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

[Van Disaster]

Now, for your problem. It is pretty much "normal" for FAR to calculate drag at engine mount like in picture. Don't know exactly how much of realism FAR caluclates, but it is not entierly unrealistic. When any body travels fast in fluid it leaves area with low preasure in rear area. That low preasure creates forces that pulling back that body - that is why it is called drag.

Not only surfaces in front, but surface area at back is equaly important when comes to drag. You will find more info and better explanation on wikipedia for this than I am able to provide.

Perhaps ferram can give more insight what is calculated and what not in FAR.

That's all very well, but a) there's an engine behind each outer area ( ignoring the tail node ) so that's what should be causing the drag - this presumably is what FAR's open node checking is meant to be simulating, and there's an engine behind each outer area filling the low pressure area with high pressure exhaust

It could be that either it has a bad FAR config or for some reason it's not getting FARBasicDragModel or whatever it was called applied ( or it's being applied and removed ), I've had that a few times in the past - pop the debug menu up & check the part configs. If it's neither, then you'll have to put up with it & wait for Ferram's new drag model.

RE: control surfaces, if they don't have a FAR config they'll work just like stock ones, bugs & all.

[smjjames]

Hm, I see a new thing about sync costs in the github, could those sync costs that cause unplayability also cause this problem http://forum.kerbalspaceprogram.com/threads/109995-Sudden-computer-slowdown-after-a-long-while-of-playing-continues-after-I-close-KSP-have-to-restart?p=1723388#post1723388 ?

[kcs123]

So it looks like I may have discovered a bug in AJE which was capping the thrust of the Sabre M much lower than it should be.

Still though, it's worth asking. Reaction Engines Limited has a simulation of the Skylon's ascent trajectory (xls file), which suggests that the Skylon manages to keep Cd low even at Mach 5 with very high dynamic pressure. Are they not accounting for skin drag, or is there something fundamentally different about the assumptions they're making vs the assumptions that FAR is making?

Also it looks like the Cd value given in the FAR editor doesn't always match what's given in flight under the same conditions. At high mach numbers, the in-flight Cd value is much higher than what's given in the editor. Does the editor analysis not account for skin drag?

Sabre M engine should be like 4 x Sabre S, at least what is in B9 descriptions, don't know why AJE have nerfed that. Main purpose is to have less parts attached on plane for same result in thrust, it is not impossible to make such engines and it is good gamebalance idea from b9 side.

Don't know much about Skylon and what was supposed to be when comes to drag values. Perhaps they have not calculated skin drag at all due to high polished surfaces, who knows.

What I know is that FAR editor analysis shows skin drag properly. It is more noticable if you can compare graphs from previous FAR versions and latest FAR on the same plane. More easy to see if you use low AoA area like up to 10 degree on higher speeds like 2 or more mach. L/D ratio is noticable lower in new versions and red line is visible lot above X axis on graph, meaning some amount of drag is present even if plane is near zero AoA angle.

There is still some bug present with FAR stability analysis, especially if you have some control surfaces with AoA(%) set with other values than zero.

Previously that was problem with flaps/spoilers when you have to click on analysis button couple of tmes until you always have same numbers.

In latest version problem with flaps is solved, but with other control surfaces with some AoA(%) it is still present. Need to make some screenshots and upload them to show.

- - - Updated - - -

That's all very well, but a) there's an engine behind each outer area ( ignoring the tail node ) so that's what should be causing the drag - this presumably is what FAR's open node checking is meant to be simulating, and there's an engine behind each outer area filling the low pressure area with high pressure exhaust

It could be that either it has a bad FAR config or for some reason it's not getting FARBasicDragModel or whatever it was called applied ( or it's being applied and removed ), I've had that a few times in the past - pop the debug menu up & check the part configs. If it's neither, then you'll have to put up with it & wait for Ferram's new drag model.

RE: control surfaces, if they don't have a FAR config they'll work just like stock ones, bugs & all.

Good point. I didn't put in consideration engines and gases from engine exhaust, don't know if FAR calculates that either. I just wanted to say that is not totally unrealistic to have high drag in rear craft area. Problem is that FAR can't show lift/drag values on small part of part area, it can show only for part as whole.

Also, you have to keep in mind that as much as FAR tends to be close to real life aerodynamic as possible some things are not calculated at all. For various reasons, too much effort to calculate it realistically for little influence on overall craft behaviour, too much calculations that could kill game FPS or influence is just too minor to be noticeable in game etc.

[The Space Man]

Can someone explain what the voxelize window thing is for?

[tetryds]

Can someone explain what the voxelize window thing is for?

You should know to expect weird stuff from a dev version.

It will be used for... aerodynamics lol.

Drag, AFAIK.

[BudgetHedgehog]

Can someone explain what the voxelize window thing is for?

Ooh! I know this!

The current [aero model] is a part-based approach, modelling each part as a section of an elliptical cone and approximating properties based on that, with some fudges for the ends of each one. It works, but it's not what I'd like.

Soon though, I'll use a voxel representation of each vessel so that I can do Slender Body theory on them for Mach from ~ 0.6 - 5, use Newtonian Impact above that, empirical values below 0.6, and blend them at the edges to make everything smooth.

It will be awesome.

Edited February 9, 2015 by ObsessedWithKSP

[blowfish]

Sabre M engine should be like 4 x Sabre S, at least what is in B9 descriptions, don't know why AJE have nerfed that. Main purpose is to have less parts attached on plane for same result in thrust, it is not impossible to make such engines and it is good gamebalance idea from b9 side.

Yes, well that's how I discovered the bug. It was a cfg error so easy to fix.

[Talavar]

Was wondering at what altitude you stop getting micro-drag and orbital decay around kerbin?

[smjjames]

Was wondering at what altitude you stop getting micro-drag and orbital decay around kerbin?

Isn't 80km the absolute top of the atmosphere?

[blowfish]

The atmosphere ends at the same place in FAR as in stock - 69 km.

[Talavar]

The atmosphere ends at the same place in FAR as in stock - 69 km.

Can't be.. I'm still getting micro at 70.25 It's exactly why I came here to ask.. lol .. It's nothing major, just very (VERY) small amounts. lol

Bah.. I'm using "NEAR" however I was under the assumption that it used the same atmospheric cap as FAR, since it's just a stripped down version. I'll head over there.

Edited February 10, 2015 by Talavar

[Drew Kerman]

Can't be.. I'm still getting micro at 70.25 It's exactly why I came here to ask.. lol .. It's nothing major, just very (VERY) small amounts. lol

Bah.. I'm using "NEAR" however I was under the assumption that it used the same atmospheric cap as FAR, since it's just a stripped down version. I'll head over there.

your orbit's always going to change slightly unless your ship is on rails (time warp or unloaded, like viewed from the Tracking Station). You'll generally notice your orbit being perturbed more when you move your craft around. If it's highly eccentric, you could easily lose a kilometer or two off your periapse when up near your apoapse and moving around a lot

[Van Disaster]

Yes, well that's how I discovered the bug. It was a cfg error so easy to fix.

Oddly when I last tried AJE it gave me 2.1MN Sabre Ms, so I get the feeling it's not immune to being messed with by something else ( I never did find out what was doing it )

[blowfish]

Oddly when I last tried AJE it gave me 2.1MN Sabre Ms, so I get the feeling it's not immune to being messed with by something else ( I never did find out what was doing it )

The bug I'm talking about is specific to 2.0.x, so if you were using AJE before that different results would have occurred.

[Herra Tohtori]

Hey, would it be at all possible to add hydrodynamics to FAR? It wouldn't need to be anything particularly complicated - just a boundary layer that changes "air" density to 1000 kg/m^3 when a part's altitude coordinate is equal or lesser than zero (sea level).

Basically I was thinking of doing some experiments with aquatic vessels, like hydrofoils, propellers, submersibles with control surfaces, even sailboats with the wind mod - you'd need hydrodynamics to get the keel and rudder working, so that you would be able to sail into the wind.

Accuracy-wise, the biggest difference would be that a simplistic implementation wouldn't take into account things like cavitation. Programming-wise, the boundary between air and water could be problematic if a part is partially submerged, since it would still only have one altitude value and it'd be either above or below the surface, but at the same time, I'm fairly sure that things would work well enough to be useable...

[BudgetHedgehog]

You might like Better Buoyancy, also by ferram4.

[Van Disaster]

Also iirc aerofoils ( and that would mean keels and rudders ) do still work underwater as if it's incredibly dense air.

[Herra Tohtori]

I need to test this, but I'm not sure about that. It might only apply to drag, but if it already works for dynamic lift as well, then that's awesome.

[KerbMav]

I always assumed FAR computed all drag and lift values from the parts physical form and thereby accounting to the lift of the planes body etc. too.

http://forum.kerbalspaceprogram.com/threads/110432-So-are-you-practicing-Pointy-Rockets?p=1731918&viewfull=1#post1731918

Reading this - will I get into trouble with unsupported mod parts then?

Does this plugin work properly with other mods / part packs?

Sure; FAR figures out what the properties of the part should be based on its dimensions and some basic aerodynamic assumptions. If you use a mod and suspect that it causes unrealistic behavior, search the thread to see if it has been brought up / addressed by the latest release; if it hasn't, feel free to bring it to my attention. The only exception is with wing parts, which are more complicated and currently must have their properties specified manually.

Hm ... ?

Edited February 13, 2015 by KerbMav

[NathanKell]

The quote seems very clear to me:

1. Non-wing parts are supported automatically, but can sometimes violate the assumptions FAR makes. If something is behaving "weird" search this thread and, if nothing found, bring it up.

2. Wing parts (wings/control surfaces/etc) *do* need configs written for them, but FAR ships with configs for many parts.

[LayTzz_swe]

Does anyone know how to make FAR use default thrust?

[Svm420]

Based on the last few post i thought id bring this up here too. I am trying to build a ssto with that parts pack. Everything works except the cockpit has crazy lift as you will see if you check the post I have linked any help would be greatly appreciated. Thanks in advance! Edited February 13, 2015 by Svm420

[Undecided]

Newbie to FAR here, trying to adjust to the new mechanics as compared to stock KSP. So I have a few question, if it isn't too much of a bother: When building SSTOs, does stock policy of "move all mass as far forward/down as possible, move lift as far backward/up as possible" still apply? Also, does FAR's game mechanics account for the fact that swept wings (e.g. wings slanted towards the rear, rather than perfectly straight and perpendicular to the hull) are more stable, or is this not in the simulation?

Thanks in advance for answering all the questions. Most literature regarding KSP and aerospace mechanics only applies to stock KSP, so it's harder to learn what works in FAR. If anyone knows any good links for FAR-specific SSTO design, I'd appreciate it as well.

[Crzyrndm]

Wanderfound's guides are probably what you want

http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1353891&viewfull=1#post1353891

[BudgetHedgehog]

Newbie to FAR here, trying to adjust to the new mechanics as compared to stock KSP.

I was never successful with making SSTO planes in stock, so I can't compare it to FAR but in general, the further apart your CoM and CoL, the more stable and less maneuverable your plane is, sometimes to the point of it being nothing more than an uncontrollable dart. In absolute terms, they can be wherever you want, but relatively, they should be close to each other. About 1 balls distance away works well and be sure to check dry CoM as well as it may surprise you.

Swept wings aren't more stable (though, it depends what you mean by stable..), they create less drag and lift in general and as such, are more suited for high speed designs. Yes, FAR simulates any and all wing sweep. The best advice I can give you is to make your plane look like it's supposed to do what you want it it to do. Take inspiration from real life designs. Install pWings to help get that perfect shape and size with none of the messiness and part count stock wings give.

As for FAR-specific craft, as I said, I've only made SSTO planes with FAR so they could be classed as such. I know forum user Wanderfound does the same thing and has many styles and downloadable craft files available. I have a couple on KerbalX that you can try or take ideas from, but really, Wanderfound is your main FAR SSTO guy to see.