DCS: F-14 Development Update - AIM-54 Phoenix Improvements & Overhaul - Guided Discussion

[Machalot]

On 9/21/2024 at 2:12 PM, Gareth Barry said:

what do you see as the limitations, perhaps in terms of context, of the nasa test?

First, I would encourage everyone on this thread to stop using the term "NASA test".  They ran simulations only, no tests.

Second, NASA used publicly available data, no doubt full of estimates and assumptions, along with an in-house trajectory optimizer to fly "zero-lift trajectories ... with no missile guidance (fixed fins)".  When I contacted the lead author through a mutual acquaintance to ask for more details, he described the results as "far from being definitive."

[Machalot]

On 9/21/2024 at 8:47 PM, Katsu said:

Here we have the results from Nasa: 


Please note that NASA indicates that the missile has a window above mach 4 of 41 seconds, what we have today replicating the same launch is a window of around 2 to 3 seconds.
which in my opinion seems to be a very significant “lack” of performance, since all that less speed will interfere with the missile's travel time, leading to various consequences. 

The NASA figure caption says that trajectory was optimized for maximum Mach number, which is not what the DCS missile guidance is optimized for.  Also note in your DCS data the missile has a little hiccup between 5 and 10 sec after launch, increasing only 0.1 Mach over 7 sec, while its average acceleration over the rest of the burn is more like 0.5 Mach every 5 sec (0.1 Mach/sec), and only exceeds Mach 3 nearly 18 sec after launch.  The NASA sim has no such lull period, reaching Mach 3 within 10 sec or so, which shows (no surprise) that its guidance (and possibly the aerodynamic model) is better optimized for the initial pull up.  With a pull-up guidance optimization, I think the DCS missile would gain roughly 0.5 peak Mach -- but this is not necessarily the guidance the missile should use tactically.

Edited September 24, 2024 by Machalot

[Katsu]

17 hours ago, Machalot said:

The NASA figure caption says that trajectory was optimized for maximum Mach number, which is not what the DCS missile guidance is optimized for.  Also note in your DCS data the missile has a little hiccup between 5 and 10 sec after launch, increasing only 0.1 Mach over 7 sec, while its average acceleration over the rest of the burn is more like 0.5 Mach every 5 sec (0.1 Mach/sec), and only exceeds Mach 3 nearly 18 sec after launch.  The NASA sim has no such lull period, reaching Mach 3 within 10 sec or so, which shows (no surprise) that its guidance (and possibly the aerodynamic model) is better optimized for the initial pull up.  With a pull-up guidance optimization, I think the DCS missile would gain roughly 0.5 peak Mach -- but this is not necessarily the guidance the missile should use tactically.

 

Well, I do believe that there is something wrong with the missile's current state of energy and I don't think that the loft trajectory alone is the main problem (but it is one of them).

And no, it's not based on thoughts but on various facts that have been presented here, simulations, reverse engineering of the original blueprint from the engine, calculations, documents etc...

The biggest fact that exists is that there is no consistency in the Phoenix information, various sources contradict each other, which is why I believe that the best way to reach a consensus on what the missile's performance would be calculations, CFD, and the kind of study that Tavarish has provided us with all this time that he does us the favor of trying to help.

an interesting fact is that Tavarish's data comes close to what the missile's performance could be is that with some programming knowledge and the data he provided, I was able to get close to the performance that Nasa simulation "presumably" presented.

null                   

If we get the first 60 seconds of flight we will a have a very close performance, just with minors tuning at thrust and burn time.

Just for comparision what we have now: 



But the problem seems to be on drag, missile seems to have a lot of drag to accelerate, and in the window between mach 4 and 3 seems to be the main issue.

Im studing the drag coefficients applied in the game and i'm looking for a solution, even if there is no official fix (if is there) I will make my own. 

Edited September 25, 2024 by Katsu

[Machalot]

6 hours ago, Katsu said:

But the problem seems to be on drag, missile seems to have a lot of drag to accelerate, and in the window between mach 4 and 3 seems to be the main issue.

Im studing the drag coefficients applied in the game and i'm looking for a solution, even if there is no official fix (if is there) I will make my own. 

The DCS missile may well have too much induced drag, if that is what's causing the 7 second period of very low acceleration I mentioned. 

[tavarish palkovnik]

Reconstruction of DCS model trying this 48kft 2M launch

Document1.pdf

And soon I will again be without possibility to upload files because reaching given maximum 

[tavarish palkovnik]

7 hours ago, Machalot said:

The DCS missile may well have too much induced drag, if that is what's causing the 7 second period of very low acceleration I mentioned. 

In these 7 seconds missile is overloaded of course. Overload is α*Cy*v˄2*ρ*A/(2*m*g) and because motor doesn't have enough kilograms of thrust, starting velocity gain is lower. Lower velocity for same G overload means higher angle of attack and then induced drag is coming in game Cx ind = Cx (α=0) + α˄2*Cy/57,3

Everything is connected. Simply, motor as in game and in addition for high altitudes where atmosphere is rare will give such velocity ''brake'' ... which is not normal. Normal motors, by the way, usually have higher thrust in start, just take a look how thrust-to-time curve of Sidewinder looks like. Also one of steps in chain of arming missile in flight is data from motor, data about chamber pressure, usually data is when pressure get down to some fixed value  

Edited September 25, 2024 by tavarish palkovnik

[Hotel Tango]

Tavarish is doing some god's work here, fellas.

[tavarish palkovnik]

Come on man, don’t exaggerate it now. Let’s stay on ground, this is after all just mathematics and mechanics

[draconus]

19 hours ago, Katsu said:

CFD

HB did CFD on Phoenix before F-14 release, afair.

[tavarish palkovnik]

What about Cy function (lifting coefficient and static stability) ? Did he make it as well?

You know, with these loft scenarios Cx is not going alone without Cy. And….uffff…this so complex, writing in not native language is difficult especially when subject is not so basic. Shortly, their AIM-54 could easily in this scenario become “AIM-180” , how I from jokes use to say, when angle of attack just turns nose of missile backwards 

 

Edited September 25, 2024 by tavarish palkovnik

[tavarish palkovnik]

Let’s put this stability aside for now, that would be really too much, of course if someone is interested in this also, no problem whatsoever to go through it. Subject is indeed interesting, not so easy to follow for those not having basics of mechanics but everything can be learned.

I said it’s maybe time to leave all this…but no, I will not leave it yet. Few days ago I saw, some guys on WT forum where they also talk about Phoenix, found some document and, step by step, some ideas started to rise, and it is a matter of time when that will flood internet again over already flooded internet when Phoenix is topic. Anyway…OMG how young’s like to say, somehow (I don’t know how) they got to theory that rocket motor burn time depends of altitude and that Phoenix’s motor depending of that can burn even up to 40 seconds  …  and that’s not all, at lower altitudes when paper work is read literally motor burns 15 seconds 

Let’s make retrospective and let’s see what mathematics says what is burn time, although it is already clear, at least to me, not from my calculations but NASA’s data, I’ve just calculated it my way to see does it fit, and it fits perfectly, from 20 to 30 seconds depending on temperature.

Now let’s use formulas involving neither specific impulse neither coefficient of thrust, but let’s go over characteristic velocity c*
 

 

What we have fixed:

Tc=5400-5700F … average 5550F what is 3339K

At=0,002408m^2 (Dtr=2,18”)

m=168kg

pc=700psi=4,8MPa (Rocketdyne’s data for reference chamber pressure)

k=1,2 … like most of Flexdyne CTPB compositions as well as R=307,9

And what would be characteristic velocity from Tc, R and k -> 1563 m/s

Now from characteristics velocity, chamber pressure and throat area mass flow rate goes -> 7,395 kg/s

And of course end is 168/7,395=22,7 seconds as total burn time, average at normal temperature. With mine 4,5 MPa as normal average it goes 24,2 seconds. 22,5 23,5 or 24,5 it doesn’t matter…but fifteeeeeeen to foooorthy 

 

 

[draconus]

26 minutes ago, tavarish palkovnik said:

average at normal temperature

Now calculate it at 10km alt where temp is -50C and the point will be proven

[tavarish palkovnik]

Already done at -40degC but seems you didn’t read it 

Fun fact…fuel in wings of passengers airplanes…outside hard minus, but friction which rises with altitude in same time warm that same fuel and fuel is in normal liquid state and there’s no need for heating elements  

Of course when aircraft goes in supersonics that change everything but not in way that minus will overcome but aircraft is going to be overheated 

[draconus]

44 minutes ago, tavarish palkovnik said:

Already done at -40degC but seems you didn’t read it 

I did but still can't find it.

[tavarish palkovnik]

Keep trying, there is somewhere in these last pages 

PS: short input, short output 

[Machalot]

10 hours ago, draconus said:

Now calculate it at 10km alt where temp is -50C and the point will be proven

The motor propellant temperature starts at roughly sea level ambient where it was stored.  It takes hours and hours for the propellant to equilibrate with the outside temperature, certainly not within a single F-14 sortie.

[Machalot]

For a bit more background on the effects of propellant temperature, see chart 7 here: https://web.stanford.edu/~cantwell/AA284A_Course_Material/Karabeyoglu AA 284A Lectures/AA284a_Lecture11.pdf

 

 

Edited September 27, 2024 by Machalot

[Viper33]

On 9/26/2024 at 6:59 PM, tavarish palkovnik said:

Let’s put this stability aside for now, that would be really too much, of course if someone is interested in this also, no problem whatsoever to go through it. Subject is indeed interesting, not so easy to follow for those not having basics of mechanics but everything can be learned.

I said it’s maybe time to leave all this…but no, I will not leave it yet. Few days ago I saw, some guys on WT forum where they also talk about Phoenix, found some document and, step by step, some ideas started to rise, and it is a matter of time when that will flood internet again over already flooded internet when Phoenix is topic. Anyway…OMG how young’s like to say, somehow (I don’t know how) they got to theory that rocket motor burn time depends of altitude and that Phoenix’s motor depending of that can burn even up to 40 seconds  …  and that’s not all, at lower altitudes when paper work is read literally motor burns 15 seconds 

Let’s make retrospective and let’s see what mathematics says what is burn time, although it is already clear, at least to me, not from my calculations but NASA’s data, I’ve just calculated it my way to see does it fit, and it fits perfectly, from 20 to 30 seconds depending on temperature.

Now let’s use formulas involving neither specific impulse neither coefficient of thrust, but let’s go over characteristic velocity c*
 

 

What we have fixed:

Tc=5400-5700F … average 5550F what is 3339K

At=0,002408m^2 (Dtr=2,18”)

m=168kg

pc=700psi=4,8MPa (Rocketdyne’s data for reference chamber pressure)

k=1,2 … like most of Flexdyne CTPB compositions as well as R=307,9

And what would be characteristic velocity from Tc, R and k -> 1563 m/s

Now from characteristics velocity, chamber pressure and throat area mass flow rate goes -> 7,395 kg/s

And of course end is 168/7,395=22,7 seconds as total burn time, average at normal temperature. With mine 4,5 MPa as normal average it goes 24,2 seconds. 22,5 23,5 or 24,5 it doesn’t matter…but fifteeeeeeen to foooorthy 

 

 

Why don't you apply to one of HB's open positions? Otherwise nothing will change if you just keep posting your math.

Edited September 28, 2024 by Viper33

[tavarish palkovnik]

1 hour ago, Viper33 said:

Why don't you apply to one of HB's open positions? Otherwise nothing will change if you just keep posting your math.

 

Already have employment, this is just a hobby and interest of mine. There is no enough math, never been, math is all around us.

Tell me please, what do you think, do you have in this simulation correctly modeled Phoenix in mechanical point of view?

This second sentence of yours is very worrying. Looks like situations in corrupted states where if you are not member of ruling political party you can bark all days long and nothing will change  Just joking, now honestly, like said it doesn’t bother me at all if anything get changed or not. But if I manage to make, at least few to get interested in all this and to start asking questions and perhaps express doubts, the goal is done.

Once I said here, flood on internet of wrong information about Phoenix is coming mostly from DCS and WT. And  I stand by that, simply, I don’t know terminology for that, but googling commonly sends you to related pages and then it is to late 

So, math will continue  , maybe guys from HB, or HB is a guy, I don’t know, will join us in this conversation, who knows. Stay well all of you 

 

[SSP1]

Я заметил что ракеты только в двух случаях ведут себя как надо.

[draconus]

2 hours ago, tavarish palkovnik said:

But if I manage to make, at least few to get interested in all this and to start asking questions and perhaps express doubts, the goal is done.

You already did much and even helped making some DCS missiles better when devs took your advices and help ex. for the HARM FM so it is appreaciated and a tribute for yourself. I even learned literal rocket science from you, thanks

Now about math in simulations - it's only good up to a point because it takes CPU time. Coders sometimes have to avoid too much math because of this. They make tables or roundings wherever possible and applicable instead.

Having said that I always prefer more realistic representation whenever feasible. As we learned HB men like to go back and update their work with new ideas and info when they have time so it's possible we may see some improvement in the future, although the new missile guidance, behavior and "API" is much more needed for Phoenix right now.

[JupiterJoe]

• Weapons. AIM-54C does not reacquire lock to target if lost while active - fixed.

[Viper33]

4 minutes ago, JupiterJoe said:

• Weapons. AIM-54C does not reacquire lock to target if lost while active - fixed.

Thats it??

[DoorMouse]

On 9/30/2024 at 12:50 PM, Viper33 said:

Thats it??

Fixing the Phoenix would require ED finishing their work on the aim-120 and new API that has been in development for.... Checks notes.... Four years?

 

So. Nearly almost not finished 

[Viper33]

1 hour ago, DoorMouse said:

Fixing the Phoenix would require ED finishing their work on the aim-120 and new API that has been in development for.... Checks notes.... Four years?

 

So. Nearly almost not finished 

Indeed. And there has been seemingly no improvement over those 4 years. AMRAAM's still fly formation with a cold target or switch to pure pursuit when a target does a hard 180 turn instead of aiming for the rate centroid. Not to mention the lack of actuall mass and inertia simulation making the missile pull way too much instant AOA and G for no reason.

Edited October 1, 2024 by Viper33