reported AN/APG-63 range is under-represented

[Harlikwin]

2 hours ago, nighthawk2174 said:

Maybe maybe not we don't know for sure, for now the range is fine, minus the MPRF thing.

 

What we do know is that the APG-73 has a larger dish aperture, and likely has more power available, and more cooling. Yet the Jeff outranges it pretty dramatically. No one with any basic knowledge of radar thinks thats actually credible. FWIW take a look at scale drawings of an F16 and JF17 and tell me which nose is bigger.

 

 

[Mike_Romeo]

There is a interview with a Pakistan JF-17 pilot who says that he could lock on a Su-30 "beyond 50-60NM" which are more or less 105km.



https://hushkit.net/2019/07/19/flying-fighting-in-the-jf-17-thunder-interview-with-pakistan-air-force-fighter-pilot/

PS: Its a very intrestring interview.

Edited October 4, 2021 by Mike_Romeo

[nighthawk2174]

12 hours ago, Harlikwin said:

 

What we do know is that the APG-73 has a larger dish aperture, and likely has more power available, and more cooling. Yet the Jeff outranges it pretty dramatically. No one with any basic knowledge of radar thinks thats actually credible. FWIW take a look at scale drawings of an F16 and JF17 and tell me which nose is bigger.

 

 

tbf the current DCS implementation of the -73 is almost certainly shorter ranged than it should be.

[Harlikwin]

3 minutes ago, nighthawk2174 said:

tbf the current DCS implementation of the -73 is almost certainly shorter ranged than it should be.

 

I mean this is gonna become a big problem for DCS unless its sorted out, maybe not in the absolute sense, but in the sense that we need some rough rank ordering of performance. 

7 minutes ago, Mike_Romeo said:

There is a interview with a Pakistan JF-17 pilot who says that he could lock on a Su-30 "beyond 50-60NM" which are more or less 105km.



https://hushkit.net/2019/07/19/flying-fighting-in-the-jf-17-thunder-interview-with-pakistan-air-force-fighter-pilot/

PS: Its a very intrestring interview.

 

 

Missing any actual details like target aspect or the radar mode used. Could have just been VS on a hot target (most likely IMO). And well, pilots lie. 

[KenobiOrder]

1 hour ago, Harlikwin said:

Could have just been VS on a hot target (most likely IMO

Most likely not. It states greater than or equal to 105km, ranges over that probably being VS. If your going to assume that publically listed max ranges are for vs, you have to do that for the 68 as well.

 

It's not the jf17 that's overmodeled, it the hornet and viper that are under modeled.

1 hour ago, Harlikwin said:

or the radar mode used

You can't lock on in VS

[henshao]

Let's not forget the Su-30 in reality, especially with stores is going to have a colossal RCS, probably at least 30. I think this is at least half of the problem with DCS radar implementation. Yes a hornet, jf-17, f-16 whatever might "only" be able to see a 5 m² aircraft from ~50 miles but very things worth worrying about are actually below that 5 m² rcs mark and even then only clean. Hang a bunch of nice reflective pylons missiles and fuel tanks etc off a nominally 20m² Flanker and even an F-16 should be able to see it from the moon, let alone a Tomcat or Eagle. But I digress

[nighthawk2174]

13 hours ago, henshao said:

Let's not forget the Su-30 in reality, especially with stores is going to have a colossal RCS, probably at least 30. I think this is at least half of the problem with DCS radar implementation. Yes a hornet, jf-17, f-16 whatever might "only" be able to see a 5 m² aircraft from ~50 miles but very things worth worrying about are actually below that 5 m² rcs mark and even then only clean. Hang a bunch of nice reflective pylons missiles and fuel tanks etc off a nominally 20m² Flanker and even an F-16 should be able to see it from the moon, let alone a Tomcat or Eagle. But I digress

From the front missiles are relatively small RCS targets but yes 8 R-27's and you'd probably add a few m^2 from the front and a whole lot more from the side.  The su30 itself I agree shouldn't be 5m^2 far closer to 15-20 while clean.  Maybe closer to 20-30 from the front when fully loaded, probably tending towards the bottom of those ranges for average numbers over the frontal arch.

Edited October 5, 2021 by nighthawk2174

[okopanja]

On 10/5/2021 at 4:41 AM, nighthawk2174 said:

From the front missiles are relatively small RCS targets but yes 8 R-27's and you'd probably add a few m^2 from the front and a whole lot more from the side.  The su30 itself I agree shouldn't be 5m^2 far closer to 15-20 while clean.  Maybe closer to 20-30 from the front when fully loaded, probably tending towards the bottom of those ranges for average numbers over the frontal arch.

Very funny.

And F16/F18F15 would also have invisible fuel tanks which do not inrease RCS?

 

[Airhunter]

54 minutes ago, okopanja said:

Very funny.

And F16/F18F15 would also have invisible fuel tanks which do not inrease RCS?

 

 

No one said that. Stores should all have an effect on RCS, but the baseline RCS of large fighters (14, 15, Flankers) should all be way greater than 5-10m^2. This is the most limiting factor when it comes to various radar detection ranges. DCS also only has one fixed value for RCS, frontal aspect and no stores - really the limiting factor here is the engine itself. 

[okopanja]

On 10/4/2021 at 6:10 PM, Mike_Romeo said:

https://hushkit.net/2019/07/19/flying-fighting-in-the-jf-17-thunder-interview-with-pakistan-air-force-fighter-pilot/

PS: Its a very intrestring interview.

There is lots of national pride there, and it is questionable if the interview was made with a real pilot. At the times it sounded as if teenager was giving this interview describing has newest gaming console. Sorry I can not accept that somebody mature would be entrusted with $30M+ machine.

 

On the other side...

On 10/4/2021 at 5:49 AM, Harlikwin said:

What we do know is that the APG-73 has a larger dish aperture, and likely has more power available, and more cooling. Yet the Jeff outranges it pretty dramatically. No one with any basic knowledge of radar thinks thats actually credible. FWIW take a look at scale drawings of an F16 and JF17 and tell me which nose is bigger.

I feel that comparing APG-73 with KLJ-7 is not an easy task since they belong to different era:

1. APG-65

- is the product of late 70s, with AGP-73 being according to Wikipedia:

a late 1980s "upgrade of the APG-65 that provides higher throughputs, greater memory capacity, improved reliability, and easier maintenance".

- weights 154kg, has volume of 0,126 m³ without Antenna

Pictures:

 

 

2. KLJ-7

- is a product of mid 2000. More than 15 years of difference for basic building pieces! Just thing how much CPUs improved in that time range!!! E.g. top off the line CPUs did have 500x more discrete elements than counterparts from late 80s. We are talking about jump of 80-100MHz to well above 1.5 GHz  within this time period. (Compare your old 486, with e.g. P3/Centrino/Intel P4)

- weights less than 120kg in total

- if AESA version was used in incident, this is an extra point for superiority over AGP-65/73

Pictures:

https://www.pakdefense.com/wp-content/uploads/2020/01/PAKISTAN-JF-17-Thunder-Block-III-AESA-Radar-700x467.jpg

 

More here:

https://www.pakdefense.com/blog/pakistan-air-force/pakistan-finalized-chinese-klj-7-aesa-radar-for-initial-batch-of-jf-17-block-iii-aircraft/

 

So when you consider all of this, it should be concluded that KLJ-7 easily outperforms the the APG-73 easily due to much newer technology base. And this is not only raw CPU processing power, but also advances made in Antenna design. The size of a cone does not mean much in this case, since it is obvious the KLJ-7 has much lower volume, likely way less needs for cooling and sits well behind in respect to the cone.

 

[Harlikwin]

2 hours ago, okopanja said:

There is lots of national pride there, and it is questionable if the interview was made with a real pilot. At the times it sounded as if teenager was giving this interview describing has newest gaming console. Sorry I can not accept that somebody mature would be entrusted with $30M+ machine.

 

On the other side...

I feel that comparing APG-73 with KLJ-7 is not an easy task since they belong to different era:

1. APG-65

- is the product of late 70s, with AGP-73 being according to Wikipedia:

a late 1980s "upgrade of the APG-65 that provides higher throughputs, greater memory capacity, improved reliability, and easier maintenance".

- weights 154kg, has volume of 0,126 m³ without Antenna

Pictures:

 

 

2. KLJ-7

- is a product of mid 2000. More than 15 years of difference for basic building pieces! Just thing how much CPUs improved in that time range!!! E.g. top off the line CPUs did have 500x more discrete elements than counterparts from late 80s. We are talking about jump of 80-100MHz to well above 1.5 GHz  within this time period. (Compare your old 486, with e.g. P3/Centrino/Intel P4)

- weights less than 120kg in total

- if AESA version was used in incident, this is an extra point for superiority over AGP-65/73

Pictures:

https://www.pakdefense.com/wp-content/uploads/2020/01/PAKISTAN-JF-17-Thunder-Block-III-AESA-Radar-700x467.jpg

 

More here:

https://www.pakdefense.com/blog/pakistan-air-force/pakistan-finalized-chinese-klj-7-aesa-radar-for-initial-batch-of-jf-17-block-iii-aircraft/

 

So when you consider all of this, it should be concluded that KLJ-7 easily outperforms the the APG-73 easily due to much newer technology base. And this is not only raw CPU processing power, but also advances made in Antenna design. The size of a cone does not mean much in this case, since it is obvious the KLJ-7 has much lower volume, likely way less needs for cooling and sits well behind in respect to the cone.

 

 

Ok, I'll play.

 

The APG-73, much like the APG-63 was continuously upgraded over its lifetime. The version we have is from the very late 90's. And things like processors sit on LRU's, line replaceable boards/units. Fun fact, the APG-63 didn't actually have TWS capability till the mid-late 80's. 

 

Talking about the KJL-7

 

Here is the radar equation for range. 

 

Terms can be found here: https://www.radartutorial.eu/01.basics/The Radar Range Equation.en.html

 

So in DETAIL, explain to me where having a 386 vs a pentium is making a massive difference in the RANGE/raw detection capability of a radar? And moreover, why does no actual Radar textbook talk about that specifically, and why isn't it a major term in the radar equation?

 

As for antennas, there have been no major advances in antenna design technology since NEC methods have become common place in the 70's and 80's. If you want to point me to the specific research paper from 2000-2007 or so that explains that "magic" advance in the design of planar array antennas I'm more than happy to read it. 

 

That being said antenna size is the dominant part of the radar equation that effects detection range. Thats why the F15, F14, SU-27 all have huge ~1 meter antennas. Or maybe those designers were idiots and didn't need to have massive antennas, and could have dramatically reduced their airframe drag by having a smaller cross section. 

 

This might be helpful, its from a textbook on radar physics. 

 

I don't see 386 or 486 or pentium mentioned there.

 

In conclusion. I certainly do not believe that the KJL has more range than a hornet radar, IMO it should be maybe on par with APG-68. The APG-73 has a significantly larger radar dish diamater (the dominant term in the equation as it scales with only the square root, not the 4th root like everything else). And being a much larger fighter the Hornet can generate more power for that radar, and likely provide more cooling for that radar as well which is an important consideration when alot of the energy put into those systems is lost as waste heat. 120kg, most likely means its air cooled much like the APG-66/68, not water cooled like the APG-73 as well which again limits peak output power. 

 

TLDR?

"Processing power" and "components" to explain "range" is some pure handwavium explanation that ignores all basic radar physics. 

 

 

 

Edited October 10, 2021 by Harlikwin

[henshao]

Look at it another way, the KLJ-7 is way newer than the AWG-9 but we know for sure which of them can detect a fighter further

[Harlikwin]

8 minutes ago, henshao said:

Look at it another way, the KLJ-7 is way newer than the AWG-9 but we know for sure which of them can detect a fighter further

 

Size matters.

 

Edited October 10, 2021 by Harlikwin

[okopanja]

3 hours ago, Harlikwin said:

Ok, I'll play.

Sorry but I have to respectable decline to play.

 

The radar equation that you posted, should be actually applied multiple times in the case of AESA depending on how many tiny antennas you have in place of the traditional dish. Typically these are large numbers 1000-2000+ elements. Each of them needs to be processed in order to derive the tracks. Needles to say this means lots of real-time processing is needed, hence my claim on the CPU power and Moore's law. Similar techniques were applied in astronomy for long time, with benefit that they did not need to processes the signals in realtime.

 

3 hours ago, Harlikwin said:

Size matters.

Size mattered in the past, the number of discrete elements is what matters today.

 

This means reduction in power requirements (thus cooling requirements as well), weight, size. AESA can be smaller, lighter and also mounted in other places than the nose of aircraft. I posted the pictures to show how tiny KLJ-7 is compared to APG-65/73.

 

Simply put: AESA radars have made virtually everything prior to them obsolete. The idea behind the operating principle was know for a long time, what was missing was the actual technology to build them to be of practical use.

 

The radar game has been played for a long time, and no doubt it will not end with AESA radars.

 

For the end I found this neat calculator which can be used to estimate the AESA capability, written by someone who understands the topic better than me:

https://www.researchgate.net/publication/341764957_AESA_Radar_Calculator_ver2

 

[henshao]

4 minutes ago, okopanja said:

Sorry but I have to respectable decline to play.

 

The radar equation that you posted, should be actually applied multiple times in the case of AESA depending on how many tiny antennas you have in place of the traditional dish. Typically these are large numbers 1000-2000+ elements. Each of them needs to be processed in order to derive the tracks. Needles to say this means lots of real-time processing is needed, hence my claim on the CPU power and Moore's law. Similar techniques were applied in astronomy for long time, with benefit that they did not need to processes the signals in realtime.

 

Size mattered in the past, the number of discrete elements is what matters today.

 

This means reduction in power requirements (thus cooling requirements as well), weight, size. AESA can be smaller, lighter and also mounted in other places than the nose of aircraft. I posted the pictures to show how tiny KLJ-7 is compared to APG-65/73.

 

Simply put: AESA radars have made virtually everything prior to them obsolete. The idea behind the operating principle was know for a long time, what was missing was the actual technology to build them to be of practical use.

 

The radar game has been played for a long time, and no doubt it will not end with AESA radars.

 

For the end I found this neat calculator which can be used to estimate the AESA capability, written by someone who understands the topic better than me:

https://www.researchgate.net/publication/341764957_AESA_Radar_Calculator_ver2

 

 

Ok which jet has an AESA in DCS

[Harlikwin]

1 hour ago, okopanja said:

Sorry but I have to respectable decline to play.

 

The radar equation that you posted, should be actually applied multiple times in the case of AESA depending on how many tiny antennas you have in place of the traditional dish. Typically these are large numbers 1000-2000+ elements. Each of them needs to be processed in order to derive the tracks. Needles to say this means lots of real-time processing is needed, hence my claim on the CPU power and Moore's law. Similar techniques were applied in astronomy for long time, with benefit that they did not need to processes the signals in realtime.

 

Size mattered in the past, the number of discrete elements is what matters today.

 

This means reduction in power requirements (thus cooling requirements as well), weight, size. AESA can be smaller, lighter and also mounted in other places than the nose of aircraft. I posted the pictures to show how tiny KLJ-7 is compared to APG-65/73.

 

Simply put: AESA radars have made virtually everything prior to them obsolete. The idea behind the operating principle was know for a long time, what was missing was the actual technology to build them to be of practical use.

 

The radar game has been played for a long time, and no doubt it will not end with AESA radars.

 

For the end I found this neat calculator which can be used to estimate the AESA capability, written by someone who understands the topic better than me:

https://www.researchgate.net/publication/341764957_AESA_Radar_Calculator_ver2

 

 

You do realize the kjl we have in game isn't an AESA? It's an apples to oranges comparison then. Maybe AESA kjl vs APG79 which also would win and again for the similar reasons. And guess what, you can fit more elements on a larger dish than a smaller one so size still matters. 

 

And really the realistic comparison 2007 DCS aside would be a mech scanned KJL vs a super hornet with an APG79. 

Edited October 10, 2021 by Harlikwin

[GGTharos]

23 minutes ago, okopanja said:

Size mattered in the past, the number of discrete elements is what matters today.

 

Same difference, and irrelevant to the subject.  The modeled KLJ-7 is an MSA.

 

23 minutes ago, okopanja said:

This means reduction in power requirements (thus cooling requirements as well), weight, size. AESA can be smaller, lighter and also mounted in other places than the nose of aircraft. I posted the pictures to show how tiny KLJ-7 is compared to APG-65/73.

 

No, AESA requires more power and more cooling for the range advantage that it provides.  The whole thing continues to follow physics as known

Edited October 10, 2021 by GGTharos

[okopanja]

Just now, Harlikwin said:

You do realize the kjl we have in game isn't an AESA? It's an apples to oranges comparison then. Maybe AESA kjl vs APG79 which also would win and again for the same reasons. 

 

4 minutes ago, henshao said:

Ok which jet has an AESA in DCS

None yet, but I would suggest you simply calculate the APG-73 range and argue based on that.

[Harlikwin]

2 hours ago, okopanja said:

 

None yet, but I would suggest you simply calculate the APG-73 range and argue based on that.

 

I'm not even sure what you are arguing about this point.

 

The original argument was that currently the JF-17's KJL is overperforming. 6 months ago it was borderline, but since then the Hornet radar got adjusted, and the Viper radar is now adjusted down to more realistic ranges as well, while the JF-17 is still sitting at the old values which were on par with the then overperforming APG-68. So All deka really needs to do is to put the KJL around the 68 again.

 

You seem to think the fact that the current Jeff happens to maybe have an AESA radar circa the late 20teens is somehow relevant in DCS? Or perhaps the real world? In the real world the US had AESA radars 20 years prior and has mastered the tech, whereas China and other nations are playing catch-up. 

 

Maybe your argument is the jeff is somehow relevant to a modern battlefield? Its not, it will be brutally murdered by 5th or soon to come 6th gen fighters cuz its basic tech is at least 30 years out of date IRL. The reason any nation buys it is because they can't buy anything better, and can't develop anything better on their own.

 

 

 

 

Edited October 10, 2021 by Harlikwin

[KenobiOrder]

The problem is the modeling of the APG-73 and 68, not the KLJ-7. Especially the 73. It has the same range as the 68 right now which seems just as silly assuming the dish size argument

[Harlikwin]

13 minutes ago, KenobiOrder said:

The problem is the modeling of the APG-73 and 68, not the KLJ-7. Especially the 73. It has the same range as the 68 right now which seems just as silly assuming the dish size argument

 

It is a bit more complex than that but yes I think in HPRF modes the 73 should be more. The 66 and 68 were more optimized for MPRF with HPRF being a "bolt" on mode for the V5 which reputedly didn't work well.

 

Edited October 12, 2021 by Harlikwin

[henshao]

Not only dish size, but even PRF mode. Even optimized for MPRF, it should not match an HPRF radar all else being equal. It's just physics.

 

all in all between the various over and under performing radars in the game it is strange to me they need decades to sort this kind of simple thing out

[okopanja]

On 10/10/2021 at 9:19 PM, Harlikwin said:

You seem to think the fact that the current Jeff happens to maybe have an AESA radar circa the late 20teens is somehow relevant in DCS? Or perhaps the real world? In the real world the US had AESA radars 20 years prior and has mastered the tech, whereas China and other nations are playing catch-up. 

 

Maybe your argument is the jeff is somehow relevant to a modern battlefield? Its not, it will be brutally murdered by 5th or soon to come 6th gen fighters cuz its basic tech is at least 30 years out of date IRL. The reason any nation buys it is because they can't buy anything better, and can't develop anything better on their own.

I can not be sure what was before, since I started flying recently. Flying with "air superiority" Su-27 shocked me, but F-15c shocked me even more. Maybe I did not try hard enough, but I could really not make F-15c work for me in the modern environment.

 

However, it was an interesting speculation if the JF17 in the incident mentioned by @Mike_Romeo was indeed flying with AESA or not. Perhaps it was a bit offtopic to discuss it here.

 

Btw @GGTharos was right: the thing with the AESA is that it produces lots of heat. On average 4 times energy goes into the heat compared to useful signal. Its actual performance depends largely on the spare cooling capacity. It appears none of them operate nowhere near theoretical full capacity. Here is an interesting paper (F-16 discussed) to read and I think it complements nicely the calculators: https://www.researchgate.net/publication/337985199_On_the_use_of_AESA_Active_Electronically_Scanned_Array_Radar_and_IRST_InfraRed_SearchTrack_System_to_Detect_and_Track_Low_Observable_Threats/link/5df8fdb7299bf10bc3633791/download

 

IMHO: in future radars designers will probably try to find the substitute for GaAs TRMs, which develops less heat and/or they may focus on advances in cooling systems (e.g. graphene, etc).

 

[Harlikwin]

2 hours ago, okopanja said:

I can not be sure what was before, since I started flying recently. Flying with "air superiority" Su-27 shocked me, but F-15c shocked me even more. Maybe I did not try hard enough, but I could really not make F-15c work for me in the modern environment.

 

However, it was an interesting speculation if the JF17 in the incident mentioned by @Mike_Romeo was indeed flying with AESA or not. Perhaps it was a bit offtopic to discuss it here.

 

Btw @GGTharos was right: the thing with the AESA is that it produces lots of heat. On average 4 times energy goes into the heat compared to useful signal. Its actual performance depends largely on the spare cooling capacity. It appears none of them operate nowhere near theoretical full capacity. Here is an interesting paper (F-16 discussed) to read and I think it complements nicely the calculators: https://www.researchgate.net/publication/337985199_On_the_use_of_AESA_Active_Electronically_Scanned_Array_Radar_and_IRST_InfraRed_SearchTrack_System_to_Detect_and_Track_Low_Observable_Threats/link/5df8fdb7299bf10bc3633791/download

 

IMHO: in future radars designers will probably try to find the substitute for GaAs TRMs, which develops less heat and/or they may focus on advances in cooling systems (e.g. graphene, etc).

 

 

I mean don't consider DCS to be at all real picture of the real world, thats mistake #1. There is too much missing or done wrong for it to be anywhere near RL. That being said the F15C and Su27 are both FC3 lo-fidelity models compared to any of the FF modules, and their radar modeling as drastically oversimplified with the nod going to the flanker for realism. But at the end of the day you are new to DCS and well, the problem is most likely that you don't really have much experience and alot to learn how on how DCS works. Both of those platforms are employed in online PVS with good success even though they are "outdated" but are still kinematically superior to the enemy which if you know what your doing is what actually counts in DCS if not RL.  If the F15C was done right radar wise, you wouldn't drop tracks anywhere like you do since notching the radar wouldn't be as easy because it has a coast mode and the overall range would be better, so you would have a launch advantage on anything currently in DCS. The other big "failing" of DCS is that it doesn't really model how SA is built in the real world, i.e. GCI/AWACS, so the DL equipped planes are at a major advantage there since A) ED doesn't model GCI or Awacs well if at all. B) DCS doesn't model any limitations/problems with the actual datalinks. That can be overcome online on servers that "Fix" that problem through scripts or with Human GCI's which is where those platforms come into their own. 

 

And yeah, thats an old paper. If you want to readup on some radar design stuff you can search for the APG-66 papers that are out there that will explain the design choices and tradeoffs that radar made relative to others.

 

 

 

[henshao]