Still struggling to understand new missile mechanics

[maverick3320]

My F-14, traveling at 920 knots, fires 2x AIM-54C Phoenix at a range of roughly 25nm. The target is a Su-27S Flanker B traveling at 920 knots heading straight toward my Tomcat and then slows down to 350 knots after firing its own missiles. There is no evasive action (that I can see) by the Flanker.

Two missiles with a range of 99nm peter out after traveling 20nm or so, with no maneuvers by the target or the missile, just a straight head shot.

What am I doing wrong?

Save attached.

[Dimitris]

What am I doing wrong?

You are firing the Phoenix at 25nm, at a much lower target.

1) The Phoenix missile has a low auto-calculated burn time thanks to its lofting and substantial mass (ie. it rapidly builds up speed again on the post-loft dive). This is how it achieves its nominal long range. I have explained in a previous post how the boost times are auto-calculated.

2) The Phoenix missiles are fired on a very shallow loft angle, to avoid the "target sprinting-under during the loft" problem.

3) The combination of minimal loft at a much lower target (ie. high drag) and short burn time means that the missile will rapidly run out of speed.

See what happens if you fire the Phoenix at long range, the way God and Hughes' designers meant it to.

Also read the last few posts here: https://www.matrixgames.com/forums/view ... 1&t=393989

[Dimitris]

Also, at the save you provided I see the missiles consistently impacting.

[CV60]

Interesting BLUF, you may want to put in a bug report. Here is what I see: At time 0, your AIM-45 are at approximately 1 descending from 50K feet to attack the target Su-27s, which are at 18K feet, 14.5nm distant. AIM-54s are travelling at 2110 knts (approx. Mach 3.1). Aspect is head-on. See attached image

Just before impact 29 seconds later, after traveling approximately 10 nm horizontally, the AIM 54s are traveling at only 761 knots. This may be bleeding speed too quickly, especially since the missile is still descending. (I placed "Xs" (reference points) on the images at 5 mile steps).

As far as the engagement itself, an AIM-54 is not particularly effective against an Su-27. It was designed for use against bombers. In the log, it has only a 38% chance of a hit. My log is below:

The big issue is the speed loss of the AIM-54, which IMHO, is excessive.

Time 0.jpg (335.59 KiB) Viewed 3349 times

T029.jpg (342.87 KiB) Viewed 3349 times

[Edit: I see Dimitris' response and I don't disagree with his logic: you are firing the missile too close, which is your fundamental problem. My question is on the rate the speed bleeds-off. I'm not saying it is wrong (and I personally don't have real world experience with the AIM-54). I just question it.

Edit2: Just to clarify my comments: the actual combat result is believable: a 38% of hitting a Su-27 with a AIM 54 in a HtH engagement is probably right. It is the bleeding of the speed I question. Although I do not have experience with the AIM-54, I do have real world experience with high power rocketry, (which is admittedly not the same thing as an AIM-54), and from that experience believe the AIM54 shouldn't bleed off speed that quickly. With that said, it won't make any difference in the combat result that you got in this engagment.




2/24/1994 7:48:08 AM - Decoy (Generic Chaff Salvo [4x Cartridges]; Tech: N/A) from IceMultiRole #8 is attempting to seduce sensor: Active Radar Seeker (Tech: Late 1980s)(Guiding weapon: AIM-54C Phoenix #7679). Final probability: 15%. Result: 29 - FAILURE

2/24/1994 7:48:08 AM - Defensive jammer (SPS-171/172 Sorbtsiya [L-005]; Tech: Early 1980s) on IceMultiRole #8 is attempting to spoof sensor: Active Radar Seeker (Tech: Late 1980s)(Of: AIM-54C Phoenix #7679). Final probability: 15%. Result: 35 - FAILURE

2/24/1994 7:48:08 AM - Weapon: AIM-54C Phoenix #7679 is attacking IceMultiRole #8 (Su-27S Flanker B) with a base PH of 85%. PH adjusted for weapon speed: 54% (pure-aerodynamic attitude control). IceMultiRole #8 has nominal agility: 4.5, adjusted for altitude: 3.8. Agility adjusted for proficiency (Veteran): 3.8. Aircraft has a weight fraction of 0.49 - Agility adjusted to 2.67. Agility adjusted for head-on impact effect: 1.6. Final agility modifier: -16%. Final PH: 38%. Result: 62 - MISS

2/24/1994 7:48:08 AM - [WP] WP :Contact MISSILE #212 has been lost.

2/24/1994 7:48:08 AM - [No Strike List] WP :Contact MISSILE #212 has been lost.

2/24/1994 7:48:08 AM - [No Strike List] No Strike List :Contact MISSILE #212 has been lost.

2/24/1994 7:48:09 AM - Decoy (Generic Chaff Salvo [4x Cartridges]; Tech: N/A) from IceMultiRole #8 is attempting to seduce sensor: Active Radar Seeker (Tech: Late 1980s)(Guiding weapon: AIM-54C Phoenix #7680). Final probability: 15%. Result: 12 - SUCCESS

[Dimitris]

From Heatblur's research paper on the AIM-54: https://web.archive.org/web/20230213075 ... AIM-54.pdf

[CV60]

From Heatblur's research paper on the AIM-54: https://web.archive.org/web/20230213075 ... AIM-54.pdf

I agree the combat results are likely close to what you would get it real life, and that is the main thing. It is impressive what you have been able to do with the game, and modeling all the various factors. My issue is based not on data, but on a "feel" i have from experience in firing unguided, high power amateur rockets. And even if I am right, and the AIM 54s are losing energy too fast, it wouldn't have made any difference in the combat results: They missed an Su-27, which is entirely believable, and since they can't turn around, any errors in calculating the energy loss doesn't make any difference. The only possible importance is in how quickly the engagement is made.

I don't doubt the AIM-54 will be making sea-skimming intercepts at sub-Mach. After the 8 sec. boost, the missile will rapidly lose energy in the low atmosphere. But in the scenario the missile was rapidly loosing energy at 40+ feet, and finally dropping to Mach 1 at about 18,000. I think this is too quickly, as I have shot rockets from SL with 8 sec. burns. After accelerating to M 1.3, at about 1000 feet, they coast straight up to about 11,000 feet before coming over, losing 866 knots in a straight 10,000 foot climb (in admittedly thinning atmosphere). Here, in a descending profile (in very thin atmosphere) the missile using a approximaty 2:1 glideslope lost 1600 knots. That feels excessive. However, I admit it doesn't make a difference from the simulation point of view: You have created a wonderful simulation of modern naval/air operations, and not an aerodynamic simulator.

[artao5]

Perhaps slightly off topic here --
As a model rocket enthusiast who designs his own rockets, I'm curious if you're using the extended Barrowman equations to simulate missile motor burn and vehicle coast.
Most missiles have what rocketeers would consider an insane amount of fin area. Not to mention the no-no of putting fins so far forward. But then, missiles have advanced active guidance systems while model rockets use passive fin-only guidance. And the more total area of an missile/rocket, the faster it loses velocity and momentum; particularly when the airstream isn't 100% running straight down the length axis of the rocket. And that loss in momentum sure isn't linear, but I'm not sure if it's exponential or what off the top of my head. (RE: surface drag and AOA drag, not to mention interference drag)

[CV60]

Perhaps slightly off topic here --
As a model rocket enthusiast who designs his own rockets, I'm curious if you're using the extended Barrowman equations to simulate missile motor burn and vehicle coast.
Most missiles have what rocketeers would consider an insane amount of fin area. Not to mention the no-no of putting fins so far forward. But then, missiles have advanced active guidance systems while model rockets use passive fin-only guidance. And the more total area of an missile/rocket, the faster it loses velocity and momentum; particularly when the airstream isn't 100% running straight down the length axis of the rocket. And that loss in momentum sure isn't linear, but I'm not sure if it's exponential or what off the top of my head. (RE: surface drag and AOA drag, not to mention interference drag)

I'm lazy. I use Rocksim. https://www.apogeerockets.com/index.php ... ts_id=2636
It See: https://apogeerockets.com/education/dow ... ter238.pdf for a paper on Rocksim v. Barrowman. However, Rocksim does allow you to solve using Barrowman, but in my experience, I think Rocksim gives better results. As far as the simulating the motor burn out, the thrust curves are actually published, so Rocksim uses the actual data for those. Cf. https://www.thrustcurve.org/motors/AeroTech/L339N/

You are absolutely correct that an air-to-air missile is different than a high power rocket in a number of ways, so I'm not saying that the simulation is wrong on its modeling of the loss in velocity of the AIM-54. It just "feels" wrong based on my hobby experience, for the reasons I described earlier. And while the 2:1 glideslope in the problem we are discussing would cause drag, the high altitude of the missile for much of the flight would minimize some of the effects of the drag, and the fall from 50K to 18K would provide some energy. [Edit: I just rechecked my math, and it is closer to a 1:1 glide ratio, so that would increase the drag] My opinion is based entirely on a gut feel, based on experience, but I could easily be wrong. If anyone has any access to any actual unclassified flight data for medium range air to air missiles, it might shed light on the issue.

With that said, the discussion is academic, assuming the range of the missile and the combat results are accurate. In the scenario that started this thread, I believe the combat results are in the bounds of what we know about the AIM-54. It wasn't intended to be used against fighter aircraft, so a 38% kill rate against a Su-27 sounds about right, as is having a reduced range in the geometry of the engagement that we were discussing.

[thewood1]

I posted a podcast a few months ago with a retired F-14 pilot. Their perspective on the AIM-54 is pretty different than the discussions on this board. I'll see if I can track it down again.

edit: https://www.matrixgames.com/forums/view ... 9#p5078249

Listening to that video and podcast gives a new appreciation of the myths that surround a A2A missile combat.

[artao5]

I'm lazy. I use Rocksim. https://www.apogeerockets.com/index.php ... ts_id=2636
It See: https://apogeerockets.com/education/dow ... ter238.pdf for a paper on Rocksim v. Barrowman. However, Rocksim does allow you to solve using Barrowman, but in my experience, I think Rocksim gives better results. As far as the simulating the motor burn out, the thrust curves are actually published, so Rocksim uses the actual data for those. Cf. https://www.thrustcurve.org/motors/AeroTech/L339N/

Oh geez, I don't do such by hand. Hell no! I use Open Rocket. I can't afford RockSim, and I'll never do high power rocketry so I'd never use its full feature set. Open Rocket fully suffices for me; tho it does "only" use the extended Barrowman equations, which totally ignore the airframe as far as drag goes. Rocksim can if you set it to, which is pretty important for HPR.
Once a body is free-flying and has no thrust, it's totally at the whim of gravity and drag. As The Tick once said, "Gravity is a harsh mistress."

[maverick3320]

Also, at the save you provided I see the missiles consistently impacting.

I think you are forgetting the part where the target is not taking evasive action. If it had, A phoenix flying at 760 knots after only "flying" 25nm isn't going to impact anything.

[maverick3320]

Perhaps slightly off topic here --
As a model rocket enthusiast who designs his own rockets, I'm curious if you're using the extended Barrowman equations to simulate missile motor burn and vehicle coast.
Most missiles have what rocketeers would consider an insane amount of fin area. Not to mention the no-no of putting fins so far forward. But then, missiles have advanced active guidance systems while model rockets use passive fin-only guidance. And the more total area of an missile/rocket, the faster it loses velocity and momentum; particularly when the airstream isn't 100% running straight down the length axis of the rocket. And that loss in momentum sure isn't linear, but I'm not sure if it's exponential or what off the top of my head. (RE: surface drag and AOA drag, not to mention interference drag)

I'm lazy. I use Rocksim. https://www.apogeerockets.com/index.php ... ts_id=2636
It See: https://apogeerockets.com/education/dow ... ter238.pdf for a paper on Rocksim v. Barrowman. However, Rocksim does allow you to solve using Barrowman, but in my experience, I think Rocksim gives better results. As far as the simulating the motor burn out, the thrust curves are actually published, so Rocksim uses the actual data for those. Cf. https://www.thrustcurve.org/motors/AeroTech/L339N/

You are absolutely correct that an air-to-air missile is different than a high power rocket in a number of ways, so I'm not saying that the simulation is wrong on its modeling of the loss in velocity of the AIM-54. It just "feels" wrong based on my hobby experience, for the reasons I described earlier. And while the 2:1 glideslope in the problem we are discussing would cause drag, the high altitude of the missile for much of the flight would minimize some of the effects of the drag, and the fall from 50K to 18K would provide some energy. [Edit: I just rechecked my math, and it is closer to a 1:1 glide ratio, so that would increase the drag] My opinion is based entirely on a gut feel, based on experience, but I could easily be wrong. If anyone has any access to any actual unclassified flight data for medium range air to air missiles, it might shed light on the issue.

With that said, the discussion is academic, assuming the range of the missile and the combat results are accurate. In the scenario that started this thread, I believe the combat results are in the bounds of what we know about the AIM-54. It wasn't intended to be used against fighter aircraft, so a 38% kill rate against a Su-27 sounds about right, as is having a reduced range in the geometry of the engagement that we were discussing.

The "process is wrong, but the result is okay, so just leave it" only works because the target was not taking evasive action.

[Dimitris]

Also, at the save you provided I see the missiles consistently impacting.

I think you are forgetting the part where the target is not taking evasive action. If it had, A phoenix flying at 760 knots after only "flying" 25nm isn't going to impact anything.

We just released a new update which tweaks the drag and burn auto-calc mechanics for (among other things) edge cases like this. Give it a try.

[Tcao]

I tried the BVR quick battle, my own testing scenarios and maverick3320's couple times under 1307.12

AIM-54C improved a lot. I think that should be thank to the increased fuel.
AIM-54C's motor burnout time increased from 7s in pervious version to 48s.


AIM-120-C7's performance worsened. Boost time increased from 5s to 8s but previously it can hit an evasion target when fired at 20nm, now it is better to let the target get closer into 15nm range.

So far this is my first impression with a limit number of tests.

[Aaroneyeball]

I have struggles with PL-15 missile when playing PLAA
The missile travel at 62k ft all the way towards target no matter what altitude the target is. as the target plane always took dive, the missile will just fly past the target then decrease altitude.

PL-12 seems alright with its arc of travel. but PL 15 is really useless with this "bug"

does anyone have the same issue?

[thewood1]

What version are you using? Is it Steam? A little detail might help.

[Dimitris]

I have struggles with PL-15 missile when playing PLAA
The missile travel at 62k ft all the way towards target no matter what altitude the target is. as the target plane always took dive, the missile will just fly past the target then decrease altitude.

PL-12 seems alright with its arc of travel. but PL 15 is really useless with this "bug"

does anyone have the same issue?

This is a known issue that has already been fixed: https://www.matrixgames.com/forums/view ... 1&t=394677 (pay attention to bold line)

[Aaroneyeball]

Thanks Dimitris. I was playing the steam version and 1307.13 is the latest version for now. I've manually updated it to 1307.14 and hope it get solved.