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

[ss8913]

@ferram4 I'm still having that "craft A gets within physics range of craft B and the craft that I'm not controlling spikes all of its heat values to well over 100% and the entire craft explodes" bug.  Now, it is true that I am using CKAN.  Would you suggest I simply do a clean install of FAR (and modular flight integrator?  or is that one ok from ckan?) and test it again, or is this still a known outstanding issue in the code somewhere?  I pretty much have to always use the ALT-F12 cheat for "Ignore max temperature" or I can't do any docking maneuvers or visit any bases.  and it causes other problems, like reactors will shut down even with the ALT-F12 enabled and I'll lose electric charge and life support/attitude control on the station/ship being docked with.

[blowfish]

17 minutes ago, ss8913 said:

@ferram4 I'm still having that "craft A gets within physics range of craft B and the craft that I'm not controlling spikes all of its heat values to well over 100% and the entire craft explodes" bug.  Now, it is true that I am using CKAN.  Would you suggest I simply do a clean install of FAR (and modular flight integrator?  or is that one ok from ckan?) and test it again, or is this still a known outstanding issue in the code somewhere?  I pretty much have to always use the ALT-F12 cheat for "Ignore max temperature" or I can't do any docking maneuvers or visit any bases.  and it causes other problems, like reactors will shut down even with the ALT-F12 enabled and I'll lose electric charge and life support/attitude control on the station/ship being docked with.

Here's what you can do:

1. Create a clean install of KSP (this can be separate from your main install)
2. Verify that the issue does not exist without FAR
3. Install FAR and MFI (and MM) without CKAN
4. Verify the issue now exists

Edited January 21, 2016 by blowfish
lists are a thing

[ss8913]

3 hours ago, blowfish said:

Here's what you can do:

1. Create a clean install of KSP (this can be separate from your main install)
2. Verify that the issue does not exist without FAR
3. Install FAR and MFI (and MM) without CKAN
4. Verify the issue now exists

I will try that right now.  Will just move GameData out of the way and create a new save after installing those mods manually.

[ferram4]

@Rodhern: Derivative calculation is set for level flight (read: 1g, unless velocity is large enough for planetary curvature to reduce the need for positive lift) and control surfaces are at neutral for that.  Generally, >0 deg means the solver failed in some way, and that's normally what happens when you try to do an stability deriv analysis of a vehicle that is incapable of creating enough lift for level flight at the conditions you gave.

 

@ss8913: Yes, there was another issue, try with the dev build (which you can't get through CKAN).  The dev build will be released (and CKAN will hopefully do whatever shenanigans they do as middlemen) in a few days, I just want to make sure that nothing horrible broke from my changes.

[ss8913]

7 hours ago, blowfish said:

Here's what you can do:

1. Create a clean install of KSP (this can be separate from your main install)
2. Verify that the issue does not exist without FAR
3. Install FAR and MFI (and MM) without CKAN
4. Verify the issue now exists

I have done this.  I can confirm that FAR definitely causes a heating issue.  It's not explosively evident with only stock parts since there's fewer things onboard capable of generating heat (no nuclear reactors), but note these two screenshots.  First is with FAR, second is without FAR, all else being equal: ... ok so i can't upload screenshots.  anyway I had 2 identical stock craft with stock radiators (small).  without FAR, when craft B approaches craft A to dock, both ships have cool radiators.  With FAR installed, when B approaches A, A's radiators are glowing orange-hot despite having been in orbit for over a day with nothing generating heat, whereas craft B has just arrived and finished a heat generating burn and has cool radiators.

 

@ferram4thanks for the tip, I will go find the dev build and install it into my normal GameData.  Also I didn't realize until now that you did KJR - doing my experiment just now without KJR was.. interesting.  You *really* miss KJR when you no longer have it   great mod on both counts.  FAR is my #1 won't-play-KSP-without-it mod.  KJR is on the top 10 list also.

 

 

[ss8913]

I'm having some trouble finding the dev build.. where's the best place to get it?  Thanks

[blowfish]

13 minutes ago, ss8913 said:

I'm having some trouble finding the dev build.. where's the best place to get it?  Thanks

Go to the main page on Github -> click "Download repository"

[ss8913]

you mean 'download zip'?  from master branch, yes?

[blowfish]

8 minutes ago, ss8913 said:

you mean 'download zip'?  from master branch, yes?

Correct

[ss8913]

got it, installed it... unfortunately I ran into some fatal issues - Mechjeb won't recognize a target set, loading a save puts the camera out near an asteroid field, and then trying to exit the game after those 2 things happened caused a CTD.  Putting back my old FerramAerospaceResearch dir (the release version) fixed all of these problems, but I still have the original one.  do I also need a dev build of ModularFlightIntegrator, perhaps?

[kcs123]

@ss8913

You should be fine with official release of ModularFlightIntegrator. However when you have made new fresh install, be sure that prior install of new versions you have deleted all folders in GameData except SQUAD.

Sometimes even empty folder may confuse MM to apply some patch that otherwise should not be. That is most common faults of CKAN installs too.

Try to reproduce bug on new game, hyperedit might help to put crafts in orbits - to do testing less tedious and in faster way, that one should not interefere with FAR and cause extra issues. There is high chance that you older save game is corrupted somehow due to various (un)installs.

[NathanKell]

@ss8913 remove the MJFARExtension dll. It's for the release version not the dev version, and will break MJ if you have dev FAR installed.

...ask me how I know. >.>

[ss8913]

3 hours ago, NathanKell said:

@ss8913 remove the MJFARExtension dll. It's for the release version not the dev version, and will break MJ if you have dev FAR installed.

...ask me how I know. >.>

*this* .. this makes a ton of sense and I should have thought of it, and yes I do have the .dll installed.  I will try this tonight after work.

[Razum]

Hello everyone!

I encountered a strange problem with FAR and B9 procedural wings. When installed separately, these mods work just fine, but together they give a problem with a mass of procedural wings and, as result, CoM position. In the simple case, the mass of the Pwings doubles(stock wings work normally), but if a lot of wings, in more complex craft and if used mass/strenght tweaks, Pwings masses can even become negative in some cases.

All mods(except for KER) installed via CKAN on fresh KSP 1.0.5.1028 instance, created especially for test this issue.

Mods: B9 Aerospace Procedural Parts - fork 2.1; Crossfed Enabler 3.3; FAR 15.5.4; ModularFlightIntegrator 1.1.2.0; Module Manager 2.6.18; Kerbal Engineer Redux 1.0.19.2.

https://drive.google.com/open?id=0B-lHX_zOo86nbW13REZGdHlIZzg - Doubled mass.

https://drive.google.com/open?id=0B-lHX_zOo86nZGJRcGlQNUlaSHc - Negative mass, but this screenshot from heavily modded version, where it was discovered...

Please anyone help! I can`t play KSP without FAR and Pwings anymore!

P.S. I`m sorry for the grammar mistakes, I used google translator...

Edited January 23, 2016 by Razum

[kcs123]

15 minutes ago, Razum said:

Hello everyone!

I encountered a strange problem with FAR and B9 procedural wings. When installed separately, these mods work just fine, but together they give a problem with a mass of procedural wings and, as result, CoM position. In the simple case, the mass of the Pwings doubles(stock wings work normally), but if a lot of wings, in more complex craft and if used mass/strenght tweaks, Pwings masses can even become negative in some cases.

All mods(except for KER) installed via CKAN on fresh KSP 1.0.5.1028 instance, created especially for test this issue.

Mods: B9 Aerospace Procedural Parts - fork 2.1; Crossfed Enabler 3.3; FAR 15.5.4; ModularFlightIntegrator 1.1.2.0; Module Manager 2.6.18; Kerbal Engineer Redux 1.0.19.2.

https://drive.google.com/open?id=0B-lHX_zOo86nbW13REZGdHlIZzg - Doubled mass.

https://drive.google.com/open?id=0B-lHX_zOo86nZGJRcGlQNUlaSHc - Negative mass, but this screenshot from heavily modded version, where it was discovered...

Please anyone help! I can`t play KSP without FAR and Pwings anymore!

P.S. I`m sorry for the grammar mistakes, I used google translator...

Don't trust KER or stock engineer reports. Use right click on PW part to show proper mass. Stock engineer report is bugged as hell, especialy when comes to procedural parts. Reason is that stock engineer report use info about mass from editor pool, rather than part attached to craft. It misscalculate even stock wings too if you alter strength/mass on wings.

For KER I can't say much but I think that it also didn't updated properly for KSP 1.0.5. MechJeb can be good alternative for info purposes against KER. You don't have to use autopiloting features of MJ if you don't like them.

[Razum]

1 hour ago, kcs123 said:

For KER I can't say much but I think that it also didn't updated properly for KSP 1.0.5.

In my case KER show same as right click menu. What is most important, the CoM position calculated wrong, accordingly FAR analysis, it seems, was totally wrong too. I have to test this...

In flight, it seems(I not tested much), everything is fine, but without FAR analysis plane construction will be very hard.

1 hour ago, kcs123 said:

It misscalculate even stock wings too if you alter strength/mass on wings

Hmmm... I thought stock engineer reports has been relatively adequate in this case... I never noticed problems with stock wings before...

 

 

Edited January 23, 2016 by Razum

[kcs123]

Forgot to say, try latest FAR devbuild from github. Latest official FAR release have some bugs that were discovered and solved in developer version. Few posts back there is info how to get it.
It was most obvious fix for data analysis and graph issues, but there is couple of other bugfixes too.

When comes to FAR, it is better to install it by yourself along with modular flight integrator to be sure it is one of numerous CKAN messup with installs.

[tetryds]

@Razum you say that all of your mods but one were installed via CKAN.

This thread does not support CKAN issues, reproduce the issue with a fresh manual install or move the report to the FAR CKAN thread.

[AK von Bismarck]

Question: Does FAR model factor in the individual prop blades in KAX and then figure out the thrust from there, or are the engines assigned some sort of thrust value unrelated to aerodynamics?

[blowfish]

2 minutes ago, AK von Bismarck said:

Question: Does FAR model factor in the individual prop blades in KAX and then figure out the thrust from there, or are the engines assigned some sort of thrust value unrelated to aerodynamics?

FAR doesn't change how engines behave at all.

[ss8913]

So.. installed dev build of FAR without mechjebfarext.dll - works great and it *fixes the rendezvous overheating problem 100%*.  Whatever you did in this dev build has definitely solved it and I no longer need any ALT-F12 cheats to dock without explosions   Hopefully it makes its way into an official build soon.

[Shadowmage]

Hi all, had a few questions regarding inter-mod compatibility.

First -- what is the proper way to trigger a re-voxelization of a part/the craft?  I have many 'modular' parts in my mod, where their models may change substantially compared to the prefab part;  others have plugin-driven animations that I would like to make sure are calculated properly.  From my brief time playing with it, most seemed to function appropriately, but there were a few problematic parts.

Second -- how do you properly make a command-pod stable for a heat-shield-first re-entry with FAR?  I have several single-part command pods (with integrated heat shield and parachutes) that work fine in stock, but in FAR they want to stabilize nose-first.  I tried re-exporting the part with the COM moved down lower, but the apparent center of lift merely moved down as well.  I also attempted to play around with the part-config CoL / CoP / CoM / centerOfBuoyancy / centerOfPressure fields, but none seemed to have any effect.  Is there a specific config setting / FAR module patch / geometry setup that needs to be done?  (I also tried the part after removing all other modules, to make sure it wasn't some cross-module conflict; there was zero change to its behaviour after the modules were removed).

Third -- third-party parachute support?  I have a custom parachute module, which works fine in stock, but in FAR they do... nothing.  Likely related to my parachute module using the stock drag-cube system.  A brief look over the included RealChutesLite implementation showed that instead of using drag cubes, it actively adds drag as a physical force on every tick.  Is this method compatible with stock as well? (I'm fine with changing implementation, as long as it will work in both FAR and stock (even if it requires different values to be patched in)).

Thanks in advance for any help and/or information,

Shadowmage

 

[blowfish]

I can answer one of those...

29 minutes ago, Shadowmage said:

First -- what is the proper way to trigger a re-voxelization of a part/the craft?  I have many 'modular' parts in my mod, where their models may change substantially compared to the prefab part;  others have plugin-driven animations that I would like to make sure are calculated properly.  From my brief time playing with it, most seemed to function appropriately, but there were a few problematic parts.

It's this part message.  I have it checking for FAR before sending it, but I don't think that's strictly necessary.

[Shadowmage]

1 hour ago, blowfish said:

I can answer one of those...

It's this part message.  I have it checking for FAR before sending it, but I don't think that's strictly necessary.

Thanks for the info    I probably would have never found that while looking through the FAR source code.  That is one problem solved.

 

Command Pods -- I'll post some pics/images that explain this a bit better after I get home from work tonight.  It seemed really strange that even when I put the COM below the pod geometry that the COL would follow it (but always remained below the COM).  Stock pods have the COL above the COM, which gives them a bottom-down stable orientation in uncontrolled flight.  Stranger yet, -some- of my pods work properly, but others do not;  I have a pod with two versions, one with heat-shield geometry and one without; the one with heat-shield is messed up, while the one without heat-shield has the COM/COL set properly.

 

Edited January 25, 2016 by Shadowmage

[kcs123]

43 minutes ago, Shadowmage said:

Thanks for the info    I probably would have never found that while looking through the FAR source code.  That is one problem solved.

 

Command Pods -- I'll post some pics/images that explain this a bit better after I get home from work tonight.  It seemed really strange that even when I put the COM below the pod geometry that the COL would follow it (but always remained below the COM).  Stock pods have the COL above the COM, which gives them a bottom-down stable orientation in uncontrolled flight.  Stranger yet, -some- of my pods work properly, but others do not;  I have a pod with two versions, one with heat-shield geometry and one without; the one with heat-shield is messed up, while the one without heat-shield has the COM/COL set properly.

 

Just guessing here, but thing of influence is only COM. Center of lift may work only for stock parts. FAR calculates his own COL based on mesh trough voxelization process. Other reason why it does not behave as expected could be wrong mesh collision. Try to take a screenshoot with transonic curves visible in editor. Also screenshoot with showed voxels in editor might help to detect faulty collision. Hopefully, with those you might get better answer what is going on.