This chapter aims to provide a broad understanding of the characteristics of the “new” AIM-54 Phoenix against manoeuvring targets. The results are probably the most vivid representation of the effectiveness of the changes to the missile.
Dynamic Scenario
The problem of dynamic scenarios is that the AI can behave differently every time, so I had to concoct a replicable scenario. I admit it looks a bit silly, but the results kind of work.
So far, we have seen that above 35,000ft and below 15,000ft, the Phoenix either behaves similarly to even higher altitudes, or it is employed in conditions highly adverse to its characteristics. Therefore, I limited the tests to 35,000ft, 25,000ft, and 15,000ft.
As you can see from the TacView tracks displayed, the target performs a series of different turns, with the tightest being close to 9G, and accelerations from M.7 to almost M1.6. These are alternated with continuous turns aimed to wear off the energy of the Phoenix by inducing constant drag. The target can also turn cold, and then reverse again.
Results
Recordings of how the Phoenix behaves are available in the video linked above. Different speed and techniques results in different behaviours.
35,000ft
At 35,000ft or 10.7km and long range, the new Phoenix outperforms the older version, regardless of the employment technique or speed. Manually lofting the missile provides even better results, with an impact speed of M2.13 compared to the M1.73 of the old Phoenix. Perhaps even more importantly, the average speed of the updated missile is higher, allowing the missile to impact even more than 20 seconds earlier.
Tacview clearly shows the trajectory of the Phoenix as it chased the target. Due to the range and the altitude, the effects are not as detrimental to the missile as we might expect.
Moving closer, the 60nm set sees the target pulling a 5G turn, and its speed changing from M1.2 to M.7, and then M1.5, before performing another turn in the opposite direction. Nevertheless, the Phoenix did not particularly struggle, and the results of every series are more or less comparable. The only set that stands out is the manual loft, which arrived earlier, faster, and meaner.
The 40nm test shows results comparable to the previous, with one big exception.
YouTube/Patreon timestamps will count as proof. Deadline is the 31/03/2026.
Back to the charts, as mentioned, the trend observed so far is visible once again. It is fascinating to see how the Phoenix seems almost impervious to the manoeuvres of the target. The secret is in its trajectory: since it flies at a very high altitude during the cruise, corrections are not as expensive energy-wise. A non-lofting missile would struggle way more in these scenarios.
Last test at 35,000ft: 20nm. As discussed in the previous parts several times, the old AIM-54 did not loft within 21nm, whereas the new one does. This is already a huge advantage, although we have seen how this matters less at high altitude. However, the scenario was straightforward and static or almost static earlier, whereas in this case, the target is really pulling some Gs. The effect is immediate: the difference between the most performing new Phoenix and the old missile is almost one full Mach. More precisely, M1.72 for the old, and M2.66 for the updated AIM-54. This is, once again, a remarkable way to highlight the importance of the altitude in the Phoenix performance equation.
25,000ft
At 7.6 km or 25,000ft, the behaviour of the Phoenix changes again. As discussed in Part II and Part III of this series, changing the engagement speed and adding manual loft are both effective methods to compensate for the lower altitude reached by supersonic launches. However, as expected, pitching up the Tomcat provides the best results, increasing the impact speed by M.4 over the supersonic launch of the new missile and M.15 over the M.9 employment.
The curves also indicate something different. The old AIM-54 did not connect, and the altitude chart shows that the Phoenix sank like a brick after circa 3 and a half minutes, possibly due to the battery running out. If that’s the case, this is another example of why the new, faster Phoenix is a great improvement.
At 60nm, we see that something odd happened again. The explanation is the same as before, and the offer still stands!
Ignoring the M.9 launch of the new missile, the anomaly is the manual loft data, which provides an absurd amount of additional energy. The impact speed increased by M.6 to M1.7, and the time to impact decreased by 23 seconds compared to the supersonic launch of the new Phoenix.
The 40nm range shows the results almost converging again, besides two main points:
- In primis, we observe another “speed inversion”, so to speak, after circa 50s, where the fastest missile became the slowest and vice versa. Manual loft is not affected at all.
- Next, speaking of manual loft, the altitude curve looks quite peculiar, with a sudden dive onto the target. However, such a manoeuvre did not result in any speed lost by the missile.
Last set of the 25,000ft block: 20nm. The Phoenix performed poorly in this scenario. The old version did not loft, and it rapidly ran out of energy. The M.9 launch of the new missile did not fare much better, as it was dragged by the manoeuvre of the target until it started to fall. The M1.2 and the manually lofted sets instead connected at M1.5 and M1.75, respectively. Surely not much, but it is still a respectable speed given the scenario.
15,000ft
4.5km or 15,000ft is an altitude at which the Phoenix performs poorly in the simpler scenarios discussed before. We can expect it to struggle, and the reason you do not see the M.9 series is that none connected. Ouch. But let’s see what happens. Remember that scenarios should not be compared across altitudes, as the target behaves slightly differently.
At 70m, the results do not look great: every missile failed to connect. The altitude curves are quite interesting, though, as we can see how the WCS tried to salvage altitude when the target turned in order to maintain energy. It did not work.
At 60nm, one missile managed to connect: the manually lofted new AIM-54, finally hitting the target at a speed of M1.3. The outcome is probably caused by the ability of this Phoenix to cruise at higher speeds. Checking both altitude and speed, we see the cost of this feature: a marked dive.
At a closer range, 40nm, once again, grim results. No Phoenix managed to hit its target. However, we see a curiosity: the manual loft set was extremely close to being successful. However, the last hard correction dissipated too much energy, from circa M1.9 at the beginning of the correction, to barely supersonic close to the target. Yes, the drag has been improved, but our missile is a big boy anyway.
Last set, 20nm. No Phoenix connected once again. What drove the missile off is the target’s acceleration from subsonic to M1.5 whilst performing an easy turn towards the beam. The AIM-54 simply lacks the energy to keep up with the target and soon falls into the sea below.
Different altitudes
We have seen what a huge difference employing from 35k makes compared to 15k. High up, the performance is remarkable against several manoeuvres. But what if the target is lower than us? What happens if we launch from 35k and the target is at 15k? How will the Phoenix deal with the prolonged dive? Let’s have a look!
70nm ΔALT
70nm is a very long distance to cover, and the resulting curves are definitely weird. The missiles launched from 35,000ft started well, reaching well into 80,000ft and a speed over M3.6. However, after circa a minute, the Phoenix dived and eventually ran out of energy, dragged by acceleration and the turn of the target. Looking at TacView, the explanation of the missed hit is eloquent. The target pulled an eye-watering 8.7G turn from M1.2 to M.8 and then reached M1.55 at the 90s mark. I doubt any conventional missile can connect in such a situation.
60nm ΔALT
Whilst the target continued its arc and accelerated to M1.5, the second Phoenix was on its way. Out of the five AIM-54 Phoenix launched from 60nm, only one from the co-altitude hit its target, and only when manually lofted. The pair tested from 35,000ft, both connected. This is quite a remarkable achievement, as the target was in a M1.50+ easy left-hand turn.
Tacview shows well how the Phoenix had to correct to compensate for the 8.7G-turn of the target, and then chased it, hitting it after a long chase.
40nm ΔALT
This scenario is not particularly interesting when tested at co-altitude, but it can be useful to have a general idea of how the altitude difference affects the behaviour of the missile.
The results are somewhat surprising, as none of the co-altitude AIM-54s hit their targets. The manually lofted test arrived very close to it, a matter of a few meters, but then it ran out of energy.
20nm ΔALT
At short range, for a Phoenix, that is, the picture changes once again. No missile managed to score a hit in this scenario, no matter if it was launched from a higher altitude or manually lofted.
The target’s behaviour resembles the 60nm test: after a hard turn, this time pulling around 6Gs, the target abruptly accelerated to M1.5. Since the missiles had time to climb and “store” energy, they all got dragged by the turbofan-powered squirrel.
Conclusions
This set of scenarios is definitely unusual, but it helps to better convey the differences between the old and the new AIM-54 Phoenix in more practical terms. The 0TA/ATA geometry, in fact, will likely never happen. Moreover, in US Navy service, the Phoenix was not the primary weapon against fighters until the mid-1980s. It was later greenlit, more or less, as the AIM-54C arrived. In Iranian service, instead, it was freely used, but they had limited numbers. Still, the performance of the Phoenix against highly manoeuvrable targets is not bad at all, and it is worth remembering that the biggest issue the missile encounters is the game itself. Unless the magical AI and Radar Warning Receivers are made more realistic, defeating this weapon will always be too easy.
Leave comments and feedback in the Discussion Forum thread.