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

[Romby]

Hey.

Im thinking about installing FAR in my current carrer save. Have been playing a little already and want to know what FAR does to my current rockets design. I haven’t made any planes yet and want to install FAR before doing so. Designed all my rockets as realistic as possible (slim, only boosters hanging of the side and always with nosecones). I remember that before version 1.0 you needed less Delta-V to get rocket into orbit because of less drag when using FAR compared to stock (would reverse this by using the KIDS mod).

How is it looking now after version 1.0 with the new stock aerodynamic s? How big is the difference in delta-v req to orbit/drag for rockets (vertical takeoff) between stock and FAR?

Asking because I don't what my current rockets to change to much. Keep using the same lifters over and over again.

[cantab]

I haven't really tested, but in general I expect a rocket that works well in newstock to also work well in FAR. The reverse may not be the case though - a rocket reliant on FAR's whole-vessel approach to drag could have problems in newstock.

[sagirfahmid3]

I noticed the flaps' controls are reversed. Is this a bug or do I have to manually set which direction control surfaces pivot to?

For example, when I press the "S" key, the plane wants to do a nosedive instead of going up, and vice versa. When I press the "A" button, the plane rolls to the right instead of left, and vice versa...

[Wanderfound]

Hey.

Im thinking about installing FAR in my current carrer save. Have been playing a little already and want to know what FAR does to my current rockets design. I haven’t made any planes yet and want to install FAR before doing so. Designed all my rockets as realistic as possible (slim, only boosters hanging of the side and always with nosecones). I remember that before version 1.0 you needed less Delta-V to get rocket into orbit because of less drag when using FAR compared to stock (would reverse this by using the KIDS mod).

How is it looking now after version 1.0 with the new stock aerodynamic s? How big is the difference in delta-v req to orbit/drag for rockets (vertical takeoff) between stock and FAR?

Asking because I don't what my current rockets to change to much. Keep using the same lifters over and over again.

FAR rocketry isn't too tricky, but if you do run into trouble, it's more likely to be related to stability than ÃŽâ€V. So: gimbal. The more the better.

If you find that your established lifters have stability issues, slapping on a few Thuds or equivalent has a good chance of providing a quick solution. It's possible to build successful FAR rockets without either gimbal or control surfaces, but it requires a fair bit of finesse.

With enough vectored thrust, you can launch quite ungainly things without much trouble:

[Svm420]

What are real life gimbal ranges like?

[Wanderfound]

What are real life gimbal ranges like?

10.5° on the Shuttle's SSMEs. That's at the upper end of the range, although the high-gimbal verniers used on some rockets probably exceed that.

[BudgetHedgehog]

I noticed the flaps' controls are reversed. Is this a bug or do I have to manually set which direction control surfaces pivot to?

For example, when I press the "S" key, the plane wants to do a nosedive instead of going up, and vice versa. When I press the "A" button, the plane rolls to the right instead of left, and vice versa...

Just a a thought - you haven't set the surfaces to deflect -100%, have you? 0 is the middle of the slider, not the left.

[Vegatoxi]

My fifth try to deal with FAR failed again...

I mean, really? Build design prototype which fly perfectly in real life, in FAR it flies like a brick and even worse...

Also it fly almost perfectly in stock AD.

Javascript is disabled. View full album

Edited October 8, 2015 by Vegatoxi

[tetryds]

@Vegatoxi: that's a very common misconception, just because it "seems" like it would be good doesn't mean that it is.

FAR is more about balancing than the design itself, and all you have to do is to balance them right.

Check stability derivatives, your static curves and fine tune your control surfaces, it's certainly going to work.

Also, how come it flies irl, did you make an RC version of it?

Anyway, remember that KSP parts have different weight and shape if compared to most rl craft so looking similar is far from enough.

Edit: just a detail, in real life modern fighters are often unstable, even if you make a perfect replica you will have to balance some stuff.

Edited October 8, 2015 by tetryds

[Van Disaster]

This is probably mostly a question about aerodynamics in general and me being unfamiliar with the current FAR, but for a given AoA does the centre of lift move depending on airspeed ( or rather dynamic pressure )?

I have this little guy:

Who's quite happily stable pulling 20 degrees AoA in a roll at 40k kPa, but try the same thing at 15kPa and it'll go off into that wierd reverse dutch roll that forward sweep craft do. Before I go and spray it with vertical surfaces ( it has yaw brakes already, they're not terribly effective ) am I trying to fix the wrong thing, is it just getting unstable because CoL is moving forwards?

@Vegatoxi - don't forget to pull the gear up when you're using the analysis tabs. Transonic stuff doesn't really matter that much until you're nearing critical mach - so not at all for a prop plane - best to sort basic stability out first, the second tab is your best friend most of the time. One day I'll learn how to use the simulation tab...

You can't build anything in KSP that's *remotely* close enough to a real craft to be a replica aerodynamically, you just don't have the parts or the design tools. You'd need X-plane and some good blueprints, and even then it'd only vaguely be a replica when it comes to actual flight.

Edited October 8, 2015 by Van Disaster

[tetryds]

@VanDisaster: yep, check your Mw for different Mach speeds.

It usually moves backwards as mach speed gets higher, but while lift moves back, the drag starts acting more on the front part.

Transonic also doesn't mean that much after you pass the transonic region.

The simulation tab is for when everything seems fine but is not, I have never used it for real.

But replicas should behave close to their counterparts if the most important parameters are close, you cannot get it to behave "exactly" like the real life version but you would be surprised at how good FAR is.

The trick is getting those to be the same, we are limited on shapes and weights don't always match.

[capi3101]

I've used the simulation tab for reals my own self. It uses the principles described on this web page regarding aircraft stability. The sim tab is for testing dynamic stability (whereas the derivatives tab tests static stability), and I have had instances where I had green static numbers across the board and still wound up with a divergent oscillation on the sim. Doesn't take much to use the sim tab - just grab the derivative values for the speed and altitude for which you're having problems, then switch over and start plugging in numbers into the sim; a value of 1 is usually sufficient to test a given axis, and I usually check the axes one value at a time. If you see a divergent pattern, that's something that needs to be fixed. ß and W are the ones you really need to keep your eyeballs on, in my experience.

[Van Disaster]

I can generally see dynamic issues just by flying the plane - plus if you plug a range of mach + alt numbers in the static tab you'll probably pick up instability anyway. Also I don't usually have issues flying in straight lines - not sure banking the plane in the hangar & rechecking derivatives is going to help much with a turn issue, but then the gravity vector can be all over the place anyway.

@tetrydis: it's not FAR that's the issue, it's KSP parts

[ZenoEval]

Why is it that everything I try to land after undocking just slams through the atmosphere with no slowing down? If I quicksave/reload it works fine, but forgetting sucks, lol.

[ferram4]

@Vegatoxi: Well, part of it might be that you installed FAR incorrectly or that you're using mod parts that still make use of the stock aerodynamics. There should be no arrow on the CoL with FAR installed, and you have one, so you've broken something horribly.

@Van Disaster: Mach number is more likely to have an effect than dynamic pressure. Possibly what's happening at lower speeds is that you've got more inertia relative to aerodynamic forces so you're reaching a higher AoA and getting some stalling on one of the control surfaces.

@ZenoEval: That was an old bug that was fixed awhile ago. I have been unable to reproduce those results in my install regardless of what I try, so unless you can provide a simple, FAR-only-with-no-other-mods save that can reproduce the issue on the most recent version of FAR, I'm going to consider this to be either already fixed or an issue that cannot be reproduced.

[ZenoEval]

Okay, thanks. Using the newest version, so it must not work with other mods. I guess I'll just deal with remembering to save. Great mod by the way!

[Van Disaster]

Just want to confirm observations: in old versions stalling a wing panel would stall the ones behind it - does stalling one at the back ( a control surface, say ) stall the panel in *front* now? I've suspicions it's mostly changing the aoa of the panel the surface is attached to but I'm watching this wing light up red from the back in sequence as it stalls.

I can understand flow seperation starting at a leading edge, not quite sure about the other way - some sort of local area of high pressure in front of the control surface?

Edited October 9, 2015 by Van Disaster

[baldamundo]

What's the performance hit from FAR like these days?

[ferram4]

@ZenoEval: Alright, good to know the issue (if it's even in FAR) isn't getting fixed except by luck then.

@Van Disaster: It depends. Trailing edge separation is a thing that can occur, and it's generally the preferred way of stalling a wing, since it tends to be more gentle than leading edge separation. But no, a wing part stalling will not stall the parts in front of it.

@baldamundo: It's there, but no, it is very unlikely to have anything to do with the issue you're looking at in support. FAR does not produce tons of garbage every few seconds unless it is commanded to recreate all of its aerodynamic data on a regular basis by other mods.

[baldamundo]

Oh no, that's not why I ask - I was just considering installing FAR (after increasingly realising that 1.0.4's physics are apparently still kind of atrocious), but weighing up how much more FPS I can afford to lose

[m4ti140]

How do you guys deal with Dutch roll using FAR built-in yaw damper? I built an airliner with Mk3+B9 wings and I keep getting into an irrecoverable dutch roll at around 7500-8000m (I'm playing with 64k) and I can't find the proper parameters for the yaw damper to stop this. Is there an aerodynamic solution that could reduce the risk of Dutch roll and works in FAR?

[CrisK]

Ferram, the latest FAR release has been incredibly stable. I think that the current (Github) version is the most stable FAR has ever been.

[capi3101]

How do you guys deal with Dutch roll using FAR built-in yaw damper? I built an airliner with Mk3+B9 wings and I keep getting into an irrecoverable dutch roll at around 7500-8000m (I'm playing with 64k) and I can't find the proper parameters for the yaw damper to stop this. Is there an aerodynamic solution that could reduce the risk of Dutch roll and works in FAR?

The page I linked yesterday deals with Dutch Roll a good deal. Looking at the design history of the two RL planes specifically mentioned on that page that had notably bad Dutch Roll tendencies - the Boeing 707 and the KC-135 - would give you some clues as to how to fix the problem design-wise; the long and the short of it boils down to increasing yaw stability (the ß parameter) with MOAR FIN (the fin surface area I usually go for is 10% the area of the main wing, if you need a working guideline). And then that page I linked will tell you how an RL pilot would handle the Dutch Roll on your existing design (it's down at the bottom).

Edited October 9, 2015 by capi3101

[Van Disaster]

How do you guys deal with Dutch roll using FAR built-in yaw damper? I built an airliner with Mk3+B9 wings and I keep getting into an irrecoverable dutch roll at around 7500-8000m (I'm playing with 64k) and I can't find the proper parameters for the yaw damper to stop this. Is there an aerodynamic solution that could reduce the risk of Dutch roll and works in FAR?

Bigger vertical tail iirc or maybe less dihedral, I think it's caused by having more roll stability than yaw. You could perhaps look at reducing the roll part rather than the yaw component. Also maybe winglets ( little wingtip vertical stabs ) which will help reduce the sideslip.

Edited October 9, 2015 by Van Disaster

[baldamundo]

Ferram, the latest FAR release has been incredibly stable. I think that the current (Github) version is the most stable FAR has ever been.

Honestly, it's the FPS more than the stability I'm concerned about! Game's already a bit slowed from all the other mods, just wondering if it's worth having FAR on top as well