Jump to content

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


ferram4

Recommended Posts

the 15.9 .version file still states 15.8.1 as its build. so avc is throwing a false positive.

 

Spoiler

{
  "NAME"     : "Ferram Aerospace Research",
  "DOWNLOAD" : "https://github.com/ferram4/Ferram-Aerospace-Research/releases",
  "URL"      : "https://raw.githubusercontent.com/ferram4/Ferram-Aerospace-Research/master/GameData/FerramAerospaceResearch/FAR.version",
  "GITHUB" : {
    "USERNAME"   : "ferram4",
    "REPOSITORY" : "Ferram-Aerospace-Research"
  },
  "VERSION" : {
    "MAJOR" : 0,
    "MINOR" : 15,
    "PATCH" : 8,
    "BUILD" : 1
  },
  "KSP_VERSION_MIN" : {
    "MAJOR" : 1,
    "MINOR" : 2,
    "PATCH" : 0
  },
  "KSP_VERSION_MAX" : {
    "MAJOR" : 1,
    "MINOR" : 2,
    "PATCH" : 2
  }
}
 

 

Link to comment
Share on other sites

I have an issue in 1.3.1. I have ABSOLUTELY symmetrical rocket, one center booster and four boosters around. But when I lounch it it starts to spin around in many dirrections. Had to uninstall this great mod. Does someone encoutered this before?

Link to comment
Share on other sites

2 hours ago, Battou said:

I have an issue in 1.3.1. I have ABSOLUTELY symmetrical rocket, one center booster and four boosters around. But when I lounch it it starts to spin around in many dirrections. Had to uninstall this great mod. Does someone encoutered this before?

Please post a picture of this rocket, and a picture in VAB with the FAR derivatives window.  And / or the craft file, if there aren't lots of mods needed.

Edited by lordcirth
Link to comment
Share on other sites

16 hours ago, Battou said:

I have an issue in 1.3.1. I have ABSOLUTELY symmetrical rocket, one center booster and four boosters around. But when I lounch it it starts to spin around in many dirrections. Had to uninstall this great mod. Does someone encoutered this before?

Let's see a screenie with the debug voxels on. Some parts (even some of the stock parts) make pretty pictures sometimes - which means they ain't gonna make space...

Link to comment
Share on other sites

I have a bit of a problem with parts (a specific part) failing due to atmospheric stresses. Is there a way to increase a part's strength through a MM config?

I have KJR installed and this is in a RO/RP-0 game. The issue happens quite often during reentry (doing a Falcon 9 landing).

I'd mostly be looking for a part config since this pretty much always happens to the center engine but any suggestions are welcome.

 

Thanks in advance

Link to comment
Share on other sites

Is there a known problem with B9 Procedural wings interacting poorly with FAR? I seem to remember there used to be some issue that was solved a long(-ish) while back. I am having a strange issue when building a certain style of plane. When i build certain delta wing design with procedural wings it yaws and/or rolls uncontrollably, sometimes with nice green derivative numbers. Also, without changing anything, the derivative numbers seem to change depending on whether it is a fresh load of the craft or a revert to hanger situation. I haven't quite nailed this down to a specific behavior yet so I'm not going to say its a bug or not. I'm still collecting info to see if it's something on my end and that's the point of this post.

So this plane yaws left hard without any input.

4A7DE6342298851DF833F12E982F7558157D37E9

And this one rolls right even from a 90 degree left bank. It also rolls right when pitching up or down.

584925FA14B156B6FC7053EB743C69753369B5F2

But these 2 fly normally

D1885CA35E0A256D89B2526CFFAB814ECAFA0538

The first two planes have something in common the other two do not. They have a section (or 2) of near body length short wings before the main wings start. The plane in the second picture is the most problematic of all the planes that I have created. The Mw number(and others) is red and, as I understand it, moving the wings up or back should help. When I move the wings up or back Mw gets worse. Almost everything I do to this plane has the opposite effect that it did on every other plane I have built. All 4 of these planes have been created in KSP 1.3.1 with FAR 0.15.9 and many other mods installed.

So, what do you think? Is this more likely a FAR problem with B9 procedural wings or something else?

Link to comment
Share on other sites

More likely design issue than any bug with PW or FAR. Hint: check stability derivates at altitude and speed when you experience unstability with craft. Default speed at sea level is good to reveal takeoff/landing issues, but might not show issues at supersonic/hypersonic velocity and higher altitude.

Link to comment
Share on other sites

There could be a design issue as kcs123 said, but there's definitely a problem with B9 PWings and FAR somewhere. A few pages back we had a discussion about it, and I created an experiment with a plane which had one B9 wing and one stock wing, with as close to identical size as I could get. It took quite a lot of control input to get it to fly straight.

No clue about which side has the bug, or if it's related to what you're seeing. I think the person who brought up the issue reported the issue on both sides, so hopefully they're at least aware.

Link to comment
Share on other sites

3 hours ago, kcs123 said:

More likely design issue than any bug with PW or FAR. Hint: check stability derivates at altitude and speed when you experience unstability with craft. Default speed at sea level is good to reveal takeoff/landing issues, but might not show issues at supersonic/hypersonic velocity and higher altitude.

I design a lot of planes, it is my favorite thing to do in KSP. I look at the derivative numbers a lot and also tab out to google so I can remember what half of them represent and the normal fix for a bad one. Which one is the roll with respect to accel?...Well OCD just got me, now I have to look it up.

This problem exist from take off (~150m/s) all the way out to mach ~7 or 8 (when the wings try to melt). It also will not climb without rolling right so, higher than about 9 km is impossible. As far as derivatives go, they go green at 0.45 mach (just barely). At mach 1.0 they look really good. I have a couple of other planes that have very similar derivative numbers and they fly just fine. I'm currently trying to redesign the wings into many different sections but achieving the same rough shape to see that has any effect of my problem.

@Maeyanie I looked a couple of pages back and didn't see anything which is why I posted. But, that happens to me often I definitely will go look again, ty for the info.

Link to comment
Share on other sites

Only "issue" with PW is their weight. People does not pay attention to it and because of it whole craft does not behave the same as craft with same wing area and same shape, but made with stock parts.

As for roll/yaw issue for shown crafts, it is due to craft designs. You got some strange dihedral and anhedral angles on wings and control surfaces and too small vertical surfaces to help you to overcome any possible issues. My best guess is that you created it like this more for aestetic purposes than for any other functionality reason. Too much of dihedral angle can be overkill and tend to alway roll your plane on one side or other to establish equilibrium on roll axis. How much of dihedral angle is enough and what is too much depends from craft to craft, depending on craft weight, location of COM, wing surface etc.

@tetryds, or some other moderator, can you move all of our posts to FAR craft repository thread ? There is no reason to clutter this threads with any further posts here, because there is not bug involved here, just craft design flaws.

Edited by kcs123
Link to comment
Share on other sites

6 hours ago, kcs123 said:

Only "issue" with PW is their weight. People does not pay attention to it and because of it whole craft does not behave the same as craft with same wing area and same shape, but made with stock parts.

No, it's not. This is from the previous discussion: https://www.dropbox.com/s/bzyyos0q51p6dlm/Kerbal Space Program 2017.07.05 - 23.51.46.23.png

CoM is pretty close to the middle, so the weight is (roughly) the same on both sides. But CoL is off to the left, which has the stock wings.

Then in flight: https://www.dropbox.com/s/2stujdvyhfe2sio/Kerbal Space Program 2017.07.05 - 23.58.37.24.png

It needs large input of right roll and yaw to keep straight. If the B9 wing was aerodynamically the same except heavier, it would need roll to the left to compensate.

Link to comment
Share on other sites

@Maeyanie While that seems to prove there is some issue, why would a plane built by mirror symmetry(all from the left side) exhibit uneven results? Your last post prompted me to stare hard at the voxels and I don't see anything. I also took a shot of the plane in flight and on the runway with the aero forces display on. Now that is a bit interesting and proves there is something wrong.

The voxels and CoM/CoL. The CoM and CoL are perfectly aligned. The voxels around the S2 crew cabin seem weird but I don't often look at voxels, so I don't know.

BD2084F13773E86BA65E508B64CA6FD35DFFB887

B8CAA85827EE7C57B2BEA75FD3D7BE00B2548946F6C675DC30F2ACA1EA0B162D26F924EA959C131F

The middle pic is with no control input except throttle but SAS is on. The last pic I tried to get straight and level without SAS and no engine output but that was not happening even when the speed was ~mach 3. If anyone has an idea of a diagnostic measure, I would surely try it. I'm also not seeing any exceptions in the log.

@kcs123 While I normally appreciate any opinion, even (or especially) if it proves me wrong, your post seemed kind of hostile. It also was not helpful in the slightest and gave me the impression you did not read what I had posted. As for the strange dihedral angles, the main wings don't have that that much and not out of line with other craft I have created that fly quite well. If you are referring to the center "vertical" wings, the angle there is to dial out roll when yawing. Also, not strange or out of line with real world aircraft (see sr-71, f/a-18, F-4 flying bandaid phantom, etc). And, finally, yes I create planes with aesthetics in mind for the original concept after, of course, intended purpose but, we all know(or should), that the difference in concept and the actual thing that works can be very different. This particular plane has been redesigned several times trying to dial out roll right under ALL circumstances.

 

Link to comment
Share on other sites

Yeah, unfortunately I can't really help much with your specific problem, as it's beyond my limited understanding of these things. :)  All I can say is that there is some compelling evidence that there is an aerodynamics bug somewhere with B9 PWings and FAR... which is a shame, because they're really nice wings.

Link to comment
Share on other sites

1 hour ago, Maeyanie said:

Yeah, unfortunately I can't really help much with your specific problem, as it's beyond my limited understanding of these things. :)  All I can say is that there is some compelling evidence that there is an aerodynamics bug somewhere with B9 PWings and FAR... which is a shame, because they're really nice wings.

I mainly use them for part reduction.....but they also look really cool. There is, I believe now, a bug which makes it beyond anyone's ability to fix other than the mod creators/maintainers of either or both B9 Pwing and FAR. Your example is easily replicable. Mine, on the other hand, is strange enough to make it hard. It at least starts with creating a stack of short span, body length wings and then the problems start from there. I did finally throw together a test. I recreated the problem craft with more small wings stuck together. I didn't recreate the exact geometry but I got close enough. The canards I used are from the first try at the original, the body is exactly the same (started from the same craft file). the new plane actually flies straight, returns to 0 roll without input, and pitches up without roll(although yaw still needs work). The mass of both planes are close, 148t problem, 146t new. ref area is 216 m2 vs 179 m2, scaled chord 6.05 vs 3.5, scaled span 2.83 vs 2.65. with those numbers the problem plane should fly much better. More span, more wing area and such for the same mass, right? The new plane is more stable and flies as expected at all speeds and at altitude.

For now, I will just use more, smaller wing parts to make up larger, longer wings since that seems to assuage my issue. I will keep the problem craft file around if anyone wants it to test. 

The new plane for reference.

15F1AB8BF3CB1FBA0F4C81B0D1801B61DC90E2A6

Link to comment
Share on other sites

17 hours ago, Maeyanie said:

No, it's not. This is from the previous discussion: https://www.dropbox.com/s/bzyyos0q51p6dlm/Kerbal Space Program 2017.07.05 - 23.51.46.23.png

CoM is pretty close to the middle, so the weight is (roughly) the same on both sides. But CoL is off to the left, which has the stock wings.

Then in flight: https://www.dropbox.com/s/2stujdvyhfe2sio/Kerbal Space Program 2017.07.05 - 23.58.37.24.png

It needs large input of right roll and yaw to keep straight. If the B9 wing was aerodynamically the same except heavier, it would need roll to the left to compensate.

I was not active at the time when that discussion was actual on forum, so I missed details about it. But, what I know is that data for PW is calculated in different way than it is for stock a like wing parts. For stock wings it is enough to provide data in MM patch like it is described at FAR github page. It might involve importing to blender to get more accurate data for it. PW use plugin to provide those data to FAR each time you change shape, length or width of wing. I can only assume that equation involved in this provide similar but not exactly same results. Info about it is buried somewhere in PW code, I bumped by accident on it while looking for something else on github page and there was recent change on that part of code in pull request. People involved in development of PW can tell more about it.

So, some differences from stock part can be expected, but is difference is really that much ? If you use PW on both sides of craft there will be no asymmetry in lift. There was issues in the past with PW symmetry, IIRC it happened along with KSP update to higher version, but that was solved long time ago. There was bug with wrong flap deflection on symmetry PW part too, but it is also solved, there is no longer issue with it in KSP 1.3.1 and latest available PW version on github.

Way too litle info is provided to tell exactly is it PW bug involved or not, I will not say exclusively that there is no any, but I can tell why I think that is not. While screenshot tells that there is asymmetry in lift, it does not tell good enough how much asymmetry is in wing weight for one side or another. It is hard to tell with eyeball and fuel tank as middle part is quite heavy compared to wing parts to figure out weight difference. On first sight I was thinking that PW is on left side, not on right side, but regardless, you need to have symmetric weight as much as possible to exclude it as possible reason of instability.

For more accurate data use only lightweight as possible "hull" part, make it as root part of craft, something like I-beam, truss part or hollow Mk1 struct part. Less weight as possible for that central part would give more accurate data. Then attach wings on such part and adjust weight so both sides weight the same. Put that aside or create subassembly of it. After that create craft using such adjusted parts when you are sure that there is no weight asymmetry involved. Another thing to consider is to get at least one structural part more on hull behind main wings, to make whole craft more sideslip/yaw stable. You need to test such things on stable craft designs in first place, to be able to tell if bug is involved or not. For now, based only on provided picture I can only speculate and tell that craft on picture is yaw instable and when you try to compensate yaw you usualy get "parasite" roll involved too and vice versa.

15 hours ago, AngrybobH said:

 

@kcs123 While I normally appreciate any opinion, even (or especially) if it proves me wrong, your post seemed kind of hostile. It also was not helpful in the slightest and gave me the impression you did not read what I had posted. As for the strange dihedral angles, the main wings don't have that that much and not out of line with other craft I have created that fly quite well. If you are referring to the center "vertical" wings, the angle there is to dial out roll when yawing. Also, not strange or out of line with real world aircraft (see sr-71, f/a-18, F-4 flying bandaid phantom, etc). And, finally, yes I create planes with aesthetics in mind for the original concept after, of course, intended purpose but, we all know(or should), that the difference in concept and the actual thing that works can be very different. This particular plane has been redesigned several times trying to dial out roll right under ALL circumstances.

AngrybobH, what part of my previous post you found to be "hostile" ? I readed and looked on provided pictures before posting. I provided link for better details how dihedral and anhedral wing angles influence craft behaviour. If you readed it, you should know what is off on your craft designs.

This craft:

4A7DE6342298851DF833F12E982F7558157D37E9

It have too much dihedral angle that push craft in roll to one side or another to establish equilibrium, it often require only small input from pilot to disturbe it. Usually it would put one wing side to be aligned to horizont or close to it, while other would be up in air. Example of dihedral overkill.

Following craft:

584925FA14B156B6FC7053EB743C69753369B5F2

It is hard to tell for me only based on this picture is it exactly same or not as picture from flight. On above picture it looks like outher wing part is cranked, but look can be deceiving dut to change in shape at same time. It does not look same as craft on following picture, but it can be due to camera angle, it is hard to tell.

B8CAA85827EE7C57B2BEA75FD3D7BE00B2548946

I assumed that you eyeballed real craft designs for aestetic purposes and you think that it is "right" thing to do. It is not always a case, you need to understand why some real life crafts looks like they looks in the first place and reasons behind it before you start to copy such designs. For example, SR-71 have anhedral or "inwards" angle on vertical tail surfaces not due to better aerodynamic stability, but to help with radar "stealth" performance by deflecting incoming radar beams upwards as much as possible instead to reflect it to beam source. It does not help to stability at all. A lot of other fighter planes you mentioned have anhedral tails on purpose to make it more maneuverable and that is often not on the same road as stable craft designs. Some of planes could not be controlable without help of fly by wire systems.

To improve craft on above picture you can do following things:

  1. get rid of most vertical surfaces you have put there for aestetic reasons, especially those on bottom part of craft. Those only reduce tail clearance when comes to takeoff/landings.
  2. Put light dihedral angle on main wing, 1-3 degree should be enough
  3. put small vertical surfaces on wing tips strictly perpendicular to whole craft hull, not with wing. In comparison to main wing you should have 91-93 degree, but 90 degree in comparison to hull
  4. leave only vertical tail surfaces that are above engine, you dont need ones in the middle at all, but make those above engines approximately 1.5 - 2X larger than those ion the middle currently are

Those are only design flaws I was able to detect from provided pictures. And I don't even consider myself as expert, there are much smarter people than me around KSP forums that can you tell more about it. I wrote this in more details to help you. Would you listen provided tips or not is up to you, but It is not intended to be hostile by any means to you or any other user. Just pointing out design flaws that were obvious to me on the first sight.

Link to comment
Share on other sites

@kcs123 If you did read my posts, you did not understand them. The pics in your last post show the 2 planes (yes the middle and lower pic are exactly the same plane) that I have trouble with but the first one is not a roll instability it is a yaw left at all times which I have fixed by removing the white section of the wing right up next to the body and that makes no sense as to why that fixed it (wings are still the same dihedral angle). The second plane rolls right for no good reason without input at all speeds and rolls right with only pitch (up or down) input. I fixed that plane by making the wings in smaller sections (more than double the parts, see my previous post). It seems there is a problem with the way FAR handles body length B9 Pwings or the way B9 gives data to FAR. I'm starting to think the old symmetry bug has returned. 

1 hour ago, kcs123 said:

If you use PW on both sides of craft there will be no asymmetry in lift.

You would think but the craft in question exhibits asymmetrical lift. As for the 1-4 or your post, 2. main wings have 4-5 degrees dihedral, 3. the outer vertical surfaces are exactly vertical. 1 and 4, there are no "aesthetic" only parts, clearance is not an issue and the vertical wings are shorter on the bottom because the CoM is higher than the true center line, and the perfect yawing control surface would be directly inline with the CoM with enough distance to have the required force. That set up would give only yaw without any roll. I use downward surfaces to balance roll out of yaw and provide sideslip reduction when flying at extreme altitude. This works in KSP with FAR installed and is most notable with heavy planes that can get huge sideslip problems. This might even work in real life but my engineering prowess lies in wheeled vehicles not aerospace.

Link to comment
Share on other sites

On 19/12/2017 at 1:03 PM, kcs123 said:

So, some differences from stock part can be expected, but is difference is really that much ?

You can see the control position in the screenshot, there is quite a bit of right rudder and aileron to keep it flying straight. A lot more than I'd expect from just some slight discrepancy in calculations.

Quote

You need to test such things on stable craft designs in first place, to be able to tell if bug is involved or not.

It was reasonably stable, actually. I flew it around in patterns both directions for a while before landing it, and once it was trimmed to fairly hard right rudder and aileron as the screenshot showed, it was completely flyable. Aside from not having enough control authority to do right turns very well.

Quote

For now, based only on provided picture I can only speculate and tell that craft on picture is yaw instable and when you try to compensate yaw you usualy get "parasite" roll involved too and vice versa.

The rudder is above the CoM, so yes, likely some of the leftwards roll is from that. The discussion at that time was more based around whether or not the drag (and L/D ratio) on PWings was significantly worse than stock parts, so I was more testing for if it would yaw to one side. Which it did, but the opposite side of what would be expected if they had worse drag.

 

Not saying this is necessarily related to the problems AngrybobH is having. It doesn't sound like it is, from his posts since then. This was also a while back, so the problem may have been fixed by now for all I know, I haven't re-tested it recently... I don't recall seeing CKAN offer me any new versions since then, but, that could be me failing on the "recall" part of that. :)

Edited by Maeyanie
Link to comment
Share on other sites

Sometimes I get this warnings on screen (debug messages activated) - copied them from my log file.

[LOG 17:45:44.617] [PR] Generating data for Pioneer 1.6 C Gael Sample Gore Debris
[LOG 17:45:44.705] Voxel Element CrossSection Area: 0.0131687066177136
Std dev for smoothing: 3 voxel total vol: 377.793683103545 filled vol: 65.4658101529774

[LOG 17:45:44.730] [12/23/2017 5:45:44 PM [x] Science!]: <Trace> (StatusWindow) - ScienceThisBiome: 3 / 1
[LOG 17:45:44.828] Trajectories: Initializing cache
[ERR 17:45:44.834] [FAR] NaN Prediction Section Error: Inputs: AtmDen: 1.257265 Mach: 0 Re: 0 Kn: NaN skin: Infinity vel: (0.0, 0.0, 0.0)

[ERR 17:45:44.835] [FAR] NaN Prediction Section Error: Inputs: AtmDen: 1.257265 Mach: 0 Re: 0 Kn: NaN skin: Infinity vel: (0.0, 0.0, 0.0)
 

[LOG 17:45:44.617] [PR] Generating data for Pioneer 1.6 C Gael Sample Gore Debris
[LOG 17:45:44.705] Voxel Element CrossSection Area: 0.0131687066177136
Std dev for smoothing: 3 voxel total vol: 377.793683103545 filled vol: 65.4658101529774

[LOG 17:45:44.730] [12/23/2017 5:45:44 PM [x] Science!]: <Trace> (StatusWindow) - ScienceThisBiome: 3 / 1
[LOG 17:45:44.828] Trajectories: Initializing cache
[ERR 17:45:44.834] [FAR] NaN Prediction Section Error: Inputs: AtmDen: 1.257265 Mach: 0 Re: 0 Kn: NaN skin: Infinity vel: (0.0, 0.0, 0.0)

[ERR 17:45:44.835] [FAR] NaN Prediction Section Error: Inputs: AtmDen: 1.257265 Mach: 0 Re: 0 Kn: NaN skin: Infinity vel: (0.0, 0.0, 0.0)

This happens often, when leaving the atmosphere and having debries from prior stages behind me. 

 

Is there something has to be done about this?

(Further explainations: I am playing with Galileo Planet Pack in a 6.4 up scaled universe.)

Link to comment
Share on other sites

3 hours ago, New Horizons said:

Sometimes I get this warnings on screen (debug messages activated) - copied them from my log file.

I recommend only turning on the debug messages if you're trying to troubleshoot a particular issue and are looking for real-time reporting at the time of a reproducible event/bug. Otherwise, reading the log will show a lot of "errors," which are usually safe to ignore. It's just feedback in case of real bugs.

Have fun with your 6.4x scale GPP!

Link to comment
Share on other sites

1 minute ago, danfarnsy said:

I recommend only turning on the debug messages if you're trying to troubleshoot a particular issue and are looking for real-time reporting at the time of a reproducible event/bug. Otherwise, reading the log will show a lot of "errors," which are usually safe to ignore. It's just feedback in case of real bugs.

Have fun with your 6.4x scale GPP!

You're just right. With debug mode on I already found some problems in my overly modded game and one can learn a lot about the mechanics behind modding :-)

Link to comment
Share on other sites

4 hours ago, New Horizons said:

Sometimes I get this warnings on screen (debug messages activated) - copied them from my log file.

1

Are you using Trajectories? Those look like errors Trajectories reports when it tries to plot an atmospheric track and can't when using it under FAR.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...