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

[blowfish]

1 hour ago, Neutrino. said:

Hi.

I'm having an issue with FAR_0_15_6_5_Knudsen. After installation of FAR loading a plane in the SPH and opening the FAR gui all the graphs are blank, all readouts are zero. This is the case with stock planes, the planes that come with FAR, and planes I built myself.

Nothing points to an issue in the logs, but did you run the analysis?  Everything will be blank/zero when you open the window, you actually have to run the analysis to get results.  For instance, in the Data+Stability Derivatives Screen, you press "Calculate Stability Derivatives" - that would have generated a log message like this, which I don't see:

[Log]: Cl needed: 0.843210555399257, AoA: 0, Cl: 0.285476622377246, Cd: 0.532236590286424

As for flight, hard to tell if anything is wrong because I don't see you entering the flight scene at all in the logs

Edited June 14, 2016 by blowfish

[Recon777]

What am I doing wrong?

Yes, I installed manually, following the instructions to copy over all the files myself rather than use CKAN.

This is a brand new career file, so the very first rocket you build pretty much has to be just the capsule, the Flea booster, and the parachute. With FAR installed, I also had to add fins, since the rocket was not controllable without them. That's fine. But it EXPLODED when returning to Kerbin.

Also, is it normal that such a rocket only made it about 65% the height as without FAR installed? Less max height and greater terminal velocity with parachute deployed. But this can't possibly be how you intended the mod to work. Surely, the initial rocket for a brand new career game would not crash. What's going on?

 

 

 

 

[cantab]

All looks normal there to me. The Flea has an impact tolerance of 7 m/s, you touched down doing 7.2, it's gonna go boom. You probably didn't get as high because of the weight of the fins, especially since you left them on the default wing strength of 1, which is overkill for most purposes, and that extra weight might also account for your landing speed.

As I understand it Ferram is targeting realism. He's not going to fudge his aerodynamics just so that the starter rocket lands more slowly.

Edited June 14, 2016 by cantab

[Neutrino.]

Ah. I did read the wiki but didn't notice anything about having to press Calculate Stability Derivatives in order to display the information, when I do that the graphs populate correctly. Which is actually a bit depressing as I was really hoping that the problems I'm having with this plane were at least in part due to a dodgy FAR installation and it seems like that might not be the case.

So I'm left still not understanding what's going on with this plane. It's a very conventional light jet design but the behaviour is just bizarre. In straight and level flight rolling left or right causes to nose to raise by as much as you banked. Adding 20 degress of flap causes it to point 90 degrees straight up in the air. Dropping the undercarriage results in no reduction of velocity. It cruises at about 250kts with about 15% throttle and glides at about 150kts with gear down and full flaps with even the slightest amount of nose down.

The characteristics on final approach are not ideal. With flaps and gear out coming in at about 100kts with 5 degrees pitch up I would expect the throttle to control the rate of descent, and the pitch to control the speed. But it doesn't seem to be working quite right, the speed is still largely governed by the throttle and it seems as if the only way to get the thing to descend is to point the nose at the ground. It really feels as if the air resistance is missing or something.

I've done a circuit of the Space Center in this new log.

https://www.dropbox.com/s/e9t1bsgx01qxx1c/output_log1.txt?dl=0

https://www.dropbox.com/s/g1wka9luw90zro1/MK1 Small Jet1.craft?dl=0

When setting up the control surfaces under the 'Std Ctrl' heading are entries for Pitch, Yaw and Roll but also a setting for Ctrl Dflct. It appears the Pitch, Yaw and Roll values configure the extent to which an input on that axis affects the plane, but that raises two questions. What do the negative Pitch, Yaw and Roll values signify, and how do the Pitch, Yaw and Roll values differ from Ctrl Dflct?

Also is there any way to dial in a specific AOA for the wings other then rotating the whole wing with the editing tools and hoping for the best? I think the wings should have 5-6 degress AOA when the plane is on the level, but I don't see any way to adjust that.

 

[cantab]

Ctrl deflection species the maximum deflection from control inputs. The percentage sliders specify how much of that deflection is used for different control surfaces. For example my spaceplane with one pair of elevons has them set to 90% pitch and 10% roll, which means pitch inputs will use almost all the Ctrl Dflct range while roll inputs will only move the elevons a small amount.

A negative percentage on pitch, roll, or yaw will make the control surface move the opposite way to normal. I'm not sure when you'd use that, but the option is there.

[Recon777]

1 hour ago, cantab said:

All looks normal there to me. The Flea has an impact tolerance of 7 m/s, you touched down doing 7.2, it's gonna go boom. You probably didn't get as high because of the weight of the fins, especially since you left them on the default wing strength of 1, which is overkill for most purposes, and that extra weight might also account for your landing speed.

As I understand it Ferram is targeting realism. He's not going to fudge his aerodynamics just so that the starter rocket lands more slowly.

 

Wait, you can adjust the fin strength? First, I tried not having fins at all, but that made the rocket highly unstable and crash anyway. So I slapped fins on to keep it going straight. 3 doesn't work. Needed 4.

Reducing the fin weight (and limiting the thrust of the engine) got me higher, but it still landed at 7 m/s and blew up. The capsule and pilot survived, so I suppose maybe this is simply a rite of passage to have part of the first rocket explode when the career begins? It's not like there are decouplers when the game begins. Landing with that heavy booster attached to the capsule is the only choice.

Has anyone here not had that happen?

[Gryphon]

5 hours ago, Recon777 said:

Reducing the fin weight (and limiting the thrust of the engine) got me higher, but it still landed at 7 m/s and blew up. The capsule and pilot survived, so I suppose maybe this is simply a rite of passage to have part of the first rocket explode when the career begins? It's not like there are decouplers when the game begins. Landing with that heavy booster attached to the capsule is the only choice.

If you were going to drop the booster with a decoupler, it would have exploded anyway, right?
Anyway, if you want to safely land the craft with an attached booster, you will need more parachute. Have you unlocked the "Survivability" tech? That unlocks the radial chute.

Aside from the changes to aerodynamics, keep in mind that FAR also uses a "lite" version of RealChutes for the parachutes. Under RealChutes, the Mk16 parachute is just intended to land that basic capsule, not a heavier craft with a booster. However, if you use the full version of RealChutes (which gives you more control over your chute sizes), there's a "RealChute Cone Chute" part that is similar to the Mk16, but defaults to deploy a bigger chute. That's what I used at the start of my career mode game.
 

[Neutrino.]

I've been doing some more testing on my simple jet trying to get it to behave sensibly and just encountered an inexplicable and uncontrollable nosedive into the ground.

I improved the handling a bit by moving the wings and adjusting the control inputs. Lowering the undercarriage still has no discernable drag effect but I'm getting used to it.

Considering how much wing it has the plane seems to have a fairly high stall speed becoming unresponsive at about 100kts. Considering delta wing fighters typically land at about 110kts irl this also seems a bit off to me.

Currently I'm experiencing an odd issue whereby on landing, no matter how gently the plane touches down, after touchdown it spins around on the runway. It's a standard tricycle undercarriage, I have steering disabled on all wheels, braking enabled on the back wheels only. No idea what's causing that.

So I'm testing this 'spin on landing' issue flying circuits and I'm on final approach. On the run up to the threshold I have 2 notches of flap dialed in on a stable glide path at 60ms. Approaching the threshold I'm still a bit high so I lower the nose slightly picking up speed to about 65ms. All of a sudden the nose pitches down hard 20 degrees, plane speeds up to 75ms but I have no control whatsoever and it dives straight into the ground.

If anyone has any clue what might have caused that I'd be interested to know. Thanks.

https://www.dropbox.com/s/7kjyh3xb41dm21g/output_log_nosedive.txt?dl=0

https://www.dropbox.com/s/73cvt6d05h9he4o/MK1 Small Jet.craft?dl=0

Win 7 x64
KSP 1.1.2

FAR 15.6.5
Kerbal Flight Indicators
TrimPlus
Deadly Reentry
Kerbal Engineer
ShowFps
KSPRC
scatterer
Planetshine
Kopernicus

 

Edited June 15, 2016 by Neutrino.
kts should have been ms in a couple of places.

[Recon777]

4 hours ago, Gryphon said:

If you were going to drop the booster with a decoupler, it would have exploded anyway, right?
Anyway, if you want to safely land the craft with an attached booster, you will need more parachute. Have you unlocked the "Survivability" tech? That unlocks the radial chute.

Aside from the changes to aerodynamics, keep in mind that FAR also uses a "lite" version of RealChutes for the parachutes. Under RealChutes, the Mk16 parachute is just intended to land that basic capsule, not a heavier craft with a booster. However, if you use the full version of RealChutes (which gives you more control over your chute sizes), there's a "RealChute Cone Chute" part that is similar to the Mk16, but defaults to deploy a bigger chute. That's what I used at the start of my career mode game.
 

 

Well, my thought was simply that to begin the career, one has no technology unlocked, but the mod would likely have not forced the very first (only) possible rocket one can build to have an explosion. Yes, it would have been destroyed anyway if let go by a decoupler, but if we're talking about realism here, isn't it a bit dangerous to have an exploding object directly attached to your capsule when you land? Out of principle, one would think that this might have a chance at harming the capsule and pilot as well. But with FAR installed, it seems that one must simply accept "your first rocket will have an explosion - don't worry about it". Usually, when something explodes on your ship, it means you did something wrong and should rethink your design.

Out of curiosity... why does a SPENT booster explode anyway? Presumably, it's just a metal shell at this point with no fuel remaining? Wouldn't it simply be destroyed without exploding?

[Nansuchao]

4 minutes ago, Recon777 said:

Well, my thought was simply that to begin the career, one has no technology unlocked, but the mod would likely have not forced the very first (only) possible rocket one can build to have an explosion. Yes, it would have been destroyed anyway if let go by a decoupler, but if we're talking about realism here, isn't it a bit dangerous to have an exploding object directly attached to your capsule when you land? Out of principle, one would think that this might have a chance at harming the capsule and pilot as well. But with FAR installed, it seems that one must simply accept "your first rocket will have an explosion - don't worry about it". Usually, when something explodes on your ship, it means you did something wrong and should rethink your design.

Out of curiosity... why does a SPENT booster explode anyway? Presumably, it's just a metal shell at this point with no fuel remaining? Wouldn't it simply be destroyed without exploding?

That is how KSP works. You can blame SQUAD setting a so low impact tolerance on the Flea. Btw I had this happening since 0.90 with FAR, so nothing new.

[ferram4]

@Recon777: I'm not seeing where you have an issue with FAR here.

1. FAR is an aero mod, not a crash tolerance and part survivability mod.  Changing the Flea's crash tolerance is out of scope.
2. The Flea being destroyed on landing does no damage to anything whatsoever; hell, it shouldn't even explode because it lacks explosives inside it, it should just crumple.  However, those effects are also out of scope for FAR.
3. Yes, it will not go as high.  Stock does not model supersonic effects properly and short, relatively fat rockets have terrible supersonic drag.  Drop the thrust so you hit that later and you'll go higher.
4. That isn't the first craft that you have to launch.  Although it's very likely someone will rush to actually fly a rocket, nothing stops you from just launching the pod and grabbing enough science on the pad to get to the next node.
5. Editing the stock tree to make the first vehicle saner is similarly out of scope.

FAR already risks trying to handle too many things.  I will not attempt to fix stock's silly career mode as well, nor do I want to expand into making an aero mod about how things are affected when they crash.  I'm also not going to make the aerodynamics less realistic to save something stupid like that.

@Neutrino.: Perhaps only having 4 degrees of pitch control on the elevators is the problem.  It's got no problem handling 100 knots in the air with about 15 degrees AoA if you increase the pitch authority up to something higher.  The stupidly excessive dihedral and anhedral on the tail and wingtips probably hurts its stall behavior a bit, but otherwise this looks about correct for a plane 6.3 tonne plane with 39 m^2 of wing area that has this degree of pitch stability.

The spinning out is likely something to do with the landing gear changes in KSP 1.1+.  Nothing I can do, I don't change anything about those.

The pitch down, no idea; it sounds like you either stalled something or simply lack sufficient pitch control to keep it on target at low speeds.

[Recon777]

2 minutes ago, ferram4 said:

@Recon777: I'm not seeing where you have an issue with FAR here.

1. FAR is an aero mod, not a crash tolerance and part survivability mod.  Changing the Flea's crash tolerance is out of scope.
2. The Flea being destroyed on landing does no damage to anything whatsoever; hell, it shouldn't even explode because it lacks explosives inside it, it should just crumple.  However, those effects are also out of scope for FAR.
3. Yes, it will not go as high.  Stock does not model supersonic effects properly and short, relatively fat rockets have terrible supersonic drag.  Drop the thrust so you hit that later and you'll go higher.
4. That isn't the first craft that you have to launch.  Although it's very likely someone will rush to actually fly a rocket, nothing stops you from just launching the pod and grabbing enough science on the pad to get to the next node.
5. Editing the stock tree to make the first vehicle saner is similarly out of scope.

FAR already risks trying to handle too many things.  I will not attempt to fix stock's silly career mode as well, nor do I want to expand into making an aero mod about how things are affected when they crash.  I'm also not going to make the aerodynamics less realistic to save something stupid like that.

 

It's okay - I was just curious because usually "Something exploded" means "I did something wrong" and was wondering if I was supposed to be doing something differently if I use FAR. I'm not asking you to do anything.

 

[Gryphon]

14 hours ago, Recon777 said:

It's okay - I was just curious because usually "Something exploded" means "I did something wrong" and was wondering if I was supposed to be doing something differently if I use FAR. I'm not asking you to do anything.

 

Ferram's comments reminded me of another option I forgot - since you are using a mod designed to increase the realism of the aerodynamics, you might look into the various mods intended to increase the realism of the tech tree.
 

[Recon777]

Just now, Gryphon said:

Ferram's comments reminded me of another option I forgot - since you are using a mod designed to increase the realism of the aerodynamics, you might look into the various mods intended to increase the realism of the tech tree.
 

 

I'm actually trying out SETI. Hoping to, anyway. Haven't figured out how to begin with it though. No reaction wheels in the initial tech. Can't control anything. I guess I'll get as far as I can figure out and then ask. At the moment, I'm trying to figure out why I'm getting crashes every time I exit the VAB. Trying to figure out what mod is the culprit. Setting up KSP is often harder than playing it!

[TeiwazVIE]

On 13.6.2016 at 9:59 PM, Dermeister said:

These are awesome videos! plus the guys voice sounds SUper nerdy! it's fitting! i'm picturing the guy with big glasses! LOL!  I love these videos! feels very kerbalish! or guys with a white shirt and a pen in his front pocket explaining stuff!

Haha, I thought the same  Old school 80s IT nerd

[Dermeister]

2 hours ago, TeiwazVIE said:

Haha, I thought the same  Old school 80s IT nerd

YEA!  BUT it's not a bad thing! In games like these nerds are quite appreciated! He sounds like the representation of my inner nerd! 

[lasergladiatrix]

I don't know if this is a known bug, but after I installed FAR, all my vessels start torquing counterclockwise when I turn off SAS, even in vacuum. Uninstalling FAR fixes it. Does anyone know a fix that lets me keep FAR? Thanks!

[Gryphon]

5 hours ago, lasergladiatrix said:

I don't know if this is a known bug, but after I installed FAR, all my vessels start torquing counterclockwise when I turn off SAS, even in vacuum. Uninstalling FAR fixes it. Does anyone know a fix that lets me keep FAR? Thanks!

Do you have any of the FAR flight assists active (even in a vacuum)?

[Acvila]

On 6/14/2016 at 2:53 AM, Neutrino. said:

Hi.

I'm having an issue with FAR_0_15_6_5_Knudsen. After installation of FAR loading a plane in the SPH and opening the FAR gui all the graphs are blank, all readouts are zero. This is the case with stock planes, the planes that come with FAR, and planes I built myself.

In flight a plane I'm testing that flies beautifully in stock is practically uncontrollable with FAR installed, (although this is my first time using FAR so I'm not sure what to expect). The plane behaves as if there is almost no drag, it cruises at 200kt with 10% throttle, accelerates towards the ground with no throttle at all with even the tiniest amount of nose down even if the undercarriage is down, and bounces wildly off the runway on even a gentle touchdown.

In level flight lowering the undercarriage has no noticeable effect (I would expect some increased drag), and rolling left or right appears to pitch the nose up too.

It's quite likely my plane is crap and badly designed, but the fact that the FAR gui is all blank is causing me to think that something else is wrong too. Since no one else seems to be having this issue I suspect I've probably done something incorrectly, but the installation instructions seem fairly idiot proof so I can't work out what I've done wrong.

This is on a clean KSP, the only other mod installed is ShowFPS.

KSP 1.1.2 x64
Win7 x64

https://www.dropbox.com/s/ah4umej3ishjy55/output_log.txt?dl=0

https://www.dropbox.com/s/73cvt6d05h9he4o/MK1 Small Jet.craft?dl=0

Hi.

i tried your model.

you should increase ctrl deflect of winglets at 18-20 and roll at 15

fat 455 spoiler at 19, rear gears brake 25

move front gear from cockpit to mk1 crew cabin and then move with slider toward cockpit. increase spring and dumper of all gear to 1.3

also center of mass is better if it's near rear gear.

activate break when u want to reduce speed  or land.

Edited June 18, 2016 by Acvila

[lasergladiatrix]

4 hours ago, Gryphon said:

Do you have any of the FAR flight assists active (even in a vacuum)?

Yes, I did have those enabled. Turning them off fixed the issue, thank you.

[cantab]

How much are struts near wings likely to affect the aerodynamics? Ie if I put a strut say half a metre in front of the leading edge, is it going to destroy the lift?

[StickyScissors]

This. mod. IS. AWESOME! I don't know why i didn't try this sooner. Much better than stock aero, and now i can glide capsules properly

[Burak]

Partially Solved

[EXC 06:17:41.178] NullReferenceException: Object reference not set to an instance of an object
    FlightGlobals.get_ActiveVessel ()
    FerramAerospaceResearch.FARGUI.FARFlightGUI.StabilityAugmentation.SaveSettings ()
    FerramAerospaceResearch.FARGUI.FARFlightGUI.StabilityAugmentation.SaveAndDestroy ()
    FerramAerospaceResearch.FARGUI.FARFlightGUI.FlightGUI.OnDestroy ()

Hi guys! I have long been trying to solve this problem but i couldn't figure it out. Because of this error, game neglects all atmospheric effects such as drag, lift etc. However, it does calculate dynamic pressure. Any suggestions? All of my mods and game's itself are up-to-date. 

One more:

[ERR 06:15:57.838] [RealChute]: Encountered an error calculating atmospheric density with FAR. Using stock values.
  at System.Linq.Enumerable.Single[KeyValuePair`2] (IEnumerable`1 source, System.Func`2 predicate, Fallback fallback) [0x00000] in <filename unknown>:0 
  at System.Linq.Enumerable.Single[KeyValuePair`2] (IEnumerable`1 source, System.Func`2 predicate) [0x00000] in <filename unknown>:0 
  at RealChute.RCUtils.get_DensityMethod () [0x00000] in <filename unknown>:0 
  at RealChute.Extensions.CelestialBodyExtensions.GetDensityAtAlt (.CelestialBody body, Double alt, Double temperature) [0x00000] in <filename unknown>:0 

Edited June 19, 2016 by Burak

[subyng]

Whoops, wrong thread. 

Edited June 19, 2016 by subyng

[visivante]

I'm still getting the weird seemingly no-mass levitation bug when using FAR and Tweakscale together. This had appeared to be fixed a version back but has since cropped up again.

I made sure to test the issue in a clean install with only FAR, Tweakscale and Engineer installed. Downscaling parts to 50% in VAB generates this NRE:

Spoiler

[EXC 14:32:26.601] ArithmeticException: NAN
    System.Math.Sign (Single value)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorAeroCenter.UpdateAerodynamicCenter ()
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorAeroCenter.UpdateAeroData (System.Collections.Generic.List`1 aeroModules, System.Collections.Generic.List`1 aeroSections)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.Simulation.EditorSimManager.UpdateAeroData (FerramAerospaceResearch.FARAeroComponents.VehicleAerodynamics vehicleAero, System.Collections.Generic.List`1 wingAerodynamicModel)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorGUI.FixedUpdate ()
[LOG 14:32:26.857] TweakScale massScale for FAR usage: 1
[LOG 14:32:26.866] TweakScale massScale for FAR usage: 1
[LOG 14:32:26.875] TweakScale massScale for FAR usage: 1
[LOG 14:32:26.883] TweakScale massScale for FAR usage: 1
[LOG 14:32:27.381] TweakScale massScale for FAR usage: 0.25
[LOG 14:32:27.393] TweakScale massScale for FAR usage: 0.25
[LOG 14:32:27.402] TweakScale massScale for FAR usage: 0.25
[LOG 14:32:27.410] TweakScale massScale for FAR usage: 0.25
[LOG 14:32:28.117] Updating Test V1
[LOG 14:32:28.233] Voxelization Time (ms): 116
[EXC 14:32:28.234] ArithmeticException: NAN
    System.Math.Sign (Single value)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorAeroCenter.UpdateAerodynamicCenter ()
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorAeroCenter.UpdateAeroData (System.Collections.Generic.List`1 aeroModules, System.Collections.Generic.List`1 aeroSections)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.Simulation.EditorSimManager.UpdateAeroData (FerramAerospaceResearch.FARAeroComponents.VehicleAerodynamics vehicleAero, System.Collections.Generic.List`1 wingAerodynamicModel)
    FerramAerospaceResearch.FARGUI.FAREditorGUI.EditorGUI.FixedUpdate ()
[LOG 14:32:28.236] Std dev for smoothing: 0.5 voxel total vol: 2249447.11192044 filled vol: 5315.33348912286

And then when launching the vehicle the Launch Clamps will detach, the engines will fire, yet the craft will just float there as if nothing is happening.

If I reset the downscaled parts to their original size the craft will fly normally. If I remove FAR and try to fly the craft with the downscaled parts it will fly normally. So only when using FAR in conjunction with downscaled parts does this happen.

Any ideas as to why or how to fix this? Apologies if this is a known issue with the latest version, but I'd browsed back quite a few pages and people seem to have this issue fixed already.

Thanks in advance!