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

[DD_Fenrir]

An "AIM-54" did reach Mach 5... sort of.

The Mach 5 figure comes from here:

NASA's plan to use an F-15 to launch hypersonic Phoenix missiles - Sandboxx

But when you read just how many modifications NASA made to the missile to get it to reach Mach 5 you start to understand just how little this test missile has in common with a US Navy fleet missile.

1. All of the internal components related to the missile’s guidance system and explosive payload were completely removed, including its guidance computer and radar tracker, leaving just its propulsion and control sections at the rear of the missile intact
2. a new nose with slightly more sloping angles was added to what would now be primary and secondary payload sections with the same 15″ diameter as the original components. The primary payload section measured about 57 inches long and, based on the weight of the guidance section it replaced, could carry approximately 184 pounds worth of testing equipment.

It's an AIM-54 in name only.

Then there's the flight profile - the fact that the launch was committed at Mach 2 and that the missile was not required to make any steering corrections as it had no target widens the gap between it and a fleet Phoenix; there's so little in common here with an operational launch of an AIM-54 that it might as well be a completely different missile.

Edited November 23, 2022 by DD_Fenrir

[IronMike]

1 hour ago, Fliegerkalle said:

Hi ironmike, okay I understand - I asking only my self as a completely noob in military aviation. I am only a gamer which became interest in these topics because of your wonderful game / module DCS F14.  But why are so different  information in the www. At least from a US Aviation History museum I would expect valid informations regarding F14 and Aim54 . They also stated in their videos that the aim 54 was flying Mach 5  

 Sad to hear from gamer perspective that heatblur has different informations so that the mach5 info seems to be fake news. 

 

Yeah, as @DD_Fenrir points out above.

Specific tests with very specific conditions are often proof of concept, an exploration into what could be if needed, etc. - but not necessarily representative of the in service normal use. Like the NASA Mach 5 test. It proved that it could potentially go mach 5. But in service ofc trading the warhead for speed makes little sense. NASA is naturally less interested in the "killing your target" part and more interested in the "how do things fly, how fast could they fly" etc part. Then ofc these kind of tests are often a bit abused in congress reports, like the mach 4.3 proof of concept, to satisfy committees, to secure funds, etc etc... And lastly, if you achieve a high number of something, it's not unwise to not recall or disprove the claim, because if your enemy thinks "oh wow it does mach 5" that in itself is like a small win. Those are just some reasons, why different testnumbers are being tossed around, which then get a life of their own on the internet. And it is completely normal to fall for them, I did, too, when I first heard about the phoenix. "A mach 5 missile - wow sounds cool." That's very natural and human to go for that.

[Machalot]

1 hour ago, DD_Fenrir said:

An "AIM-54" did reach Mach 5... sort of.

The Mach 5 figure comes from here:

NASA's plan to use an F-15 to launch hypersonic Phoenix missiles - Sandboxx

But when you read just how many modifications NASA made to the missile to get it to reach Mach 5 you start to understand just how little this test missile has in common with a US Navy fleet missile.

1. All of the internal components related to the missile’s guidance system and explosive payload were completely removed, including its guidance computer and radar tracker, leaving just its propulsion and control sections at the rear of the missile intact
2. a new nose with slightly more sloping angles was added to what would now be primary and secondary payload sections with the same 15″ diameter as the original components. The primary payload section measured about 57 inches long and, based on the weight of the guidance section it replaced, could carry approximately 184 pounds worth of testing equipment.

It's an AIM-54 in name only.

Then there's the flight profile - the fact that the launch was committed at Mach 2 and that the missile was not required to make any steering corrections as it had no target widens the gap between it and a fleet Phoenix; there's so little in common here with an operational launch of an AIM-54 that it might as well be a completely different missile.

 

Excellent summary. Also worth pointing out the NASA paper resilts are simulations with no classified input data, not actual test flights. The simulated shots in Figure 7 that reach Mach 3.7-3.8 were launched at M1.2, 45 kft, in a 30-45 deg climb, and the missile does no additional lofting.  The level shot only hits M3.3 and doesn't loft even after launch. So the typical level shot at M1.2, 45 kft with lofting from the missile guidance would fall between those in performance. 

Edited November 23, 2022 by Machalot

[tavarish palkovnik]

This is how it looks in my ''shots'', AIM-54C ballistic flight, 45000ft, 1,2M, 0;30;45 deg, constant thrust of 13595N in 27 seconds

 

 

 

But, thrust of 13595 N is something about what I would like to talk. 

13595*27/163=2252 kg*s/kg ... it means impuls is 2252 or 229,5 s what seams is quite less then regulary used 250 or 245 or 260 etc

One example for start, rocket motor of Neva 5V27 rocket

 

 

 

Fuel in this motor is known, composite with maximal theoretical impulse of 245s at pressure ratio pk:pa=40:1

Diagram says 345300/151,4=2280 kg*s/kg or 232 s what is 5% less then theoretical maximum at 40:1. It can be seen that average pressure in chamber is 50 bar and test is performed at pa=1 bar. So mathematic will say for this ->

I=245*9,81*0,96*0,97*0,98+190,3+76*5-3,058*5^2-7000*0,1+25484*0,1^2=2242 or 228,5 s

In practise, motor with fuel with maximal impulse 245s at 40:1 will give real 228,5 s at 50:1 with combined losses in burning process.

 

What about AIM-54C and these ''shots'' with 13595N of thrust. I have one fuel composition which I believe is or it is close to what is inside. It is fuel with maximal theoretical impulse of 247 s at ratio pk:pa=70:1, density is 1,62 kg/dm^3 and burning rate at 70 bar is 5,3 mm/s. Let's say nozzle throat is 55mm, chamber pressure should be around 4 bar -> 13595/1,4*4/3,14/55^2

I=247*9,81*0,96*0,97*0,98+190,3+76*4-3,058*4^2-7000*0,1+25484*0,1^2=2211 or 225 s

 

So, this motor in practise at sea level will have thrust as result of impulse of cca 225-230s. But these shots are up there at high. Configuration of nozzle is such to allow expansion of gases to the pressure quite less then 1 bar, based on some calculations I'm getting 25000 Pa what is in line of ambient pressure up there. So what exactly should be impulse of motor burning there ->

I=247*9,81*0,96*0,97*0,98+190,3+76*4-3,058*4^2-7000*0,025*25484*0,025^2=2498 or 255 s 

2498*163/27=15080 N or 10% more then at sea level

Higher altitude -> higher impuls

Higher altitude -> higher Cx

In some cases what impuls gives drag takes in roughly same percentage but here taking in consideration nozzle exit and expansion, I belive impuls will give a little more then what drag will take and by that, mach numbers should be more then as presented at the beginning, by my opinion...unless flight is not ballistic but with kinematic overload

 

 

 

 

 

 

 

Edited November 29, 2022 by tavarish palkovnik

[Machalot]

1 hour ago, tavarish palkovnik said:

I=245*9,81*0,96*0,97*0,98+190,3+76*5-3,058-5^2-7000*0,1+25484*0,1^2=2242

Can you provide the general form of this equation? 

In my understanding the DCS aero model does not change the base drag when the rocket motor burns out, and the effect of ambient pressure at the nozzle exit is disabled for the AIM-54 because the nozzle area is set to 1e-6 sq m. 

Even with these limitations, you can go back in this thread and find a benchmark case posted by Klarsnow that shows a velocity match of the current DCS AIM-54(C? I think)  within a few percent of the public NASA paper. 

[tavarish palkovnik]

I will later today...in the meantime just to make one correction, it was with typing mistake ->

I=245*9,81*0,96*0,97*0,98+190,3+76*5-3,058*5^2-7000*0,1+25484*0,1^2=2242  

[tavarish palkovnik]

Base drag is something else and actually not related to this topic which is more about internal ballistic.

Here I made some simple text about base drag or better to say, about difference of drag coefficients in active and passive stages if someone finds interesting ->

But to return on equation, it is empirical equation in form

Ical=Th*Inom+190,3+76*pk-3,058*pk^2-7000*pa+25484*pa^2

pk- chamber pressure (MPa)

pa- ambient pressure (MPa)

Inom- nominal impulse, theoretical impulse, maximal impulse. Fuels of USA origin are usually given in ratio of 70:1 while of Russian origin in ratio 40:1 (40 bars in chamber vs 1 bar ambient)

And Th are coefficients of losses which are always present. I took some maximal values, losses of unburned fuel, losses of friction in nozzle and losses due to effect of condensation phases in exhaust

 

[Katsu]

On 11/29/2022 at 10:29 AM, tavarish palkovnik said:

This is how it looks in my ''shots'', AIM-54C ballistic flight, 45000ft, 1,2M, 0;30;45 deg, constant thrust of 13595N in 27 seconds

 

 

 

But, thrust of 13595 N is something about what I would like to talk. 

13595*27/163=2252 kg*s/kg ... it means impuls is 2252 or 229,5 s what seams is quite less then regulary used 250 or 245 or 260 etc

One example for start, rocket motor of Neva 5V27 rocket

 

 

 

Fuel in this motor is known, composite with maximal theoretical impulse of 245s at pressure ratio pk:pa=40:1

Diagram says 345300/151,4=2280 kg*s/kg or 232 s what is 5% less then theoretical maximum at 40:1. It can be seen that average pressure in chamber is 50 bar and test is performed at pa=1 bar. So mathematic will say for this ->

I=245*9,81*0,96*0,97*0,98+190,3+76*5-3,058*5^2-7000*0,1+25484*0,1^2=2242 or 228,5 s

In practise, motor with fuel with maximal impulse 245s at 40:1 will give real 228,5 s at 50:1 with combined losses in burning process.

 

What about AIM-54C and these ''shots'' with 13595N of thrust. I have one fuel composition which I believe is or it is close to what is inside. It is fuel with maximal theoretical impulse of 247 s at ratio pk:pa=70:1, density is 1,62 kg/dm^3 and burning rate at 70 bar is 5,3 mm/s. Let's say nozzle throat is 55mm, chamber pressure should be around 4 bar -> 13595/1,4*4/3,14/55^2

I=247*9,81*0,96*0,97*0,98+190,3+76*4-3,058*4^2-7000*0,1+25484*0,1^2=2211 or 225 s

 

So, this motor in practise at sea level will have thrust as result of impulse of cca 225-230s. But these shots are up there at high. Configuration of nozzle is such to allow expansion of gases to the pressure quite less then 1 bar, based on some calculations I'm getting 25000 Pa what is in line of ambient pressure up there. So what exactly should be impulse of motor burning there ->

I=247*9,81*0,96*0,97*0,98+190,3+76*4-3,058*4^2-7000*0,025*25484*0,025^2=2498 or 255 s 

2498*163/27=15080 N or 10% more then at sea level

Higher altitude -> higher impuls

Higher altitude -> higher Cx

In some cases what impuls gives drag takes in roughly same percentage but here taking in consideration nozzle exit and expansion, I belive impuls will give a little more then what drag will take and by that, mach numbers should be more then as presented at the beginning, by my opinion...unless flight is not ballistic but with kinematic overload

 

 

 

 

 

 

 

 

if this information turns out to be true it would be interesting to be able to have this thrust correction, since for such a big and heavy missile anything that helps it to have some extra thrust is a big help.

Now it's up to heatblur if this data is valid or not.

[DD_Fenrir]

42 minutes ago, Katsu said:

Now it's up to heatblur if this data is valid or not.

 

On 11/18/2022 at 6:01 PM, IronMike said:

It's hard to stress how close the current missile is to irl performance, likely among some of the closest end-consumer missile simulations we have to date. This is just to underline once more, that we have zero intentions on changing the performance anymore, whatsoever.

 

 

I'd say that about covers it.

 

 

[BubiHUN]

1 hour ago, DD_Fenrir said:

It's hard to stress how close the current missile is to irl performance, likely among some of the closest end-consumer missile simulations we have to date. This is just to underline once more, that we have zero intentions on changing the performance anymore, whatsoever.

 

[DD_Fenrir]

Meme away mate; the issues that trash most otherwise legit Phoenix shots are partly to do with guidance (EDs responsibility), partly to do with the AI's superhuman ability to know exactly where an attacking missile is and defend perfectly (again, EDs responsibility) and partly with a whole lot of users not understanding that Phoenix performs best where the air is thin (cos it's one thicc momma) and expecting it to still be a 100nm shot missile whilst launching it against manoeuvring targets at medium to low altitude (disappointment due to own over-expectations/ignorance).

Edited November 30, 2022 by DD_Fenrir

[BioZ]

Hi all, haven't played DCS for a bit. What is better right now AIM-54C MK60 or MK47? 

And are AIM-54Cs a good opponent for F-16/18/15 at the moment?

[Callsign JoNay]

2 hours ago, BioZ said:

Hi all, haven't played DCS for a bit. What is better right now AIM-54C MK60 or MK47? 

And are AIM-54Cs a good opponent for F-16/18/15 at the moment?

C47 is the best, IMO. Best counter measure resistance, goes active on it's own if you drop a lock/track, and mostly smokeless. It's a good opponent against other Fox-3 carrying targets, but we don't dominate them like we used to. AIM-120C out classes all 54s. The biggest problem is how easily the missile is notched. I still don't understand why the AWG-9 PDSTT can't be notched (perpendicularly), but AIM-54s are so easily notched.

[Naquaii]

20 minutes ago, Callsign JoNay said:

C47 is the best, IMO. Best counter measure resistance, goes active on it's own if you drop a lock/track, and mostly smokeless. It's a good opponent against other Fox-3 carrying targets, but we don't dominate them like we used to. AIM-120C out classes all 54s. The biggest problem is how easily the missile is notched. I still don't understand why the AWG-9 PDSTT can't be notched (perpendicularly), but AIM-54s are so easily notched.

Because one of those we have control over and the other not.

[Hardcard]

9 hours ago, DD_Fenrir said:

Meme away mate; the issues that trash most otherwise legit Phoenix shots are partly to do with guidance (EDs responsibility), partly to do with the AI's superhuman ability to know exactly where an attacking missile is and defend perfectly (again, EDs responsibility) and partly with a whole lot of users not understanding that Phoenix performs best where the air is thin (cos it's one thicc momma) and expecting it to still be a 100nm shot missile whilst launching it against manoeuvring targets at medium to low altitude (disappointment due to own over-expectations/ignorance).

 

Who expects 100nmi hits at medium / low altitude?

Afaik, the main issues are that phoenixes perhaps struggle to accelerate too much and perhaps lose steam too soon.

It would be nice to get a list of sources that HB devs consider to be valid, since there are publicly available documents stating stuff that is being denied here (max speed, for instance).

The phoenix whitepaper says that burn length should be 27 seconds for the mk47 and 30 seconds for the mk60: http://media.heatblur.se/AIM-54.pdf (page 4)

This seems to be consistent with, at least, one of its source documents (approved for public release, unlimited distribution):
 https://apps.dtic.mil/sti/pdfs/ADA142508.pdf (page 14) 

However, that's not what we see in DCS, mk60, for instance, has a burn time of only 21 seconds.

@IronMike @Naquaii
What am I missing here?

Edited December 1, 2022 by Hardcard

[Machalot]

3 hours ago, Hardcard said:

Who expects 100nmi hits at medium / low altitude?

Afaik, the main issues are that phoenixes perhaps struggle to accelerate too much and perhaps lose steam too soon.

It would be nice to get a list of sources that HB devs consider to be valid, since there are publicly available documents stating stuff that is being denied here (max speed, for instance).

The phoenix whitepaper says that burn length should be 27 seconds for the mk47 and 30 seconds for the mk60: http://media.heatblur.se/AIM-54.pdf (page 4)

This seems to be consistent with, at least, one of its source documents (approved for public release, unlimited distribution):
 https://apps.dtic.mil/sti/pdfs/ADA142508.pdf (page 14) 

However, that's not what we see in DCS, mk60, for instance, has a burn time of only 21 seconds.

@IronMike @Naquaii
What am I missing here?

 

In this thread we have a comparison of acceleration (via the velocity curve at zero alpha) between the DCS Phoenix and a NASA simulation based on public data. They match very close.

https://forum.dcs.world/topic/308085-dcs-f-14-development-update-aim-54-phoenix-improvements-overhaul/?do=findComment&comment=5044246

 

The burn times cited in the white paper are superseded by the recent changes to the rocket motor parameters based on new sources.  That is also explained further up in the thread, I think.  The burn times are not that important though, it's matching the total impulse, which seems to be very accurate now. 

Edited December 1, 2022 by Machalot

[Naquaii]

13 hours ago, Hardcard said:

Who expects 100nmi hits at medium / low altitude?

Afaik, the main issues are that phoenixes perhaps struggle to accelerate too much and perhaps lose steam too soon.

It would be nice to get a list of sources that HB devs consider to be valid, since there are publicly available documents stating stuff that is being denied here (max speed, for instance).

The phoenix whitepaper says that burn length should be 27 seconds for the mk47 and 30 seconds for the mk60: http://media.heatblur.se/AIM-54.pdf (page 4)

This seems to be consistent with, at least, one of its source documents (approved for public release, unlimited distribution):
 https://apps.dtic.mil/sti/pdfs/ADA142508.pdf (page 14) 

However, that's not what we see in DCS, mk60, for instance, has a burn time of only 21 seconds.

@IronMike @Naquaii
What am I missing here?

 

Iirc (I'm not the best guy to ask about the missiles) the motor performance of the mk60 was found to be exaggerated. They should be very similar, if slightly different, with one burning with more power but shorter duration and the other longer but weaker. And iirc we could disprove the initial findings that informed our initial mk60 model.

[GGTharos]

15 hours ago, Hardcard said:

Afaik, the main issues are that phoenixes perhaps struggle to accelerate too much and perhaps lose steam too soon.

What is your expectation?   The Phoenix accelerates at around 4g, which means at the top end of the speed it is harder for it to accelerate compared to say a sparrow which accelerates at 10-11g (but doesn't have enough fuel in its booster to reach a high speed).  The sustainer on the sparrow actually can't overcome the drag at low altitudes (it's acceleration would be around 2-3g in vacuum) ... not sure why anyone would expect the fat Phoenix to do anything useful in this case, which is why Phoenixes are meant to be 'death from above' and even then they will hit the brakes.

 

15 hours ago, Hardcard said:

The phoenix whitepaper says that burn length should be 27 seconds for the mk47 and 30 seconds for the mk60: http://media.heatblur.se/AIM-54.pdf (page 4)

This seems to be consistent with, at least, one of its source documents (approved for public release, unlimited distribution):
 https://apps.dtic.mil/sti/pdfs/ADA142508.pdf (page 14) 

However, that's not what we see in DCS, mk60, for instance, has a burn time of only 21 seconds.

Burn time depends on temperature and altitude.   The texts you're reading don't tell you if the number is average burn time, cold/hot burn time, and at which altitude.   DCS also doesn't really model this anyway AFAIK.   The impulse of these rocket motors is nearly the same, so you'll get nearly the same performance.

[tavarish palkovnik]

For start one fun fact…commercial jets which usually (from clear technical reasons) fly at 11km at 0,8M where temperature is roughly -60degC don’t have artificial heating of wings where fuel tanks are located because air friction energy do that job instead.

On burning rate of solid rocket fuels, primarily pressure in rocket motor’s chamber influence. Temperature of fuel grain also. But pressure primarily.

Different temperatures and accordingly variable pressure will make one differential, different pressures and accordingly variable temperature other differential. This other is significantly bigger.

Examples…this is 9M330 motor of TOR, but from time when 9M330 of that time was in form as it was. Later 9M330 and it’s characteristics is changed but what I want to show is not changed.

 

As you can see, total impulse (суммарный импульс as written, for those who struggle with Russian language) is quite close at +50 ; +15 and -50degC

Another one, 5V55 of S-300

тяга двигателя -> motor thrust and this is in k(1000) of kilos. Same story.

And third sample is archaic R-3… 

 

 

 

All right, temperature made influence. Total impulses should be roughly in line but altitudes influence also. Not on burning rate actually but like altitude does, with pressure.

And by the way, this motor, this nozzle has very poor expansion compared to some other “church bells”

[cheezit]

Nb. that DCS also doesn't model the change in Isp based on atmospheric pressure at all.  While a typical missile has a nozzle that's optimized for low altitudes (and thus with a relatively low spread between the sea level ISP and the vacuum ISP), there's still probably a 7% or so difference between a sea level shot and a 35k foot shot that Heatblur can't do anything about, with the difference being a bit more for a higher shot, if we use other SL-optimized solid rocket motors with similar propellant chemistry and known characteristics as a model.  I'm guessing the Isp used in the in-game model is correct for intermediate altitudes, causing the thrust to be slightly higher than it should be for the portion of the motor burn that occurse down low and slightly lower than it should be way up high, bounding the worst-case error, but I'm obviously not privy to that.

 

Another oddity: do the Mk47 Mod 0 and Mk47 Mod 1 really have the same Isp and total impulse despite having different propellant chemistry?  The whitepaper (I know, I know, it's "no longer operative" and whatnot) claimed the Mod 1 had a higher Isp and thrust, the previous in-game model had the Mod 1 with *lower* Isp and thrust, the current in-game model has the Mod 1 and Mod 0 with the same Isp and thrust (unless I'm missing something in the files, in which case please correct me).  It's probably a marginal difference, but it seems really odd that a change to a propellant with a different energy density wouldn't change the total impulse.

[tavarish palkovnik]

On 12/1/2022 at 8:26 PM, cheezit said:

While a typical missile has a nozzle that's optimized for low altitudes

More or less but mostly  everything can be calculated with just simple involving elementary study of rocket motors. Luckily there are plenty of motors sharing same diameter 380mm or 15’’ and having similar composite fuels so even visible differences forcing directions.

All these three are in 380mm

 

 

5V27 of S-125 Neva (SA-3), R-33 and Kh-58…all in same diameter as AIM-54

5V27 is primarily for low and near to intermediate altitudes and nozzle is sized respectively 

 

 

Kh-58 is more for low intermediate and near to low altitudes

This motor’s cross section is my creation but I stand for it very firmly 

And this is Mk47 Mod 0

with nozzle bell sized, obviously not optimized for sea level or intermediate altitudes but for up there altitudes just as R-33 

 

[tavarish palkovnik]

Some mathematics about this...

 

 

 

 

So based on this, I'm getting that this motor is optimized for altitude of 10km with pressure at nozzle exit of 26 kPa. With same principles motor of Neva 5V27 is with pe=80 kPa or optimized at altitude of 2km

[GGTharos]

On 12/1/2022 at 2:26 PM, cheezit said:

Nb. that DCS also doesn't model the change in Isp based on atmospheric pressure at all. 

It used to since I don't know, a decade ago?  Maybe half a decade.   Has it been removed?

[Machalot]

7 hours ago, GGTharos said:

It used to since I don't know, a decade ago?  Maybe half a decade.   Has it been removed?

This may be referring to the fact that the Phoenix has a nozzle exit area of 0.000001 m^2 (apparently at ED's direction), which effectively zeroes out the ambient pressure correction. 

[cheezit]

14 hours ago, Machalot said:

This may be referring to the fact that the Phoenix has a nozzle exit area of 0.000001 m^2 (apparently at ED's direction), which effectively zeroes out the ambient pressure correction. 

I don't think the nozzle_exit_area business is related to this - my understanding is that that coefficient is only used for correcting base drag while the burn is ongoing (less weird turbulence when there isn't a low pressure area right behind the base) and does not affect thrust.  Happy to be corrected if I'm wrong here.

 

 

 

21 hours ago, GGTharos said:

It used to since I don't know, a decade ago?  Maybe half a decade.   Has it been removed?

What coefficients control this for the DCS engine?  I don't see anything with a name that looks related.  Obviously applying a standard curve won't work correctly, as depending on nozzle design, propellant chemistry etc. the difference between sea level and vacuum thrust and Isp can plausibly be anywhere from 1.1x to 2.5x.  If there's some curve being applied that has no coefficients I'll go back and edit my post to correct it, but I've never heard of this before your post.

 

 

 

22 hours ago, tavarish palkovnik said:

Some mathematics about this... [snip to not flood thread]

 

Thanks for running this analysis.  Did you export that from TeX btw?