intake at high altitudes are NOT an issue in real life. Jets can function fine at super high altitudes, its just that there is a problem with high TEMP (which saber engines have resolved).

#3. Air intake DRAG does not balance with air intake AMOUNT. So your system punishes use of a lot of air intakes with increased drag but still keeps the stock air intakes unit scoop volume? How is that fair? This is why I complain. Unless you are using saber engines with B9 pack, you wont get into orbit with a stock SSTO very easily, especially SMALL sstos (low mass). 'Cause whenever a low mass ssto reaches thin air and flames out, that plane is basically a coffin.

1- actually they do. Having flown a few sims and had some real experience in an twin engine aircraft, nothing sucks more then one engine suddenly loses power, and I am not talking 1 or 2 % but like 50+% of its power at once. You can recover but it takes a great deal of effort.

2- ORLY? Then why can't the F-15C Eagle fly higher than 65kft? *rhetorical question* Because the engines won't breath that high.

3- Actually that is true in real life to. Why don't you see aircraft with a MASSIVE intake in the front anymore or why we quit making planes like the F-84 and F-86. Because that massive nose intake was causing more drag than it was worth. If you notice the F-22 and the F-35 have much smaller intakes than even the F-15C and F-16 fighters.

I have built the fastest, and the challenge shows it, FAR mod Stock aircraft here, and it isn't an intake spam yet reaches Mach-6. Don't believe me look at this thread. http://forum.kerbalspaceprogram.com/threads/27296-The-MachingBird-Challenge! Who is on top of the Stock+FAR mod?

I have also built over two dozen SSTO space planes, all using FAR, some using Arcturas thrust corrector, and they all work, even the few that are using stock parts. So I know it is completely possible. I think your problem is, really two fold, but I will deal with the problem in KSP+FAR as the other problem I can't fix. You are taking a two aggressive ascent profile for that craft. Each one of my SSTOs work differently, some I can shoot right off the end of the runway and go straight into a 60deg climb till 20km, others I have to ease up to 15km then build up speed at a 5-10 deg climb rate till 25-30km.

But to end this little debate, it is all speculation without actually seeing your craft.

[palker]

If you open the flight data UI part you will never flame out again because there is an indicator of how much excess air do you have.

#1. Asymmetric flameouts in REALITY DONT DO THAT. So don't blame my piloting or design skills. Vertical stabilizers at those high speeds should keep my plane from flat spinning during asymmetric flameout. This thing literally frisbeed.

The aircraft on the picture would frisbee as well and the vertical stabilizers would not help cause their stabilizing effect is dependent on Q not on speed only Q=1/2*density*speed^2 . And do not tell me that stock aircraft are immune to this i had my share of asymmetric flameouts in stock that caused me to spin out violently.

#2. Air intake at high altitudes are NOT an issue in real life. Jets can function fine at super high altitudes, its just that there is a problem with high TEMP (which saber engines have resolved).

Intake air is not an issue in real life cause real life jet engines work differently from the KSP. Also they have automatic control that reduces fuel flow as air becomes thinner while in KSP you have to throttle down to stop engines from flaming out. I do not know how FAR and Taverio Pizza and aerospace change the jet engine characteristics but i guess they are far closer to reality than stock which is not that close at all.

[White Owl]

I love how so much of this enormous thread is taken up by people telling ferram4 how to make the mod work differently, and yet to my knowledge not one single competitor mod for realistic aerodynamics has ever appeared.

Ferram4, I think you're doing something right. This labor of love you've put together and continue to support is indispensable to flying in KSP. Thanks for that.

[ferram4]

Well, there was the old Raycast Drag Experiment that I built for 0.19; I'll probably look into getting that working using the KSPAddon entry point instead of the ad hoc one I used before. That way there will be an answer for the aerodynamic equivalent of the, "we could model Lagrange points with a small SOI at each point," suggestion.

But that doesn't count because it's mine.

I don't begrudge people who come in here and yell at me about how things should or shouldn't work; there's always the possibility that they're right and I'm wrong. If they're right, I get to fix things and the mod gets better; if I'm right, they learn something. Either way, someone directly involved in the exchange benefits, while anyone looking in from the outside can also learn. And if none of that happens, it probably ends up being funny at least.

[localSol]

You know it's interesting, last week I remember some good conversation about re-entry at high angle of attack, lifting bodies, SSTOs and that sort of thing.. So today I decided to try an SSTO with minimal wing, it went really well! Static analysis for mach 1 to 2 made me think an angle of attack of 23 degrees would be best..haha! So it was fun, so I made a video of it. Things get much easier after I get over mach 2 or so. Also minimal plane otherwise. :

(Video sped up to 4x, long parts of flight at another 2x in KSP) (720p)

http://www.youtube.com/watch?v=2oQ4bj-T1WY

craft (needs TV, B9 and FAR..TV has comparable landing gear also) : https://dl.dropboxusercontent.com/u/6231140/KSPstuff/minimalairfoils1.craft

I don't know aerodynamics really, but here : I have been using dihedral wings, they're supposed to give a little more stability and lose a little maneuverability?, seems stable to me..I guess these flat-ish mk2 body sections are making more lift than I thought they would. And that's good. Though still the engines do most of the work, L/D is very low compared to my other "planes". Keeping an eye on my air requirement met % and angle of the intakes relative to airflow makes it easy. VOID also helps you track your vertical climb/drop in m/s numerically to help you keep stable especially during transonic flight. Jeb could be eating spiced soup while doing this.. While absent mindedly piloting at mach 3 in there. I want some soup, Jeb. I like soup.

Don't forget to wear your pilot's scarf and aviator goggles!

Edited November 11, 2013 by localSol
VOID mention

[Snjo]

Interesting fact about flat spins in KSP (and maybe unity in general).

InfiniteDice did some test to try to reduce the tendency to spin out when losing one of two engines. He did this by moving the thrustTransform position away from the engine model towards the CoM.

Imagine the span between the CoM and the engine as a range of 0-10000, where 0 is CoM and the engine is at 10000. Even when the thrust happened at 1, i.e. some millimeters away from the CoM, you would still flat spin if you lost an engine.

Now, I didn't write the code, so I can't be sure how the positioning worked out, but if it's accurate, it means there is a bug in how force applied asymmetrically affects angular velocity on a rigidbody (or set of rigidbodies as in KSP).

[Surefoot]

About twin engine flameouts, my current optimal design for Kerbin -> Laythe stays stable whenever an engine goes down at mach 5. It is also quite difficult to throw it into a spin. Downside is landing requires flat terrain and low AoA, but otherwise it flies wonderfully.

I also made a pure atmo design, that looks and handle like a real jet fighter. Again, twin engine.

Upon flameout they will veer to one side but not go into flat spin.. Of course they lose speed quickly and i have to either switch the engines to rocket mode (for the SABRE jet) or throttle down.

So, it's definitely possible. Just read a bit more about aerodynamics, wing configurations, and real life examples (like SR-71, Mig 29, F-16..). For example handling the SR-71 without SAS required a lot of piloting talent as it was quite unstable, and flameout had quite a quite brutal result..

Edited November 11, 2013 by Surefoot