Van Disaster

At the risk of repeating many comments, it's already been done & available for you to try. Link to Principia is in the second post in the thread, copy your KSP install & drop it in to test it.

I think this is the only thing I have from ... probably 0.22? - when did those wheels appear? - which still works. Things from older versions definitely don't. There's an all-stock one in the rover thread somewhere. I've been building B9 S2 forward-swept spaceplanes for years, but not reusing the same design - I guess that is sort-of a signature. also variants of this thing although I've never got it to look as good as the first one which I don't think I have anymore.

Yeah, there may be an advanced prop engine mod I'm not aware of with torque, but traditional prop engine mods are just visual effects on top of normal engines - by which I mean fuel in, thrust out, end of matter. FS props were certainly like that, I've used KAX in the past but can't remember - the SXT props in my screenshots are certainly that way too. Anyway you can check very simply by just replacing the props with small jets & checking if it still has aero issues. Or flipping the props to pushers, I suppose. However - you didn't accidentally thrust-reduce one engine at some point? I usually have mine slightly assymetrical so they sound like multiple engines if you start them seperately, but never more than a couple of % which doesn't really affect flight much. Edit: and either I forgot how to delete posts with this board software or you can't, and I also forgot to quote again & you can't insert quotes to trigger notification later

Sweep does not have to look like an EE Lightning I'm going to say you will probably never tune all roll/spiralling out of your aircraft - this is not something we've ever tuned out of real aircraft completely, which is why we invented autopilots instead of just using trim controls. That it's this bad *is* a build problem.

Unless something has changed recently, that is not true for FAR - the aerofoil section is a mashup of several NACA profiles suitable for supersonic flight ( which is why they're not fantastic for flying very slowly - we had a bit of that last page ). Ferram mentioned which ones they are somewhere in the main thread, I'm not about to dig through that behemoth again though. What dihedral actually does is worth reading up on (spiral mode & divergence in general too ) - it's far more complicated than the wing having a vertical component, and anhedral will tend to promote rolling into sideslip unless there's a compensating factor like the wings being shoulder-mounted & the fuselage quite heavy, or a lot of sweep, etc.

I don't ever get notified anyway, so I don't think of that... Before you start fiddling with the sim I'd try some simple tests with wing shape first. Dihedral would be a start - actually perhaps pulling the wings off & reattaching might be the place to start just in case it's a symmetry error. Back sweep is self-stabilising in yaw because the inner wing is temporarily less swept than the outer one, so it has a little more lift & rolls the plane out of the sideslip ( the fun corollary is that forward swept wings roll *into* the sideslip ) which is why I suggested a tiny bit of sweep, but try dihedral first. Most of my planes seem to ( slowly ) fall off their heading if I don't trim them or use a course-hold tool like pilot assistant, I suspect we're never going to get it absolutely perfect.

I didn't quote you because I was the next post anyway, the aoa/mach graph is all about pitch stability & the static panel assumes the craft is in the position it's in in the hangar as far as I remember, which is generally perfectly level unless you've tilted the entire craft yourself. I keep forgetting about the dropdown list & the transonic panel - previously the simulation tab was the last one from left to right & I always remember it there... You have to preload the sim by hitting the calc derivatives button in the static panel ( I suspect a lot of people don't do this ), then you can set your starting state - the coefficients are the same ones as in the static derivatives, so the explanation of them is the same - & for a stable plane hope the graph trends towards zero rather than away. I generally don't touch the panel & try and eyeball problems, but if you're attempting to get the last bits of performance out of a craft, say for a contest, it's sometimes something to spend time in.

RE: planets on rails, and given the existence of an n-body mod already ( if you're worried about performance, go and test performance already ) this is a relevant question: can you save systems from Universe Sandbox in a convertable format? you can pre-test your system for stability then.

Well, there's ways around the fact that craft are a bunch of parts flying around together - I'm not sure how often you're going to have more than one body exerting enough force on a craft to think about *pulling it apart* ( as opposed to just one body, which is already covered and tends to only pull craft apart when they're in contact with it ) so most of the time you could treat the whole craft as one item; if there's something that delicate on it then you can precalc forces for the whole ship, and if there's something which looks like a stressful conjunction of force then only apply it to parts below a threshold, and whatever they're attached to. Of course if someone's built a completely mad system then the threshold might cover the entire ship & things get expensive, but that does seem something of an edge case.

It's still possible the issue is *starting* with sideslip and you just have enough vertical stab to point the nose into it fairly fast, but that means you're turning - try a little dihedral, and perhaps just a couple of degrees of sweepback, to start with. You might end up with dutch roll instead, but if you do at least that will keep you pointing roughly in the right direction. Also note symmetry isn't completely precise, and you do get some issues from that ( I suspect that is actually the problem here ). Analysing this properly is what the final FAR panel which hardly anyone uses is for. I have used it in the past to try & find oscillations, but I can't really remember what I'm doing with it at the moment. One last possibility is you added the wings in rotational symmetry mode by accident, added a degree of incidence on one wing which added a degree of opposite incidence to the other... Going back to post/during stall controllability, I found this in some screenshots: Not sure if there's some washout on the tip panels ( I suspect so ) or if it's a result of the trailing edge controls on the centre panel stalling the whole panel, but 45 deg AoA, mostly stalled, and still mostly controllable is not terrible you can see the drooping leading edges aren't stalled, nor is the tail which I'd hope I'd left a little roll control in. That engine is the old Wheesley, which had no gimbal whatsoever.

As always, there is a mod for it- Principia - albeit n-body simulation of everything rather than the limited one you propose.

In as far as stalling from the correct direction for sweep, yes it does. You'll also find the right sort of instability issues due to sideslip.

FAR does handle body lift - but I don't think you can make an aerofoil shape out of body parts ( ask in the main FAR thread how sophisticated body lift is, I don't know the details ). Easiest thing to do is just have longer proc wings & clip the roots into the fuselage. If you want to build a full replica of something at a structural level try X-plane. Supermaneuverability is having full control of the aircraft post-stall. I don't know the details of how they do other than thrust vectoring, but presumably there's still use of control surfaces for differential drag. That we can't really do unless you differentially unstall a surface, I think. KSP vectoring engines are strong enough to build a working aircraft. Sticking AoA into leading edges works to delay stall of the wing surface. Other things to do are mixing AoA into control surfaces so they stay unstalled.

I can't look at it in game right now, but it might simply need lower wing loading - the advantage real aircraft have is they can shape fuselage areas to properly act as lifting bodies, and we can't really even if FAR could pay attention properly ( fuselage stuff is part of voxel-FAR so presumably it could in theory - that doesn't change that we can't sculpt our fuselages ). If you make it light enough you can pull extreme G even with low structural strength, ( a rather old example ) but I doubt you're going to manage that with two heavy engines

"Cockpits are heavy" - because the mass distribution of kerbal parts is not like real ones. Drag-wise, don't forget Kerbin's atmosphere stops at 70km so it is all compressed vertically. Tweakscale at times has caused chaos, but generally it's been fixed eventually - you'll have to go check the mod threads. As for control authority, defining it only by the part of the envelope you find yourself in a normal landing *at sea level* is pretty restrictive - supersonic flight might need more authority than that, given the CL shift rearwards. CL shifts fore-aft in subsonic flight, too.

( Response to Xavven's post): I feel you're taking the stock elitist ego-stroking approach that mods must by nature make the game easier, and that anyone using mods is an unskilled scrub. That's an awfully long way from true ( quite a few make it considerably harder ) & you're probably cheating yourself out of experiences nicely put together mods would provide you; for instance, I've had a whole bunch of things relating to life-support mods over the years I'd never have got with plain stock. You're also implying that everything in stock is carefully balanced, and while it's better now than a couple of years ago I'm not sure I'd agree. I don't actually find this a particularily good game without mods.

Yes. It's more than that though, there's much less energy bleedoff when you're maneuvering too, a good sense of how much energy you're really carrying in the craft, and the overall shape really does matter. Current stock is at least lightyears away from the insanity of pre 1.0 stock aero where either drag or lift depended on mass, if I remember properly... this was the sea level max speed, and yes that was a viable "aircraft"

Yes, it's too soupy ( it was too soupy in the initial revised aero too ) - mostly you can see the problems when you maneuver aircraft. I've used FAR since a month after I started playing & won't leave it, and most of the time I just throw aircraft or spacecraft together & they work, it's not any more complicated than building something stock. Only time I use all the scary number tools is when something is wrong and I can't work out what by looking, if I was running stock I'd be really wishing I had the same tools...

As the others said, there is no such thing. However I like any parts & behaviours to seem coherent & pay some service to the idea of real physics ( hard for a theoretical warp drive, sure ) or I feel I'm cheating myself out of some experience. Cheating in a challenge is obvious, it has rules, follow them.

Well, two things. 1) the aerofoil profile is a NACA supersonic profile, so it's not optimised for slow flight. 2) You'll have to ask Ferram for more because I don't actually use the panels all that much, I tend to just build things & only look at the numbers when there's a gross handling problem - but the graph has no altitude so I suspect it's under best possible conditions. At sea level my craft is hitting flow separation issues at 37m/s & 4 deg AoA, that seems reasonable to me given the wing profile. If we had the wing profile of a glider it would be less so. This is the slowest flying aircraft I can find, and again around 4 deg AoA in level flight is it's sea-level minimum. The wings are similar but not quite the same shape & I think this one has more incidence.

If I have the choice I set text area backgrounds to a very pale cream, something like slightly old paper - warm ( yellows are higher colour temperature ) light colours are easy on the eyes & it's still good contrast if the text is dark. Image issues are solved by a very small border ( preferably one without a sharp outer edge, then it doesn't shout "BORDER" ). I've come directly from android to Windows again & the default is actually kinda usable on my tablet, excess wasted space notwithstanding - under windows on this calibrated monitor is still a blinding lack of contrast & rapidly deteriorates into a morass of swimming dots. Keep going, something is getting there slowly!

If one part of that panel is garbage ( like the out-of-bounds AoA ) then I wouldn't trust any of it. This is as slow as it will fly with those flap settings ( actually flies slower with no flaps, I don't remember what I was doing there ). Cl is way below maximum possible because it doesn't need any more lift to stay level at that altitude. At 8km it's a bit different - I think that is actually the stall speed given a fraction above it gives a lot of control back. This would be the lowest usable speed I guess:

Hum. Can you take a shot of the statics panel with it all zeroed? caveat - I'm still running 1.1, because I haven't played since 1.2 came out & I haven't upgraded my own mod yet - so there could be some new behaviour in the last two FAR versions. If the panel goes *completely* blank then you want to check your logfile for NREs I suspect, because that sounds like a bug & probably not one in FAR. Note the >0 AoA in the derivatives, iirc that means it's gone out of bounds which I'm pretty sure *is* 30 deg AoA but not 100%. It does look like the craft is stalled though given the enormous drag. I presume that isn't what you're seeing.

Zero on the info panels means you've stalled ( or perhaps gone out of bounds of that particular simulation - I think AoA over 30 might do that, I'd have to dig up a plane with the ability to pull really huge AoA to check it & I can't remember what I did with one ). Flaps are currently just modelled as wing pieces iirc, so they will stall like a wing piece - I don't think I've had usable flaps with more than about 40-45 deg deflection, and there's a small part of the envelope you can use that sort of deflection. If you want to make them more effective, make them larger. I believe in your flaps 3 pic you've actually stalled the flaps, given your cD is triple that of flaps 2, that would explain the loss of lift & higher AoA. If you get nothing else, I would get a procedural wing mod for building FAR craft. Flaps on this are limited to 20 degrees, but they're so large they work very well. Any more deflection & they'll stall, especially with the aircraft itself pulling AoA. ( I probably wouldn't fly this slowly though - nice to know it can do it ).

Indeed. However the grey is very grey on android ( despite sending a windows useragent ) but only a little grey on a desktop, but the white areas ( very light grey now I guess ) are about the same. This is what I have... I'm not sure it's actually meant to look like this. Loading it in IE gives gigantic text ( if I tell FF to zoom the font is actually harder to read ), but the colour ( or lack of it ) is the same. Back to figuring out how to squash the user info into something less wasteful, & then get rid of the outer borders...