CFD of R-27ER

[Cmptohocah]

I hope it's ok to share a post from a different category here.

Basically I've completed the first round CFD trials for the R-27ER missile and the results can be seen here:

It would be great if people can comment, verify and/or criticize my results/experiment.

The goal is get the most accurate results as possible, within reason that is.

Thank you in advance.

[nighthawk2174]

I ran my own CFD some time ago on a full 3-D model with tapered wings feel free to compare.  Now this was one of the first missiles I did which was on much older hardware so the mesh resolution was nowhere near as high for my latter stuff like the HARM and SD-10 which were both within 3%'ish of ED's stuff.  Additionally I didn't run the intermediate values around M1.0 (.95,1.05,1.1,1.15) so the shape of the curve left of the peak at M1.2 is probably not correct either.

27R

27T

Edited January 15, 2021 by nighthawk2174

[Cmptohocah]

14 hours ago, nighthawk2174 said:

I ran my own CFD some time ago on a full 3-D model with tapered wings feel free to compare.  Now this was one of the first missiles I did which was on much older hardware so the mesh resolution was nowhere near as high for my latter stuff like the HARM and SD-10 which were both within 3%'ish of ED's stuff.  Additionally I didn't run the intermediate values around M1.0 (.95,1.05,1.1,1.15) so the shape of the curve left of the peak at M1.2 is probably not correct either.

27R

27T

 

This looks awesome. Do you have any 3D models to share? The one I used is accurate, dimension-vise, but as you have mentioned the flight surfaces are not fully modeled.

I did my tests in Autodesk CFD.

Also hit me up, if you are interested in contributing in any way.

[85th_Maverick]

On 1/14/2021 at 4:07 PM, Cmptohocah said:

I hope it's ok to share a post from a different category here.

Basically I've completed the first round CFD trials for the R-27ER missile and the results can be seen here:

It would be great if people can comment, verify and/or criticize my results/experiment.

The goal is get the most accurate results as possible, within reason that is.

Thank you in advance.

Welcome, man!

 

           First of all, thank you very much for also taking your time and passion to try obtaining some aerodynamic data estimations for the missiles in DCS (mostly A-A, at least...).

 

I've been battling this for more than 1 year on a different thread and just thought that I'm fighting wind mills as people behind modeling DCS missiles didn't want to accept that how they've modeled the missiles (especially air to air), made them all have a way too great minimum drag coefficient (all the missiles decelerated like crazy before 2.7 version), a too low induced drag exponent power (most certainly well below 2) which made all missiles loose airspeed at a too low rate for a given instantaneous airspeed and AoA, and in the case of the AIM-120, also an incredibly high maximum lift coefficient (due to the lift slope itself, not critical AoA (which is otherwise correct at about 28-30)), as they were simply able to make almost 2 turns (720 degrees) in subsonic while chasing a target..., that's how much incredibly high lift slope and very little drag vs AoA that missile had. I have dedicated a good time (being an ex-aerodynamicist) in order to determine more realistic CD vs AoA and CL vs AoA functions for most missiles and I've even offered them a modified "missiles_data.lua" file, which has now been rendered useless in version 2.7, file which offered the most accurate missiles aero data (within the limitations of the flight model in DCS) ever seen in DCS, determined through a part CFD and results corrections. Basically, a missile is an aircraft with a very high wing! So, by definition, as the burner no longer contributes to the lifting force by the sine of AoA, it would turn much worse than any fighter at the same IAS as well as decelerate slower than that fighter for the same G-load. Although I hope that I've somewhat contributed/managed to "wake them up" and help them understand that their AIM-120 was having a very unrealistically high lift as well as minimum drag and too little drag vs AoA exponent power, the 2.7 version still gives this missile as well as most others, a way too great maximum CL and still..., a CD0 (zero lift drag coef) much higher than in reality..., at least for both supersonic and subsonic. For transonic, it's quite hard to determine an accurate value just by CFD (that's the most inaccurate part for the results), yet if they'd accept to use a gross transonic CD0 of double the low subsonic one, it wouldn't hurt the accuracy unacceptably much!

So, my hope is, that if we can help them understand it and come with very consistent estimations (we'd most probably not have access to real wind tunnel tests data for AA missiles) and they'd accept it, they'd actually be able to make this sim make one step further in realism.

Thank you and keep it up:).

[GGTharos]

 

Quote

So, my hope is, that if we can help them understand it and come with very consistent estimations (we'd most probably not have access to real wind tunnel tests data for AA missiles) and they'd accept it, they'd actually be able to make this sim make one step further in realism.

 

You're way off base.  You're not going to force them to do something that's not on their schedule.  CFDs of these missiles done internally by ED was planned and they have carried this out.   People are upset that it will take a longer time to CFD and @Cmptohocah decided to run his own CFD to make a comparison.

 

ED will no more take those results than they did IASGATGs results (which resemble their own) for the AIM-120, not yours which came much later, as you've noticed.   And to that extent ED is being consistent and therefore everything is subject to the available manpower inside ED.

 

That doesn't mean that work by people has no value, even if ED in the end doesn't directly accept it:  With well documented and explained research, it will at least poke ED to perform a review of their own data and of course, the community can be informed of those findings as well.  So yep, if your work pointed them to a flaw that's great.  If that wasn't that did it (I mean it's not guaranteed that we'll ever find out), thanks for sharing your knowledge.

Edited April 28, 2021 by GGTharos

[nighthawk2174]

13 hours ago, 85th_Maverick said:

also an incredibly high maximum lift coefficient (due to the lift slope itself, not critical AoA (which is otherwise correct at about 28-30)), as they were simply able to make almost 2 turns (720 degrees) in subsonic while chasing a target..., that's how much incredibly high lift slope and very little drag vs AoA that missile had. I have dedicated a good time (being an ex-aerodynamicist) in order to determine more realistic CD vs AoA and CL vs AoA functions for most missiles and I've even offered them a modified "missiles_data.lua" file, which has now been rendered useless in version 2.7, file which offered the most accurate missiles aero data (within the limitations of the flight model in DCS) ever seen in DCS, determined through a part CFD and results corrections. Basically, a missile is an aircraft with a very high wing! So, by definition, as the burner no longer contributes to the lifting force by the sine of AoA, it would turn much worse than any fighter at the same IAS as well as decelerate slower than that fighter for the same G-load. Although I hope that I've somewhat contributed/managed to "wake them up" and help them understand that their AIM-120 was having a very unrealistically high lift as well as minimum drag and too little drag vs AoA exponent power, the 2.7 version still gives this missile as well as most others, a way too great maximum CL and still..., a CD0 (zero lift drag coef) much higher than in reality..., at least for both supersonic and subsonic. For transonic, it's quite hard to determine an accurate value just by CFD (that's the most inaccurate part for the results), yet if they'd accept to use a gross transonic CD0 of double the low subsonic one, it wouldn't hurt the accuracy unacceptably much!

And as was shown to you in the other thread the current lift values lineup with what was measured in a few wind tunnel tests that we have.  The current values really don't seem out of line at all.

[Cmptohocah]

Could someone explain to me what is the relationship between lift-induced drag and speed loss (non-motor burning part of the missile flight).
In order to turn, the missile needs some sort of force to create a change in flight path. This force is achieved by increasing AoA which in turn produces additional lift.

Is there any clear relationship between this added drag force and kinetic energy loss?

 

What I am getting at is: will a missile that pulls certain amount of G in a turn, have more/less/same speed loss than a fighter that is pulling exactly the same G? Thrust is to be ignored.

[GGTharos]

You would have to calculate it - having a CFD handy, you're in the best position to actually 'do the math' on it.

[Exorcet]

On 4/29/2021 at 5:06 AM, Cmptohocah said:

What I am getting at is: will a missile that pulls certain amount of G in a turn, have more/less/same speed loss than a fighter that is pulling exactly the same G? Thrust is to be ignored.

It depends. Smaller, lower aspect ratio lift surfaces on a missile will tend to be draggier than a plane (but this also depends on speed), but if the missile is light enough that can be overcome because it will only take a small amount of force to turn. You can't answer the question purely with aerodynamics.

[Cmptohocah]

On 5/1/2021 at 3:51 AM, Exorcet said:

It depends. Smaller, lower aspect ratio lift surfaces on a missile will tend to be draggier than a plane (but this also depends on speed), but if the missile is light enough that can be overcome because it will only take a small amount of force to turn. You can't answer the question purely with aerodynamics.

I have a feeling that this might be a dynamics problem. I would have to do more research, but I think it would have to do with the object's impulse and the magnitude of the drag force produced.