Jump to content

On radar performance in DCS full fidelity modules


TLTeo

Recommended Posts



 
Hm... not sure of that. 
 
As for the 68v5. I mean those numbers don't seem terribly wrong, but I also have docs on the APG-66 that put it there too and then the 68 is 33% better than the 66. So its like ish, since it doesn't specify which mode or anything like that for the numbers. The 66 at least has few modes so you can guess. The 68 its harder.
 
And then of course you have the magic KJL-7.... Which IMO should more or less be like a 68v5. 
 


Well, the 742-100 only talks about RWS, TWS and VS, as A/A master modes. Other than those, you have ACM, STT, RAID and auto-acquisition modes.

This document is for the post-MSI APG-65 though, so maybe the APG-73 RUG II is different.

Yeah, I also looked and found info on the APG-66 that says ~30 NM. Then there's the question of the mode, look-up or look-down etc... But at least in DCS, we should be discussing the Viper's RWS and TWS modes, nothing else. If ED wants to add other modes that offer increased detection range, such as LRS, then they should model them properly.

I got no comment on the KJL-7V1, really. It was introduced more recently than the APG-68(V)5, I'd say that it could be more comparable to the V8 or V9. But that's all speculation, as I haven't really spent much time looking into it. It really would depend on how much it improved over the older Phazotron designs it was based on.
  • Like 1

The vCVW-17 is looking for Hornet and Tomcat pilots and RIOs. Join the vCVW-17 Discord.

CVW-17_Profile_Background_VFA-34.png

F/A-18C, F-15E, AV-8B, F-16C, JF-17, A-10C/CII, M-2000C, F-14, AH-64D, BS2, UH-1H, P-51D, Sptifire, FC3
-
i9-13900K, 64GB @6400MHz RAM, 4090 Strix OC, Samsung 990 Pro

Link to comment
Share on other sites

My tip: Dig through Newspaper reportings and other public sources about shots fired in anger by F-16s (say Pakistan) .. look at the distances involved .. come to your own conclusion.

And bare in mind that the USAF didn't upgrade the F-16s they had to the (9) radars, because they were happy with what they had. Is that something that has a range equal to my car reverse-gear distance-warn-beeper? ... I am sceptical.

Why am I defensive about this? Because of things like how Chaff is modelled, because of things like how SAM magically makes RWRs not give a dime is modelled, etc. etc. all of it yells: "I don't understand radar or at least can't explain obvious weird things to others" (which I don't either) ..

 

I wanted to rejoin with a tanker in DCS at 20.000ft (tanker was at 21.000ft) and it was notching me at around 15-20 nm .. I could see it .. it was big .. it was surely reflecting some of that magic radar-stuff.

 

So yeah ..

1.) Whatever ED does .. we will have to love it, change it or leave it.

2.) ED should explain very very detailed why they came to whatever conclusion.


Edited by deadpool

Lincoln said: “Nearly all men can stand adversity, but if you want to test a man's character, give him power."

Do not expect a reply to any questions, 30.06.2021 - Silenced by Nineline

Link to comment
Share on other sites

52 minutes ago, Harker said:


 

 


Well, the 742-100 only talks about RWS, TWS and VS, as A/A master modes. Other than those, you have ACM, STT, RAID and auto-acquisition modes.

This document is for the post-MSI APG-65 though, so maybe the APG-73 RUG II is different.

Yeah, I also looked and found info on the APG-66 that says ~30 NM. Then there's the question of the mode, look-up or look-down etc... But at least in DCS, we should be discussing the Viper's RWS and TWS modes, nothing else. If ED wants to add other modes that offer increased detection range, such as LRS, then they should model them properly.

I got no comment on the KJL-7V1, really. It was introduced more recently than the APG-68(V)5, I'd say that it could be more comparable to the V8 or V9. But that's all speculation, as I haven't really spent much time looking into it. It really would depend on how much it improved over the older Phazotron designs it was based on.

 

 

Yup, the publicly available docs I have from IEEE basically have a chart for 85% detection probability in MPRF vs a "small" target (my guess 3m2 or less), in the 20-30nm range depending on aspect, and there are a bunch of caveats that go into that. So I'd use that as the "floor" for any APG-68 detection range in MPRF. And various sources say its 33% better than the 66, of course not specifying mode or anything else so that gets us to like 30-40 if thats MPRF. And then rule of thumb is HPRF might get you 20% more. 

 

As for the 73 I'm less familiar but I heard a long range search mode mentioned as well.

 

As for the KJL, bottom line there have been no major advancments in radar physics that warrant a huge leap in performance. The older phazatron more or less had like half the ranges listed for the KJL, could it have been improved some, sure... But there is no way the boosted power enough to make a huge difference, the antenna size is what it is. And maybe they got some mileage out of lower noise components or processing, but every western radar would have likely had those as well, they are constantly updated. I recall the conversation I had with an F15 guy, and his comment was software/processing was updated every 6 months in his day (cold war).

 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

50 minutes ago, deadpool said:

My tip: Dig through Newspaper reportings and other public sources about shots fired in anger by F-16s (say Pakistan) .. look at the distances involved .. come to your own conclusion.

And bare in mind that the USAF didn't upgrade the F-16s they had to the (9) radars, because they were happy with what they had. Is that something that has a range equal to my car reverse-gear distance-warn-beeper? ... I am sceptical.

Why am I defensive about this? Because of things like how Chaff is modelled, because of things like how SAM magically makes RWRs not give a dime is modelled, etc. etc. all of it yells: "I don't understand radar or at least can't explain obvious weird things to others" (which I don't either) ..

 

I wanted to rejoin with a tanker in DCS at 20.000ft (tanker was at 21.000ft) and it was notching me at around 15-20 nm .. I could see it .. it was big .. it was surely reflecting some of that magic radar-stuff.

 

So yeah ..

1.) Whatever ED does .. we will have to love it, change it or leave it.

2.) ED should explain very very detailed why they came to whatever conclusion.

 

 

What mode were you in, and yeah in a notch its reflecting energy, but the radar is filtering it out thats the defenition of "the notch". 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

20 minutes ago, Harlikwin said:

 

What mode were you in, and yeah in a notch its reflecting energy, but the radar is filtering it out thats the defenition of "the notch". 

 

Mode: STT (and after lost lock, RWS with a blip there, but "locking it" would just reset the cursor .. )

 

The AN/AWG9 was able to automatically deactivate the MLC filter when the target was above the horizon. 

That what you want to filter out should not be there at this altitude in a look-up condition. Or with range gating .. or whatever ..

So the USAF .. years later .. tossed that capability away? I am unconvinced.


Edited by deadpool

Lincoln said: “Nearly all men can stand adversity, but if you want to test a man's character, give him power."

Do not expect a reply to any questions, 30.06.2021 - Silenced by Nineline

Link to comment
Share on other sites

9 minutes ago, deadpool said:

 

Mode: STT (and after lost lock, RWS with a blip there, but "locking it" would just reset the cursor .. )

 

The AN/AWG9 was able to automatically deactivate the MLC filter when the target was above the horizon. 

That what you want to filter out should not be there at this altitude in a look-up condition. Or with range gating .. or whatever ..

So the USAF .. years later .. tossed that capability away? I am unconvinced.

 

 

Nah, notching shouldn't work very well in lookup, but its a function of main lobe clutter, which in lookdown is the main issue, in look up its significantly less depending on how far above the horizon as I understand it and also ownship altitude but its not entirely gone. DCS models this very simply as above horizon or not AFAIK so not sure why it was doing that maybe its more complicated. 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

On 6/11/2021 at 8:57 AM, GumidekCZ said:

When testing ranges, I also come to one weird thing, and thats upper hemisphere detection with AN/APG-73 is NOT automaticly striped of Doppler speed limits. Which are now 100kph or 27,78m/s.

With such "modern" radar which also served shortly on Super Hornet, I would expect some automatic function like MLC AUTO in F-14 tomcat had many years before APG-73.

Link to comment
Share on other sites

14 minutes ago, GumidekCZ said:
On 6/11/2021 at 12:57 AM, GumidekCZ said:

When testing ranges, I also come to one weird thing, and thats upper hemisphere detection with AN/APG-73 is NOT automaticly striped of Doppler speed limits. Which are now 100kph or 27,78m/s.

With such "modern" radar which also served shortly on Super Hornet, I would expect some automatic function like MLC AUTO in F-14 tomcat had many years before APG-73.

 

Yeah, thats weird broken if its notchable in lookup, but how much look up are we talking? 3 degrees above the horizon, or 45? The second half of that 100kph is lik 53kts which is about right from what I've read. 


Edited by Harlikwin

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

21 minutes ago, Harlikwin said:

 

Nah, notching shouldn't work very well in lookup, but its a function of main lobe clutter, which in lookdown is the main issue, in look up its significantly less depending on how far above the horizon as I understand it and also ownship altitude but its not entirely gone. DCS models this very simply as above horizon or not AFAIK so not sure why it was doing that maybe its more complicated. 

 

From what little I understand distant ground clutter is also less of a problem. Look up would be mainly influenced by bloomed chaff and ECM, I understand.

1 minute ago, Harlikwin said:

 

Yeah, thats weird broken if its notchable in lookup, but how much look up are we talking? 3 degrees above the horizon, or 45? The second half of that 100kph is lik 53kts which is about right from what I've read. 

 

 

15-20nm distance, he at 21000 me at 20000 .. I'd need to do trigonometry .. but it should be workable .. but especially with the curvature .. ground was a nofactor .. I have little faith in what ED passes for radar stuff tbh.

Lincoln said: “Nearly all men can stand adversity, but if you want to test a man's character, give him power."

Do not expect a reply to any questions, 30.06.2021 - Silenced by Nineline

Link to comment
Share on other sites

9 minutes ago, deadpool said:

 

From what little I understand distant ground clutter is also less of a problem. Look up would be mainly influenced by bloomed chaff and ECM, I understand.

 

15-20nm distance, he at 21000 me at 20000 .. I'd need to do trigonometry .. but it should be workable .. but especially with the curvature .. ground was a nofactor .. I have little faith in what ED passes for radar stuff tbh.

 

Yeah, as I understand it the general rule of thumbs are radar range goes up the higher you are (in look up), but gets sort of worse in look down. 

 

I mean I'd take that as co-alt situation so if your radar is centered its gonna get some returns from clutter since its a "cone" I'd imagine, so IDK. And yeah who knows how ED is handling it. 

 

As for Chaff and ECM thats a whole other story, Chaff in DCS is just a "radar flare" the way it works now, not realistic at all. 

 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

On 6/9/2021 at 4:35 PM, dundun92 said:

Another note on the F-16 MPRF, and why its MPRF is so good. Because it doesnt use HPRF, they can get away with quite high peak powers compated to HPRF radars while keeping average power low. For reference, the APG-66 operates at a 21.5 kW peak power, and the 68 is at 17.5 kW. This contrasts to the AGP-63 which is 13 kW, with the AWG-9 in the same range. APG-65 is around 5 kW (hard to find an exact number though), as is the N001. So thats part of the tradeoff of optimizing for MPRF.

MPRF will never match the detection range performance of HPRF.. Ever.

 

A typical MPRF waveform only get to integrate 1/8 of the total 'time on target' due to the use 8 separate PRFs (typically). Each of the those PRFs gets sampled/integrated individually. 

 

HPRF (with FMR for ranging; ie STT, RWS, TWS) gets to integrate about 1/3 of the total 'time on target' since it only has three segments in its waveform. (3 stage FMR)

 

On average, MPRF waveforms transmit about 20,000 pulses per second and HPRF about 200,000 pulses per second. Thus, HPRF can integrate many more pulses given the same integration period as MPRF. But as I explained above, HPRF has much longer integration periods than MPRF.

 

Without going into detail, a HPRF waveform will integrate on the order of many thousands of pulses, while MPRF will integrate on the order of hundreds of pulses.

  • Like 2
  • Thanks 2
Link to comment
Share on other sites

1 minute ago, Beamscanner said:

MPRF will never match the detection range performance of HPRF.. Ever.

 

A typical MPRF waveform only get to integrate 1/8 of the total 'time on target' due to the use 8 separate PRFs (typically). Each of the those PRFs gets sampled/integrated individually. 

 

HPRF (with FMR for ranging; ie STT, RWS, TWS) gets to integrate about 1/3 of the total 'time on target' since it only has three segments in its waveform. (3 stage FMR)

 

On average, MPRF waveforms transmit about 20,000 pulses per second and HPRF about 200,000 pulses per second. Thus, HPRF can integrate many more pulses given the same integration period as MPRF. But as I explained above, HPRF has much longer integration periods than MPRF.

 

Without going into detail, a HPRF waveform will integrate on the order of many thousands of pulses, while MPRF will integrate on the order of hundreds of pulses.

I know, my comparison was not HPRF to MPRF, but rather the high peak power MPRF of the APG-68 to the lower peak MPRF of other radars, specifically the APG-73

Eagle Enthusiast, Fresco Fan. Patiently waiting for the F-15E. Clicky F-15C when?

HP Z400 Workstation

Intel Xeon W3680 (i7-980X) OC'd to 4.0 GHz, EVGA GTX 1060 6GB SSC Gaming, 24 GB DDR3 RAM, 500GB Crucial MX500 SSD. Thrustmaster T16000M FCS HOTAS, DIY opentrack head-tracking. I upload DCS videos here https://www.youtube.com/channel/UC0-7L3Z5nJ-QUX5M7Dh1pGg

 

Link to comment
Share on other sites

20 minutes ago, dundun92 said:

I know, my comparison was not HPRF to MPRF, but rather the high peak power MPRF of the APG-68 to the lower peak MPRF of other radars, specifically the APG-73

This means nothing if you dont know the duty cycle. 

 

Peak power isnt even in the radar range equation. Average power is. If your radar has 20kw peak power at a 10% duty cycle, and my radar has 15kw peak power at a 20% duty cycle. My average power (3kw) will be higher than your average power (2kw). 

 

Duty cycle can be calculated via Pulse Duration/Pulse Repetition Interval. Good luck finding those for the -68 and -73. 

 

Also, power is one of the least significant variables to detection range. A common saying in the radar community is, "If you want to double your range, but you can only control one variable, you would need to increase your output power by 16x. But only 4x for antenna gain."

 

At the end of the day, antenna size, integration count, and sidelobe cancelation are more important than total output power. Especially if you're only talking about a few extra kw. 


Edited by Beamscanner
  • Like 2
Link to comment
Share on other sites

3 hours ago, Beamscanner said:

This means nothing if you dont know the duty cycle. 

 

Peak power isnt even in the radar range equation. Average power is. If your radar has 20kw peak power at a 10% duty cycle, and my radar has 15kw peak power at a 20% duty cycle. My average power (3kw) will be higher than your average power (2kw). 

 

Duty cycle can be calculated via Pulse Duration/Pulse Repetition Interval. Good luck finding those for the -68 and -73. 

 

Also, power is one of the least significant variables to detection range. A common saying in the radar community is, "If you want to double your range, but you can only control one variable, you would need to increase your output power by 16x. But only 4x for antenna gain."

 

At the end of the day, antenna size, integration count, and sidelobe cancelation are more important than total output power. Especially if you're only talking about a few extra kw. 

 

 

I was hoping you'd show up.

 

so question... The other part where it seems like you'd get some bang for your buck is the sensitivity of the receiver. Is it known how much of a difference you could expect from say something built with different generations of components i.e. GaS components vs GaN or so forth? Or put another way how much more sensitive would something built in the 80's or 90's be vs the 2000's.

 

Similar question on signal proc, like how much more does having 80's vs 90s 2000's hardware/sofware do for you on that end?

 

 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

6 hours ago, Harlikwin said:

 

I was hoping you'd show up.

 

so question... The other part where it seems like you'd get some bang for your buck is the sensitivity of the receiver. Is it known how much of a difference you could expect from say something built with different generations of components i.e. GaS components vs GaN or so forth? Or put another way how much more sensitive would something built in the 80's or 90's be vs the 2000's.

 

Similar question on signal proc, like how much more does having 80's vs 90s 2000's hardware/sofware do for you on that end?

 

 

Sensitivity does improve over the decades, but increased sensitivity doesn't mean much if you dont improve the signal to noise ratio. the "sensitive" part of the modern digital receiver is the ADC. They have gotten more sensitive over the decades. But this benefit is useless if the background noise remains loud. Thus DSP noise reduction techniques are required to take any advantage of a more sensitive ADC. 

 

Phase coded pulses (pulse compression) can be used to increase your SNR via a phase match filter, which can add upwards of 13dB of gain. However, those are mainly used with long range MPRF modes. Not HPRF. So if ur radar has HPRF, pulse compression wont be the deciding factor of your max detection range (HPRF > MPRF with phase coded pulses).

 

Larger FFTs (integrations) can increase SNR. For every pulse coherently integrated (pulse doppler radar), your gain increases by ~1dB. About .3 dB per pulse integrated non-coherently (non-pulse doppler radar).

 

Sidelobe cancelation increases SNR.

 

Minimizing signal loss can increase gain. 

 

Radar cooling reduces heat at the receiver, decreasing noise, thus increasing SNR. 

 

GaAs and GaN are semiconductors that perform well at high power with microwave (and millimeter wave) frequencies. I dont think they directly improve sensitivity per say. But they certainly enable AESA architectures which can employ advanced signal processing (DBF, TBD, adaptive nulling, STAP, etc) and provide low receiver loss since the ADCs are located close to the antenna. (waveguide antennas, like planar array and PESA, experience a lot of loss due to waveguide frequency response curves, internal reflection (non-perfect coupling), heat, and path loss) So in a sense, they indirectly improve receiver performance. 

 

 

A pretty cool radar with a VERY sensitive receiver I know of is a actually an analog photonic radar. This radar doesn't down-convert the signal at all, thus reducing a lot of noise. The microwave energy is upconverted to near optical light, and processed via beam-slitters, fiber optics, etc.

 

 

  • Like 2
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...