Current version radar test, PRF disambiguation

When first starting with the FA-18, it is not clear what some of the radar settings do, or what factors will affect detection ranges in DCS. Some factors may be modeled, not modeled, be weird, or change as the game is updated.

I did some tests under various conditions to hopefully make it a little clearer, particularly about things like look-down, closing speed, loadout and PRF setting.

Target is Mirage 2000 in clean configuration. Version 2.5.5.41371 as of March 27, 2020.

Range is the horizontal distance reported by the ruler on the map.

----

Detection range at same altitude (20,000ft):

Hot, Vc 1100kts, MED PRF: 37nm

Hot, Vc 1100kts, HI PRF: 71nm

Hot, Vc 1100kts, HI PRF, Mirage is loaded with bombs and missiles: 71nm

Target beaming, Vc 500kts, MED PRF: 37nm

Target beaming, Vc 500kts, HI PRF: 58nm

Cold, Vc -500kts MED PRF: 24nm

Cold, Vc -500kts HI PRF: 24nm

Detection range while looking down (35,000ft looking down to 1,000ft):

Hot, Vc 1100kts, MED PRF: 47nm

Hot, Vc 1100kts, HI PRF: 47nm

Target beaming, Vc 550kts, MED PRF: 22nm

Target beaming, Vc 550kts, HI PRF: 31nm

Cold, Vc -420kts MED PRF: 16nm

Cold, Vc -400kts HI PRF: 16nm

----

Conclusions:

-Target loadout does not affect detection range (at least for Mirage 2000).

-HI PRF always detects at equal or further range than MED PRF (at least for Mirage 2000).

-HI PRF is way better at detecting targets moving toward you.

-Looking down might slightly reduce detection range, or at least erase the HI PRF advantage. It is hard to tell because looking down also changes the target aspect like beaming does. In any case, it does not have much effect even in this extreme case.

edit: To underscore the importance of PRF, if you are using INTL sometimes when you STT a target the radar will randomly pick MED. If the target is too far away, this will cause the STT lock to be lost, and you have to start over. Strongly recommend exclusively using HI. Keep an eye on it, because sometimes it'll switch back to INTL or MED when you switch modes.

edit: To underscore the importance of PRF, if you are using INTL sometimes when you STT a target the radar will randomly pick LO. If the target is too far away, this will cause the STT lock to be lost, and you have to start over. Strongly recommend exclusively using HI. Keep an eye on it, because sometimes it'll switch back to INTL or LO when you switch modes.

There is a bug/WIP currently in STT where the radar logic won't decide what the best PRF is for the given situation (aspect, range, etc.). So yes, you are correct.

-HI PRF always detects at equal or further range than MED PRF (at least for Mirage 2000).

-HI PRF is way better at detecting targets moving toward you.

-Looking down might slightly reduce detection range, or at least erase the HI PRF advantage.

To my (not expert) knowledge, this is all totally correct for HPRF radar systems, though I'm a bit surprised to see HPRF detecting beaming targets at longer ranges.

However, coming to the conclusion that you should exclusively use HPRF would be incorrect. Because of range ambiguity and nyquist sampling, HPRF has a higher range binning error and does not provide as accurate target range information. Once you're inside the range where MPRF will reliably track a target, switching over to Medium provides the most accurate weapons guidance.

Each mode has it's place. Initial detection at long range is best done with HPRF, while weapon launch is best done with MPRF.

where can I read up on the physics? I'm really impressedd by the knowledge some of you guys exhibit here and would like to read up

To my (not expert) knowledge, this is all totally correct for HPRF radar systems, though I'm a bit surprised to see HPRF detecting beaming targets at longer ranges.

 

However, coming to the conclusion that you should exclusively use HPRF would be incorrect. Because of range ambiguity and nyquist sampling, HPRF has a higher range binning error and does not provide as accurate target range information. Once you're inside the range where MPRF will reliably track a target, switching over to Medium provides the most accurate weapons guidance.

 

Each mode has it's place. Initial detection at long range is best done with HPRF, while weapon launch is best done with MPRF.

I agree with you here, but I think OP made the suggestion in the context of DCS only. Although I have to say that at least for the Mirage, selecting Low PRF allows better detection (not tracking) of cold targets, as it should.

These are interesting findings, it's be good to test with other aircraft as well, see if it's specific to the Hornet or to the Mirage as a target.

To my (not expert) knowledge, this is all totally correct for HPRF radar systems, though I'm a bit surprised to see HPRF detecting beaming targets at longer ranges.

 

However, coming to the conclusion that you should exclusively use HPRF would be incorrect. Because of range ambiguity and nyquist sampling, HPRF has a higher range binning error and does not provide as accurate target range information. Once you're inside the range where MPRF will reliably track a target, switching over to Medium provides the most accurate weapons guidance.

 

Each mode has it's place. Initial detection at long range is best done with HPRF, while weapon launch is best done with MPRF.

no, its definatly wrong.

High PRF due to high powerr output with high Vc and no clutter will provide substantially greater detection distance, though range resolution is poor relstive to mprf due to pulse time. Requires more filtering due to high power, particularly against clutter, thus more susceptible to notching.

Med prf is better at all aspect/lower closure rates and requires less filtering, less susceptible to being notched, better range resolution. This comes at expense of high Vc detect range Vs high prf.

There are too many variables to discuss numbers, and OP did not include any info about range resolution. However, should be clear that, at least according to OP results, proper relative peformance is wrong. RCS of ordijance doesnt appear to be modelled (i've obsserved thisnresting as well - hornet rcs can easily double with moderate payload vs clean).

So many things to improve in the radar modelling. ED needs to make a good compromise because the subject is tricky and hugely difficult to make right, but definetely there are some scripting or tweaking that without going too far would increase playability and better represente real tactical options, some examples for me are:

-RCS multiply factor regarding payload (you could be quite conservative/simplistic and at least increment 20/30 % RCS when you are fully loaded)

-Radar cell resolution: far away target flying very close one to the other could be represented as one.

-SPJ on fighter are not noise jammers, they could be better represented as a 0.x% of probability of breaking lock, being the probability less as you get closer. They should act similar for deceiving radar guided missiles.

Logically MED has to be good for something IRL, if it exists at all.

I'm just talking about when you are jousting in-game and want the lock yesterday, without any glitches. And MED's got half the detection range on hot targets, dang.

Logically MED has to be good for something IRL, if it exists at all.

 

I'm just talking about when you are jousting in-game and want the lock yesterday, without any glitches. And MED's got half the detection range on hot targets, dang.

my late night phone post was a bit of a mess, but med prf is pretty good at everything... high is better at long range detection of high closure rates/frequency shifts, but that's it. And that is just detection, high duty cycle results in significantly more range ambiguity.

Not criticizing - standard WIP disclaimer...

-RCS multiply factor regarding payload (you could b quite conservative/simplistic and at least increment 20/30 % RCS when you are fully loaded)

Not picking but just fyi RCS for modest load can more than double RCS of Hornet.

Also, figured my "its wrong" was a little vague, and implied more was wrong than really is. I cannot provide a source, not here, so you'll just have to take (or not) my word for this.

Hot, Vc 1100kts, MED PRF: 37nm

Hot, Vc 1100kts, HI PRF: 71nm

Hot, Vc 1100kts, HI PRF, Mirage is loaded with bombs and missiles: 71nm

I won't nit pick the difference in detect range except to state the obvious, that, provide the detection occurs at less than the theoretical maximum, larger RCS should be detected sooner. The relative difference between detection ranges i'll defer to game mechanics. A 50% Pd range is not realistic but not going to nitpick.

Target beaming, Vc 500kts, MED PRF: 37nm

Target beaming, Vc 500kts, HI PRF: 58nm

Looks to me like they are applying a proportional benefit to high PRF based on Vc against a clean background. So, same reason as above, chalk this up to game design or whatever.

Cold, Vc -500kts MED PRF: 24nm

Cold, Vc -500kts HI PRF: 24nm

This is where it starts to go wrong. Vc is same as above just opposite vector, yet identical detect range? No reason to even have high prf on at this range, but in terms of radar physics frequency shift occurs on positive and negative closure. Should be no difference here.

Detection range while looking down (35,000ft looking down to 1,000ft):

 

Hot, Vc 1100kts, MED PRF: 47nm

Hot, Vc 1100kts, HI PRF: 47nm

Look down is where it really comes off the rails. HPRF produces more nose that must be filtered but it won't filter 1100 Vc. Absolute value depends on a lot of variables but high PRF, while more susceptible to filtering returns, won't filter 1000+ kts Vc.

Target beaming, Vc 550kts, MED PRF: 22nm

Target beaming, Vc 550kts, HI PRF: 31nm

Bonkers. Vc is almost irrelevant looking down, what matters is relative background. If target's in the notch HPRF won't pick this up at all, MPRF could miss this as well though Vc range would be smaller.

Cold, Vc -420kts MED PRF: 16nm

Cold, Vc -400kts HI PRF: 16nm

Looks consistent with theory above that game is interpreting a difference in negative closure vs. positive, and while there is a difference in radar modes not yet implemented, for purposes this discussion its not relevant. +500 or -500 doppler shift still works, and 500kts is not going to get filtered out at 15 nm. Even where Vc = 0 and there is no frequency shift, radar will pick this up against moving background (earlier in MPRF than HPRF). So this is broke too.

Nice work though, results do seem to suggest source of the problem (or problems, as the case may be).