ferram4

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

Recommended Posts

Ok now, I tried doing this yesterday but the forum was offline for some reason. @ferram4, We have a slight symmetry problem in voxelization. Let me show you:

usq3Irv.png

See the weird voxels where the left wing connects to the fuselage? And on the leading edge? So ok, this is my modded game, but I could reproduce the exact same result in a pure stock install, just FAR (Kuchemann).

Well, what's going on you say? I experimented a little, and found out everytime I attached a wing low or high on the fuselage, that would happen. It doesn't matter if I use the offset gizmo or if I place the wings underneath the fuselage and then rotate them, happens anyway. If I smack them dead center with angle snap, parallel to the ground, the voxels seem ok and the plane flies nicely. Remember that issue with the pWings? Seems I had two problems at the same time, but the pWings masked this one.

Ok now, steps to reproduce: build basic plane, starting with fuselage. Attach wings normally, then offset them high or low on the fuselage. Check voxel debug. Fly plane and confirm assymetric behaviour.

@the_machemer check this out.

Share this post


Link to post
Share on other sites

@Vegetal I think Ferram has mentioned there are problems with the code for wings, and that he's planning to overhaul that part of FAR.

If this is a bug he hasn't seen, I am pretty sure he will want a craft file that shows the issue, preferably built with only stock parts (or with as few other mods as possible, along with a list of what other mods are required).

Oh, and the log file too, probably. If you look at many of his posts in here, you can see the info he needs to make your report useful to him.

ETA: and definitely exact reproduction steps - and boy, howdy, he means EXACT.

Edited by Gryphon

Share this post


Link to post
Share on other sites

@Gryphon I know he plans to do that, I'm just trying to help because this bug is a sneaky b4stard, it flew below the radar until now. And he's seen the bug before, even I saw it, but we thought it was related to a pWing mod I was using at the time. As I said, I could reproduce the bug on a pure stock install, so, no weird mod interactions here.

If he wants a craft file, fine, but the bug is so simple to reproduce, my steps are *quite* exact. In fact, you could even test it yourself to see if you confirm it. If you don't confirm....well....then there's some weird stuff going on...

Share this post


Link to post
Share on other sites

@genbrien: Found that and got it sorted; some code I thought was firing wasn't.  Until the next release, do the workaround you identified.

@Vegetal: Hrm... I don't think that small asym error would cause a massive issue.  It might create some slightly-off-center forces, but I doubt it would be really noticeable unless something else was off.  I'll investigate; mind sending me the craft file so I can be absolutely sure of what's going on?

Share this post


Link to post
Share on other sites

Hi, I'm having some serious issues with FAR. Most of my rockets are suddenly really unstable after updating everything to 1.1.3 (I know how to build them, it's not an aerodinamic problem). I should mention that I play RSS/RO.

I've realised that the centre of lift wasn't calculated properly or refreshed whenever I added parts in the VAB. Also FAR crashed during "Calculate stability derivatives" (blank FAR window) and couldn't display any of the curves, nor debug voxels. Mach 1 Wave Drag-Area said: NaN. Removing part by part, I found the cause of all this: engines!!

I don't know what mod or since when, added the "shroud" option to add an adapter shroud on top of some engines. I'm guessing RO did this. Well, some engines with this option (e.g. RD-253/RD-275 or RL10 Series) if added to the craft mess with FAR and aerodinamic during gameplay! If I use another engine without this new feature (just like it was on 1.0.5, e.g. ), everything is perfectly fine.
Also, RD-103 and Agena have this feature but do not cause trouble.

Let me know if I need to post this in the RO topic...
 

Edited by Oromis

Share this post


Link to post
Share on other sites

Thanks to discussions with @politas, @pjf, @TeddyDD and the rest of the CKAN crew my biggest objection to CKAN is now resolved, and so CKAN installs of FAR are now supported again. [insert rejoicing]  As of this writing (June 27, 2016) the debugging and diagnostic tools that I hope CKAN to have soon have not been created yet, so I will not be able to easily tell if it is a CKAN metadata issue solely from logs.  If after providing debug logs and repro steps I still can't create your issue, please try a manual install if you haven't already to see if it was a CKAN metadata issue.  If that fixes it, the issue is possibly CKAN-related in some way, and the CKAN guys would like to hear about it in their thread or their github so they can investigate.

@Vegetal: Sorry about the delay, I'll get to looking at that soon.  Things have been happening, as you might have guessed from above. :P

Share this post


Link to post
Share on other sites

I found a conflict with Ferram Aerospace and Kopernikus. The problem is that solar panels do not send energy. They receive sunlight but do not work.

Share this post


Link to post
Share on other sites

Yay! Now I can use 1.1.3. Thanks for the mod, Ferram! It's became a core part to every install of my KSP. 

Share this post


Link to post
Share on other sites

@Vegetal: I did some testing and did fix some asymmetry bugs, but those didn't fix the extent of your craft's unevenness.  Turns out that the vertical tail really is just angled a fraction of a degree off center, with the leading edge a tiny bit further to the left than the trailing edge is.  Swap this for your FAR dll, grab the rotation gizmo in the editor, hit F to switch it to absolute mode, turn on angle snap, select the vertical tail and click on any of the rotation rings to get it to snap and then try it and see if it works.

Once that's done I'll push a new FAR version bundling the new MFI and those fixes.

Share this post


Link to post
Share on other sites

@ferram4 good news. Just played a little and it seems the problem is 90% fixed. I used Atmo Autopilot to keep the plane perfectly level, and it's heading drifted something like .1 degree every 20 seconds or so, measured using MJ.

So, still a tiny bit asymmetrical, but hardly noticeable and not a problem. The front voxels at the wing connector B seemed a tiiiiiiny bit off.

The weird thing is, you tell me my tail stabilizer was somewhat off. I *did* use absolute snap to position it, so I can only assume it's KSP's weridness....oh and I did snap it again just in case, for the test.

Share this post


Link to post
Share on other sites

@Vegetal: If I find anything else to try and fix, I'll look into it.  It doesn't make any sense that these things happen.

Anyway, FAR v0.15.7.1 "Kutta" is out with slight fixes to voxel asymmetry, bundling MFI 1.1.6 to fix the Kopernicus issue and a change to the license.  All art assets and shaders are now licensed ARR, everything else is GPL V3, as it was before.

Share this post


Link to post
Share on other sites
9 hours ago, ferram4 said:

@Vegetal: If I find anything else to try and fix, I'll look into it.  It doesn't make any sense that these things happen.

Anyway, FAR v0.15.7.1 "Kutta" is out with slight fixes to voxel asymmetry, bundling MFI 1.1.6 to fix the Kopernicus issue and a change to the license.  All art assets and shaders are now licensed ARR, everything else is GPL V3, as it was before.

This newest update breaks parachutes, they deploy as usual but dont slow down the craft at all, it works normally in stock but when I install the newest version of FAR and its dependencies the parachutes dont slow down the craft.

Share this post


Link to post
Share on other sites

@rndomguy: That would be a minor problem, wouldn't it?  I managed to find what it is, but it should have affected more than just v0.15.7.1, it should also have affected v0.15.7.

Anyway, to fix that FAR v0.15.7.2 "Lanchester" has been released with the fix to that.

Lanchester is rather notable in that he came up with the concepts behind the circulation theory of lift and the vortex-based lifting line theory for wings before Ludwig Prandtl would, (though it was Prandtl who would mathematically verify it) and also contributed a fair bit to the analysis of airplane stability.  Unfortunately, he was enough ahead of his time that his contributions weren't recognized until after he died.

Share this post


Link to post
Share on other sites

Looking at the last commit, due to the find-and-replace to change the release name FAR apparently now uses Runge-Lanchester integration :)

Share this post


Link to post
Share on other sites
15 hours ago, ferram4 said:

...of course that would happen.  Welp, I need to fix that then. :P

Does the newest update work as attended? btw love your mod :)

 

Share this post


Link to post
Share on other sites
38 minutes ago, ferram4 said:

@rndomguy: That is why Lanchester was released, it's my post right above the last one. :P

The mistaken find-and-replace won't have actually broken anything then? (I suppose if it's everywhere, then it shouldn't...)

Share this post


Link to post
Share on other sites

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.