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

[nitebomber51]

nitebomber51: by far the easiest way is to install Linux and run 64-bit Linux KSP. It's also the easiest way to avoid getting yelled at by modders.

Notepad++ or the easier option to install linux. Using mods on x64 windows ksp only increases instability, its bad to the point where squad has pulled the windows x64 version. This should tell you something, if the original devs dont support something why should the modders.

I was afraid I would have to use linux for this, I've got a dualboot of GNUbuntu but I would prefer finding where ferram has disabled x64 support. I understand the instability of win64 unity players but I dont think modders should actively shun the use of an unstable project.

On the other hand, I do enjoy being able to whirl kerbals around at 30kmASL at 6km/s xD

[ferram4]

Alright, so FAR v0.15.4 "Glauert" is out, which includes internal ducting (for intakes -> engines) in its calculations of cross-sectional area. This should make realistic looking supersonic vehicles a lot easier. As a result most vehicles that have been area ruled extensively with the previous FAR builds will need to be redesigned for this one.

Also lots of bugfixes, changelog has all of them.

[Temeter]

Even better, thx for the update!

[Nansuchao]

Great Ferram, it seems also much more RAM friendly, and less overheathing issue!!!

[taniwha]

I understand the instability of win64 unity players but I dont think modders should actively shun the use of an unstable project.

It is the unfortunate result of certain members of the community using 64-bit Windows KSP and insisting that bugs that were not reproducible on 32-bit Windows KSP, 64-bit Linux, 32-bit Linux, or MacOS were somehow the fault of the modder. (I am not saying you are such a person).

[Crzyrndm]

@ferram4

    * Supports stock intake part + airbreathing engine part setups, AJE intake part + airbreathing engine part setups, and combined intake + engine part setups

Is there a particular reason these are done explicitly by hardcoded module type? I can understand AJE having a special case since it doesn't use IntakeAir, but this doesn't cover any modules deriving from the stock ones. Wouldn't it be more robust to get all modules that are or derive from the base class (ModuleEngines, ModuleResourceIntake). The difference I see is that contains should be faster, but this is all setup stuff that only gets run once so the impact is minor.

[COLOR=#008000]// Get the first engine module using intake air as a resource[/COLOR]
PartModule module = part.Modules.OfType<ModuleEngines>().FirstOrDefault(e => e.propellants.Exists(p => p.name == "IntakeAir"));[COLOR=#008000] // broken into proper loops when not tab restricted...[/COLOR]
[COLOR=#008000]// NOTE: I cast it all the way back to PartModule to be sure the moduleName field was correct, but ModuleEngines would probably work too...[/COLOR]
if (module != null)[COLOR=#008000] // Found an air breather[/COLOR]
{
    engineType = module.moduleName
    // engine ducting setup
}

[nablabla]

**Fixed NRE with Launch Clamps **

what is an nre?

[Nigeth]

Ferram, what will you do with your naming scheme once you've reached Weber, Weissenberg or Womersley and have run out of dimensionless physical constants? (AFAIK there's no dimensionless quantity beginning with a "Z"

- - - Updated - - -

**Fixed NRE with Launch Clamps **

what is an nre?

Null Reference Exception I'd guess. Accidentally using a reference to an oject that doesn't point to an actual object (a so called NULL pointer or NULL reference)

[nablabla]

ohohoh, I am a programer, shame on me

so was it the reason that launch clamps flew around in the sky?

[Feradose]

I have a problem

it seems to either show the results twice or do the calculations twice then show the results twice

performance drop included

thanks for the help crzrndm

I have yet to figure out the performance drop

Edited July 13, 2015 by Feradose

[Crzyrndm]

That indicates the intake module getting added twice to the part (someone messed up a MM patch). Since FAR doesn't add any stock modules to parts it isn't coming from FAR...

Edited July 13, 2015 by Crzyrndm

[sarbian]

so was it the reason that launch clamps flew around in the sky?

No, that one is stock.

[eddiew]

Alright, so FAR v0.15.4 "Glauert" is out...

Sweet, thanks Ferram! I took the dev version for a drive a couple of days ago, and had 50% more delta-v at LKO than either the previous release, or stock aero, could deliver. Nice trade off between working harder in the SPH and getting greater efficiency for the effort

[hippomormor]

I don't know if this is relevant in this thread (i also wrote in the KAX-thread): FAR_0_15_4 breaks the Kerbal Aircraft Expansion Tail Rotor. It shows a black screen when launching a craft.

Great mod and great work giving people a reason to build airplanes in KSP!

Best regards

[Xgkkp]

Does anyone know of any good guides to creating decent/working airplanes? I've seen http://forum.kerbalspaceprogram.com/threads/52080-Basic-Aircraft-Design-Explained-Simply-With-Pictures but it says a lot about what should work and seems low on actual practical design advice.

Ideally, something that involved using the FAR GUI and learning how to design, and then tweak based on the simulation results. Even more ideally, something like a series that starts with the most basic design with a minimal of parts and building up from there.

Anything like this around?

[tetryds]

Does anyone know of any good guides to creating decent/working airplanes? I've seen http://forum.kerbalspaceprogram.com/threads/52080-Basic-Aircraft-Design-Explained-Simply-With-Pictures but it says a lot about what should work and seems low on actual practical design advice.

Ideally, something that involved using the FAR GUI and learning how to design, and then tweak based on the simulation results. Even more ideally, something like a series that starts with the most basic design with a minimal of parts and building up from there.

Anything like this around?

Not yet, but I am working on it since a very long time ago.

Now I just need time to record and edit the video series, the scripts are all finished.

You can find some specific stuff on my signature.

Wanderfound also has some guides, they may be linked on his sig.

[Wanderfound]

Does anyone know of any good guides to creating decent/working airplanes? I've seen http://forum.kerbalspaceprogram.com/threads/52080-Basic-Aircraft-Design-Explained-Simply-With-Pictures but it says a lot about what should work and seems low on actual practical design advice.

Ideally, something that involved using the FAR GUI and learning how to design, and then tweak based on the simulation results. Even more ideally, something like a series that starts with the most basic design with a minimal of parts and building up from there.

Anything like this around?

See post #3 in the Kerbodyne thread and some of the build videos on the Youtube channel. This one in particular is likely to be useful:

They're all a bit out of date now (particularly in regard to area ruling), but the basic principles still apply.

Edited July 14, 2015 by Wanderfound

[ferram4]

@Nigeth: Well, it's not dimensionless constants; there's no Glauert number, after all. But when I run out of dead aerodynamicists and fluid dynamicists, I'll come up with something.

@Cryzyrndm: Because a more comprehensive method would also handle other resource types as well. I just haven't figured out how to do that yet, but this works for 99% of the stuff that will show up.

@hippomormor: I think what's happening is that it has an intake module defined, but it doesn't actually have a transform for its forward direction. If that's the case, I need to add a workaround for utterly broken modules like that. If that's not the case, then I don't know what's going on.

[Nigeth]

Then I'm looking forward to FAR Navier, and FAR Stokes because many fluid dynamicists died to bring us this SW

[Svm420]

Thermo uses cubes whether or not FAR is installed, though FAR will set the areas based off what it calculates.

Would FAR run into the same issue with conformal radiators seeming larger to the game that the model? Or would FAR "correct" this because it sees the model through voxelization orr something else

[hippomormor]

@hippomormor: I think what's happening is that it has an intake module defined, but it doesn't actually have a transform for its forward direction. If that's the case, I need to add a workaround for utterly broken modules like that. If that's not the case, then I don't know what's going on.

[Voculus]

Thanks again, for the hard work, Ferram. Much appreciated!

[Xgkkp]

So, i'm attempting to understand plane construction - without stepping up to spaceplanes, for now. It'd be nice to just get around on Kerbin with the simplest possible configuration, and build up from there. I've made two similar planes with different wing configurations -

Javascript is disabled. View full album

,

one with the swept wings and one with the wing parts (so, rectangular). It was my understanding that the bigger wing should give a larger lift; however, it seems to require a larger take-off speed, and the data from FAR seems to match this - both in the graphs and the stability derivatives giving a lower Cl (I understand to effectively be the lift) for the rectangular - 0.278 vs 0.369, despite being a larger area. Why is this the case?

In addition, I seem to be getting lots of swaying side-to-side when attempting turn (with joystick - I can't seem to find the documentation any more that explains the FAR assistance options and keyboard just seems to give massive control overkill) and am not sure which numbers I should be reading to describe this - I tried a lateral simulation just setting the Beta value but it just ended up diverging. Probably lots of other control issues but this is the main one I have noticed.

Also; landing - I haven't been able to actually get the planes to slow below ~150m/s. I assume I either need

• More elevons to push the angle of attack into something higher to cause drag when landing (it can't seem to turn faster than a few degrees)
  
• Spoilers (which I cannot do at the moment because of the tech level of the SPH, I am trying to use minimal starting parts/configurations)
  

Any advice on my pitiful attempts at getting Jebediah to fly?

[ferram4]

You have less pitch authority on the rectangular wing. Why? Because you're relying on the control surfaces attached to the wing to take off, rather than a separate canard or horizontal tail. So when you switch to the rectangular wing, you have the control surfaces closer to the CoM, so they can't act on it as well. This is also the reason why you have trouble landing. Make the plane longer, add a separate surface for pitch control, make sure that the main gear are only just behind the CoM so that it can rotate well. The fact that you need a lower lift coefficient for flight at the same speed indicates that the wing is behaving as it should, but that the issue is elsewhere.

For turning, I'm assuming that you're rolling and then pitching up to turn; in that case, the problem is almost certainly that your vertical tail is too close to the CoM, so you have basically no yaw stability or damping at all. It's a good thing you haven't tried to take this thing into high supersonic speeds, because it'll reach some point where it'll just go completely sideways.

Basically, make your design look more conventional, less like a mini-plane.

[VentZer0]

I have this observation that my crafts do something weird when I pitch up at > mach 1 without SAS they tend to roll to the left or right (most times to the right it seems) while pitching up. So basically I wanna do a loop at mach 1 and the craft performs effectively a spiral or barrelroll. With SAS on this doesnt happen and the effect is exaggerated when I have BDarmory ordnance on my wings.

The crafts are totally symetrical, I made sure they are and even built a simple low part count plane to test this, I even checked if my joystick is the culprit - nope it isn't.

Does it have to do with aerodynamic stresses being a factor here?