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

[Chirality]

Can you post a picture of the Ugly Duckling lifting body craft?

[ferram4]

@DaMichel: I don't know, the GUI doesn't seem to be clear with it set up that way.

I think there is a bug somewhere in the CoL code, but I haven't been able to track it down. I'll see what I can do.

@jrandom: If camlost is talking about the book I think he's talking about, that's one of the ones I have lying around. It's got the limitations of an introductory book (not very in-depth) but is otherwise quite good.

@Bouncingbunny: From earlier in the thread:

You could just load it in-game and see the horror for yourself.

[Andon]

This is what happens when you mix playing Darts with airplane designs...

[jrandom]

@jrandom: If camlost is talking about the book I think he's talking about, that's one of the ones I have lying around. It's got the limitations of an introductory book (not very in-depth) but is otherwise quite good.

Introductory seems like a good place to start. I've got a small collection on the way, and "Aircraft Dynamics: From Modeling to Simulation" just arrived at my doorstep today.

Edit: oh wow, I have completely forgotten how to do integrals. Crap. Now where'd I stow that calc book...?

Edited December 31, 2013 by jrandom

[DaMichel]

@DaMichel: I don't know, the GUI doesn't seem to be clear with it set up that way.

I agree but i figured the convenience would make up for it

Another question ... Does FAR use any of the stock aerodynamic parameters?

Edit:

//These parts don't have their attach nodes at the correct size, so drag isn't calculated correctly on them

@PART[solidBooster1-1]

{

@node_stack_bottom = 0.0, -3.914617, 0.0, 0.0, 1.0, 0.0, 1

@node_stack_top = 0.0, 3.939497, 0.0, 0.0, 1.0, 0.0, 1

}

...

What does correct size mean?

Edited December 31, 2013 by DaMichel

[ferram4]

@jrandom: Wolfram Alpha is your friend. Trust me, the number of ugly integrals that can't be solved easily that will pop up in aerodynamics is amazing; you'll just have to bite the bullet and do a lot of it numerically.

@DaMichel: FAR only leaves the parachutes to do their stock aerodynamic properties. Everything else is cleared out.

The attach nodes are supposed to be set up so that the last number controls the size of the node, which is used by FAR to figure out very, very sudden changes in cross-sectional area. It's what's used to determine most of the drag and the lift of a reentering command pod. There were a few stock parts that didn't have the last number set or had it set to the wrong value, causing weird things to happen.

[jrandom]

@jrandom: Wolfram Alpha is your friend. Trust me, the number of ugly integrals that can't be solved easily that will pop up in aerodynamics is amazing; you'll just have to bite the bullet and do a lot of it numerically.

I don't doubt it. But at the very least I need to understand why they're used where they're used. Once I understand, I can let machines do all the hard grunt-work.

[DaMichel]

Ah, thanks. I was debugging a issue with the drag on a part which had the intake transform oriented wrong.

[SnappingTurtle]

Would it be possible for this mod to make the compression heating effects more indicative of the energy lost to the environment?

[taniwha]

SnappingTurtle: Use DRE for more realistic heating effect indication. The intensity of glow is purely a function of temperature, and that is, surprisingly, pretty much temperature (K) = velocity (m/s). The energy loss does not go so much into heating the air, but rather into accelerating it, and that acceleration causes the compression. The heating is from the compression (try playing with a bicycle pump: it's possible to get minor burns). Thus, really, most of the energy becomes sound (eg, the sonic boom) rather than heat.

[LoSboccacc]

question: why stock sas wobble so much with far installed?

[ferram4]

Because SAS is not designed to deal with lots of aerodynamic control authority.

Try upgrading to v0.12.4; I've increased the mechanical lag of the control surfaces to be closer to what stock has. People have reported that it reduces the wobbling, but I'm not exactly sure why, since it should cause more control problems. Nevertheless, upgrade and see if that fixes it. If it doesn't, get rid of some roll control surfaces.

[LoSboccacc]

thanks that would explain why it was wobblier since updating. removing some surfaces and adding more sas worked.

edit I am quite dull this evening what are the flaps key now? can't find them in changelog and 9/0 doesn't work for me

[ferram4]

You have to set them manually now using action groups.

[LoSboccacc]

tried that and can't select the control surfaces from the action group view

[ferram4]

What? You should be able to click on them just like any other part and set the action groups for them. I just tried it in my install and it worked fine. You're not getting any errors in the output_log, correct?

[Surefoot]

I usually turn off SAS when flying in atmo, it cannot handle properly whatever is happening with FAR. I think it still does with 12.4, i'll have to try again later.

[LoSboccacc]

removed and reinstalled FAR and module manager now it works thanks

[jrandom]

Any news on if the supersonic roll-twitch is fixable? That's about the only thing standing in my way of long-term sustained supersonic flight. (I did manage a hands-off 8-hour subsonic flight that got almost a quarter of the way around RSS Kerbin before KSP decided to crash. That was pretty neat to watch -- I'd check in every once in awhile to see how it was doing.)

[ferram4]

I'm not entirely sure where it comes from to be honest. Since you're apparently really good at causing it, can you test two cases for me? The first one is to just fly along between Mach 0.8 and Mach 1.2 (closer to Mach 1 the better) and see if the twitch occurs; that will just see if it happens at transonic speeds as well. The second one is to try and keep a wing at 2.8 or -2.8 degrees angle of attack at a few different Mach numbers; that will test if the error is in the switch between a shock on the upper surface and an expansion fan on the upper surface of the airfoil, right behind the leading edge, like so.

Otherwise I'm not sure what could be causing it to happen.

[jrandom]

You bet. I have a little aircraft I made today out of B9 and stock parts that seems supremely susceptible. I'll go fly it now at those conditions.

[jrandom]

Mach 0.8 - 1.2 at AoA of 0.3 degrees: No roll twitches observed.

Mach 0.8 - 1.08 at AoA of 2.8 degrees: No roll twitches observed.

Mach 1.5-4.8 at AoA of 1.0-3.3 degrees: Roll twitches observed, usually somewhere between 2.5 and 3.0 AoA. (Aircraft is sensitive to pitch, but I'm trying to pause as soon as a twitch occurs so I can read off the AoA.) A few occurred around AoA of 2.2. Severity increases as the mach number increases.

[ferram4]

Okay, it sounds like the leading edge shock / expansion fan is where things are freaking out. I'll see if I can fix it.

[jrandom]

What setting in the .xml config file would I need to change (and to what value) to get back to the previous control surface change/speed setting? I think I preferred that over the new slower one.

[AbeS]

Modify this in the config:

<string name="ctrlSurfTimeConstant">0.25</string>

I THINK it used to be 0.05, not really sure, test a bit