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

[ferram4]

@AccidentalDisassembly: Okay, so the way that's running, there are really only two possibilities: either part is null, or part.Modules is null, and I really need that to be only called if part.Modules isn't null, since that's the only way to resolve a proper drag model for intakes. Anyway, try the latest dev build and see if that works. If it does, then that will be 0.14.4.1.

[AccidentalDisassembly]

@AccidentalDisassembly: Okay, so the way that's running, there are really only two possibilities: either part is null, or part.Modules is null, and I really need that to be only called if part.Modules isn't null, since that's the only way to resolve a proper drag model for intakes. Anyway, try the latest dev build and see if that works. If it does, then that will be 0.14.4.1.

Going to go find that and give it a try -

EDIT: Aaaaaaaand same issue with the version updated about 30 mins ago. =(

Edited November 26, 2014 by AccidentalDisassembly

[thorfinn]

The first point allows pitch-controlling surfaces to deflect some percentage of the current AoA, from -200% to 200%.

Edited November 26, 2014 by thorfinn

[HoneyFox]

I flew in Falcon 4.0, FreeFalcon and FalconBMS, and they are good sims but not 100%, neither is FAR for that matter.

But you have to take into account the max pitch angle of your eleverons on the rear of the craft. If they have a 20deg sweep then they may have more or less then the real F-16. Also you have to look at the speed you are attempting to perform the maneuver at. The F-16 handles best around 350-550kts.

I know my Mirage knockoff handles about like a heavy Mirage, as it is a heavy twin engine mirage. It turns on a dime when doing 400-500kts. It turns inside of the state of Texas when it is going supersonic, and can turn quite well when going less than 400kts but has stall issues.

Here is the reference:

http://www.benchmarksims.org/forum/content.php?149-Flight-Model-(FM)-Developer-s-Notes-Part-4

Check page 4 of the pdf, the graph is for F-16 at low speed. Should be lower than the corner speed of F-16 (corner speed should be somewhere between 350 and 400knots I think)

The max deflect angle of the eleveron is 25 deg according to that material.

[HoneyFox]

With a negative setting, this could make mildly pitch-unstable design viable, right? Does it work on the other axes too? A yaw damper would be really useful, considering that almost all KSP aircraft are at least supersonic. I don't know what the best strategy for tuning would be though.

I can't see the utility of the positive setting, can anyone explain me?

Currently only on pitch axis.

The utility of positive setting is, well...

First, you might need it because sometimes the deflection is on the opposite side as you want, you then need to negate that value. (@ferram4: I guess we need to calculate an "AoALocation" instead of using "PitchLocation" for AoA% because it shouldn't be affected by whether the control surface is in front of or behind the CoM.)

Second, for unstable aircraft, you would need positive setting for the elevator so that it will deflect to increase more local AoA increment than aircraft AoA's increment in order to provide more nose-down moment to create artificial pitch stability.(just like the F-16 I posted) Although that will make the elevator stall earlier and more quickly than the wings of the aircraft.

Edited November 27, 2014 by HoneyFox

[Hodo]

HoneyFox, I am not saying you are wrong but this craft right here can pull 13Gs sustained in FAR+DRE, and is still VERY flyable.

Perhaps the F-16 you are using isn't balanced right, or something is off slightly?

[HoneyFox]

HoneyFox, I am not saying you are wrong but this craft right here can pull 13Gs sustained in FAR+DRE, and is still VERY flyable.

Perhaps the F-16 you are using isn't balanced right, or something is off slightly?

Ah, I'm not arguing. Just want to provide more supplementary references for people that are interested.

For your aircraft, is it an static-unstable one? (Judging from your screenshot it seems not)

if it's not, usually the eleverons won't stall when you are pulling a high AoA. But if it is, eleverons might need to deflect downwards instead of upwards to compensate the increasing nose-up moment due to increasing AoA and thus will stall easily.

[Einarr]

Is there any reason why the FAR windows may refuse to show up in flight for vessels created in the SPH and launched from the Runway, yet work normally for vessels created in the VAB and launched from the Launchpad?

Edit: Seems there may be some conflict with the Stock Bugfixes. I removed the SymmetryActionFix.dll and the problem I was describing has vanished.

Edited November 27, 2014 by Einarr

[icecubecookie]

Hey ferram thanks again for the awesome mod =)

Appreciate the recent update but i have problem to report - since 14.3.2 (maybe earlier?), when i fly from launch, cargo bays closed, my Cd is 0.015, but once when i open it, the drag goes up to 0.045 - thats perfectly normal.

But when i close the doors once again in flight after opening it once, the drag stays at 0.045, speed is confirmed lowered so its not an error in displays, and when i turned on visualisations, i confirmed that the shielded parts are indeed producing drag (glowing yellow) in a closed space (cargo bays are on locked status and are closed).

The cargo bays that are confirmed to be affected are from B9 and the new stocked space plane plus cargo bays for Mk2 size.

Hope this helps to improve this awesome mod =)

And also thank you for your time =D

[thorfinn]

Second, for unstable aircraft, you would need positive setting for the elevator so that it will deflect to increase more local AoA increment than aircraft AoA's increment in order to provide more nose-down moment to create artificial pitch stability.(just like the F-16 I posted) Although that will make the elevator stall earlier and more quickly than the wings of the aircraft.

OK, it's mostly a matter of convention: I would have called that "negative" because the elevator commands pitch down when the AoA goes up. Then the negative setting would be for canards, with your convention.

But, would there be a use for the "make this less stable" setting?

[HoneyFox]

OK, it's mostly a matter of convention: I would have called that "negative" because the elevator commands pitch down when the AoA goes up. Then the negative setting would be for canards, with your convention.

But, would there be a use for the "make this less stable" setting?

I set the flaperons of that F-16 to deflect up a little bit when AoA increases. This might make the entire aircraft a little bit more unstable but these flaperons won't stall as easily as they were so I can have slightly better lift and still have sufficient roll control when at medium-to-high AoA.

Other than that kind of usage, perhaps there's no other.

[ferram4]

@Einarr: I cannot reproduce the issue, and I've been using Stock Bugfix Modules for some time now. I suspect that you're either using an older version of the modules or another mod is the actual source of the conflict.

@icecubecookie: I cannot confirm; all cargo bays show the expected Cd of 0.003 for me. I'll need full reproduction steps and craft files using as few non-stock parts as possible.

[steve_v]

Don't suppose we could have some usage hints in the OP for these new control surface tweakables?

I've been playing with them, but I'm still not entirely sure how to use the AoA% to do anything usefull. Is this auto deflection in response to AoA changes? If so, do I need to set the corresponding control axis to non-zero to get it to work?

It looks handy but apparently I have idea how to use it, since I can't seem to get it to do anything.

Second, for unstable aircraft, you would need positive setting for the elevator so that it will deflect to increase more local AoA increment than aircraft AoA's increment in order to provide more nose-down moment to create artificial pitch stability.(just like the F-16 I posted) Although that will make the elevator stall earlier and more quickly than the wings of the aircraft.

Sorry, I can't make heads nor tails of this.

[icecubecookie]

okok heres how to reproduce the *edit* falsified* drag (will upload pics later)

EDIT : The Cd is a lie! at terminal velocity, bays closed, the plane flying level, once opening the bays the plane immediately slows down, but when closed again, the speed picks up, but the Cd in the flight data and the visualisations are still the same as open bays

so do treat this as a small bug, not a major mod breaking and sorry to have made any form of panic or inconvenience >.<

1. have things in the cargo bays (turning off aerodynamic failure for this would make things easier)

2. take off and look at the flight data, also turn on visualisations, look into the cargo bays and see nothing is glowing

3. open the cargo bay in flight @ 200m/s (higher would make things more pronounced), and look at the flight data and visualisations, especially the stuff inside the bays

5. everything shows to be producing drag - normal as they are no longer shielded

6. close the bays

7. flight data shows no change in Cd (staying at 0.028 for this plane), looking into the plane's cargo bays, everything shielded within the bays are glowing yellow

works for all stock or mod parts

thing to note is that the Cd of the parts in the bays are not changing dynamically like all the other parts, they just stay there

no red or orange messages are in the debug menu.

Edited November 28, 2014 by icecubecookie
added pics, found the problem

[HoneyFox]

Don't suppose we could have some usage hints in the OP for these new control surface tweakables?

I've been playing with them, but I'm still not entirely sure how to use the AoA% to do anything usefull. Is this auto deflection in response to AoA changes? If so, do I need to set the corresponding control axis to non-zero to get it to work?

It looks handy but apparently I have idea how to use it, since I can't seem to get it to do anything.

Sorry, I can't make heads nor tails of this.

You don't need other axis set to non-zero if you simply want a control surface to auto deflect due to AoA.

Two usages for normal aircrafts would be: LeadingEdgeFlap/Slat that can imcrease the critical AoA of the main wing it's installed onto. and a way to make pitching up milder so that you feel smooth when you hold the S key for a period of time during flight.

What I said in your quotation is for unstable aircraft design, which is unusual design in KSP though, so don't care about that too much.

[BudgetHedgehog]

You don't need other axis set to non-zero if you simply want a control surface to auto deflect due to AoA.

Two usages for normal aircrafts would be: LeadingEdgeFlap/Slat that can imcrease the critical AoA of the main wing it's installed onto. and a way to make pitching up milder so that you feel smooth when you hold the S key for a period of time during flight.

You lost me at 'axis' and only got me back at 'smooth'. It's times like this when I wonder if I'm a bit too simple for FAR..

[Wanderfound]

Just to test if I'm understanding it right...a 100% negative AoA setting will attempt to hold the control surface parallel to the airstream regardless of the AoA?

[Starman4308]

Hey Ferram, I'm beginning to work on a contract payout adjuster for RSS, which basically grabs orbital and atmospheric parameters, figures out how much dV certain maneuvers would require, assumes a given Isp, and appropriately adjusts it from stock.

Orbital maneuvers are easy. Ascents and descents, not so much, but I've figured out a general framework. I'll simulate a repeated TWR curve up to the edge of atmosphere (straight vertical), multiply accumulated aero drag by 1.5-2.5, and call it the atmosphere drag for that body. My question is: would the ideal drag equation with a reference cross-sectional area and drag coefficient* be a reasonable representation of FAR/NEAR aerodynamics, particularly as one goes hypersonic (I can approximate out the transonic region by ignoring it)?

*Looking at Wikipedia, its "streamlined object" has a Cd of 0.04; I might go with 0.10 as my default, so as to be generous about the quality of rocket design. It probably doesn't matter a huge deal if I cut out the thrust when drag-adjusted apoapsis hits the edge of atmosphere, or have the reference vehicle thrust all the way to the edge: the second option will probably not add much aero drag.

EDIT: Solved my own problems. Found your Mach effect code, and based on building a smallish rocket, I think I'll go with a cross-sectional area of 5.5 m^2 and a Cd of 0.18 (had one 1.25m stack with three fins, with three 1.25m SRBs on decouplers, all with aerodynamic caps). I'm thinking that the 5.5 m^2 area came out of having the fins, and Cd went down significantly after staging because much less of that 5.5 m^2 area was actually filled up?

Edited November 28, 2014 by Starman4308

[steve_v]

You don't need other axis set to non-zero if you simply want a control surface to auto deflect due to AoA.

Two usages for normal aircrafts would be: LeadingEdgeFlap/Slat that can imcrease the critical AoA of the main wing it's installed onto. and a way to make pitching up milder so that you feel smooth when you hold the S key for a period of time during flight.

What I said in your quotation is for unstable aircraft design, which is unusual design in KSP though, so don't care about that too much.

So if I set, say, an elevator to some postitive AOA% it will deflect *up* as the *aircrafts* AOA increases and the inverse for a negative setting? this sounds extremely usefull

Or is this based on the AoA of the control surface itself?

[HoneyFox]

So if I set, say, an elevator to some postitive AOA% it will deflect *up* as the *aircrafts* AOA increases and the inverse for a negative setting? this sounds extremely usefull

Or is this based on the AoA of the control surface itself?

It's of course based on aircraft's AoA (which is shown in FAR's flight data window as well), not control surface's local AoA. Talking about whether the AoA% should be positive or negative, well you can always change the sign if you find that it's deflecting to the opposite side.

[HoneyFox]

Just to test if I'm understanding it right...a 100% negative AoA setting will attempt to hold the control surface parallel to the airstream regardless of the AoA?

If the control surface is installed on a zero-sweep-angle wing, it should be. (thought I'm not completely sure whether it will always deflect to the correct direction so perhaps in some cases you would need to set +100% instead of -100% for AoA%)

[steve_v]

Awesome, this should make my marginally stable aircraft a bit easier to fly

[Jovus]

Minor potential bug report with a bit of the new UI: it seems impossible to select anything below Custom07 for default keys for spoilers or flaps. When you move that far down, the window loses focus and the menu disappears.

I've tested this both in the default space centre scene and in the contract building scene (due to a misclick). It seems pretty easy to reproduce, and I'm using a fresh install (albeit with a lot of other mods). If this were at all a complicated workflow I'd pin down the bug further before reporting, but considering how simple it is to check reproducibility I figured it wasn't too much of a sin to report first.

Ferram4, if you have trouble reproducing it, I'll dig further.

[Iskierka]

My input: I feel like +100% should mean that control surfaces' zero-input line is at 0 AoA on the control, as that is the most common actual set-up for high AoA or high-manoeuvrability aircraft. This is so that the control surfaces will never stall and retain similar authority over the entire flight envelope. It's notably more common for them to track AoA like that than attempt to fight it, though the latter does exist.

But, that's just what my brain instinctively says, I've not yet had chance to test this new feature. (Which will hopefully be useful regardless of whether its numbers make sense, for the aforementioned reason of retaining authority at high AoA.)

Edited November 28, 2014 by Iskierka

[TheReaper]

i cant really see my center of lift in hanger bay, its just a small blue ball, smaller then usual, usualy its at the same size with center of mass, but its just a small blue ball with no arrows attached to anything, is this normal?

Edited November 28, 2014 by TheReaper