Wings over Korea all-in-one pak?

[Do335 382]

Reference area applies only to the wings, as per TK.

The question then is how are forces/moments calculated on other components, if that is the case:|

 

Moreover how does the FM engine tell if it is the "wing", vs "stab", "verttail", etc, afterall it is all just char type variable to the code, these component names?...

 

 

 

 

If your plane is departing too soon, remember that actual pitching moment is your Cm * ReferenceArea * ReferenceChord... or, in case of all-moving stabiliator, CL * Xac (aerodynamic center position) * ReferenceArea * ReferenceChord. So if you changed any of the refence geometry, your aircraft will fly differently unless the coefficients, or Xac position of the tail, are also adjusted.

http://bbs.thirdwire.com/phpBB/viewtopic.php?f=14&t=5658&p=34556&hilit=referencearea#p34556

Edited April 17, 2016 by Do335

[+baffmeister 1,205]

That's an interesting and , for me, complicated link. For your situation with the 6-3 wing, shouldn't the change in behavior be welcome if it reflects the real aircraft behavior? Not sure exactly what the results would be but if I remember correctly the addition of the 6-3 wing did aggravate the pitch up behavior during take off. Assuming you did increase the reference area to represent the 6-3 wing area I think it would be appropriate to move the XAC position slightly forward as well which should aggravate the pitch up.

 

I was wondering about how the game does the calculations for the vertical tail and horizontal stabs as well. I think for some components it uses actual model geometry in the calculations. Maybe the bounding boxes?

 

You do dig up some good old threads, Do335. Will try to digest that one tomorrow.

[Do335 382]

BM: yes indeed for the real aircraft...have done some deep google-fu and seems a few issues with the 6-3 on the real thing as well. Funnily apparently the real guys changed the fuel sequencing on the fuselage fuel cells (aft cell empties first) as an attempt to move the CG forward a bit......

 

 

For bounding boxes it is for a while I suspected, TK use them for [system] hitbox damage modelling like ailerons etc. But then there are cases when a component doesn't even have a node name. Haven been wondering for months of how it was and at last realized it's occam's razor as one Referencearea fits all. It then all fell into place. Simple example i guess CLa of [left/rightstab], they're all scaled down to fit the global ReferenceArea.

 

 

 

 

edit: Wait.... I think you knew this one BM? In your FMs you usually note CD0 total = xxx and stuff. CD0 of all parts summed together = CD0 of entire AC so all components would use the same cross section there. I think i used it as evidence to confirm my understanding.

 

but while I reckon i've come to good terms of game engine in these parts, my knowledge of real aerodynamics is meh... probably a good idea to leave real aero there to you guys... they are certainly complex!

Edited April 17, 2016 by Do335

[Fubar512 1,350]

The question then is how are forces/moments calculated on other components, if that is the case:|

 

Moreover how does the FM engine tell if it is the "wing", vs "stab", "verttail", etc, afterall it is all just char type variable to the code, these component names?...

 

 

 

 

http://bbs.thirdwire.com/phpBB/viewtopic.php?f=14&t=5658&p=34556&hilit=referencearea#p34556

From the positions of the Xac tables, for one.  Other examples; "vertail" is not a lifting surface, and therefore has no Xac tables associated with it, and wings don't have downwash tables associated with them, (etc.).

[Do335 382]

Fubar: as forces go they would all need a section number. The fuselage may not be LiftSurface for example, but still generate up/down/left/right forces. So they would need a section value there??.... Not to mention stabs as they indeed are Liftsurfaces....

Xac is lift center and determine location of the force --> then produces moment no? How does Xac come into play with section value here? You are not making sense....

Edited April 17, 2016 by Do335

[Fubar512 1,350]

The Xac used by the series is not based on the overall aerodynamic center of a given flight model.  It is instead based on the aerodynamic center of each surface.

 

 Xac for each surface is determined along a line that runs fore and aft at the chord of each lifting surface.  Generally, for simplicity's sake, we start the 'Xac Progression" at 25% of the distance back from the leading edge along the chord line, and progress back to approximately 70%, at max value.

 

Experiment with this, and see what happens.

[Do335 382]

The Xac used by the series is not based on the overall aerodynamic center of a given flight model.  It is instead based on the aerodynamic center of each surface.

 

 Xac for each surface is determined along a line that runs fore and aft at the chord of each lifting surface.  Generally, for simplicity's sake, we start the 'Xac Progression" at 25% of the distance back from the leading edge along the chord line, and progress back to approximately 70%, at max value.

 

Experiment with this, and see what happens.

eh good to know Fubar. I had only known the aero center usually bases around 25% mean aero chord and goes from there. As said my aerodynamics knowledge is crappy:X I think you and BM have good stuff to talk about!

 

 

 

 

 

oh PS! mac length I've been searching for a long time on the net.

F-86A/E (early slat wing):

MAC 97.03 inch @ wing station 98.71 inch

 

F-86F (6-3 early)

MAC 101.94 inch @ wing station 103.1 inch

 

F-86F (6-3 with 12in tip ext. and slats)

MAC 100.66 inch ...wing station i didn't bother as not present in korea..

 

and Stabs

MAC 34.71 inch @ h. tail station 33.54 inch

Edited April 17, 2016 by Do335

[+baffmeister 1,205]

BM: yes indeed for the real aircraft...have done some deep google-fu and seems a few issues with the 6-3 on the real thing as well. Funnily apparently the real guys changed the fuel sequencing on the fuselage fuel cells (aft cell empties first) as an attempt to move the CG forward a bit......

 

 

For bounding boxes it is for a while I suspected, TK use them for [system] hitbox damage modelling like ailerons etc. But then there are cases when a component doesn't even have a node name. Haven been wondering for months of how it was and at last realized it's occam's razor as one Referencearea fits all. It then all fell into place. Simple example i guess CLa of [left/rightstab], they're all scaled down to fit the global ReferenceArea.

 

 

 

 

edit: Wait.... I think you knew this one BM? In your FMs you usually note CD0 total = xxx and stuff. CD0 of all parts summed together = CD0 of entire AC so all components would use the same cross section there. I think i used it as evidence to confirm my understanding.

 

but while I reckon i've come to good terms of game engine in these parts, my knowledge of real aerodynamics is meh... probably a good idea to leave real aero there to you guys... they are certainly complex!

 

 

My approach to doing FMs is quite simplified. If an FM expert took a detailed look at one of mine I think they might fall out of their chair laughing.   I make no apologies though, the main reason I started doing them was because of the extreme over performance I was finding in some of the add on planes. Also, a lot of them flew in a very strange fashion. Given limited knowledge, the safest approach to take was to try and find the best fitting ThirdWire tables based on wing thickness, wing position, sweep angle etc. and just use those. Likewise, I would usually just try to find a TW plane with a similar tail plane set up and graft that on as a starting point then adjust as required to get the desired behavior. 

 

I do try to calculate the lift coefficients and  CDL tables for the planes with cambered airfoils as well as setting up the XAC points for wings and tails. For the prop planes, a lot of effort goes into research on the engine performance at various altitudes. ThirdWires flight engine seems to be quite a good wind tunnel. I find if I put together a good engine table and make a decent CDL table the resulting speeds end up very close to book. If I can't find actual CD0 numbers for the aircraft in question I start with a best guess and then adjust the headline CD0 numbers as required to get the desired speeds at altitude. In that case, the CD0 values are a resultant, and I do a sanity check by comparing the numbers to some reference values I have for similar aircraft. This acts as a cross check against the engine table and CDL table as well. I also check the low speed turn times against some info I have for some WW2 airplanes but sometimes I have to use "anectodal information" and do it on a comparative basis.

 

An interesting situation I came across when doing the Grumman Tigercat FM. The ThirdWire "wind tunnel" was working very well up to a certain altitude but the maximum speed at altitude was less than book. The reason was I didn't model any ram air boost into the engine table. Most engine data available comes from tests in altitude chambers so the amount of ram air boost above full throttle height isn't noted. In general, I find to get the book value maximum speed at altitude I have to extend the full throttle height upwards by 2000 to 3000ft.

 

When doing FMs for subsonic jets I just use one of the ThirdWire engine tables which are a bit generic, not type specific. I make a specific CDL table for the cambered airfoils and adjust the CD0 values to get the proper maximum speed at sea level only. The resultant CD0 value acts as a sanity check similar to the prop planes. Both the subsonic jets and higher speed prop planes get a generic "mach tuck effect" that I can tune reasonably well to a specific mach number so that something bad happens if you exceed the maximum permitted speed. The mach tuck effect is generated by reducing lift and increasing drag in the wing mach tables. I don't mind messing around with things like that if it's outside of the approved operating envelope for the aircraft. I think it makes things a bit more interesting for the player.

 

Something I've noticed recently while doing a couple of biplane FMs. ThirdWires wind tunnel is malfunctioning!. If i use what I think are realistic CD0 values on some late 1930's biplanes [probably around 0.038] while keeping gear drag at zero, the planes end up being too slow. Not sure whats going on but I had to fudge the numbers to hit the book value speeds. I noticed some similar issues, although not so bad, on a draggy monoplane, probably the Boulton Paul Defiant.

 

Regarding the wind tunnel, the only other issue I've seen is getting some of the late WW2 high performance monoplanes up to speed. I don't think its a problem with the wind tunnel though, it seems more likely related to the CDL table. For the cambered airfoils I just make 4deg tables but when doing that you can end up with no CDL = 0 point. To solve that issue with the Seafury I just hacked the table and put in a CDL=0 at zero deg alpha. A bit ugly but it fixed the problem. Likewise, for ThirdWires under performing P-51, I think I did a similar hack to get it up to speed. [among other things.] Probably, for the high performance WW2 planes, the best solution would be to make a 1deg CDL table and ensure a CDL=0 point. The high performance prop planes seem to be quite sensitive in that regard.

 

Anyway, when I started doing FMs the plan was "learn as you go." That approach worked to a certain extent but it's probably exhausted at this point. Any improvements will probably require some formal study on my part.

[Do335 382]

Well keep up the good work bm. I occasionally check out Il2bos foras and take in some of the FM complaining and thought thankfully TW FMs are completely open, we're not as pro but we make it as seen fit, it's like Ivan Kozhedub said if you want something done, better do it yerself.....:D

Something about TW CDL tables i found they are Y=b(X-a)^2 tables. If the component has Cl0 then 'a' would be none zero. So it's not hard to convert 4 deg tbl to 1 deg tbls and such. Also probably just formatting this one, all the mach table data, the value at mach 0.4 is always 1.0 so TK probably normalizes it that way. I dunno if it has any significance but interesting.
About the mach tuck, i think it is a real problem.(?) It would be great if TW stock FMs all simulated it, it would make it easier to model sabre's 6-3 wing effect but it was not to be..
I think yes learning by doing can be quite time consuming... It is ok if you are just doing a few during a couple years time but more would mean not very time efficient, as i've found out myself it takes a lotta time to test stuff out if i don't have an academic basis for the numbers i'm using. So... depends on what one likes, i like campaigns and while FMs are integral part of it, dedicated aerodynamics expert is certainly beyond me... But if u have the capacity to go further, it certainly is extremely cool to say the least!

Edited April 19, 2016 by Do335

[Do335 382]

A small Mig-15 cockpit avionics revision by Coupi and me (original is ofc Stary). Warning: Backup original files!
 
*ADI is changed to soviet style "viewed from ground", similar to modern Russian jet HUD's attitude indicator (demo image by coupi). ADI pitch degrees is measured.
*Gunsight along with cockpit viewpoint is moved to the right so the ADI is not obstructed.
*Gunsight reticle pitched down 0.3x degrees to converge with rounds ballistics at ~250meters.
*Gun indicator lights on the left hand canopy side are illuminated when the cannons have rounds; extinguished when the guns are dry.
*The 2 red lights on the lower console is used as external fuel tank indicators. The lights are illuminated when external fuel tanks are dry or jettisoned.
*Fuel gauge measured, max volume indicated 1050 liters.
*Low fuel light illuminates when fuel quantity is below 300 liters. Low altitude light fixed. Bearing marker fixed.
*EGT gauge measured and functional. Oil pressure small gauge measured and functional.
*After repeated no joy in finding a suitable mach number indicator... a pseudo mach indicator is setup using the fuel pressure gauge. sigh... but fuel pressure is non-existent in SF anyway.
*Cleaning of redundant entries.

 

and update

*VVI reconfigured with 0 ~ +-70m/s and measured.

*Accelerometer measured.

 
To use the oil press and EGT gauge, add the following to the Mig-15/15Bis_data.ini's [engine] section

GasTempMaxRPM=690.0
GasTempIdleRPM=450.0
GasTempChangeRate=5.0
OverheatTemp=790.0
DamageTempDelta=300.0
OilPressMaxRPM=35.0
OilPressIdleRPM=10.0
OilPressChangeRate=0.3
LowOilPress=2.0
LowOilTempDelta=400.0

Edited April 24, 2016 by Do335

[Do335 382]

More edits to the pit... Used up every needle except the volts gauge. Liking the result! Won't bother UL the bits again, gonna put in the final mod.
 

 
1. G meter
2. Pressure Altitude
3. Airspeed gauge
4. Altimeter
5. Radar altimeter
6. ADI
7. Low altitude light
8. Slip indicator
9. Clock
10. External fuel tank lights
11. Low fuel warning light
12. VVI
13. Engine fire light
14. Fuel gauge
15. Engine overheat light
16. Bearing marker
17. Compass
18. Fuel flow gauge
19. Oil pressure gauge
20. Engine pressure ratio gauge
21. Engine RPM gauge
22. Exhaust gas temp gauge
23. Mach indicator
24. Cockpit pressure differential gauge

25. Gun indicator lights

Edited April 25, 2016 by Do335

[+The Trooper 104]

Actually MiG-15/17's attitude indicator is quite different from that used on the more modern Russian jets. It is somewhat similar to the western approach, but verticaly inverted. The sphere is fixed relative to the ground and the aircraft "rotates" around it, so when you pitch up, you see more of the bottom half of the sphere. That's why the top is painted brown and the bottom - blue. We've discussed this earlier in the SF1 forums:
 
 

You know the funny thing about this whole discussion is that even as a western-trained flyer, I can see why this was done from an engineering standpoint.

Now, in the ADI that's being discussed, the idea is that the horizon on the 'ball' is ALWAYS level with the horizon on the planet. In 2 axes (roll and pitch) control, that makes perfect sense. If you think about the ball being 'fixed' in space, and the aircraft rotating around it, it makes sense why it reads the way it does.

It's funny, but when you think about western designed ADIs, they do the same thing...but in only one axis.

Think about it. A western 'ball' has the blue on top, brown on bottom...when you roll, the ball rolls away so it's always level with the horizon, exactly the same as a eastern 'ball'. The horizon's match (inner and outer).

However, when you pitch, the ball 'over rotates' to give you the pitch picture needed. An example...if you pitch 10 degrees nose high, the ball 'pitches' 10 degrees relative to the aircraft datum line...meaning the ball actually pitches 20 degrees upward relative to the horizon.

Which would make no sense on first reading of it, until you thought about it for a bit.

With the Soviet design, at least it's consistent in both axes.

Ironically, the early Su-27 HUD did the exact opposite of western designs. It behaves like you'd expect in the pitch axis, but 'over rotates' in the roll axis...ie in 90 degrees of bank, the aircraft symbol is 180 degrees to the horizon.

Weird stuff I think about while jogging...

FastCargo

Edited April 26, 2016 by The Trooper

[Do335 382]

oh yes I had searched "the sim that focuses on real life switchology", starting @ 18min

https://www.youtube.com/watch?v=DTnrwoa4zeI&feature=youtu.be&t=18m

 

so it works like that. But with the 3d model of the pit set as we have, if the ADI line moves vertically especially downwards it looks weird like objects in space. so opted for a vertically fixed ADI line while the ADI ball rotates western style.

I *think* when Stary made the pit, the intention was a completely western style ADI so, the current setup is a compromise... But I don't mind whichever way, as long as when you overshoot a F9F panther in a descending spiral, you know which way is up, so he won't have enough time to reverse and bring those lethal radar-ranging quad 20mm to bear

 

Actually MiG-15/17's attitude indicator is quite different from that used on the more modern Russian jets. It is somewhat similar to the western approach, but verticaly inverted. The sphere is fixed relative to the ground and the aircraft "rotates" around it, so when you pitch up, you see more of the bottom half of the sphere. That's why the top is painted brown and the bottom - blue. We've discussed this earlier in the SF1 forums:
 
 

 

 

PS for me the Su-27 etc style HUD is just counter-intuitive, reason is simply explained. With the western style HUD attitude indication, you only have to use one reference plane, which is the ground, either viewing outside or at instruments. With russian style HUD, when you look outside, your reference plane is the ground; but when you look at instruments, your reference plane is the HUD itself. -> 2 different reference planes. It doesn't help that HUD is focused at infinity so it wants you to reference the ground. One can get used to it certainly, but there is always an extra thought process required in that loop.

Edited April 26, 2016 by Do335

[+The Trooper 104]

I see... I was under the impression you had the source 3ds max file.

[Do335 382]

Negative. I only deal with database.

 

 

 

 

 

And finally remembered to do this test, so posting for document.

Theory: "Normal" FM do not depart/spin, whatever you do, but this changes after damage is taken even on normal FM -> once you overshoot AlphaDepart, the plane does enter flat spin.

Tis a campaign test flight and got shot by a 50cal AAA. Remembered this on the way back and so tried it. Spin is easily recoverable by player but AI can't do it.

(last part of vid was checking if damage reduces AlphaStall which it does, probably also AlphaDepart.)

https://vimeo.com/164273840

Edited April 26, 2016 by Do335

[Do335 382]

ayyy dunno whether to laugh or cry. As the united forces of NK/China/Soviet close on the pusan perimeter, the UN carrier groups somehow got full detection. So naturally this happened...

 

it is certainly exciting to watch but maybe full fledged red side campaigns will need to be separate ones after all, and no more SF2na carrier groups.

[Nicholas Bell 117]

I'm missing something here DO335.  Is the problem that the CV TF's act as a strike magnet resulting in the annihilation of the Communist Airforces?  No matter, the images look terrific.  Couldn't help but think "Kamikaze" 

[Do335 382]

I'm missing something here DO335.  Is the problem that the CV TF's act as a strike magnet resulting in the annihilation of the Communist Airforces?  No matter, the images look terrific.  Couldn't help but think "Kamikaze" 

ahahahah yeah it was also what i thought:D still contemplating if just leaving it as is for a bit of fireworks. will see!

[Do335 382]

1. G meter

2. Pressure Altitude

3. Airspeed gauge

4. Altimeter

5. Radar altimeter

6. ADI

7. Low altitude light

8. Slip indicator

9. Clock

10. External fuel tank lights

11. Low fuel warning light

12. VVI

13. Engine fire light

14. Fuel gauge

15. Engine overheat light

16. Bearing marker

17. Compass

18. Fuel flow gauge

19. Oil pressure gauge

20. Engine pressure ratio gauge

21. Engine RPM gauge

22. Exhaust gas temp gauge

23. Mach indicator

24. Cockpit pressure differential gauge

25. Gun indicator lights

Just a follow on to this one. After 50+ campaign missions in the mig with this pit setting aka proper testing and a few more fixes, it is finished far as I'm concerned. So I'll just UL it here.

The completed avionics makes me quite like the mig atm.

Could be better is the edits to the textures... the font is not an exact fit. (It's the same as the pit diagram above) It doesn't affect functionality any, but maybe someday Stary or someone may make it more proper, that would be great.

 

Everything is in metric so it might take some mental conversion from knots to km/h and feet to meters.

Edited May 6, 2016 by Do335

[serkarajan 2]

Brilliant work i like this and i have one question

Will there will be an Opfor campaign or something ? 

[Do335 382]

Will there will be an Opfor campaign or something ? 

Redfor campaign is already functional in uploaded 1.1.

 

 

 

 

i'm just making some adjustments so it can last longer i.e. more missions for campaign victory. but as you can see there's some side effect but.. isn't hard to fix this bit yessss.

Edited May 8, 2016 by Do335

[serkarajan 2]

Redfor campaign is already functional in uploaded 1.1.

 

 

 

 

i'm just making some adjustments so it can last longer i.e. more missions for campaign victory. but as you can see there's some side effect but.. isn't hard to fix this bit yessss.

great work thank's wish you all best man with your current progress !

[Do335 382]

Lights up a smoke....

 

 

Other than bug fixes, minor tweaks, and adding possible future new mods, work on this project is I reckon finished finally. All my thanks goes to the KAW team and the dedicated modders at CA that produced the korean war era mod series.

 

 

Far as I can see a 1950s theater is unique and a harder job to tackle because it is an era earlier and outta scope from the official TW games.

 

Out of 40 aircraft in the install, only 7 is from the default game and the remaining 33 relies solely on 3rd party mods. Outta 8 combat ships only 2 of them are TW default. And so more stuff to start from scratch and less stuff to directly reference. Not to mention the modern and Vietnam era theme is obviously more popular due to the intrinsic setting of SF.

Also, a guns only environment differs from the missile-era and requires different configs that can't be directly transferred over from the stock game.

Last but not least the terrain is completely new and campaign testing, especially the strategic nodes took up about half the time for my own part. <- Probably made it too complex myself so was barely able to get it finished...

 

 

Some rumblings... probably should setup a personal blog or summin' but at the moment this will do i reckon.

 

My own motivation has always been a lack of Korean war era themed flight sims. -- living "in theater" the korean war has been sort of a big deal even today, there is also a very large south korean expat community here although they mostly keep to themselves except the... restaurants... uhmm -- Mig Alley by rowan soon became abandon ware and the outdated graphics and some parts of game engine made it difficult to get into. The jet war mod for IL2 was great, but among various issues the most glaring is that IL2's AI is just way too problematic at the root level and 3rd party mods couldn't properly address it. DCS has the Sabre and Mig addon but the lifelessness and mod unfriendliness has always been a big no. There is a korea addon for CFS3 but it's rough around the edges from the looks of it. And so over the years despite an interest in this specific air war, there was never a good enough flight sim to fill this void:|

But SF fills it quite nicely. The graphics is not bad, the AI is kickass vs other sims, the campaign is not full blown Falcon level DC but surpasses everybody else that's still current, the FMs although a bit lacking can be openly modded. The game engine, "miraculously" it seems, supports this theater very well.

 

Since the CA modders made these great single mods, it just felt compelled to put them together or... what if TK did a SF2:Korea module.

 

For myself, while ULing a big mod took some effort, if a young "me" were to look for a korean war addon, he may stumble on this and have some enjoyment outta it, while I can somehow find a sense of closure. I think tis the best way to put it.

 

 

And so, on to the next one....

Edited May 21, 2016 by Do335

[Wrench 9,461]

Good on you, brudda!!

[Do335 382]

eh big thanks to you and the team Wrench!