SuicidalInsanity

DJA 1946 BDAc AI Dogfight Tournament

Recommended Posts

Posted (edited)

Whoa... this is brutal...

Granted these are illegal 160-point heavy Me-262-alikes that put 120 of that into weapons but still... murder beam.

In a few seconds I lost the AI module, both engines, and an intake, as well as parts of the tail.

E6vEV2b.jpg

Edited by Pds314

Share this post


Link to post
Share on other sites
2 hours ago, Pds314 said:

Whoa... this is brutal...

That moment you realize that your FAR plane-building skills are crap and you don't stand a chance against the aircraft entered. :D

(At one point I thought I might be able to enter. That illusion has long since passed.)   ~ The Dunatian

Share this post


Link to post
Share on other sites
Posted (edited)
11 hours ago, SuicidalInsanity said:

I don't really expect such a design strategy to offer substantially more longevity

after some hit tests, 20mm rounds can one-shot b9 control surface, 2-3 shot main wings, while stock type C wing complex can withstand much more than that. A direct hit from 3 30mm shots results in losing two wing components, moderately damaged the surrounding one. Proc wing complex is as fragile as proc wings alone, 1 23mm round can destroy half of the wing. I tried to install 30mm cannon on the small fighters, it is extremely hard for 30mm shells to score any hit on a maneuverable target.  using 40mm is completely overkill and equivalent to gambling. If you cannot finish your opponent within the first head-on pass, you are finished due to inferior TWR and wing loading. 

I don't think critical Mach number really matters,  as dogfights of early jets are mainly in the subsonic regime. The one really matter is the Clmax. With proc wings, you can hit 2 while stock wings only 1-1.5. But the effect is really limited as you will break your wings if pulled to critical AoA.

Edited by Me1_base

Share this post


Link to post
Share on other sites
Posted (edited)
4 hours ago, The Dunatian said:

That moment you realize that your FAR plane-building skills are crap and you don't stand a chance against the aircraft entered. :D

(At one point I thought I might be able to enter. That illusion has long since passed.)   ~ The Dunatian

I mean, those planes are hardly "engineered." More like ,"slapped together with moderate concern for aerodynamics." I didn't even touch the FAR tab until after they were built to check that their stability was okay. I don't even know what their stability derivatives are doing and their transonic drag area is an unexceptional 0.7 m^2. vs. like 0.25 for optimized transonic fighters, and the only thing I did to mess around with the CL curve was preemptively give the wings 100 points of AOA deflection before even checking the CL curve.

Edited by Pds314

Share this post


Link to post
Share on other sites
5 hours ago, The Dunatian said:

That moment you realize that your FAR plane-building skills are crap and you don't stand a chance against the aircraft entered. :D

(At one point I thought I might be able to enter. That illusion has long since passed.)   ~ The Dunatian

Building a craft in FAR really isn't all that different from stock. Just pay lip-service to aerodynamics, make sure CoM is about where the main wing is, and that your control surfaces don't have crazy high deflection, and the craft should fly. The main 'difficulty' with FAR is relearning how wings and control surfaces work (in the context of KSP mechanics, not IRL) - wings are no longer magic speed-powered antigravity devices, so some attention will need to be paid in regard to wing segments that are not perpendicular to prograde, in tweaking control surfaces so hard turns don't rip wings off, which will require a test flight or two, and making sure you've paid sufficient heed to aerodynamic principles when trying to build things like spaceplanes that require stability in both subsonic and supersonic flight. (And even then it pretty much boils down to "Do I have a big enough vertical stabilizer/is it far back enough?" and "Is my CoM more or less where it needs to be?")

No matter how bad you think your craft is, you are still able to enter. If your craft has issues, post a pic. We can probably tell you what's wrong and how to fix it.


Pds314: Kerbal G-limits will be off for this one.
 

Share this post


Link to post
Share on other sites
On 5/26/2019 at 2:27 AM, Me1_base said:

1 vs 1, stock wings are kind of weird. BTW does anyone consider posting their design here other than me and dundun92? 

Stock wings have 500-1000 hit points while b9pw only have 100-300 depends on size, will there be any balance on that?

 

I will post some planes later for download but they're not gonna be entries or finalized designs.

Share this post


Link to post
Share on other sites

Despite vast differences in the characteristics of these planes, the AI is as ram-happy as ever lol.

8VBK3mi.jpg

Share this post


Link to post
Share on other sites
Posted (edited)

Some of my heavy planes:

From left to right:

Horton 229 analogue but with the top tier axial engines. AI doesn't know how to fly it, obviously.
Armament: none. Would be dual 30mm if AI could use.
Top speed: ~270 m/s on a good day. Transonic drag is not kind to it. Nor is mach tuck.
Turning performance: stalls easily but turns sharp.

Dual Phantom Heavy. Loosely inspired by a very large version of a Yak-23. Currently most-promising.
Armament: Dual 30mm.
Top speed: 308 m/s.
Turning performance: It can do a 9-G turn at 100 m/s.

Speed Demon. Loosely inspired by experiments like the X-3 Stiletto.
Armament: Dual 30mm. Formerly 40mm + triple UBK 12.7mm but that made it 150.1 points.
Top speed: 377 m/s.
Turning performance: good at high speeds. Bad at low speeds.

Me-262 analogue using Gremlins. 160-point illegal fighter.
Armament: Quad 30mm murder beam of death.
Top speed: 286 m/s.
Turning performance: reasonably-good at all speeds but bleeds energy and can't get it back thanks to anemic TWR.

C4qopvP.png

Edited by Pds314

Share this post


Link to post
Share on other sites

Has anyone managed to get the Warlock to actually get 28 kN of thrust? I haven't even managed 24 including with the oxidizer-hungry water injection mode.

Share this post


Link to post
Share on other sites
19 hours ago, SuicidalInsanity said:

If your craft has issues, post a pic. We can probably tell you what's wrong and how to fix it.

Since you have offered I'll take you up on it.

This is the bones of my aircraft. It's a fairly sleek design and should have enough thrust from the dual engines. Unfortunately I just can't figure out what to do for a wing. Everything that I try seems to fail. Either there isn't enough lift or the wing creates too much drag and stalls.

xuTYKAe.jpg

Share this post


Link to post
Share on other sites
Posted (edited)
4 hours ago, The Dunatian said:

Since you have offered I'll take you up on it.

This is the bones of my aircraft. It's a fairly sleek design and should have enough thrust from the dual engines. Unfortunately I just can't figure out what to do for a wing. Everything that I try seems to fail. Either there isn't enough lift or the wing creates too much drag and stalls.

xuTYKAe.jpg

Narrow the area of fuselage around the wing?

 

In any case very few of the designs here are supersonic (most of what people have posted are obvious turnfighters with massive wings) and out of all the dogfights my supersonic plane with wave drag area of 0.12 m^2 has done, only in two of them did it break 340 m/s.

Edited by Pds314

Share this post


Link to post
Share on other sites
3 minutes ago, Pds314 said:

<snip>

Thankyou for taking the time to try and help me, but what I really need to know is how large should a wing be in FAR? I always tend towards cosmetics over function, which I suspect hurts my designs. Should I use a highly swept/delta wing or a regular straight wing?

Share this post


Link to post
Share on other sites
Posted (edited)
5 hours ago, The Dunatian said:

Thankyou for taking the time to try and help me, but what I really need to know is how large should a wing be in FAR? I always tend towards cosmetics over function, which I suspect hurts my designs. Should I use a highly swept/delta wing or a regular straight wing?

It depends on whether you want good speed, roll performance, stability, turning?

 

Larger wing area gives you better acute turning performance.

 

Wider wings give you better energy retention in a turn.

 

Deeper wings give you high angle of attack capabilities but slightly reduce lift coefficient for low angles of attack.

 

Highly swept wings give you better supersonic and transonic L/D because they are mostly inside their own mach cone and therefore they do not need to try to redirect shockwaves, just regular air pressure. Not that it's impossible for a wing to redirect the explosion-like shockwave of supersonic air, but it's certainly less efficient.

 

Lightly-swept wings help at transonic speeds because they prevent acute changes in wave drag area.

 

Sharp leading edges and low camber (flat airfoils) help delay transonic effects to higher mach. E.G. mach 0.75 instead of Mach 0.5.

 

Counterintuitively, negative AOA% on the front leading edge can massively delay stall and increase high AOA performance while reducing drag.

 

Deep wing roots and extended leading edges like on an F-18 delay full stall. I forget exactly how though.

 

FAR does model wing fences and wingtip devices but I don't remember the exact reasons those help. I do know winglets reduce wingtip vortices but I forget what MiG-17-style wing fences do.

Square, board-shaped wings are inefficient and produce more lift-induced drag than trapezoidal ones. The ideal profile is an ellipse but don't bother because you'll get >90% efficiency from a good trapezoid shape.

 

So if your goal is an area-ruled transonic fighter I would probably give it a ~45 degrees swept wing that's fairly deep but not exceedingly long (i.e. 11 meter wingspan is probably ok. 19 meter wingspan less so). Then use other tricks I mentioned like leading edge -% AOA control to give it decent turning performance below that speed without bleeding too much energy.

 

Lastly, remember that you could end up fighting someone who says "I don't care if my top speed is 240 m/s I'm maximizing turning performance and turning performance alone." Unless you are doing the same or manage to get an initial speed advantage, out-turning them is unlikely.

Edited by Pds314

Share this post


Link to post
Share on other sites
16 hours ago, Pds314 said:

Speed Demon. Loosely inspired by experiments like the X-3 Stiletto.
Armament: Dual 30mm. Formerly 40mm + triple UBK 12.7mm but that made it 150.1 points.

Weapons round down to nearest integer for DJA, so it would be 148 points for 40mm+3xUBK.

 

5 hours ago, The Dunatian said:

Thankyou for taking the time to try and help me, but what I really need to know is how large should a wing be in FAR? I always tend towards cosmetics over function, which I suspect hurts my designs. Should I use a highly swept/delta wing or a regular straight wing?

Swept wings. Straight wings don't do as well at higher speeds, but the BADT Jets the competition is using will struggle to hit Mach 1, so a delta isn't really necessary.

As Pds314 said, wing shape and size is highly dependent on desired flight characteristics and envelope. that said, gut feeling for that fuselage and engines (assuming ~5 ton mass) would be something like a pair of Wing Connector Es with a Structural Wing C in front of those, an elevon 1 and 2 behind, mounted on a wing root composed of a fore Structural Wing D and an aft strake, which should give something like a 30 deg sweep, and give the whole thing a wing incidence of 1-2 deg. Swap out the strake for a second SW D if you want a sweep closer to 45 deg. Maybe add fore elevons, either for slats or simply to increase wing area, and increase the size of the rear control surfaces if you want maneuverability over pure speed.

Share this post


Link to post
Share on other sites
Posted (edited)

A new single engine design I built:
zxQ36SO.png

Pulling a 9G turn at 100 m/s.
WxIgsEI.png

It has about 0.50 m^2 of wave drag (the original version, before optimizations, had like 1.3). Top speed is 282 m/s.
It's armed with six 12.7mm Berezin UB. It fires "warning shots" 1500 meters out with two of them to get the enemy to try to dodge and bleed energy (or who knows, maybe take out a 100 hp part?). Then waits until 500 meters to open fire with all of them.

Edited by Pds314

Share this post


Link to post
Share on other sites

I have a video uploading of a test fight between two of my own creations.

Share this post


Link to post
Share on other sites
Posted (edited)

My craft so far. Some need tweaking to fight efffectively. Some need tweaking to take off effectively.
Vulture Mk1 (Single engine plane): https://kerbalx.com/pds314/Vulture-mk1
Boom A (Supersonic plane): https://kerbalx.com/pds314/Boom-A
Horton (Flying wing, this isn't a good fighter. I doubt the AI could even get to 150 meters before crashing): https://kerbalx.com/pds314/Horton
Mescherschmisn't Shwallon't (Me 262 analogue): https://kerbalx.com/pds314/Mescherschmisnt-Shwallont
Minifast (attempt to make an axial flow single engine plane go fast. Not that fast but probably and ok fighter): https://kerbalx.com/pds314/Minifast
Centrifast (attempt to make a tiny centrifugal flow single engine plane go fast. Very slightly faster than Minifast but not a good fighter. Warranty void if stared at): https://kerbalx.com/pds314/Centrifast
Fat 23 (dual engine plane with Yak-23-like engine placement): https://kerbalx.com/pds314/Fat-23

Testing two of my aircraft. The AI on the Fat 23 needs work.

 

Edited by Pds314

Share this post


Link to post
Share on other sites
Posted (edited)

Another test with improved AI tuning.
 

 

Edited by Pds314

Share this post


Link to post
Share on other sites
Posted (edited)

@Me1_base BTW, Simple Violence 3 has a MechJeb part in it that I can't open.

And just a reminder WRT Simple Violence 4: running out of ammo means you lose, and running out of fuel means you get shot down. My testing shows that your craft literally runs circles around my heavier fighter, but often does superficial damage if it only manages wing hits or something, then runs out of ammo, then fuel, then this happens:

f8N9NVo.png

Edited by Pds314

Share this post


Link to post
Share on other sites
Posted (edited)

@Me1_base

However, if it gets a good center hit on anything, well obviously it's a Vickers gun.

WRyryUd.png

Edited by Pds314

Share this post


Link to post
Share on other sites

oswald.jpg

18 hours ago, Pds314 said:

Square, board-shaped wings are inefficient

Taper ratio is not important for low/medium AR wings.:D

Share this post


Link to post
Share on other sites
8 hours ago, Me1_base said:

oswald.jpg

Taper ratio is not important for low/medium AR wings.:D

Wait what value is Y axis? Is that e?

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.